Came across this while looking for people using AMD GPU encoding. If quality is negligible between methods, what is most important to you? Size or Speed?
He deserves a few views for his efforts.
TLDR; Don’t use AMD if h.264 is your focus.
My AMD R9 380 works fine for 264 and 265 as well.
(in the RARE event something has to transcode)
When I’m “Encoding” - I don’t use the GPU at all.
That would be foolish.
I typically don’t rely on YouTube Celebrities for my technical research.
Pot Brownie Recipes? Maybe. Depends on the Pot.
You rely on pot brownies for your technical research? That explains SO MUCH!
You must be on about a half a tray right now…
I didn’t say anything like that.
I love that Aunt Linda gave him a shout out in the comments. It is a decent video overall.
For streaming (or transcoding) it’s impressive how fast that AMD is. It’s also really impressive how fast that Laptop 1050ti is.
I haven’t seen a recent quality comparison, maybe I’ll go look. It’s frustrating to see comparisons like this that include speed & size but not quality or any BD-rate evaluation.
I know that NVENC H.265 quality has gotten remarkably good - significantly better than NVENC H.264 quality used to be. I wonder how AMD compares today.
He references his previous video where at the end he did a quick quality comparison on a 4k monitor and could not identify any differences.
I actually saw that when it came out. I went back and watched again.
To be an armchair quarterback about it, he loses credibility by being vague about quality in that way. My interpretation wasn’t that they were all “good enough” but that he didn’t know what he was doing.
I mean something like VMAF or ideally BD-Rate.
For streamers I imagine sheer speed is the metric above all others, or that it is a minimum threshold criteria. For archival there are people who choose quality or prioritize size, with different secondary preferences.
I don’t think it makes sense to measure speed and size and then make a recommendation. If quality isn’t measured or controlled that’s a big gap.
Software encoder performance changes dramatically when quality and “effort” settings are changed. The FPS of the hardware encoders changes dramatically with settings as well.
Some discussion of the ways to measure video encoder algorithms:
Here’s an example. I’d like to see something like this for the newest hardware:
Oh, and to answer your actual question! 
If quality is the same, I prioritize file size.
