Rebuilding old Gretag profiles in i1 Profiler

[pfigen]

I would love to try rebuilding a few of my older ProfileMaker press profiles in i1Profiler to see if there's any improvement in the Perceptual conversions on a few specific types of images. This used to be super easy to do in ProfileMaker - just drop the old profile into PM, reset the parameters and re-generate the profile.

It seems to me that I've read that people have done this very think but I can't for the life of me figure out how to do this, and the X-Rite documentation hasn't exactly been very helpful. You certainly can't drop a PM profile anywhere in Profiler and extracting the data file doesn't seem to work either, as it doesn't seem to be recognizable in either of the formats that Measure Tool exports.

Maybe I'm just missing something obvious, but if someone can point me in the right direction, it would be greatly appreciated.

Thanks

[Scott Martin]

If you've got the measurement data in a compatible format, simply drag and drop it onto the 'Measurement' Printer Profiling Workflow icon. It will load the reference color patches and the measured patches and you can proceed from there. In some cases, you can even drag and drop a profile onto the Measurement icon and it will extract the reference and measurement data in the same fashion. Your results may vary as there are many variables.

As a side note, for the last 14 years I've kept the measurement data for all the printer profiles I've created. I didn't keep the profiles but I kept the much smaller measurement data. This way I can re-generate profiles as needed and keep a huge amount of measurements without massive storage capacity.  I miss Monaco Profiler's ~56K LAB only measurement files...

[pfigen]

Well, I've already tried every file format for the measurement files that were saved out of Measure Tool (and yes, I've always saved those data files as well) but nothing I've tried has worked. Dropping a ProfileMaker profile into i1Profiler only results in a "This is not an X-Rite profile and we can't modify it" message. Measure Tool only seems to be able to save out in two different text formats, neither of which is compatible. Oh well.

[Rhossydd]

I've just tried Scott's method mentioned above using drag and drop onto i1Profiler (1.4.2) and it works fine with measurements made with GMB Measure tool (5.0.1). i1Profiler complains there's no spectral data and needs a declaration of format, so declare it as a Gretagmacbeth format file (GMDI) and it imports fine.

If I remember correctly, this didn't work in some earlier versions of i1Profiler, so check you're running the latest version.

A potentially useful workflow, thanks to Scott for pointing it out.

[pfigen]

Well, I don't have the latest version, so I'll download it and try it out. A possible glimmer of hope...

[Scott Martin]

Well, I don't have the latest version, so I'll download it and try it out. A possible glimmer of hope...

Ahh, that's why it's not working for you. CGATS handling has seen several significant improvements in nearly every version. When you drag and drop your profiles onto the "Measurement"workflow icon in 1.4.x you should see a rotating arrow animation on the "Patch Set", "Test Chart" and "Measurement" workflow icons indicating that it's loading those data sets from your file.

Of course you could drag and drop your file onto the "Patch Set" icon and it would only load the reference data without the measured data but you want both, so dropping onto the "Measurement" is the way to go. Since measured data is worthless without the corresponding reference data it loads both. 

You will get a "This profile was not created by X-Rite and cannot be modified or optimized further" message but hit OK it will still load the reference and measurement data sets as indicated by the helpful animation. You don't really care about modifying or optimizing the profile - you're wanting to make a new profile anyway, and THAT profile i1P can modify or optimize further.

i1Profiler isn't the most intuitive app out there but I find it surprisingly flexible and the color science under the hood is unparalleled, IMO.

[Scott Martin]

I would love to try rebuilding a few of my older ProfileMaker press profiles in i1Profiler to see if there's any improvement in the Perceptual conversions on a few specific types of images.

Monaco Profiler and now i1Profiler (which represents the evolution of Monaco's engine) have always had better perceptual rendering characteristics than Gretag (ProfileMakerPro and i1Match). It's a crown jewel! Crank up that saturation to 45. :-] I gave up on ProfileMakerPro in 2005 and chuckled at those who kept using it for so long. I think you'll also see some improved paper white and black approach handling with all intents. There are lots of differences when you really dig into it. Your measurement mode can make a difference as well...

Your CMYK profiles based on an IT8 target will generate excellent profiles but RGB profiles should certainly be remade using new i1P targets that take advantage of it's unique engine. Don't use non-i1P patch sets (like BA's 918 patch set) in i1Profiler. Either way it's not a bad idea to print fresh i1P targets if possible to make sure you're characterizing your device in a recent state. Getting a full set of measurements (M0, M1 and M2) could also have some advantages which your legacy measurements won't have.

[pfigen]

Of course, I've made new all i1 profiles from all my RGB inkjet profiles. No problem there, but it's not always possible to do that with someone's press or proofing system, hence the original question. I'll download tomorrow when I get back to the studio and see what happens. Thanks for the help.

[Scott Martin]

Of course, I've made new all i1 profiles from all my RGB inkjet profiles. No problem there, but it's not always possible to do that with someone's press or proofing system, hence the original question.

Ah of course. That's the perfect situation to generate new profiles from existing measurement data. Hope it works well for you!

[pfigen]

Thanks for all your help Scott. It's working now. I can see it will take a few tries to get the parameters where I want/need them to be, but that won't take long. The X-Rite rep at Samy's response to asking for help in this was "why would you want to do that?" Oh well...

[Scott Martin]

Thanks for the follow-up - glad its working!