Installing LinuxCNC 2.9 or Master on Linux Mint 21

Will let it run in this test for night

Soo. Only test one process on single core.

Looking good, thank you.

tommylight wrote:

Looking good, thank you.
 

And it have stable result now.  After restart. Around 7k. You should not test it with both threads. All optimization is done for single core only.

Could anybody explain - are there any difference between servo and base thread on testing ? If servo is run at same period as base ?

 

Been to long, but i do recal one of those does floating point calculations while the other does not. No idea as to why or how, i forgot.
One remark, if you plan on using the parallel port, test for base period as that limits the amount of steps per second produced, coupled with step timings.
Last 3 years i have moved to Mesa ethernet boards for everything, so no need for base period and latency is not an issue anymore as most new-ish PC's are OK, although some are useless, had one of those.
Still, the amount of tweaks you and NTULINUX figured out is astonishing, thank you all.

Re-run servo only with 25000 period - core load is same as from base thread if run together with serve thread. Around 35-36%. So it may be the same.

Why i am on parport ? Because i can. And it do remember me, 20 years ago - you can't scroll pdf document fast (On celeron 300 ). Today - also. And adobe reader is a big Reader today.. but pc power is much better today. Reason - too much crap in software, because it still work. And it can be easier to sell. You buy for 80% of features which will not be needed
F35 have ~11 milion lines of source code and runs on three paralell os..

Sarcastic developer remembering his childhood.