What's the best deal in scanner-land with no compromises in quality?

[Juergen Sattler]

Frank is 100% correct. You need to learn the basics first. Scanning and printing does not happen on the push of a button.

[PaulSchneider]

Hey guys, thank you for your advice!

I'm more collection knowledge than immediately pulling the trigger on a huge financial decision.

I was just wondering how much does a ICG 380 cost and how much do refurbished ones cost such as the 370 HS? I would like to know how far away this stuff is from reality!

Regards

Paul

[Lenny Eiger]

Aztek is great and all but their scanners are slightly limited in D-range. From my experience the Tango/Primescan and Eversmart Supreme both have a slight edge in shadow performance.
Good luck,
Ed

Ed, this is an unsupportable claim. I've tested this, the Tango can't really compare, and the Eversmart is a CCD. Even tho' its a good one, it can't compare with a PMT. That's simply physical realities with different technology.

I think the Premier is the top drum scanner. I say this with the caveat that I haven't even seen the top ICG (the 380?), which might work as well. Both have a 3 micron capability, which forces everything in the instrument to a higher standard.

If you are close, you can come visit and I'll give you a tour, show you where to look for the differences.

Lenny

[8x10 user]

Lenny I have used 6 high end scanners in the last two years which I have done many many comparison tests with. Two of the devices were Azteks; they are good machines and all but the D-max is a bit limited. I know this from first hand experience. Even Aztek has 3.88 down as the D-max on their drum scanner. The tests scans on this forum show the difference in shadow performance, it is clearly there for anyone to see. Of course there is a noise eliminator option in DPL that uses a noise histogram to deal with noise / grain issues but that is a more of a "lossy" way of doing it.

I do think something is up on the Eversmart scan and that it should be sharper (should be equal to Supreme II)... There are a lot of factors that can effect the quality of an Eversmart. None of the scans comparison page were done by me or with my equipment.

BTW, you have one of the earliest premiers right? Even told me that there were many revisions to the line and that the newest ones have a better board. It would cost you 15k to upgrade to the newer one.

As for the 3 microns. What films do you use it for? 10 microns and above is ideal for the majority of film types. Plus there is a lot more to a scanner then just the aperture of the scanning head.

Finally I want to add that mounting 8x10s on the premier would be substantially more difficult then a drum scanner or flatbed with a larger drum/bed.

Regarding CCDS if you want I can get into to some advantages of CCDs from a scientific standpoint.

[8x10 user]

I have not used the ICG 380 or Screen 8060P mark II but it would be interesting to see how they compared. So far I find advantages and disadvantages to every scanner that I get to use. Maybe one scanner is not the best for all uses?

[Lenny Eiger]

Lenny I have used 6 high end scanners in the last two years which I have done many many comparison tests with. Two of the devices were Azteks; they are good machines and all but the D-max is a bit limited. I know this from first hand experience. Even Aztek has 3.88 down as the D-max on their drum scanner. The tests scans on this forum show the difference in shadow performance, it is clearly there for anyone to see. Of course there is a noise eliminator option in DPL that uses a noise histogram to deal with noise / grain issues but that is a more of a "lossy" way of doing it.

I do think something is up on the Eversmart scan and that it should be sharper (should be equal to Supreme II)... There are a lot of factors that can effect the quality of an Eversmart. None of the scans comparison page were done by me or with my equipment.

BTW, you have one of the earliest premiers right? Even told me that there were many revisions to the line and that the newest ones have a better board. It would cost you 15k to upgrade to the newer one.

As for the 3 microns. What films do you use it for? 10 microns and above is ideal for the majority of film types. Plus there is a lot more to a scanner then just the aperture of the scanning head.

Finally I want to add that mounting 8x10s on the premier would be substantially more difficult then a drum scanner or flatbed with a larger drum/bed.

Regarding CCDS if you want I can get into to some advantages of CCDs from a scientific standpoint.

[Lenny Eiger]

Lenny I have used 6 high end scanners in the last two years which I have done many many comparison tests with. Two of the devices were Azteks; they are good machines and all but the D-max is a bit limited. I know this from first hand experience. Even Aztek has 3.88 down as the D-max on their drum scanner. The tests scans on this forum show the difference in shadow performance, it is clearly there for anyone to see.

The scans on this forum are no way to judge shadow detail. They are severely downsized, compressed, sharpened and all kinds of changed. I think it is a great exercise to send the same piece of film around - and I remain appreciative of all the work Leigh put into it - but I remain unimpressed with the ability to make distinctions between the scanners.

I did the Premier example and it looks way better over here. There is a ton more shadow detail, the colors aren't washed out, etc. The shadow noise is ridiculous.

Aztek made a conscious decision many years ago to report the actual DMax instead of the theoretical. On their comparison, the Tango's listed at 3.0, for example.

BTW, you have one of the earliest premiers right? Even told me that there were many revisions to the line and that the newest ones have a better board. It would cost you 15k to upgrade to the newer one.

It isn't one of the older ones - its got a new board in it. The revisions were mostly to combine the boards into a single board so that there were less connector errors.

As for the 3 microns. What films do you use it for? 10 microns and above is ideal for the majority of film types. Plus there is a lot more to a scanner then just the aperture of the scanning head.

There aren't any films that I use it for. It's about quality. It's the same with lenses. If you use a lot more glass, you have to make sure every part of it is perfect. If you have a car that can go over 150 mph, then you have to have tires that can handle it. For a scanner to be able to scan at 3 microns, it has to have a mechanism that is sensitive enough to move the bed over only 3 microns at a time. That requires an extra amount of quality, which filters itself through the entire machine.

Finally I want to add that mounting 8x10s on the premier would be substantially more difficult then a drum scanner or flatbed with a larger drum/bed.

That ridiculous. A Premier is the same as any other drum scanner, even easier than some. You mount on a mounting station. I do 8x10's all day long. I shoot 8x10. I don't have bubbles or any other trouble. It doesn't sound like you have run one personally as you suggest. Using someone at A&I, or even at Aztek, to run a scan for you isn't the same as testing things yourself.

Lenny

[8x10 user]

The scans on this forum are no way to judge shadow detail. They are severely downsized, compressed, sharpened and all kinds of changed. I think it is a great exercise to send the same piece of film around - and I remain appreciative of all the work Leigh put into it - but I remain unimpressed with the ability to make distinctions between the scanners.

I did the Premier example and it looks way better over here. There is a ton more shadow detail, the colors aren't washed out, etc. The shadow noise is ridiculous.

Lenny

I agree, it probably didn't help that the aperture was set too small, IMO.

Aztek made a conscious decision many years ago to report the actual DMax instead of the theoretical. On their comparison, the Tango's listed at 3.0, for example.
Lenny

Do you think the tangos D-max is really 3.0? I think that would be a claim that you would have trouble supporting.

There aren't any films that I use it for. It's about quality. It's the same with lenses. If you use a lot more glass, you have to make sure every part of it is perfect. If you have a car that can go over 150 mph, then you have to have tires that can handle it. For a scanner to be able to scan at 3 microns, it has to have a mechanism that is sensitive enough to move the bed over only 3 microns at a time. That requires an extra amount of quality, which filters itself through the entire machine.

Lenny

I see what you mean. Your talking about precision. I can't commit too much on how the different drum scanners compare there. I would trust the Aztek scanner more then a Howtek in that regard, if it is well maintained.

It isn't one of the older ones - its got a new board in it. The revisions were mostly to combine the boards into a single board so that there were less connector errors.Lenny

Which board do you got? I think there were like 6 or 7 in the premier series and a few more in the DPL8000.

That ridiculous. A Premier is the same as any other drum scanner, even easier than some. You mount on a mounting station. I do 8x10's all day long. I shoot 8x10. I don't have bubbles or any other trouble. It doesn't sound like you have run one personally as you suggest. Using someone at A&I, or even at Aztek, to run a scan for you isn't the same as testing things yourself. Lenny

I used one of the earlier DPL scanners many times in the past. It's close to being the same scanner as the premier with an identical D-max. I found the 8x10 difficult but do-able on the smaller drum. There is not much wiggle room on there compared to some other drum scanners. But at least you don't have to mask it off like you should with a flatbed. BTW, Any interest in buying my old drums? I still got a couple lying around for the DPL/premier and one for a 4000 or 4500 (I can't tell).

As I said before there are many things that I like about the Aztek scanner. I could see one being used besides a tango or eversmart for when you want to do really light stuff (should be an advantage of DPL due to bypassing the log amps) or for films that have a fine enough grain for a 3 or 6 micron aperture capture. Although I would worry loosing that little bit of extra shadow detail that the Aztek cant seem to reach. I would say that the Aztek is a good drum scanner with some unique innovations, it just doesn't need to be hyped up with a bunch of marketing crap about apertures and race cars... Anyone with operational experience would see through for the reasons I already mentioned (not practical for most film types).

[8x10 user]

WARNING!

Theoretical ideas and opinions on scanning technology to follow. I could know more about this stuff so please feel free to correct me if I got something wrong.

About the technology:

PMT (drum scanners) - The photons are converted to electricity by a photocathode and then are amplified in a vacuum tube to increase the voltage. A drum scanner uses PMT’s to create an image by measuring the values obtain as it spins the film around the sensor in one axis (referred to as the fast scan), the sensor is moved perpendicular to the rotation to create the second axis (referred to as the slow scan).

The light signal is amplified without added noise from convectional thermal sources; although heat should be able to affect the value before and after it is amplified. Other forms of noise should affect the data at varying degrees throughout the process. In additional to thermal noise can think of four other types of noise that could be an issue.

1. Burst or Popcorn noise, that’s the stuff you hear on the radio my guess is that it comes from space radiation as well as earth originating singles that bounce around our atmosphere?

2. Shot noise: Light and electrons travel in the form of single “packets” (electricity or photons) the slower the rate or total amount that is being sampled the less stable the reading will be. This is called “shot” noise in physics and can be a factor to other “packets” besides then light and electrons. Imagine the varying rate of salt particles from a shaker.

3. Flicker noise: Is increased as the single goes up? If so must be some type of electronic feedback?

4. Vibration:

Could be more of a factor with some scanners over others.

Possible Advantages of PMT technology for image scanners:

• Easy to adjust the sampling area separately from the scanning resolution
• Drum scanners are resistant to long distant flare issues, however more local flare can be an issue if something is not right in the scanner.
• Low noise for a single pass scanner
• Strong performance in the shadows and highlights.
• Handles contrast well.
• High signal gain without added thermal noise

Possible Disadvantages of PMT technology:

• Extremely small amounts of light are being sampled at the dark areas, especially at small apertures…
• If the signal is weak from the PMT then dark current in the scanner can become more of an issue.
• Very high gain; would amplify any noise at or before the photodector…
• There might be some natural “shot noise” from the lower frequency of photon’s being detected at low apertures and high densities (light behaves more sporadically when there is less of it).
• Single pass only – no multisampling
• Some of them seem to produce a little bit less detail in low contrast deep shadow regains or detail is overshadowed by noise (IMO/E).
• Can be tricky to get the photons to hit it just right so the efficiency of the photon detection is lower then other technologies
• Fixed ISO / amplification

CCD scanners – light rays bend through a lens and strike a photoelectric material connected to a capacitor that stores the energy up for a period of time. At the end of the cycle the charge from each capacitor travels separately down the array until the end where it can be amplified if needed. The length of the cycle can be changed to adjust the ISO.

Possible Advantages of CCD technology::

• All pixels illuminated at same time (For example 8000 RGB samples at once with some scanners)
• Receives more light= less “shot” noise, less amplification required, stronger signal’s help to overshadow white noise in the electronics.
• Thermal noise at the CCD can be reduced to one photon per hour by cooling the CCD head. (Supreme I and II have thermoelectric cooled CCD heads but they are not as cold as the ones with one photon per hour noise).
• Iso can be altered by adjusting the cycle frequency or charge amplification.
• Images can be scanned multiple times then averaged for more pure values... For a greater effect the ISO can be changed between passes.
• High percentage of photons detected versus film or PMT (more efficiency) roughly 70% of light that strikes the CCD is detected
• Flatbed scanners can scan flat objects
• Good shadow performance when the right ISO is used in addition to multisampling.

Possible Disadvantages of CCD technology:

• Flare can be an issue if there are flaws in the optical path. Generally the optically path for CCD scanners has more surfaces that can get out of alignment or collect dust.
• For the most part you cannot change aperture or sampling size of scanner separately from the sampling frequency (resolution).
• More noise when done at high ISO or with single pass.
• Dust on the CCD head will create major problems.
• Slightly lower D-Min?

Conclusion:

Overall I think both technologies are great. Obviously there are differences from a high-end CCD and a lower end one. With high-end scanners the condition of the device is very important. My opinion is that both technologies offer benefits for certain applications, and that there isn’t a single ultimate scanner for all uses without any compromises what-so-ever. Most of what I have said here is theoretically based but it seems to back up what my own personal experiences have been. Some of the things that I said I am less sure about so I added a question mark. Also I think there might be more factors in PMT technology that might give some drum scanners better shadow performance then others. I need to make up a special ultra dark scanning target so that the density ranges of scanners can be more accurately tested. I do honestly feel that there is too much marketing of scanners for ones own self-interest on this forum. I suppose that I am guilty of it sometimes so I will have to try to cut back on it but I am not the only one. I just hope people realize there is some bias out there… It is true thou that it is hard to speck from your own perspective without being at least little bit bias.

[Matus Kalisky]

8x10 user,

I guess that with "shot noise" you simply mean statistical fluctuation.