Pretty nice article, but could use some polish..

First thing I see that is misleading:

" However, if your timeline playback is 29.97 fps in Best/Full quality with the video card’s GPU selected, then you just also increased your render speed to 29.97 fps. So you can see that increasing your playback rate definitely helps speed up your output rendering speed."


Full rate playback just means your GPU/CPU can preview at the full rate. The render rate could me much more than 29.97, since it is not limited by the media frame rate..

For example, my system takes 13 seconds to render the Sony Demo Benchmark to .MPG, which is almost 150 fps..

Thanks for the input. I know it needs more polish I just posted and wanted to get some feedback. Thanks for corrections I will update the article tomorrow.

As for the Royalty Free Music, I did fix that not sure why the corrected version upload. I will fix that in the morning.

Dave

[I]NOTE 3: I also want to point out the both Sony Vegas Pro and Sony Movie Studio only support using one GPU. So getting a second video card or a video card with dual GPUs will not improve performance.[/I]

Not quite true. I use two R9 290 in my system and Vegas does make use of both; strange but it does. The performance difference is in the order of 10% between one or two cards, which may not be sufficient for some to spend the extra money. However, NeatVideo for example does fully support both cards and there the improvement is much higher.

What Vegas does not support is CrossFire or SLI, there you will find no difference or sometimes worse performance. But you where right that dual GPU cards will not improve performance over a single GPU card.

I think it is also important to note that this not only applies to the desktop cards but also to the workstation cards.

There is another misconception when it comes to rendering. Only MC AVC and Sony AVC codecs have options to choose additional GPU acceleration during rendering and that only works with the older models as you mentioned, desktop and workstation cards alike.

I still believe the best way to tune your system is SCS own Benchmark project as will give a baseline for everyone. There is an old thread in here where one of the users even made an Excel sheet noting down many users input on their render times with different hardware all related to the SCS benchmark project.

"Also, include are the AMD HD 4000, HD 5000 and HD 6000 video cards" should maybe be "included"

"A lot of changed since these "

maybe should be "a lot has changed"

"Sony Vegas Pro and Sony Movie Studio is based on "

should be "are based on"

" any where " should be "anywhere"

"examples later in the article" should change the "the" to " this' so it matches your next similar statement

"cards, it just that with Sony Vegas Pro and Sony Movie Studio, modern AMD video cards will give you a much better " don't need the comma after STUDIO but do need an "is" before just

"MainConcept only support GPU " should be "supports"


" Here is an example of how playback of the timeline affects the rendering for export your video." I would say "affects the rendering of your video". Don't need the "export"

"take 25 minute " should be "minutes"

"then a 5 fps playback speed" should be "than"

"(fps = frame per second)" probably should be "frames"

"modern AMD video cards will give you a much better performance over modern NVIDIA graphic cards." you call one a Video card and the other a Graphic card



"which help greatly " should be "helps"

"To use CUDA with the MainConcept AVC encoder you can only a GT/GTX200, GT/GTX400 and GT/GTX500 series of video cards. " you can only what?

"Why they would hardcode the list of video card into the MainConcept AVC encoder is beyond me" should be "video cards"

" encoder you an only use the" should be "can"

"you can still find them on the used market fairly inexpensively"
you can buy them inexpensively but you probably find them inexpensive

"it effectively block " should be "blocks"

"It effectively block all newer video cards such as the NVIDIA GT/GTX600, GT/GTX700, GT/GTX900 and AMD HD 7000, R7 and R9 series of video cards from having GPU acceleration when using the MainConcept AVC encoder." you said "video cards" twice in the same sentence. You only need one of them.

"Since they are two different encoders, you will see a quality difference between the two, especially at lower bit rates. The GPU encoder is not as good as the CPU encoder, especially at lower bit rates." you said "especially at lower bit rates" twice in the same context.

"between the two encoder," should be "encoders"


" Just remember to keep the timelime playback set to your video card under Options/Preference/Video, as this helps with the speed for rendering out the final video." Check this statement. there are two places to set GPU acceleration. One only affects the timeline playback, the other affects rendering.

"we have also seen report the when some " huh?

"Some 3rd party plug-ins uses OpenGL and not OpenCL. Some of these effects can be quite intense and simply won't playback in real time no matter what you do." "some" is redundant. Drop the "some of" in the second and it reads better. Also change "uses' to "use".


"However, with 3rd party plug-in that do supports OpenCL" either needs to be "plug-ins that do support" or "with a 3rd party"

"It use to slow down timeline " should be "used"

"So as you can see by now the AMD video cards will give you much better performance with Sony Vegas Pro and Sony Movie Studio than the NVIDIA video cards will." drop the "will" at the end.

"take in to consideration " "into" is one word.

Here is a link to the spreadsheet with the performance numbers with the Vegas Benchmark Project

http://www.hyperactivemusic.com/vegaspro/vegaspro.html

Note: The best performance when the spread sheet was last updated was with an R9 290 GPU and an i7 4930K CPU with XDCAM renders of 30 and 132 seconds respectively. ( 132 seconds without GPU support )

With my R9 290X and 5960X @ 4.3ghz I get 20 & 114 seconds respectively.

Note: The comments at the spreadsheet site include " Timeline performance without GPU support is not worth reporting."


P.S. Here is a link to the Benchmark Project:

download.sonycreativesoftware.com/whitepapers/vp11_benchmark.zip