24x30 From a 4x5 Scanned On A 4870?

[neil poulsen]

I have a 4870 scanner. I finally got one.

Can this scanner, or it's 3200 nephew, support a 24x30 print from a scanned 4x5 with first rate print quality? What resolution should I select for such a print? I can do the arithmetic to obtain the "correct" resolution, but I'd like to have the viewpoint of someone who's actually tried this.

Ultimately, I'm wondering about purchasing a 7600. It'll be weeks before the 4000s become available. I'm getting impatient and am wondering if perhaps the 7600 will be a better printer, regardless of the 4000's schedule.

I want to do fine art, as well as professional architectural photography.

[sanking]

In my opinion, absolutely.

In fact, the Epson 4870 will support a print of this size (24"X30") from a 6X9 cm negative.

For 4X5 you can almost certainly get the required quality from the less expensive Epson 3200. but for 35mm and medium format scanning the 4970 is clearly superior to the 3200.

My work method is to always scan at the maximum dpi file storage conditions allow, and then sample down to 360 dpi for printing.

[Bruce Watson]

It depends entirely on how you define "first rate print quality."

I have a 2450. It does a pretty good job. I used it to scan 4x5 Tri-X for prints up to about 22x17.5 on my 7600. The sharpness is sufficient, and tonality is also pretty good. The Epson scanners are, I think, surprisingly good. I had to work the images in Photoshop a fair amount to get what I wanted, but I did get it. Still, the images didn't work very well at 22x28. They seemed sort of veiled somehow, and not as crisp.

I wanted to make bigger prints. I determined that I needed a better scanner, and ended up with a used drum scanner. Not for everybody, but it works for me. The scans I get from the drum just blow the flat bed's scans away, as you would expect from a piece of equipment that cost more than 10x the cost of the 2450. I don't have any problem with prints up to 40x50. Haven't gone any bigger than that (yet).

Part of the equation is your film and your subject matter. If you are doing Velvia trannies of highly detailed landscape, you'll be pushing the limits of what the CCDs can do and you might be somewhat disapointed. If you are doing negative film portraits, it might work out fine.

So... will the 4870 give you scans that let you make 24x30 prints with "first rate print quality?" It might. YMMV.

[Leonard Evens]

The largest print I've made so far is 13 x 17 because that is as far as my Epson 1280 printer will go. But I think I can extrapolate from that,and, by doing some additional arithmetic you may not have thought of , I may be able to add some more insight.

I just looked at a 13 x 17 print on my wall under a 2 X magnifier. The only loss of detail I can see is what I would expect to see anyway, however the print were made. Certainly the sharpest parts of the image still look very sharp. That suggests a 24 x 30 print made from the same scan would still look very good when viewed closeup.

The actual photographic resolution in the usual sense that the Epson 3200 can deliver is about 30 lp/mm. (The theoretical maximum for a scanner which scans at 3200 ppi is about 63 lp/mm.) From the reports I've seen, it appears that the Epson 4870 can do at most 10 to 20 percent better.

If you use the simple rule of dividing by 300, an appropriate ppi to send an inkjet printer, you will get one answer, but the photographic resolution number is a better guide to how large a print you can make. The usual rule is that a typical viewer can distinguish about 5 lp/mm at 10-12 inches. That means you ought to be able to enlarge a 4 x 5 negative by 5-6 times with little or no loss of fine detail, so a 24 x 30 print viewed closeup should look fine. Viewed at normal viewing distances, there should no problem even with a larger print.

However, some people claim that the 5 lp/mm print standard is too low and they would use 8 lp/mm or even higher. For people with such discerning vision, a 24 x 30 print made from a 4 x 5 scan would be deficient when viewed closeup.

See www.math.northwestern.edu/~len/photos/pages/e2450.html for an example of a scan of a 4 x 5 negative made with an Epson 2450, including a blow-up of a small section. The 3200 does better by about 10 percent, and the 4870 should do even better.

If you scan at 4800 ppi, your image files are going to be enormous. There is no point in keeping such a high resolution since the scanner probably can't deliver much better than about 30-35 lp/mm, which is what a perfect 1800 ppi scanner would give you. So if you rescale to 1800-2400 ppi, you will almost certainly not lose any relevant information.

[Darin Cozine]

At 1200 dpi, you get a 4800x6000 image. This will give you a 24x30 inch image at 200 dpi.

At 2400 dpi, you get a 9600x12000 image. This will give you a 24x30 inch image at 400 dpi.

Now I've scanned some sharp 4x5 slides and negs. Going from 1200 to 2400 dpi really doesnt give you any more detail that wasnt there before. However I noticed that small details were less pixelated on the higher res scan. I really dont see much use on going beyond 2400, so I havent upgraded my scanner. Also, the files are huge, I reccomend 1 gig memory if you are working with these.

[Ken Lee]

Just to augment what Lenoard has said: from 4 inches, 24 inches represents an enlargement of 6x. If your output is 300 dpi, you will need to scan at 1800 ppi, since 6 * 300 = 1800.

The question is, does the Epson really give you 1800 pixels per inch ? According to Leonard (a professor of mathematics in addition to everything else) the 3200 gives around 30 line pair per milimeter. Since there are around 25 milimeters in an inch, this means 30 * 25, or 750 line pairs per inch, or 1500 lines per inch. Add another 10%, and you get something like 1650.

This suggests that Epson, bless their hearts, are over-estimating their numbers by quite a bit. Apparently, the makers of flatbed scanners consistently overestimate, and the true numbers are usually around 60% of the claimed ones. Go figure.

Personally, I like to send 360 ppi to my printer. Since I like my images to be very sharp, I never print beyond 12 x 15. At that size, they are what I like to call "insanely sharp" - because once you add a little of Photoshop's "unsharp mask", the results often exceed what one could get... uh oh - I don't want to start a war of emails about sharpness and digital versus analog methodologies.

I suggest that before you purchase the scanner, you get one made on the scanner - Only you can decide if the quality is appropriate. As Leonard suggests, sharpness is also a function of viewing distance. In addition, certain images look sharper than others, due to the presence of edges, lighting, subject matter, etc. There is sharpness, and then there is the *impression* of sharpness.

[neil poulsen]

Thanks for all the responses. The obvious finally occurs to me. It would be easy to conduct a simple test.

Scan a 4x5 at about 1800 to 2400, crop to an appropriate size, and do a 6x or 7x enlargement to a 2200 printer. The 2200's have about the same print quality as a 7600, but to a smaller print. One could judge the quality by that print.

As far as flatbed scanners not living up to their claims, I'm wondering if one can compensate for some of that through averaging multiple scans. I have the Silverfast AI software, and it allows for this. I'd rather do a one-time wait for the scan, versus an extended wait after every command. I'll give this a try. Ideally, one should only need do an 1800 dpi scan to achieve what I want.

[Bruce Watson]

Neil,

All other things being equal, 1800 ppi might do what you want. But there is more to a scan than ppi numbers. Nothing is 100% effecient. All the component parts of the scanner (any scanner) loose information and/or add noise. Some scanners hold the negative in the focus plane better than others, yada, yada, yada.

By all means, do the tests. It's the only way to find out if the total system (scanner, photo editor, printer, inks, paper, etc.) will do what you want, and provide the quality level you need.

[Ken Lee]

To get the real 1800 resolution, you will probably need to scan at 3200 - Either that, or once you get past 1800, you may only see redundant pixels... Why not post your results ?

[Frank Muscroft]

The answers given here, particularly that by Leonard Evans, have prompted me to ask another question. Leonard says that the theoretical maximum resulotion of a 3200 ppi scanner is around 63 lp/mm and a practical limit of around 30 lp/mm. Okay I can see this as (3200/25.4)/2=62.99.... or 63 lp/mm. However, I have always considered pixels to be square or rectangular and so the maximum dimension is not the linear width or height but the diagonal. Consequently, I have always thought that the theoretical maximum resolution should be (3200/25.4)/(2x2^0.5)=44.548... lp/mm for square pixels instead of 63 lp/mm.

This then needs to be reduced by the Nyquist limit which basically asserts that unless the lines we are trying to resolve are exactly the same size and are aligned exactly with the pixels, we need around 10 pixels to resolve 7 lines. Therefore the 44.548... lines has to be multiplied by 0.7 for this modification which gives 44.548x0.7=31.18... lp/mm which is around the 30 lp/mm that Leonard suggests we can get from a 3200 ppi scanner. Of course we have not talked about the contrast ratio between the lines but that's another story.

Does this make sense to anyone else?

[Leonard Evens]

Frank,

Let me explain where I got the numbers. 3200 ppi is about 126 pixels per mm. Digital sampling theory says that if you collect 126 pixels per mm, then the most you can resolve in line pairs per mm is half of that, which is what is usually called the Nyquist limit. Roughly speaking, you need two pixels for each line pair. But the digital theory is actually much more complicated, and despite several attempts, I don't claim to have mastered it yet, but I am still working on it. It invovles some advanced mathematical techniques like Fourier analysis, which fortunately, like any well trained mathematician, I do understand. But there are many other subtleties, and the texts on the subject tend to not to be models of exposition. Personally, I find that naive attempts to understand it via simple intuitive models seldom work for me until I understand the actual theory. The problem is that there are numerous ways to construct simple intutitive models, and they give different answers, so one can argue interminably about which is right. When all is said and done, you have to go to real digital theory, confirmed by observation, and done by experts. Then you can get a better understanding by an appropriate intutitive model. I don't know just how the shape of the sampling elements and the fact that it is stepped in half steps and anything else may affect the results, but I am doubtful that one can work it out by such a naive model.

As an example, one point I didn't mention is aliasing. Any complex signal can in principle be decomposed into periodic signals of different frequencies. Frequencies above the Nyquist limits are aliased to lower frequencies and create artifacts which can degrade the image. So scanners sometimes have methods to filter out frequencies higher than the Nyquist limit. The Epson scanners actually use a staggered set of sensor elements, each half the nominal scanning frequency, and this supposedly reduces aliasing. But how it might modify the "true" sampling frequency, I don't know.

I hadn't seen a Nyquist limit referring to the shape of the sensors that you refer to. Can you give me a reference?

P.S. My name is spelled "EVENS".

[Frank Muscroft]

Leonard,

Sorry for misspelling your surname. Getting back to the topic, I must have explained poorly. Firstly, using the shape of the pixel to arrive at a reolution is purely intuitive on my part. To my knowledge CCD sensor specifications are usually given as X by Y pixels and consequently when calculating the number of pixels per mm the result is the number of pixels per mm in the direction of the major or minor axis. Photographs are made up of diagonal lines as well as horizontal and vertical lines and along the direction of the pixel diagonal the dimension of a square pixel is greater than the horizontal or vertical direction by 1.414.., therefore in this direction there are 1.414.. less pixels per mm and the reolution number is therefore effected by a factor of 1.414.... Even if the pixels are not square, there will still be less pixels per mm in the diagonal direction than in the horizontal or vertical direction. Thus, the resolution of example reduces to 48.5 lp/mm in the diagonal direction from 63 lp/mm in the horizontal or vertical direction.

My reference to the Nyquist limit was confusing, the reference to using 10 pixels to resolve 7 lines is actually in reference to aliasing at frequencies above the Nyquist limit. An explanation of this is given in the Schneider white paper "Optics for Digital Photograpy" available on the Schneider web site http://www.schneideroptics.com/info/white_papers/optics_for_digital_photography.pdf .

[byard pidgeon]

A basic rule for scanning to large format printers is: one pixel in the file for each pixel in the output. You can sometimes get away with less, you can use more, but I've seen surprisingly "photographic" results, even on lower resolution older printers, using this rule. Printer manufacturers, like scanner makers, tend to blatantly lie (oh, alright...exaggerate) the resolution capabilities of their devices, because that's what the market is attuned to. The optimum printing resolution is usually a fraction of the advertised resolution...like maybe a fifth or sixth...but as far as I know, only a few large format printer manufacturers will tell you what it is.