So Our PC's are experiencing some issues and i finally tracked it down to be SLI related.
Here is the problem. With SLI on, we get no crashes and everything is wonderful there, but on large files, we get 2 hour renders a frame on frames that should only take 13 minutes. It starts out fine, but eventually the frames take longer and longer to render. Since, it is known that SLI can cause issue's with Octane... no biggy. We'll just turn SLI off.
So we turn SLI off and we get frames rendering as they should, but Octane hard crashes the computers every once in a while. I just had 4 crashes on a 300 frame render at 260 spp. Very quick render and it hard crashes the computer 4 times. Turn SLI back on and it renders fine all the way through. This file is small enough that we did not experience the longer render times using SLI.
This is not an issue with our Mac's FWIW.
Any insight? Any graphics card options i am missing?
Specs:
1.00 RC2
Win 8 (not 8.1)
r15
Machine specs in my sig.
SLI and non SLI issues
Moderators: ChrisHekman, aoktar
1 Dell Z840 with 2x Titan XP's
2 Digital Storm Hailstorm/Win8/32GB/EVGA Titan x4
2 Server blades with 4 Tesla K80's
2 Digital Storm Hailstorm/Win8/32GB/EVGA Titan x4
2 Server blades with 4 Tesla K80's
Hi,
this issue is out of plugin. Octane team maybe give some advices. But sounds like a hardware problem
this issue is out of plugin. Octane team maybe give some advices. But sounds like a hardware problem
Octane For Cinema 4D developer / 3d generalist
3930k / 16gb / 780ti + 1070/1080 / psu 1600w / numerous hw
3930k / 16gb / 780ti + 1070/1080 / psu 1600w / numerous hw
I'd call it a hardware problem if it wasn't both machines. OK. We'll see if anyone else has an issueaoktar wrote:Hi,
this issue is out of plugin. Octane team maybe give some advices. But sounds like a hardware problem
1 Dell Z840 with 2x Titan XP's
2 Digital Storm Hailstorm/Win8/32GB/EVGA Titan x4
2 Server blades with 4 Tesla K80's
2 Digital Storm Hailstorm/Win8/32GB/EVGA Titan x4
2 Server blades with 4 Tesla K80's
hi,
a couple of questions:
did you have also connected the cards with hardware sli?
which kind of exporting per frame are you using?
ciao beppe
a couple of questions:
did you have also connected the cards with hardware sli?
which kind of exporting per frame are you using?
ciao beppe
bepeg4d wrote:hi,
a couple of questions:
did you have also connected the cards with hardware sli?
which kind of exporting per frame are you using?
ciao beppe
EXR float files
With or without hardware bridge, turning off SLI results in random crashes.
1 Dell Z840 with 2x Titan XP's
2 Digital Storm Hailstorm/Win8/32GB/EVGA Titan x4
2 Server blades with 4 Tesla K80's
2 Digital Storm Hailstorm/Win8/32GB/EVGA Titan x4
2 Server blades with 4 Tesla K80's
it crash when saving?
at what resolution?
are you voxeling the entire scene or only some parts?
ciao beppe
at what resolution?
are you voxeling the entire scene or only some parts?
ciao beppe
it only randomly crashes when rendering the image sequence. Send all data. 1080ibepeg4d wrote:it crash when saving?
at what resolution?
are you voxeling the entire scene or only some parts?
ciao beppe
1 Dell Z840 with 2x Titan XP's
2 Digital Storm Hailstorm/Win8/32GB/EVGA Titan x4
2 Server blades with 4 Tesla K80's
2 Digital Storm Hailstorm/Win8/32GB/EVGA Titan x4
2 Server blades with 4 Tesla K80's
If these crashes are hard crashes (ie: machine totally locks up or spontaneously restarts) then i would start by suspecting some kind of hardware/firmware or driver issue rather than octane. SLI generally just causes issues, so i'm guessing it's something that SLI is causing (lower card usage, different VRAM usage or some such) that is saving it from crashing (just a random guess - i really don't know).
What kind of PSU wattage do you have?
Have you had any hard crashes in any other software?
4x Titans can generate a lot of heat, so keep an eye on that (Titans overheating is far less likely than something like 4x 590's, but it's still worth checking).
(Our test machine here has the side panel and some of the front drive bay blanks removed to help in airflow.)
Thanks
Chris.
What kind of PSU wattage do you have?
Have you had any hard crashes in any other software?
4x Titans can generate a lot of heat, so keep an eye on that (Titans overheating is far less likely than something like 4x 590's, but it's still worth checking).
(Our test machine here has the side panel and some of the front drive bay blanks removed to help in airflow.)
Thanks
Chris.
I really appreciate the help FooZe. I'll check these things out. My initial thought is, these are gaming machines essentially. I'm guessing that they are just put together to work with SLI. I'm not PC-savy at all (I'm a Mac guy), so i'm not sure what software i should be looking at. The nvidia settings have a lot in there. All i have done is turn off SLI. I may contact Digital Storm to see if they have any insight.FooZe wrote:If these crashes are hard crashes (ie: machine totally locks up or spontaneously restarts) then i would start by suspecting some kind of hardware/firmware or driver issue rather than octane. SLI generally just causes issues, so i'm guessing it's something that SLI is causing (lower card usage, different VRAM usage or some such) that is saving it from crashing (just a random guess - i really don't know).
What kind of PSU wattage do you have?
Have you had any hard crashes in any other software?
4x Titans can generate a lot of heat, so keep an eye on that (Titans overheating is far less likely than something like 4x 590's, but it's still worth checking).
(Our test machine here has the side panel and some of the front drive bay blanks removed to help in airflow.)
Thanks
Chris.
We only use these for Octane and Octane only, so i'm not sure if this issue happens with anything else.
Taking off the side panel on the PC's actually created more heat according to the temp monitors. The machines are liquid cooled, with about 20 fans and we have external fans strategically placed on the machine as well.
Machine specs are below. Since both of our machines are identical and have the identical issues, i would say it's a driver/firmware issue or what i mentioned above. The cards and processor are NOT over-clocked outside any factory over clocking.
Chassis Model: Special Deal Hot Seller - Hailstorm II Edition
Processor: Intel Core i7 3930K 3.2GHz (Unlocked CPU for Extreme Overclocking) (Six-Core)
Motherboard: ASUS Rampage IV Extreme X79 (Intel X79 Chipset) (Features USB 3.0 and SATA 6Gb/s)
System Memory: 32GB DDR3 1866MHz Corsair Dominator Platinum DHX (Extreme-Performance)
Power Supply: 1500W Silverstone (Dual/Triple/Quad SLI Compatible) (Recommended)
Hard Drive Set 1: Operating System: 1x (480GB Solid State (By: Corsair) (Model: Neutron GTX Series) (SATA 6Gbps) Optical Drive 1: DVD-R/RW/CD-R/RW (DVD Writer 24x / CD-Writer 48x)
Internet Access: High Speed Network Port (Supports High-Speed Cable / DSL / Network Connections)
Video Card(s): 4x SLI Quad (NVIDIA GeForce GTX TITAN 6GB (EVGA Super clocked Edition)
Sound Card: Integrated Motherboard Audio
Extreme Cooling: H20: HydroLux Level 4: Digital Storm Exotic Custom Cooling System (4x Video Cards + CPU + Chipset) H20 Tube Color: Red Tubing with High-Performance Fluid (UV Lighting Reactive)
Chassis Airflow: Upgrade All Fans to Corsair Airflow 120mm Performance Edition (Up to 6 Fans)
CPU Boost: Stage 1: Overclock CPU 4.0GHz to 4.4GHz
Graphics Boost: - No Thanks, Please do not overclock my video card(s)
Memory Boost: - No Thanks, Please do not overclock my memory
OS Boost: - No Thanks, Please do not tweak the services on the operating system
Windows OS: Microsoft Windows 8 Professional (64-Bit Edition)
1 Dell Z840 with 2x Titan XP's
2 Digital Storm Hailstorm/Win8/32GB/EVGA Titan x4
2 Server blades with 4 Tesla K80's
2 Digital Storm Hailstorm/Win8/32GB/EVGA Titan x4
2 Server blades with 4 Tesla K80's
Hi,
A good test would be to try an animation that you know is most likely to hang the machine and just use various combinations of the GPU's (rather than all four) just by un-selecting them in octane.
Your doing the right thing by disabling multi-gpu in the nvidia control panel, this should be all you need to touch.
If you find it crashes with a specific GPU enabled, then this would suggest it is a problem with a specific GPU.
Another thing to check is the windows event viewer under System, see if you can find any information about the crashes.
If it is a device or driver issue, usually windows will be able to write a log to the system event log before the machine restarts.
If it's hardware problem like the power supply or motherboard spontaneously restarting, then windows will not be able to write anything before the restart.
But when windows comes back up again it will put in a critical level log about it.
ie: if i hit the reset button on my PC when windows boots back up it will have a log:
level: Critical
Source: Kernel-Power
EventID: 41
Task Category: 63
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
Thanks
Chris.
A good test would be to try an animation that you know is most likely to hang the machine and just use various combinations of the GPU's (rather than all four) just by un-selecting them in octane.
Your doing the right thing by disabling multi-gpu in the nvidia control panel, this should be all you need to touch.
If you find it crashes with a specific GPU enabled, then this would suggest it is a problem with a specific GPU.
Another thing to check is the windows event viewer under System, see if you can find any information about the crashes.
If it is a device or driver issue, usually windows will be able to write a log to the system event log before the machine restarts.
If it's hardware problem like the power supply or motherboard spontaneously restarting, then windows will not be able to write anything before the restart.
But when windows comes back up again it will put in a critical level log about it.
ie: if i hit the reset button on my PC when windows boots back up it will have a log:
level: Critical
Source: Kernel-Power
EventID: 41
Task Category: 63
The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
Thanks
Chris.