|
|
02-06-2009, 08:08 AM
I was wanting to take a survey and see about the stability of REDrushes for people. I have found that it works well with a few clips at a time but put on a large batch it it will often crash or just stall. Sometimes on a stall it will finish the clip but give no indication that is is doing so.
so this thread is just asking for comment on how stable REDrushes is for most users.
02-06-2009, 08:31 AM
perhaps add a poll for number of crashes?
02-06-2009, 08:48 AM
Thanks! I went to the FAQs and figured out how to do it. I made a new thread since I couldn't add it to this one. I would delete this whole thread but there doesn't seem to be an option to do it!Originally Posted by USlatin
02-06-2009, 12:49 PM
My experience - only crashes when/if close to running out of HD space.
Other than that - if I use 8 cores and process 6 clips at once - it's hit or miss wether it will work. Normally, use 8 cores, process 2 clips. This is most reliable for me.
But I stick with REDCINE.
02-06-2009, 02:20 PM
I get best results on an 8-core system by doing 4 clips at once and setting the cores per clip at 2 and then walking away. Or if I'm going to continue working on the machine, do 3 clips at once and leave the # of cores at auto.
Very stable here. Have not had any redrushes issues to speak of. Always reading / writing to faster RAID volumes. Drive systems are not a bottleneck in any way on these systems and plenty of space to operate in. 16GB RAM on both of my 8-core boxes. But redrushes is a 32bit app anyway. However, I think it spawns multiple instances of RedLine, but I'm not completely sure. Looking forward to Windows version of redline so i can put it on my render farm. 2 x 8-core Mac Pros is fast enough for half-hq debayer 4K and full from 3K most of the time.
02-06-2009, 02:41 PM
Precisely my experience. It has been fine with 4/2.
02-07-2009, 08:16 PM
i have never had mine crash - but it seems like the quality and gamma do not equal that of the other apps.
02-08-2009, 03:14 AM
RR tends to crush on our 8cores 3.0 and 3.2 Ghz machines, when encoding just to anything. One cannot just run a render for longer than 20hours, and be certain, that it will not crash after 5 hours.
We now went over to creating scripts with RR, and split them onto 2 terminals of redline, until our renderfarm distributor is ready. That will allow to setup 50 macs to render overnight with one click.
Redline render in the terminal is about 10% faster, than with RR for some reason. All 8 cores are between 95-98% all the time. It is sweet, how you see the speed difference between 3.0 and 3.2 Ghz that way.
We never experienced a crash from the terminal and redline. I just have a render on 4 8core machines running that way for over 40 hours, and they are all online.
| « Previous Thread | Next Thread » |
