anyway we love octane running as a devil
Purchase advice for network rendering
Forum rules
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
@glimpse big respect for all of you, rappet,smicha,etc. I love to follow how are you doing (ultrasilent,overclock) but i'm not so confident with myself. Some bad experiences in the past. In some cases, as me, it't better to pay more
anyway we love octane running as a devil
anyway we love octane running as a devil
hi, Salmon.
First I try to answer Your Questions,
about Thunderbolt - there is a possibility to add a card that will translate Your GPU signal to thunderbolt - but You need to look for compatibility as not every system allows that =)
Also I would stay away from wireless solution as it will probably bottle-neck network rendering =) I've done tests while beta testing 2.0 with wireless (300mb - sorta midrange) & it caused slave's GPU not being utilised 100% all the time..so personally I would rather have cable connection or even use remote desktop (like VNC) to fire up render task natively to avoid any network rendering issues (as there might be some)..
about guaranties..well, You will always get friendly reply & help if You encounter any problem. if You wish to pay a bit extra we can deal like if You have problems You will get part replacement before it will be RMAed (that's a topic to be talked about)
We all have bad experiences - that's how You build knowledge..we learn from our mistakes & move one trying to find the best possible way out =) Consider simple fact..if You go boutique build option You'll pay for piece of mind =)
You can do the same Yourself, just by ordering good parts & spending an evening having fun while building Your rig & then if something fail along the way You should have warranty to cover every part (if You choose good quality SKUs, You'll usually get few years at least =)
This forum will give You plenty of guidelines if You want to build something on Your own =) & talking 'bout Smicha, we are working on extended article based on his build that soon will surface - You'll find a lot of interesting info there =)
First I try to answer Your Questions,
about Thunderbolt - there is a possibility to add a card that will translate Your GPU signal to thunderbolt - but You need to look for compatibility as not every system allows that =)
Also I would stay away from wireless solution as it will probably bottle-neck network rendering =) I've done tests while beta testing 2.0 with wireless (300mb - sorta midrange) & it caused slave's GPU not being utilised 100% all the time..so personally I would rather have cable connection or even use remote desktop (like VNC) to fire up render task natively to avoid any network rendering issues (as there might be some)..
about guaranties..well, You will always get friendly reply & help if You encounter any problem. if You wish to pay a bit extra we can deal like if You have problems You will get part replacement before it will be RMAed (that's a topic to be talked about)
We all have bad experiences - that's how You build knowledge..we learn from our mistakes & move one trying to find the best possible way out =) Consider simple fact..if You go boutique build option You'll pay for piece of mind =)
You can do the same Yourself, just by ordering good parts & spending an evening having fun while building Your rig & then if something fail along the way You should have warranty to cover every part (if You choose good quality SKUs, You'll usually get few years at least =)
This forum will give You plenty of guidelines if You want to build something on Your own =) & talking 'bout Smicha, we are working on extended article based on his build that soon will surface - You'll find a lot of interesting info there =)
-
JessicaVines
- Posts: 69
- Joined: Thu May 08, 2014 7:52 am
Hi guys,
I have more or less similar questions regarding this type of slave setup in a multiuser work environment. We have started rolling out the new tin can mac pros which are fitted with ATI cards; useless for Octane. Some questions if I may?
Presuming we buy a linux box with 4 GPUs and an additional octane license, and I am working on a new mac pro without ANY Cuda support,
1) Can I run a Octane cinema 4d plugin and utilise these slave GPUS as if they were internal cards, or do I need to have at least one internal card?
2) I understand how team render queuing or 3rdparty cmdline render works, but how does Octane slaves go with multiple users iterating at the same time? Is each user notified though dialogue which GPUS are already in use over the network?
3) Is it possible to run Octane slave node on a linux box, while all our work boxes are on macs using say cinema 4d plugins?
Thank you.
I have more or less similar questions regarding this type of slave setup in a multiuser work environment. We have started rolling out the new tin can mac pros which are fitted with ATI cards; useless for Octane. Some questions if I may?
Presuming we buy a linux box with 4 GPUs and an additional octane license, and I am working on a new mac pro without ANY Cuda support,
1) Can I run a Octane cinema 4d plugin and utilise these slave GPUS as if they were internal cards, or do I need to have at least one internal card?
2) I understand how team render queuing or 3rdparty cmdline render works, but how does Octane slaves go with multiple users iterating at the same time? Is each user notified though dialogue which GPUS are already in use over the network?
3) Is it possible to run Octane slave node on a linux box, while all our work boxes are on macs using say cinema 4d plugins?
Thank you.
Interesting Questions, for some of them (Q1) I want to to know answers too, 'cos I'd like to hook ultraportable (MacBookAir) with render slave..sometimes workstations is bit busy..JessicaVines wrote:Hi guys, ..
1) Can I run a Octane cinema 4d plugin and utilise these slave GPUS as if they were internal cards, or do I need to have at least one internal card?
2) I understand how team render queuing or 3rdparty cmdline render works, but how does Octane slaves go with multiple users iterating at the same time? Is each user notified though dialogue which GPUS are already in use over the network?
3) Is it possible to run Octane slave node on a linux box, while all our work boxes are on macs using say cinema 4d plugins?
as for Q2, at least as I understand, if someOne is using them, there's no way for other to use the same slave..or GPUs. There's technology, company called Renegatt is wring that allows GPUs to be virtualise, & then You can manage GPUs as You like in multi user environments =) but Virtualised GPUs goes agains OTOY's EULA (that doesn't allow to run Octane on such "devices")
Q3, I'm pretty sure it is, but ask Aoktar (developer of C4D to be sure =)