I found the best graphics card for Vegas Pro 12.0

JohnnyRoy wrote on 4/2/2014, 6:14 AM

I know that Vegas Pro 13.0 is around the corner and maybe SCS will upgrade their GPU support to use with newer Kepler and ATI architectures but for the time being, the questions that is still on everyone’s mind is, “What’s the best GPU for Vegas Pro 12.0?” I think I may have found an answer.

I recently purchased a 2008 Mac Pro 2.8Ghz 8-Core Xeon with 16GB memory, 128 SSD boot, 2TB RAID 0, and ATI Radeon HD 5870 on eBay for only $740! I configured it to dual-boot with BootCamp and Windows 7 Pro 64-bit and I was really please with the performance I was getting in Vegas Pro 12.0. So much so that I thought I would perform some benchmarks against my Intel Core i7-3930K Sandy Bridge-E 3.2GHz with 16GB memory, 256GB SSD, and Quadro 4000. I was shocked by what I found. (I also posted this on the Creative COW so apologies to those of you who read booth forums. This is a summary of several of my COW posts.)

Here is my little render test: rendertestjr.veg (Vegas Pro 12.0 only)

The purpose of this test is to measure the difference between GPU acceleration for both timeline playback and rendering. The project is 15 seconds long. It has two tracks:. The lower track contains a Generated Media NTSC Color Bars that rotates 360 degrees in 15 seconds. The upper track had Generated Media Noise Texture with the Progress animated so that it would move. I added Sony Bump Map and Sony Glow (both GPU accelerated FX) to the noise texture and the composite level of the event was dropped to 60% so that the rotating colors bars would show through. Compositing random movement ensures that every frame would need to be rendered during the test. This project requires Vegas Pro 12.0 to open.

In order to test on your computer be sure to do the following:

(1) Set your RAM Preview to 200MB (the default) so that frame cacheing doesn’t skew the results.
(2) Set your preview window to Best(Full) and resize to 900x506 for measuring playback fps
(3) Don’t forget to restart Vegas Pro between turning the GPU on and off
(4) For Sony AVC use the Internet 1920x1080-30p template (turn GPU on/off as appropriate)
(5) For MainConcept AVC use the Internet HD 1080p template (turn GPU on/off as appropriate)

That’s it for the ground rules. Now for the results…

ATI Radeon HD 5870

With GPU turned OFF, that project played back at 0.5 fps on my 8-core Mac Pro with Bootcamp (btw, even on my new 6-core/12-thread Intel Core i7-3930K it only played at 0.7 fps with no GPU). With GPU turned ON it plays back at 29.97 fps. So the GPU is giving me a 60x improvement in playback (from 1/2 frame per second to 30 frames per second). That's pretty amazing! :-)

Here are the timings that I recorded for the Radeon:



-------------------------------------------------------

                  ATI Radeon HD 5870

-------------------------------------------------------

Timeline GPU Acceleration OFF (Playback 0.5fps)

-------------------------------------------------------

Sony AVC Internet 1920x1080-30p   (CPU Only) . . . 1:34

Sony AVC Internet 1920x1080-30p   (OpenCL) . . . . 1:30 (0x)

MainConcept AVC Internet HD 1080p (CPU Only) . . . 2:40

MainConcept AVC Internet HD 1080p (OpenCL) . . . . 1:23 (2x)

-------------------------------------------------------



-------------------------------------------------------

Timeline GPU Acceleration ON (Playback 29.97fps)

-------------------------------------------------------

Sony AVC Internet 1920x1080-30p   (CPU Only) . . . 0:31

Sony AVC Internet 1920x1080-30p   (OpenCL) . . . . 0:29 (0x)

MainConcept AVC Internet HD 1080p (CPU Only) . . . 0:54

MainConcept AVC Internet HD 1080p (OpenCL) . . . . 0:15 (3.6x)

-------------------------------------------------------

Apparently, Sony AVC does not take advantage of OpenCL with this card. There was no appreciable improvement (just a few seconds) between CPU Only and Use OpenCL. Maybe if I make the test project longer (like one minute) but the GPU Load meter did not show the Sony encoder using the GPU at all (maybe one or two spikes).

MainConcept was another story entirely. The difference with MainConcept AVC GPU is incredible. MainConcept AVC saw a 3.6x improvement for GPU rendering and 10x overall between no GPU for timeline or render (2:40) and only 0:15 with both turned on!!! On the MainConcept AVC render my CPU was about 29% and my GPU was about 71% utilized. That is closer to what I expected.

At no time did the GPU slow things down. It's important to note that the CPU and GPU are pretty evenly matched. They are both circa 2008/2009 so they compliment each other well so I'm not sure how much having the CPU/GPU combination evenly matched adds to the performance boost. I did use the latest Catalyst drivers from AMD and not the older Apple drivers from BootCamp (which didn’t recognise a GPU with Vegas Pro).

NVIDIA Quadro 4000

Then I set out to do the same tests with my Quadro 4000. Remember, the Quadro 4000 is in my fairly new Intel Core i7-3930K Sandy Bridge-E 3.2GHz (6-core/12-threads) so the CPU Only times are better than my 2008 Mac Pro 2 x 2.8 GHz Quad Core Xeon E5462 (8-cores/8-threads) but much to my surprise, the GPU renders were actually slower on the NVIDIA Quadro 4000 than the ATI Radeon HD 5870!

Here are the times for the Quadro 4000:



-------------------------------------------------------

                  NVIDIA Quadro 4000

-------------------------------------------------------

Timeline GPU Acceleration OFF (Playback 0.7fps)

-------------------------------------------------------

Sony AVC Internet 1920x1080-30p   (CPU Only) . . . 1:10

Sony AVC Internet 1920x1080-30p   (CUDA) . . . . . 1:08 (0x)

MainConcept AVC Internet HD 1080p (CPU Only) . . . 1:17

MainConcept AVC Internet HD 1080p (CUDA) . . . . . 1:07 (0.1x)



-------------------------------------------------------

Timeline GPU Acceleration ON (Playback 2.0fps)

-------------------------------------------------------

Sony AVC Internet 1920x1080-30p   (CPU Only) . . . 0:27

Sony AVC Internet 1920x1080-30p   (CUDA) . . . . . 0:36 (0x)

MainConcept AVC Internet HD 1080p (CPU Only) . . . 0:40

MainConcept AVC Internet HD 1080p (CUDA) . . . . . 0:33 (0.2x)

I'm not even sure what's going on here. First I was shocked that the Quadro 4000 could not play the timeline better than 2 fps with GPU ON when the Radeon HD was a solid 29.97fps at Best(Full) all day long. This is why the render times are so poor. My Radeon HD 5870 rendered MainConcept AVC with GPU twice as fast as the Quadro 4000! (0:15 vs 0:33). I'm beginning to think that I should be using my 2008 Mac Pro as my primary Vegas Pro editing workstation because the Radeon performs better than the Quadro with Vegas Pro 12.0.

Obviously this was an artificial test. Generated Media is uncompressed in Vegas Pro so I eliminated any lag due to decoding video on the timeline. But it did isolate just the timeline GPU acceleration and rendering GPU acceleration in a project that doesn't play back smoothly without GPU on any PC.

Real-world Tests

So how about real-world rendering? I thought I would test this with Sony’s own “Red Car” test that is posted on their GPU Acceleration page. So I downloaded the 2GB project and ran it against the two computers. I should mention that Sony did their measurements with the preview set to Best(Full) at 900x506 also (which is why I used that in my previous tests) but they used different render templates so please read the Sony instructions and use the same render templates as below.

Here’s what I measured with the “Red Car” project:



------------------------------------------------

           SONY “Red Car” Project

------------------------------------------------

NVIDIA Quadro 4000 w/Core i7-3930K 3.2Ghz 6c/12t 

------------------------------------------------

MainConcept AVC Internet 1080-30p . . . . . 1:34 (94 seconds)

XDCAM EX HD 1920x1080-60i 35Mbps  . . . . . 1:42 (102 seconds)



------------------------------------------------

ATI Radeon HD 5870 w/Xeon 2.8Ghz 8c/8t

------------------------------------------------

MainConcept AVC Internet 1080-30p . . . . . 0:57 (57 seconds)

XDCAM EX HD 1920x1080-60i 35Mbps  . . . . . 1:15 (75 seconds)

What amazed me is that Sony's "Red Car" project plays back at full frame rates 29.97 Best(Full) all the way through using the Radeon HD 5870 but the Quadro 4000 could not maintain the 29.97 fps throughout the timeline like the Radeon HD 5870 did. Again quite shocking to me, because the Radeon is in a slower computer but I guess it might have more to say about Sony making better use of OpenCL with ATI cards than NVIDIA cards for timeline GPU acceleration.

Conclusions

Observation 1: GPU acceleration absolutely works! I got a 10x improvement with timeline GPU on + MainConcept AVC

Observation 2: It appears that Sony makes better use of OpenCL than CUDA, therefore ATI cards give better GPU performance than NVIDIA

Observation 3: The ATI Radeon HD 5870/6970 is about the fastest GPU that works with Sony’s requirement for older cards. (newer cards are actually slower with Vegas Pro 12.0)

From these tests I have concluded that the ATI Radeon HD 5870 is the sweet spot for GPU acceleration in Vegas Pro. Of course the 5870 is no longer available but the Radeon HD 6970 is a newer model of this card and still for sale. Nothing more powerful (i.e., 7xxx series+) will work as good.

Now I realize that the Quadro 4000 isn’t the fastest card from the Fermi line up. I’d be interested in hearing what timings others are getting with their GeForce cards. Maybe there is one that's faster than the Radeon HD 5870. I'm sure everyone would like to know.

I know we’ve been “benchmarked out” on this forum but I thought having a benchmark that we can run in Vegas Pro 12.0 and then in Vegas Pro 13.0 when it is released will tell us if Vegas Pro 13.0 has improved anything in the GPU dept.

Bottom line for me: I got a whole 8-core/16GB Mac Pro with Radeon HD 5870 for less than my $800 Quadro 4000 alone and it performs better than my current computer in real world playback and render tests with Vegas Pro 12.0. I don’t think I would ever waste my money on a Quadro card again.

~jr

Back to post

Comments

Cliff Etzel wrote on 4/14/2014, 6:39 PM

I ran the Red Car project on my main workstation - I only got about 20fps on average with the full/best in preview mode - sometimes dropping below 12fps

I ran JR's rendertest veg - I was at best 7fps :-(

I'm somewhat frustrated by this as my specs, although not cutting edge by any means, aren't too shabby for editing Canon DSLR footage. The Red Car project is nowhere near the level of complexity of content I produce so I'm trying to figure out where in my pipeline is the bottle neck. Given what JR has posted initially, I'm beginning to wonder if it isn't my nVidia GTX550Ti graphics card.

I hate saying it, but PPro CS6 plays back smooth as silk (I"m guessing due to the optimized code for CUDA) - wishing Vegas Pro 12 could compete head to head. Instead I'm left right where I was back when I switch to Adobe's NLE 2 years ago

ANYONE?!

OldSmoke wrote on 4/14/2014, 6:59 PM

Cliff.

I see you are using one of the 3xx.xx drivers, have you tried changing to 296.10? That is one of the fastest drivers for Nvidia cards.; the difference can be as much as 30% in performance. You must however completely remove the current driver. The 550Ti is an ok card for Vegas but not as good as higher 500 like the 560Ti up to 580. AMD cards seem to have better preview performance, 6950 and 6970 are reported to have better timeline performance then Nvidia; I personally haven't tested it.
Another driver is 334.89 with a "tweak" in the Vegas Internal settings; here is a link to a previous post of mine: http://www.sonycreativesoftware.com/forums/ShowMessage.asp?Forum=4&MessageID=885329

If you want to stick with Nvidia, here is a link that will show all the different 500 series and their configuration http://en.wikipedia.org/wiki/Comparison_of_Nvidia_graphics_processing_units#GeForce_500_Series. The general rule is more CUDA cores is faster and for VP11&12 use only FERMI based cards.

GJeffrey wrote on 4/15/2014, 7:54 AM

I've been tested the veg files below on my system (I7-3930K + GTX570)
GPU ON
- Playback 17fps
- Render Sony AVC GPU 21s
- Render Sony AVC CPU 19s
- Render MC CPU 36s
- Render MC CUDA 15s

GPU OFF
- Playback 0.6fps
- Render Sony AVC GPU 1min02s
- Render Sony AVC CPU 1min03s
- Render MC CPU 1min16s
- Render MC CUDA 1min02s

So it confirms that GPU acceleration on timeline preview is better with a ATI card. The difference between the Quadro 4000 and GTX570 is quite huge though.
But with a very similar system as John I have faster render...

OldSmoke wrote on 4/15/2014, 9:07 AM

I am quite tempted to put an additional 6970 together with my 2x GTX580; I just wonder what kind of headache that will give me during installation.

NormanPCN wrote on 4/15/2014, 10:13 AM

So it confirms that GPU acceleration on timeline preview is better with a ATI card. The difference between the Quadro 4000 and GTX570 is quite huge though.

Vegas only uses OpenCL for its video engine. This is used in both playback and encoding.
The Sony AVC and MC AVC encoders both have CUDA versions for their use of GPU on Nvidia.

It seems that Nvidia has problems implementing good performance in OpenCL.

Cliff Etzel wrote on 4/15/2014, 2:05 PM

Old Smoke - I uninstalled the current driver and replaced with the 296.10 driver per your suggestion. I did see a better imrpovement in performance so I"m glad to see that helped. Are there any other tweaks I should apply to the Internal Preferences? I did increase the meomry allocation for 384 to 768 - it seemed to help a little but unsure if I should allocate the full 1GB memory on the card - your thoughts? I'm beginning to think I need to look for an AMD card, but that doesnt' help me on my Laptop since it has a Quadro FX880M card. Would rolling back to the 296.10 drivers make a difference in performance as well on my Dell Precision M4500 laptop?

VidMus wrote on 4/15/2014, 3:35 PM

OldSmoke said, "I am quite tempted to put an additional 6970 together with my 2x GTX580; I just wonder what kind of headache that will give me during installation."

A HUGE driver headache! But I am sure you already knew that.

JohnnyRoy wrote on 4/15/2014, 4:47 PM

> "JohnnyRoy - is there really that big a performance difference in your testing that points to AMD being a better solution for GPU/performance as a general rule for Vegas Pro 12?"

I'll tell you honestly Cliff, I think general rules are hard to extrapolate from this data. This week someone posted some footage on the COW from a Blackmagic ATEM Switcher that was bogging down Vegas Pro. It was a QuickTime MOV file and strangely enough the codec only said QuickTime in Vegas but in QuickTime Player is said AVC/H.264.

I tested it on my 8-core Xeon with ATI Radeon HD 5870. There is no GPU activity at all and I was getting 6-7 fps with only 25% CPU load. Clearly the QuickTime playback wasn't using more than 2 of the 8 cores. :(

On my 6-core/12-thread Core i7 with NVIDIA Quadro 4000 I was getting a respectable 24 fps at Preview Auto and 20 fps at Best Full. That's not bad. The CPU is again at only 25% but my GPU is at 45% which is why I'm getting better frame rates.

So it appears that for this particular footage (QuickTime not identified as AVC), Vegas isn't using my ATI card at all but it is using my NVIDIA card. How do you make any general rules when all the rules are broken going from one media file to the next? This GPU stuff just gets more and more confusing every day.

~jr

Cliff Etzel wrote on 4/15/2014, 7:02 PM

VegasPro is a frustrating piece of software for me. I've said it before and I'm saying it again - I love the concept, the problem is the execution - that being consistency in performance. As you've pointed out JR, you discovered by accident that an older video card performs better than a so called "professional" video card. What else is missing in the equation that we don't' know about?

Granted, my nVidia GTX-550ti video card is considered so so, but the reality is at the time I purchased it, it was a decent card. Now I learn that AMD provides a better experience performance wise, yet all my other software more or less doesn't' take advantage of the Catalyst drivers. I will say that I was able to get more or less full frame rate if I dropped the quality to Best/Half setting for the Red Car project in VP12 using the 334.89 nVidia drivers Old Smoke Recommended. Not bad TBH. I also made changes to the OpenCL memory config to the full amount of ram on my card - from 384 to 1024 and that seemed to be the ticket for getting the Best/Half performance to improve. Even at Preview Full setting, I still could not get full frame rate.

So I'm left with the fact that my newest documentary project that I'm about to begin production on will be edited once again in Premiere Pro CS6/Audition CS6/Photoshop CS6.

Should I look at transcoding everything to either Cineform or DNxHD thereby leveling the playing field with regards to footage on the timeline? All my footage is going to be Canon DSLR acquired footage.

If I thought for a moment I could confidently rely upon Vegas Pro for the project I would but given this revelation of issues related to performance, I have my doubts. Then again, maybe VP13 when it's released will have made the changes necessary to allow me to reconsider SCS as my primary post production software company. For now, Adobe, even though it takes more steps to accomplish the same thing compared to Vegas Pro, will be my default set of tools.

Any thoughts?

OldSmoke wrote on 4/15/2014, 7:33 PM

Cliff

Sorry, but the GTX 550Ti (192 CUDA cores) isn't even as good as an older GTX460 (336 CUDA cores). To see better timeline performance from a Nvidia cards you need a good 400 cores and more. Memory bandwidth is also important. Did you have a look at the Wikipedia link?

mdindestin wrote on 4/15/2014, 8:55 PM

And I suspect Cliff, that you're probably not enamored with the color correcting/grading/matching tools in Vegas as compared to CS6 and FCPX.

Cliff Etzel wrote on 4/15/2014, 10:00 PM

Old Smoke - yes I looked at the info - finny thing is, PPro CS6 doesn't exhibit any of the performance issues that Vegas Pro does so I haven't felt any need to upgrade to a newer card. Seems if I want to use Vegas Pro, it's going to cost me and I'm someone who won't upgrade unless I have a truly compelling reason - especially since PPro CS6 works pretty well with the card. But then again, it seems the more CUDA cores, the even better performance - especially since I'm considering using grading software - either Speedgrade CS6 or Resolve Lite.

Guess it means weighing the pros and cons

mdindestin - you're right, the grading tools in CS6 offer alot more IMO compared to Vegas Pro, but I guess using software like Resolve Lite is the workaround for that.

Would transcoding to an intermediate like Cineform or DNxHD alleviate the performance issues on the GPU side of things?

OldSmoke wrote on 4/15/2014, 10:08 PM

Cliff

If you are in the US, eBay is your friend. You can get a GTX580 for $150.00 or a HD6970 for similar or less. I am not sure, but doesn't PPro use proxies? The part I don't like about PPro is rendering speed which is far behind Vegas on my machine. Also HD - DVD works better with Vegas in my opinion. The whole editing experience in Vegas is so much simpler and intuitive even less restrictive. There are plenty of grading tools that work well with Vegas if the build in ones aren't sufficient for your work.

Cliff Etzel wrote on 4/15/2014, 10:18 PM

Old Smoke - I"m not sure TBH about PPro using proxies - I think if you do a render of the timeline because of using non-RT plugins like Magic Bullet Looks, it creates proxie mpeg2 files for RT playback. I don't typically use MB as I find it slows down the editing experience.

Given your recommendation, if it were you, which card would you go with? I've only used nVidia so I have no clue about AMD, and as JR has pointed out, the variables with hardware vary to such a large degree, 2 identical computers typically behave differently with the same software installed.

Grading tools - Yeah - I've been looking at Resolve Lite but that's a whole different animal in post production.

ushere wrote on 4/15/2014, 10:39 PM

@ oldsmoke - whilst i admire and appreciate your contributions and views, along with your well proven case for the 580, and old drivers for it. i do think it's time we moved on....

if nearly every other nle is outperforming vegas in gpu, timeline performance why are we being directed to obsolete technology in order to get the best out of vegas?

i run vegas alongside cs6 and other software which takes advantage of new advances in gpu technology and 580 would be a huge step backwards.

please do not take this as criticism, i really do understand where you're coming from...

Cliff Etzel wrote on 4/15/2014, 11:19 PM

Leslie - That's my thinking as well - albeit the fact I can run PPro CS6 with my current low end card and have it utilize the GPU and be in what appears to have better performance compared to Vegas just has me confounded.

I really have grown tired of the constant hardware upgrade cycle and I'm sort of resisting the notion of upgrading unless it will provide a marked improvement in the overall editing experience. As JR has pointed out - his expensive Quadro card was lackluster compared to an older AMD card. TBH, if I were to upgrade the card to that level, it would be directly because of Premiere Pro CS6 - which has been shown to improve the editing performance especially with plugins like Magic Bullet Looks, not Vegas.

Seems a catch-22 situation.

ushere wrote on 4/15/2014, 11:48 PM

cliff, i was happily running with my 550 when i was offered a 650 for next to nothing (great having nerdy friends who constantly upgrade;-)).

ppr saw an immediate improvement, not great, but obviously a step forward (wished he'd bought a 680...).

i neither like, nor want to be constantly looking to hardware upgrades to improve performance. vegas suffers from a vfw base that is so obsolete it's beyond a joke. that said, i still love vegas for editing, and though i've tried the rest i still keep coming back to it.

OldSmoke wrote on 4/16/2014, 9:46 AM

Maybe I come from a different place. When VP11 was first released I immediately jumped onto it as my current system at that time, VP10 with a GTX460, was still not what I expected in terms of timeline playback. With VP11 I immediately could get Good/Full with all the FXs applied, "32bit video levels only" was still a no go. About a month after VP11 was released, I studied the SCS GPU acceleration page, the documentation that was included with their benchmark project and decided to upgrade to a GTX570 because that is what SCS used to do all their testing and I am a Nvidia guy. The GTX570 at that time was a fairly new card and I was so glad I went that way.
So for me, I never had the feeling of buying old hardware. Last year when the prices on GXT580s finally dropped down to $150 for a used one I jumped on it and got two GTX580 and all the hardware to get it watercooled. I sold my GTX570s, I bought a second one earlier last year, and the total upgrade was only $150 for two watercooled 580 in exchange for two air cooled 570s.
I personally don't care whether a hardware is outdated or not as long as it supports the software or the software requires it then that is the way to go. Why would I try and change something that is out of my control, like demanding from SCS to support 600/700 series, when there is a solution that I can control. In the past three years, GPU acceleration has been my friend all the way and I would have dearly missed it would I have insisted only to buy the latest hardware for my system.

Cliff
FYI, depending on your motherboard and other hardware you have in your system, you can either sell your 550 or have both cards in the system to get even better performance or support an additional monitor.

BruceUSA wrote on 4/16/2014, 10:28 AM

My obsolete cards AMD 6970 will stay with me with V11& V12. Its performs just like the way I like it, good render times and smooth playback preview at full frame rate. Try that on the new card with 1000+ large still image from a 5D mark III (no resize) on the timeline, see what happen?.. http://www.sonycreativesoftware.com/forums/ShowMessage.asp?

Cliff Etzel wrote on 4/16/2014, 4:43 PM

Old Smoke - would you recommend getting a 570 card? My current 550ti drives my 23" monitor and the second display is running off of the HDMI port on the card.

I guess I"m at a point now where I'm tired of trying to keep up with hardware changes and just want a solid machine to provide the performance I need/want when editing DSLR video footage.

OldSmoke wrote on 4/16/2014, 4:49 PM

Let's wait for VP13. Maybe it will finally support 600/700 series cards but I personally wont keep my hopes high.
GTX570 and 580 are almost the same price on eBay and Asus cards are always a bit more expensive. I had 2 ASUS GTX570, those where 3 slots wide and got 2x EVGA GTX580 because these are only 2 slots wide and once watercooled are only 1 slot wide.
The 580 is another 10-15% faster compared to the 570. What kind of Power Supply and motherboard do you have?

Edit: there are some really good ones on eBay! http://www.ebay.com/itm/EVGA-NVIDIA-GeForce-GTX-570-1-25-GB-GDDR5-SDRAM-PCI-Express-x16-/231201054279?pt=PCC_Video_TV_Cards&hash=item35d4a81e47

Cliff Etzel wrote on 4/16/2014, 6:39 PM

MoBo - EVGA X58 Series running a Xeon 3520 processor with 18GB Triple Channel RAM

Power supply is a Corsair 750

OldSmoke wrote on 4/16/2014, 9:20 PM

Cliff

That is a great board, much better then the newer Z68 and later ones. The reason I am saying this is because you can actually run 2 graphic cards at x16 which is not possible on the socket 1155. So keep the 550 and put in another 570... depending on what VP13 has to offer. Corsair is a good brand too and 750W should be sufficient to run both cards.

NickHope wrote on 4/19/2014, 12:40 AM

So, now that it seems that VP13 isn't much different in terms of GPU performance from VP12, it seems that one of these or one of these is currently the "best" graphics card for Vegas.

Can I run 3 screens simultaneously (in Win 8.1) off that one card (using the 1 x HDMI and the 2 x DVI sockets)?

1 2 3 4 … 5

of 5