Digital BW, The Print

--- In DigitalBlackandWhiteThePrint@y..., "Antonis Ricos" <
antonisphoto@y...> wrote:
> --- In DigitalBlackandWhiteThePrint@y..., "Martin Wesley" 
> <mwesley250@e...> wrote:

Antonis, it's sounding to me like you have a lot of quad experience 
yourself, certainly more than I.
> 
>> ...Hopefully with the MIS VM I will be able to dial in a tone 
> > that suits. I was thinking that the full-blown RGB profile would 
> > allow me to adjust the color cast on screen.

True, but you need to also be adjsting the non-toner inks for best 
dither and accurate grayscale reproduction, and mixing the toner ink 
in to suite, then your overall densities have changed so the non-toner 
inks have to be readjusted for density accuracy which changes the hue 
again...and doing it all in a way that forces the dirver to gring the 
K ink in apropriately...
My experience with a good RGB profile, and good CMYK setups is that 
subtle hues and near neutral colors are hard for monitors to preview 
dead on. You have to go the last 5% by eye anyway.
> 
> OK, here is where I am being a bit theoretical about this and will have to 
> respect Dan's far more practical experience: I think you can separate color 
> from density in this case. Once you settle on the color you like, you don't have 
> to deal with it again. This may be a simplified use of the VT possibilities 
> (deliberate selective toning of tonal regions etc), but it's a start. Then you deal 
> with spreading out the inks over the print densities using just densitometry.

One conclusion from the above is that, even with as RGB preview 
profile, good sound quad separation methods necessary for any of these 
workflows still apply. Right, the profile won't do that for you.

> > So if you are just measuring density then the workflow would be to 
> > print and measure a test wedge, make an adjustment curve using the 
> > measured values to push tones in the proper direction, apply the 
> > curve as a final step prior to printing, print and repeat until you 
> > have an accurately printed step wedge that divides the range evenly 
> > from Dmax to Dmin?

You have to do that anyway. In a sense, printing and measuring a chart 
and building a profile is providing you with the same information. But 
more elaboratly (and expensively). Developing and applying the curves 
to print remains the same. The profile simply helps you make good 
curves quicker, and see it on the monitor.
> 
>...My own experience assumes that you have access to 4 
> individual "plates" and can write separation actions. I have not, muself, done 
> this through an RGB driver nor with 6 inks. I am still thinking that I would rather 
> have a large format Epson with a RIP that I can talk postscript to, than anything 
> with an RGB driver and a black box of internal separations. But that's another 
> story...

That's exactly right Antonis. But if one IS using the driver, the 
profile has the advantage of reporting to you exactly what the driver 
is doing. You probably know from your experience that having the exact 
density and dot gain of each ink does not necessarily tell you how 
they will combine at different densities. Certain combos of the MIS 
set actually reverse out, without knowing what tests to run you'd 
never know it and be trying to undo it. Since the profiler is 
reporting info about 729 different mixtures, it's pretty accurate. 
Also it does a remakable job of predicting exactly how the black ink 
replacement will effect the high CMY values, which was one of the 
major difficulties with the Epson driver before, requiering trial and 
error.
In Allen's case, a lot of reverse engineering.
> 
> ....
> > What if you jump from the $900 BW densitometer to the $1400 
> > spectrophotometer? Well I guess if you are just measuring density you 
> > don't gain anything.
> 
> You have to add the cost of the software to make the 1,400 investment 
> worthwhile. Of course, if the argument is that the spectro can do double duty 
> and that you can buy the software as needed later, then, sure, why not. Just 
> keep in mind that the $1400 doesn't buy you a transmission unit (whereas the 
> $900 does) - just in case you were going to read your negs too and calibrate 
> your processing or those digital negs you are looking to buy !......  <g>
> 
> That's a loooong way to go to avoid the warming of piezo. And even then, 
> aren't Paul's curves for the VT inks pretty much a done deal out of the box? (at 
> least the ones completed).
> 
> Antonis

I agree with you Antonis. The only thing I can say is that since I 
already had the spectro and software, I had to try Dan's idea. I've 
now used it in a few different situations and have been really 
impressed by how well it works as a solution for using the Epson 
driver for quad ink sets.
But it won't replace good sep methods, you still have to learn all 
that!
Tyler