Scanning 4x5 transparencies with Epson 3200

[michael_joubert]

I'm on a tight budget and can't afford the highend Imacon and Nikon

scanners so I am considering purchasing Epsons new 3200 flatbed

scanner. If you have expereince with this I'd really appreciate your

feedback. I've never scanned 4x5 before so I'm not aware of any

problems one may encounter.

Thanks, Michael.

[leonard_evens]

I have a 4 x 5 example at

 

math.northwestern.edu/~len/photos/pages/e2450.html

 

The Epson 3200 should be a bit better. You should be happy with the results from this scanner with 4 x 5 film. I have yet to see a negative comment from someone who has been using it for 4 x 5.

 

I scan mainly negative film in color and bw. With transparency film, it is possible you may encounter some dynamic range problems. Supposedly the Epson 2450 has a dmax of 3.3 and the 3200 a slightly higher 3.4. My experience with transparency film is that the maximum densities I need to deal with are less than that but not by much. Conventional wisdom is that it is better to underexpose transparency film slightly than to overexpose it, so if you are careful about exposure you may not have any problems. But you may also want to consider using color negative film. I've seldom encountered densities higher than 2.0 with negative film. Personally, I don't see any real advantages to using transparency film if you are going to scan.

 

You should also keep in mind that although you probably will scan at 3200 ppi, you will want to rescale to a lower resolution. Otherwise your files will be too large for convenient processing in a photoeditor. I have a 1.4 Ghz Athlon and 1.5 Gb of ram, but I still rescale my 2400 ppi files to 1800 ppi. As far as I can see there is no loss of detail in so doing. You can probably rescale to 1600 or possibly even 1200 ppi. But I think you still will do better than you would using a scanner whose hardware tops out at the lower scanning resolution.

 

From what others have said, it seems that the Epson 3200 probably provides better than 30 lp/mm in terms of real photographic resolution. That should allow up to 6 X enlargments which can be viewed reasonably close up and still hold up, but perhaps not as well as conventional prints made with an enlarger or from drum scans.

 

At the price, the Epson 3200 should be an excellent choice for 4 x 5.

[ken_lee11]

If you are going to shoot black & white negatives, then you might find it helpful to produce negatives that are a little softer than otherwise. Scanners have a hard time punching through dense areas of the film, which correspond to highlights in the final image. A development scheme of N-1 will be helpful if you want to retain detail in those areas.

 

In the digital process, one can always increase contrast, ad infinitum. However, if your image contrast exceeds the capacity of the scanner, then you may have to resort to 2-pass scanning, and that's a nuisance.

 

If you've never scanned 4x5 before, then you're in for a treat.

[vijay_ramakrishnan1]

I second Leonard's comment about having adequate computer hardware to do the job. Think carefully about the print size you are going to want, and whether you want to work with 48-bit images or are willing to work with 24-bit images. I decided I wanted to be able to print my 4x5's at 24x30 inches and work in 48-bit. 512 Megs of RAM was not enough. I needed to upgrade my computer because 512 Megs was the maximum memory it could support. You need to figure this into your budget as well.

 

Because I want 24x30 prints, this means sending files out for printing. So now I need to add another set of items to my budget to enable me to set up a color-managed workflow.

[morey_kitzman]

If you are shooting 4 x 5 it does not make much sense to process it through a low end scanner with dubious dmax. I have my slides scanned at The Slideprinter where they have a new rate of $30.00 flat, whether you have a 35, medium, panoramic or large format. Best price I have seen and it is one of those $200,000. scanners. So why compromise? I have pushed my Microtek 1800f and it does not do the slides any justice. I imagine the Epson's result are as marginal. Epson and Microtek are great for proofing and the ocassional inkjet, but no where close to high end scanners. Hope this helps.

[jonathan_ratzlaff]

I don't think you will have much trouble scanning negative film with this scanner, as the dynamic range of both colour and B&W negative film is considerably lower than that of transparency film. You will find that negative scans are considerably flatter and lower in contrast than you may be used to. I have scanned 4x5 negatives on a perfection 1200 and did not have any trouble getting detail out of the dense parts of the negatives, however the resolution left a bit to be desired.

[ralph_barker]

I've had my 3200 for about a week, and have been impressed so far. While there may be occasions that you'll want to have a specific neg or transparency drum-scanned for hi-res output, I think you'll find the 3200 is adequate for most purposes that can be handled on a reasonably configured desktop PC.

[ralph_barker]

<p>Here's something more concrete to work with, Michael. The first is a reduced size, adjusted image of the full B&W negative scan. The second is a full-sized, unadjusted, unsharpened clip from the shoreline at the left end of the light-colored trees. The full scan was done to create an 11x14 print at 300 PPI, producing an image that is 3300 x 4186 pixels and 39.5MB in size. </p>

<center>

<img src="http://www.rbarkerphoto.com/misc/Travel/YVmirrorlake72-600mfbw.jpg">

</center>

<p>The raw clip:</p>

<center>

<img src="http://www.rbarkerphoto.com/misc/Travel/YVmirrorlake72-EP3200raw-clip.jpg">

</center>

[leonard_evens]

Great picture, Ralph. And you've done an excellent job of preserving tonal values in the scan. It can't all be the scanner.

[kelly_flanigan1]

My Epson 2450 photo scanner was used to scan a reference negative; of an old lens test that resolves 80 lines/mm . When the negative is scanned with a Epson 2450 scanner; the resolution is 31 lines/mm along the direction of travel; and 28 lines/mm 90 degrees from direction of travel. A perfect non obtainable scanner at 2400 pixels/inch should approach 2400/(25.4 * 2)= 47 lines/mm . My epson scanner's resolution is 66% ; and 60 % of the theoretical number. <BR><BR>The same reference negative scanned with my Canon FS4000US film scanner reads 63 lines/mm. A perfect 4000 pixel/inch scanner should approach 4000/(25.4*2)= about 79 lines/mm. My Canon FS4000US film scanner resolution is 80% of the theoretical number. <BR><BR>The Reference negative is of a high contrast USAF test target; and is Panatomic-X film; and is about 30 years old.<BR><BR>Typical 4x5" negatives are not of ultra contrast perfect test targets; but of real world grey scale images. <BR><BR>In many clients 4x5 color tranies; the trany is the limit; and not the flatbed scanner. For a good sharp 4x5; the Epson 2450 flatbed is clearly the limit; and a drum scan will yield a much sharper image. For mild enlargements; the print fom the scan is not much better than the plain jane flatbed @2400 ppi. I have nver had a problem with the Dmax issue; but I do use prescan curves alot to get a better scan.. The 2400 ppi 4x5 color scans I do are about 290 Megs in RGB color; if this file size scares you off; then a 4000 ppi scan of a 4x5 should be a real problem.

[kelly_flanigan1]

For mild enlargements; the print fom the <b>drum</b> scan is not much better than the plain jane flatbed @2400 ppi

[kelly_flanigan1]

I usually fill up one CD per 4x5 image; and save several different scan resolutions; and scans done differently. The volume of the CD's generated is about 5 to 10 times the volume of the 4x5 trany and sleeve.

[michael_joubert]

Thanks SO much for the info and taking the time to share your experience with me. I truly appreciate it.

 

I am little confused on some issues though. I was under the impression that transparency film had a narrower dynamic range than negative film. Do I have it the wrong way around?

 

I'm going to be shooting in color exclusively and will be sending the film to a lab to be developed or processed. This is one of the reasons I'd prefer to stick with transparency because I need to maintain color accuracy as much as possible and was under the impression that there's more risk of color shifts when having a negative developed at a lab than transparency being processed. What are your thoughts on this?

 

I would have to say that color accuracy takes precedence over dynamic range for the work that I'll be doing since most of the prints will be large and probably viewed from about six feet away. If there isn't a tremendous amount of detail in the shadows I think it'll be ok. How do you find the accuracy of the Monaco profiles for those of you that are using the software?

[tim_atherton2]

"I am little confused on some issues though. I was under the impression that transparency film had a narrower dynamic range than negative film. Do I have it the wrong way around?"

 

 

Unfortunate confusion of terms - use of the same or similar term to mean something very different. The "D" in DMax when used with regard to scanning isn't referring to dynamic range but density. Here is a definition I keep around to help me make sense of it...:

 

"First dRange... The "d" means density (and has nothing to do with dynamic

range), and when measuring density that film records (as in contains valid

image data), there is a minimum value (dMin), and a maximum value (dMax).

The range between these two density points is the density "range", or

dRange.

 

Positive (transparency) film has a clear base...so the dMin for it is going to be quite

low...as compared to negative film, which has a cloudy base...so the dMin

for negative film will be quite a bit higher than positive film. Both films

will pretty much have the same max density, black is still black, whether

it's positive or negative film.

 

Well, let's say the dMax (blackest part) of both films can be measured at

3.6...and the positive films dMin is .2; and the negative films dMin is

.8... That gives a dRange for the positive film of 3.6 - .2 or 3.4, and for

the positive film 3.6 - .8 or 2.8.

 

It's purely the film base "offset" that creates the difference in density

range.

 

(and so you don't need the same density range to get the info out of a negative film as you do from a transparency film)

 

For another discussion let's say that the

same range of image tonality could be recorded on either film...just that

negative film would have the range compressed, film density wise that is,

compared to positive film. Also, the "dynamic range" of the film is not the

same as the density range...and the term dynamic range is often misused when

talking about density range."

 

I'll add that there is no set of standard for manufacturers to measure the Density Max on their scanners - so one manufacturers 3.9 might not be the same as anothers 3.9...

[john_hennessy1]

One issue not yet mentioned is the Epson software. You may find it will not do the job so you�ll have to use the scanner with other�s software. It comes with Silverfast light (or whatever it�s called) and the light version of that too is of limited use. VueScan full version is $79 and Silverfast is about $200 to $250�I cannot remember. VueScan works with many scanners for the same $79 but you�ll need another copy of Silverfast should you get another scanner. So you should factor in those potential costs too.

 

The basic issue with the software it comes with (either one) is that it will not use a profile other than its own. As the light source ages, beginning as soon as you turn it on, the built-in profile will get out of kilter. And of course, unlike higher priced scanners, it includes no program to make a new profile. The better versions of VueScan and Silverfast do.

[ardea]

I read with interest John's post regarding Vuescan and Silverfast profiling of the scanner. I have a Epson 3200 that I puchased last week for scanning 4x5 chromes but have not used it for that purpose yet. I did use it with Monoco EZ Color to profile my printer for some new paper though. I also scan 35mm stuff with a Nikon Super Coolscan 4000ED. Nikons software will not let me embed a IT-8 profile. Sort of what I believe John is saying about the Epson 3200.

Assuming that one is taking the scanned data and running it into a photo manipulation program like Photo Shop I wonder how important it is to have a profiled scanner. One can have a most difficult time if your monitor and printer are not profiled but I see the scanner as a digital camera. When we take a photo with a digital camera or my Shen Hao the availible light varies in intensity and temperature which is like saying, that the light source of the scanner changes with age. I think this can be compensated for in PS.

At least this is the way I see it, but if I'm way off base maybe John could explain further. I'm a long way from being a whizz at this stuff. I hope others will jump in and clarify this.

This is a good thread and I would like to see more on scanning with the 3200. I do have one foot in a wet (real)darkroom and the other in a digital darkroom......We live in interesting times, which at times is not a curse.

Regards, Richard

[tim_atherton2]

I find the new IT8 profiling options on Vuescan a real boon - I happened to have both 35mm and 4x5 targets.

 

Both my current scanners are now profiled. While they weren't bad before (my monitor is failry well calibrated via Optical) it just makes that bit of difference now (my 35mm scanner always seemed to scan a bit on the dark side for one thing). Now, 9 times out of 10 using Vuescan, transparency scans are so close colourwise, that I don't hqave to spned much time getting that side of it in the ballpark - it's already there.

 

But as I already had the targets and wath "grandfathered in" to Vuescan, it didn't cost me anything...

 

tim

[michael_joubert]

Thanks John for your post. I just checked Epsons website and I found out that the 3200 �pro� version ($600) comes bundled with SilverFast Ai which is the full version of SilverFast as well as MonacoEZ. The cheaper option of the 3200 ($400) comes bundled with the entry level SilverFast SE which cannot be upgraded to the full version. I just wanted bring this to the attention of those following this string.

 

From what I�ve read on the internet SilverFast is touted the best scanning software available by a respected independent research and review group that specialises in digital capturing devices, and inkjet technology for professional use, http://www.flatbed-scanner-review.org/FLAAR_Digital_Imaging_Center/Digital_Imaging_Technology.html

 

The following (very nice) review of the 3200 has a comparison between the Epson driver and the SilverFast driver for large format film and you can see what a difference the driver alone can make. http://www.photo-i.co.uk/Reviews/interactive/Scanners/Epson_3200/page_11.htm

 

With MonacoEZ and SilverFast Ai one should be able to establish a pretty good color calibrated workflow however you�ll need to buy the OPTIX sensor to be able to calibrate your monitor, an extra $225 which is essential to a color workflow. In fact, calibrating your monitor is the 1st place to start. Personally I�m considering a NEC/Mitsubishi or a Sony Artisan monitor which are fairly spendy but there�s a reason for that and that is that it will calibrate the monitor hardware unlike MonacoEZ and other popular low-end color management software which will profile the output of your video card and reduce its gamut in the process. (There�s a string somewhere on Photo.net that discusses this).

 

No doubt in my mind that the �pro� bundle equips you with some of the best tools available on the market that will push the scanners potential to the max at a VERY aforable price. Well worth the extra $200 in my opinion. (MonacoEZ alone costs $300.)

[john_hennessy1]

Richard, I think serious workers profile everything and often. In the same way one uses a warming filter on the camera to compensate for early or late in the day light, my goal is to land in photoshop with the image closest to the scene as possible without the scanner trying overcome shortcomings in the exposure of the film and PS trying to make head or tale of what it gets from the scanner. Then the fun starts as I decide to faithfully reproduce the scene or radically change it. I may have already decided at the scene what to do, or at least try in PS. If the film is too blue and the scanner tends to cyan in the high values and red in the low, and the monitor is wacked and who knows what the printer will make of what it gets, etc. then I handicapping myself at every stage. In other words I want all the variables to be in PS under my control (with a hope and a prayer) and all other pieces to just do what they�re told.

[tim_atherton2]

I've had a couple of scanners come with Silverfast and I've never liked it. I've certainly never found it worth the extra charged for it. Although not as slick looking, I've always found Vuescan does a much better job. Even with what is now just the basic version. Profiling with IT8 targets in the newer pro version has just been an added benefit. I've also found Pantone Colorvision Optical to do a very good job of monitor calibaration

[michael_joubert]

Tim, I'm interested in what you've experiened with Vuescan and Optical. Obviously it makes sense to go for the lower price if it does just as good a job as something more expensive. Just asking but, is there any way could do scan of some film with both the Vuescan and SilverFast driver of the same piece of film and post a portion of it to this string with some comments? I know it's asking a lot so if it's not possible that fine.

[kelly_flanigan1]

Here is the scan of the same Konica Auto S2 @F5.6 Panatomic-X negative; which reads 80 line pairs/mm very clearly on the negative. This negative was shot over 30 years ago; and was the sharpest aperture for this camera; and directly on axis. I use this negative frame of a USAF test; to test enlarger lenses of mine; and scanners. <BR><BR>As mentioned above; "the negative is scanned with a Epson 2450 scanner; the resolution is 31 lines/mm along the direction of travel; and 28 lines/mm 90 degrees from direction of travel. A perfect non obtainable scanner at 2400 pixels/inch should approach 2400/(25.4 * 2)= 47 lines/mm . My epson scanner's resolution is 66% ; and 60 % of the theoretical number."<BR><img src="http://www.photo.net/photodb/image-display?photo_id=1471278&size=lg"><BR><BR>Here is the same negative scanned with a 4000 ppi film scanner; <b>It resolves more than double the resolution; thus the Epson flatbeds are leaving alot of detail out; that a drum scan can stil gather.</b><BR>"reference negative scanned with my Canon FS4000US film scanner reads 63 lines/mm. A perfect 4000 pixel/inch scanner should approach 4000/(25.4*2)= about 79 lines/mm. My Canon FS4000US film scanner resolution is 80% of the theoretical number"<BR><BR><img src="http://www.photo.net/photodb/image-display?photo_id=1471262&size=lg"><BR><BR>The photo.net image is not as sharp as the original tif file; which clearly resloves the numbers quoted above; using a very conservative "easy to resolve" criteria. <BR><BR>The photos are highly cropped sections of a lens test ran at a 1:50 standard magnification.

[kelly_flanigan1]

Full 35mm frame; of 35mm Konica Auto S2 negative; scanned for above cropped images of center USAF test target. The lens has a slight barrel distortion.<img src="http://www.photo.net/photodb/image-display?photo_id=1471363&size=lg">