Changing from v700 to IQSmart2 for 8x10?

[Peter York]

An 8x10 and an Epson is a powerful combination. That said, my Cezanne smokes my Epson 4990.

My first impression when comparing the two was that the Epson scans using default software and medium sharpening were almost as good as the Cezanne. Now I know that I screwed up the focus in the shots I took, or the film sagged in the holder, or the wind was too strong, etc. No scanner can fix technical issues with the negative/positive, and I think these issues happen quite regularly to even the best large format photographers. So, given that you probably won't scan an 8x10 at resolutions greater than 2000 spi, the Epson is, as stated above, very good provided you solve scan height issues and newton ring issues.

Resolution is not the only factor at play, however. Tonality and color are superior to my eyes with the Cezanne. For myself this is most critical in B&W landscapes with subtle gradations in skies (which even my plastic yellow filter can muddy), and interestingly with shots taken w/ a soft-focus Verito where resolution is irrelevant.

When the Cezanne dies I will switch to digital.

[Pali K]

Pali, those are pretty awesome test targets; where did you get them?

Thank you Ari! I got it from Marin (username: LF_rookie_to_be) who was selling these on the forum last year. I believe he has the original glass negative and may be able to make you a contact copy if you are interested. I used it to calibrate my enlarger but it also comes in handy for tests like these.

Pali

[Pere Casals]

An 8x10 and an Epson is a powerful combination. That said, my Cezanne smokes my Epson 4990.

My first impression when comparing the two was that the Epson scans using default software and medium sharpening were almost as good as the Cezanne. Now I know that I screwed up the focus in the shots I took, or the film sagged in the holder, or the wind was too strong, etc. No scanner can fix technical issues with the negative/positive, and I think these issues happen quite regularly to even the best large format photographers. So, given that you probably won't scan an 8x10 at resolutions greater than 2000 spi, the Epson is, as stated above, very good provided you solve scan height issues and newton ring issues.

Resolution is not the only factor at play, however. Tonality and color are superior to my eyes with the Cezanne. For myself this is most critical in B&W landscapes with subtle gradations in skies (which even my plastic yellow filter can muddy), and interestingly with shots taken w/ a soft-focus Verito where resolution is irrelevant.

When the Cezanne dies I will switch to digital.

Hello Peter,

From the 4990 to the V750 there is quite a remarcable difference, being the 4990 very good.

BW tonality is just the same, as it is calibrated IT8.

You can prove it to yourself. Take an old scan of the 4990, then scan again the negative with Cezanne.

Place both scans in layers in PS, auto-align and substract both layers, move contrast and curve until the result of the substraction is a near black image. This will show you that it cannot be any difference in BW tonality that cannot be solved with the simple controls we always touch.

If there is a difference was because ICC profile handling.

[Pere Casals]

color are superior to my eyes with the Cezanne.

With color you can also make same process than I showed it to you with BW.

Just take a V750 and Cezanne scan, layers, autoaling, then first convert color to BW and adjust curves until the substraction match to black image.

Then remove the BW conversion and touch colour controls until subtraction is black.

Also there are software that makes a conversion LUT from color of a image to another.

Color profile is about flavours, for V750 users, for portraits I'd recommend the settings shown by Bohman in the 3rd post of this link:

http://photo.net/film-and-processing-forum/00Zdeb

Regards.

[Peter De Smidt]

Pere,

A few points:

An IT8 target and calibration is not the ne plus ultra of color and tonal reproduction, just as CRI is a very limited characterization of the quality of a light. Higher end scanners use Hutch targets, and no doubt even those have limitations. While profiles can be helpful, it's even better to have an excellent native response that doesn't need as much tonal re-mapping to get a good result. For example, take two high quality monitors, say an NEC and an Eizo. Use their respective profiling systems. Now compare the monitors. The color and tonality will still differ between the two monitors.

Second, using auto-align layers will inherently cause a loss of detail. As the geometry of an image is adjusted there will always be pixels from the original positioning that will not make it completely to a new pixel location in the new geometry. This will cause averaging, which inevitably involves a loss of information. That's why it's best to modify image geometry as few times as possible, and also why it pays to get it right in camera or in the scan.

Third, you are dead wrong about the dmax capability of an Epson. Using it as a densitometer doesn't prove a thing, as for all you know it's a very crappy densitometer. A scan of a calibrated step wedge is the correct way to discover in-use dmax.

[Peter York]

I disagree w/ Pere as no amount of post could achieve the same tonality and color with the Cezanne. This should not be surprising as we are comparing a unit with a plastic lens, fixed focus and a cheap stepper motor. Given the constraints the performance is really quite good, but in my opinion in a distinctly lower league than the high-end flatbeds.

I trust the OP can form their own opinion.

[Pere Casals]

Pere,

A few points:

An IT8 target and calibration is not the ne plus ultra of color and tonal reproduction, just as CRI is a very limited characterization of the quality of a light. Higher end scanners use Hutch targets, and no doubt even those have limitations. While profiles can be helpful, it's even better to have an excellent native response that doesn't need as much tonal re-mapping to get a good result. For example, take two high quality monitors, say an NEC and an Eizo. Use their respective profiling systems. Now compare the monitors. The color and tonality will still differ between the two monitors.

IT8 targets are very good because from barcode calibration sofware downloads a file from the target manufactured with the pecise calibration for each cell of that individual target, as from one to one there are differences.

Problem comes from that the R,G and B dyes on the pixels of the linear sensor are different from one scanner to the other, as dyes for each color are different from Portra to Ektar, or from Canon to Nikon DSLRs.

No dye is better to the other, but they have a different look, and then one can like Imacon, or Cezanne, or Epson.

At the end what counts is post process and image enhancing software, IMHO there is a lot of unjustified hype to the "colors" of scanners, because the important thing is your color profiles for each film. As Jose Villa say.

Second, using auto-align layers will inherently cause a loss of detail. As the geometry of an image is adjusted there will always be pixels from the original positioning that will not make it completely to a new pixel location in the new geometry. This will cause averaging, which inevitably involves a loss of information. That's why it's best to modify image geometry as few times as possible, and also why it pays to get it right in camera or in the scan.

The reference layer do not lose detail, the other can, but this test is not about detail, but about tonality, just to check that there are no difference.

Third, you are dead wrong about the dmax capability of an Epson. Using it as a densitometer doesn't prove a thing, as for all you know it's a very crappy densitometer. A scan of a calibrated step wedge is the correct way to discover in-use dmax.

IT8 As a densitometer is perfect, I've the particular calibrated densities of my IT8, I've cut the calibrated BW strip and I scan it at the same time that the negative, 16 bit, then I compare and interpolate. Perfect. I fact I use it as the calibrated step wedge yo say.

[Pere Casals]

I disagree w/ Pere as no amount of post could achieve the same tonality and color with the Cezanne. This should not be surprising as we are comparing a unit with a plastic lens, fixed focus and a cheap stepper motor. Given the constraints the performance is really quite good, but in my opinion in a distinctly lower league than the high-end flatbeds.

I trust the OP can form their own opinion.

About the resolution performance it's clear by the USAF 1952 tests,

http://archivehistory.jeksite.org/ch.../appendixc.htm


2900 dpi in the vert axis and 2300 in the Hor, this is a lot in true terms of performance. real photography have shake, subjects moving, lenses at not ideal stop etc

About colors, you know it's all software, if you don't agree please review color theory, a LUT converts anything to anything, even film is perfectly emulated by VSCO and other. See this : http://www.vision-color.com/impulz/

what counts for color is having a good profile for each film we use, as Jose Villa says.

[Pere Casals]

Pali, have you tried to sharpen + contrast both targets to best point and compared ???

[Peter York]

Its not all matrix transformations (software). The information in the original matrix that is to be transformed (hardware) is very, very important.