Espon V-700 vs. Nikon Coolscan LS-9000 Film Scanner

[bherman]

All;

I'm sorry if this has been posted already, but I've been trying to get down into the weeds and possibly justify a major expenditure.

I already own an Epson V-700 and have purchased Doug Fisher's (betterscanning.com) improved 120 inserts as well and the ANR Glass option.

While I don't shoot professionally anymore, I am perhaps more critical than most and see this is an 'affliction' of sorts, especially when it comes to large expenditures such as this.

I shoot promarily with a DSLR, but also shoot a few rolls of 120/220 a month with my Fuji RFs. I shoot Color negative as well as C-41 B+W. When I scan my 6x7 or 6x9 negatives, I would classify the results as 'pretty good'. Not great and not as good as a direct 'wet-print' off of the negative.

I've been told that a negative scanner such as the Nikon Coolscan ED LS-9000 will yield results that will "blow you away". While I'm inclined to believe that it's true, the scanner is approx. $1800.00, plus $200.00-$300.00 additional if you wan the upgraded negative carriers with the ANR Glass inserts.

Has anybody really compared these? I mean, someone who has the V-700 or V-750 flatbed and actually copmpared it to the Nikon LS-9000?

Quite a bit of $$ is on the line here, and I'm interested in real comparisons that might translate into a justification for a possible purchase.

Thanks;

Brad

 

[Thardy]

There was a thread and examples here a couple of months ago. The Epson v700 or 750 example was really good, but the Nikon gave exceptional results.

 

[Harry Lime]

I have the 9000ED and have experience with flatbed scanners. While the Epson can probably produce decent results I very much doubt that it can match the 9000ED. But then again the 9000 costs a truckload more money.

The 9000ED is in the same ballpark as the low end Imacon 343. Entry level pro units.

 

[Dr. Strangelove]

All;

Has anybody really compared these? I mean, someone who has the V-700 or V-750 flatbed and actually copmpared it to the Nikon LS-9000?

Thanks;

Brad

Very professional and comprehensive reviews on the net:

• Epson V-700
• Epson V-750

Those will cover many questions, including ways to get most out of the scanners and comparison to the Nikon LS-9000 and Canoscan 9950F. The short story is that the Nikon is better, but not much better if you know how to unleash the full potential of the Epsons. This is pretty much the same old story: good results with flatbeds requires a lot of tweaking and adjusting, but it can be done.

Please note that you must read the whole reviews in order to get the full picture. The initial results were not nearly as good as tweaked results.

 

[mfogiel]

bherman
look here:
http://www.flickr.com/photos/59177039@N00/1195860909/in/set-72157602208709970/
to make it really simple: Epson resolves real 2100dpi (at 6400 dpi setting) while Nikon resolves 3800dpi ( at 4000dpi setting), so with Epson you can enlarge 6 times before quality starts to drop, while with Nikon you can enlarge 12 times. If you have Doug's glass, keep it, so if you get the Nikon I will tell you how to use it with standard Nikon holder to keep the film flat. BTW if you scan slides, Nikon gets you more shadow details and truer color to start with, if you only do B&W you have to be careful though which film you use and how dense the neg is. All in all, if you only shoot 6x7 or bigger and do not print bigger than 30x40cm, then Epson is the solution, for smaller negatives or bigger prints, you have to get a real film scanner.

 

[Thardy]

bherman
look here:
http://www.flickr.com/photos/59177039@N00/1195860909/in/set-72157602208709970/
to make it really simple: Epson resolves real 2100dpi (at 6400 dpi setting) while Nikon resolves 3800dpi ( at 4000dpi setting), so with Epson you can enlarge 6 times before quality starts to drop, while with Nikon you can enlarge 12 times. If you have Doug's glass, keep it, so if you get the Nikon I will tell you how to use it with standard Nikon holder to keep the film flat. BTW if you scan slides, Nikon gets you more shadow details and truer color to start with, if you only do B&W you have to be careful though which film you use and how dense the neg is. All in all, if you only shoot 6x7 or bigger and do not print bigger than 30x40cm, then Epson is the solution, for smaller negatives or bigger prints, you have to get a real film scanner.

Aha! This is the test I was talking about.

 

[srtiwari]

Brad,
Your question is very timely (for me). I happen to have both, the V700 (with the Fisher holder) and the Nikon 9000ED (with the AN glass holder). I am going to scan a 6X6 negative (shot on a Rolleiflex TLR 50 years ago) on both scanners, and post the findings , with small 100% crops. Of course, my skills, though adequate will not necessarily squeeze every bit of Info that an expert might be able to. And I will use the Nikon 4 software for one and the Epson software for the other. I do have Vuescan, and it may be a more 'true' test, but I am not as comfortable with all the options. It just confuses me (for now).
Alternatively, if you have a negative that you 'know' well, send it to me, and I can try doing the same test for you.
I should be able to swing all this, this weekend.
mfogiel, I hope you will teach us the Fisher-glass-in-Nikon method.

 

[srtiwari]

Also, Brad, I was able to grab a "Like New" Nikon 9000ED for $1250 not too long ago on the oBoy site. The prices are dropping.
Subhash

 

[mfogiel]

Here is an old thread where you will find the tips:
http://photo.net/bboard/q-and-a-fetch-msg?msg_id=00J7Lp

 

[Dr. Strangelove]

bherman
look here:
http://www.flickr.com/photos/59177039@N00/1195860909/in/set-72157602208709970/
to make it really simple: Epson resolves real 2100dpi (at 6400 dpi setting) while Nikon resolves 3800dpi ( at 4000dpi setting), so with Epson you can enlarge 6 times before quality starts to drop, while with Nikon you can enlarge 12 times. I.

Well, according to this resolution test:

http://www.photo-i.co.uk/BB/viewtopic.php?t=4149

Your estimation of 2100 dpi seem much too low. 56 lp/mm is not achievable with 2100 ppi no matter how much unsharp masking is applied. And you really should USM your flatbed scans; there's nothing wrong with it. Every digital camera picture we see has some sharpening applied or otherwise they would look a lot softer.

 

[oscroft]

Your estimation of 2100 dpi seem much too low. 56 lp/mm is not achievable with 2100 ppi no matter how much unsharp masking is applied. And you really should USM your flatbed scans; there's nothing wrong with it. Every digital camera picture we see has some sharpening applied or otherwise they would look a lot softer.

I've tried scanning with my V700 at 6400dpi, with and without USM, and it doesn't seem to give me any more detail with USM than without. I can see the grain in the results, which suggests that there isn't any more actual resolution to be had. But then, even with USM turned off, I do see something that looks like anti-aliasing at contrast borders, so maybe it's doing some degree of USM even when I tell it not to? (I've been using EpsonScan rather than Silverfast or anything like that, because I need the bulk scanning facility).

 

[shadowfox]

Also, Brad, I was able to grab a "Like New" Nikon 9000ED for $1250 not too long ago on the oBoy site. The prices are dropping.
Subhash

Any chance that we'll see some of the pictures from the scanner posted here for us to look at ?

 

[Dr. Strangelove]

I've tried scanning with my V700 at 6400dpi, with and without USM, and it doesn't seem to give me any more detail with USM than without. I can see the grain in the results, which suggests that there isn't any more actual resolution to be had. But then, even with USM turned off, I do see something that looks like anti-aliasing at contrast borders, so maybe it's doing some degree of USM even when I tell it not to? (I've been using EpsonScan rather than Silverfast or anything like that, because I need the bulk scanning facility).

The true resolution of the V700 is of course nowhere near the 6400 dpi*, mfogiel is quite right about that. 6400 dpi should be able to resolve 63 lp/mm under all conditions, which the V700 clearly does not do. It's probably somewhere around 3000 dpi along sensor axis, which should be enough to resolve 56 lp/mm under most conditions. The resolution along motor axis is lower (45 lp/mm).

Seeing the grain does not mean that there is no resolution to be had though, but you are quickly reaching diminishing returns. On the other hand high resolution film such as Velvia 50/100 or Provia 100F has so much resolution that 8000 dpi scan can actually give slightly better results than 5000 dpi. But not much better.

* Should of course be SPI (Samples Per Inch). DPI is very incorrect unit with scanners. PPI is also wrong, but not as much, since the output file has pixels and not dots, which are exclusively printer units.

 

[srtiwari]

Will, I do plan to test/upload this weekend.
Subhash

 

[oscroft]

The true resolution of the V700 is of course nowhere near the 6400 dpi*, mfogiel is quite right about that. 6400 dpi should be able to resolve 63 lp/mm under all conditions, which the V700 clearly does not do. It's probably somewhere around 3000 dpi along sensor axis, which should be enough to resolve 56 lp/mm under most conditions. The resolution along motor axis is lower (45 lp/mm)

Yep, that sounds perfectly reasonable - thanks for the info.

Seeing the grain does not mean that there is no resolution to be had though, but you are quickly reaching diminishing returns

True. I guess I might just be seeing the larger grain clumps, when there are also smaller ones (and some actual picture resolution that is finer than the larger grain clumps) that could potentially be resolved. But as you say, it's certainly getting into diminishing returns.

On the other hand high resolution film such as Velvia 50/100 or Provia 100F has so much resolution that 8000 dpi scan can actually give slightly better results than 5000 dpi.

Hmm, and I've got a load of Kodachrome 25 slides from the past too - nah, I'm not even going to think it

The one thing I do like about the V700 when compared to my previous Epson 3200 is the far superior shadow detail I get from it (even at the same nominal sampling resolution), and I believe that's an area where the better dedicated film scanners score over flatbeds. I'd love to be able to have a play with a good dedicated film scanner (or then again, maybe I really don't want to - my bank account is hurting enough already )

 

[nightfly]

If you only scan 35mm black and white with fairly dense, traditional emulsion (Tri-X, Neopan 400) negatives, does this change the equation at all?

I've been fairly happy with my Epson 4990 Photo but I would like to upgrade at some point. My main problem is keeping the negatives flat with Epson's crappy negative holders rather than the technical limits of the scanner.

I had one of those Medium Format film holders for an older Epson scanner and found them really fiddly with lots of small plastic parts to loose. Have they improved?

 

[oscroft]

If you only scan 35mm black and white with fairly dense, traditional emulsion (Tri-X, Neopan 400) negatives, does this change the equation at all?

With 400 ASA b&w (so far I've only scanned Delta 400 that was developed in HC-110), I find that a sampling resolution of 4800dpi is more than enough for me to see the grain in some detail, and I'm sure that's beyond diminishing returns and that something lower would be plenty good enough.

 

[MikeL]

mfogiel, I hope you will teach us the Fisher-glass-in-Nikon method.

I purchased a piece of anti-newton glass from focal point (Focal Point), and it's worked great for me. I removed the holders on the medium format tray and just taped one side of the glass piece to the tray. The weight of the glass flattens most negatives. I added another piece of tape and a tab to hold the other end of the glass for really bent negatives. It's nothing fancy, but it works. The difference in sharpness at the negative edges is very noticeable. If you don't get the negative flat, I don't think you'll see any benefit of the 9000 over the Epson.

I had the Epson 750 for a week with Silverfast, and after a week of tweaking, there was a difference between the two. The 9000 can capture detail that the 750 was not able to, despite trying all the different settings for the trays and resolution. The difference was noticeable in 8x10s. Whether the improved resolution is worth the price is definitely up for debate.

 

[espressogeek]

I spent hours with the better scanning.com tray on my v700 and my 645 negs look very soft compared to a 6mp R-D1 file! I just dont get it. Now I did not scan at 6400 because it takes a LONG time and the files are very hard to open even with 2 gigs of ram. Would a 9000 give me the quality I want and will be easy to use or should I go for a 343? I really want to keep shooting film but wasting hours twiddling in front of the computer for a decent scan seems like a frustrating endeavor.

 

[rogue_designer]

Most scans will require some degree of sharpening. Also, I found my results varied wildly depending on what software I was using to run the scanner. With the 3200 vuescan gave me better results, but with the v700 I was testing, I had better luck with Silverfast. Either one was better than Epson's scan software - YMMV

Did you get the adjustable height better scanning tray?

I think you will see a huge difference between the v700 and the 9000 though - and not as big a difference between the 9000 and the 343 - not enough to justify the cost anyway. Now the 848 - now theres a scanner... sheesh.