I don't know. I don't own either, so it was an informational post.

But a link to your informational post would certainly seem helpful to assist folks to make up their own minds, wouldn't it? [wink /]

;?)

@Musicvid Yes indeed, I was just waiting for the HORQ test to finish so that I could update here accordingly, with the new hevc data. Your post here reminded me that I hadn't done the Nvenc hevc test for both RTX and GTX, although I did previously already do both for Nvenc h264.

Link to previous tables here. Scroll near to the bottom of the page for them.

Link to dropbox zip with screenshots and spreadsheets here.

FHD below.

Thanks for your thoroughness, as always.

I regretably didn't mention that the Vegas posts I have been responding to are all related to GPU/CPU usage ratio, a presumed collateral effect of the card's basic functionality; in fact, those posters never mentioned quality metrics (to which they seem oblivious), in deference to their beloved speed comparisons.

Any chance that trivial usage information surfaced in your tests, as well?

@Musicvid Delighted, as I said, your prompted me to fix a hole in the data.

I intentionally never attempt to derive anything re: gpu usage because I suspect that its a little bit like studying the basic dimensions of an aeroplane and from that inferring something about its characteristics, capabilities.

Cpu usage, yes, it can tell you when its maxed out at say 100%, as expected while doing a cpu only render.

The problem is that with say Nvenc, HW encoding, there are so many things at play. If there are few FX that use the gpu, then the asic with the cpu will do most of the heavy lifting, and then people wonder why their gpu appears lifeless. I leave that to those more knowledgeable.

I've found this HW GPU review for streamers recently:

One more time: it is for streamers. Small bitrate required for streaming and encoding must be faster then realtime.

I was a bit surprised that based on the review, HW encoder in some Nvidia GPU (Turing?) overperformed in quality x264 Slow.

Unfortunately, single number (netflix metric) was used for comparison. I'd prefer to see graph instead.

I agree with all of the above caveats. I also am pretty convinced that these cards will not render as efficiently as their more modern counterparts, and try to get their owners to focus on that rather than Vegas as some sort of victimizing force.

The following post appeared on Handbrake yesterday. In light of many recent complaints focused on Vegas' performance with these cards, you may want to consider this:

To get usable hevc encoding with Nvidia card you need the Turing NVEnc, this means it has to be a 1650 Super or better (1660/20xx). The non-Super 1650 don't have this, nor do any of the older Nvidia cards like your 1080 and there's currently no AMD cards with what I would consider an usable hevc encoder.
 

I googled that quote. https://forum.handbrake.fr/viewtopic.php?f=7&t=40055

The Poster is complaining about Handbrake Nvenc HVEC creating much larger files than handbrake x265 with hugely compressible video such as video with many still pictures . That seems to be a handbrake fault as I use FFMPEG Nvenc HEVC at constant quality 24 to backup h264 1080p files and works great. average bitrate can go as low as 2MB/s for video with a lot of low motion and static scenes, up to 7MB/s with constant high movement. It works exactly as I would hope, and this is with a previous generation 10 series Nvidia card.

Only the current Turing Nvenc supports HEVC B Frames, It is said by Nvidia that Turning HEVC has up to 25% more efficiency in file size over previous HEVC so 25% smaller files?

The problem the poster was talking about was the file size being over 400% larger when using Pacal HEVC Nvenc compared to x265. That would seem to be a handbrake fault, and not a fault of NVENC or FFMPEG