v3 release date

[philliplakis]

Any one have any idea in regard to a release date?

Here is my concern atm

I just upgraded to a nMP, and atm I'm modelling on the nMP deactivating and taking it to my cMP for render.

I want to network it,..
But don't want to spend 300e on it now and then have to spend addition money to upgrade to v3.

Can Otoy offer a free upgrade to v3 if its coming soon?

[philliplakis]

also just adding:


Is anyone else rather curios on the OpenCL adaptation

with the R9 390x with 8gb Ram blowing a Titan X in OpenCL for a 3rd of the price, What happens to the people who bought them and 980ti's
How will Otoy adapt OpenCL with us who have spent all this money on Cuda when OpenCL is implemented??

Will they restrict it? Im awaiting answers but 3 questions have been left un-answered from Otoy

[Rikk The Gaijin]

The closed beta should start soon, but the final release will probably be around the end of the year.

[ten]

also just adding:


Is anyone else rather curios on the OpenCL adaptation

with the R9 390x with 8gb Ram blowing a Titan X in OpenCL for a 3rd of the price, What happens to the people who bought them and 980ti's
How will Otoy adapt OpenCL with us who have spent all this money on Cuda when OpenCL is implemented??

Will they restrict it? Im awaiting answers but 3 questions have been left un-answered from Otoy

Why on earth would they restrict anything? They will want to give the maximum performance in every varient on as many different pieces of hardware as possible to maximize sales potential.
"Hey lets spend all that R&D time on openCL implementation and then gip it so it runs slow as to not annoy cuda people" That makes good business sense, right?

Although I don't think cuda peeps should worry anyway. For the first revisions I seriously doubt the opencl version will match the cuda versions speed. Octane has been cuda since day1 so it has years of coding and tweaking under the hood. Maybe in time the opencl version could rival or surpass it, although personally I doubt it.

ten

[philliplakis]

Why on earth would they restrict anything? They will want to give the maximum performance in every varient on as many different pieces of hardware as possible to maximize sales potential.
"Hey lets spend all that R&D time on openCL implementation and then gip it so it runs slow as to not annoy cuda people" That makes good business sense, right?

Although I don't think cuda peeps should worry anyway. For the first revisions I seriously doubt the opencl version will match the cuda versions speed. Octane has been cuda since day1 so it has years of coding and tweaking under the hood. Maybe in time the opencl version could rival or surpass it, although personally I doubt it.

ten

So if the OpenCL version comes out, and a R9 390x beats your Titan X you would be ok with that?

All I'm saying is if they release a openCL version that beats Cuda then how bout people like you with 2 titan x's you could of bought 5-6 R9 390x's for the price and your getting beat... I think its unfair.

[glimpse]

So if the OpenCL version comes out, and a R9 390x beats your Titan X you would be ok with that?

All I'm saying is if they release a openCL version that beats Cuda then how bout people like you with 2 titan x's you could of bought 5-6 R9 390x's for the price and your getting beat... I think its unfair.

Yes!!! I would be totally fine =) if You're not, who on earth forced You to buy TitanX?

upcoming Octane Render v III was introduced March 17

TitanX surfaced at GDC ~March 2nd, but was detailed at Nvidias GTC ~April 4th soon after comming to market..

So, You knew about Version III before having a possibility to buy TitanX..- what it has to do with OTOY, if let's say You make a stupid decission & buy soemething You don't need? It's Your choise & that's it..

Titan X has 12 GB of vRAM & it's ment for those who need that much (& if You need 12 GB You would not start this conversation in first place), otherwise get 980Ti..or buy older 780 6GB if You can find them..

I would be totally fine if OTOY would charge separatelly for OpenCL version, 'cos I bet it has been hell of a work to translate all the code.

Guys put tremendous amount of work to bring products without any guaranties that their hours are going to payoff in the end - users on the other hand make all the decicions about purchases..& as mentioned I don't see OTOYs responsibility to think for users that invest into anything (there's a reason why press releases are made not the last day, so any user would have enough time to get information they need).

BUT =) I do see Your point, just..if that matters for You, simply hold horses for half a year..or build something You can easily resell if needed & that's all.

[r-username]

any thoughts if v3 public beta will be released in the cloud in advance of the downloadable standalone ones? I could be wrong but it seems that ORC users will get the new toys first.
https://orc.otoy.com/

or if cuda and opencl can work on one render at the same time
(x cuda cards and x opencl cards in the same box)

[ten]

Why on earth would they restrict anything? They will want to give the maximum performance in every varient on as many different pieces of hardware as possible to maximize sales potential.
"Hey lets spend all that R&D time on openCL implementation and then gip it so it runs slow as to not annoy cuda people" That makes good business sense, right?

Although I don't think cuda peeps should worry anyway. For the first revisions I seriously doubt the opencl version will match the cuda versions speed. Octane has been cuda since day1 so it has years of coding and tweaking under the hood. Maybe in time the opencl version could rival or surpass it, although personally I doubt it.

ten

So if the OpenCL version comes out, and a R9 390x beats your Titan X you would be ok with that?

All I'm saying is if they release a openCL version that beats Cuda then how bout people like you with 2 titan x's you could of bought 5-6 R9 390x's for the price and your getting beat... I think its unfair.

I wouldn't give a hoot. It's called progress (and I would upgrade my system accordingly if I thought I needed it) . If ever a company ever thought like that it would be the end of them.

Anyhow, once again, the opencl version will NOT beat the cuda version!!! I guarantee it! [emoji6]

Ten

[sgtprobe]

So if the OpenCL version comes out, and a R9 390x beats your Titan X you would be ok with that?

All I'm saying is if they release a openCL version that beats Cuda then how bout people like you with 2 titan x's you could of bought 5-6 R9 390x's for the price and your getting beat... I think its unfair.

Why would it be unfair? It's not a competition or anything. If AMD cards are faster in Octane3 I would simply sell all my Nvidia cards and buy AMD instead. Sure, a lot of money invested in my Titan Blacks, but that's how it is.

[aoktar]

also just adding:


Is anyone else rather curios on the OpenCL adaptation

with the R9 390x with 8gb Ram blowing a Titan X in OpenCL for a 3rd of the price, What happens to the people who bought them and 980ti's
How will Otoy adapt OpenCL with us who have spent all this money on Cuda when OpenCL is implemented??

Will they restrict it? Im awaiting answers but 3 questions have been left un-answered from Otoy

Nothing is unfair. You may want to say thanks if they can bring a good alternative on opencl. It's very big job to provide two different kernels for Cuda+opencl. Double work. Also don't except much words about it until things become more presentative. Also i'm thinking that will be good answer to NVIDIA's efforts for being a monopoly in computational area.