Until now I’ve been using the full resolution of the tif scans Cone Editions does for me - 36000 x 15000 pixels no matter the print size and small prints look incredible. It has always worked great under windows but when I tried printing them on an older iMac running 10.14 (I think Mojave) the whole thing just freezes up with these gigabyte sized files. Is there a file size or resolution limit on the Mac?
And on the same topic - what resolution should I be using ? Should I resample in Photoshop to 720 and set QTR to 2880 dpi ? I assume how this all works is that QTR creates a dither of 4 dots for each image pixel?
I think what you explained gives you a “Filter failed” issue in the print box.
I had this error and discussed about it with Roy Harrington, it is due to a large pixel file problem.
After some investigation we arrived to the conclusion that there’s a 32 bits limitation somewhere in Mojave, wich was not there before Mojave (I don’t know on the more recent system).
In fact :
16 384 x 16 384 pixels : it prints
16 385 x 16 385 pixels : it doesn’t (filter failed error)
That’s the 32 bits limitation (31 bits exactly, as one bit is used for negatives numbers).
What you could do is force QTR to 8 bits mode and it will print fine (see the printscreen).
I don’t know much about your second question as I never found a fully explained paper on this.
What I tested is with a vectorized file and QTR at 2880DPI :
Difference is visible between 360 and 720 dpi.
Difference is visible between uni and bidirectionnal.
When you upscale/downscale the file on a real image, it not as easy to see as a vectorized file.
Yes I believe that is the error message I received. Does anyone know if this limitation has been resolved in more recent versions ? This is a serious limitation since my files are roughly 350 dpi at 42" x 100".
Won’t printing at 8-bit have an negative impact on print quality ? I have lots of smooth gradients in my images and if I print them as 8-bit FILES I get lots of banding. Or am I confusing two 8-bit concepts ?
I didn’t see differences but the files I printed at the moment were targets with patches (huge one), no gradient at all.
I asked the question to Roy when I had this problem about the downsides of forcing the printing to 8 bits and here is what he answer : “For 16-bit I’ve never seen any way to see a difference - loupe/no loupe.
Everything on a print is simply a dot of ink or a missing dot – so basically 1-bit. There maybe multiple inks (2 or 3) at any
place, and dot res is higher than pixel res. But all in all printers don’t come close to representing those max levels and detail
at the same time. The final bit depth is going down to less than 8 as it is.”
Hope it helps, I don’t know much about these geeky numbers
