Help me please.
I Can use GMG Profiles in EFI Best Color RIP?
Help me please.
I Can use GMG Profiles in EFI Best Color RIP?
you can import any standard RGB icc profiles into EFI, but when using a CMYK profile a linearization will have to be made in EFI first, then the target will need to be printed through EFI-- after this the software used to create the ICC doesn't matter
The same is true of RGB profiles in EFI - linearization first (though obviously not an actual, true linearization), then profile target, then profile.
But the point is, you can use any profile in any program, that's not the problem. The problem with using a GMG profile in EFI is that it will be incorrect as EFI is not GMG, therefore they will have different print processing pipelines, different linearization process, different ink-limiting, etc.
So there is really no use in using a GMG profile in EFI - or any program other than GMG. The question remains, why would you even consider doing this sort of mismatched profiling/printing if you are going to the extent of using a RIP in the first place? Seems a complete waste of time and effort as the whole point of using a profile and indeed a RIP is for improved print quality. Using the wrong profile is a sure-fire way to be sure that won't happen.
I haven't been able to figure out what the linearization for RGB profiles in EFI actually DOES (my guess is... nothing) but I've been using this method and getting much better results than with RGB profiles created through efi.
One word of warning though-- for one very specific job printing vector graphics (from rasterized tiffs) I found that neutral tones that were very close to black went black when printing from EFI, but not when using the exact same profile through photoshop. I think there is a slight difference in the black point compensation being used when printing with relative colorimetric, but in general printing photographic images I have had no problem using this method, and the advantages of EFI's auto layout have been great
The RGB Profile Connector is supposed to be used for times when a profile is built using any app other than ColorManager in EFI, to connect linearization and profile. Quite obviously, the problem with your method (of not printing target through EFI) is that the profile is now completely inaccurate as you are not profiling a specific printer, paper, inks and printing pipeline as the last stage has been broken. If (for example) the target was printed through Ps, how could the profle possibly know what the printer will do or what corrections will be needed when printing through EFI? Just makes no sense to me at all. Why would someone bother going to the lengths of installing and learning to use the RIP, buying expensive spectro hardware, learning to use it and it's profiling software, going to all these lengths, usually with the hopes of getting higher quality print output only to cripple their printer by using a profile which does not describe the printer they are currently using?
This would seem to me to be exactly what's happening with the problem you're experiencing with dark, neutral tones going to black. Of course, EFI will lay down the ink and have different cut-off points than whatever you used to print the target, so it isn't really surprising to see this happening.
I'm sure you know this well, but if you create a profile (ordinary RGB, target printed through Ps, normal profiling app, no RIPs or anything fancy) and then change just one singal setting in the drivers, lets say for an Epson Pro x880/x900, clicking on the MicroWeave or altering the Density setting, that completely invalidates the profile because something has changed from the way things were when the profile was made. So think about how crappy things will be when you change the entire printing engine and printing application if just changing a single minor setting throws off the accuracy of the printed output?
I don't think your problem is being caused by a difference in the BPC between the 2 profiles/profile creation apps. BPC's whole purpose is to allieviate pretty much exactly the problems you're experiencing. Seems much more likely that the problem is being caused by using the invalid profile, the one which had it's target printed outside of the EFI pipeline, I think you said Ps was used.
I have noticed something I'm definitely not happy with in EFI though, and it is somewhat similar in that the darkest tones seem to be blocking up. I tried printing a B&W test image through EFI, something I'd not done before this morning actually - had only ever printed RGB or CMYK files through EFI in the past but I wanted to see what the EFI B&W output is like. It does seem quite neutral and not much metameric stuff going on, seems neutral under sunlight, fluro and ordinary household bulbs, which is a positive sign. But I was very disappointed with the overall look of the print, WAY too dark and lots of dark tones all being pushed to black, AFAICT. Not sure if I might have to fiddle with the brightness slider or try a different RI. Also, was using the canned EFI created Epson Premium Semi Gloss profile on that paper.
Come to think of it, I've not seen anything in the entire PDF or HTML manuals written about creating B&W profiles. Does anyone know if it's even possible in EFI or does one just use their standard RGB/CMYK profile and simply click on the 'Convert to greyscale' button to make B&W prints? Have to say, QTR seems heaps better for B&W output, especially once the printer has been both linearized and profiled in QTR.
Regarding the RGB lin, as far as I can tell, it is used to let EFI know if the print/driver/media settings chosen are in the ballpark, in that the lin target doesn't have huge wet pools of ink on it, doesn't completely block up from, say, anything higher than 60%, etc. I've not tried to purposely use a setting which would make this happen, but just thinking logically, there's really not a lot else the RGB lin could be accomplishing AFAIK.
As far as BW printing with EFI, there is no way to create a BW profile through it-- What I do is similar to what we've been talking about-- print the QTR target thru efi once you have a Lin, create the profile, then convert your images to that profile and print in efi w color management turned off-- the only difference is EFI won't let you import a B&W profile directly, so you have to convert each individual image.
I use a similar method when proofing BW images, but in that case I am converting to my QTR profile, then using a custom separation which emphasizes the black channel heavily and only includes CMY info in the darkest tones-- then I am sending this to EFI with color management off. This works great on the x880 series (as it does on press) but when I have used this method with our 9900 the black is very green, which is simply how the black inks look on that machine
In answer to the posters question no, you best bet is to recreate the profile either directly from the source data i.e. fingerprint chart if available. If not then get the source of the GMG profile to print out a suitable chart for example ECI2002 along with some test images for example Altona test suite and recreate the profile.