But you don’t want to do that in engine() as it will cause a huge number of 
threads to be spawned.


> On Feb 8, 2015, at 6:31 AM, Mike Wong | ax.gmail <[email protected]> wrote:
> 
> or you want some speedup when you actually pull the data ?  
> 
> e.g. create a Tile from upstream ?
> 
> http://docs.thefoundry.co.uk/nuke/90/ndkreference/Plugins/classDD_1_1Image_1_1RowCacheTile.html
>  
> <http://docs.thefoundry.co.uk/nuke/90/ndkreference/Plugins/classDD_1_1Image_1_1RowCacheTile.html>
> 
> the 'mt' thread is your friend.
> 
> 
> On Sat, Feb 7, 2015 at 8:01 AM, Johannes Saam <[email protected] 
> <mailto:[email protected]>> wrote:
> Hey!
> 
> does iop-request internally cut the requested area up and thread it properly? 
> I am in a situation where i want the entire image requested. And thought i 
> could speed it up by threading it using Thread::spawn and then call request n 
> times ( one block per thread ) but i am not seeing much improvement. Is that 
> handled by nuke? As in does it get cut up and then passed to _request?
> 
> If not i can post the threading example code here and people could do that if 
> need be.
> 
> Thanks
> jo
> 
> _______________________________________________
> Nuke-dev mailing list
> [email protected] <mailto:[email protected]>, 
> http://forums.thefoundry.co.uk/ <http://forums.thefoundry.co.uk/>
> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-dev 
> <http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-dev>
> 
> 
> _______________________________________________
> Nuke-dev mailing list
> [email protected], http://forums.thefoundry.co.uk/
> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-dev

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

Reply via email to