Tweet
Hi,
My workstation has recently been upgraded to a 6core hyperthreaded i7, and it has revealed a bit of an issue...
When doing a DR render, the slaves (which are all intel quad cores), join in the final render REALLY late, sometimes too late to actually do anything. According to the vray log, the slaves were still calculating the light cache, while the workstation finished it ages ago and got on with the other passes.
A workaround is to render the lightcache seperately, and then direct all the slaves to use the saved pass. this way all the nodes join in on the other passes.
Is there any way of automating this in one render? making the slaves wait for the workstations light cache rather than rendering their own?
cheers,
Nat.
My workstation has recently been upgraded to a 6core hyperthreaded i7, and it has revealed a bit of an issue...
When doing a DR render, the slaves (which are all intel quad cores), join in the final render REALLY late, sometimes too late to actually do anything. According to the vray log, the slaves were still calculating the light cache, while the workstation finished it ages ago and got on with the other passes.
A workaround is to render the lightcache seperately, and then direct all the slaves to use the saved pass. this way all the nodes join in on the other passes.
Is there any way of automating this in one render? making the slaves wait for the workstations light cache rather than rendering their own?
cheers,
Nat.
Comment