Ad
Ad
Ad
Pages: [1] 2 3 4 »   Bottom of Page
Print
Author Topic: Epson Scan Software  (Read 6036 times)
wlemann
Jr. Member
**
Offline Offline

Posts: 50


« on: December 26, 2013, 03:27:57 PM »
ReplyReply

I am using the V-700 Perfection Epson scanner and am confused about the Epson Scan software in Professional Mode.
I think I get the resolution issue. 
As to the "Target" what does this mean?
Same with the size boxes?
Thanks, Walter
Logged
xocet
Newbie
*
Offline Offline

Posts: 26


« Reply #1 on: December 26, 2013, 04:15:40 PM »
ReplyReply

You can do the scanning in one of two ways - scan to a specific output (target) size and resolution, or scan to original size.

With the first option, the software works out what resolution to scan at in order to produce a file that will print at the specified output size.

Your best option is to scan at 2400 or 3200 dpi for film (or 200/300 for document scans), output set to original size.  Then process the scanned file in your favourite image editor, and resize to suit your target output size.
Logged
wlemann
Jr. Member
**
Offline Offline

Posts: 50


« Reply #2 on: December 27, 2013, 09:58:34 AM »
ReplyReply

Thanks for your response:

So for full frame 35 mm film, the 3200 res should be plenty to allow a "full size" print (24" wide X x") on my Epson 7900, correct?

How would you compare the quality achieved with the Epson Perfection V-700 compared to an older Hasselblad/Imacon Flextight?  Are they close?  The Epson is much easier to use, certainly.

Thanks,  Walter
Logged
Peterretep
Jr. Member
**
Offline Offline

Posts: 85


WWW
« Reply #3 on: December 27, 2013, 10:08:40 AM »
ReplyReply

I've never used a Imacon Flextight but I very much doubt you'll get as good of a quality scan from the Epson, you get what you pay for often holds true.Try Vuescan software!

Peter
Logged

TonyW
Full Member
***
Offline Offline

Posts: 120


« Reply #4 on: December 27, 2013, 11:24:32 AM »
ReplyReply

Thanks for your response:

So for full frame 35 mm film, the 3200 res should be plenty to allow a "full size" print (24" wide X x") on my Epson 7900, correct?

How would you compare the quality achieved with the Epson Perfection V-700 compared to an older Hasselblad/Imacon Flextight?  Are they close?  The Epson is much easier to use, certainly.

Thanks,  Walter
I doubt very much that the image quality of the Epson is anywhere near that achievable with the Imacon Flextight although the specifications for maximum spi may be similar.

The problem is that although the maximum optical resolution of the Epson is quoted as 6400 spi for the Epson it is only able to resolve detail at around 2300 spi.  Therefore while you will get the added pixels and a bigger file size there is no real increase in the ability to resolve the finest detail past 2400 spi.

The cheaper Imacon X1 claimed 6300spi is the highest resolution and from test I have seen indicate it gets close to this at around 6150 spi. Therefore it should be possible to resolve finer detail that may be in the original.

Testing is fine of course and the result quoted above are based on others testing using USAF test charts but of course in the real world we are usually more interested in how the images look and print and it is quite possible to get very satisfactory results at even lower sampling rates.  For instance I use the V500 and its real limits to resolving detail peak at about 1500 spi but still provide an acceptable result - most of the time!  Obviously there is a limit to what can be achieved with the cheaper flat bed scanners over a dedicated film scanner and you may have the need to go to a film scan provider for the ultimate scan using an Imacon or similar

Logged
rgs
Sr. Member
****
Offline Offline

Posts: 359



WWW
« Reply #5 on: December 28, 2013, 09:32:18 PM »
ReplyReply

I use the Epson software for reflective scans and it's "copier" functions. When scanning film, I use VueScan or SilverFast (not the cut down version that comes with the scanner), usually SilverFast. I use multi-scan (does the Epson software do that?) and avoid all sharpening, digital ICE, or other manipulation. I do those in LR of PS where I can control them more precisely. I usually scan at 2400-3200 and then down rez as needed. My 4x5 and 6x7 chromes scan nicely (negs are a little trickier), but 35mm is not so good and would be better with a film scanner or good scanning service.
Logged

artobest
Sr. Member
****
Offline Offline

Posts: 258


WWW
« Reply #6 on: December 29, 2013, 10:42:50 AM »
ReplyReply


The problem is that although the maximum optical resolution of the Epson is quoted as 6400 spi for the Epson it is only able to resolve detail at around 2300 spi.  

Actual resolution is very much dependent on the film's proximity to the scanner's true focus plane, and to improve this you need the betterscanning.com holders - Epson holders can adjust up or down, but only in crude increments. If you get custom holders and spend an hour or two getting them set up to your particular scanner, you should be able to exceed 2300. Furthermore, the scans have, to my eye, pleasing colour and tonality, although they do show chromatic aberrations that clean up nicely in Lightroom or Camera Raw.

Here's a scan made on a V750 from 6x6 Portra, downsized to 3000 pixels/side: http://www.flickr.com/photos/61171860@N05/10711368395/sizes/o/
« Last Edit: December 29, 2013, 11:18:17 AM by artobest » Logged

TonyW
Full Member
***
Offline Offline

Posts: 120


« Reply #7 on: December 29, 2013, 05:46:51 PM »
ReplyReply

Actual resolution is very much dependent on the film's proximity to the scanner's true focus plane, and to improve this you need the betterscanning.com holders - Epson holders can adjust up or down, but only in crude increments. If you get custom holders and spend an hour or two getting them set up to your particular scanner, you should be able to exceed 2300. Furthermore, the scans have, to my eye, pleasing colour and tonality, although they do show chromatic aberrations that clean up nicely in Lightroom or Camera Raw. ...
Very good point and while I agree that you should establish the scanners true focus point by raising and lowering the film holder using test images over the scanner bed the fact remains that there is a limit to what the scanner is capable of resolving once you have set the sweet point - this will almost always (for flatbed scanners) be much less than the manufacturers stated maximum optical resolution. 

Dedicated film scanners such as the now deceased Nikon and Minolta range are capable of resolving detail equating to close to their stated optical resolution of around 4000 SPI

Typically it seems that flatbed scanners ability to resolve real detail is limited to 1/3 to 1/2 of the stated optical resolution and in the case of the Epsons from low to high price range somewhere between 1500 SPI to 2300 SPI vs the 6400 quoted maximum (uninterpolated).  While there may be small differences between scanner samples with some possibly scoring a little higher than others the ability to resolve the finest detail contained within transmissive media hits this limit. 

Going further than this point will give you more pixels to play with a larger file size but no increase in the scanners ability to actually resolve detail  that may be contained within the film or transparency.  Downsampling may give the impression of a better sharpness and will certainly aid in disguising noise

Testing a small section of your own images at varying sampling rates should reveal the point where there is no increase in the scanners ability to resolve detail.  However this is really so dependent on how much detail is in the original and how well you have recorded the detail so if you really want to determine the actual resolution of a scanner this will require a dedicated test target.  One that is frequently quoted as useful is the USAF-1951 target.  Scanning this target as per the instructions and then using the supplied charts will reveal the true resolution of your scanner

As I understand it the USAF 51 target it is not as accurate as required by the ISO standards but should show the differences in quality between scanners and the ability to resolve detail.  The ISO standards used to measure resolution are I believe ISO 16667 and ISO 14773 not sure how they differ and I have never gone down this route so cannot comment.
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8848


« Reply #8 on: January 07, 2014, 06:07:16 PM »
ReplyReply

The problem is that although the maximum optical resolution of the Epson is quoted as 6400 spi for the Epson it is only able to resolve detail at around 2300 spi.  Therefore while you will get the added pixels and a bigger file size there is no real increase in the ability to resolve the finest detail past 2400 spi.

Going further than this point will give you more pixels to play with a larger file size but no increase in the scanners ability to actually resolve detail  that may be contained within the film or transparency.  Downsampling may give the impression of a better sharpness and will certainly aid in disguising noise

Testing a small section of your own images at varying sampling rates should reveal the point where there is no increase in the scanners ability to resolve detail.  However this is really so dependent on how much detail is in the original.....

I  recently bought the Epson V700 with the hope of being able to finally digitize all the thousands of negative and slides I have that go back to the school days of my father. My previous scanning attempts started with the Nikon LS-2000 (a mere 2700 dpi, but no doubt less in practice) and ended with the Minolta Dimage 5400 II. I concentrated only on a select few slides and negatives, re-scanning them as better and more affordable scanners became available.

I'm disappointed that this process of continual development of affordable, dedicated film scanners seems to have stopped. I too read that the real resolution of the V700 is claimed to be only 2300 dpi, which is a long way off 6400 dpi. However, what I've found is that there does appear to be an advantage in scanning at the full resolution of 6400 dpi. One might not initially notice any additional detail in the 6400 dpi scan, compared with a 3200 dpi scan, or even 2400 dpi scan, but the larger file of the 6400 dpi scan allows for better sharpening during post-processing.

One technique for sharpening I read about years ago (on this forum) is to interpolate an image to several times its size, apply the more aggressive sharpening that the larger image can take, then downsample the image to its original size using Bicubic Sharper. The results do seem to be noticeably sharper without increasing the noise.
However, upsampling an images tends to introduce a certain degree of softness, which limits the benefits of such a method.

By scanning at 6400 dpi, as opposed to 2400 or 3200 dpi, one is certainly not introducing any softness. In fact one might get a more accurate result due to an absence of 'grain aliasing', in the same sense that a very high-resolution sensor has less need of an anti-aliasing filter.

For this reason I'm using the full resolution of 6400 dpi when scanning. Hard drives are so affordable nowadays, so disc storage should not be a problem. How many 6400 dpi scans can one fit onto a 2 TB pocket drive?
Logged
TonyW
Full Member
***
Offline Offline

Posts: 120


« Reply #9 on: January 07, 2014, 06:52:05 PM »
ReplyReply

...I too read that the real resolution of the V700 is claimed to be only 2300 dpi, which is a long way off 6400 dpi. However, what I've found is that there does appear to be an advantage in scanning at the full resolution of 6400 dpi. One might not initially notice any additional detail in the 6400 dpi scan, compared with a 3200 dpi scan, or even 2400 dpi scan, but the larger file of the 6400 dpi scan allows for better sharpening during post-processing.

One technique for sharpening I read about years ago (on this forum) is to interpolate an image to several times its size, apply the more aggressive sharpening that the larger image can take, then downsample the image to its original size using Bicubic Sharper. The results do seem to be noticeably sharper without increasing the noise.
However, upsampling an images tends to introduce a certain degree of softness, which limits the benefits of such a method.

By scanning at 6400 dpi, as opposed to 2400 or 3200 dpi, one is certainly not introducing any softness. In fact one might get a more accurate result due to an absence of 'grain aliasing', in the same sense that a very high-resolution sensor has less need of an anti-aliasing filter.

For this reason I'm using the full resolution of 6400 dpi when scanning. Hard drives are so affordable nowadays, so disc storage should not be a problem. How many 6400 dpi scans can one fit onto a 2 TB pocket drive?

Hi Ray,
Interesting method and certainly grain aliasing needs to be accounted for.  I accept that you are seeing an improvement with this method.  I wonder though how much benefit is gained when downsampling particularly when downsampling by large amounts.  My concern would be the fact that we have no control over the pixels to be discarded therefore perhaps it would be better to gradually downsize in 10-20% steps and choosing an algorithm that adds no sharpening to the image until the last step to final size (I read this somewhere but cannot remember the source).   

I have not done any film scanning for a long time and have not tried this or another technique that I have seen recommended.  This one came from the Vuescan bible.
Set the scan resolution to the maximum optical resolution e.g. 6400 and set the scale to 50% (3200).  It was stated that the effect could not be replicated in an editing application after but has to be done during scanning.  I will probably give this method a try next time I scan a film and hope it is not a voodoo move
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8848


« Reply #10 on: January 07, 2014, 09:06:05 PM »
ReplyReply

Hi Ray,
Interesting method and certainly grain aliasing needs to be accounted for.  I accept that you are seeing an improvement with this method.  I wonder though how much benefit is gained when downsampling particularly when downsampling by large amounts.  My concern would be the fact that we have no control over the pixels to be discarded therefore perhaps it would be better to gradually downsize in 10-20% steps and choosing an algorithm that adds no sharpening to the image until the last step to final size (I read this somewhere but cannot remember the source).   

I have not done any film scanning for a long time and have not tried this or another technique that I have seen recommended.  This one came from the Vuescan bible.
Set the scan resolution to the maximum optical resolution e.g. 6400 and set the scale to 50% (3200).  It was stated that the effect could not be replicated in an editing application after but has to be done during scanning.  I will probably give this method a try next time I scan a film and hope it is not a voodoo move


Hi Tony,
There are so many different options to compare. I haven't yet downloaded the Vuescan software for the V700. Maybe I should give it a try. When the Minolta Dimage 5400 II was no longer supported by the manufacturer regarding updated drivers for the new 64 bit operating systems, I found that Vuescan did include a driver that worked. I now find that Nikon do not provide drivers for my Coolscan 8000ED that allow me to use the scanner with Windows 7, but Silverfast does.

The reason I bought the V700, despite still having a couple of functioning, dedicated film scanners, was the potential for speeding up the scanning process. The 35mm film holder for the V700 can hold up to 24 frames. What this means in effect is, I can spend just a few seconds making individual adjustments for each frame, say  15 or 20 minutes in total, then batch scan all 24 negatives, leaving me free to do other things whilst the scanner's doing its job.

As a rough estimate, if it takes 20-30 minutes to make the adjustments in the scanner's software for each individual frame, including the insertion and removal of the 4 strips of negatives in the film holder, and if it takes an hour to batch scan all 24 negatives, I should be able to scan about 168 negatives during a 10 hour period, but the amount of work required from me during that 10 hour period might be as little as 3 hours. That's the appeal.

However, so far it hasn't quite worked out like that because I've spent so much time trying out the different scanning software and the different options in Silverfast. I was particularly interested in Silverfast's RAW HDR and ME (multiple exposure) scan method which claims to capture all the detail one's scanner is capable of delivering, whilst also preserving the negative aspect of the negatives.

This seemed a great idea because one doesn't have to spend time making individual adjustments for each frame in the scanner software, and one can process the images later as time permits.

Having downloaded the trial version of Silverfast HDR, which is specifically designed for converting the RAW negatives to positives and making the sorts of adjustments one might have made previously if one hadn't scanned in RAW HDR mode, I find that the entire process is unfortunately more time-consuming and that it's much more difficult to get a satisfactory color balance through that route, even when including further processing in Photoshop.

I've now reverted to using the Epson Scan software bundled with the scanner. I find it quicker and simpler than Silverfast, and any deficiencies in the color and contrast of the scanned result are usually fixed instantly with a single click on Auto Color or Auto Tone in Photoshop.

If anyone's interested, my method at this stage is to first click on Color Restoration, which dramatically changes the appearance of the image but also appears to blow out the highlights and block the shadows. I then bring up the histogram, which is similar to the 'levels' histogram in Photoshop, which allows me to unblock the shadows and reduce the highlights, causing the image to look flatter and less contrasty, but never mind because there are other controls in the Epson Scan software that are similar to the Color balance tool and Brightness/Contrast tool in Photoshop, as well as a general Saturation slider.

By playing around with these sliders for just a few seconds, moving Blue towards Yellow and/or Red towards Cyan etc, and increasing saturation and brightness, I find I can get the image in the ballpark so that often (not always) a single click on Auto Color in Photoshop results in a satisfactory outcome, or gets me 90% of the way there.

For some reason, I find it easier and quicker to unblock the deepest shadows using Epson Scan.
Logged
TonyW
Full Member
***
Offline Offline

Posts: 120


« Reply #11 on: January 08, 2014, 09:47:12 AM »
ReplyReply

Ray,
That is the problem, finding the time to do full comparison tests of all options available to us particularly with pressure to produce results and trying to archive 1000s of captures.

If you can find the time it may be worthwhile testing the demo from Vuescan.  It does include some of the features of the full blown version of Silverfast including Multi scanning, Multi exposure and a raw option. 

Vuescan I understand is also compatible with the Nikon CoolScan 8000ED and if not one person at least has found a way to fool the system into submission
http://photo.net/digital-darkroom-forum/00Ryck

Like you I did have a large collection of family images and professional work portfolio that amounted to 1000s of images sadly the vast majority lost around 15 years ago during a house move.  By the time the images were noted as missing several months had gone by and my request to the movers to find what they had lost failed.  Still although the loss was upsetting at least I did not have to start the laborious archiving!

My reason for buying the V500 rather than the V700 or V750 (apart from being naturally tight fisted with my money!) was for photo restoration work.  The vast majority of this work is reflective media and I could not see any quality difference between the Epson scanners V500 vs the V750. 

My thoughts at the time were that as Nikon etc  no longer manufactured film scanners that there would be a glut of used models for sale on ebay etc. at good prices how wrong was I, the prices reached amazed me. 

In the event I was really surprised at the quality achievable with the cheap Epson for film in spite of its modest resolving power in comparison to the higher end models.  I have managed to produce some very pleasing results from old negative and slides far from great but in many cases quite acceptable.  When really needed I have gone out to a scan only service that use an Imacon although prices are a little steep IMHO, ranging from 7 -10+ depending on file size up to a maximum of 100MB

Seems we have had similar experience I started with of course the Epson software, and then tried the Silverfast light? Then the full blown expensive version and finally Vuescan.  Like you I somehow reverted to Epson scan, I quite like the UI and do find it simpler and quicker to use might be a different story with difficult negs or trans.

Good luck with your task of archiving your old images and if you do come up with a better/faster way I hope you will post
Logged
Alan Klein
Sr. Member
****
Offline Offline

Posts: 637



WWW
« Reply #12 on: January 12, 2014, 11:09:22 AM »
ReplyReply

Tony:  Although you said you already purchased the V500, just a comment about it vs. the V600.  Although both units can apply ICE to film scans, the V600 can also apply ICE to prints.  That helps a lot with old prints that have suffered creases, marks and other defects.
Logged
Alan Klein
Sr. Member
****
Offline Offline

Posts: 637



WWW
« Reply #13 on: January 12, 2014, 11:10:27 AM »
ReplyReply

The V600 is only slightly more expensive than the V500 and a lot cheaper than the V700 or V750.
Logged
TonyW
Full Member
***
Offline Offline

Posts: 120


« Reply #14 on: January 12, 2014, 11:15:32 AM »
ReplyReply

Alan, thanks for the heads up I was not aware of the changes between models.

Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8848


« Reply #15 on: January 19, 2014, 08:18:43 AM »
ReplyReply

The problem is that although the maximum optical resolution of the Epson is quoted as 6400 spi for the Epson it is only able to resolve detail at around 2300 spi.  Therefore while you will get the added pixels and a bigger file size there is no real increase in the ability to resolve the finest detail past 2400 spi.

Hi Tony,
I wonder how accurate this assessment of the V700 resolving power is. I've seen only one site claiming this figure of 2300 dpi.

Having got around to scanning some 6cmx6cm B&W MF film on the V700, which is a real joy because of the less significant grain and smoother tonality compared with 35mm, I wondered how much better such images would appear if i were to scan them on my dedicated Medium Format Nikon 8000 ED, which I bought several years ago with the intention of scanning the old MF film I have in my possession, as well as the new MF film I was using before I bought my first cropped-format DSLR.

I never did spend much time scanning MF film on that scanner because the price of DSLRs fell so quickly, and my time then was taken up exploring the new medium. However, Nikon no longer support this scanner with drivers for recent operating systems, so my 8000 ED has remained dormant for a few years.

I recently discovered that Silverfast is selling a version of their software that has Nikon D8000 ED drivers for modern operating systems, so I downloaded the trial version and have spent a bit of time comparing scans of the sharpest of these 6x6 format B&W film I have, on both scanners, using the Epson Scan software with the V700.

I was expecting to see noticeably more detail from the Nikon 8000 ED. Imagine my surprise when I sometimes saw less detail.

One of the complaints about the V700 is it's lack of manual focussing. If the focussing is off, one has to change the height of the film holder. However, in my experience one doesn't have to change the height of the film holder for each individual frame. Once it's right, it remains right. The default height works best for me.

The sort of thing I was seeing in the Nikon scans was the top of the image appearing out-of-focus, and significantly less sharp than the V700 scan, but the bottom of the scan appearing equally as sharp as the V700 scan.
In order to get a good scan out of the Nikon, for comparison purposes, I had to make several scans with different focusing points, then choose the best. It was a long and tedious process, but I had to determine if this 4000 dpi scanner would provide any advantage.

I'm surprised it doesn't. In fact, the Epson V700 flatbed sometimes has a slight edge.

The attached image, with 100% crops, was taken over 50 years ago in Bangkok, with a Rolleiflex. I've chosen it for this demonstration because it's quite sharp and has good DoF. The Epson scan is at full optical resolution of 6400 dpi, and the Nikon scan at full optical resolution of 4000 dpi. No noise reduction was used with either scanner, because this is a test of detail and resolution.

Sharpening has been adjusted to approximately equalize noise in each image. The V700 scan, after sharpening, has been downsampled to 4000 dpi.

The Epson V700 was first released many years ago. Perhaps they have gradually improved the scanner without any major announcement.  Wink
Logged
Doug Fisher
Jr. Member
**
Offline Offline

Posts: 65


WWW
« Reply #16 on: January 19, 2014, 12:21:53 PM »
ReplyReply

>>However, in my experience one doesn't have to change the height of the film holder for each individual frame. Once it's right, it remains right. <<

Correct.  While it is good to reconfirm maybe once a year, unless something like damage/a large shock occurs to your scanner, you will probably never see a change in the optimal film suspension height.

Doug
Logged

TonyW
Full Member
***
Offline Offline

Posts: 120


« Reply #17 on: January 19, 2014, 01:55:09 PM »
ReplyReply

Hi Ray,

Good to see your comparisons but quite disappointed that you are not getting better from the Nikon 8000ED.  Just a few thoughts and trying to defend the Nikon  Smiley

Quote
I wonder how accurate this assessment of the V700 resolving power is. I've seen only one site claiming this figure of 2300 dpi.
I know what you mean about the accuracy of testing, but I tend to believe that the figures quoted are meaningful.  But ultimately it is the human eye that is counting the LP and unless the same eye each time and the individual is sufficiently rested then results can differ sometimes by noticeable amounts I have witnessed this first hand with resolution testing and evaluation in medical imaging.  The USAF 51 target is also based on LP in the horizontal and vertical only perhaps the inclusion of diagonal LP and circles would offer even more insight?   Most testing using this target seems to confirm that the resolution will differ from the horizontal to the vertical with horizontal lines displaying better than the vertical and the results are normally just averaged.

Quote
I was expecting to see noticeably more detail from the Nikon 8000 ED. Imagine my surprise when I sometimes saw less detail... I'm surprised it doesn't. In fact, the Epson V700 flatbed sometimes has a slight edge..
I agree and I am surprised too that you even at times saw less apparent detail. Still I have some theories why this may be the case.

Great image I loved my old Rolliecord so light and easy to use, no batteries, no autofocus, no inbuilt metering still I would miss some of the features of the new DSLR if I had to go back.

Just a few observations on your images and the comments you have made about the quality of the 8000ED vs the v700.  I have to stress that this is just a personal opinion (read as guessing!  Grin) trying to reason why the 8000ED did not come up to expectations mine as well as yours.

Although not exactly the best example if you go to http://zeux.zlakfoto.ch/scanvergleich/index.html and click on the v700 in the left panel and the 8000ED in the right panel you will see a comparison of similar images and the difference I would have expected in quality between the two scanners.

First thing that struck me is that the focus is out overall and of course you have experienced this with uneven sharpness.  One report http://forum.silverfast.com/post26877.html

The other thing reported using MF is the quality of the film holders being poor/makeweight and leading to unsharp/soft scans.  I understand that Nikon also offered a better quality film carrier which also included glass to enable the film to be held flat.  I think the Nikon item numbers  FH-869 GR and FH-869G

I also wonder about sharpening and how it is being applied in these products.  Even though we may think that we have turned all the bells and whistles off for shapening noise reduction etc to get a 'pure' scan there is no way I can think of to easily check what the manufacturer maybe applying via their software or indeed any third party software.  Although I suppose that scanning via different applications may reveal differences - but then how to interpret such?

« Last Edit: January 19, 2014, 02:04:51 PM by TonyW » Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8848


« Reply #18 on: January 19, 2014, 08:12:41 PM »
ReplyReply

Although not exactly the best example if you go to http://zeux.zlakfoto.ch/scanvergleich/index.html and click on the v700 in the left panel and the 8000ED in the right panel you will see a comparison of similar images and the difference I would have expected in quality between the two scanners.

Thanks for the link, Tony. That site offers an excellent method of comparing different scanners. But I see two glaring faults with this particular comparison of the V700 and the 8000 ED.

First, the V700 crop is significantly smaller than the 8000 ED crop. Why is that? It should be the other way round. A 6400 dpi scan is significantly larger than a 4000 dpi scan. One should either show both crops at their original size, or downsize the larger file to the same size as the smaller file, for comparison purposes. It makes no sense to show the originally-larger file at an even smaller size than the originally-smaller file.

To confuse matters, not only are the image elements in the V700 crop smaller, but the degree of cropping is greater so that one gets an initial impression is that the reason for the V700 image being smaller is that it has been cropped more. This appears very devious to me.  Wink

The second glaring fault is that the negatives being compared are not the same. The lowest of the 3 crops shows different people passing by in the street.

A third fault, although less glaring, is an appearance of noticeably brighter highlights in the V700 scan, which of course reduces the amount of visible texture on the highlighted surfaces.

Quote
The other thing reported using MF is the quality of the film holders being poor/makeweight and leading to unsharp/soft scans.  I understand that Nikon also offered a better quality film carrier which also included glass to enable the film to be held flat.  I think the Nikon item numbers  FH-869 GR and FH-869G

Any system that can ensure the film is held flat has to be an improvement, whatever the scanner. It so happens I have quite a few old 35mm Kodachrome slides that have been mounted between glass plates. I've just tried squeezing one into the Epson frame holder for slides, and it fits, but it's a tight squeeze. This should be a better test for the two scanners. I guess I should have used those glass-mounted slides in the first instance, but I was in MF scanning mode, or mood.  Wink

Quote
I also wonder about sharpening and how it is being applied in these products.  Even though we may think that we have turned all the bells and whistles off for shapening noise reduction etc to get a 'pure' scan there is no way I can think of to easily check what the manufacturer maybe applying via their software or indeed any third party software.  Although I suppose that scanning via different applications may reveal differences - but then how to interpret such?

I guess the bottom line is the best result one can get from the resources available. I rescanned this particular image, shown above, several times with the Nikon 8000 ED, not only using different focussing points but different amounts of sharpening, from the maximum amount of 500 at 4 pixels to 100 at 1 pixel. A sharpening amount of 200 at 2 pixels width seemed about right for the MF film I was scanning with Silverfast and the 8000 ED. and roughly equivalent to the 'high' sharpening setting in the Epson Scan software which gives one a choice of only low, medium, or high. Sharpening was further adjusted for each scan later in Photoshop using Smart Sharpen, with the object of getting maximum detail in both images with equal noise, and without applying noise reduction to either image.

I'm particularly pleased that the shadow detail in the Epson V700 scan is at least the equal of the 8000 ED scan. I think the DMax is around 4.00 for each scanner, or is it 4.2 for the Nikon?

So far, I'm very satisfied with this flatbed scanner. Of course, for even better results I guess 'wet mounting' would be the choice, but this sounds like a very time-consuming and fiddly process. Also, the Epson V750, designed to include wet mounting, is not available in Australia. There'd be no warranty if I imported it.

Regards,
Logged
artobest
Sr. Member
****
Offline Offline

Posts: 258


WWW
« Reply #19 on: January 20, 2014, 05:40:50 AM »
ReplyReply

I pay absolutely no attention to resolution tests of the V700 because there is no way of knowing how well the test scanner is set up. I have seen the difference a betterscanning.com film holder has made to scans from my V750, and I'm pretty sure that I'm getting equivalent resolution to that from my 3200 dpi Konica-Minolta film scanner. Furthermore, it can dig out detail from thin negs that is just unobtainable from the dedicated film scanner.

I have concluded that the resolution issue is a red herring, as camera shake, mirror slap, film grain etc are frequently the main brake on ultimate detail resolution. More important to me now is the quality of the scan, and I like the output from the Epson (with negative film, and after judicious sharpening and CA reduction) better than that from my film scanner.
Logged

Pages: [1] 2 3 4 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad