Hi, I’m trying to google individual stages of chia plotting. I can not find it. I do not use a plotman, so far I am defending its use.
Can you please explain to me how many phases the interlacing has and which phases use how much RAM, CPU, I / O?
Ideally a nice picture that will explain the course to me. Thank you very much.
BTW: I ran two fences in parallel, one with bitfield and the other without. Without BF, he was faster at first, but the one with BF then caught up with him and both fences ended at the same time.
I dont know any good picture or guide that explains the phases and their implications on CPU, RAM, I/O.
What is known is that there are 4 Phases. The first is able to leverage Multi-Threading, while the others arent. Usually, phase 1 and 3 take the most time.
As theoretically, this is exactly what I was looking for, but in practice it is nonsense. Since 10 hours with 5 GB of RAM and 4 threads plot to the K-32 it has probably the slowest SSD that can be bought.
In addition, I’m most interested in disk I / O activity, space availability is not important to me. I need to know the IO at certain stages in order to better optimize parallel processing.
Even so, your post is very interesting, thank you.
