I am back again ;D Sorry.

We have just changed and started using local caching.
After some more use of this function we are having the issue, that disk caching 
takes very different long times to cache.

Usually a artist is initializing a shot and does then "update all" for his 
source (usually just one read node dpx 10bit).
The progressing time is ok and depends of course on our network.
But what then happens is a little bit strange for me to understand.

Artists usually just connect a viewer to the readnode and want to play the 
scene back. Until now the first playback was about 20-24fps what was great.
But now artists claim that their first playback is about 1fps...

What we ve done than was cleaning all caches and restart but this didnt help.
It s still the same project and same footage where fast playback was even 
possible till now.

I really cant see a reason why this is happening now.

Our Local Cache and Disk Cache are on a extra HDD (common hdd, no ssd or 
something) that is "separated" from the OS and other programms.
We are using Nuke 6.3v2 and CentOS 5.4

Would be great if anyone has an idea what s going on...
Is it too much to put both caches on the same HDD, because read and write 
processes are using the same device and block each other?
What setups are you using? SSD? Raid? Several divided storages for the Caches?

This is a very "mysterious" topic because i cant find too much information 
about that and im not sure who to ask about....
Any Hints are really much appreciated.
Thanks in advance.

Best,
Christopher



_______________________________________________
Nuke-users mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users

Reply via email to