I was one of the original beta testers for VRay GPU—in fact, I was telling Vlado at a conference about the nVidia GPU cores and rendering at a conference many years ago.
I got out of vray because the software expanded to other modeling apps and GPU support was buggy many years ago so I had switched.
now I’m back testing out Vantage (and Omniverse) and I’m impressed. I feel F-Storm and Octane will fall behind.
but getting to my questions:
1. What are the main differences between VRay GPU and Vantage. Why even have 2 different GPU renderers? Is one strictly RTX?
2. Is there any advantage with VRay in using RTX GPUs or CUDA processor AND GPUs?
3. Do you see a roadmap in which VRay GPU or Vantage replaces each other?
im sure these have been discussed in pieces before, before there’s too much information to look through. Thanks.
The main difference is that V-Ray GPU uses offline rendering and Vantage is a Real time solution, Vantage uses near 100% raytracing and makes use of dedicated raytracing hardware on Nvidia’s Turing and Ampere GPUs
Vantage(Real time solution) is able to achieve high frame rates, it focuses on performance in all respects and offers easy, intuitive animation tools
V-Ray GPU(offline rendering) is able to achieve very high trace depth for shaders, and by default uses 100 bounces for global illumination. It is equipped with advanced shading features like RandomWalk SSS, advanced hair shading, randomization controls..etc these shading effects are expensive on performance, so they are not meant for Real time rendering(for now at least)
That being said, Vantage doesn’t really replace V-Ray GPU, both products compliment each other and in many studios are used together in the same pipeline.
This depends on the use case, do you need multi-pass rendering, advanced shading effects like hair and Random Walk SSS, high trace depth and GI depth?
Vantage is still able to produce great results, for large-scale exterior scenes specially and for the easy and fast animation rendering
For me personally, I come from a concept design and Print background. I have used V-Ray GPU in many Print studios, usually shading would usually take 60 to 70% of time and effort in Print projects, hence oneself needs shading features. Specially for massive resolutions, where the camera can be very close to a product, high trace depth and GI depth come handy as well
On the other hand, someone that spends more time on design than shading or post-production, they will spend a lot more time in Vantage. I did some Archvis in the past as well, Vantage could have been handy in many situations(it was not released yet)
V-Ray and V-Ray GPU are part of a DCCs workflow, they are deeply integrated in Maya and Houdini’s workflows for example. This is something to consider for the use case I think
No, as mentioned above the 2 products don’t really compete with each other, in some cases they are used together in the same pipeline for different purposes.
We have separate V-Ray GPU and Vantage teams
Thank you for taking this time to explain the differences. It would appear (for now) that VRay GPU has more support and would be geared towards more detailed animations or renderings, in particular scenes with extremely heavy geometry. It could be used towards character animations, etc.
Vantage appears to meet less strict requirements and is handy for things such as archviz or even some product renderings.
Is Vantage convergent/tandem in development with OmniVerse or meant to compete with it? I have tried Omniverse, and quite frankly it works but isn’t as easy to use and as fluid. I understand that part of Omniverse is to use Open Source to switch between different modeling packages. I also realize that part of their goal is to allow for remote collaboration. Is that something that Vantage plans to implement in the future?
A few more questions:
1. Will there be network rendering/distributed rendering or is that limited (for now) by most network transfer speeds?
2. Will multi-gpu systems still be limited in the foreseeable future?
3. Do you plan on implementing NVLink for 3000 series and up for added memory?
Thank you for your interest in Vantage, as well as V-Ray GPU. As Muhammed explained, V-Ray GPU is really the GPU version of V-Ray and aims to solve all the classic 3D production rendering challenges. In contrast, Vantage is all about being real-time while still 100% ray tracing, with many people finding that a perfect fit for pre-visualization and look-dev, with many people no longer working in 3ds Max without it. At the same time, others find Vantage’s real-time capture great for “play blasts” and its final (off-line) rendering sufficient for many projects. Regardless, if you need more quality or capabilities when using Vantage, V-Ray and V-Ray GPU are always there for you. Vantage is very much a companion tool to your V-Ray workflow.
In regards to your questions -
0. We are expanding our overall USD support in Chaos products as quickly as the host applications support USD themselves, with this currently being best from Houdini, then Maya, and lastly 3ds Max. We are considering having Vantage support USD directly as this overall support becomes more complete. Until then, the vrscene support from all the V-Ray products into Vantage accomplishes most of what a USD workflow would do as well (at least within Vantage). We are examining ways for Chaos products to interact with Omni, and you’ll probably see the first signs of that at GTC this month.
1. Vantage does work in headless mode today via command line, and can be automated in a network rendering manner by a queue manager, so the likes of Deadline, etc. could support it if they wish. As with V-Ray and Corona, we look to the makers of such managers to be adding support in their solutions, so don’t hesitate to request it of them.
2. We are looking at how best to support than 2 GPUs in the future with Vantage. We haven’t made this a priority due to everyone’s difficulty in purchasing GPUs for over a year.
3. We are also looking at how best to support memory pooling via NVLINK in the future as well. This priority got reduced when NVIDIA removed NVLINK from GeForce except at the top 3090 level.
1. Vantage does work in headless mode today via command line, and can be automated in a network rendering manner by a queue manage
This is a queue function - not networking rendering utilizing other GPUs on different PCs which is what I was more interested in.
2. We are looking at how best to support than 2 GPUs in the future with Vantage. We haven’t made this a priority due to everyone’s difficulty in purchasing GPUs for over a year.
Understood. Many of us still have 2080s in our pipelines and spare rendering machines/backup PCs that could be put to good production use. Please see above 1.