A few questions about the Epson V750M Pro

[timbo10ca]

1) What is the maximum real dpi for scanning (beyond which there is no improvement in scan quality, just increased file size)? I've been using 4000dpi for my Nikon V ED, but the files are still quite manageable coming from 35mm film. A scan of a 5x7 sheet film in 16 bit greyscale at 4800 with the V750 gives a huge file of 1.4 GB- my computer can't handle this!

2) what is its true OD? (specs say 4.0 but I don't know if I believe it).

3) Are the Betterscanning sheet and roll film holders necessary if one was to use the included V750 Fluid Mount kit and only scan using fluid, not the film holders?

4) What is the recommended fluid for fluid mounting

5) is 3mm Mylar ok for fluid mounting, or should it be 1mm?

Thanks,
Tim

[Brian Ellis]

Since nobody else has answered - as I recall from the tests the late Ted Harris did that were published in View Camera a few years ago, the 700/750 is capable of resolving about half its stated capability, i.e. about 2000 dpi. I don't know about the rest of your questions, sorry.

[Steve Wadlington]

[QUOTE=timbo10ca;424652]1) What is the maximum real dpi for scanning

>I have a Epson 4990 which is simular. About max 1500-1800dpi.

2) what is its true OD? (specs say 4.0 but I don't know if I believe it).

>I have scanned a Stouffer step tablet 3.0D. The scanner will capture the range although there is roll off on the high end. I made a spreadsheet to correct to linear for density measurements.

3) Are the Betterscanning sheet and roll film holders necessary if one was to use the included V750 Fluid Mount kit and only scan using fluid, not the film holders?

>You will need to determine your optimum height. On the 4990, I scanned on the glass, 1.2mm up, 3.0mm up. On mine 1.2mm was best. I have a Betterscanning roll holder but don't like it. I use a mask cut from matteboard 1.2mm thick and a glass plate on top for wetmounting.

4) What is the recommended fluid for fluid mounting

>I use Aztec fluid.

5) is 3mm Mylar ok for fluid mounting, or should it be 1mm?

>I use Aztec mylar.

Steve

[Ed Richards]

One caveat - Ted did not run it at the highest rez, which engages the second optical system. I have not seen any of the LF reviewers do this, so it is unclear if it can do better than about 2000. (Not that 2000 is bad at all for a consumer scanner.)

The best review I found is here:

http://www.photo-i.co.uk/Reviews/interactive/Epson%20V750/page_1.htm

[timbo10ca]

One caveat - Ted did not run it at the highest rez, which engages the second optical system. I have not seen any of the LF reviewers do this, so it is unclear if it can do better than about 2000. (Not that 2000 is bad at all for a consumer scanner.)

The best review I found is here:

http://www.photo-i.co.uk/Reviews/interactive/Epson%20V750/page_1.htm

I thought this was a misnomer, and that both lenses could function as "high res". I thought that one lens was to scan on the glass and the other identical lens was to scan film in the holders, as they are fixed lenses working with 2 film positions. I figured that 6400dpi was high res, and scanning at 2000dpi wouldn't be considered high res, regardles of the lens in use. I'm not sure what this whole "high res' thing is all about, other than marketing.

[PenGun]

I thought this was a misnomer, and that both lenses could function as "high res". I thought that one lens was to scan on the glass and the other identical lens was to scan film in the holders, as they are fixed lenses working with 2 film positions. I figured that 6400dpi was high res, and scanning at 2000dpi wouldn't be considered high res, regardles of the lens in use. I'm not sure what this whole "high res' thing is all about, other than marketing.

The "low res" lens scans the entire area and is needed for 8x10, full scale suff.

The "hi res" lens scans about 5.7" down the center where the film holders place the 35mm, MF and 4x5 film.

[PenGun]

Got my V700. It is interesting that my machine is sharpest, 6400 dpi 35mm in the strip holder, with the feet off, at it's lowest position in other words. Others have found the highest setting is sharpest. It's obvious these machines need to be individually dialed in for optimum performance. I'm not sure even lower would not be sharper but that's enough for tonight's testing.

This with Vuescan in Linux.

[Peter De Smidt]

PenGun,

Try scanning with the emulsion directly on the scanning bed. Now flip the negative and scan with the base side down. Which looks best? This'll tell you whether the best position is on the bed or above the glass. If it's above the glass, then you'll need to check various heights. One my Canon, it was best with the emulsion .5mm above the bed.

[PenGun]

PenGun,

Try scanning with the emulsion directly on the scanning bed. Now flip the negative and scan with the base side down. Which looks best? This'll tell you whether the best position is on the bed or above the glass. If it's above the glass, then you'll need to check various heights. One my Canon, it was best with the emulsion .5mm above the bed.

I have just begun testing this puppy. I will find it's sweet spots and use them. Going to test in windose today.

AFAIKT the "low res" lens system focuses on the glass and the "hi res" lens focuses somewhere above it.

[Gene McCluney]

Various Microtek scanners have this low-res, hi-res feature, which scans the full size of the scanning area for "low-res" and then for "hi-res" switches to another lens (internally) and scans a smaller area on the scanning stage. My Microtek Artixscan 2500f will only scan 1/2 the area for "Double Optical Resolution" mode.

[Ed Richards]

Tim,

As I understand, and this is just from reading some of the technical manuals, the second lens system is used only at 6400 DPI. Unless you use Vuescan, 6400 DPI with 4x5 will choke most systems, so no one fools with it. (Vuescan uses a virtual file for big scans, and can be set to reduce the file size before it saves it to disk, which gives the effect of overscanning.)

[Gene McCluney]

The point to be made here is that "some" flatbed film scanners have two lens systems, and two physical optical resolutions, one for each lens system. Usually the higher-resolution lens assembly only scans a portion of the field scanned by the lower-resolution lens system.

What is tricky is that "some" software allows you to set a resolution beyond the hardware resolution you are using, thus the file is upscaled, and not a true higher-resolution file. On my Silverfast software running my Artixscan 2500f, I can dial in 2500 ppi regardless of the physical settings of the scanner, and it will produce a 2500 ppi file but it is really a 1250 ppi file upscaled, however it does have a menu item called "DOR", which I interpret to mean "Double Optical Resolution" and in this mode (when switching and initiating the first scan) you can hear the scanner making several mechanical adjustments inside to switch to the smaller field higher resolution optical system. It then produces a tru 2500 ppi scan of about 1/2 the whole platen area. My experience is just with the Microtek scanner, but perhaps the Epson is similar?

[PenGun]

Tim,

As I understand, and this is just from reading some of the technical manuals, the second lens system is used only at 6400 DPI. Unless you use Vuescan, 6400 DPI with 4x5 will choke most systems, so no one fools with it. (Vuescan uses a virtual file for big scans, and can be set to reduce the file size before it saves it to disk, which gives the effect of overscanning.)

Vuescan will scan to a .jpg which is certainly useful for focus issues. I'm used to Linux which seldom has memory/display problems. Sometimes we hang display for a while if the environment is really harsh.

[jb7]

I've just unpacked my V750, and I've come across the 8x10 film area guide-
Now, I don't have any 8x10's to scan yet, but it looks like I simply place the film within the guide area, directly on the glass- would this be correct?

I've read here, probably, that a sheet of ANR glass placed on top of the back of the neg should be used to keep the neg flat-
However, I've also read (elsewhere) that some brands of anti reflective picture framing glass might perform a similar function for much less money- as well as being much easier to source (for me)

Might I be on the right track?

The film area guide doesn't exactly fill me with confidence-
but presumably it is necessary to use it- for the calibration dots?

j

[Doug Fisher]

>>As I understand, and this is just from reading some of the technical manuals, the second lens system is used only at 6400 DPI. <<

This is not true. Some bad Epson marketing materials started this idea and then the misinformation has taken on a life of its own from there. You can scan at one of many resolution levels with both lenses.

Doug
---
www.BetterScanning.com

[zedxed]

I am looking at getting the v750 scanner and I live in Australia, but epson does not import this product into australia, so I am going to have to get it from the US, Over here in aus our power is 240volt, is anyone able to tell me if the ac power adapter supports an input of 240 volt. Thanks heaps

[rdenney]

I am looking at getting the v750 scanner and I live in Australia, but epson does not import this product into australia, so I am going to have to get it from the US, Over here in aus our power is 240volt, is anyone able to tell me if the ac power adapter supports an input of 240 volt. Thanks heaps

I will look when I get home this evening. Even if it doesn't, however, a routine transformer/adapter should do the trick.

Rick "wondering if Epson markets it under a different name in Oz" Denney

[Paul H]

I am looking at getting the v750 scanner and I live in Australia, but epson does not import this product into australia, so I am going to have to get it from the US, Over here in aus our power is 240volt, is anyone able to tell me if the ac power adapter supports an input of 240 volt. Thanks heaps

You could just buy the V700, add a third party wet mounting kit (Doug's universal kit maybe), and upgrade the Silverfast software yourself. You'd then have a V750 in all but name. At least that way you'll have no warranty worries.

[jb7]

My transformer is rated at 220-240v input-
which makes me think that the US version will have a 110v version-

You could always price the power supply from Epson, I suppose-

[zedxed]

You could just buy the V700, add a third party wet mounting kit (Doug's universal kit maybe),.

Thanks for the replies.
Where do I find (Doug's universal kit maybe)?
I'm thinking though If I just get the V700 here in aus and the software and the wet mount kit for the cost of all of that i may as well get it imported and chase up a power adapter.

I can purchase the V750 from the UK and that would have the correct voltage, but after looking into this a bit more i discovered that in the UK epson forgot to include the wet mount adapter in a heap of the box's so it can be hit or miss to get one, if its not in the box you have to contact epson to get them to send it to you but they will only ship to a UK address.

[Brian Ellis]

I've just unpacked my V750, and I've come across the 8x10 film area guide-
Now, I don't have any 8x10's to scan yet, but it looks like I simply place the film within the guide area, directly on the glass- would this be correct? . . . The film area guide doesn't exactly fill me with confidence-
but presumably it is necessary to use it- for the calibration dots?

j

That's how my 4990 works with 8x10 film - no holder, you just put the film on the glass. I imagine the 750 is the same but someone who owns a 750 should be able to tell you for sure. I never found it necessary to use the 8x10 guide. On the 4990 it's easy to see where the film needs to be placed without the necessity of the guide. But I don't what the "calibration dots" are, maybe that's something peculiar to the 750. If they're on the 4990 guide I've never seen them.

[jb7]

Thanks Brian, I kinda figured it out-
The guide is just that- a guide, and not particularly accurate-
but it works-

[Paul H]

Thanks for the replies.
Where do I find (Doug's universal kit maybe)?
I'm thinking though If I just get the V700 here in aus and the software and the wet mount kit for the cost of all of that i may as well get it imported and chase up a power adapter.

I can purchase the V750 from the UK and that would have the correct voltage, but after looking into this a bit more i discovered that in the UK epson forgot to include the wet mount adapter in a heap of the box's so it can be hit or miss to get one, if its not in the box you have to contact epson to get them to send it to you but they will only ship to a UK address.

Doug's holders can be found on his website, http://www.betterscanning.com/

You don't have to get the upgraded Silverfast software - you may find the Epson software does all you need (it's all I use), and you also have the option of buying Vuescan as well.

With something like this, I'd buy locally - I prefer to spend a little extra, and have the "comfort" of the local warranty.

Good luck!

[percepts]

you can only scan at 6400dpi with the super resolution lens. The spec sheet is quite clear on this and it says that when using the film holder, the scan width is 149mm = 5.866ins.

Then if you look at the sensor spec it says that when scanning at 6400dpi you get two lines of 18880 pixels. Those two lines are offset by aprox half a pixel so that you get a scan sensor width of 18880 x 2 = 37760 pixels. So 37760/5.866 = 6436spi.

The so called super resolution lens is just a lens with a narrower field of view which is always spread over 37760 sensor pixels. The net result is apparent higher optical resolution but that is the trick of flatbed scanners.

If you are using the so called low res lens, then its field of view is wider and it's always projected over 20400 x 2 = 40800 sensor pixels. Its scan subject width is always 216mm = 8.504ins and 40800/8.504 = 4797spi.

So the v750 has two native scan resolutions. 4797 (4800 rounded) and 6436 (6400 rounded).

Any physical scan no matter what you set the scan resolution to will be done at one of those two resolutions as the scanner is hardware limited at those two resolutions. The internal software will set the output at what you tell it to scan at but it can't physically alter the number of hardware pixels can it, and each lens only has one focal length. That is why the usual recommendation is to scan at native resolution because you then have control over the up or down sizing in PS which means you can possibly do a better job than the default algorythms in the scanner internal software.

[Paul Kierstead]

Although I have not been methodical about it, I have found better results in Vuescan by setting scan resolution to 6400 and output resolution to whatever I want (letting vuescan down-res) then by setting scan resolution to what I want.

But, as I said, I haven't been methodical about it, and the change is not ground shaking (if it was, I would be methodical to verify and check if something was wrong).

[Keith S. Walklet]

Nicely explained percepts!

[percepts]

I would add that I don't know the exact method the scanner uses to size for output.
For example, if it is using the wider angle lens (4800dpi) and you scan at 2400dpi or less, it is quite possible that the scanner will only use 1 line of 20400 so the native res becomes 2400. And if using the high res lens and you scan at 3200dpi, or less, the same might be happening. That suggests that there might not be software down res in some circumstances. And infact it may always be throwing away pixels rather than using software down res. But however you look at it, it will never be as good as a drum scan which can adjust the physical sampling size to suit the required output size (within limits of available sample aperture sizes).

[Keith S. Walklet]

it will never be as good as a drum scan which can adjust the physical sampling size to suit the required output size (within limits of available sample aperture sizes).

Would it be more accurate to say, "It will not be able to match the limits of a drum scan, which can adjust the physical sampling size to suit the required output size..."?

I think a large part of the debate about the quality differences in the two technologies resides in their limits. There is a significant disparity in the quality achievable at the maximum resolution of a drum vs. a consumer flatbed. Simply put, the drum has higher achievable limits. What seems to be at issue is whether the quality difference comes at a point beyond which it is noticeable in print form?

If the standard scan resolution recommended for a drum scan of a 4x5 is 2000spi, and a flatbed can approximate that resolution when run at its maximum optical setting (say 6400spi, with measured true output of 2200-2400lpi), does the extra resolution achievable with the drum matter?

Not as much perhaps? Certainly DMAX does matter, especially for "challenged" originals. And the "feel" of the scan, which is something that is much more difficult to quantify without observing work first-hand.

Also, which tool to use can be influenced by what media is being scanned, with grain seeming to be less desirable in most instances for chrome, but clearly a big part of the B&W experience. A drum will resolve the grain much more easily, if that is what you're after.

Again, terrific descriptions.

[percepts]

Well yes all machines have limits and to go the extra mile you have to pay the price. Flatbeds are pretty cheap considering what they can do when compared to the price of a drum scanner.

One thing the flatbed sensor does imply is that by having 2 lines with an offset of 1/2 to 1 pixel between the lines, is that there is spacing between the pixels which makes the second line give some useful additional information. If there is not spacing, then that would explain why no more than 2400dpi resolution is achievable or 3200dpi in the case of the 6400dpi lens. I haven't pulled one of these machines apart and inspected the sensor under magnification but if anyone has one which is shot, then maybe they would to do so and report back. It could be that the real purpose of the second line is to give pixel overlap which would provide smoother tonal change between pixels rather than real additional resolution.

The manufacturers should say "optical path resolution" because the quoted figures always seem to be interpreted as lens resolution which they are clearly not. They are sensor resolution for scan width. Not the same thing.