It doesn't seem to exist yet. The specifications are not specifications, they are design goals. I don't see how they can get the color coverage they're claiming with RGB LEDs.
Seems about as credible as a lot of the crowdfunded stuff.
Strobing RGB LEDs is how Epson's professional flatbeds do color. That's one of the more plausible aspects of the design, I think. No mention of IR though, so I suppose that means no dust correction.
The custom software package is clearly in its foobar stage. Loving the word "TextLabel" surrounded by a bunch of padding.
The issue with LEDs, is very pure colors. That’s actually a bit of a problem, with film scanners. You need a smooth curve, and it needs to extend out a bit. You don’t want areas of color being missed.
The Coolscans had a light color response (think the “levels” screen, in Photoshop) that looked like three steep hills, with minimal overlap, but they were able to make them wider than a “pure” LED. Coherence is a feature of LED lighting.
Most previous light sources used filters over a white light, and they looked “sloppier,” with a lot more overlap, so there was more coverage. We had to correct for the unusual color coverage of LEDs.
One of the things that the Coolscan did well from a hardware perspective was that it made the transport sprocketless which allowed damaged film to be easily scanned, and it also allowed non-standard formats to be scanned easily as well. I’m curious if they have a sprocket driven system for this or if it utilizes a similar system as the Coolscan - I’ve used many scanners and the Coolscan is still the best/most convenient because of being able to just sequentially scan an entire roll of negatives.
I assume the LEDs were matched to the typical pigments used in films though? Because otherwise metamerism just wouldn't work, RGB mixed to some CCT is not white light and can't illuminate arbitrary pigments with any kind of good color reproduction.
I assume so. The folks that designed the scanners were no slouches. I suspect that they never completely turned off any LEDs, so there was always some deliberate “slop.” With LEDs, however, you can explicitly control that. They probably had some kind of filter, also. I never took one apart, though.
I got the response curves by feeding in a special slide with a diffraction grating.
The curves were markedly different from an incandescent light source.
It seems like folks buy a used Coolscan, scan their stuff, then sell it. They seem to last pretty well. I'm about to buy a used one to scan my Dad's old slides. And then sell it.
The slide feeder is good but it's worth being aware that if you have slides mounted on cardboard (I had a lot of old family photos like this I used it for) it will often grab a couple at once. You can fix that by clipping eg a driver's licence in the right place to narrow the gap it pulls the slides through, but it will still need some manual supervision.
If you get one, have a look at VueScan on the software side - the original software needs (I think) a Windows XP virtual machine to drive it.
They also sucked. I had an LS-2000 and the images were noisy as hell and it couldn't scan negatives for shit. I sold it on eBay. It's incredible how overrated the Nikon scanners were.
In the end I found a new in-box Minolta Dimage Scan Elite 5400 II, pretty much the end of the line for film scanners. I haven't tried it yet!
The 4000-5000 series Nikon Coolscans sell for about the same price they did 20 years ago because they still produce excellent scans and there’s nothing quite as good for that $1000-$15000 price out there.
For example I wanted to look at the first picture in the horizontal gallery that scrolls horizontally when you scroll vertically. However, there is no way for me to view the whole image. Either it is cutoff at the bottom, or it starts horizontally scrolling. Switching from vertical to horizontal scrolling is awkward and I just want to skip the gallery.
scrolling on that page feels slow, sluggish, and if you switch to spacebar, you actually miss significant content since it only loads/becomes visible halfway into the page.
Like others have said, dust is a huge issue. Some film labs cut film into short strips. some film is just a single image (for example if previously cut to fit into slides).
The film is designed to form into a coil. So, if there's grit or any hard material you'll end up with scratches on the negative itself.
--is it only 35mm as well? I don't think I see any mention of formats it supports. So I can only assume it's just 35mm.--
EDIT: found the 120mm section in the FAQ.
In my recollection, NYT started/popularized the whole "reportage as a interactive story with stuff moving as you scroll" on the web, but I think it was web developers wanting to emulate the Macbook/iPhone inertia scroll that started the whole "lets override scroll behaviour" trend.
The first I ever saw of it was the NYT story about the deadly avalanche at Stevens Pass in WA state. I liked it for that story but not so much for this product site.
As someone who has a mirrorless scanning setup for my film, and pondered getting a dedicated scanner... the price of this is quite steep given how inflexible of a tool it is.
A second hand DSLR setup is going to be roughly the same price or less. I'm also not sure what kind of workflow improvements it actually offers. If you want fancy and experimental, filmomat has arguably a more interesting but pricier offering.
But naysaying aside, I hope they manage to find a niche that allows them to survive as a company, and keep the analog photography revival alive.
I bought some time on a hasselblad medium format scanner (took fucking ages)
The results are good, as you'd expect. However can I tell the difference between that and me putting the negatives on a decent softbox and using a fancy camera to take a picture? yes, but not by much.
I think the main issue is film registration, that is getting the film to be flat and "co-planar" to the lens so the whole frame is sharp.
My negatives are slightly warped, so they really need a frame to make sure they are perfectly flat. But for instagram, they are close enough.
However scanning more than a few pictures is a massive pain in the arse. If I was scanning film regularly, then this is what I'd want, and its cheaper than the competition.
Assuming that its actually any good, I haven't seen any scans yet.
It'd be nice if they were able to adapt the Hasselblad/Imacon "virtual drum" concept and curve the film underneath the sensor for side-to-side flatness. I wonder if that's feasible with a 2D sensor.
Thats a good question. I wonder if the "virtual drum" was there to get over film holding issues (as in it physically bends the film) or that its a line scanner
> Thats a good question. I wonder if the "virtual drum" was there to get over film holding issues (as in it physically bends the film) or that its a line scanner
It's not - the issue that still remains is keeping the film flat, and this is especially problematic with smaller formats. With current solutions you can get the resolution but not the flatness, or you sacrifice something to get the flatness (e.g. ANR glass holders). It's the old glass vs glassless carrier debate, applied to a modern workflow.
I repeat myself: focus, DPI / resolution, dynamic range - these are the solved problems. In fact, modern medium format digital cameras are superior on all these factor. Keeping the film flat, however? Only drum scans and the Imacon "Flextight" solution do this well.
Of course, it depends on what you plan to do with the scans and for 99% of people the solution in the link above is more than good enough.
I've written about this previously https://leejo.github.io/tags/scanning/ # I'm going to add the fourth, and hopefully final, part in a couple of months time.
I almost missed the price. Wow you're right that's a lot! And the final retail price is 1599 euros. I have a good Plustek that cost me $300 or $400. Automated transport and unantiquated software sounds nice, but those features are not worth an extra $600-1,100 to me.
Am I missing something or is this supposed to be in another tier of image quality?
Honestly I feel like anything beyond 5 megapixels per frame is pushing beyond reasonable expectations with 35mm. This is certainly the case with any kind of available light or high speed work in silver-halide process, the area where I figure most people are going to be using this device. Lab-work in C41 and E6 is definitely possible at home but must account for single digit percentage of the home analogue market.
A 4000 DPI scan of 135 gives you 21 megapixels. So 36MP with a good lens will easily resolve just as much detail. There is not 60-70MP of information in a 4000 DPI scan, period.
For most films, anything beyond 4000 DPI is just going to help resolve the grain particles or dye cloud shapes. You have to be shooting slow fine grained BW with the best lenses to need more.
"…beyond 5 megapixels per frame is pushing beyond reasonable expectations with 35mm."
Well, as I mentioned elsewhere old fashioned Kodachrome resolves ~100 lines/mm and some newer color emulsion are considerably higher, and of course B&W ones have even higher resolutions.
Given that a 35mm frame is 36x24mm even Kodachrome achieves 8.64 megapixels. OK, let's allow for an overgenerous Kell factor of say 0.8, this figure will drop to ~6.9 megapixels. Given the ready availability of emulsions with higher resolutions, especially the best B&W ones then a figure well in excess of 5 megapixels is relizable in practice.
Of course, that doesn't take into account the image chain as a whole, lenses, displays, compression, etc. which would reduce the effective resolution. That said, these days the typical image chain can easily achieve much higher pixel throughput than 5 megapixels before bandwidth limiting so the effective Kell derating factor could easily be kept quite small.
I think I see what you mean. It’s the difference between having an image showing the shape and texture of each film grain, and an image which looks like what I saw in the camera and which isn’t going to be any sharper. The former has value but the latter was always good enough for me and, surprisingly, rather low in resolution compared to subsequent DSLRs and mirrorless cameras I bought in the 2010s.
Ilford Delta 400 pushed two stops to 1600 ASA in a 1970s Asahi Pentax SP1000 was always going to produce… artistic results, requiring as much imagination as acuity to appreciate the subject. (Read: see past the blur.)
I wondered what the price was. 1599 seems pretty decent.I was expecting about 4k This is about the price of a venerable Nikon 5000. Some of the setups use film mounts that cost as much as this whole unit.
> A second hand DSLR setup is going to be roughly the same price or less.
And if you get one with Pixel Shift, you can get way higher resolutions than the 22MP they're offering (e.g. my cheapo Olympus gets 40MP JPEG or 64MP RAW from a 16MP sensor.)
You’re for sure exceeding the linear resolving power of 35mm film at 40MP or 64MP.
However, a Bayer-filtered sensor has lower color resolution, since each pixel only sees one color. So the pixel shift really helps quite a bit here since the sensor (and Bayer array) are shifting relative to the film multiple times per exposure.
High-quality film scanners maintain color resolution by using linear sensors without Bayer filtering. But they’re slow and expensive.
All the current Nikon Z bodies (and probably other brands too) have different levels of pixel shift where it’ll take 4 or 8 images and basically cancel out that it’s a bayer sensor. The bayer array is a 4 pixel pattern, so it moves one pixel to the right then one down and then one back to capture all 3 channels for each individual pixel. For things like film scanning it works flawlessly, I use it all the time.
Then it’ll do a 16 or 32 shot stack in order to do the same thing but with more resolution.
The scan is the least of the problems - good luck getting to that level of detail with mostly vintage lenses, balancing depth of field and diffraction, keeping the film perfectly flat, on a stable enough tripod with no vibration whatsoever; developing perfectly in the dedicated developer. Yes, it's impressive but no, it's not relevant to the average user or hobbyist.
Filmomat looks fun. Many money. Love the hipster flex with the Weber HG-1 in background of the demonstration video. I do own an Intrepid enlarger (sort of experimental?), and I used to live near Ars Imago in Zurich who sell a "lab in a box", similar to Filmomat's Light system. The independent dev scene is pretty great, though none of it is particularly cheap and is rarely open source, which is disappointing.
> I'm also not sure what kind of workflow improvements it actually offers.
The obvious one is auto-feeding and portability, but without using it who knows. It doesn't offer IR, but even Filmomat's system needs a modified camera. You get that with most flatbed and Plustek-style scanners. I have a V850 Pro which wasn't cheap either, but it'll do a full roll in one go and I can walk away. Even if I shot a roll a day it would be more than fast enough. It has occasional focus issues, and you need to be scrupulous about dusting, but it works well enough. I've never been a huge fan of the setup required for copy-stand scanning and it's tricky getting the negatives perfectly flat in/frame. The good carriers are also not cheap, look at Negative Supply for example.
Frankly it also looks great, like the Filmomat. I think some of the appeal is a chunk of modern looking hardware and also the hope that it's maintained? My Epson works well, but I ended up paying for VueScan because the OEM software is temperamental.
The biggest pains of film scanning are in the post process color balance and dust removal. Unless this can improve those parts of the workflow, it's only going to be a minor improvement. I like the continuous reel movement of this scanner vs a flatbed though, that can improve the physical workflow quite a bit.
PS to add more - I am unable to scroll, all I see is the picture with the dark background. If I use arrow keys instead of the touchpad, I can scroll a bit then after a second or so the page snaps back to the top. I have Firefox on MacOS.
(I know the HN rules say that we should focus on the contents rather than criticising the technical aspects of a website, but in this case the contents are not accessible).
> Yes. We’re collaborating with several film labs in Berlin to benchmark Knokke against Fuji Frontier and Noritsu scanners.
> Sample results will be published before the Kickstarter campaign, so you can make a fully informed decision.
I don't care how cool your scanner looks or how "modern" the workflow is - it's samples or nothing. Additionally, if they were really smart, they'd collaborate with a well known film photographer instead of using someone's walk-around point-and-shoot photos.
I'm glad this exists but at ~1k EUR I would be interested if it could scan 120 medium format negatives...but the fact that it does not is an absolute deal breaker for me. It seems like they are considering it. I hope they do figure that out sooner than later.
Epson stopped making their flatbeds that do film, reportedly because they can’t get the CCDs anymore. That may be a rumor.
The result is they go for 2x MSRP on eBay for models that are many years old. Because that’s all that exists.
Without that, you can buy the kind of scanner meant for a photo lab ($$$$$), DIY it with a DSLR ($$$ if you don’t have one), or pay your a lab a lot per roll and hope they do a good job.
I’m not saying it’s a giant market but it certainly seems to me like there’s enough of one that it could support a small product.
You can get brand new Plustek OpticFilm scanners for 35mm and smaller starting around $300, and there are plenty of other options above that. Plus the DIY.
I’m sure 35mm is easier to make and certainly a bigger market but it’s also a lot more crowded.
I expect their specs are far better than the $300 one I’ve mentioned, I don’t know enough to know. But medium format people are desperate for anything.
> Epson stopped making their flatbeds that do film, reportedly because they can’t get the CCDs anymore. That may be a rumor.
Wow, you weren’t kidding, I completely missed this. I bought one, sold it, then bought and currently own another. I better baby it, there’s really nothing like it out there.
I tried some scanning on a Plustek 8300, which is supposed to be the fastest. The process is still extremely manual/slow and I don't think it's practical on a large scale. Many families who owned cameras in the 60s-70s-80s-90s will have potentially thousands of negatives to scan, but I don't see a solution that will automate that digitalization process.
Software could also use some improvement. Automating batch correction and clean up should be easier, IMO.
This really isn’t my area but it sounds like nothing is fast. DSLR may be fastest without just flat out hiring someone else to do it. But even with thousands of shots that would still take quite a lot of time.
And yeah, workflow is the thing that seems the worst. That seems like a great place to try to improve things to get a sale.
Not that they're cheap, but you can get Imacon scanners for much less than they retailed for. I inherited a Flextight Precision II and it still does a great job.
Do you use it with an old computer, or do you have a good way to interface a SCSI scanner with a modern machine? I tried to get my Precision II up and running but the SCSI card driver would crash Windows at random intervals.
This is way out of date. I have since been able to get it working on a Windows 11 4th gen Intel machine with 64-bit drivers cobbled together from a couple of versions of FlexColor and some .inf modification. It's not flawless, there's some major corruption that can occur when trying to use certain operations, but overall it works for my needs.
From what I’ve seen people mostly used the Epson scanners like the V600, V700, V850, etc.
They stopped making them early this year. Only the top end model for $1500 still exists and I don’t know if that’s because they still make it or just that there is still stock left at Amazon/etc.
Are you taking about other models like the v600 and v700? Because I’m not. I’ve never really seen those models with film scanning tools, it was usually only the v850 that came with film scanning tools. So I have always expected to have to stump up for a v850 if I wanted to do film scanning.
I scanned a lot of 35mm film, of various kinds, using a high-end flatbed scanner (EPSON V700 Photo). The biggest problem is not the optical quality, but mechanics: flatbeds can only scan strips. And if your film is very old and has been stored in a roll, you might not want to cut it, and even if you do, getting the strips to stay flat is nearly impossible.
I tried various fancy holders, but in the end decided that I'll likely have to make my own holder from aluminum or steel sheet metal. And even then you run into the problem of lengthwise curvature. For those that are unaware of the problem with this, these scanners have a very limited depth of field, in the range of 1mm or less. So if your film is bent, some of it will always be out of focus.
I can't see much on this fancy webpage, because they made it so fancy that some of the images do not load and those that do load are oh so mysteriously dark. But if their scanner can scan both heavily curved rolls and strips, I will be buying it.
As to optical quality, if you can get your film to stay flat, this is a solved problem, that Epson mentioned above can produce fantastic results (more pixels that you want, generally).
Doesn't always work. I've got old Agfa negatives I developed from my grandpa in Korea in the 50s. Developed them after finding them in his attic maybe 10 years ago now. They sat between two panes of glass for 5 years with volumes of books on top, not a single change toward flat. I finally gave up and just put them in the archival sleeves and in the binder with the curve.
I'd never run film at claimed top speeds unless it was a disposable copy print, but seems exciting. Long way have we come from linear array cameras and telecine to these. Different light sources would be useful for dust and scratches mask. $1k price is also insane. For full mechanical assembly with casters etc it probably comes within five or ten, but still a bargain. I wrote here years ago where I had a side hustle scanning and processing ye olde reels with an absolute beast of a scanner (think room sized) that did 2k in real time with an SGI and Hippi fiber network. Tech almost interesting as films themselves. :)
Funny thing is, in general, low and midrange desktop scanners that public can generally buy, haven't changed much in 10-20 years since they started using led lights and IR dust removal (Canon Fare, Digital ICE or similar stuff). Some are even the same hardware just slightly rebadged or with a different USB connector. But they're the same price or more expensive.
And, at a different level, professional film scanners are EXPENSIVE. Lots of people are now scanning their film using a digital camera and led backlight (now that there is affordable good quality led lights) instead of a dedicated scanner. But that's not very fast and requires some extra manual work. If this scanner offers reasonable quality and a good workflow (that not very proprietary or closed), 1000-1500 dollars is a reasonable price, especially if you have lots of film coming in, or an old collection to scan.
I could imagine my dad buying one of this to scan his hundreds and hundreds of rolls from the 70s/80s and then selling it once he finish. It would be like 1-3 USD per rolled scanned :)
In the "things that fit on your desk" category, slow is pretty relative - people on ebay still want thousands for Nikon 9000 scanners, and those were not fast if you were scanning high-res 16-bit. And always needed time in post too (cropping/straightening/tweaking exposure and color). An ILC kit can go a lot more quickly.
This one would be much faster than the plustek though, if they keep their promises. I use a 8200. It takes more than 1 hour to scan a 36 frame color negative with dust removal on, and it requires constant manual input. Pushing the holder to the next frame, unloading and loading the holder, etc.
It's fine, sure. For the price I paid for it and the image quality I'm getting, I have no complaints. On the other hand, a new device that can cut the time down to 5 mins with modern software support (silverfast is kind of dated and VueScan will run you another 100), while priced at 1000 EUR, is not cheap, but also not that unreasonable tbh.
For a studio or a professional - sure. For me, an amateur, who simply does film photography for his own amusement, it makes no sense. I also use standing development process that takes around 40 mins. So the speed is irrelevant for me.
When it comes to the light source, it's a huge missed opportunity. The next wave of film scanning involves capturing Red, Green and Blue separately (separate narrow-band LEDs) to better isolate C-41 layers and counteract crossover on the image sensor. And yes, it should have an IR pass as well, for dust removal.
With that said, I'm happy to see new film products released in 2025/2026. Hopefully this is just the first at-bat.
I think the problem is that it sounds like you get worse results for slide film with RGB than you get with C41 and white light. So the tradeoff is only worth it if you shoot no slide and C41.
I still have an expensive Canon dedicated slide/film scanner from 20 years ago.
IIRC at some point their value started going up as they became rare.
Mine did something like 50MP scans of 35mm film/slides. The quality was more than enough.
But it was painfully slow.
This thing is not 100x faster, so I think it's still painfully slow. If it takes 5 minutes to do a roll of 24 that still means someone with hundreds of rolls needs to have a lot of time on their hands.
Not sure I can actually figure out software to get my old one to work FWIW, but I don't think I care to deal with it, I have a big enough mess dealing with the ~200k digital photos that are already on disk.
Seems weird. The fact it talks about rolls of film and reading DX codes suggests that it's able to scan undeveloped film. Maybe that's possible, but I've never heard of that before.
If it only scans developed film, then it's unlikely to still be in the cannister with DX codes, and I've never seen that film delivered to a customer in a roll - it's normally cut up into strips so they can be stored flat.
Developed film has a bar code type of encoding for DX info. The talk about the whole roll is the ability to scan the entire thing in one go vs having to load in multiple strips.
Any suggestions for a scanner meant for bulk scans of old family photos (think a few thousand images)? I bought, what I thought, was a reasonably solid scanner, the Pacific Image Powerfilm scanner but the software is so janky that it hangs every two strips and has to be restarted making the entire process super labor intensive. Also the entire "bulk feature" where it's meant to pull the strips one at a time iis not even close to working.
Granted my film scanner (epson v750) has gotten on in years the more prosumer-professional Epson scanners were good, software was still a bit janky but IIRC there is aftermarket scanner software for them.
It can do 36 exposures but you have to cut them into strips and place them in a carrier but it isn't terrible and if you store your negs in film protectors you are cutting them down anyway.
I am fairly sure the newest version (V850) is the same but be aware they aren't cheap, at least $1k+ USD but still cheaper than the next level up which are pro drum scanners and they are many orders of magnitude more expensive.
I applaud their intent to be repairable, but would really like to see a commitment to open sourcing the software under specific conditions (time, EoL, acquisition, closing down, etc).
> “Is Knokke open, repairable, and long-term supported?”
> “Absolutely. We're committed to building a scanner that lasts decades.
All schematics and repair manuals will be publicly available, replacement parts can be purchased directly, and the software will remain supported for as long as possible.”
Yes. Our control application, Korova, will be fully open source and maintained long term. It’s a native, lightweight application for Windows, macOS, and Linux.
I for example develop my (black and white) films myself and leave them to hang dry uncut, then proceed to store them in a box. When I have the time, I "scan" them using a MILC on a tripod, a 3d printed holder and a LED backlight panel. Afterwards I cut the strips up and file them properly for long term archival. Many of us do it this way.
Disappointed to see that the first reactions on HN are so dismissive.
I'm both amazed and really pleased to see anyone attempting to launch a totally new scanner in 2025, and genuinely hoping the actual scans are really made at the resolution and color-depth claimed in the text: too many recent scanners are simply upscaled, lower bit-depth devices marketed with exaggerated specs.
I also have a Nikon Coolscan 9000, so I'm not immediately in the market for this. But I don't expect the Coolscan to last forever, and the Firewire connections on the machine are already abandoned by Apple, who chose not to support the cables in their latest Operating System - so eventually I won't be able to connect it to a new computer.
The open source aspect seems worthy of attention if nothing else. Even if this is a middle-of-the-road scanner - the community being able to customize, improve and support it would be incredible. Especially considering your scanner is considered one of the best despite being over 20 years old.
For the "this is crazy expensive" crowd: this competes with the Pakon F135 - an ancient lab scanner that involves booting up Windows XP, finding old drivers on a CD, etc
No, this competes with a lightsource, some stand and a macro lens+DSLR/M. It's a good price if don't have the later ones. But chances are high that you do if you are into photography.
5-10 minutes depending on equipment/skill. You "lose" most of the time if your strip is cut into stripes of 4-6 images, otherwise you can do it in 2 minutes.
Seems like too much time was spent on the concept and website and not that much on the product. I can't help myself but always think those projects are too marketed and engineered to be honest. Film scan should be simple and straight to the point, we don't need this Apple aesthetics and ultra marketing.
"Designed by photographers, for photographers." Nice. Would love to see your pictures then.
Without any samples it’s hard for a $999 kickstarter project, considering a Epson V750 scanner costs much less than that but already provides great quality and supports more formats
I’ve been using an inexpensive Epson V600, I think I paid under $300 and it came with a license for the scanning software which was another $50-$100 value. I also got $25 anti newton ring inserts. The process is SLOW but the quality I think is excellent unless you need to print XL and show your work at MOMA.
I thought you use a camera on stand is the common way now. Did one in Edinburgh and ordered one back home waiting for delivery. After decade of epson it is not comparable for speed and option …
for just contact sheet fast … you can just move and push the blue tooth button … for one particular treasure slow … you can do pixel shift and focus stacking …
Scanning 135 format at home is pretty much a solved problem right? The home made solution to this costs $0 if you own any DSLR and some other basic photography gear.
I think the product would be more compelling and worth it or even a good deal at the price they are offering if it offered drum scan-quality for larger formats.
The workflow for this scanner would allow you to thread an uncut roll of 35mm film through it. You'd have to spend more than $0 to get that kind of speed on a DSLR rig.
1. I had never even heard of an uncut developed roll of film before, so I guess it's useful for that.
2. Time is money, but who is honestly shooting that much 135 film that it's worth 1600 Euros to buy a faster scanner for it? I don't think a museum wants to feed degraded film through a fast scanner, and surely pros who still shoot film would use a larger format, since that's where it has some differences / advantages compared to digital?
> I had never even heard of an uncut developed roll of film before
That's how film is developed. Someone at a lab has to cut it.
> who is honestly shooting that much 135 film
How about a film lab? A place where "uncut developed film" is extremely common.
>it's worth 1600 Euros to buy a faster scanner for it
Price is 999 euro.
> pros who still shoot film would use a larger format
Some do, some don't. It depends on the project. I'm a little surprised by your comment looking at your history. You say you're a retired professional photographer and you've never heard of "uncut developed film" before? If you're retired in 2025, you must have been working when all photography was on film. You never developed a roll of film before?
Some influencers that make money directly from their photos could find it beneficial. Although as the saying goes, the fastest way to make money with photography is to sell your gear!
This is cool. I spent ~$1k on a Pakon F135+ a number of years ago, and the workflow was indeed extremely frustrating and the results not that great. If I were still shooting a lot of film I’d definitely pick this up. But we need to see sample images!
Can I ask why it will produce terrible color rendering? In addition to commercial scanners that used narrowband trichromatic (RGB) light sources, hobbyists are creating their own RGB light sources to digitize color negative film claiming superior results and putting forward arguments why this is better:
(NB: Most film I shoot is slide film, which I’ve been told doesn’t benefit from RGB light sources because it’s intended viewing was projected with a broad-spectrum white light [likely a warmer than daylight (but color temperature isn’t much of a concern for digitizing slides)] so I haven’t dug into this much.)
RGB should be fine, especially if you use a genuine ultra-high CRI source. A few companies make them, I think Waveform is one of the more well-known. If you really want to spend money, the optics suppliers like ThorLabs sell broadband LED sources. In an ideal world you could calibrate the image sensor against a known spectrum so you'd know its response. If you can estimate colour to a reasonable degree then you can transform it to what it "should" look like. Nevermind that pixels are single-wavelength as well.
Negative Supply use something similar in their light tables, though I don't know exactly what the source or spectrum is. They're highly regarded enough that I think it's not an issue.
You can also use LEDs for enlarging, but you need to be careful about buying the right bands for the paper. I've used Luxeon SunPlus with some success as you can buy the correct green/blue for the different contrast layers. Though for B&W, even a random 5500K module from Cree worked quite well.
I got one of these from the latest batch last week and I’m not entirely convinced by it yet. I need to experiment some more but I went back and did a couple rolls from this summer and so far I think the cslite warm setting + negative lab pro results are better and more consistent. I’m still getting some wonky colour casts with it. It’s nice that the control app lets you change the power of each LED colour separately, so that’s the next thing I’m going to experiment with.
I’ll also note that negative lab pro hates negatives that are scanned with it. They don’t turn out at all. If you’re using it, you should expect to be inverting them manually, which is kind of a pain. I was quietly hoping (but not expecting) to still see some of the benefits of it when passing them through NLP.
In case anyone is wondering, the current meta in home 35mm film scanning is divided:
Option 1 is to get an Epson Perfection series flatbed scanner. V800 or V850. This approach is highly automated and you get automatic dust correction with color film. But, leading software packages don't support Linux, and the quality for 35mm negatives is just okay. Performance on medium and large format is SOTA.
Option 2 is to assemble a scanning rig with a DSLR and a light table. This approach is fiddly and requires a lot of space, but with some tuning, the 35mm scan quality can beat flatbeds.
There are some other, more obscure approaches, like vintage Minolta and Nikon scanners, but unless you have a PC with a parallel port laying around, you're gonna have a hell of a time getting those working.
None of these options are good, and if this thing can really perform, I'd happily drop $500+ on it.
Side note: Those little toy scanners like the Kodak branded ones on Amazon are atrocious. Avoid them. If you need to scan some family photos and you don't want to break the bank, go to your local photography store. They could really use your business.
DSLR is the most flexible way, cost wise. You’ll likely have a DSLR already if you’re a film photographer, and You can get a decent scanning setup for under $200. Being able to edit RAW files is also more flexible. It’s a lot more fiddly though so not ideal for high volume work.
Don't count out the flatbeds. I don't have the space (or the patience) for a DSLR rig, and I've been quite happy with the output from my Epson V800. I bought mine used for $700 or thereabouts. For the amount of time it's saved me, the added cost over a DSLR and bellows/macro lens was worthwhile.
Hmm, does seem pretty expensive but sounds interesting. I've got an old Canon FS4000 for 35mm, which works ok for me. I'm curious what people recommend for 4x5 film.
I don't think you want a drum scanner, look up a video of the process for wet mounting negatives... It's super time consuming, cumbersome and messy. Something like an Imacon looks like the best middle-ground, shame they don't make them anymore...
I've been camera scanning 4x5 and I'm happy with the results. Take two offset photos and stitch them in post. Mind you, I scan with pixel shift for higher res.
Maybe a mirrorless, a Macro lens, and the Valoi adapter. This is what I use now after having painfully used a Flatbed Epson for years. This set up is fast and sharp. Best of both worlds.
Presumably the negatives, but not the pictures, as this is intended to scan only negatives. You'd probably have better luck with a cheap flatbed scanner that comes with adapters for slides and negatives. (I've used both dedicated film scanners and flatbed scanners to scan thousands of rolls of film and prints taken over a span of 50 years.)
No. 4x6 negatives are large format and requires a different type of scanner (flat bed). If you look at the photos the 35mm roll slides into the device so you can’t fit a larger format sheet into the feed.
Film doesn't really make sense anymore outside the realm of luxury-budget art. And if so, why 35mm? Why not medium format / Hasselblads, or large format? Why not actually go for the format sizes where there's a reasonable argument that film can produce a better end result compared to DSLRs?
35mm wins by a mile on price per shot. And for web scans (where most photos end up), it's got more than enough resolution.
35mm film and 120 film are a similar cost per roll, but with 35mm you get 36 exposures vs 8-16 exposures on 120 film (6x12 and 6x4, with square 6x6 in the middle at 12 exposures). And if you shoot half-frame, the cost/shot really goes in the 35mm direction.
That said, I have a handful of of 35mm cameras (all fixed lens vintage rangefinder) and a post-war Zeiss Super Ikonta IV (6x6 120 format). The Olympus 35DC is my favorite of the bunch - it's automatic except focus - really sharp and fast lens - just a pleasure to use. And a Polaroid Go 2 because it's just dumb fun (way overpriced for the quality, and sensible people buy Instax cameras instead, but the Polaroid form-factor was just too much for me to pass up).
I shoot film because it makes me slow down and think a bit. With my mirrorless cameras, I'm too prone to spray and pray and sorting through hundreds of shots can kill the fun for me. That, and the film look is nostalgic for me - sometimes I just want rough snapshots - feels more like a memory vs the crystal clear high res digital output.
> Why not medium format / Hasselblads, or large format?
That has also been superseded in the digital realm. I can use a top DSLR from the last decade to blow film medium format cameras away under most conditions, especially in low light. If I just use a digital medium format camera, it invalidates almost any rationale for film medium format (there is a minor, minor argument to be made about depth of field when the lens is wide open).
Large format can squeeze slightly more resolution out of an image, but the ability to actually use that extra resolution is rarely satisfied. Again, a MF digital subs in for almost every conceivable use case.
The reality is that people do this because it is fun, or a challenge, or for a feeling. Basically, its interesting. The technical excellence of the 'end result' is secondary.
There's tons of things that people do that don't make sense from a purely pragmatic point of view, and that's what makes the world so much more fun to live in.
But these cost several thousand dollars, due to them still being relevant for contemporary professional applications. Medium format bodies can be found for hundreds of dollars on eBay. Even at $1/shot, if as a hobbyist even if you shot 4 rolls (48 shots total) in a weekend, you'd be shooting that much every weekend, every weekend straight for years before a top digital body would break even for you. And you would have more fun and a more meaningful time due to the more deliberative process.
35mm is super cheap to get into if you just want to experiment. Old 35mm cameras and even home darkroom equipment are gathering dust in tens of thousands of households, cheap on FBM or thrift shops.
It's expensive compared to digital for snapshots, but if you enjoy working in the darkroom as a hobby, you can probably get everything you need for free or cheap.
I don't see any infrared sensor. This makes it drastically worse than the competitors on the market, as it's an essential sensor for dust and scratch removal.
Did they not research the competition?
I can buy a brand-new 7200 (virtual) DPI machine with infrared, proper color metering, wide software support, and multi-exposure system for $400, less than half the price of this offering.
It also supports slides and single frames, whereas this has a min. of 3 in a strip.
Yes. From the article: "It's [sic] final retail price is set at 1599€"
I was actually excited from the title but upon reading further I got really disappointed. It's basically wishful thinking right now, not sure if it's even going to be released. Not sure why this is posted here.
yeah static mostly. You can get a pizeoelectric antistatic gun to neutralize it. You can also get antistatic brushes that use alpha emissions from polonium 210 to do this.
every optical system attracts dust - dslrs / mirrorless cameras with removable lenses have extensive hardware and software systems to handle dust detection and removal from the sensor.
I gave up trying to read the article when animations started happening when I scrolled. It's annoying and I have better things to do than waste time waiting for your animations to stutter and finish moving around when I'm trying to scan the article. Good luck with whatever you're trying to do.
JFC, for that price I can get a NiB Epson Perfection v850 Pro, and have something that can scan film almost as effectively and is a full-blown generalist scanner, to boot. The best consumer scanner ever released, IIRC.
I doubt whether this product will usher in a new era of film scanning for many photographers. For starters, as it's only aimed at 35mm negatives it's unsuitable for my needs.
As with many photographers, my collection consists of B&W and colour prints of various sizes and formats, 35mm B&W and colour negatives in both rolls and cut strips, 35mm slide/reversal material both in rolls and as mounted slides. Film stock covers many brands including Agfa, Ansco, Fuji, Kodak—including its Eastman movie emulsions—and others. Kodak holds special place, with Kodacolor, Ektachrome (including infrared versions) and Kodachrome. At a guess, I've have about 30,000 Kodachrome slides alone. And that's not all, I've also larger format photos, prints, B&W and colour negatives and reversal stock.
Most of this material has still not been scanned because of the challenges involved, for instance those in the know will be aware of the difficulty of scanning Kodachrome slides because of residual silver that's still in the processed emulsion. Then there are scanning difficulties, mounting various formats (slides, rolls of negatives, etc.) and technical difficulties such as focus adjustment, avoiding Newton's rings, etc. Simply, I've not been able to get the tech necessary to do what I consider an adequate job.
Restricting my comments to just 35mm I can confidently say there is NO 35mm film scanner on the market today that can do full justice to a large range of film types—except perhaps exotic and expensive drum scanners which are unavailable to the vast majority of photographers including many professionals. (Drum scanners are only found in high-end professional and technical environments, they cost upwards of tens of thousands of dollars and are a damn pain to use.)
Fact is there is NO film scanner on the market today that can faithfully reproduce in digital form the full dynamic range and resolution† of old fashioned chemical film emulsions. I say 'old fashioned' because modern digital photography, HDR etc., is capable of much wider dynamic range, resolution and colour gamut than film emulsions, so it's not a technology limitation (converting the limited dynamics of old film ought to be easy but no manufacturer makes equipment that does). It's really shameful that no manufacturer has stepped in to fill this technical gap when clearly the technology is available to do so.
Below, I've kept to the basics, an in-depth comment would be much more detailed:
• Argument goes that no one would pay for a film scanner with those specifications—as its manufacture would require precision/exotic tech, and anyway it's doubtful anyone would notice the difference with currently available scanners. I question both those assumptions as I'll explain.
• Leaving drum scanners and a few very expensive ones aside, in the past the best 35mm scanners on the market were the Nikon COOLSCAN range but Nikon discontinued them some years back and nothing has equalled or replaced them since. They were not perfect but they had the best optics and overall provided the best resolution and dynamic range available of any scanner. The COOLSCAN's most significant limitation was its incredibly slow scanning speed (nothing much has changed here with the possible exception of this soke engineering device, film scanners have always had snail-like speeds for seemingly inexplicable reasons).
• Nowadays, for most photographers the best compromise between quality and usability are Plustek scanners, whilst they have neither as good a resolution nor the dynamic range of the Nikon COOLSCANs they are about the best available. I'd add neither are Plustek's mechanics for scanning films as good as the COOLSCANs (that said, in this regard the Nikons weren't much better than just adequate).
• So is there really a noticeable difference between a Plustek and a COOLSCAN? Yes there is, COOLSCANs have noticeably greater dynamic range in dark shadowy areas, and despite the Plustek having comparable resolution specifications with the Nikons the COOLSCANs produced visually sharper scans.
• Why are all film scanners so pathetically slow? A good question I cannot fully answer. Perhaps 20 or so years ago there may have been some excuse but even back then I'd argue they should have been much faster. For argument's sake even if the electronics had slewing limitations and had difficulty in processing images—which wasn't the case—then scanners could have been made much faster by simply increasing the number of rows of sensors—for example, increasing the rows from one to 10 and stepping 10 pixels at a time would increase scanning speed by 10. This is so obvious that it's mindboggling that it hasn't been incorporated into scanners previously. (Note, the other obvious option of photographing an image as does a camera has serious quality limitations.)
There's much more to say about speeding up scanners which I cannot cover here except to say have you noticed that scanners still use USB-2 and not USB3-3? Why?
• There are other significant issues that haven't been addressed adequately in many scanners such as colour calibration. For instance, every type of colour negative has a unique set of parameters often referred to as 'film terms'. In short, these parameters define how the destructive colour mask should be decoded (that's the orangy mask that's incorporated in all colour negatives). Many scanners only approximate or guess these parameters and expensive third party proprietary software such as SilverFast is needed to correct these limitations.
If I didn't know better I'd reckon the lack of a competitive range of high performance film scanners on the market was some form of conspiracy—electronics designers having an intrinsic distain for old fashioned analog film technology or such but clearly there's more to it than that. Whilst I can surmise reasons I'd only be guessing but for sure it has little to do with technical limitations.
Why the scanner crisis hasn't been a much hotter topic amongst serious photographers and professional reviewers has perplexed me. Perhaps if nothing else this scanner from soke engineering might fan the debate, it could perhaps force scanner manufacturers such as Plustek to upgrade their long-stagnant designs.
_
† Kodachrome has a resolution of 100 lines per mm which roughly equates to an image with 3600x2400 pixels (a frame being 36x24mm). Some films have even higher resolutions. Nyquist math says that the sampling rate should be doulde which means a scanner should be able to resolve to 7200 lines per frame but in practice no commonly available scanner comes anywhere near this figure. Diehards note, I'm aware this isn't a precise calculation but it'll do for argument's purpose.
The interesting thing is w gen Ai images, a reliance in terms of photographic evidence may come down to analog. Plus Kodak returning to film stock, the future looks bright for 35mm
The fact that there is still no sample scans has me heated - instead of showing us all these specs, how about some sample images!!
It doesn't seem to exist yet. The specifications are not specifications, they are design goals. I don't see how they can get the color coverage they're claiming with RGB LEDs.
Seems about as credible as a lot of the crowdfunded stuff.
Strobing RGB LEDs is how Epson's professional flatbeds do color. That's one of the more plausible aspects of the design, I think. No mention of IR though, so I suppose that means no dust correction.
The custom software package is clearly in its foobar stage. Loving the word "TextLabel" surrounded by a bunch of padding.
The Coolscans used RGB LEDs.
Can confirm. I used to write software for them.
The issue with LEDs, is very pure colors. That’s actually a bit of a problem, with film scanners. You need a smooth curve, and it needs to extend out a bit. You don’t want areas of color being missed.
The Coolscans had a light color response (think the “levels” screen, in Photoshop) that looked like three steep hills, with minimal overlap, but they were able to make them wider than a “pure” LED. Coherence is a feature of LED lighting.
Most previous light sources used filters over a white light, and they looked “sloppier,” with a lot more overlap, so there was more coverage. We had to correct for the unusual color coverage of LEDs.
One of the things that the Coolscan did well from a hardware perspective was that it made the transport sprocketless which allowed damaged film to be easily scanned, and it also allowed non-standard formats to be scanned easily as well. I’m curious if they have a sprocket driven system for this or if it utilizes a similar system as the Coolscan - I’ve used many scanners and the Coolscan is still the best/most convenient because of being able to just sequentially scan an entire roll of negatives.
I assume the LEDs were matched to the typical pigments used in films though? Because otherwise metamerism just wouldn't work, RGB mixed to some CCT is not white light and can't illuminate arbitrary pigments with any kind of good color reproduction.
I assume so. The folks that designed the scanners were no slouches. I suspect that they never completely turned off any LEDs, so there was always some deliberate “slop.” With LEDs, however, you can explicitly control that. They probably had some kind of filter, also. I never took one apart, though.
I got the response curves by feeding in a special slide with a diffraction grating.
The curves were markedly different from an incandescent light source.
It seems like folks buy a used Coolscan, scan their stuff, then sell it. They seem to last pretty well. I'm about to buy a used one to scan my Dad's old slides. And then sell it.
What's a reasonable model with a semiautomatic workflow (that I don't have to work manually frame by frame) for that? A 5000ED or something like that?
I could do that for my dad too...
Slides are going to manual because they’re (usually) individually mounted.
I haven’t seen any consumer scanner that has an auto feed. Good ones have a nice sprocket wheel but you still feed manually with a wheel.
You can get a slide feeder for the “modern” (early aughts) Nikon Coolscans, the SF-210:
https://studio-supplies.com/products/nikon-sf-210-239995
The slide feeder is good but it's worth being aware that if you have slides mounted on cardboard (I had a lot of old family photos like this I used it for) it will often grab a couple at once. You can fix that by clipping eg a driver's licence in the right place to narrow the gap it pulls the slides through, but it will still need some manual supervision.
If you get one, have a look at VueScan on the software side - the original software needs (I think) a Windows XP virtual machine to drive it.
They also sucked. I had an LS-2000 and the images were noisy as hell and it couldn't scan negatives for shit. I sold it on eBay. It's incredible how overrated the Nikon scanners were.
In the end I found a new in-box Minolta Dimage Scan Elite 5400 II, pretty much the end of the line for film scanners. I haven't tried it yet!
The 4000-5000 series Nikon Coolscans sell for about the same price they did 20 years ago because they still produce excellent scans and there’s nothing quite as good for that $1000-$15000 price out there.
I think the word "modern" gives away that it's going to be marketing vaporware.
There are a few scans on Instagram, I'd charitably describe them as "dogshit".
Ah, so bad. I got excited for a second.
I wish the fancy scroll nonsense would go away.
For example I wanted to look at the first picture in the horizontal gallery that scrolls horizontally when you scroll vertically. However, there is no way for me to view the whole image. Either it is cutoff at the bottom, or it starts horizontally scrolling. Switching from vertical to horizontal scrolling is awkward and I just want to skip the gallery.
scrolling on that page feels slow, sluggish, and if you switch to spacebar, you actually miss significant content since it only loads/becomes visible halfway into the page.
Like others have said, dust is a huge issue. Some film labs cut film into short strips. some film is just a single image (for example if previously cut to fit into slides).
The film is designed to form into a coil. So, if there's grit or any hard material you'll end up with scratches on the negative itself.
--is it only 35mm as well? I don't think I see any mention of formats it supports. So I can only assume it's just 35mm.-- EDIT: found the 120mm section in the FAQ.
https://dontfuckwithscroll.com
It’s a cancer on the web. Apple started it and I hate them for it. And I’m an Apple fan.
If I wanted to wait 1/2 second for each part of the page to load I’d have stayed on dialup.
Wasn't the NYT that started this?
In my recollection, NYT started/popularized the whole "reportage as a interactive story with stuff moving as you scroll" on the web, but I think it was web developers wanting to emulate the Macbook/iPhone inertia scroll that started the whole "lets override scroll behaviour" trend.
The first I ever saw of it was the NYT story about the deadly avalanche at Stevens Pass in WA state. I liked it for that story but not so much for this product site.
https://www.nytimes.com/projects/2012/snow-fall/index.html#/...
Oh that does sound familiar. I think you’re right.
Apple certainly made it popular for product pages though.
Nike mainstreamed it, I think - https://onepagelove.com/nike-better-world
On Firefox Mac, it doesn't scroll with the scroll wheel, though it does with the down arrow key. Go figure.
Looks cool, but I'll stick with my DSLR scanning setup.
Annoying as hell. WTF.
As someone who has a mirrorless scanning setup for my film, and pondered getting a dedicated scanner... the price of this is quite steep given how inflexible of a tool it is.
A second hand DSLR setup is going to be roughly the same price or less. I'm also not sure what kind of workflow improvements it actually offers. If you want fancy and experimental, filmomat has arguably a more interesting but pricier offering.
But naysaying aside, I hope they manage to find a niche that allows them to survive as a company, and keep the analog photography revival alive.
I bought some time on a hasselblad medium format scanner (took fucking ages)
The results are good, as you'd expect. However can I tell the difference between that and me putting the negatives on a decent softbox and using a fancy camera to take a picture? yes, but not by much.
I think the main issue is film registration, that is getting the film to be flat and "co-planar" to the lens so the whole frame is sharp.
My negatives are slightly warped, so they really need a frame to make sure they are perfectly flat. But for instagram, they are close enough.
However scanning more than a few pictures is a massive pain in the arse. If I was scanning film regularly, then this is what I'd want, and its cheaper than the competition.
Assuming that its actually any good, I haven't seen any scans yet.
It'd be nice if they were able to adapt the Hasselblad/Imacon "virtual drum" concept and curve the film underneath the sensor for side-to-side flatness. I wonder if that's feasible with a 2D sensor.
Thats a good question. I wonder if the "virtual drum" was there to get over film holding issues (as in it physically bends the film) or that its a line scanner
personally I think that technology has come on enough to move on from the imacon/hasselblad: https://emulsive.org/articles/opinion/scanning-film-the-20k-...
> Thats a good question. I wonder if the "virtual drum" was there to get over film holding issues (as in it physically bends the film) or that its a line scanner
Both.
> personally I think that technology has come on enough to move on from the imacon/hasselblad: https://emulsive.org/articles/opinion/scanning-film-the-20k-...
It's not - the issue that still remains is keeping the film flat, and this is especially problematic with smaller formats. With current solutions you can get the resolution but not the flatness, or you sacrifice something to get the flatness (e.g. ANR glass holders). It's the old glass vs glassless carrier debate, applied to a modern workflow.
I repeat myself: focus, DPI / resolution, dynamic range - these are the solved problems. In fact, modern medium format digital cameras are superior on all these factor. Keeping the film flat, however? Only drum scans and the Imacon "Flextight" solution do this well.
Of course, it depends on what you plan to do with the scans and for 99% of people the solution in the link above is more than good enough.
I've written about this previously https://leejo.github.io/tags/scanning/ # I'm going to add the fourth, and hopefully final, part in a couple of months time.
You could, but it probably makes more sense to do focus stacking in that case
I almost missed the price. Wow you're right that's a lot! And the final retail price is 1599 euros. I have a good Plustek that cost me $300 or $400. Automated transport and unantiquated software sounds nice, but those features are not worth an extra $600-1,100 to me.
Am I missing something or is this supposed to be in another tier of image quality?
It's actually a lower DPI and no IR sensor.
Honestly I feel like anything beyond 5 megapixels per frame is pushing beyond reasonable expectations with 35mm. This is certainly the case with any kind of available light or high speed work in silver-halide process, the area where I figure most people are going to be using this device. Lab-work in C41 and E6 is definitely possible at home but must account for single digit percentage of the home analogue market.
I digitise a lot of 35mm.
A 36Mp camera is not enough to best a 4000ppi scanner. You need about 60-70 mp to resolve the detail of a scan to similar level .
Even a layman can see the difference at 100%
A 4000 DPI scan of 135 gives you 21 megapixels. So 36MP with a good lens will easily resolve just as much detail. There is not 60-70MP of information in a 4000 DPI scan, period.
For most films, anything beyond 4000 DPI is just going to help resolve the grain particles or dye cloud shapes. You have to be shooting slow fine grained BW with the best lenses to need more.
For me real grain looks much better than digital smoothness. That is important for big prints.
"…beyond 5 megapixels per frame is pushing beyond reasonable expectations with 35mm."
Well, as I mentioned elsewhere old fashioned Kodachrome resolves ~100 lines/mm and some newer color emulsion are considerably higher, and of course B&W ones have even higher resolutions.
Given that a 35mm frame is 36x24mm even Kodachrome achieves 8.64 megapixels. OK, let's allow for an overgenerous Kell factor of say 0.8, this figure will drop to ~6.9 megapixels. Given the ready availability of emulsions with higher resolutions, especially the best B&W ones then a figure well in excess of 5 megapixels is relizable in practice.
Of course, that doesn't take into account the image chain as a whole, lenses, displays, compression, etc. which would reduce the effective resolution. That said, these days the typical image chain can easily achieve much higher pixel throughput than 5 megapixels before bandwidth limiting so the effective Kell derating factor could easily be kept quite small.
Never. 20 Mp if you want "lossless".
I think I see what you mean. It’s the difference between having an image showing the shape and texture of each film grain, and an image which looks like what I saw in the camera and which isn’t going to be any sharper. The former has value but the latter was always good enough for me and, surprisingly, rather low in resolution compared to subsequent DSLRs and mirrorless cameras I bought in the 2010s.
Ilford Delta 400 pushed two stops to 1600 ASA in a 1970s Asahi Pentax SP1000 was always going to produce… artistic results, requiring as much imagination as acuity to appreciate the subject. (Read: see past the blur.)
You are right in that a lot of advertised DPI over 5MP is interpolated and not actual sensor DPI.
I wondered what the price was. 1599 seems pretty decent.I was expecting about 4k This is about the price of a venerable Nikon 5000. Some of the setups use film mounts that cost as much as this whole unit.
Are there any sample images
That was over twenty years ago! Why should it cost that now?
Dynamic range is much better (120dB is ~20 stops, your plustek is ~12 stops.
If that matters, I’m not sure.
> A second hand DSLR setup is going to be roughly the same price or less.
And if you get one with Pixel Shift, you can get way higher resolutions than the 22MP they're offering (e.g. my cheapo Olympus gets 40MP JPEG or 64MP RAW from a 16MP sensor.)
You’re for sure exceeding the linear resolving power of 35mm film at 40MP or 64MP.
However, a Bayer-filtered sensor has lower color resolution, since each pixel only sees one color. So the pixel shift really helps quite a bit here since the sensor (and Bayer array) are shifting relative to the film multiple times per exposure.
High-quality film scanners maintain color resolution by using linear sensors without Bayer filtering. But they’re slow and expensive.
All the current Nikon Z bodies (and probably other brands too) have different levels of pixel shift where it’ll take 4 or 8 images and basically cancel out that it’s a bayer sensor. The bayer array is a 4 pixel pattern, so it moves one pixel to the right then one down and then one back to capture all 3 channels for each individual pixel. For things like film scanning it works flawlessly, I use it all the time.
Then it’ll do a 16 or 32 shot stack in order to do the same thing but with more resolution.
It’s been a feature of Olympus (now OM System) high-end cameras for years. I did not realize that Nikon had picked it up as well.
Some modern 35mm emulsions can record ~500 megapixels worth of detail, but good luck getting all that detail in a digital scan.
https://www.adox.de/Photo/films/cms20ii-en/
500 megapixels can have less detail than an old 1 mpix digital from 2001.
[Image resolution is a very complicated topic](https://en.wikipedia.org/wiki/Image_resolution) and megapixel count, or even lines/mm does not tell the full story.
The scan is the least of the problems - good luck getting to that level of detail with mostly vintage lenses, balancing depth of field and diffraction, keeping the film perfectly flat, on a stable enough tripod with no vibration whatsoever; developing perfectly in the dedicated developer. Yes, it's impressive but no, it's not relevant to the average user or hobbyist.
I wonder how this compares to Technical Pan, which I imagine it was modeled after.
The Olympus pixel shift bodies are underappreciated stand cameras. The quality is just bananas.
Filmomat looks fun. Many money. Love the hipster flex with the Weber HG-1 in background of the demonstration video. I do own an Intrepid enlarger (sort of experimental?), and I used to live near Ars Imago in Zurich who sell a "lab in a box", similar to Filmomat's Light system. The independent dev scene is pretty great, though none of it is particularly cheap and is rarely open source, which is disappointing.
> I'm also not sure what kind of workflow improvements it actually offers.
The obvious one is auto-feeding and portability, but without using it who knows. It doesn't offer IR, but even Filmomat's system needs a modified camera. You get that with most flatbed and Plustek-style scanners. I have a V850 Pro which wasn't cheap either, but it'll do a full roll in one go and I can walk away. Even if I shot a roll a day it would be more than fast enough. It has occasional focus issues, and you need to be scrupulous about dusting, but it works well enough. I've never been a huge fan of the setup required for copy-stand scanning and it's tricky getting the negatives perfectly flat in/frame. The good carriers are also not cheap, look at Negative Supply for example.
Frankly it also looks great, like the Filmomat. I think some of the appeal is a chunk of modern looking hardware and also the hope that it's maintained? My Epson works well, but I ended up paying for VueScan because the OEM software is temperamental.
The biggest pains of film scanning are in the post process color balance and dust removal. Unless this can improve those parts of the workflow, it's only going to be a minor improvement. I like the continuous reel movement of this scanner vs a flatbed though, that can improve the physical workflow quite a bit.
Archive link for those who, like me, are unable to scroll on that page
https://web.archive.org/web/20251111210606/https://www.soke....
PS to add more - I am unable to scroll, all I see is the picture with the dark background. If I use arrow keys instead of the touchpad, I can scroll a bit then after a second or so the page snaps back to the top. I have Firefox on MacOS.
(I know the HN rules say that we should focus on the contents rather than criticising the technical aspects of a website, but in this case the contents are not accessible).
Just to add to it: I found out that disabling the plugin Logitech SetPoint allows me to scroll (it's slow and unpleasant but at least I can scroll)
> Yes. We’re collaborating with several film labs in Berlin to benchmark Knokke against Fuji Frontier and Noritsu scanners. > Sample results will be published before the Kickstarter campaign, so you can make a fully informed decision.
I don't care how cool your scanner looks or how "modern" the workflow is - it's samples or nothing. Additionally, if they were really smart, they'd collaborate with a well known film photographer instead of using someone's walk-around point-and-shoot photos.
I'm glad this exists but at ~1k EUR I would be interested if it could scan 120 medium format negatives...but the fact that it does not is an absolute deal breaker for me. It seems like they are considering it. I hope they do figure that out sooner than later.
Medium is the problem. There’s nothing.
Epson stopped making their flatbeds that do film, reportedly because they can’t get the CCDs anymore. That may be a rumor.
The result is they go for 2x MSRP on eBay for models that are many years old. Because that’s all that exists.
Without that, you can buy the kind of scanner meant for a photo lab ($$$$$), DIY it with a DSLR ($$$ if you don’t have one), or pay your a lab a lot per roll and hope they do a good job.
I’m not saying it’s a giant market but it certainly seems to me like there’s enough of one that it could support a small product.
You can get brand new Plustek OpticFilm scanners for 35mm and smaller starting around $300, and there are plenty of other options above that. Plus the DIY.
I’m sure 35mm is easier to make and certainly a bigger market but it’s also a lot more crowded.
I expect their specs are far better than the $300 one I’ve mentioned, I don’t know enough to know. But medium format people are desperate for anything.
> Epson stopped making their flatbeds that do film, reportedly because they can’t get the CCDs anymore. That may be a rumor.
Wow, you weren’t kidding, I completely missed this. I bought one, sold it, then bought and currently own another. I better baby it, there’s really nothing like it out there.
I tried some scanning on a Plustek 8300, which is supposed to be the fastest. The process is still extremely manual/slow and I don't think it's practical on a large scale. Many families who owned cameras in the 60s-70s-80s-90s will have potentially thousands of negatives to scan, but I don't see a solution that will automate that digitalization process.
Software could also use some improvement. Automating batch correction and clean up should be easier, IMO.
This really isn’t my area but it sounds like nothing is fast. DSLR may be fastest without just flat out hiring someone else to do it. But even with thousands of shots that would still take quite a lot of time.
And yeah, workflow is the thing that seems the worst. That seems like a great place to try to improve things to get a sale.
Not that they're cheap, but you can get Imacon scanners for much less than they retailed for. I inherited a Flextight Precision II and it still does a great job.
Do you use it with an old computer, or do you have a good way to interface a SCSI scanner with a modern machine? I tried to get my Precision II up and running but the SCSI card driver would crash Windows at random intervals.
https://web.archive.org/web/20250118205639/http://pathar.tl/...
This is way out of date. I have since been able to get it working on a Windows 11 4th gen Intel machine with 64-bit drivers cobbled together from a couple of versions of FlexColor and some .inf modification. It's not flawless, there's some major corruption that can occur when trying to use certain operations, but overall it works for my needs.
> I expect their specs are far better than the $300 one I’ve mentioned
It's not, it's actually quite a bit worse, especially with color reproduction.
I don't shoot 120, only 35mm. But I thought you could get away with a high end flatbed scanner for 120 negatives?
From what I’ve seen people mostly used the Epson scanners like the V600, V700, V850, etc.
They stopped making them early this year. Only the top end model for $1500 still exists and I don’t know if that’s because they still make it or just that there is still stock left at Amazon/etc.
>Epson stopped making their flatbeds that do film
I don’t understand this… the V850 Pro is still being publicly sold with film and slide trays… what am I missing?
And even eBay is selling them at only about $1,200-1-700 CAD, which is considerably cheaper than the 2,000+ CAD MSRP.
All the cheaper models that used to exist are gone. So the entry price went from like $300 MSRP to something much higher.
Are you taking about other models like the v600 and v700? Because I’m not. I’ve never really seen those models with film scanning tools, it was usually only the v850 that came with film scanning tools. So I have always expected to have to stump up for a v850 if I wanted to do film scanning.
I scanned a lot of 35mm film, of various kinds, using a high-end flatbed scanner (EPSON V700 Photo). The biggest problem is not the optical quality, but mechanics: flatbeds can only scan strips. And if your film is very old and has been stored in a roll, you might not want to cut it, and even if you do, getting the strips to stay flat is nearly impossible.
I tried various fancy holders, but in the end decided that I'll likely have to make my own holder from aluminum or steel sheet metal. And even then you run into the problem of lengthwise curvature. For those that are unaware of the problem with this, these scanners have a very limited depth of field, in the range of 1mm or less. So if your film is bent, some of it will always be out of focus.
I can't see much on this fancy webpage, because they made it so fancy that some of the images do not load and those that do load are oh so mysteriously dark. But if their scanner can scan both heavily curved rolls and strips, I will be buying it.
As to optical quality, if you can get your film to stay flat, this is a solved problem, that Epson mentioned above can produce fantastic results (more pixels that you want, generally).
Optical quality is kind of a problem with a flatbed; even with it perfectly flat, it looks a lot better coming out of a Coolscan.
Seems like an obvious question, but why not sandwich it between panes of glass?
Newton's Rings. Anti-newton glass is frosted, loses resolution.
Professional drum scanners would immerse the film in mineral oil. Epson used to ship a kit for their flatbeds. Popular with 8x10 photographers.
Example Newton's rings:
https://shootitwithfilm.com/what-are-newtons-rings-and-how-t...
Doesn't always work. I've got old Agfa negatives I developed from my grandpa in Korea in the 50s. Developed them after finding them in his attic maybe 10 years ago now. They sat between two panes of glass for 5 years with volumes of books on top, not a single change toward flat. I finally gave up and just put them in the archival sleeves and in the binder with the curve.
I'd never run film at claimed top speeds unless it was a disposable copy print, but seems exciting. Long way have we come from linear array cameras and telecine to these. Different light sources would be useful for dust and scratches mask. $1k price is also insane. For full mechanical assembly with casters etc it probably comes within five or ten, but still a bargain. I wrote here years ago where I had a side hustle scanning and processing ye olde reels with an absolute beast of a scanner (think room sized) that did 2k in real time with an SGI and Hippi fiber network. Tech almost interesting as films themselves. :)
Related, not a 35 mm scanner but a super-8 scanner with true hacker ethos: https://tscann8.torulf.com/
(I have submitted it earlier but no traction)
This is crazy expensive. I built my own copy stand recently from this post's specs and even adding in a digital camera, it's still a lot cheaper than 999 euros. https://alexandermatragos.com/blog/2023/1/15/building-a-copy...
I got my relatively portable scanner for around ~100 euros, plus the software for 100 as well if I’m not mistaken. And it works fine.
It’s weird for me that with the advancing technology people keep coming up with higher price with an excuse of approach, design, whatever.
This seems like an overpriced piece of tech for niche connoisseurs. And I don’t like it.
What scanner are you using, out of curiosity?
Funny thing is, in general, low and midrange desktop scanners that public can generally buy, haven't changed much in 10-20 years since they started using led lights and IR dust removal (Canon Fare, Digital ICE or similar stuff). Some are even the same hardware just slightly rebadged or with a different USB connector. But they're the same price or more expensive.
And, at a different level, professional film scanners are EXPENSIVE. Lots of people are now scanning their film using a digital camera and led backlight (now that there is affordable good quality led lights) instead of a dedicated scanner. But that's not very fast and requires some extra manual work. If this scanner offers reasonable quality and a good workflow (that not very proprietary or closed), 1000-1500 dollars is a reasonable price, especially if you have lots of film coming in, or an old collection to scan.
I could imagine my dad buying one of this to scan his hundreds and hundreds of rolls from the 70s/80s and then selling it once he finish. It would be like 1-3 USD per rolled scanned :)
In the "things that fit on your desk" category, slow is pretty relative - people on ebay still want thousands for Nikon 9000 scanners, and those were not fast if you were scanning high-res 16-bit. And always needed time in post too (cropping/straightening/tweaking exposure and color). An ILC kit can go a lot more quickly.
it's plustek OpticFilm 7200, I bought it second-hand.
This one would be much faster than the plustek though, if they keep their promises. I use a 8200. It takes more than 1 hour to scan a 36 frame color negative with dust removal on, and it requires constant manual input. Pushing the holder to the next frame, unloading and loading the holder, etc.
It's fine, sure. For the price I paid for it and the image quality I'm getting, I have no complaints. On the other hand, a new device that can cut the time down to 5 mins with modern software support (silverfast is kind of dated and VueScan will run you another 100), while priced at 1000 EUR, is not cheap, but also not that unreasonable tbh.
For a studio or a professional - sure. For me, an amateur, who simply does film photography for his own amusement, it makes no sense. I also use standing development process that takes around 40 mins. So the speed is irrelevant for me.
When it comes to the light source, it's a huge missed opportunity. The next wave of film scanning involves capturing Red, Green and Blue separately (separate narrow-band LEDs) to better isolate C-41 layers and counteract crossover on the image sensor. And yes, it should have an IR pass as well, for dust removal.
With that said, I'm happy to see new film products released in 2025/2026. Hopefully this is just the first at-bat.
Only for C41…. For slide film that sets you back.
I think the problem is that it sounds like you get worse results for slide film with RGB than you get with C41 and white light. So the tradeoff is only worth it if you shoot no slide and C41.
This sounds really interesting! Do you have any info about that scanning process?
I should have dropped a link! Check out this project:
https://jackw01.github.io/scanlight/
No IR sensor for dust removal?
I still have an expensive Canon dedicated slide/film scanner from 20 years ago.
IIRC at some point their value started going up as they became rare.
Mine did something like 50MP scans of 35mm film/slides. The quality was more than enough.
But it was painfully slow.
This thing is not 100x faster, so I think it's still painfully slow. If it takes 5 minutes to do a roll of 24 that still means someone with hundreds of rolls needs to have a lot of time on their hands.
Not sure I can actually figure out software to get my old one to work FWIW, but I don't think I care to deal with it, I have a big enough mess dealing with the ~200k digital photos that are already on disk.
This and Vuescan. I once lost the license file, wrote to the author and got a new one.
Seems weird. The fact it talks about rolls of film and reading DX codes suggests that it's able to scan undeveloped film. Maybe that's possible, but I've never heard of that before.
If it only scans developed film, then it's unlikely to still be in the cannister with DX codes, and I've never seen that film delivered to a customer in a roll - it's normally cut up into strips so they can be stored flat.
there is no such thing as a scanner for undeveloped film nor will there ever be.
Developed film has a bar code type of encoding for DX info. The talk about the whole roll is the ability to scan the entire thing in one go vs having to load in multiple strips.
Somewhat related question:
Any suggestions for a scanner meant for bulk scans of old family photos (think a few thousand images)? I bought, what I thought, was a reasonably solid scanner, the Pacific Image Powerfilm scanner but the software is so janky that it hangs every two strips and has to be restarted making the entire process super labor intensive. Also the entire "bulk feature" where it's meant to pull the strips one at a time iis not even close to working.
Granted my film scanner (epson v750) has gotten on in years the more prosumer-professional Epson scanners were good, software was still a bit janky but IIRC there is aftermarket scanner software for them.
It can do 36 exposures but you have to cut them into strips and place them in a carrier but it isn't terrible and if you store your negs in film protectors you are cutting them down anyway.
I am fairly sure the newest version (V850) is the same but be aware they aren't cheap, at least $1k+ USD but still cheaper than the next level up which are pro drum scanners and they are many orders of magnitude more expensive.
I applaud their intent to be repairable, but would really like to see a commitment to open sourcing the software under specific conditions (time, EoL, acquisition, closing down, etc).
> “Is Knokke open, repairable, and long-term supported?”
> “Absolutely. We're committed to building a scanner that lasts decades. All schematics and repair manuals will be publicly available, replacement parts can be purchased directly, and the software will remain supported for as long as possible.”
In the FAQs:
Is the software open source?
Yes. Our control application, Korova, will be fully open source and maintained long term. It’s a native, lightweight application for Windows, macOS, and Linux.
No video demo, no scan restults, focusing too much on design and buzz words. Thanks I'll stick to my $400 epson scanner
Is there anyone in the world who hasn't cut their negatives into strips? https://www.beauphoto.com/product/printfile-archival-negativ... The ability to scan an entire film seems rather useless to me in light of that.
I for example develop my (black and white) films myself and leave them to hang dry uncut, then proceed to store them in a box. When I have the time, I "scan" them using a MILC on a tripod, a 3d printed holder and a LED backlight panel. Afterwards I cut the strips up and file them properly for long term archival. Many of us do it this way.
Disappointed to see that the first reactions on HN are so dismissive.
I'm both amazed and really pleased to see anyone attempting to launch a totally new scanner in 2025, and genuinely hoping the actual scans are really made at the resolution and color-depth claimed in the text: too many recent scanners are simply upscaled, lower bit-depth devices marketed with exaggerated specs.
I also have a Nikon Coolscan 9000, so I'm not immediately in the market for this. But I don't expect the Coolscan to last forever, and the Firewire connections on the machine are already abandoned by Apple, who chose not to support the cables in their latest Operating System - so eventually I won't be able to connect it to a new computer.
The open source aspect seems worthy of attention if nothing else. Even if this is a middle-of-the-road scanner - the community being able to customize, improve and support it would be incredible. Especially considering your scanner is considered one of the best despite being over 20 years old.
I used a semi pro Plustek model to scan hundreds of negatives and was able to sell it off for same price as I paid.. that was nearly 8 years back.
Guess this can't improve on that lol. But by the look of it, negatives that's already cut into small strips of 4-6 frame each wont be easy to load?
I think software is the key. While the bundled one was ok to do basic stuff, figuring out stuff was complex. In the end I just used default.
For the "this is crazy expensive" crowd: this competes with the Pakon F135 - an ancient lab scanner that involves booting up Windows XP, finding old drivers on a CD, etc
This sounds excellent to me, personally.
No, this competes with a lightsource, some stand and a macro lens+DSLR/M. It's a good price if don't have the later ones. But chances are high that you do if you are into photography.
How long does a roll take you to scan that way?
5-10 minutes depending on equipment/skill. You "lose" most of the time if your strip is cut into stripes of 4-6 images, otherwise you can do it in 2 minutes.
Why are you getting downvoted? You’re absolutely right. My mind instantly went to “I would have bought this instead of the F135+ in a heartbeat”
Folks who've never scanned film in bulk, I guess
Seems like too much time was spent on the concept and website and not that much on the product. I can't help myself but always think those projects are too marketed and engineered to be honest. Film scan should be simple and straight to the point, we don't need this Apple aesthetics and ultra marketing.
"Designed by photographers, for photographers." Nice. Would love to see your pictures then.
I whole heartly agree, especially
> Would love to see your pictures then.
Without any samples it’s hard for a $999 kickstarter project, considering a Epson V750 scanner costs much less than that but already provides great quality and supports more formats
I’ve been using an inexpensive Epson V600, I think I paid under $300 and it came with a license for the scanning software which was another $50-$100 value. I also got $25 anti newton ring inserts. The process is SLOW but the quality I think is excellent unless you need to print XL and show your work at MOMA.
I thought you use a camera on stand is the common way now. Did one in Edinburgh and ordered one back home waiting for delivery. After decade of epson it is not comparable for speed and option …
for just contact sheet fast … you can just move and push the blue tooth button … for one particular treasure slow … you can do pixel shift and focus stacking …
Scanning 135 format at home is pretty much a solved problem right? The home made solution to this costs $0 if you own any DSLR and some other basic photography gear.
I think the product would be more compelling and worth it or even a good deal at the price they are offering if it offered drum scan-quality for larger formats.
The workflow for this scanner would allow you to thread an uncut roll of 35mm film through it. You'd have to spend more than $0 to get that kind of speed on a DSLR rig.
1. I had never even heard of an uncut developed roll of film before, so I guess it's useful for that.
2. Time is money, but who is honestly shooting that much 135 film that it's worth 1600 Euros to buy a faster scanner for it? I don't think a museum wants to feed degraded film through a fast scanner, and surely pros who still shoot film would use a larger format, since that's where it has some differences / advantages compared to digital?
> I had never even heard of an uncut developed roll of film before
That's how film is developed. Someone at a lab has to cut it.
> who is honestly shooting that much 135 film
How about a film lab? A place where "uncut developed film" is extremely common.
>it's worth 1600 Euros to buy a faster scanner for it
Price is 999 euro.
> pros who still shoot film would use a larger format
Some do, some don't. It depends on the project. I'm a little surprised by your comment looking at your history. You say you're a retired professional photographer and you've never heard of "uncut developed film" before? If you're retired in 2025, you must have been working when all photography was on film. You never developed a roll of film before?
Some influencers that make money directly from their photos could find it beneficial. Although as the saying goes, the fastest way to make money with photography is to sell your gear!
At this point I'm mystified at any film scanner that doesn't also offer an advance mechanism for motion-picture film.
Otherwise you can use a slide-copier attachment on a DSLR, taking multiple exposures if necessary to achieve whatever dynamic range you want.
I don't get it.
This is cool. I spent ~$1k on a Pakon F135+ a number of years ago, and the workflow was indeed extremely frustrating and the results not that great. If I were still shooting a lot of film I’d definitely pick this up. But we need to see sample images!
tlxclient they say
The industrial design is really nice! I like the fun LED display as opposed to an LCD. I am surprised there doesn't seem to be any physical controls!
I'm sure this will be on every photography youtuber's channel shortly, can't wait to see it in action.
It seems to me like they spent most of their budget on things that don't really matter.
I have 3 Nikon Coolscans and I also repair them…
I would love a new scanner for 21st century but there just no way anyone serious is trading CCD (or PMT if you got the cash) for CMOS.
But I applaud the initiative and will definitely buy it to try but not to keep.
RGB LED backlight is a terrible choice. Wide gamut but terrible color rendering.
Can I ask why it will produce terrible color rendering? In addition to commercial scanners that used narrowband trichromatic (RGB) light sources, hobbyists are creating their own RGB light sources to digitize color negative film claiming superior results and putting forward arguments why this is better:
https://jackw01.github.io/scanlight/
(NB: Most film I shoot is slide film, which I’ve been told doesn’t benefit from RGB light sources because it’s intended viewing was projected with a broad-spectrum white light [likely a warmer than daylight (but color temperature isn’t much of a concern for digitizing slides)] so I haven’t dug into this much.)
RGB should be fine, especially if you use a genuine ultra-high CRI source. A few companies make them, I think Waveform is one of the more well-known. If you really want to spend money, the optics suppliers like ThorLabs sell broadband LED sources. In an ideal world you could calibrate the image sensor against a known spectrum so you'd know its response. If you can estimate colour to a reasonable degree then you can transform it to what it "should" look like. Nevermind that pixels are single-wavelength as well.
https://store.waveformlighting.com/collections/led-strips/pr...
Negative Supply use something similar in their light tables, though I don't know exactly what the source or spectrum is. They're highly regarded enough that I think it's not an issue.
You can also use LEDs for enlarging, but you need to be careful about buying the right bands for the paper. I've used Luxeon SunPlus with some success as you can buy the correct green/blue for the different contrast layers. Though for B&W, even a random 5500K module from Cree worked quite well.
I got one of these from the latest batch last week and I’m not entirely convinced by it yet. I need to experiment some more but I went back and did a couple rolls from this summer and so far I think the cslite warm setting + negative lab pro results are better and more consistent. I’m still getting some wonky colour casts with it. It’s nice that the control app lets you change the power of each LED colour separately, so that’s the next thing I’m going to experiment with.
I’ll also note that negative lab pro hates negatives that are scanned with it. They don’t turn out at all. If you’re using it, you should expect to be inverting them manually, which is kind of a pain. I was quietly hoping (but not expecting) to still see some of the benefits of it when passing them through NLP.
In case anyone is wondering, the current meta in home 35mm film scanning is divided:
Option 1 is to get an Epson Perfection series flatbed scanner. V800 or V850. This approach is highly automated and you get automatic dust correction with color film. But, leading software packages don't support Linux, and the quality for 35mm negatives is just okay. Performance on medium and large format is SOTA.
Option 2 is to assemble a scanning rig with a DSLR and a light table. This approach is fiddly and requires a lot of space, but with some tuning, the 35mm scan quality can beat flatbeds.
There are some other, more obscure approaches, like vintage Minolta and Nikon scanners, but unless you have a PC with a parallel port laying around, you're gonna have a hell of a time getting those working.
None of these options are good, and if this thing can really perform, I'd happily drop $500+ on it.
Side note: Those little toy scanners like the Kodak branded ones on Amazon are atrocious. Avoid them. If you need to scan some family photos and you don't want to break the bank, go to your local photography store. They could really use your business.
DSLR is the most flexible way, cost wise. You’ll likely have a DSLR already if you’re a film photographer, and You can get a decent scanning setup for under $200. Being able to edit RAW files is also more flexible. It’s a lot more fiddly though so not ideal for high volume work.
Don't count out the flatbeds. I don't have the space (or the patience) for a DSLR rig, and I've been quite happy with the output from my Epson V800. I bought mine used for $700 or thereabouts. For the amount of time it's saved me, the added cost over a DSLR and bellows/macro lens was worthwhile.
What about option 3: drum scanner?
I said "home" 35mm film scanning! Sure, you could go find yourself a used Heidelberg Tango, but it's the size of a fridge, and heavier than a fridge.
Granted, I've read it as "home film" , but have also read stories of people using their garages for photoproductive use getting themselves used ones!
Hmm, does seem pretty expensive but sounds interesting. I've got an old Canon FS4000 for 35mm, which works ok for me. I'm curious what people recommend for 4x5 film.
Is there such a thing as a cheap drum scanner.
I don't think you want a drum scanner, look up a video of the process for wet mounting negatives... It's super time consuming, cumbersome and messy. Something like an Imacon looks like the best middle-ground, shame they don't make them anymore...
I've been camera scanning 4x5 and I'm happy with the results. Take two offset photos and stitch them in post. Mind you, I scan with pixel shift for higher res.
I haven't done 4x5 in a while, but I have an HP flatbed with a 4x5 adapter (purchased used on ebay) which does an OK job for the price.
The Epson scanners are supposed to be nice but they stopped making them and they’ve shot way up in price second hand.
Yeah I do remember hearing good things about the Epson scanners, will have another look, thanks.
I’m not sure if they can do 4 x 5. I know they can do 120.
I got excited and thought this would support 35mm stills (e.g. Kodachrome).
Off-topic: does anyone know the best tool for scanning old stills in 2025?
Maybe a mirrorless, a Macro lens, and the Valoi adapter. This is what I use now after having painfully used a Flatbed Epson for years. This set up is fast and sharp. Best of both worlds.
I wish they'd make something for auto-feeding slides. Slide projectors all have auto-feed, but none of the scanners do.
Granted, I got sick of doing this myself so I ended up happily forking over the money for the local lab to do it on their Macon
I’ve got a decade’s worth of 4x6 pictures and negatives from my childhood. Would something like this help me digitize them?
Presumably the negatives, but not the pictures, as this is intended to scan only negatives. You'd probably have better luck with a cheap flatbed scanner that comes with adapters for slides and negatives. (I've used both dedicated film scanners and flatbed scanners to scan thousands of rolls of film and prints taken over a span of 50 years.)
No. 4x6 negatives are large format and requires a different type of scanner (flat bed). If you look at the photos the 35mm roll slides into the device so you can’t fit a larger format sheet into the feed.
You might be confusing 4x6 inch prints with 4x5 inch film.
35mm typically uses 2:3 ratio images, often printed as 4x6 inch pictures.
Well, this 35mm scanner will be helpful for neither 4x6cm medium format film, nor 4x6in pictures :)
Doh! You’re right.
Korova, the name of their software, means "cow" in Russian.
Film doesn't really make sense anymore outside the realm of luxury-budget art. And if so, why 35mm? Why not medium format / Hasselblads, or large format? Why not actually go for the format sizes where there's a reasonable argument that film can produce a better end result compared to DSLRs?
35mm wins by a mile on price per shot. And for web scans (where most photos end up), it's got more than enough resolution.
35mm film and 120 film are a similar cost per roll, but with 35mm you get 36 exposures vs 8-16 exposures on 120 film (6x12 and 6x4, with square 6x6 in the middle at 12 exposures). And if you shoot half-frame, the cost/shot really goes in the 35mm direction.
That said, I have a handful of of 35mm cameras (all fixed lens vintage rangefinder) and a post-war Zeiss Super Ikonta IV (6x6 120 format). The Olympus 35DC is my favorite of the bunch - it's automatic except focus - really sharp and fast lens - just a pleasure to use. And a Polaroid Go 2 because it's just dumb fun (way overpriced for the quality, and sensible people buy Instax cameras instead, but the Polaroid form-factor was just too much for me to pass up).
I shoot film because it makes me slow down and think a bit. With my mirrorless cameras, I'm too prone to spray and pray and sorting through hundreds of shots can kill the fun for me. That, and the film look is nostalgic for me - sometimes I just want rough snapshots - feels more like a memory vs the crystal clear high res digital output.
I don't feel like you really countered my argument. If you're concerned about cost/shot, digital beats 35mm, no question.
> I shoot film because it makes me slow down and think a bit.
100%, because of the higher cost per shot compared to digital. The higher still cost per shot of medium/large format enhances that effect.
Cost per shot and availability (far more 35mm stuff out there than larger format). That explains 35mm.
The look and the process explains film over digital.
> Why not medium format / Hasselblads, or large format?
That has also been superseded in the digital realm. I can use a top DSLR from the last decade to blow film medium format cameras away under most conditions, especially in low light. If I just use a digital medium format camera, it invalidates almost any rationale for film medium format (there is a minor, minor argument to be made about depth of field when the lens is wide open).
Large format can squeeze slightly more resolution out of an image, but the ability to actually use that extra resolution is rarely satisfied. Again, a MF digital subs in for almost every conceivable use case.
The reality is that people do this because it is fun, or a challenge, or for a feeling. Basically, its interesting. The technical excellence of the 'end result' is secondary.
There's tons of things that people do that don't make sense from a purely pragmatic point of view, and that's what makes the world so much more fun to live in.
> top DSLR / digital medium format
But these cost several thousand dollars, due to them still being relevant for contemporary professional applications. Medium format bodies can be found for hundreds of dollars on eBay. Even at $1/shot, if as a hobbyist even if you shot 4 rolls (48 shots total) in a weekend, you'd be shooting that much every weekend, every weekend straight for years before a top digital body would break even for you. And you would have more fun and a more meaningful time due to the more deliberative process.
35mm is super cheap to get into if you just want to experiment. Old 35mm cameras and even home darkroom equipment are gathering dust in tens of thousands of households, cheap on FBM or thrift shops.
It's expensive compared to digital for snapshots, but if you enjoy working in the darkroom as a hobby, you can probably get everything you need for free or cheap.
Would've been an auto-buy at $500. Looks cool though I hope they have success
I don't see any infrared sensor. This makes it drastically worse than the competitors on the market, as it's an essential sensor for dust and scratch removal.
Did they not research the competition?
I can buy a brand-new 7200 (virtual) DPI machine with infrared, proper color metering, wide software support, and multi-exposure system for $400, less than half the price of this offering.
It also supports slides and single frames, whereas this has a min. of 3 in a strip.
Looks interesting but €999? Is that a placeholder value?
Yes. From the article: "It's [sic] final retail price is set at 1599€" I was actually excited from the title but upon reading further I got really disappointed. It's basically wishful thinking right now, not sure if it's even going to be released. Not sure why this is posted here.
Pretty bit of hardware. Need to show samples though.
This site seems kind of pointless. There are no examples of the product's output or any demonstration of how it compares with other film scanners.
Why does film attract so much dust?
No expert but I'd guess "thin shiny material being rubbed against things -> STATIC AHOY -> clingy dust".
yeah static mostly. You can get a pizeoelectric antistatic gun to neutralize it. You can also get antistatic brushes that use alpha emissions from polonium 210 to do this.
every optical system attracts dust - dslrs / mirrorless cameras with removable lenses have extensive hardware and software systems to handle dust detection and removal from the sensor.
Why not use a blower fan?
Because dust sticks strong enough to withstand the blow
I gave up trying to read the article when animations started happening when I scrolled. It's annoying and I have better things to do than waste time waiting for your animations to stutter and finish moving around when I'm trying to scan the article. Good luck with whatever you're trying to do.
JFC, for that price I can get a NiB Epson Perfection v850 Pro, and have something that can scan film almost as effectively and is a full-blown generalist scanner, to boot. The best consumer scanner ever released, IIRC.
I doubt whether this product will usher in a new era of film scanning for many photographers. For starters, as it's only aimed at 35mm negatives it's unsuitable for my needs.
As with many photographers, my collection consists of B&W and colour prints of various sizes and formats, 35mm B&W and colour negatives in both rolls and cut strips, 35mm slide/reversal material both in rolls and as mounted slides. Film stock covers many brands including Agfa, Ansco, Fuji, Kodak—including its Eastman movie emulsions—and others. Kodak holds special place, with Kodacolor, Ektachrome (including infrared versions) and Kodachrome. At a guess, I've have about 30,000 Kodachrome slides alone. And that's not all, I've also larger format photos, prints, B&W and colour negatives and reversal stock.
Most of this material has still not been scanned because of the challenges involved, for instance those in the know will be aware of the difficulty of scanning Kodachrome slides because of residual silver that's still in the processed emulsion. Then there are scanning difficulties, mounting various formats (slides, rolls of negatives, etc.) and technical difficulties such as focus adjustment, avoiding Newton's rings, etc. Simply, I've not been able to get the tech necessary to do what I consider an adequate job.
Restricting my comments to just 35mm I can confidently say there is NO 35mm film scanner on the market today that can do full justice to a large range of film types—except perhaps exotic and expensive drum scanners which are unavailable to the vast majority of photographers including many professionals. (Drum scanners are only found in high-end professional and technical environments, they cost upwards of tens of thousands of dollars and are a damn pain to use.)
Fact is there is NO film scanner on the market today that can faithfully reproduce in digital form the full dynamic range and resolution† of old fashioned chemical film emulsions. I say 'old fashioned' because modern digital photography, HDR etc., is capable of much wider dynamic range, resolution and colour gamut than film emulsions, so it's not a technology limitation (converting the limited dynamics of old film ought to be easy but no manufacturer makes equipment that does). It's really shameful that no manufacturer has stepped in to fill this technical gap when clearly the technology is available to do so.
Below, I've kept to the basics, an in-depth comment would be much more detailed:
• Argument goes that no one would pay for a film scanner with those specifications—as its manufacture would require precision/exotic tech, and anyway it's doubtful anyone would notice the difference with currently available scanners. I question both those assumptions as I'll explain.
• Leaving drum scanners and a few very expensive ones aside, in the past the best 35mm scanners on the market were the Nikon COOLSCAN range but Nikon discontinued them some years back and nothing has equalled or replaced them since. They were not perfect but they had the best optics and overall provided the best resolution and dynamic range available of any scanner. The COOLSCAN's most significant limitation was its incredibly slow scanning speed (nothing much has changed here with the possible exception of this soke engineering device, film scanners have always had snail-like speeds for seemingly inexplicable reasons).
• Nowadays, for most photographers the best compromise between quality and usability are Plustek scanners, whilst they have neither as good a resolution nor the dynamic range of the Nikon COOLSCANs they are about the best available. I'd add neither are Plustek's mechanics for scanning films as good as the COOLSCANs (that said, in this regard the Nikons weren't much better than just adequate).
• So is there really a noticeable difference between a Plustek and a COOLSCAN? Yes there is, COOLSCANs have noticeably greater dynamic range in dark shadowy areas, and despite the Plustek having comparable resolution specifications with the Nikons the COOLSCANs produced visually sharper scans.
• Why are all film scanners so pathetically slow? A good question I cannot fully answer. Perhaps 20 or so years ago there may have been some excuse but even back then I'd argue they should have been much faster. For argument's sake even if the electronics had slewing limitations and had difficulty in processing images—which wasn't the case—then scanners could have been made much faster by simply increasing the number of rows of sensors—for example, increasing the rows from one to 10 and stepping 10 pixels at a time would increase scanning speed by 10. This is so obvious that it's mindboggling that it hasn't been incorporated into scanners previously. (Note, the other obvious option of photographing an image as does a camera has serious quality limitations.)
There's much more to say about speeding up scanners which I cannot cover here except to say have you noticed that scanners still use USB-2 and not USB3-3? Why?
• There are other significant issues that haven't been addressed adequately in many scanners such as colour calibration. For instance, every type of colour negative has a unique set of parameters often referred to as 'film terms'. In short, these parameters define how the destructive colour mask should be decoded (that's the orangy mask that's incorporated in all colour negatives). Many scanners only approximate or guess these parameters and expensive third party proprietary software such as SilverFast is needed to correct these limitations.
If I didn't know better I'd reckon the lack of a competitive range of high performance film scanners on the market was some form of conspiracy—electronics designers having an intrinsic distain for old fashioned analog film technology or such but clearly there's more to it than that. Whilst I can surmise reasons I'd only be guessing but for sure it has little to do with technical limitations.
Why the scanner crisis hasn't been a much hotter topic amongst serious photographers and professional reviewers has perplexed me. Perhaps if nothing else this scanner from soke engineering might fan the debate, it could perhaps force scanner manufacturers such as Plustek to upgrade their long-stagnant designs.
_
† Kodachrome has a resolution of 100 lines per mm which roughly equates to an image with 3600x2400 pixels (a frame being 36x24mm). Some films have even higher resolutions. Nyquist math says that the sampling rate should be doulde which means a scanner should be able to resolve to 7200 lines per frame but in practice no commonly available scanner comes anywhere near this figure. Diehards note, I'm aware this isn't a precise calculation but it'll do for argument's purpose.
[dead]
[dead]
Stick to my Nikon scanner until this comes down in $
It's quite punchy in price for a niche market. I'm struggling to see much of a USP for me.
The interesting thing is w gen Ai images, a reliance in terms of photographic evidence may come down to analog. Plus Kodak returning to film stock, the future looks bright for 35mm
Quite expensive though...