Digital BW, The Print

Clayton Jones wrote:
> Hello Ernst,
> 
> 
>>I'm still convinced that much of those BO banding issues can be 
>>diminshed to zero if the black ink can be squirted from many 
>>nozzles. 
> 
> 
> I can't argue with the idea, not having tried it, but I wonder,
> because last summer when I did the analysis of the tech print exchange
> one of the things I graded on was dither banding, and even included a
> report for each print on which sections of the ramp it appeared. 
> Every system was represented in that exchange and they all had it to
> some degree (one Septone print had none, but others did have some).  I
> have also seen it in BW prints from my 2200 made with the full UC ink
> set. 
> 
> 
>>With QTR it must be possible to use all the heads of say a C84 
>>with black and reduce the amount of ink on all heads where 
>>needed. 
> 
> 
> In a recent print exchange I received a QTR/Grayscale print that has
> some of the worst dither banding I've seen.  So it would seem that
> having all nozzles firing doesn't seem to be related...but it was also
> done with an "Epson Photo EX" - I don't know what that is, could it
> have caused it?  
> 
> Flash! - I just now rec'd in the mail a QTR-XP/2200/UC print that has
> some very minor DB.

All workflows that partition the greyscale on the available grey 
inks.  True there's overlap but less than twice the head nozzle 
count. Toner may add to that sometimes. You didn't get samples 
that either used 4 heads for black or had the black accompanied 
by the grey of the UC over the entire range like Tyler suggests. 
In all partioned workflows there are no overlaps on the lightest 
grey and the black or the last gets a small boost from the next 
grey (that's what I have done on the  Wasatch SoftRIP curves as 
the 9000 doesn't print a BO without some banding). On top of that 
the nozzle count per head is on all the models 1/2 or less of the 
4000 and 10X00 headnozzles.  I guess the 4000 also replaces 
failing nozzles in its printing like the 10X00 does.

In the early 6 head Piezography arrangement were two heads busy 
with the same task if I remember it correctly. But at that time 
the heads were even more limited in nozzle count.

>>the 4000 does a good job with its 180 nozzles 
> 
> 
> There also seems to be two different kinds of DB.  Carl Schofield sent
> me some 4000 BO prints that had no DB throughout the print (covering
> all Zones) except for a barely visible amount (with 8x loupe) about
> 3/4 inch along one end (either the start or the end of the print, but
> only on one end).  This DB also crosses all Zones, which means it is
> more likely from a physical cause such as paper feed, etc.  Much DB
> appears only in certain Zones, which would seem to be software
> related.

Often caused by a head that starts. As I explained to Antonis 
some time ago it can be handy to print a CcMmYK bar at the edge 
of the paper which starts earlier than the print and ends later 
to get/keep the head open in time. With more prints after another 
on the roll and lots of margins or the greys distributed unevenly 
along the printrun the chance that nozzles fall away can be 
diminished by a bar like that. Antonis used a small greyscale bar 
along the length so all the channels were busy. I've used it for 
years on 9000's with Generation 4.  As I understand it there are 
now RIPs that have it built in.

> 
> I also received a 4000 BO print from someone else that has quite a
> lot.   I just went back through a dozen prints from a BO exchange and
> found two without any DB (both from 2200s) and three more with
> miniscule amounts in a few places.
> 
> At least we know it's possible not to have any dither banding with BO.
> What do you make of all this?

I would love to see a C84 with an all black cartridge driven by 
QTR.  BO certainly has its qualities but can't depend on our luck 
with printers.

Ernst