Has anyone tried Silverfast AI Studio IT8 w/ printer calibration?

[dgberg]

Have been looking at upgrading my Silverfast software for my Epson v750 pro. Found a few items that interest me HDRi and especially the AI Studio IT8 with printer calibration and icc profile wizard. Now I know its not an Xrite i1 but has anyone seen a review on this. They show a short video of scanning the i8 targets then making a new icc profile. Its about $400 and with discount I think its about $300. The question is how good of a scan of I8 targets do you get with this scanner vs. with an i1 spectro and its software?  Would like to see several profiles make by each to compare.

PS. Just found a good review in NADIR Magazine but no comparisons.

[Mark D Segal]

Dan, first could you please post a link to the NADIR magazine review, if this is on the internet.

Second, you may wish to see my recent article on Silverfast on this website: Old-New, where I discuss a number of substantial features in the latest vesion of the program, including one-click profiling, HDR and the infra-red channel. However, in this article I did not compare profile making via the scanner versus using a spectro.

[terrywyse]

The question is how good of a scan of I8 targets do you get with this scanner vs. with an i1 spectro and its software?  Would like to see several profiles make by each to compare.

Short answer:
A scanner is NOT a spectrophotometer and probably doesn't even make that good of a colorimeter which is essentially how it would be used in this application. I've never heard good things about those packages that claim to be able to use a scanner for measuring printer targets. At the very least, you'd need to first profile (characterize) the scanner itself and then measure the print target. I can *imagine* taking the scan of the print target, assigning the scanner profile to it and then attempting to extract L*a*b* measurement data from that....but I seriously doubt that it would work all that well.

Considering the $300 price tag, the Eye-One Basic package isn't all that much more expensive (about $1K or so) but I'll bet it would make profiles that would blow away the "scanner" profiles. Would the Eye-One profiles be 3-4x better? I'd say yes.

Just my 2 cents,
Terry

[dgberg]

Here is the article from Nadir.

[Mark D Segal]

Dan,

Thanks very much for the article.

Terry,

Yes I agree with you completely if we are talking printer profiles. Not clear to me whether Dan was referring to profiling the printer or profiling the scanner, because the SF package allows you to do both. If he was referring to profiling the scanner, of course one must generate the reading of the target from the scanner, and then one can explore stuff like the quality of alternative targets and their reference files for doing this.

[dgberg]

Heres the video. Sorry I wasn't clear but I was referring to making individual paper icc profiles just like in this video.

[Mark D Segal]

OK, in that case, what Terry said.

[Czornyj]

Heres the video. Sorry I wasn't clear but I was referring to making individual paper icc profiles just like in this video.

This is my most favorite quote from "Real World Color Management"  :

There are profiling packages that use a flatbed scanner to measure the printed targets, but in the final analysis, scanner-based profllers are like the talking horse: what's amazing isn't how well they do it, bur that they do it at all

[probep]

Short answer:
A scanner is NOT a spectrophotometer and probably doesn't even make that good of a colorimeter which is essentially how it would be used in this application. I've never heard good things about those packages that claim to be able to use a scanner for measuring printer targets. At the very least, you'd need to first profile (characterize) the scanner itself and then measure the print target. I can *imagine* taking the scan of the print target, assigning the scanner profile to it and then attempting to extract L*a*b* measurement data from that....but I seriously doubt that it would work all that well.

Considering the $300 price tag, the Eye-One Basic package isn't all that much more expensive (about $1K or so) but I'll bet it would make profiles that would blow away the "scanner" profiles. Would the Eye-One profiles be 3-4x better? I'd say yes.

Just my 2 cents,
Terry

Yes. Or at least it's better to buy ColorMunki spectrophotometer ($350-$380) than to use a scanner (with software for $300).

[BobFisher]

I did try the paper profiling option in Silverfast.  I asked for a refund.  It's not good.

[Mark D Segal]

I did try the paper profiling option in Silverfast.  I asked for a refund.  It's not good.

I wouldn't have recommended buying a program like SilverFast for the sole purpose of profiling a printer in the first place. This is not a primary purpose of SilverFast or its strong point. It's a convenience to users who are buying the program mainly for scanning and want a near zero-cost option for profiling their printers. That said, no matter what the application, using a scanner to read a printer profiling target is going to be less satisfactory than using a spectrophotometer. As is often the case, one gets what one pays for, and you are not really paying much for printer profiling when you buy SilverFast - you are mainly paying for a scanning application.

[BobFisher]

I didn't say I bought it solely for that purpose.  I already had a license for the scanning software.  I was upgrading my license for a couple other features after buying my new scanner and decided to add on the print profiling feature.  It was the print profiling feature I got a refund on, not the entire package.  Normally I wouldn't even bother for the obvious reasons but since it was Lasersoft I decided to give it a go.  Figured maybe they'd found a way to build a better mousetrap.  They haven't.  Simple.

[Mark D Segal]

I didn't say I bought it solely for that purpose.

That's true - you didn't say what you bought it for, but what you did is now clarified and makes sense. I think the weak part of the mousetrap is the use of a scanner for reading the target.

[dgberg]

Thanks Bob and Mark,
I too have the software and was looking to upgrade for a better scanning solution when I saw the profiling part of this program. Was more curious to how good the scanner handled the IT8 scans and now I know.
Thanks