Flatbed vs dedicate negative scanner...

[jasonjoo]

Which one is the one for me? So I just got into the film game and have been thinking if I should invest in a dedicated negative scanner such as a Nikon Coolscan 9000 or perhaps a flatbed Epson V750. The main reason for scanning is to archive and back up digitally. I would also post process through Photoshop, but I'm also still leaning towards making prints in an actual dark room. To me, prints that are made in a dark room evoke a different emotion and just look different... but thats another story. I have a few other questions, so if you guys could answer them, I would appreciate it greatly. I'm new to all of this, so please be patient and go easy on me

1) I'll be shooting MF through a Rolleiflex and also 35mm film with an EOS 3 and Canon L lenses. Image quality is important to me, but while I did pixel peep during my digital days, I havent been doing so much of it recently. While my main reason is to make archives and backups, if need be, I'd like to make prints as well. Probably no bigger than 13x19, but occasionally I WOULD like to make 20x30 prints. Is it absolutely necessary to have a Coolscan 9000 then?

2) In terms of "dynamic range," I was always confused if a digital scan would pick up all the "DR" of the film negative. For example, negative film can usually be overexposed and then pulled during development and even burn/dodge during print to bring back blown out highlights, etc. If you take a scan of the negative, do you still have this latitude?

3) I am MOST DEFINITELY not a professional. I do not sell prints, but I do take pride in my work and would like good image quality. However, price is also of a concern. If the Epson V750 could produce decent results, I would jump for it in a heartbeat. However, many people say it's a waste to use a flatbed scanner to scan medium format (or even 35mm) negatives because you're not even taking advantage of those negatives 100%. Does anyone have comparisons between a Coolscan 9000 (or similar) and a high end flatbed scanner?

I've set aside the money for a Coolscan 9000, but again, I really don't know if I can justify owning one. At most, I can only imagine shooting 1 or 2 rolls of 120 film and maybe a couple of 35mm rolls. If I go on a photo trip or any trip for that matter, I'll probably shoot a lot more. I travel at least once or twice a year though...

Again, I appreciate your help! Any suggestions would be awesome.

Thanks,

Jason

 

[Mike Richards]

Nikon 9000 vs Epson 4490

I don't have all the answers, but here is a comparison that speaks for itself. Nikon 9000 on the top, Epson 4490 on the bottom. Scala developed in Rodinal as a negative.

View attachment 205

 

[Pinholemaster]

See if you can find a used Nikon Coolscan 8000.

I have the V750 and the 8000. I'd never scan 35mm on the V750. I only scan MF 6x12s on the V750 'cause the 8000 and 9000 only handle up to 6x9.

 

[jasonjoo]

Walt, what's the difference in quality like? Does the 8000 blow the V750 out of the water?

I'm beginning to think that the Nikon Coolscan 9000 or other similar models maybe something that is just "too" much. However, I am still concerned about the scan quality. If anyone could post up examples of scans from the 9000/8000 and V750, that would be very helpful.

Also, any answers for the DR portion of the question?

 

[toyotadesigner]

I have the V750 and the 8000. I'd never scan 35mm on the V750. I only scan MF 6x12s on the V750 'cause the 8000 and 9000 only handle up to 6x9.

Wrong. If you use VueScan you can scan the 6x12 in 2 parts and just put them together without a stitching software. VueScan now scans up to 89 mm per image...

 

[Pinholemaster]

Yes Jen, I could use the Nikon software or the Silverfast software to scan sections of my 6x12, and assemble them in Photoshop, but I usually don't 'cause I like the custom made anti-newton ring glass holders I have for 6x12 on the V750.

Call me lazy in that regards.

Jason, yes the 8000 blows the V750 out of the water for small formats. Other then the 6x12 strips, I usually only scan 4x5, Type 55 and 8x10 on the V750. I have a large piece of anti-newton ring glass to hold the film flat to the scanner glass.

 

[lonelyboy]

I used an EPSON 4990 for 2 years and I changed it to a Nikon LS9000 one month ago. A very significant is the sharpness is much much better in LS9000 than Epson 4990.

 

[Marco B]

I own a Canon 9950F flatbed, which was the former top of line flatbed / negative scanner of Canon.

I have always been disappointed with image sharpness ABOVE 1200 dpi (up to 1200 dpi is OK, but that is not enough for 35mm). Actually, my big suspicions were recently confirmed in an article in a big German photographic magazine (C'T):

They concluded, in a test of 6 different combined "flatbed / negative" scanners of different brands (EPSON, HP , Canon, Microtek), that NONE were able to scan more than a meager 1200 dpi (ranges from 220(!) - 1552 dpi at best!!!)

They even went as far as stating "That up to now they had *NEVER* seen a flatbed scanner capable of doing 2400 dpi" (that most likely includes the V750 many magazines have been raving about, it wasn't included in this particular recent test, but I am sure they have tested it before)...

I have made excellent high resolution (up to 3600dpi) and SHARP scan's on a IMACON Flextight 848 scanner (up to 4x5 format), but that is a professional machine featuring internal focusing and costing maybe 15.000 euro!

My conclusions:
- Despited bloated claims of "4800 dpi" or higher resolutions, NO flatbed has actually more than about 1200 dpi. It's a shame, because 2400 dpi minimum is needed for acceptable 35mm scanning.
- The manufacturers are blatantly, and knowingly, lying about the true optical resolutions, higher than 1200 dpi is *interpolated* on flatbed...
- Flatbed at 1200 dpi IS usefull for 4x5 image scanning, but probably not even enough for MF.
- Get a film scanner for 35mm and MF.

 

[Marco B]

In terms of DR of scanned images, even my in terms of resolution disappointing 9950F (with a dMax of 3.2 or 3.4, which is not the highest) has NOT disappointed me. It will pick up an amazing amount of detail... so I guesse you shouldn't worry to much about that aspect.

 

[donbga]

I like the custom made anti-newton ring glass holders I have for 6x12 on the V750.

Walt,

How did you make them?

Don

 

[Greg_E]

Google brought up this:
http://fpointinc.com/glass.htm

Keep in mind that anti-newton glass is really a diffuser.

 

[rgacpa]

You Don't Need A Scanner

Jason,

Based on your detailed post, I'd say you don't need a scanner, period. Your negatives/positives are your best archive. Your into darkroom printing so you don't need to digitize for printing.

If you just want to do web display, a dedicated film scanner is over kill. Scan your prints! A flatbed will do...

If you ever want a high reso scan of a frame, you can outsource that or rent time on an Imacon scanner somewhere.

I shoot film and own a Nikon CS8000 (I shoot 6x6 and 35mm). I print digitally (I enjoy that and don't have room for a darkroom), so I need the scanner. Otherwise I wouldn't have one.

That's my $0.02.
Bob

 

[donbga]

Google brought up this:
http://fpointinc.com/glass.htm

Keep in mind that anti-newton glass is really a diffuser.

Hi Greg,

I'm aware of that source for AN glass, I just thought Walt might have something extra to share with us.

Don

 

[Marco B]

Jason,

Based on your detailed post, I'd say you don't need a scanner, period. Your negatives/positives are your best archive. Your into darkroom printing so you don't need to digitize for printing.

If you just want to do web display, a dedicated film scanner is over kill. Scan your prints! A flatbed will do...

If Jason indeed only wants to back-up and do some webdisplay of darkroom printed images, you are spot on and I agree fully, but if he, as he seems to be wanting, also wants to "post-process in PS" (don't know exactly what it means), than a good filmscanner might be useful.

 

[sanking]

My conclusions:
- Despited bloated claims of "4800 dpi" or higher resolutions, NO flatbed has actually more than about 1200 dpi. It's a shame, because 2400 dpi minimum is needed for acceptable 35mm scanning.
- The manufacturers are blatantly, and knowingly, lying about the true optical resolutions, higher than 1200 dpi is *interpolated* on flatbed...
- Flatbed at 1200 dpi IS usefull for 4x5 image scanning, but probably not even enough for MF.
- Get a film scanner for 35mm and MF.

Marco,

I have tested a number of scanners with high resolution chrome on glass targets. The best Epson flatbeds (V700 and V750) are capable of resolution on the order of 2200 ppi. The Epson 4990 gives about 1800 ppi. With both scanners you can get *almost* maximum resolution with the scanner set for 4800 ppi.

By contrast, the Nikon 9000, which is also a flatbed, though a dedicated one, delivers about 2600 with the setting at 4000 ppi.

My Leafscan 45, now about thirteen years old, delivers about 4000 ppi when the scan is at 5040 ppi. This is only available for 35mm. For medium format it delivers about 2400 ppi when the scan is done at 2540 ppi. My Scitex/Creo EverSmart Pro, which is a high end flatbed, delivers about 3000 ppi when the scan is done at 3175 ppi.

Others may have different results, but this is what I have found in testing the above equipment.

Sandy King

 

[toyotadesigner]

Marco,
By contrast, the Nikon 9000, which is also a flatbed, though a dedicated one, delivers about 2600 with the setting at 4000 ppi.

How did you conduct the tests? Can you give me some details? I'm just curious, because our LS 9000 delivers 3.980 ppi when set @ 4.000 ppi.

...My Scitex/Creo EverSmart Pro, which is a high end flatbed, delivers about 3000 ppi when the scan is done at 3175 ppi.
Sandy King

But you know that the Creo is even more expensive than the Imacon? You compare apples and pears in this aspect.

 

[sanking]

How did you conduct the tests? Can you give me some details? I'm just curious, because our LS 9000 delivers 3.980 ppi when set @ 4.000 ppi.



But you know that the Creo is even more expensive than the Imacon? You compare apples and pears in this aspect.

I conducted the tests by putting a 2X2" high resolution target (250 lppm potential) in a slide carrier and focusing the unit. You can then easily evaluate the effective resolution of the scanner by looking at the image on screen at high resolution. Even with optimistic eyes the effective resolution of the LS 9000 barely reaches 50 lppm, which is about 2600 ppi of effective resolution. Of course, that is not bad, in that it is about 70% efficient, which is a lot better than the Epson flatbeds.

I will say for fact certain that no LS 9000 will deliver 3980 ppi of *effective resolution* when the theoretical maximum is 4000 ppi. Not even drum scanners and high end flatbeds are capable of that kind of efficiency. So if your tests actually indicate effective resolution of 3980 ppi you need to find a better method of testing, or change the methodology, because the LS 9000 is simply not capable of 80 lppm of resolution.

Not sure what you mean by apples and pears comparison. I did not compare a Creo to an Imacon, don't even have access to an Imacon for that matter.

Sandy King

 

[jd callow]

I conducted the tests by putting a 2X2" high resolution target (250 lppm potential) in a slide carrier and focusing the unit. You can then easily evaluate the effective resolution of the scanner by looking at the image on screen at high resolution. Even with optimistic eyes the effective resolution of the LS 9000 barely reaches 50 lppm, which is about 2600 ppi of effective resolution. Of course, that is not bad, in that it is about 70% efficient, which is a lot better than the Epson flatbeds.

I will say for fact certain that no LS 9000 will deliver 3980 ppi of *effective resolution* when the theoretical maximum is 4000 ppi. Not even drum scanners and high end flatbeds are capable of that kind of efficiency. So if your tests actually indicate effective resolution of 3980 ppi you need to find a better method of testing, or change the methodology, because the LS 9000 is simply not capable of 80 lppm of resolution.

Not sure what you mean by apples and pears comparison. I did not compare a Creo to an Imacon, don't even have access to an Imacon for that matter.

Sandy King

Sandy have you tested any of these scanners for true dynamic range, specifically dmax?

 

[sanking]

Sandy have you tested any of these scanners for true dynamic range, specifically dmax?

JD,

No, I have not tested any of these scanners for dynamic range. Being able to capture high Dmax is not a major concern for me since I use primarily color and B&W negative film.

Sandy

 

[Ted Harris]

If ypo are only scanning up to 6x9 then the Coolscan 9000 is the answer. It has a real resolution of ~ 4000 spi. None of the flatbed consumer scanners come close to that. The 750 is slightly less than 2400. Additionally the 9000 has better Dmax than the 750.

The consumer flatbeds do an acceptable (not fabulous) job for 4x5 and larger film, ok for some 6x9 too but for 35mm.

 

[sanking]

If ypo are only scanning up to 6x9 then the Coolscan 9000 is the answer. It has a real resolution of ~ 4000 spi. (not fabulous) job for 4x5 and larger film, ok for some 6x9 too but for 35mm.

Ted,

I think you need to revisit your tests of of the Nikon 9000. No way it will deliver an effective resolution of 4000 spi, or close to it. I generally respect your test results, but in this case you are way off from my own tests of several Nikkon 9000 units. I never saw real resolution of more than 3000 ppi.

Sandy King

 

[Jeff Searust]

I am amazed at the distinct difference in the scans posted of the 900 vs. the 750.

 

[sanking]

I am amazed at the distinct difference in the scans posted of the 900 vs. the 750.

The comparison I saw was between a Nikkor LS 9000 and an Epson 4990, not a 750. However, there is not a lot of difference between a 750 and 4990 in terms of sharpness.

The comparison does not look very realistic to me. There is no question but that the LS 9000 is a much sharper scanner than an Epson 4990 but the difference is nowhere near as great in my experience as the comparison would suggest.

Sandy King

 

[Carl Radford]

I've been offered an old flexitight scanner for near enough the same money as a v750. I want to scam medium format negs. The 750 would allow me to scan some lf stuff as well but I get from this thread that I'd be better of getting the flexitight than a new 750 and use a friends 750 or 4990 when I need to?

 

[Daniel Balfour]

Wow, I was actually going to pick up an Epson V-700/750 tomorrow - probably still will. The idea of being able to do post-process editing digitally & output near-perfect digital negatives for traditional & alt photography processes is too tempting to pass up.

In no way do I doubt the results posted by folks like Sandy King, I can't help but wonder. What difference does it make if this car goes 300 mp/h and that one only goes 225 mp/h. I'm not going to be driving any faster then 75-85 mp/h!

I've done most of my shooting in MF and some in 35mm. My enlargements were generally 8x10 and some even smaller, with the odd 11x14. The losses in quality inherent to the enlargement process seem to exceed the differences in the quality of the scans that can be had with ANY of these scanners!

Correct me if I'm wrong - I'm just starting out with digital negatives.