RandomControl's Arion Hybrid CPU+GPU renderer Web Ad 3

Arion is the hybrid-accelerated and physically-based light simulator developed by RandomControl. It inherits all our expertise in light simulation and makes it run on steroids, thanks to our very unique approach to massive CPU+GPU+LAN computation.

Arion is:
an interactive WYSIWYG editing application.
a super-high performance production renderer.

Arion's unique approach

Arion uses all the GPUs -and- all the CPUs in your system simultaneously, not wasting a single flop available. Additionally, Arion can use all the GPUs and all the CPUs in all the other computers in your network forming a cluster for massive computation.

In that regard, Arion is a grid-computing solution to the problem of light physics simulation. Simultaneous GPU+CPU+LAN, combined with our exhaustively optimized code, is what we call hybrid acceleration.




Some facts about Arion

Uses all the GPUs -and- CPUs in your system simultaneously.
Plugging an extra GPU delivers an immediate performance boost.
20x to 100x speed up per machine (hardware-dependent).
Strictly physically-based.
Fully unbiased (no cheats, no shortcuts).
Does not need any kind of precomputations.
Easy-to-use interactive WYSIWYG editing app.
The real-time viewport -is- final quality.
Production render engine features (commandline renderer, compositing channels, etc...)
Compatible with NetWarrior for hyper-fast cooperative still frame and animation rendering.

Category:

Science & Technology

Tags:

• GPU raytracing
• CUDA raytracing
• GPU
• CUDA
• NVIDIA
• NVIDIA CUDA raytracing
• raytracing
• RandomControl
• Arion
• Arion render
• fryrender
• real-time raytracing
• real time raytracing
• realtime raytracing
• interractive rendering
• interractive raytracing
• realtime

[PotadoTomado]

@anttik1979 That makes one clear frame in about 10 seconds on a cutting edge computer. In games you need to render 60 frames per second on regular consumer computers....

[tangletailpro]

@gobacktorussia continuation with hopefully less grammar errors. With game's the work is split. The real drawing happens on the GPU, which you will only see the real benefit if you have an actual card not a chipset. Because of that, the CPU has less overhead and can calculate more float-points faster for physics and locations, then share information.

Production is slower to reduce errors, and increases accuracy, because it's all on the CPU, or the GPU, work is hardly divided between them.

[tangletailpro]

@gobacktorussia The rendering engine is different compared to production based.

Game rendering is far more erroneous than production based. Their are very few poly's seen at real time, and you have many "Obstructive" geometry lying around to enforce back-face culling. In real time you only see the faces that your vision "touches"

Production based is slower because it actually calculates the lighting, poly locations, and all poly's are drawn logically to calculate light vectors.

[gobacktorussia]

@Tibike77ro of course true, i have always wondered though, how games can render so fast e.g crysis which although inferior can easily render 30 frames at high res in one second... I guess its just the level of detail in lighting and materials right?

[Tibike77ro]

@anttik1979 Because you can't play a game at "one frame every 2 to 5 minutes" :)

[Tektechno]

absolutly amazing !!!!!!

[anttik1979]

omg! Why dont you let that engine for game devs to have a game that good looking !