4990 versus V700

[Paul Warner]

Thinking of graduating to the V700 from a trusty 4990. Any views regarding quality gains I might expect....if any? Those who have read my views elsewhere on Silverfast (see Plustek forum) will know that I don`t regard the inclusion of Ai as justification for spending £150 or so extra on the V750. (The only other variable, as far as I can ascertain, is in the anti reflective coating and I`ve yet to see evidence that this makes any difference!)
I will use it for rollfilm negatives principally, but 35mm scanning is also important.

[bez]

Paul
As I think I’ve already said, you will see a difference when moving up, as I did. I also decided the V750 is probably not worth the extra.
The V700 is not quite as good as a dedicated scanner, but unless you’ve got £2000 for a Nikon 9000 ...

[Doug Fisher]

To me, the difference in the images betweeen the 4xxx and V Series in Vincent's test is so small that I am pretty sure most people could acheive similar results (in terms of sharpness) with a 4990 if they took the time to test and shim their holders. Try shimming your holders and see if you get any gain. As posted here before, try cut business cards or something like a thin plastic report cover.

Doug

[bez]

Hmm, perhaps I didn’t get the best out of the 4990. It never occurred to me to try changing the height, until I got the V700.

Doug – if it’s only got one lens, where is it focused? and what happens if you need to lower the holder as I would expect, if the single lens does reflective as well...

[Paul Warner]

I am using A Doug Fisher height adjustable film holder for rollfilm on the 4990. When used with the glass insert it is pretty good, maintaining quite critical flatness. However, when carrying out recent tests between Vuescan and Silverfast (see Plustek forum) I was amazed to find a difference in sharpness between those scan programmes using the same negative. This suggested to me that either S/fast use some sharpening whether you like it or not, or the seemingly built in micro contrast of that programme had an effect rather like that of sharpening (One of the reasons why I have more or less decided not to use S/fast any more!). Has anybody tested similarly using the 4990 and the V700?
The only difficulty with the DF holder is that there are several height adjusters ( 8 in all) and adjusting them precisely and in parallel is rather tedious and a little haphazard since it is quite easy to damage the rather fragile nylon screws even when you are accustomed to working with tools at a fairly exacting level. In short, it is very easy to chew them up! At least the Epson `V` film holders have click settings.
The other oddity of the 4990 is the appearance of `fog` beneath the glass. (I`ve heard several reports of this). Kev tells me that you can get at this to clean it (thank you, Kev), but apparently it is likely to reappear. Whether it makes any difference I don`t know. Any observations?
I suppose I also considered changing to the 700 series of scanners because I fear the demise of my Plustek. It gives me great monochrome scans....but I am on at least my third scanner within the year. Datamind are great and have readily replaced faulty units, but there has to be a limit!

[Doug Fisher]

Bez - it is supposed to be at 1 mm off the glass but you will find variance. I think they error to the high side just to make it less likely they have any that focus below the glass. There are a few people who have tested who found their scanner focused at the glass bed level or lower. Unfortunately, there isn't much you can do for that except hope you are still under warranty! I can make holders that suspend at .5 mm but then you start running into more Newton Ring issues.

Paul - if you need more set screws, just let me know. Yes, I could have probably gotten by with fewer screws but I opted for more stiffness/over-engineering since in reality once the height is set you will probably never need to change it.

Doug

[Paul Warner]

Hi Doug,
Nice to see you are keeping a proprietorial eye on things. The film holder is generally superb (I unhesitatingly recommend them!) and, yes, I agree that the 8 adjusters are needed given the overall size of the device. I think I might re-calibrate for Vuescan as I am still puzzled about the focus deficiency between S/fast and Vuescan and am becoming quite disilluasioned with the former. In which case I would very much appreciate some more screw adjusters.
Incidentally, does the 4990 film holder also fit the V700, or do I need to start again?

[Doug Fisher]

Yes it can with a small modification to the set screws. Watch for an email from me.

Doug

[bez]

A bit late in the day, but I’ve just managed to find some old 4990 scans on a CD.
Both have been sharpened about as much as possible before artefacts start to appear. It’s also interesting to look at the grey buildings on the skyline, where there’s window detail from the V700 and not much on the 4990.

Of course we don't know what results would have been like if I had experimented with the 4990 before passing it on, although I would have had to make custom holders (or buy Doug's) to get down to 1mm or less...


Original


4990


V700

[grantp]

bez,

Some significant differences between the scan samples there. Are you confident that the scanning application and settings were the same for both? In other words is there possibility that the differences are down to more than just the scanner and the height/etc. adjustment of the original?

Grant

[Paul Warner]

Interesting Bez. There probably might have been some variation in scan procedure and settings, of course. Wonder if someone can make a side by side comparison. If the results were still the same I`d be convinced!

[bez]

Both scans definitely @ 2400dpi with Epson Scan, but I don’t think a slightly different resolution would make very much difference anyway.
I have never used any extras (Sharpening, ICE, Dust Removal etc) and certainly never used SilverFast apart from some tests.

The difference must be due simply to the optical quality of the scanners, or possibly because I never found the sweet spot of the 4990, as Doug suggests, but then it never occurred to me to try...

The 4990 is at a colleague’s house 2 miles away, so if I get the time I will try some more scans at different heights.

[grantp]

The difference must be due simply to the optical quality of the scanners, or possibly because I never found the sweet spot of the 4990, as Doug suggests, but then it never occurred to me to try...

The 4990 is at a colleague’s house 2 miles away, so if I get the time I will try some more scans at different heights.

bez,

Thanks for the confirmation.

Same version of Epson Scan?

What puzzles me a little is that when I tested my 4990 at different heights (you may recall the thread some time ago with Larry's test method using shims) I could find no identifiable difference until I got to 3.5 mm above the bed, i.e. higher than the glass surround and about as far as I could sensible go. So on the glass up to 3.5mm were all the same in my view.

This does not mean that all 4990s would be the same - mine might be good or bad - I have no way to know though I guess we could check that by scanning the same images on the same settings on two or more machines to see what if any differences become apparent.

If my machine is typical of the model and given all of your other settings are consistent then the differences you have identified must be related to the the physical components of the scanner or, just possibly, the way the scanner components works combined with the scanning software.

If you do get a chance to re-run the scans with height adjustments I for one would be very interested in the outcome and further comparisons with the V700.


Grant

[bez]

Grant
Not sure about Epson Scan version as there’s no easy way of checking now, but do you think that would really make a difference? I have never bothered to update it, just used the bundled CD version with both scanners, on the basis of ‘if it isn’t broke...’

As to height, if I read you right you’re saying the 4990 was all the same from 0 to 3.5mm, then it got worse? Compared to the V700 this seems strange, where I can see differences with adjustments of 0.5mm or less, unless the 4990 has a lens with much greater depth of field, which is a possibility given the same lens has to do both reflective and film.
But as I said last week somewhere, I always thought the 4990 produced good results for the money, even though it was my very first scanner...

[Paul Warner]

Hi Bez. I have just been studying your examples again. Impressively better results from the V700 it seems; a much smoother and more revealing scan all round. Tell me- was the original rollfilm, 6X7 perhaps? If you were using the same input and output requirements there is little left to be said, except perhaps to wonder whether your 4990 was performing as it should. Your V700 certainly seems to be.

Kev, I take your point about the reviewer. I was reading it in a fairly pragmatic and `non-tech`way and don`t use my Plustek in the way he seems to. If you scan at full res. all the time everything in the system is likely to get overloaded. 7200 dpi scans seem somewhat coarse anyway. 3600 seems to be about the performance optimum and an A4 file or thereabouts is nicely sized up by Qimage to A3 without detectable problems providing the neg quality was there in the first place (or more if you feel that`s o.k for a 35mm neg!).