Epson paper settings

[Cornelius]

Hi guys

Does anyone knows what kind of information is stored in the Epson driver paper settings?
Quite a while ago I chatted with the Epson support in germany and they told me, the paper settings basically just hold the information about thickness, platen gab and with the smaller printers (like 3880 etc.) also which paper feed is used.

But what about the paper settings on the plotters?
I tried some test prints with similar settings and my own ICC profiles on my Epson 9890 with Tecco PFR295 FineArt Rag paper.

Basic settings:
1440dpi
BiDi with Super MicroWeave
Paperthickness: 3
Platen Gap: Auto
Ink density: standard

These are the results
Enhanced matte (what Tecco recommends) --> all good
Archival matte --> fine bandings in solid black regions
UltraSmooth FineArt Paper -> all good, except the colors look much warmer then they should
Textured FineArt Paper -> similar to USMP

For me it seems, even if I use OSX ColorSync and my ICC profile, the paper setting from the Epson driver still has an influence in colors.
Also I'm not sure why I get the faint banding in solid black regions with Archival matte setting but not with the others?

Looking forward for your explanations. I don't understand it.. :-(

[Mark D Segal]

The ink lay down is in part determined by the Media Type setting the profile maker selects when making the profile for any given printer/paper combination. Tecco probably made their profile for that paper using the Enhanced Matte Media Type setting, therefore they correctly recommend that you use that Media Type setting when you use that profile. Outcomes cannot be assured when you mismatch the Media Type between what the profile was made for and how you are using it.

[Cornelius]

Hi Mark

Thanks for your feedback. I found this information also in the meantime. That's the explanation of the color shifts if have seen.
But what is still not clear for me, is the banding I can optain, depending on the paper I choose in the driver.

[Mark D Segal]

Wrong media type may be a factor. Another factor could be if you are not sending 16-bit data to the printer, or in particular if you are sending 8 bit data in a wide colour space such as ProPhoto; for wide colour spaces you want to be in a 16-bit workflow.

[Rand47]

Wrong media type may be a factor. Another factor could be if you are not sending 16-bit data to the printer, or in particular if you are sending 8 bit data in a wide colour space such as ProPhoto; for wide colour spaces you want to be in a 16-bit workflow.

Mark,

Sorry to invade this thread, but what you say about 16 bit is alarming to someone who uses Windows.  As I understand it, Windows can only send 8 bit data as there is no 16 bit option.  Since LR uses a form of ProPhotoRGB as its native colorspace, anyone printing their edited raw files out of LR in Windows has a problem???

Rand

[digitaldog]

Sorry to invade this thread, but what you say about 16 bit is alarming to someone who uses Windows.  As I understand it, Windows can only send 8 bit data as there is no 16 bit option.  Since LR uses a form of ProPhotoRGB as its native colorspace, anyone printing their edited raw files out of LR in Windows has a problem???

Correct but don't be alarmed! I don't think there's any difference or benefit sending 8-bits per color of good data versus high bit to the Epson driver. Tests I did on the Mac showed zero difference under a very good loupe. Conversations about this years ago with the product manager at Epson (at the Epson Print Academy) backed this up but he pointed out, someday, that could be different.
FWIW, by the time the driver gets the data in it's output color space, it's NOT ProPhoto RGB but (at least going through the PCS) Lab.

[Mark D Segal]

Yes, I've heard that too. But my advice applies to the whole workflow (I wasn't clear about that); and regardless, I assume there is some reason why both LR and PS have a 16-bit option one can select in the Print menu.

[digitaldog]

Yes, I've heard that too.

Test it, look at it (measure it). I see no difference either way.

But my advice applies to the whole workflow (I wasn't clear about that); and regardless, I assume there is some reason why both LR and PS have a 16-bit option one can select in the Print menu.

The fact is, anyone on Windows sending high bit data to the Epson driver, despite these settings isn't getting high bit data to the driver. And again, on the Mac, I don't think there's any difference, not with the Epson's I tested.

[Mark D Segal]

Andrew, I'm not disagreeing with your observations. But I was also talking about the pre-print workflow settings for which I expect we agree that if using a wide colour space one should also edit in 16-bit where the choice exists. And again, I ask why in LR and PS there is an option to "Send 16 bit data" for the Print stage if it makes no perceptible difference to results?

[digitaldog]

Andrew, I'm not disagreeing with your observations. But I was also talking about the pre-print workflow settings for which I expect we agree that if using a wide colour space one should also edit in 16-bit where the choice exists. And again, I ask why in LR and PS there is an option to "Send 16 bit data" for the Print stage if it makes no perceptible difference to results?

Again, by the time the driver gets the data, ProPhoto RGB has long left the building. We edit in high bit to reduce rounding errors upon the data from the edits. Doesn't matter the working space or it's gamut. YES, the rounding errors are colorimetrically larger in a wider gamut space but that's not why one edits in high bit. You should do the same with sRGB.

[Mark D Segal]

Yes I know what happens by the time the data gets to the print stage; but before that stage is the 16-bit mode not also giving us far more levels which help with tonal transitions? So yes I agree, good to do in any working space, but all the more so in very wide ones. I believe I've read you yourself making this point in the past.  And why do LR and PS give us the option to send 16-bit data through the Print pipeline?

[digitaldog]

Yes I know what happens by the time the data gets to the print stage; but before that stage is the 16-bit mode not also giving us far more levels which help with tonal transitions? So yes I agree, good to do in any working space, but all the more so in very wide ones. I believe I've read you yourself making this point in the past.  And why do LR and PS give us the option to send 16-bit data through the Print pipeline?

Why? Because there is no reason to spend the proceeding to truncate this data! A driver may or may not accept it. The bigger question you need to yourself is, does it matter? Thus far, I say no.

[Mark D Segal]

Why? Because there is no reason to spend the proceeding to truncate this data!

Parsing this sentence, do you mean to say that if working in 16 bit up to that point, there is no point reducing the data to 8 bit for printing? If so, that's fine.

[digitaldog]

Parsing this sentence, do you mean to say that if working in 16 bit up to that point, there is no point reducing the data to 8 bit for printing? If so, that's fine.

Yes!

[Rand47]

Mark, Andrew...

Thank you for this conversation.  Very helpful.

Rand

[Sbarroso]

Smooth gradation of a single color (e.g. clear skies), especially in BW, may show the difference between 8 and 16 bits. I've seen it with my pixma pro 1 (bw, not yet color) . In 8 bits subtle banding appears because one can appreciate the difference between one tone x, and the next one (x+1). In 16 bits, this difference is too small to see.

Best
Santiago