Designing a new film scanner; need your help

[Ranchu]

Ranchu: Basing the profiles on LUT (i.e. tables instead of formulas) is a good idea that I'll experiment with.

I think a good way would be to build a LUT profile of the scanner with something known, one of those calibrated slides or whatever, and use that to convert everything to the output space, and then have a curves and then to the output space again. Make the profile for the negatives with the came calibration slide, but through a piece of blank developed portra 400. The second color space in the chain a LUT color space. I would lean toward inverting negatives while it's linear but I don't know... Adobe98 is a terrible color space, you might try colormatch...but what do I know.

edit: you could probably just do the first conversion, and leave out the curves etc. as long as it kicked it out in 16 bit and the negatives were inverted, an accurate (LUT) conversion is all you would need. There are other programs for curves, and you wouldn't have to have any user interface with that, just a go button and sizes. It'd be cool, though, to be able to adjust the light intensity.

 

[x-ray]

Prophoto is probably the desired color space to work in. Lightroom works in Prophoto only which is to my knowledge has the widest color gammut. Beyond color space it should scan in a raw format for example like DNG and give Tif 16 bit and jpg as an option. I often scan to 3F format with my Imacon. For editing it gives the most flexibility vs Tif or Jpg.

 

[DNG]

Don't make it too limited in what it can do.. that will decrease sales

Don't make it too limited in what it can do.. that will decrease sales

I own a V700, I got it after my Dad left me over 10k in slides, negatives from 35mm to 4x5.

Before that I had a Plustek 7200 for my personal 35mm B&W, 6 at a time, and very time consuming between each negative. About 4 minutes per negative from prescan to Adj to final scan. That is 24 minutes for 6 scanned negs..
The V700 does 24 negatives in the same time!!

I got the V700 over other flatbeds because it automatically framed strip film it the OEM film holders, This allowed me to prescan 24 negs in about 2 minutes and then about 10m minutes to make any small adjustments in Epson Scan, then scan them (all 24 as tiffs) in about 10 minutes at 3500dpi (Computer CPU and Memory dependent)

I have heard a lot "make it for me" features.... like B&W only, or up to 4x5 only, and the like... not a good business model.
You want to sell this to a "W I D E" audience right away so your 1st Q is in the Black.

I use a scanner for scanning my stuff, and my Dad's stuff from 1940s on. You just don't know if one day you will have a need for more features.

Some only need it for their stuff... great, so having a scanner that can meet "everyone's" needs will have more buyers on a world market.

• 35mm to 8x10 Film (neg or pos) (though document would also be nice for scanning old prints from family inheritance etc).
• Auto Framing Holders with your software (similar to Epson)
• RGB Adj,Levels Adj, Curves Adj, WB dropper, Exposure Adj, Contrast Adj etc
• Color Recovery for old positives that have tinted over time
• 16bit grayscale option
• Sharpening, Dust Control, etc
• TRUE 3500dpi minimum
• Adjustable Focus Plane with verification (Glass top to 5mm above)

I am sure they're more that may be needed (wanted).......

A scanner has to be able to get any media (if faded or tinted over time, or just A BAD exposure) to a place to make a decent scan as a base for editing later.

I chose $800, I feel that though many RFF's may have better resources than most here, and chose over $1500 and suggested even higher (that is the minority of buyers) it should be priced to a more financially diverse market to sell more units. But, $1000 would be a good price point also.

If it is too expensive, you will price out a lot of folks who can afford $1000, but not much more. I assume you want a long haul company, so, sales is a big deal in the beginning, but so is quality and longevity. If, this is successful, you can always offer a more specialized unit for the those that can afford it. And offer less featured units for a limited film selection for those that can't afford $1000 (maybe 35mm to 4x5)

 

[bence8810]

Very exiting project, wish you the best of luck.

As for me - perhaps a minority here - I only care for the speed and ease of scanning therefore I use a Pakon 135+ which is perfect. Single roll uncut fed into the machine and 36 frames are scanned in 3 minutes. The only problem is that it isn't being made anymore and if fails you are in a lot of pain as no one can fix it.
If anything new came around that would be similar, I'd be in for one!

Reason I am not too interested about resolution etc is that I only scan to archive and when I like something, I wet-print it and scan the print itself.

Ben

 

[nongfuspring]

It sounds like this is quite a specialist piece of equipment. I imagine that there would be uses for such a scanner outside of film scanning also (digitising of art and, small museum artefacts etc.) so I think it a worthwhile project, but unless the price is very low I would be skeptical as to how worthwhile such a scanner would be to the general (and let's remember - very small) film shooting community.

For me personally - and I know this is the case for a number of people here; I scan using a macro reproduction set up. The one I have is one I cobbled together myself for under $100 including lens and gives me up to 32MP files from 35mm film which in general far exceeds the actual resolution of the emulsion, so I don't have any great desire for anything that produces higher resolution results. Personally I would love to buy a macro reproduction setup that is better made than my current one, but I have no interest in scanners. The only instance where I could see myself using this is if I were making enormous digital prints with extreme grain clarity from 35mm film, which I wouldn't rule out but don't have any particular interest in doing at the moment.

To echo what some others have said, make sure you use glass rather than acrylic, and be sure that it's anti-newton ring glass otherwise things will get very frustrating!

 

[Ranchu]

There's no reason to make a non-specialist piece of equiptment, there's already a v500. Or V600. If it's cost effective high quality people will buy it.

 

[Merlijn53]

What surprises me in this discussion, is that quite a lot of people seem to think you can have a good quality filmscanner for around $1000. Let's be realistic! For $1000 you can have an Epson. A Plustek 120 costs $2000. Imacon starts at $15000 and people are happy to pay $3000 for a used, 10 year old Nikon.
So what can you expect for $1000?
Frank

 

[nongfuspring]

What surprises me in this discussion, is that quite a lot of people seem to think you can have a good quality filmscanner for around $1000. Let's be realistic! For $1000 you can have an Epson. A Plustek 120 costs $2000. Imacon starts at $15000 and people are happy to pay $3000 for a used, 10 year old Nikon.
So what can you expect for $1000?
Frank

For $1000 you can buy a decent digital camera with a macro lens that will blow away any flatbed, with the added benefit of getting a digital camera and a macro lens. There are a few online comparisons of drum scans compared to stitched macro scans where the macro scans beat the drum scanner by a noticeable margin. Taking multiple overlapping shots can be tedious, but the stitching is easily automated.

IMO the only situation where a flatbed makes sense anymore is if you're scanning medium or large format film and don't want to put the effort into stitching. For the most part if puzzles me why anyone uses flatbeds for film scanning at all.

 

[Ranchu]

How do you do C41? How's the color? If a dslr is not as good and easy at C41 color as a v500 I lack interest.

"Is it too much fuss to capture color negatives? I don't think so at all. Post-processing of digital images can have just as many steps. (Either way, it's work, which is a good reason not to give your photography away for free.)

At the same time, "macro scanning" of negatives is not so much work that it would be unrealistic to do it in the first place. And, as I said, you may find it becomes a new opportunity for creativity. "

http://120studio.com/mcap-c41.htm

 

[nongfuspring]

How do you do C41? How's the color?

I've only done tests since I normally shoot BW. I've found it hard to get the colours correct as is usually the case, but no more difficult than with a scanner.

 

[k__43]

Question is: is this going to be a performance or convenience optimized scanner?
I paid 1000€ for a Nikon 8000 and I do not plan to move to LF in the moment. I'm happy with the output, but I guess when this machine dies on me I'd be in the market of something like the Plustek 120 which is a ~1700€ machine. I think the Nikon is great but scanning MF could be easier with the plustek.
So if your scanner is in the performance and convenience zone of these both I'd argue that a similar price is fair. If your scanner has a higher optical performance but no automated scanning I'm not your customer.

 

[Roger Hicks]

Sorry, I'm not a native speaker. If there is a distinction between slides and transparancies, then I mean the ones not mounted in those plastic holders for projection. But it should scan them up to 4x5 and 6x17. I never found those mountings for anything larger than 6x6 anyway, neither do I have a projector for any larger ones.

As said, there is use for scanning the mounted ones (I got thousends of them) but I agree that this might be an option.

There isn't. I'm a native English speaker and I've written for the photo press for decades. Calling mounted transparencies "slides" and unmounted slides "transparencies" might conceivably be a useful distinction but it doesn't exist.

Cheers,

R.

 

[__jc]

I know probably 75-90% of photographers here at RFF that use film at all, use Digital for color and film only for B+W.

I'm surprised by this statistic.

I tried looking for a poll for this and found this one:

https://www.rangefinderforum.com/forums/showthread.php?t=78502

Or was there another poll?

I also question that statistic. I'd like to know where you got those numbers from, f16sunshine. I would have thought most film users use film most of the time, black and white or colour.

Also, agree with Roger above that slides and transparencies, at least to my knowledge, have always meant the same thing.

 

[WJJ3]

Alright, fine, here is one. It's a tiny crop of a distant building from a Mamiya 6 50mm lens. I haven't got access to the original filmstrip at the moment, so I can't tell you what kind of emulsion it was on, or how large the crop is relative to the full frame - but the longer dimension is probably on the order of 5mm.

This image confirms that the grain is easily resolved. Imagine if you swap out the 4x microscope objective used here to something like 10x or, gasp, something closer to 40x.

Please pardon the damaged film surface, the lack of color, the use of diffused backlighting, the clipping/contrast and the fatiguing grain and tones. It was always meant to be an image for internal use to confirm focus.

But at least this proves that this scanner is not vaporware.

By the way, have a read on Kohler illumination, commonly used in microscopes. That's how the new design is illuminated, and should result in better natural contrast.

I was just scanning through this thread and have a couple comments to add.

quejai, keep up the good work. Those samples you posted are VERY interesting. My biggest gripe with all the scanners ive used is introduction of noise. The film structure revealed in your samples is simply next level. So for my list of requests in order of priority: lowest noise possible please.

I definitely support the idea of a monochome dedicated scanner. I have messed around with scanning color film quite a bit and trying to get the results I want quickly leads to less fun and more work and frustration. You could also interperet this comment as: wonderful color results out of the box or please dont bother.

Acceptibly convenient film handling that will keep up to 6x9 frames flat without the need for wet mounting. Option to wetmount would be lovely.

Film holders and scan software capable of scanning the whole frame including a margin of its unexposed border. This is not a hipster fad request. I have been going to great lengths to include frame borders in the darkroom and in my scans when I want them for some time and I value this capability.

As far as price, you should charge what is reasonable for you. If it turns out as good as it sounds people who need it will find a way to afford it (within reason：〕)

Also, one request if I may: If you are inclined to post more high-mag samples, would it be possible to include film type info, and perhaps a comparison? Say Ilford FP4 and Delta 100?

Thank you




Tapatalk を使用して私の ME173X から送信

 

[anerjee]

Isn't it better to design a high precision x-y movement rig which can mount a camera and macro lens and take multiple pictures to stitch them together?

Plus software that would process the resulting images and create "scans" out of them.

Please excuse me if this sounds naive, I'm just curious, and not technologically sophisticated.

I use a Pakon scanner today, and the software's ability to nail white balance in any lighting situation more than makes up for the lower resolution scans.

Hey everyone

I'm a keen film photographer and also a mechatronic engineering student. I'm working on a project to design, build and sell film scanners. I'm planning to host a kickstarter in a couple of months to make the first batch, but before I do that I'd like some of your advice.

Could you please make a list, ranked in order of importance down, of features you value in a film scanner? Feel free to include features that haven't been mentioned yet.

There's also a poll, where you can vote on how much a good scanner is worth to you.




More details:
I've spent an obscene ammount of effort on this. I started just over two years ago, have gone through several kilograms of paper sketches, and spent hundreds of dollars on parts for an underwhelming prototype.

Fortunately for you, that prototype inspired several more months of drastic redesigns; and my current design is more affordable, has sharper optics, is lighter, has unprecedented color fidelity and is pretty darn fast, compared to other scanners out there at the moment. I'm pretty confident that this will be the biggest development in film scanners for over a decade.

Before I commit to building a second prototype based on the evolved design, I'd like to hear from you (my target market! ) about what you value in film scanners.

Just to clarify: For each time the scanner is loaded, it will be able to scan anything from several strips of six exposure 135 rollfilm up to (hopefully) 8x10 sheetfilm.

So thanks for the help.

 

[Dwig]

There isn't. I'm a native English speaker and I've written for the photo press for decades. Calling mounted transparencies "slides" and unmounted slides "transparencies" might conceivably be a useful distinction but it doesn't exist.

Cheers,

R.

The distinction was one that was very consistently enforced in all of Kodak's publications and product descriptions. They even rejected "slide film" in favor of the more proper "film for slides" for their 35mm reversal films as the film itself didn't produce "slide", only "transparencies", though it would typically be mounted as slides.

 

[Spanik]

Isn't it better to design a high precision x-y movement rig which can mount a camera and macro lens and take multiple pictures to stitch them together?

Plus software that would process the resulting images and create "scans" out of them.

I think that is an interesting idea. And it shouldn't be that high precision as the software can correct for the last bit. Now with a fovean sensor it would also mean that colour information is gathered at the same place at the same time. And with controlled lighting the WB problems of fovean don't matter much.

 

[f16sunshine]

I also question that statistic. I'd like to know where you got those numbers from, f16sunshine. I would have thought most film users use film most of the time, black and white or colour. Also, agree with Roger above that slides and transparencies, at least to my knowledge, have always meant the same thing.

I pulled the figures out of my time on threads here.
It's not scientific. This is not a think tank or board meeting simply a conversation.
In many threads over the years rff members have stated they use film for b&w and digital for color.
The which film for travel thread pulled up as an example is biased by a condition.... One film for travel... Of course color is the most versatile choice.

Anyway, I stand by my request. A scanner and software to run it that is optimized for traditional b&w film.

 

[DNG]

But f16, his market is MUCH bigger than this forum membership.

someone else mentioned if this scanner will be for "Professionals" IE: Commercial grade high quality, high durability, or for "Advanced Consumers" who don't want to pay for heavier materials (Medal in lieu of plastic for example),
but want a long lasting, easily repairable unit that won't cost 1/2 the price to fix it. enough features that will allow a mix of formats and film types with very good resolution and noise control.

I agree that many MF dedicated scanners are in the $1500-2500 or more.... But, this a multipurpose flatbed, not a dedicated scanner for one or two film sizes.
Which are always close to double the price of a high end flatbed or more. I don't think making an industrial strength flatbed with the most expensive parts is where the OP'r is headed.
I think he wants to put a "Hurt" on Epson with a better product using newer technology. And with that said... I hope he does, and I will line to buy one myself.

 

[giganova]

Adjustable focus or autofocus
Flat scans (glass if needed to get them flat)
Scan area large enough to include film perforation