Make sure you read the instructions, the first time I tried to follow this I misread the instructions and thought my CPU didn't support it or there was some other issue but it couldn't have been easier when I actually stopped and read instead of skimming. To everyone else: If you have an Intel CPU that supports QS and you are running Plex in a docker container then I HIGHLY recommend you look into enabling HW acceleration (both in plex settings and by mounting a device, /dev/dri, into your container). Performance and quality edit Like most desktop hardware-accelerated encoders, Quick Sync has been praised for its speed. It kills me how long I waited to try this, it literally took <10min to setup and has been a huge performance boost. Quick Sync Video is available on Core i3, Core i5 & Core i7 processors starting with Sandy Bridge, and Pentium processors starting with Haswell, 3 Celeron processors do not feature Quick Sync Video. I know there has been some grumbling about QS not producing the best output but that seems to be mainly related to the earlier iterations of it, I have not noticed any issues. QuickSync is CRAZY good, I stumbled onto a post about it a month or so ago and thought "Ehh, I'm probably already using it, w/e" then saw I needed to run some commands and mount a device in my plex docker container to actually enable it and I was blown away by how little load there was on the CPU. IB average 3.7 faster than i7 SB and only 16 over i5 SB not.
I'm not against reusing the Quadro if I have to, so I'm after either a solution to get the quality of QSV to a reasonable level or a way to get the Quadro to transcode quick enough so that there is no buffering.I hadn't considered a NUC to run the plex server, right now I enjoy managing everything through UnRaid but I'm not opposed to another device that would manage plex alone. Arcsoft’s MediaConverter supports Quick Sync just fine, but the latest build doesn’t behave as well under APP or CUDA/NVEnc. 264 Encoder and max quality NVENC vs AMF/VCE vs QuickSync vs X264 - ULTIMATE. I do have a Quadro P2000 from my old server that didn't have any quality issues, but never seemed fast enough to transcode 4K remux files (yes I know this kills the hdr but my remote users don't care) and the max I could ever get from it was a rate of 0.8 in Tautulli, which would of course result in buffering.
I7 QUICKSYNC DRIVERS
and the latest Intel Graphics drivers at the time of writing. So it seems that Intel quick sync is capable of producing good quality, just that it seems to degrade or start off poor for some reason.
I7 QUICKSYNC MOVIE
If, however, I back out of the movie and then resume it from where it left off the quality is fine for a while, before then degrading back into a pixelated mess. One thing I have noticed is that playing a new movie file from the start results in poor quality.
I7 QUICKSYNC PORTABLE
Unfortunately the transcoding quality is very very poor with hardware acceleration enabled, and no matter what I do I don't see to be able to improve matters. Intel® Quick Sync Video delivers fast conversion of video for portable media players, online sharing, and video editing and authoring. This is due to bandwidth limitations on my upstream. Please contact OEM for the BIOS that includes the latest Processor configuration update. I only use transcoding for my remote users as everything direct plays locally. Some products can support AES New Instructions with a Processor Configuration update, in particular, i7-2630QM/i7-2635QM, i7-2670QM/i7-2675QM, i5-2430M/i5-2435M, i5-2410M/i5-2415M. The new server is an I5-10400 with an Intel UHD 630 graphics for hardware transcoding. Our collaboration with Intel has enabled us to optimize the DaVinci Resolve 17.1 application that is available now to take advantage of Intel’s built-in, dedicated, hardware-based Quick Sync Video technology on the latest 11th Gen processors to make these advanced media capabilities available to our customers, without compromising on performance. I recently built a new Plex server to replace my aging i7 920 box.
0 kommentar(er)
