OctaneRender® 1.024 beta 2.48 TEST (win) [OBSOLETE]

[arexma]

I am not sure why NVIDIA did this, but I assume (I am not sure though) it's a change related to trying to cripple the use of Geforce Products for CUDA computation in profesional markets.

OT:
You got to love NVIDIA.
They cripple GeForce Cards up and down in OpenGL (performance changes from driver to driveR) to sell Quadros for professional tools, then they cripple GeForce and Quadros in their GPGPU capabilities to sell more Tesla cards and all run with the exact same chip just with different limbs broken -.-
Great, soon I don´t even need to have a fortune aside for various software licenses, I also need a few computers with different cards for various tasks.

And what can we do about it? Nothing, because all 3 product lines will continue to sell.
Stupidass Nvidia.

[Jaberwocky]

Ok another small bug to add to the list.

If you open a scene and switch to daylight mode.

Notice that the lighting in it starts bluish in colour.

Now slide the day from the 1st to any other day and then back to the 1st again.The lighting changes and then remains that colour when you go back to the 1st again.

It's almost as if the starting value's for the R.G.B. colours are wrong, then they correct themselves as soon as you set a day.

Hope that makes sense.

[Jaberwocky]

Sorry I forgot.Nice job with the memory stats in the bottom RHS of the render preview window guys. I now no longer need to try and figure out how much memory i've got left on my card after powering my screens. Excellent.. Looking forward to 2.5. Should be good......

[abstrax]

Ok another small bug to add to the list.

If you open a scene and switch to daylight mode.

Notice that the lighting in it starts bluish in colour.

Now slide the day from the 1st to any other day and then back to the 1st again.The lighting changes and then remains that colour when you go back to the 1st again.

It's almost as if the starting value's for the R.G.B. colours are wrong, then they correct themselves as soon as you set a day.

Hope that makes sense.

Yup, makes sense. Thanks for the report.

Marcus

[abstrax]

I am not sure why NVIDIA did this, but I assume (I am not sure though) it's a change related to trying to cripple the use of Geforce Products for CUDA computation in profesional markets.

OT:
You got to love NVIDIA.
They cripple GeForce Cards up and down in OpenGL (performance changes from driver to driveR) to sell Quadros for professional tools, then they cripple GeForce and Quadros in their GPGPU capabilities to sell more Tesla cards and all run with the exact same chip just with different limbs broken -.-
Great, soon I don´t even need to have a fortune aside for various software licenses, I also need a few computers with different cards for various tasks.

And what can we do about it? Nothing, because all 3 product lines will continue to sell.
Stupidass Nvidia.

At least you can now use an ATI card for display and work in your 3D app with good OpenGL performance and use an NVIDIA GeForce for rendering with Octane

Cheers,
Marcus

[pixelrush]

with the rewrite then it will be possible theoretically to network pc running Octane with the same scene? ie the UI could function as a master in a wider sense than multiple gpu on the same mobo?...

[abstrax]

with the rewrite then it will be possible theoretically to network pc running Octane with the same scene? ie the UI could function as a master in a wider sense than multiple gpu on the same mobo?...

Yes it should be possible, but it's still quite a bit of work which is the reason why it probably won't be done for a while. But yes, we kept network rendering in our minds, when we redesigned the whole thing.

Cheers,
Marcus

[pixelrush]

very good. carry on then.

might be useful to have that capability - drop say a GTX460 into a spare pci-e slot on local networked pc's for a boost in rendering.
many would have enough space and power for another graphics card I think....and could probably run a cuda app in the background.

[vipvip]

Thank you Marcus for your answer.

on focusing values, you're ok, sorry, i sould did a mistake, now it works...

but for the emitive ( and non-emitive too...) animation objects, there is an heavy and BASIC problem ( yesterday i did tests xith lightwave (10.1), and today with maya (6.0) and there there's still the problem: animation of objects ( emitive or not ) doesn't function with 2.48 ( the same scene works prefectly with 2.44 cuda 3.0 version )

[face]

Thank you Marcus for your answer.

on focusing values, you're ok, sorry, i sould did a mistake, now it works...

but for the emitive ( and non-emitive too...) animation objects, there is an heavy and BASIC problem ( yesterday i did tests xith lightwave (10.1), and today with maya (6.0) and there there's still the problem: animation of objects ( emitive or not ) doesn't function with 2.48 ( the same scene works prefectly with 2.44 cuda 3.0 version )

Correct, i can confirm that. The relinking seems to be broken.

face