my network latency is not making use of local cpu usage to its maximum extent so is there anyway where the read nodes copies the data to temp and then renders the frames and after that deletes the temp...? any possibility of this thing...
|
|
my network latency is not making use of local cpu usage to its maximum extent so is there anyway where the read nodes copies the data to temp and then renders the frames and after that deletes the temp...? any possibility of this thing...
Would that speed things up at all? Wouldn't you just end up waiting longer for the data to copy before you did any rendering?
If you're planning to copy DPX or EXR frame sequences, you're copying entire the frames.
That breaks the benefit of reading partial frames. By only reading in the lines of a frame that are used, Nuke already cuts the network read time to the minimum.
there should be solution for it my cpu time is almost idle it doesnt cross up beyond 60% in much time .... mostly it will be in 10-40% best case it hits 90% but only 1% of time.....
some solution to be there for this..
It doesn't matter what your CPU is doing if your bottle neck is your network.
my network is 1Gbps it has benchmark of 60MB/sec sustained throughput for dpx sequences
| « Previous Thread | Next Thread » |
| Thread Information |
Users Browsing this ThreadThere are currently 1 users browsing this thread. (0 members and 1 guests) |