Yahoo Groups archive

Digital BW, The Print

Index last updated: 2026-04-28 22:56 UTC

Message

[Digital BW] Re: 16 bit and printer output -???

2005-01-14 by john dean

Sorry I missed out on this conversation that I started.  To me it is significant, for the 
present as well as the future. I was teaching a class on post production tonight  where I 
just explained to the students that they should always try to capture and scan a file, and 
work on it in 16 bit but, that it was totally redundant and a time drag to ever send this file 
to the Epson printer as 16 bit because it would be converted to 8 bit at printing anyway. 
So? How far off am I? Tyler suggests that it is not doing this at output rip time.

I'm still confused - if you send a fine drum or digital camera file scan to the Epson printer ( 
without a rip or specialized software ) in 16 bit color, is there any difference visible 
difference between that print and the same file going to the printer as an 8 bit color file? 

My comment to the class is that we can scan in 16 bit, now we can do all the PS 
manipulations in 16 bit, but WE can't print from it with any added value? Is this true or 
not? I would think that outputting to film would be more of a process that could utilize it 
but ink on paper? Would you even see that extra data? 

My final question is: even with this specialized rip and profiling application that this 
company claims they have, that ya'll are refering to, is the printer mechanism capable of 
utilizing this extra color ( and tonal ) depth? From the comments below I would think not, 
but I'm very sure that I have no idea. Nor do I know if rips like Studio Print or Imageprint 
have the ability to utilize 16 bit is some different way that the generic Epson drivers do. To 
me this would seem like a major thing to consider these days with everything capturing in 
16 bit, but I never hear anyone discussing it.

John



> 
> It makes good sense for the software to be completely 16 bit. That
> means that profiles and image file are 16 bit. These are most usually
> not floating point so that at the ends of the scale the relative
> errors are large if 8 bit. Having only one meaningful limitation makes
> the outcome far more predictable.
> 
> The only limitation that should be felt is that of the printer.
> 
> Here; it is not magic, a set of halftone paterns are enumerated and
> appear, when covering a reasonably sized swatch to be one density or
> another, but with different sqirt amounts and dot gain the enumerated
> list of densities may not even be monotonic; let alone able to map the
> full range of a 8 bit number. As well particulaly with pigment
> printers there is an ink maximum past which the density decreases. The
>  profiling software needs to be smart enough to reject some patterns
> to get a stable monotonic function to use in printing.
> 
> From inspection of the Epson NCA grayscale one finds a flat black toe
> so that even Epson is not finding sufficient enumerable patterns to
> fill the 8 bit space. From the length of the toe for the 2200 there
> are about 30 or so steps less than 255 available.
> 
> ...John
> 
> --- In DigitalBlackandWhiteThePrint@yahoogroups.com, Joe Berndt
> <joeberndt@v...> wrote:
> > Tyler,
> > 
> > Yes, IJC/OPM is completely 16 bit,  image files that are 16 bit
> remain 16
> > bit and all 8 bit files are internally converted to 16bit .  We
> drive the
> > printer with this 16bit data.
> > 
> > OPM uses it¹s own method to connect and print to the printer,  it
> completely
> > by-passes the Epson driver  and all  system  related printing methods.
> > 
> > OPM actually opens a direct handle to the printer and steams the data
> > directly to it.  
> > 
> > Joe Berndt
> > 
> > 
> > 
> > On 1/13/05 10:46 AM, "Tyler Boley" <tyler@t...> wrote:
> > 
> > > 
> > > I'm getting out of my element here, but my understanding is that all
> > > printer drivers make 16 bit ramps internally. Also, many profiles,
> > > even color, use 16 bit LUTs. Of course IJC/OPM probably does not use
> > > icc profiles in the strictest sense, but if their "profiles" use LUTs,
> > > that's probably where that statement applies.
> > > So some of this would not be extremely informative, until you see
> this-
> > > "(16-bit files remain 16-bit)".
> > > That answers the question.
> > > Tyler
> > > 
> > > ---- In DigitalBlackandWhiteThePrint@yahoogroups.com, "bwbonkers"
> > > <PeterDLevis@a...> wrote:
> > >> > 
> > >> > OPM/IJC from Bowhaus uses 16 bit for printing.
> > >> > 
> > >> > I quote from their website:
> > >> > 
> > >> > ...All of IJC/OPM profiles and rendering algorithms operate in 16-
> > >> > bit, over 65 million shades of gray, resulting in smoother
> gradations
> > >> > and transitions. IJC/OPM internally converts all 8-bit grayscale
> > >> > files to 16-bit during printing to take advantage of the 16-bit
> > >> > profiles and rendering algorithms (16-bit files remain 16-bit)....
> > >> > 
> > >> > Peter.
> > > 
> > > 
> > > 
> > 
> > 
> > 
> > 
> > [Non-text portions of this message have been removed]

Attachments

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.