Hi,
Does the Titan RTX have the same enhanced drivers for the Maya viewport as would be found in the RTX Quadro cards?
Is there any benefit to having a RTX Quadro 5000 vs a Titan RTX. I'm doing Maya animation, May Rendering in Arnold and Octane. I will also use Daz Studio.
From what I've read it seems as if Maya rendering in Quadro will provide similar speeds as rendering in a general RTX card, but it's the viewport response where there's a speedup when using the enhanced drivers. Not sure how true this is but wanted to get your thoughts.
I'm a noob so no flames please ๐
Thx
Hollaback
Hi!
Drivers for pro cards like NVIDIA Quadro - series are optimized for pro applications like Maya, so you can expect better overall speed/performance, feature support and stability.
Appreciate the reply mspeer!
I heard the previous generation of Titan cards did receive the enhanced (i.e., Quadro-like) version of the drivers which made the viewport experience faster than a normal GTX card. I'm wondering if this version of the Titan RTX also received the same driver enhancements as well
@mspeer wrote:
Hi!
Drivers for pro cards like NVIDIA Quadro - series are optimized for pro applications like Maya, so you can expect better overall speed/performance, feature support and stability.
Hi!
"I heard the previous generation of Titan cards did receive the enhanced (i.e., Quadro-like) version of the drivers which made the viewport experience faster than a normal GTX card"
- I never heard of this before and i was not able to find anything from NVIDIA stating this.
Actually I just did a search and actually see a post from Nvidia about it. I hadn't seen this previously. Only heard it on YouTube etc, previously. I listed 2 links below but will try to dig up other links later. Please feel free to let me know if I'm interpreting the info correctly. Thx!
1. Actually found an announcement about it from Nvidia back in July 2017
- TITAN Xp Enables New Levels of Performance for Creatives, July 31, 2017 by MATT WUEBBLING
- https://blogs.nvidia.com/blog/2017/07/31/titan-xp-drivers-new-levels-of-performance-for-creatives/
2. Gamers Nexus
- Best Workstation GPUs 2018: Premiere, AutoCAD, VRay, Blender, & More
- Me mentions it specifically (for last gen of Titan) at 4:25 in this video)
- he also shows Maya specviewperf viewport performance of Titan XP vs. Quadro P6000 at 2:54
- https://www.youtube.com/watch?v=gKBLrkZtIVk
3.
Hi!
1. For nearly every benchmark delivering a certain result you will find an other delivering a complete different result.
2. TITAN Xp "Our latest driver โ available today โ delivers 3x more performance in applications like Maya"
How was the performance before?
For me this looks more like, "oh, the performance for certain applications was so bad that something needed to be corrected".
Such in increases in performance you see mostly for very new cards with new technology and for me it sounds more like marketing.
I also found this, looks like marketing mixed up some information:
"TITAN Xp is purpose-built to deliver up to 3X the performance of previous-generation graphics cards*"
Maybe also a (marketing) reaction to the AMD enthusiast series.
3. I can't see any comparison at your links between a Gaming / Titan RTX card and the Quadro counterpart.
So i stick with my first statement / reply at this topic.
I too noticed this as soon as they released the new line of Quadro RTX cards. I think the price of the Quadro 5000 RTX is a joke. 25.00 extra dollars for the Titan RTX gets you 8 more GB of GDDR6 ram, 1536 more CUDA cores, 192 more tensor cores, and a higher clock speed. The only difference is the Quadro RTX can use a 50GB bandwidth NVLINK Bridge, that enables you to have two QUADRO cards with shared memory (for as little as 4950.00(two cards) + the cost of the NVLINK connector) still only giving you 8 more GB of ram compared to the Titan. The Quadro RTX 6000 card should be priced at 3200.00 us dollars. I would not purchase the Quadro RTX 5000
@mspeer wrote:
Hi!
1. For nearly every benchmark delivering a certain result you will find an other delivering a complete different result.
2. TITAN Xp "Our latest driver โ available today โ delivers 3x more performance in applications like Maya"
How was the performance before?
For me this looks more like, "oh, the performance for certain applications was so bad that something needed to be corrected".
Such in increases in performance you see mostly for very new cards with new technology and for me it sounds more like marketing.
I also found this, looks like marketing mixed up some information:
"TITAN Xp is purpose-built to deliver up to 3X the performance of previous-generation graphics cards*"
Maybe also a (marketing) reaction to the AMD enthusiast series.
3. I can't see any comparison at your links between a Gaming / Titan RTX card and the Quadro counterpart.
So i stick with my first statement / reply at this topic.
Appreciate your info mspeer.
I submitted a questions to Nvidia online chat earlier. Rep said he will have to check with colleagues and will get back to me by Monday. Will post whatever I hear back.
@Anifex09 wrote:
I too noticed this as soon as they released the new line of Quadro RTX cards. I think the price of the Quadro 5000 RTX is a joke. 25.00 extra dollars for the Titan RTX gets you 8 more GB of GDDR6 ram, 1536 more CUDA cores, 192 more tensor cores, and a higher clock speed. The only difference is the Quadro RTX can use a 50GB bandwidth NVLINK Bridge, that enables you to have two QUADRO cards with shared memory (for as little as 4950.00(two cards) + the cost of the NVLINK connector) still only giving you 8 more GB of ram compared to the Titan. The Quadro RTX 6000 card should be priced at 3200.00 us dollars. I would not purchase the Quadro RTX 5000
Thanks @Anifex09. The way the product line is structured, its a bit confusing to me right now. But I'm sure I'll see the light as I learn more.
Also, are specviewperf scores considered to be an accurate measure of how well the card performs?
Gamers Nexus did a benchmark of the Titan XP vs Quadro P6000 on Maya Viewport performance and the Titan XP had better results (371 to 290).
See 3:17 of https://www.youtube.com/watch?v=gKBLrkZtIVk
Also attached as a png as well.
@Anonymous wrote:
@mspeer wrote:
Hi!
1. For nearly every benchmark delivering a certain result you will find an other delivering a complete different result.
2. TITAN Xp "Our latest driver โ available today โ delivers 3x more performance in applications like Maya"
How was the performance before?
For me this looks more like, "oh, the performance for certain applications was so bad that something needed to be corrected".
Such in increases in performance you see mostly for very new cards with new technology and for me it sounds more like marketing.
I also found this, looks like marketing mixed up some information:
"TITAN Xp is purpose-built to deliver up to 3X the performance of previous-generation graphics cards*"
Maybe also a (marketing) reaction to the AMD enthusiast series.
3. I can't see any comparison at your links between a Gaming / Titan RTX card and the Quadro counterpart.
So i stick with my first statement / reply at this topic.
Appreciate your info mspeer.
I submitted a questions to Nvidia online chat earlier. Rep said he will have to check with colleagues and will get back to me by Monday. Will post whatever I hear back.
Hey @mspeer ... apologies that I missed the 3rd point from your earlier post.
The review I posted didn't do a direct compare of Titan RTX to Quadro RTX. They said one is coming but it's still not yet available. Apologies for any confusion.
I was trying to use data from the previous gen comparisons of the Titan XP to the Quadro P6000 to help me forecast what might happen with this current gen. I know that's probably dangerous.
Their comparison of the Titan XP on specviewperf had it beating the Quadro P6000 on Maya viewport specs.
3:17 of https://www.youtube.com/watch?v=gKBLrkZtIVk
I know you already weighed in with your previous comments but I didn't want you to think I ignored your last point.
thanks again for your help.
@alex.motion wrote:
Titan RTX no doubt !
Thanks @alex.motion ... and I hope that's the case since I already bought it LOL!
I have been buying Quadro Cards for several years now. I use Maya, Aftereffects, Nuke, Motionbuilder, Speedtree, ZBrush, and other heavy GPU demanding application. My reasons for choosing Quadro over gaming cards was Memory, Build quality, and drivers. The RTX Titan has the same build quality as the Quadro and has 8 more GB of RAM. We recently built a GPU render machine for small projects using 4 RTX Titans. The one thing that will allow you to have several 20k textures in a scene with millions of polygons it the onboard RAM. The more you have the longer the card will remain relevant.
Hi!
"Titan XP on specviewperf had it beating the Quadro P6000 on Maya viewport specs"
- "beating", by 300 fps vs 290 fps?
I would say they show overall the same performance for this benchmark.
But keep in mind that's just one average value for one specific test, with a really small feature set used compared to what Maya offers.
How often has the test been repeated?
What's the resolution used?
Does this test really reflect settings, scene elements, screen resolution(s) and UI/windows, as used by you in production?
- For me the answer would be clearly ... "no".
Firepro cards are much faster in Mayas virewport than Quadro cards. So I dont know why you are looking at Quadro at all (and Amd is much cheaper). Quadros are faster in 3ds Max. So iff you are working in Max, Quadros are better. But you have to rob a bank to buy them. From my perspective of views...
To moderator: this forum on mobile is falling appart.
Hello
i am interested in 3x/4x Titan RTX build. My vendor is advising installing only two.
can you share the details of how you have setup your system ?
Last year I had to buy a quadro rtx 6000 (together with a dual xeon, 28 cores). This was not my decision (and not my money). The 18 Maya years before that, I always had a geforce type of card.
what I experienced:
- rendering: ~300 instances with about 200000 coordinates and together about 3 billion triangles. arnold gpu mode (maya2020) had not the slightest chance against the cpu mode. It took like 10 minutes before the render started (data transfer), then about 30 minutes render time. But actually the system never returned from rendering and I had to kill Maya. In cpu mode, it took about 5 minutes and all was fine.
- viewport rendering: I did not run the relevant scene with my old geforce (gtx 980, 8 GB), but comparing the current scene with an old one which used the geforce, I would say the interaction speed may have doubled. But it is not fast enough for true interactivity. It would have to be like 15 times faster ( which means its still not possible to have all instances visible)
Can't find what you're looking for? Ask the community or share your knowledge.