Vegas Pro 12 slower than Vegas Pro 11 using Sony Vegas benchmark project

[Norman Black]

[Stephen Crye] “* SVP12 render CPU only – 7:35 (horrible!)
* SVP12 render with OpenCL selected in the render – 8:40 (more horrible!)
* SVP12 render with CUCA selected in the render 3:22 (pretty bad, but at least the GPU was being used more)”

I have some questions about this. The Mainconcept AVC OpenCL encoder does not support Nvidia. It only supports AMD. You use CUDA for Nvidia. This is documented.

The question is that your CPU only and OpenCL encodes should both be same time, and they are not, since CPU would be used when you selected OpenCL. This is assuming you changed nothing else other than the render as selection. e.g. Same GPU option in your video prefs. So the thing here is consistency of results.

The Vegas video composite and effects engine always uses OpenCL for both AMD and NVidia. This is independent of Mainconcept AVC OpenCL/CUDA options. The benchmark project uses lots of effects that are GPU accelerated.

[Steve Rhoden]

The integration of GPU / OpenCL / CUDA with NVIDIA or AMD Card….it’s
just a complete mess right now.

Steve Rhoden
(Cow Leader)
Film Editor & Compositor.
Filmex Creative Media.
https://www.facebook.com/FilmexCreativeMedia
1-876-461-9019

[Dave Osbun]

What’s the CPU in that system? Is it an i7? If so, is it a Sandy Bridge or Ivy Bridge chip?

Just curious……

Dave

[Stephen Crye]

Hi Norman;

During CPU only render with Mainconcept, there was GPU activity. In fact, during the 1st run after it was done and I looked, I thought I had specified some kind of GPU assist, and was surprised to see that I had three choices and it had defaulted to CPU only. My second screen shot in the original post showed that GPU activity when CPU only was selected in the render template.

So, I thought “well, I should get better numbers with OpenCL selected, after all OpenCL is hot stuff.” So, I was very puzzled that it ran slower and showed less GPU than with CPU only!

When you said “…The Vegas video composite and effects engine always uses OpenCL for both AMD and NVidia.” , do you mean during preview or render?

Thanks for your reply;

Steve

Win7 Pro X64 on Dell T7500, MultiTB SATA, 8GB RAM, nVidia Quadro 2000, Vegas 12, 11, 10, 9 DVDA 6.0 & 5.2(build 135) Sony HDR-CX550V, Panasonic GH3 with LUMIX G X VARIO 12-35mm / F2.8 ASPH, LUMIX G X VARIO 35-100mm / F2.8

[Stephen Crye]

Hi Dave;

Thanks for your interest.

Let me ‘splain something about my “new” computer. It was an end-of-sale Dell T7500 clearance priced to sell back in Nov. 2012. It sat in its box in my garage for more than a year until Dec 2013. I had no time for video fun last year because my wife had been very ill, nearly died, but the surgeons fixed her up; she is now disabled and can’t drive or walk much, but is happy and able to eat food again – yaay!

So, I finally set it up last month. “New” to me, but old in Internet time, as will be evidenced by the specs below.

I really appreciate your taking the time to reply; I’m looking to speed it up without buying a new system – I am hoping John will comment on Quadro 2000 vs Quadro 4000 because I might be able to get a K4000 cheap.

Tonight I replaced my WD Blue 500 GB c: drive (cloned it) with a new Corsair Neutron 480GB GTX. I know this won’t help Vegas but man! Talk about fast boot and program loads. It is freaky fast.

Steve
———–
CPU-Z TXT Report
————————————————————————-
Binaries
————————————————————————-
CPU-Z version 1.68.0.×64
Processors
————————————————————————-
Number of processors 1
Number of threads 4
APICs
————————————————————————-
Processor 0
— Core 0
— Thread 0 0
— Core 1
— Thread 0 2
— Core 9
— Thread 0 18
— Core 10
— Thread 0 20
Timers
————————————————————————-
ACPI timer 3.580 MHz
HPET timer 14.318 MHz
Perf timer 2.078 MHz
Sys timer 1.000 KHz
BCLK timer 133.02 MHz
Processors Information
————————————————————————-
Processor 1 ID = 0
Number of cores 4 (max 16)
Number of threads 4 (max 32)
Name Intel Xeon E5606
Codename Westmere-EP
Specification Intel(R) Xeon(R) CPU E5606 @ 2.13GHz
Package (platform ID) Socket 1366 LGA (0x0)
CPUID 6.C.2
Extended CPUID 6.2C
Core Stepping B1
Technology 32 nm
TDP Limit 80 Watts
Core Speed 2128.3 MHz
Multiplier x Bus Speed 16.0 x 133.0 MHz
Rated Bus speed 2394.3 MHz
Stock frequency 2133 MHz
Instructions sets MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, EM64T, VT-x, AES
L1 Data cache 4 x 32 KBytes, 8-way set associative, 64-byte line size
L1 Instruction cache 4 x 32 KBytes, 4-way set associative, 64-byte line size
L2 cache 4 x 256 KBytes, 8-way set associative, 64-byte line size
L3 cache 8 MBytes, 16-way set associative, 64-byte line size
FID/VID Control yes

Turbo Mode not supported
Max turbo frequency 2133 MHz
Max non-turbo ratio 16x
Max turbo ratio 16x
Max efficiency ratio 9x
TDC Limit 70 Amps
Core TDP 70 Watts
Uncore TDP 10 Watts
Power @ 9x 22 Watts
Power @ 10x 27 Watts
Power @ 11x 32 Watts
Power @ 12x 39 Watts
Power @ 13x 47 Watts
Power @ 14x 56 Watts
Power @ 15x 67 Watts
Power @ 16x 80 Watts
Max bus number 63
# of QPI Links 2
Attached device PCI device at bus 63, device 2, function 0
Attached device PCI device at bus 63, device 2, function 4
Attached device PCI device at bus 63, device 3, function 4
Chipset
————————————————————————-
Northbridge Intel 5520 rev. 22
Southbridge Intel 82801JR (ICH10R) rev. 00
Graphic Interface PCI-Express
PCI-E Link Width x16
PCI-E Max Link Width x16
Memory Type DDR3
Memory Size 8190 MBytes
Channels Triple
Memory Frequency 532.1 MHz (2:8)
CAS# latency (CL) 7.0
RAS# to CAS# delay (tRCD) 7
RAS# Precharge (tRP) 7
Cycle Time (tRAS) 20
Row Refresh Cycle Time (tRFC) 86
Command Rate (CR) 1T
Uncore Frequency 1596.2 MHz

Win7 Pro X64 on Dell T7500, MultiTB SATA, 8GB RAM, nVidia Quadro 2000, Vegas 12, 11, 10, 9 DVDA 6.0 & 5.2(build 135) Sony HDR-CX550V, Panasonic GH3 with LUMIX G X VARIO 12-35mm / F2.8 ASPH, LUMIX G X VARIO 35-100mm / F2.8

[John Rofrano]

[Stephen Crye] “During CPU only render with Mainconcept, there was GPU activity. In fact, during the 1st run after it was done and I looked, I thought I had specified some kind of GPU assist, and was surprised to see that I had three choices and it had defaulted to CPU only. My second screen shot in the original post showed that GPU activity when CPU only was selected in the render template.”

You need to realize that there are two GPU options in Vegas Pro. One is under Options | Preferences | Video and controls whether Vegas Pro uses the GPU for Video FX and Timeline playback. (note: you must play back the timeline in order to render so if this is on it will always be used while rendering)

Then there is the GPU options under rendering. This only affects using the GPU to ENCODE the video. It has absolutely no affect on Vegas using the GPU for it’s GPU accelerated FX. This is something completely distinct and different.

One is using the GPU to convert from Vegas’s internal uncompressed 444 format to what ever codec you are rendering to and the other is used to accelerate FX. So if you do a CPU only encode but you have the GPU Acceleration turned on for FX, you will see GPU activity during rendering.

If you want to measure absolutely no GPU assist, you must both turn off GPU Acceleration and use a render template that is CPU only.

~jr

http://www.johnrofrano.com
http://www.vasst.com

[Norman Black]

[Stephen Crye] “When you said “…The Vegas video composite and effects engine always uses OpenCL for both AMD and NVidia.” , do you mean during preview or render?”

The composite and effects engine is always used. Playback and render. Vegas creates a single video stream from your various sources with your edits. By definition, this is done during playback and during encode. During playback the video stream is sent to your preview window/device. During encode (render as), it is sent to the file encoder.

[Dave Haynie]

You got me curious about what my recently upgraded system would do in Vegas 12. When Vegas 11 came out, I bought both nVidia GTX570 and AMD HD6970 cards… both around my $300 budget, and did extensive benchmarking, versus each other and Vegas 10. I chose the HD6970. Have not taken a hard look recently, but I just ran a few benchmarks. Also didn’t use the accelerated Main Concept AVC encoder last time.

So this is the HD6970, an i7-3930K (6 cores + hyperthreading = 12 virtual cores), 64GB DDR3-1866. Rendering is from a 6TB RAID10 to a 960GB SSD (basically trying to eliminate any I/O-related issues).

So first, playback/preview. I set the mode to Best (Full), in a full sized 1920×1080 window on a 2560×1440 monitor. Ran these tests:

Frame Rate CPU GPU
8-bit, Good, GPU enabled 29.97fps solid 13% 36%
8-bit, Best, GPU enabled 29.97fps solid 20% 37%
32-bit, Best, GPU enabled 29.97fps solid 17% 48%

8-bit, Good, CPU-only 2.5-29.97fps 44% 11%
8-bit, Best, CPU-only 2.9-29.97fps 55% 14%
32-bit, Best, CPU-only 1.1-29.97fps 92% 13%

The effect of the GPU here isn’t even funny, how good it is. The CPU-only preview is very jumpy, the GPU smooth as glass in all cases. Though I’ll give some credit to Vegas here… in the past, Vegas didn’t make full use of a multi-core CPU for previews. Here, it’s doing pretty well.

Next is the actual rendering. Since I never render at anything less than “Best”, I didn’t bother benchmarking the “Good” setting, despite that being the default for the Sony Benchmark project. All renders are 1080i60. CPU is estimated average, GPU is peak.

Time CPU GPU
MC-AVC 25Mb/s 8-bit GPU-enabled 0:57 35% 93%
MC-AVC 25Mb/s 32-bit GPU-enabled 2:40 25% 77%
Sony AVC 16Mb/s 8-bit GPU-enabled 1:24 25% 56%
Sony AVC 16Mb/s 32-bit GPU-enabled 2:28 25% 63%

MC-AVC 25Mb/s 8-bit CPU-only 5:19 90% 11%
MC-AVC 25Mb/s 32-bit CPU-only 6:52 90% 10%
Sony AVC 16Mb/s 8-bit CPU-only 3:59 75% 24%
Sony AVC 16Mb/s 32-bit CPU-only 5:33 70% 33%

The GPU peaks for non-GPU renders are very momentary… and noticeably higher for Sony than MC. No idea about that one.

It’s interesting that the Sony AVC seems to be less computationally complex vs. Main Concept, but the latter gets more help from the GPU.

Both CODECs use the GPU better than I had seen in Vegas 11, but the Main Concept CODEC is out of control here, really shows that most of the render is happening in OpenCL, not on the CPU. Of course Vegas itself is responsible for some of this, but both CODECs get that same boost.

-Dave

[Stephen Crye]

Thanks Dave!

I really appreciate your taking the time to contribute to this thread. I know it is not a small thing to fiddle with benchmarks. Your times make me envious … but I knew my “new” system is actually 2+ year old tech. I’m somewhat hopeful that a new vid card will help SVP12 a lot.

Question: what about shelling out for a second Xeon? That would give me 8 physical cores – worth it for Vegas?

I’m heading over to a friend’s house this afternoon who is a professional video producer, he works for a local TV station and has a business on the side doing weddings, concerts, ads, etc. He recently built a new computer, not sure of the exact specs, but he has a GTX 650. I’m going to run the benchmarks and see what I get.

BTW, the new Corsair Neutron SSD for my C drive, much to my surprise, improved the SVP12 render time bit, but not the SVP11. The MainConcept render, which was 3:22, went down to 3:11 . Scratching my head over that one.

I’ll have the spec’s of my friends system and the results later tonight.

Thanks again!

Steve

Win7 Pro X64 on Dell T7500, MultiTB SATA, 8GB RAM, nVidia Quadro 2000, Vegas 12, 11, 10, 9 DVDA 6.0 & 5.2(build 135) Sony HDR-CX550V, Panasonic GH3 with LUMIX G X VARIO 12-35mm / F2.8 ASPH, LUMIX G X VARIO 35-100mm / F2.8

[Norman Black]

[Dave Haynie] “It’s interesting that the Sony AVC seems to be less computationally complex vs. Main Concept, but the latter gets more help from the GPU.

Both CODECs use the GPU better than I had seen in Vegas 11, but the Main Concept CODEC is out of control here, really shows that most of the render is happening in OpenCL, not on the CPU. Of course Vegas itself is responsible for some of this, but both CODECs get that same boost.”

Sony AVC only uses GPU for some things like motion search. Otherwise still the same encoder as CPU only. Quality should be pretty similar between CPU and GPU.

Mainconcept OpenCL (or CUDA) are completely different encoders than CPU only. They are mostly written to run on the GPU. Quality is lower than CPU only. Depends on bitrate and at higher bitrates I cannot see any difference.

The Mainconcept AVC OpenCL does not support AMD graphics core next GPUs. This means 7xxx series and later. Includes the current R8/R9. I found this out the hard way. My old 5850 flew on MC OpenCL and my new 7950 gets no use at all. Vegas itself and Sony AVC are fine with current AMD cards.