More fun with linearizing Accurip.

[ZooCity]

Today I switched our 4800 over to Chromaline's Accuink.  Wow, big difference in d-max over the bulk stuff we had prior which wasn't horrible but the Accuink is miles ahead.  This stuff is worth the cash.  Of course, time to set a new linearization curve.   Attached.  Using aforementioned ink/printer and fixxons film, it puts our film as close as I could get to Pierre's example he sent me, which I know is calibrated. So dots are behaving in each % pretty well, great. 

Now here's the problem- outputting a regular old greyscale drawing with this linearization curve yields an utter trainwreck. Detail lost everywhere in the 5-15% areas and then 97 and 99% fill areas printing like 80 or 90%, just a freaking mess. 
Yet, this is the "correct" curve (or very close to it).  I've seen this come up in the past where our linearization was dialed in but seps done for us by others and outputted with it went fubar.  Shut off the linearization and the seps print perfect.  So the solution has always been to just shut the linearization off.  While I may second guess my own separation skills I'm not second guessing a pro separator's so I don't think this is an "adjust the art" situation based of past experience.

It's almost like linearizing AR causes inaccurate output from the file- based off my test strips, a 97% area of the art should close up a bit when printed onto film and be around 99% or so (effectively solid fill on most mesh counts), not open up and print as if it were 90 or 95%. (very little difference can be made between them anyhow with AR)  I wonder if AR can only make decisions every 5% when using the dot gain linearization controls?....that would explain the mess I'm seeing pretty well.

Time to upgrade rips or am I missing something? 

[TCT]

Kinda crazy as I JUST today put the values in on our AR. We had a issue also, higher values didn't come out right. I know it is a crapy pic, but if you can see on the fade from 0-100, when you get around 97% it ends up looking like 65% or so.  I'll get a better pic tomorrow, and a screen shot of our values. 

[TCT]

So here is my values, a bit different than Zoo's, but the loss that happens around 97% is just odd.

[blue moon]

So here is my values, a bit different than Zoo's, but the loss that happens around 97% is just odd.

Alex,

that curve should be smooth, it should not have the big dip in the middle! Just change that value until it is aligned with the rest of the line.
Overall, the curve looks pretty good and is what I would expect.

not sure about the issues you are having though. I did talk to Charlie and he said everybody should run the wizard and get their density (and/or what ever it adjusts) before doing the linearization.

pierre

[ZooCity]

TCT, change your values in the tenths to factors of five, i.e., 0.5 instead of.9.  AR freaks out if you don't stick to this.

Sent from my XT907 using Tapatalk

[TCT]

TCT, change your values in the tenths to factors of five, i.e., 0.5 instead of.9.  AR freaks out if you don't stick to this.

Sent from my XT907 using Tapatalk

You are the MAN Zoo! Thanks!

[ZooCity]

Anytime TCT. 

I wanted to make sure this didn't sound like a knock on this RIP.  It's not.  We've gotten our use out of AR a million times over for the low price we paid for it, from that perspective it's an incredible product and it runs beautifully on my mac for crying out loud. 

At this point it's a matter of determining if there is an issue with the RIP's linearization feature of if I need to get savvy to adjusting art for linearized output.  Pics attached to show what I'm talking about here.  Check the red box of this image, tons of data lost with linearized output.  Looks great with the linearization off but just look at that 95% box with it's solid fill.  I'm kinda stumped with this.

Pierre, I agree with Charlie, droplet weight is the first line of attack here.  Set your ink deposit as low as you can go and still get acceptable d-max first.  Only then should the output be linearized.  If you are flooding the film with too much ink from the get go, no amount of linear curve choking is going to help you.  Ink and film characteristics come into play here.

Trouble is, AR makes mashed tater looking dots on all settings but 2880 and it has no deposit options at 2880 on any setting.  The tradeoff then is better linearization through deposit control but with fuzzy dots v. nice, clean dots and edges but less control.  This makes a steep linearization curve your only line of attach- unlike my issue which may/may not be RIP related, this is a genuine ceiling to the rip's abilities.

[TCT]

So here is my values, a bit different than Zoo's, but the loss that happens around 97% is just odd.

Alex,

that curve should be smooth, it should not have the big dip in the middle! Just change that value until it is aligned with the rest of the line.
Overall, the curve looks pretty good and is what I would expect.

not sure about the issues you are having though. I did talk to Charlie and he said everybody should run the wizard and get their density (and/or what ever it adjusts) before doing the linearization.

pierre

Pierre- I was just going off the readings you gave me, you think I should deviate from that and just make it more straight?