Epsion 4990 best resolution?

[Capocheny]

Hi Sandy,

I'm not quite visualizing how one can adjust their glass mounting stations to optimize the scanning distance and, since I'm just looking to purchase one of these scanners... does anyone have any pictures of their glass mounting station that they wouldn't mind posting here for us to see?

Thanks in advance.

Cheers

I believe the consensus is that the difference in results one would see between these three scanners using a good wet-mount system would be quite small, assuming care is taken to place the material to be scanned at the best plane of focus. The V750 and V700 scan faster than the 4990 and give slighly better results, but you probably would not see the difference in scans of LF negatives printed at 2X-4X.

The Epson consumer scanners (4870, 4990, V700 and V750) do not focus. For that reason it is important to test and find the plane of best focus, which varies slightly with different scanners. Some of the wet mounting stations provide a mechanism which allows one to adjust the height of the station above the scanner glass to obtain optimum focus. Doug Fisher's mounting station has small plastic screws which can be turned to vary the distance of the station over the glass by several millimeters. Most people find that the optimum distance is 1mm - 3mm above the scaner glass. One could obtain similar adjustments by gluing or taping spacers (small washers, for example), to the bottom of the glass wet-mounting station.

You will also get best results if the negative or transparence to be scanned is wet-mounted to the bottom of the mounting station so that the light from the negative to the CCD passes only through the glass of the scaner itself, not through the glass of the mounting station.

Sandy King

[Gordon Moat]

What is the consensus on the best Epson for wet-mounting every scan? In other words, take the three scanners: 4990/V700/V750, use the same software, and scan them using the wetmount method. Would one see a difference between any of these scanners?

This one:

http://www.epson.com/cgi-bin/Store/c...048262&ref=wcr

I doubt you would find consensus though this is the lowest cost Epson I would recommend to anyone. Basically find out what lab or service bureau scans cost, multiply by how many scans you expect to do in two years, then this should be your budget amount towards buying a scanner.

Ciao!

Gordon Moat
A G Studio

[sanking]

Hi Sandy,

I'm not quite visualizing how one can adjust their glass mounting stations to optimize the scanning distance and, since I'm just looking to purchase one of these scanners... does anyone have any pictures of their glass mounting station that they wouldn't mind posting here for us to see?

Thanks in advance.

Cheers

Go to http://www.betterscanning.com/scanning/mstation.html for a look at Doug Fisher's wet mount system. You might also look at the wet-mount forum on Yahoo moderated by Julio, who also markets a fluid-mount station. http://tech.groups.yahoo.com/group/W...guid=300049323

However, the concept is so simple a picture might acutally complicate understadnding. Basically you have the scanner glass. You don't want to mount the negative directly on the scanner glass because, 1) it likely will not be at the plane of best foucs at that point, and 2) it is messy to mount directly on the scanner glass.

So, you use a second piece of glass (plain window glass is fine), to which you place spacers on the bottom to elevate the station above the scanner glass. You determine how much elevation is needed by scanning at high resolution target with the mounting station glass at different heights above the scanner glass, say from 0.5mm to 4-5mm. You then blow the images up in Photoshop and compare resolution.

Remember, the negative is mounted on the underside of the mounting station, i.e. the side that has the spacers. To mount, you pour a small puddle of mounting fluid on the mounting station, place the negative over the fluid, and gently push or roll the fluid to the edges to remove all of the bubbles. Then tape the negative on all four sides, using a tape that will not leave residue. I use the blue-line masking tape that is available at home supply stores like Home Depot and Lowes. Wipe away excess fluid so it will not drop on the scanner glass.


Some people take a further step and sandwich the negative with a layer of thin mylar, and taping this to the station. I have not found this improves scan quality very much with LF negatives, but it would probably be worth the effort with medium format and smaller negatives.

Hope this makes sense.

Sandy King

[Josh Z.]

I personally found there to be a significant difference between 1900dpi and 2400dpi. Maybe that's just me though. I always scan at 2400dpi now (mostly Velvia 50 & 100), and have excellent results printing up to 16x20". I believe the results even at 20x25" would be satisfactory for me (I'm picky but I don't take a loupe to the print) but have yet to try it.

[sanking]

This one:

http://www.epson.com/cgi-bin/Store/c...048262&ref=wcr

I doubt you would find consensus though this is the lowest cost Epson I would recommend to anyone. Basically find out what lab or service bureau scans cost, multiply by how many scans you expect to do in two years, then this should be your budget amount towards buying a scanner.

Ciao!

Gordon Moat
A G Studio

There is no question but that the Epson 10000 is better built than the consumer level scanners (4990, V700, V750), and if memory serves it also has an auto-focus system, which is a desirable feature. However, unless you need the tabloid size I personally don't believe that it offers much over the consumer level scanners for LF negatives up to 8X10. Some months ago I did a comparison scan of a 5X7 negative with the 4990 and 10000, using Silverfast and a resolution of 2400 dpi with both scanners. I looked at the scans closely and to my eye the one made with the 4990 was at least as good as the one made with the 10000. I actually thought the one made with the 4990 was better, but will settle here for "at least as good."

And if there is no difference in the quality of the scane the price of the 4990, which can be purchased reburbished from Epson at about $280, is sure attractive compared to the $3000+ for an Epson 10000 with TMA.

Sandy King

[audioexcels]

Go to http://www.betterscanning.com/scanning/mstation.html for a look at Doug Fisher's wet mount system. You might also look at the wet-mount forum on Yahoo moderated by Julio, who also markets a fluid-mount station. http://tech.groups.yahoo.com/group/W...guid=300049323

However, the concept is so simple a picture might acutally complicate understadnding. Basically you have the scanner glass. You don't want to mount the negative directly on the scanner glass because, 1) it likely will not be at the plane of best foucs at that point, and 2) it is messy to mount directly on the scanner glass.

So, you use a second piece of glass (plain window glass is fine), to which you place spacers on the bottom to elevate the station above the scanner glass. You determine how much elevation is needed by scanning at high resolution target with the mounting station glass at different heights above the scanner glass, say from 0.5mm to 4-5mm. You then blow the images up in Photoshop and compare resolution.

Remember, the negative is mounted on the underside of the mounting station, i.e. the side that has the spacers. To mount, you pour a small puddle of mounting fluid on the mounting station, place the negative over the fluid, and gently push or roll the fluid to the edges to remove all of the bubbles. Then tape the negative on all four sides, using a tape that will not leave residue. I use the blue-line masking tape that is available at home supply stores like Home Depot and Lowes. Wipe away excess fluid so it will not drop on the scanner glass.


Some people take a further step and sandwich the negative with a layer of thin mylar, and taping this to the station. I have not found this improves scan quality very much with LF negatives, but it would probably be worth the effort with medium format and smaller negatives.

Hope this makes sense.

Sandy King

Thanks Sandy. You are a wonderful contributer to say the very least. I'm only going to be scanning "odd sized" film or if I get an 8X10 back onto the camera, will have a normal sized negative. I will be scanning 4X5 for color and a wee bit of b/w. To the contrary of my use of 4X5, I will be doing my odd sized (whole plate/4X8/6X8) in B/W with some in color. I will have to see what kind of detail I am getting with my color images with 4X5 vs. the whole plate size. If there is enough difference, I won't even bother shooting 4X5. I like to have 4X5 as a cheaper route for color work, though with paper for 8X10 (expired to even 2006) not being much different in price to new 4X5 paper, I don't see why it makes too much sense to shoot in 4X5 especially if a difference can be seen. However, if I decide to go with the HP printer, that eliminates the amount of enlargening I can do and so it may not be even possible to see a difference with the 4X5 and larger sizes in a 12X18 print or a 13X19. I'm very curious why the barrier/threshold of potential enlargements with these Epsons and HPs has that set figure of 13X19, no different than today's standard LF formats of primarily 4X5and 8X10. As one said, for 50 years in UK, Whole Plate was the primary/standard for photographers. Would be nice to see something for the price of the HP that can do 30X40 "maximum" prints (Of course 80X100 would be wonderful, but I think they could put out a machine that can do 30X40 which is basically 2 times the size of print that is currently possible with them).

Regarding wet mounting...The only kit of the two you mentioned that will do 8X10 is the scanscience? one. Doug's can do something close to 8.5 by 6 which isn't too exciting to me considering it eliminates all formats larger than 5X7 w/only exception in my case that would be scanning 4X8's or even 3.5X8.5.

[sanking]

Regarding wet mounting...The only kit of the two you mentioned that will do 8X10 is the scanscience? one. Doug's can do something close to 8.5 by 6 which isn't too exciting to me considering it eliminates all formats larger than 5X7 w/only exception in my case that would be scanning 4X8's or even 3.5X8.5.

If any of the fluid mount stations don't fit your needs it would be a simple matter to assemble your own for either the 4990, V700 or V750. Just have a piece of window glass cut to a size that will fit snugly over the scanner glass, then adjust height over the glass to the optimum plane of focus of the scanner via spacers. Then cut a mask with the same external dimensions as the mounting glass, with the frame cut to match the scanning area of the format. To use simply place the mounting glass over the frame, fluid mount and tape the negative on the under size of the mounting glass (side with the spacers), then flip and place the mounting glass, spacers down, on the scanner glass. This should allow you to scan the full area allowed by the scanner, whatever that may be.

Sandy King

[Chris Strobel]

I have the scanscience 4x5 and 8x10 kits for my 4990.They work quite well, however you could easily make up your own holders for A LOT less money.The holders are just black picture mats cut to size and glued.

[Jiri Vasina]

I personally found there to be a significant difference between 1900dpi and 2400dpi. Maybe that's just me though. I always scan at 2400dpi now (mostly Velvia 50 & 100), and have excellent results printing up to 16x20". I believe the results even at 20x25" would be satisfactory for me (I'm picky but I don't take a loupe to the print) but have yet to try it.

From a mathematical (or digital) standpoint, it's not wise to choose any resolution. The scanner has a physical number detectors (I don't know the exact number for this scanner, but let's assume it's 4800 per inch). If you select a resolution such that if you divide the physical resolution you get a whole number, the output quality is better. If you choose a different resolution, there has to be an interpolation of data between the detectors (hardware interpolation) that would (IMO) degrade the resolution. So if you choose a resolution like 2400, 1200, (maybe 3200, depends on the algorithm, if it does interpolation, or omission of 1 sensor of 4).

The in-computer/in-photoshop resampling algorithms are much better than the ones in the scanner hardware. So IMO (and from what I have read so far elsewhere) it's better to scan at 2400dpi and then resize to desired size - it could be the first step to speed the remaining steps...

[David Luttmann]

From a mathematical (or digital) standpoint, it's not wise to choose any resolution. The scanner has a physical number detectors (I don't know the exact number for this scanner, but let's assume it's 4800 per inch). If you select a resolution such that if you divide the physical resolution you get a whole number, the output quality is better. If you choose a different resolution, there has to be an interpolation of data between the detectors (hardware interpolation) that would (IMO) degrade the resolution. So if you choose a resolution like 2400, 1200, (maybe 3200, depends on the algorithm, if it does interpolation, or omission of 1 sensor of 4).

The in-computer/in-photoshop resampling algorithms are much better than the ones in the scanner hardware. So IMO (and from what I have read so far elsewhere) it's better to scan at 2400dpi and then resize to desired size - it could be the first step to speed the remaining steps...

The scanner doesn't have a resolution. Stepper motors "sample" the image. As such, there is no interpolation of the image whether you use 2400, 3200, or 4800 settings. I find good results for MF & LF at 3200 and then downrez in Photoshop or SAR.