That's far from a universal thought, though.Roland is now a shell of their former selves, and I am.far.from alone in thinking
That's far from a universal thought, though.Roland is now a shell of their former selves, and I am.far.from alone in thinking
That's too bad. I surely can't argue about your experience....I clearly must have a more active sales rep.I disagree, when we had the same issues with our roland, roland refused to speak with me, all communication had to go through the dealer,.and.ebentually they stopped responding to the dealer. Roland is now a shell of their former selves, and I am.far.from alone in thinking that.
I bet you two have some frustrating comparisons, at least for him.My cousin has a shop in Arkansas
solventinkjet,The RIP process is non-parallelizable so it can't use hyper threading or multiple cores. I assume setting it to use CPU 0 only fixes some sort of data flow bug in the software. Multiple cores only benefits the RIP process when you want to RIP multiple files at once.
There are some parts of of the process that can be parallelized but most of it can't so the RIP companies don't bother. Basically, the value of one pixel/dot relies on information from the one before it so it has to be processed sequentially.solventinkjet,
i learned something here.
i just assumed we could make use of multiple cores
i was able to make use of multiple cpu's and cores in sco unix in 1997 (you had to press f1 key to assign a process to core 0, f2 key to assign to core 1 etc
i had to load specific firmware to do the same thing in windows NT5 in 1999...
the new CHEAP dell optiplex 7020 micro we installed 10 of yesterday has 20 cores (8 physical and 12 "make believe" for running vm's)
this is shades of running 32 bit windows on a pc with 64gb of ram and finding out that the OS can only use the first 4 gb of ram
It seems strange that RIPping is the only process that can’t be hyperthreadedThere are some parts of of the process that can be parallelized but most of it can't so the RIP companies don't bother. Basically, the value of one pixel/dot relies on information from the one before it so it has to be processed sequentially.
But you can always just hit the turbo button for more processing power right!![]()