Vegas 12 rendering slower than Vegas 9?

VMP wrote on 5/31/2013, 7:20 PM
I could finally test my new Intel i7-3770 3.40GHz based (64 bit Windows 7 Ultimate ) PC that I have discussed earlier.
Overall it is rendering 4.5x faster than my previous Intel Q6600 2.40GHz (32 bit XP) PC so that's great.

I really wanted to try out Vegas12 to see how fast it renders compared to my old Vegas 9.
Downloaded the V12 trial and to my surprise it renders slower than Vegas 9!

Render time needed for 1 minute length 1080-50i 50 Mbps clip MXF file output:

Windows 7 & Vegas 9 - (GPU Accel not availb) = 48 sec (0:00:48)
Windows 7 & Vegas 12 - (GPU Accel Disabled) = 62 sec (0:01:02)
Windows 7 & Vegas 12 - (GPU Accel Enabled) = 67 sec (0:01:07)

Previous PC:
Windows XP & Vegas 9 - (GPU accel not availb)= 243 sec (04:03)

Setting the priority higher slows things down even more.

Why is Vegas 12 rendering slower than Vegas 9?
The GPU acceleration doesn’t seem to add any acceleration during rendering is that right?

Source file: HD 1080-50i '.m2ts' from Sony HXR NX5
Project setting: HD 1080-50i (1920x1080; 25,000 fps)
Render setting: HD422 1920x1080-50i 50 Mbps (under Sony MXF)

Nvidia driver: 320.18 WHQL
Release Date: 2013.05.23
Single video card 'EVGA GTX 670 FTW,4GB'

Thanks for any reply

Comments

Trainman wrote on 5/31/2013, 10:14 PM
I have also updated my system to be able to work HiDef video better. I did some time tests and found the new system running VP11 and W7 Ultimate 3x quicker than the older system. I upgraded to VP12 due to too many issues with VP11 and found 12 to be slightly quicker, whether that was due to the software finally recognizing my GTX670`s I could not say. Rendering using 12 for me is quicer than the earlier versions. BUT I did notice an issue with render speed with both GPU`s hooked together (SLI), it was slower, removing the SLI bridge rendering increased again to what I had been seeing prior to SLI. Are you running 2 GPU`s?
Speed difference for me between SLI configured GPU`s and non SLI was large. With SLI rendering1920x1080 25fps clips was 19 frames per sec, with SLI off rendering increased to 39fps which equates roughly to 38 secs for 1 min of video big difference eh!
Basically I can not tell you why VP9 was quicker than 12, it is opposite of what I experienced with the new system.
Cheers
ritsmer wrote on 6/1/2013, 4:53 AM
Just a "long shot" - but in V12 try to set Preview RAM to 30 MB and Rendering threads to 5 and test again with GPU disabled?
VMP wrote on 6/1/2013, 10:28 AM
Thank you for your reply,

@ Trainman

I am using a single 'EVGA GTX 670 FTW,4GB'

Have you tried rendering with Vegas pro 9? Maybe it's faster with your system too.
You can download a trial here:
http://www.sonycreativesoftware.com/download/link?id=4219.1
http://www.sonycreativesoftware.com/download/updates/vegasfamily

@ ritsmer

Thanks for the tip.

I have set the preview RAM to 30 MB and rendering threads to 5 with the GPU disabled.
But the render speed is exactly the same 62 sec (0:01:02).

V9 still faster = 48 sec (0:00:48)
OldSmoke wrote on 6/1/2013, 12:31 PM
Are the two rendering templates identical? It could be that some of the presets have changed between 9 & 12. I still have 9 on my system and I will give it a try too although I only got 60i footage but it does come from NX5U too.

Proud owner of Sony Vegas Pro 7, 8, 9, 10, 11, 12 & 13 and now Magix VP15&16.

System Spec.:
Motherboard: ASUS X299 Prime-A

Ram: G.Skill 4x8GB DDR4 2666 XMP

CPU: i7-9800x @ 4.6GHz (custom water cooling system)
GPU: 1x AMD Vega Pro Frontier Edition (water cooled)
Hard drives: System Samsung 970Pro NVME, AV-Projects 1TB (4x Intel P7600 512GB VROC), 4x 2.5" Hotswap bays, 1x 3.5" Hotswap Bay, 1x LG BluRay Burner

PSU: Corsair 1200W
Monitor: 2x Dell Ultrasharp U2713HM (2560x1440)

VMP wrote on 6/1/2013, 1:42 PM
Hi there Oldsmoke,

I am pretty sure they are identical.

Render as -> 'Sony MXF' -> 'HD422 1920x1080-50i 50 Mbps'

Yes please give V9 a try, maybe it's faster than V12 for everyone.
I am curious.

ritsmer wrote on 6/1/2013, 3:46 PM
Thank you for testing. I would have expected a substantial improvement - but obviously not so on your machine. Sorry.
NormanPCN wrote on 6/1/2013, 4:36 PM
The GPU acceleration doesn’t seem to add any acceleration during rendering is that right?

GPU accel in video prefs is always active in playback and encoding a file (aka render as). Of course you have to be using some effect/transitions that have GPU accel.
VMP wrote on 6/1/2013, 6:44 PM
@ ritsmer
Indeed we gave it a shot :-).

@ NormanPCN
Thanks.
Are all the video FX from Vegas 12 GPU acceleration-compatible?

Just renderd out the same clip with 'Saturation Adjust' plugin.

V9 = 50 sec.
V12 = 77 sec (GPU accel enabled).

TheRhino wrote on 6/2/2013, 9:28 AM
On my 6-core 980X Vegas 10e uses 100% of the CPU during renders. With GPU OFF, V11 & V12 only used about 60% of the CPU.

After installing a GTX570, my V12 renders now beat my best 10e scores on the types of files I render. I often render more than one file type at once since most clients want a HQ version along with a web version, etc. When I am rendering two files at once, V12 will utilize 100% of the CPU and the completion time for both renders is much faster than 10e rendering the same two VEGs at once...

5 MINUTE HD to BR RENDER (TYPICAL HD PROJECT FOR ME):
V10e = 100% of CPU = 7:52 minutes
V12 = 50%-70% of CPU, GPU ON = 3:55 minutes

I then allowed (2) 1.5 hour VEG files to render at the same time. One was HD to BR and the other was a copy the same project down-converted to DVD mpg2.

1.5 HOUR HD TO BR & 1.5 HOUR HD TO DVD (2 Renders at Once):
V10e = 100% of CPU = 3.5 hours
V12 = 70%-80% of CPU, GPU ON = 2.25 hours

The above renders are from a typical project we do which do not have as many layered events or special effects as the benchmarks.

EDIT:
V9e is slower than V10e on my 6-core because no matter the settings V9e only uses about 60% of my CPU. V10e is the only one that totally maxes-out my CPU and the render times match the % of CPU used... For instance, if V10e uses 100% of the CPU, and V9e uses 60%, then V9e takes almost 2X as long to render the same project.

In comparison, V12e only uses 60%-75% of the CPU but with GPU on, render speeds are significantly better than 10, and over double 9e... Again, this is for the work I do. I start-out with high quality AVID DNxHD, ProRes HQ or even uncompressed AVI. The output is typically back to AVID DNxHD, Blu-ray, DVD, and H.264.

Workstation C with $600 USD of upgrades in April, 2021
--$360 11700K @ 5.0ghz
--$200 ASRock W480 Creator (onboard 10G net, TB3, etc.)
Borrowed from my 9900K until prices drop:
--32GB of G.Skill DDR4 3200 ($100 on Black Friday...)
Reused from same Tower Case that housed the Xeon:
--Used VEGA 56 GPU ($200 on eBay before mining craze...)
--Noctua Cooler, 750W PSU, OS SSD, LSI RAID Controller, SATAs, etc.

Performs VERY close to my overclocked 9900K (below), but at stock settings with no tweaking...

Workstation D with $1,350 USD of upgrades in April, 2019
--$500 9900K @ 5.0ghz
--$140 Corsair H150i liquid cooling with 360mm radiator (3 fans)
--$200 open box Asus Z390 WS (PLX chip manages 4/5 PCIe slots)
--$160 32GB of G.Skill DDR4 3000 (added another 32GB later...)
--$350 refurbished, but like-new Radeon Vega 64 LQ (liquid cooled)

Renders Vegas11 "Red Car Test" (AMD VCE) in 13s when clocked at 4.9 ghz
(note: BOTH onboard Intel & Vega64 show utilization during QSV & VCE renders...)

Source Video1 = 4TB RAID0--(2) 2TB M.2 on motherboard in RAID0
Source Video2 = 4TB RAID0--(2) 2TB M.2 (1) via U.2 adapter & (1) on separate PCIe card
Target Video1 = 32TB RAID0--(4) 8TB SATA hot-swap drives on PCIe RAID card with backups elsewhere

10G Network using used $30 Mellanox2 Adapters & Qnap QSW-M408-2C 10G Switch
Copy of Work Files, Source & Output Video, OS Images on QNAP 653b NAS with (6) 14TB WD RED
Blackmagic Decklink PCie card for capturing from tape, etc.
(2) internal BR Burners connected via USB 3.0 to SATA adapters
Old Cooler Master CM Stacker ATX case with (13) 5.25" front drive-bays holds & cools everything.

Workstations A & B are the 2 remaining 6-core 4.0ghz Xeon 5660 or I7 980x on Asus P6T6 motherboards.

$999 Walmart Evoo 17 Laptop with I7-9750H 6-core CPU, RTX 2060, (2) M.2 bays & (1) SSD bay...

OldSmoke wrote on 6/2/2013, 10:57 AM
You are right that VP9 is faster for that particular codec. GPU acceleration doesn't help in VP12 as the codec is not written for it. Try an AVC codec and you will see the difference. What I found interesting is that the file rendered by VP12 reported under MediaInfo to be written by VP10.

Proud owner of Sony Vegas Pro 7, 8, 9, 10, 11, 12 & 13 and now Magix VP15&16.

System Spec.:
Motherboard: ASUS X299 Prime-A

Ram: G.Skill 4x8GB DDR4 2666 XMP

CPU: i7-9800x @ 4.6GHz (custom water cooling system)
GPU: 1x AMD Vega Pro Frontier Edition (water cooled)
Hard drives: System Samsung 970Pro NVME, AV-Projects 1TB (4x Intel P7600 512GB VROC), 4x 2.5" Hotswap bays, 1x 3.5" Hotswap Bay, 1x LG BluRay Burner

PSU: Corsair 1200W
Monitor: 2x Dell Ultrasharp U2713HM (2560x1440)

VMP wrote on 6/2/2013, 5:06 PM
Hi OldSmoke,

Thanks for doing the test.

So far only 'Main concept AVC AVCHD 1080 50i' seems to render faster.
If we just put the GPU option aside one would think that V12 should at least render at equal speed as V9.

Some more tests with different codecs:
Rendered clip length = 1 minute.

Main concept AVC AVCHD 1080 50i

Vegas 9 - 60 sec
Vegas 12 - 52 sec
---------------
Mpeg 2 BD 1080 50i

Vegas 9 - 28 sec
Vegas 12 - 37 sec

---------------
Avi HD yuv 1080 50i

Vegas 9 - 45 sec
Vegas 12 - 94 sec

---------------
Pal DV Uncompressed 720x576

Vegas 9 - 28 sec
Vegas 12 - 52 sec

MXF (from Sony) is based on Mpeg 2 even though it is not supported by GPU Accel I wonder why it is rendered slower by V12.

Have you tried other codecs as comparison?
OldSmoke wrote on 6/2/2013, 7:19 PM
VMP, I only tried AVC codecs by Sony and MC aside from the MFX 422 that you initially used. I am not sure why VP12 is slower but I do remember that VP8 & 9 where quite fast on my old laptop and now on my new one VP11 and 12 are crawling. I might consider to go back to VP 9 on my laptop which I only use to make the initial cuts at site and then transfer the project to VP12. My usual workflow includes HDV and AVCHD files from my Z5U and NX5U; delivery is usual BD, DVD and Internet; luckily those codecs are all GPU accelerated and really fast on my system.

Proud owner of Sony Vegas Pro 7, 8, 9, 10, 11, 12 & 13 and now Magix VP15&16.

System Spec.:
Motherboard: ASUS X299 Prime-A

Ram: G.Skill 4x8GB DDR4 2666 XMP

CPU: i7-9800x @ 4.6GHz (custom water cooling system)
GPU: 1x AMD Vega Pro Frontier Edition (water cooled)
Hard drives: System Samsung 970Pro NVME, AV-Projects 1TB (4x Intel P7600 512GB VROC), 4x 2.5" Hotswap bays, 1x 3.5" Hotswap Bay, 1x LG BluRay Burner

PSU: Corsair 1200W
Monitor: 2x Dell Ultrasharp U2713HM (2560x1440)

TheRhino wrote on 6/2/2013, 8:22 PM
Just thought of something - make certain you have quality settings set to BEST in V9... It's been a while since I used V9, but I think the default settings are only "GOOD" and therefore you are getting faster renders at the sake of quality. I might be wrong but once I set all project settings to identical, V9 is slower than V10e and V12 (with GTX570)...

Workstation C with $600 USD of upgrades in April, 2021
--$360 11700K @ 5.0ghz
--$200 ASRock W480 Creator (onboard 10G net, TB3, etc.)
Borrowed from my 9900K until prices drop:
--32GB of G.Skill DDR4 3200 ($100 on Black Friday...)
Reused from same Tower Case that housed the Xeon:
--Used VEGA 56 GPU ($200 on eBay before mining craze...)
--Noctua Cooler, 750W PSU, OS SSD, LSI RAID Controller, SATAs, etc.

Performs VERY close to my overclocked 9900K (below), but at stock settings with no tweaking...

Workstation D with $1,350 USD of upgrades in April, 2019
--$500 9900K @ 5.0ghz
--$140 Corsair H150i liquid cooling with 360mm radiator (3 fans)
--$200 open box Asus Z390 WS (PLX chip manages 4/5 PCIe slots)
--$160 32GB of G.Skill DDR4 3000 (added another 32GB later...)
--$350 refurbished, but like-new Radeon Vega 64 LQ (liquid cooled)

Renders Vegas11 "Red Car Test" (AMD VCE) in 13s when clocked at 4.9 ghz
(note: BOTH onboard Intel & Vega64 show utilization during QSV & VCE renders...)

Source Video1 = 4TB RAID0--(2) 2TB M.2 on motherboard in RAID0
Source Video2 = 4TB RAID0--(2) 2TB M.2 (1) via U.2 adapter & (1) on separate PCIe card
Target Video1 = 32TB RAID0--(4) 8TB SATA hot-swap drives on PCIe RAID card with backups elsewhere

10G Network using used $30 Mellanox2 Adapters & Qnap QSW-M408-2C 10G Switch
Copy of Work Files, Source & Output Video, OS Images on QNAP 653b NAS with (6) 14TB WD RED
Blackmagic Decklink PCie card for capturing from tape, etc.
(2) internal BR Burners connected via USB 3.0 to SATA adapters
Old Cooler Master CM Stacker ATX case with (13) 5.25" front drive-bays holds & cools everything.

Workstations A & B are the 2 remaining 6-core 4.0ghz Xeon 5660 or I7 980x on Asus P6T6 motherboards.

$999 Walmart Evoo 17 Laptop with I7-9750H 6-core CPU, RTX 2060, (2) M.2 bays & (1) SSD bay...

VMP wrote on 6/2/2013, 10:01 PM
@ OldSmoke, indeed.

@ TheRhino, Good point!
But I see that V9 & V12 were set to 'good' so identical.
NormanPCN wrote on 6/2/2013, 11:18 PM
Are all the video FX from Vegas 12 GPU acceleration-compatible?

Effects/transistions/media generators that are Vegas GPU accelerated are listed in the "GPU accererated" category folder(s).

Third party plug-ins may not properly identify themselves for Vegas to include it in the GPU accel category folder.

One thing that seems to be GPU accel but has no listing as it is not applicable in interpolation (crop/resize including rotate).