Least CPU/GPU intensive video format for Vegas?

essami wrote on 1/6/2016, 11:00 AM

We are doing film work on Vegas Pro 12. Dialogue, foley, music etc. and we've been doing fine with my computer until now. But we just started testing the surround sound 5.1 mixing and the computer doesn't seem to be able to handle that properly.

Now I'm wondering what is the least CPU/GPU intensive video format that Vegas Pro 12 would have no trouble crunching through. We only have one video track and have been using low quality mp4, h264 full hd renders.

We do need good quality since we must see lip sync properly. But I was wondering if a compressed h264 is more intensive on the computer then say a DNxHD?

Can anyone advice or has anyone made any tests?



musicvid10 wrote on 1/6/2016, 11:06 AM
Might look at your audio buffers and driver selection in Vegas.
Your system should handle smooth preview, but you need to render a file and use a player to check audio-video sync. It will always be a little different than you are able to preview on the timeline.
OldSmoke wrote on 1/6/2016, 11:07 AM
MPEG2 seems to be one of the best implemented ones. Try XDCAM 1280x720. What frame rate is your project and source?

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)

essami wrote on 1/6/2016, 11:54 AM
I've tried most of the possible settings I can think of. The computer handles 100+ audio tracks + the mp4 video with ease, it's just when I switch to surround sound mixing that I end up in trouble and get contast audio dropouts etc. Check for details here on that issue: http://www.sonycreativesoftware.com/forums/showmessage.asp?messageid=937734

Choosing the correct video format is just one step to making it easier for Vegas to run the material. We also need to cut down the number of tracks and effects by doing prerenders if need be before doing 5.1 mix.

The original material is 25fps Prores 422 HQ. Unfortunately it is edited on a mac based system with Premiere Pro so I'm uncertain if it can render XDCAM, hence my suggestion of DNxHD.
astar wrote on 1/6/2016, 3:25 PM
XDCAM-ex.MXF in 1080 or 720 will have the least overhead. Alternatives to try would be Cineform-low profile in .AVI, or even rendering a "PAL widescreen" DV bump using a custom profile where you specify progressive 25P. DV is also under the .AVI format selections.
PeterDuke wrote on 1/6/2016, 5:44 PM
"Alternatives to try would be Cineform-low profile in .AVI,"

Does Cineform handle more than two audio channels these days?
NormanPCN wrote on 1/6/2016, 6:56 PM
I would agree with others in this thread that XDCAM in MXF (Sony MXF) or MP4 (XDCAM EX) is about as low overhead as it gets. XDCAM is mpeg-2 video.
astar wrote on 1/6/2016, 7:27 PM
"Does Cineform handle more than two audio channels these days?"

I was assuming that the user only really needs picture, or picture with 2 channels of audio since he is mixing sound.
TheHappyFriar wrote on 1/6/2016, 7:45 PM
I've done several 5.1 mixes with DV AVI back on my old AMD XP 1800. Ran great with multiple audio & video tracks.

DV AVI is the lest intense format out there for the most part.
musicvid10 wrote on 1/6/2016, 7:57 PM
ESPECIALLY if you are doing 5.1 you need to render a file to play back on a reference player.
NO accurate preview is possible because the audio bus mapper on the timeline and ENCODED COMPRESSED 5.1 are completely different.
It's trial and error, not a one-step product.
NickHope wrote on 1/6/2016, 11:04 PM
+1 TheHappyFriar

DV AVI is intraframe, so there is no temporal compression to decompress. And the Sony DV codec is basically unchanged for over a decade, and played smoothly even on mediocre hardware back then, so it's demands are tiny for modern PCs. It can do widescreen too.
TheHappyFriar wrote on 1/7/2016, 5:56 AM
ESPECIALLY if you are doing 5.1 you need to render a file to play back on a reference player.

Accurate preview is possible, you just need to compensate for any delay with the decoder. I was under the impression that the player auto compensates though (at least one DVD player I had let you adjust 5.1 mix settings). When I've done 5.1 mixes & exported to DVD I never recall ANY playback issues and I did nothing fancy with my mixes or encodes.
musicvid10 wrote on 1/7/2016, 8:40 AM
No, accurate preview of 5.1 to be ENCODED is not possible on the timeline.

Sync, volume, frequency response, noise figures, channel separation, dynamic range, loudness contour, THD, bass response and timbre are all quite different in the ENCODED 5.1 as opposed to mapped pcm at the bus level. That assumes the editor is not tone-deaf.

Every 5.1 encoder setting (Dialog Norm, Dynamic Range Control, Bitrate, highpass and lowpass filters, phase shift) affect the parameters above. These settings are in addition to the expected compression losses from the encoding itself.

None of these factors can be backported to the timeline because there is no realtime 5.1 encoder at the editing stage (I don't think this has changed). Even at its most transparent settings, the influence of the encoder is still most obvious to the ear and measurement..

That said, for speech (no music), the timeline preview might be just fine.
The HF rolloff and limiting in the encode might actually help it a bit.

But for ANY music or dynamic audio fx, the differences should literally jump out of the speakers if one has even marginal listening skills.

So for any surround project that is to be encoded to 5.1 (excluding mapped PCM), the encode IS the preview.

TheHappyFriar wrote on 1/7/2016, 8:20 PM
So you're saying if I take a 5.1 project, render out to DVD mpeg/ac3, encode to DVD, then re-import the DVD to Vegas it won't be the same?
musicvid10 wrote on 1/7/2016, 9:12 PM
I said that if I preview (or render 6 ch discrete pcm) from the 5.1 project timeline, then render the timeline to ac3 surround, the playback will sound remarkably different over the same audio system.

My goodness. Have you actually tried this?
riredale wrote on 1/8/2016, 12:41 PM

I've done a dozen or so DolbyDigital2/2 surround-sound documentaries over the years. Never noticed any hint of difference in the final DVD versus what I previewed using Vegas7 back then. None of the various parameters seemed to color the encode in any way. Setting DialogNorm to -31 set the output gain to unity.

Am I missing something in this conversation? Wouldn't be the first time.

Regarding the OP topic, it would seem to me that doing a video render to DVavi and then using just that single video track while working on the audio would dedicate 99.9% of the CPU to just audio. But the issue must be something else--I can't imagine 5.1 bogging down ANY system, let alone a powerful one.
musicvid10 wrote on 1/8/2016, 1:02 PM
You are not getting encoded ac3 or aac off the audio bus. And at least with music, it always sounds different, sometimes unacceptably different. This can't be hard for anyone to understand.

That is true even if one goes to the expense of purchasing an audio interface with at least six discrete i/o channels (not a 5.1 sound card).
That's what I did. I still used it to record with, but I preview my surround with a dvd or media player on my home entertainment system, which I suspect is the way most people will enjoy my product.

So why would I suggest that the OP go to all the bother and $$, when the one perfect solution is a couple of DVD-RW media?

Do what you want, but in this case the price of accurate preview, whether sync or something else, is the time spent in its rendering.
Replacing an audio track in Architect is quick and doesn't involve a separate video render.
That's all I can offer to the discussion . . .