comparison between Epson V750, Imacon and drumscan

<p>Hi!<br>

Here's a comparison between Epson V750, Imacon and drumscan of 400 ASA Kodak TMAX 6x7 black and white negative.<br>

I 'm going to print a bunch of four foot enlargements from 6x7 black and white.<br>

While my Epson V750 is plenty good enough to deliver scans (At 4.600 PPI) from 100 ASA Tmax to do 4 foot prints, the 400 ASA film, scanned on the Epson, gave serious grain aliasing, and I wanted the grain to look like grain.<br>

I decided to get an Imacon scan (At 2.900 PPI) and a drumscan (At 5.000 PPI) made.<br>

Here are the results.<br>

The Imaconscan shows more shadow detail. The grain is more pronounced.<br>

The drumscan shows less shadow detail, and is a bit more fuzzy.<br>

I can make both Imaconscan and drumscan look almost identical with some photoshopping (sharpening, levels)<br>

The Imaconscan cost me 30 Euro's. The drumscan cost me 140 Euro's.<br>

The people from the Imacon scanner said in their sales pitch that drumscanning resolution is not necessary. It seems they 're right, unless I don' t look at things right. (I'm no expert on scanning)<br>

My conclusion is that I'm not drumscanning any more.<br>

Do you agree?<br>

Dirk.</p>

<p> </p><div></div>

<p>Epson detail:</p>

<p>detail of Epson V750:</p><div></div>

<p>Imacon scan:</p><div></div>

<p>detail of Imacon scan:</p>

<p>Imacon scan detail:</p><div></div>

<p>drum scan:</p><div></div>

<p>drumscan detail:</p><div></div>

<p>All sczns as I got them, no sharpening, levels, etc. TIFS grayscale.</p>

<p>Grain aliasing Epson in the sky.</p><div></div>

<p>Imacon, same image piece</p><div></div>

<p>Drumscan</p><div></div>

<p>For me the Imacon shows a visibly better result tham my V700. At up to 12" sq I can ignore the small difference. Above that I'll buy in an Imacon scan. I agree about the better shadow detail. </p>

<p>I have to say though that I've been paying a great deal less for an Imacon scan than you do (in the UK) . Sadly my scanning resource no longer provides that service but I did do some preliminary work to source an alternative if I need it and I feel confident I can get a price below £15 from a MF slide or neg. Making a 1m print from an apparently sharp 67 b&w via an Imacon scan a few months ago was possible but not easy. </p>

<p>Drumscans- well my strictly non-scientific impressions are that there's a lot of variation in scan quality. The best prints from my colour work I've ever had have been made from drum scans though I can't necessarily tell what's down to the printing and whats based on the scan. Again I think you are paying a lot of money for a drumscan, though if you're getting that locally rather than taking the risk and time to mail them , that carries a benefit. Personally and after the experience mentioned above, I'd expect to drum scan on those rare occasions where I need a print over say 30" or 36", and the one thing about your conclusions I don't really buy is the implication that Imacon scans are better than drum scans. I thinlk Imacon scans are very good and that they are better than <em>some</em> drum scans.</p>

<p>Of course there's a missing step in your process here - that of scanning with a low-end film scanner such as a Coolscan 9000. I used to own one of these till I found I could buy superior Imacon scans cheaply enough to simply eliminate the Nikons from my thinking. </p>

<p>Dirk, Thanks for posting your scan comparisons. I have had 6x7 Velvia transparencies scanned on an Imacon, which produced excellent 20x25 inch prints. I also have 6x7 b&w negatives in line for scanning for prints. Your post caused me to read more about grain aliasing, and for discussion, I will reference this tutorial by Norman Koren. The tutorial is about as technical as I wish to delve into. At the bottom of the page is a link to Neat Image, a program that Mr. Koren endorses with enthusiasm. I have had some interest in drum scans, but it would be on a very selective basis. I curtailed my thoughts of owning a scanner when the Coolscan 9000 went out of production.</p>

<p>http://www.normankoren.com/Tutorials/MTF8.html</p>

<p>http://www.neatimage.com/</p>

<p>The imacon scan here is way over sharpened for a raw scan. The grain is very ugly and I think it won't print up looking natural. I think the best scan you have here is from the drum scanner. But I imagine the Imacon could be adjusted to give a more natural looking scan.</p>

<blockquote>

<p>The imacon scan here is...over sharpened</p>

</blockquote>

<p>I appears that may be the case. Which Imacon is this? On my 848 you have to set the sharpening to -128 in Flexcolor to get true unsharpened results. Setting it to "0" still applies a considerable amount of sharpening.</p>

<p> </p>

<p>Dirk again.<br>

The drum scan DOES have lots of shadow detail.<br>

I did some sliding with the grey (middle) tones in Photoshop levels and it opened all up.<br>

Got an Email from the drum scanner and he says he looked at my neg with a 75 x loupe and the scanned grain is exactly as it is in the neg, and that the Imacon scan is way oversharpened.<br>

So, now I'd even more like to get contributions from people who know about scanning.<br>

I should get an optical print of this neg at the same enlargement; Then I'll know.<br>

Dirk.</p><div></div>

<p>Looking at your last crop I see your image was actually shot on HP5+. HP5 has larger, and less well defined, grain than TMax 400. So seeing that I think the drum scan really is really about as good as you will get from the film (as far as resolution). To tell if it's really captured the full tonal range of the negative open a levels layer and hold the option/alt key down while moving the end point sliders. Ideally you should see the edges turn black before any large areas in the image. That tells you that you metered the scene to capture full details. If the edges and areas of the image are already black with the shadow slide all the way to the left (0) then the scanner operator clipped the scan and there is likely missing shadow detail. If the edges and large ares of the image go to black at the same time it says you underexposed and there is no detail recorded in that area of the film. You may as well print these areas as black or they will just look muddy. Same idea for the highlights, but you can't really see overexposure this way since you don't have a good reference like the film edges. Even specular highlights shouldn't be pure white on a good raw scan (but you probably want to push them there in the final image). Raw scans like this must be done in 16/48 bits if you want to have room to adjust the negative.</p>

<p>Check all your scans to see if they clipped the ranges at either end. If they did, talk with the operator and see if you can get them to give you a full range and unsharpened scan. I've never used an Imacon, but I have a Scanmate 5000 drum scanner. With it I can get scans which image the grain and produce prints that look very close to my darkroom prints of the same image. Looking at your image from the drum I suspect you won't see much difference in a wet print from a well made inkjet. I bet the Imacon scan can be redone with no sharpening and an expanded range and it will be able to give you a better print.</p>

<p>Keep in mind that a good raw scan looks muddy and not quite sharp at the pixel level (unless scanned low res). But with a few levels and curves layers you can make it sing.</p>

<p><strong>Of course there's a missing step in your process here - that of scanning with a low-end film scanner such as a Coolscan 9000. I used to own one of these till I found I could buy superior Imacon scans cheaply enough to simply eliminate the Nikons from my thinking.</strong><br>

<strong><br /></strong>The real test of whether or not any of the mentioned scans are worth it would probably not be from seeing the type of enlargement posted, but from evaluating the finished print at a reasonable viewing distance.<br>

The other question is how much better either a drum scan or an Imacon scan are than a Nikon scan. I have a Nikon and have not compared to the other two, but from posted images I've seen, anything above a Nikon is diminishing returns, especially considering the cost. It seems to me that you're barking up the wrong tree--if you really need that kind of microscopic quality for giant prints, you're using the wrong film and medium to begin with.</p>

<p>Scott. Diminishing returns is not relevent when you need a scan of a certain quality to make a print of a certain size. If you wanted to make a print of say 20" x 16" then it is clear to me that assuming competence on behalf of the persons making the scans, there isn't a big benefit from going beyond a Nikon 9000. However, diminishing returns notwithstanding, if you want to make a print 4 feet across, as the OP seems to want, you'll be better off with a an Imacon (and I'm casting doubt over whether even that is really good enough) or a drum scan. </p>

<p>Note also that if I can buy an Imacon scan cheaply enough, it might not be worthwhile to use a Nikon 9000 under any circumstances. Personally I was quite happy to liquidate my investment in the 9000 and holders because I had a deal to buy cleaned Imacon scans for under $10 a time. </p>

<p>I owned a Nikon 9000 for a couple of years and I've had several hundred medium format frames scanned on both Imacons and Drum scanners. </p>

<blockquote>

<p>The imacon scan here is way over sharpened for a raw scan.</p>

</blockquote>

<p>That's very possible. The OP should check on the settings used on the Imacon as the Zero setting for sharpening is not one that turns off all sharpening! Don't tell them that, just ask "<em>what was the setting for sharpness in FlexColor</em>"<br>

As someone who's made a lot of scans on two drum scanners, a number of Imacon's and others, I'd say that for the money, the Imacon's are really great. But they do not compare to the potential quality of a good PMT scanner, driven with good software by someone who knows how to use it. The drum scan should have been oil/gel mounted. It has a wider (and real) dynamic range. When ink hits paper, the Imacon should deliver. If you're making 2nd generation digital dupes or enlargements going of to a LVT, you probably want that drum scan. I seriously hope in this day and age no one has to do this anymore. But in the old days, that's they way you had to output Photoshop manipulated documents for repro. <br>

<br>

Dirk, be real interesting if you printed the three to what you'd feel is max size from the smallest file, same crop you've already used.</p>

<p>I own an Imacon 848 and have compared it to drumscans. The problem with your "tests" is that the Imacon scan is automatically sharpened during the scan. If you know what you are doing you can control that and sharpen to your taste in Photoshop after the scan or even during it (by making the proper adjustments.) I scan to the imacon raw format (.fff) and then output using the imacon sharpening software with a small amount of sharpness. I then do the rest of the sharpening, which is determined by what size print I want to make, in Photoshop. I use a wet scanning process with the Imacon created by Scan Science (wet mounting dot com). I compared this to drum scans and this process is sharper while at the same time LESS grainy than a drum scan, which they owner told me it would be. This process is best with the 848 series and previous models, not with the newer imacons due to the nature of the new lights. As far as what you showed, the Imacon is the best. No comparison as you need that shadow detail in the face to make the photo work. Invest money in Imacon scans but talk to the lab about how to sharpen it or else ask for raw scans and use the free flexcolor software to output the scans to tifs, yourself. If you want the best, this is the only way to do it, unless your lab will work with your taste to give you what you want and need each time.</p>