which hmd and what cpu and ram you got ?
1 Like
I went from an i9-10900k to an i9-13900k, and a 3090 FE to a 4090 FE. I will say this has allowed me to max everything out in iRacing at 90 FPS/Normal FOV.
But, Iām getting stutters and the black screen flicker randomly. I have a 2073 8kx, btw. Iāve cleaned the Nvidia driver and installed the latest version, as well as deleted all the Pimax stuff and installed the latest Pitool-already had the latest firmware installed.
The stuttering is both In-game, and in Pimax Home.
Iāve tried different USB ports to address the screen flicker but it didnāt help.
Any ideas on the stuttering and screen flicker?
1 Like
Given the software angles youāve already tried, Iād suspect the tether cable.
Also try changing your HAGS setting.
Iām running an AMD 5900X with a Zotac 4090 Trinity OC. Iām on Pitool 283 and nVidia driver 526.47 (10/27/2022) and not experiencing such issues. This is with a fiberoptic tether cable.
I suspect the tether cable particularly if you have a fiberoptic because I have had a fiberoptic cable failure before just a few months ago. And its failure mode was different than Iād ever seen before for standard cables, so I didnāt recognize it as cable failure at first. On a standard cable, you usually get sparkles as it starts to go out. But on the fiberoptic cable, I think it has some sort of resend mechanism so instead you get intermittent delays.
2 Likes
Thanks for the response.
HAGS is off, but I do have the fiber optic cable, and have had it since I got the headset a couple of years ago, so you may be right.
The problem is they arenāt available for replacement.
There is no screen flicker at 75 hz if that makes a difference.
1 Like
Do you still have the stock cable the headset came with? You could try swapping to that to test.
And since you mention that this only happens at 90Hz, on the pre-2076 models I think 90Hz is iffy video card by video card. Itās pushing DisplayPort above its rating, so on some cards it will work and on some it wonāt. It seems to be a lottery. So you may have moved from a card that was able to do it to one that canāt even though itās otherwise newer and better.
On my previous GPU, I was one of the unlucky ones where 90Hz didnāt work. But then Pimax support sent me firmware P2_DEBUG_8KX_M299_t20211101115020_r85cdd55_forEnableHDCP which got it working on mine. And that firmware is continuing to work for my 4090. So that would be another thing to try. Or perhaps another variant of the 299 firmware will resolve your particular issue.
2 Likes
I have this exact same card and CPU and can confirm my 8kx worked without any issues. I also have the fiber optic cable as well, I think that is a big reason why I havenāt had any issues. I also had a EVGA 3090ti which worked out of the box as well.
1 Like
Igor didnāt really though. All he has is a theory. He did not actually test whether he could replicate the failure using that theory. And he also seems to have been dissecting what may have been a pre-release sample which is different and inferior to GNās actual production samples.
Gamers Nexus did test Igorās theory. In fact, they intentionally sabotaged their 12vhpwr adapters in various different ways and still could not get them to fail. That included even cutting half the wires and putting sharp bends on it.
I really appreciated GNās video because itās the only source of information Iāve seen on this issue with real credibility. They perform testing and a lot of it rather than speculating and making up or passing along rumors. And what I most appreciate about their coverage is that they refuse to draw any conclusions or accept anyone elseās conclusions which are not supported by data.
Even though that means their conclusion is inconclusive. They were not able to determine what caused this small number of adapters to melt yet.
However, what their testing (and all of the other actual testing Iāve seen) does show is that these small, fragile looking power connectors are actually tough little f*ckers. A lot of people speculated that they must barely be able to handle the 600W load theyāre rated for, but actual testing to failure has them handling 1400W before failingā¦ no wait. Actually, they did handle 1400W just fine. To get the connector to actually melt required 1400W AND intentionally having the connector partially seated. And that was finally enough to barely get it to start melting.
nVidia and Gamers Nexus have ongoing investigations into this issue. Iāve seen a lot of comments that it must be really bad if nVidia still hasnāt released any statement about it yet. But Iām pretty sure what nVidia is actually doing is performing engineering analysis on the problem rather than giving some jumping-the-gun PR answer. And real engineering takes time. This seems to be a difficult failure analysis because its apparently quite hard to get these adapters to melt without running them way over their ratings and/or intentionally sabotaging them.
3 Likes
Just signed up should help alot, thanks!
1 Like
I actually only plugged in 3 of the 4 power cables to my GPU. Limiting it to 450W
I see no need to go higher.im probably even going to setmy power limit to 75-80% as that drops power consumption by 100W and heat is so low fans barelyneed to run. And all of this only loses you 5% fps
All this to run at the same 350w power level as my old 3080Ti.
Gamers Nexus also thoroughly debunks alot of the hyperbole in his video
2 Likes
Then if it is only a small number without becoming a trend. Cpuld simply mean a bad batch of adapters and/or bad connector on the gpus affected. Simply defects that were not discovered during the manufacturing process.
I think that is most likely the case, yes.
1 Like
My Thermaltake toughpower gf3 is under way. It seems to be the first atx3.0 psu out in the wild (it seems a batch of the Asus Rog loki also got released some weeks ago but since then no stock anywhere) .
Going to wait for that before connecting my 4090. I donāt want to use that converter plug although Iām sure the risk is very small
2 Likes
indeed the less connections like cable adapters the better.
1 Like
Seems like this issue needs more evaluation and might be possibly an Nvidia side issue. As even new reports o Atx3.0 native 16pin connector melting.
3 Likes
LOL, damn, here I was thinking I could just bypass the whole issue but now the native connectors melt too!
2 Likes
Yeah not good. Means definitely an idea to not leave the system on 24/7. Though with the hungry power supply maybe a given with the cost of electricity. lol
3 Likes
This isnāt surprising in the slightest. Weāre going to see examples of failed 12vhpwr connectors of every kind. There has never been a power connector in the past that has had a 0% failure rate. We should not expect a 0% failure rate from the new power connector either.
The article mentions that there have been 18 āconfirmedā incidents on reddit in total. It was recently reported that 100,000 4090ās have been sold so far. So thatās a failure rate of 0.018%. And thatās not even counting all the 3090 Tiās that have already been out there for a while now which use the same power connector with the same TDP.
But this is such easy clickbait to frighten the public with. So you can bet the media are going to keep pumping those clicks by promoting panic every single time another individual power connector fails.
6 Likes