“You often feel tired, not because you've done too much, but because you've done too little of what sparks a light in you.”
― Alexander Den Heijer, Nothing You Don't Already Know
Unfortunately, I consider all of these comparisons to be flawed. The clarkvision scans are questionable scans of strange images. They were done in 2001, on a Tango. That means that they likely didn't know how to set the aperture correctly, which was figured out much later. Was auto sharpening applied? Was the scanner in good condition - we don't know. It's also a scanner that is, half as sharp as a Premier. Some operators really know how to get everything out of a Tango, and this difference can end up being very small. However, how many times have you seen a comparison and the flatbed scanner is outlined exactly (or the digital camera) but they just indicate "drum scanner" for the other side? They don't bother to tell you. This means they put all their effort into getting the best scan they could out of the flatbed, but what did they do on the drum - oh, they gave it to someone else. Who, what kind of scanner, when was it serviced, what settings were used, etc.? In the case of a recent comparison by Luminous we discovered that the scanner used was severely out of whack.
Tim's pages wouldn't load properly today (it could be me) and the largeformat comparison, despite a lot of good effort, is not very good at all. It doesn't distinguish properly between anything. There is too much post-processing, and it isn't applied evenly. The one from DPUG that Ed Bray pointed to is a good example from Castor, but it only shows the drum scan...
After my initial tests with the 750, I was so disappointed with it that I simply got rid of it. The only thing I can't do is scan something that can't be bent (I have friends that can), and for everything else, why would I start with something so blurry as an Epson - if I had another choice?
I generally don't publish because its impossible to know what one is looking at, especially when its on the web. On occasion I get one of my clients to publish their results, recently, Greg O'Hanlon put up his very fine image of the Vietnam War Memorial, printed at 8 feet, from a 6x17 chrome.
http://www.largeformatphotography.in...hlight=vietnam
Was that chrome perfect? No. Smooth and grainless? No.
Every piece of film is different. How does one know what kind of detail is possible without looking at the actual film? If I have never shot anything that looked remotely like the clarkvision image, which I haven't, how could that be a guide for me? How much smoothing of the grain is required for the image? Is there a lot of sky or a lot of water? How was the film developed, exposure for color? Not to mention how well is the drum scanner tuned. These are all questions that would have to be answered in a real comparison.I think it is an impossible task, certainly without the funding to do it properly.
I think there is only one choice for people that are interested in knowing what's possible (or not). One should take a piece of film, that they shot, preferably an image they love and know well, and get it scanned one both scanner types and look at the difference, print it out with their best PhotoShop techniques, their printer tuned, etc. Then there is something that actually makes sense to the actual situation...
I am sorry for this long-winded post. I probably could have said this in three sentences, but I worked very late last night and this is what happens to me...
Lenny
EigerStudios
Museum Quality Drum Scanning and Printing
Lenny, I agree. Running a meaningful test is no small matter!
“You often feel tired, not because you've done too much, but because you've done too little of what sparks a light in you.”
― Alexander Den Heijer, Nothing You Don't Already Know
My post was to the initiator who asked drum or flatbed. My answer would always be drum. However the initiator is a student, on budget, his output is an ink jet, appeared to be leaning away from drum scans and may be interested in a cool little project of his own. Not everyone has the money to buy expensive equipment and not everyone has the money to have the highest quality scans done. For that reason a creative workaround was suggested. It may not work for him but it's an option he's explored. He may learn allot about the process along the way which means it's not absurd.
Lenny, on the subject of aperture selection, is there an advantage to setting the scanner aperture selection to say 25 microns and scanning at 4000 dpi or does this just result in a larger file size with smoother grain but no real improvement in image detail? I guess the it could also depend completely on the image and film in question as well. I could experiment with some negatives, just haven't yet. It'd be interesting to hear a professional's take on it.
Thanks,
Ryan
The goal of aperture setting is to match the size of grain clumps in the image. If you set the size too small, you sample the grain 1 1/2 - 2 times. This results in something called grain anti-aliasing. It is usually slightly off kilter and the effect is to add much more graininess to the image. File size is not involved.
If you set the aperture too high you undersample and the effect is blurriness, or lack of sharpness. There are usually two apertures that will work best, if you have 3 micron steps, one if you have 6. What's interesting is that one aperture will appear slightly sharper and the other smoother. However, when you look at the smoother one, you can see that very small details are articulated. It's a matter of how much smoothing you want.
It's sharp to begin with, so a small amount of sharpening can be applied after the scan (radius of .2). Another factor that can toss a monkey wrench into the process is that many LCD screens add a little sharpness to their displays. On many scanners you can select a small section, perhaps an 1/8 of an inch or so, and blow it up to full size. If you do this and look at different aperture settings, you can see the result of these effects...
I scan most b&w film and color transparencies at 10 or 13 microns. Color negatives go at 16 or 19, unless it's Ektar, which is down at 13. This varies with development and exposure, if color... I look at most film, blowing it up to see the smallest section I can, and compare the different apertures. With a little experience it becomes pretty easy to choose...
Be aware that the settings I am referencing are for the Howtek and Aztek series of scanners and the specific numbers might vary considerably.
I hope this answers your question,
Lenny
EigerStudios
Museum Quality Drum Scanning and Printing
Take a look at Bruce Watson's comment in an old thread.
http://www.largeformatphotography.in...ed-to-enlarger
The title is a bit misleading because eventually he gets to talking about the relationship between aperture and sharpness.
Sandy
For discussion and information about carbon transfer please visit the carbon group at groups.io
[url]https://groups.io/g/carbon
Quoted from the Bruce Watson post linked to above:
And we haven't talked about that big variable -- film grain or dye clouds (depending on the film, and for simplicity I'm going to call it all grain). The detail in the film is carried in the grain, and in the spaces between grain. We see it as the way it attenuates light that is shown through it. That sad thing is, the size of the grain is a variable, and the amount of variance is quite large.
What this means to scanning is just this. A scanner uses a fixed aperture to measure film that has a more or less random grain size (at least it's random to the scanner). If your aperture is 6 microns and the spot you are measuring is "thin" then you might be measuring several small grains at once. If on the other hand you are measuring a spot that is "dense" then you might be looking at a grain that is huge compared to the spot size.
And this is why it doesn't matter nearly as much as you'd think. And this is why you can't see the fluctuations in optical resolution of the scanner. It's much smaller than other variations that are going on at the same time.
Scanning is nothing more and nothing less than an approximation of the original film. In the case of drum scanning, a very, very good approximation, but an approximation none-the-less. Remember that, and you'll be fine.
Thomas
One of the things that Bruce alludes to here deserves a little more discussion as the question comes up often. There is a huge difference between optical resolution and the number of pixels a scanner and its software can generate. Just because a scanner can generate 11,000 or 12,000 pixels doesn't mean it can optically resolve that much. The number of pixels it generates related to the stepper motor it uses.
Those of us older folks remember when those large schoolroom clocks changed from rotating around the full circle in a smooth fashion to stepping through each second tick on the clock face. That's a stepper motor. They apparently went insane with that technology. Aztek's Premier, the scanner I use, can do 18,000 steps for a single rotation of the drum. Tango's and ICG's can apparently do 11,000 or 12,000 (and maybe some others). That's the limit of how many samples the scanner can take, and how many pixels it can generate. This is very different from the aperture its using to take the sample.
Of course, how many line pairs you can resolve depends on the entire system. It has to do with the accuracy of all of the elements, as well as how close you can get to matching as much of the grain as possible. Generally speaking, when you see a scanner that can resolve to as little as 6 microns, you are looking at at max of about 4000-4200 ppi of optical resolution. (You may want to note that 4000 x 6 microns = 24,000 microns, which is close to an inch at 25,400. 8 x 3 is also 24.) When you see scanners that are set up to handle apertures at 3 microns you have a scanner that can theoretically resolve to 8,000. However, the ICG, as Bruce notes, resolves to around 6,000 and the Premier is about 7,400. Maybe its the ceramic bearings, or the precise angle of the paper disc with the holes in it, who knows. Scanners are a mix of amazing technology and low tech. Either one is great.
Aztek made the decision to limit its ppi results to 8,000 - vs delivering 18,000 pixels - because that's the maximum of what they could optically resolve. They figured they would rather give you 8,000 great pixels vs 18,000 ok, or "interpolated" ones. You can always interpolate in PhotoShop if you need to.
I hope this helps,
Lenny
EigerStudios
Museum Quality Drum Scanning and Printing
The size of the aperture does limit the actual resolution in pixels per inch. Here are approximate figures.
3 microns = 8000 ppi
6 microns = 4000 ppi
13 microns = 2000 ppi
25 microns = 1000 ppi
These are the maximum possible resolutions in ppi for the aperture. If you scan at 25 microns to minimize grain the maximum effective resolution will be only about 1000 ppi/dpi, regardless of how you set the resolution. So you can scan with an aperture of 25 microns at 4000 dpi, but anything beyond 1000 dpi is scanner software interpolation.
Sandy
For discussion and information about carbon transfer please visit the carbon group at groups.io
[url]https://groups.io/g/carbon
Posting Permissions
Reply With Quote
Bookmarks