If ya’ll don’t feel like reading below - Apparently it matters which USB port you use to plug your OSR into. Consult your motherboard manual and determine which ones are wired straight to your CPU and do NOT go through a port that goes through a controller like ASMedia.
For the past few months I’ve been dealing with some light jitter on my OSR2. It was KIND of like a servo potentimeter going bad, but not as severe. It was livable, but still annoying enough for me to do a lot of troubleshooting including swapping logic boards, replacing servos, changing cables, trying a different Windows install, etc. None of it fixed the issue. This was around the same time I moved over from Intel (14th gen i9) to a Ryzen 9950x3d. I thought it may have to do with the dual CCDs and messed around with Process Lasso, prioritization of MFP, and more, and still didn’t get a fix.
My OSR was directly plugged into the motherboard (no hubs in between), so I really didn’t give it much thought. A few days ago, I was researching something totally unrelated, and then realized that some of the ports on the motherboard were wired through an ASMedia controller chip. Don’t get me started on the limited lanes on Ryzen chips… I swapped my OSR to NOT one of those ports, and holy shit, things are back to normal. Now I do know that Ryzen’s architecture has a bit more latency than Intel’s, but I didn’t know that this would have manifested so overtly with OSR performance.
I never encountered this with any of my Intel systems. I was loyal to Intel since Sandy Bridge, but recently jumped ship due to the self-immolating 13th and 14th gen chips and the very weak Arrow Lake release. So maybe this is something that Ryzen dudes know, but I wanted just to throw this out there in case anyone wanted to check to see how their OSR/SR is plugged into their computer.
I know this was an issue with VR cause they recommend that if youre gonna go wired with the Quest 2, to plug the USB cable into the back directly into the mobo instead of in the front in the case ports. Didn’t know there was an even further step with the ports in the back.
Intel boards also use/used ASMedia Controllers for additional slots, but yeah gotta read the manual beforehand (e.g. also for other things, lots of boards will have some ports not working anymore if you put in an additional e.g. nvme). Most of the time not noticable cause people rarely use devices with time criticial usb packets that need low latency.
Buuut this is really important and if you as a merchant didn’t know beforehand then that should a very important point to add to any manual/tutorial etc. for anyone - maybe spread it around a bit, but it’s one of those things that can save a lot of hassle and time.
Yep, aware of this - but never had this same issue with my Intel systems. I was thinking it may be related to differences in the chip architectures, not a particular issue with the ASMedia controller themselves. Maybe a difference in how the CPU or chipsets deal with interrupts or maybe a more aggressive handling of idle periods, or something else beyond my little brain.
Yep, definitely will put it into my setup guide. Thanks!
Anything placed between the usual communication can introduce lag.
I have a ryzen, and use the front usb ports, which while they do support usb3, only support its lowest power value. And yet i dont get any issues with the OSR in performance.
Delays shouldnt be a problem if the chip in between is good. As the USB protocol does allow you to attach multiple devices to a single lane without lag issues. Thats why ryzen cpus might never have bothered going for more lanes.
But if it does give issues, its surely welcome to know the things that can cause this. The USB standard while its flexible, is also a huge mess in all its varieties.
Oh wow tried this and this fixed every bit of jitter I had for the past months. Bought a pair of new servos for cheap and thought it was just got worn out quicker than usual. Thx @g90ak for this man.
Now I need to go back and check my old servos I removed because they started jittering a bit too much lol.
Glad it may have helped. The main difference between computer induced jitter vs a bad potentiometer is a bad servo will jitter at specific points. My port related jitter was more present during slow motion and not at any specific point.
Its the Ryzen 7 3700X
From the name i would say no x3d, as a 3700x3d also exists. And no idea of CCDs.
I think the motherboard matters a lot on this. And of that, no idea which specific one i have. Its a replacement after warranty from the previous one breaking, and the company fixing it installed a significantly more expensive one (my old motherboard supported 4 nvme ssds which at the repair time was a quite rare thing - and afterward they didnt want to go through the hassle of getting the extra costs of the motherboard).
All i know is that its targeted towards gaming, which might explain USB3 ports at the front with good latencies, yet poor power output support. And on that, the ports are directly connected to the motherboard.
