• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

RTX 3080 gigabyte gaming OC stuttering and coil whine?

Status
Not open for further replies.
Jan_C said:
Update : After seeing Memtest errors, I investigated my RAM to see if they were the culprit for the stuttering and ran the Memtests for each individual stick. Strangely, although I initially got errors, I took out the ram sticks and put them back in and I got no errors in the following Memtest, honestly feels like I am going down a massive rabbit hole.

In either case, I tested the stuttering for each RAM stick individually and found that the stuttering was present for each stick, so I have ruled out the possibility of faulty RAM sticks causing the stuttering and that it's something else.

Not sure what to test anymore, I have checked SSD health, tested my RAM, tried DDU installation for GPU drivers and different cables for the GPU. Temperatures are normal, everything looks normal.

Although I cannot scientifically rule out the PSU as the faulty for the stuttering, I am fairly certain it's not the cause of the stuttering as the stuttering is only for the first minute or two of starting a game / joining a map and afterwards the performance is flawless (I think this is also indicative that the temperatures are not the cause), so it definitely feels like a driver issue for my particular setup / faulty card.

The good news is that I believe my issue is unique and not representative of the cards performance as lots of people commented using the same card and having good performance and no issues. People who have pre-ordered this card should feel excited to get their hands on it!
Click to expand...

Bit flipping in normal ram would just cause instability rather than stutters, Vram is different because it has a form of error correction. Avoids the crashing (as long as rate of errors is low enough) but can hit performance instead.

Cardinal_Fang said:
Another issue has now arisen with partner cards caused by the manufacturer using cheaper capacitors on the power supply, causing it to crash when it tries to boost.

https://www.igorslab.de/en/what-rea...tabilities-of-the-force-rtx-3080-andrtx-3090/
Click to expand...

I read about that, apparently this is a unproven theory at the moment, it could well be just the AIB's pushing chips past their stability, in the previous 3 generations this has happened, a board partner doing a shipping O/C that the chip cannot handle. I think their testing is not rigorous enough, and that is if it even existed, they might have just "assumed" the chips can handle it.
 
It's not an "unproven theory" - there are already reports from some 3080 owners about this problem. It's pretty clear that some manufacturers decided to ignore Nvidia's design and try to cheapen it, and now we see the result.
 
Cardinal_Fang said:
It's not an "unproven theory" - there are already reports from some 3080 owners about this problem. It's pretty clear that some manufacturers decided to ignore Nvidia's design and try to cheapen it, and now we see the result.
Click to expand...

I meant the theory is on the cause of the problem, not the problem itself. There has been reports of the problem on TUF cards as well. Even the title of the article states its a "possible" reason.

He acknowledges my idea as well, here is a quote.

Since the driver problem lasted until shortly before the launch, but the first wave of cards had to be produced already, the functional testing of the first models was obviously limited to power-on and thermal stability. Running, not running. However, this does not say much about the chip quality and the possible maximum frequencies that the respective chip can safely handle.

Thus, it would at least be plausible that cards could have been sold as OC cards, which wouldn’t have passed a real quality test at the manufacturer with the delivered settings. Real binning? Nothing. Subsequent selection of particularly overclocked cards? Impossible, in fact. And so it is by no means impossible that one or the other “Potato” chip could also have gotten lost on such an OC card. We know the consequences from the posts of the buyers in the relevant forums.
Click to expand...
 
Last edited:
 
Igor's lab seem to be getting to the bottom of what's causing these issues.

A lot of the cards are using mixtures of POSCAP and MLCC capacitors. The more POSCAPs, the more power draw issues or interference is occuring and it's causing a lot of these CTD issues. That could very well be the issue. I'd take the back plate off if possible and have a look at it. The cheaper models closer to MSRP appear to have POSCAPs
 
Cardinal_Fang said:
Another issue has now arisen with partner cards caused by the manufacturer using cheaper capacitors on the power supply, causing it to crash when it tries to boost.

https://www.igorslab.de/en/what-rea...tabilities-of-the-force-rtx-3080-andrtx-3090/
Click to expand...
Read that also - also this video based on this. https://www.youtube.com/watch?v=x6bUUEEe-X8&t=0s

Time will tell on this one, but might be some unhappy AIB punters around soon if mass underclocking (or rather in some situations not being able to take advantage of the boost)

Things might be interesting, in any case I might hold out for the FE card more so now. Assuming I get it before my pension is due to kick in.
 
I edited my post, to try and avoid double posting, but since you guys replied, I ask you to see my amendment, basically Igor also listed the lack of chip binning as a potential cause as well.

The reason I think chip binning is more likely, as even without the o/c, cards are still using the same amount of power, and so the components are under the same level of stress.
 
chrcoluk said:
I edited my post, to try and avoid double posting, but since you guys replied, I ask you to see my amendment, basically Igor also listed the lack of chip binning as a potential cause as well.

The reason I think chip binning is more likely, as even without the o/c, cards are still using the same amount of power, and so the components are under the same level of stress.
Click to expand...
Time will tell what the actual issue is but it's difficult to gauge as very few cards appear to actually be out in the wild. But it boils down to all the secrecy in the lead up to launch. The AIBs were not ready for production. So be it caps or badly binned chips it could have been avoided if they had time in advance to tweak their designs and test the cards with proper variable gaming loads etc.
 
weldon855 said:
Time will tell what the actual issue is but it's difficult to gauge as very few cards appear to actually be out in the wild. But it boils down to all the secrecy in the lead up to launch. The AIBs were not ready for production. So be it caps or badly binned chips it could have been avoided if they had time in advance to tweak their designs and test the cards with proper variable gaming loads etc.
Click to expand...

I think its a combination of both, I do think something is in it. Also agree on the rush of the launch has led to this mess.
 
On another note, what do people think. If I do RMA my card, would you go for a replacement or get a refund and use that money for Big Navi? I do feel my experience with the RTX 3080 has been troubling so far, and the current media coverage on all the other issues makes it tempting to ditch NVIDIA at this point.
 
Jan_C said:
On another note, what do people think. If I do RMA my card, would you go for a replacement or get a refund and use that money for Big Navi? I do feel my experience with the RTX 3080 has been troubling so far, and the current media coverage on all the other issues makes it tempting to ditch NVIDIA at this point.
Click to expand...

Well you obviously want to keep the price you paid for it, so I would wait a month or two for new production runs, then RMA the card at that point.
 
how can the AIBs not be ready for production so how come Asus was ready and is looking like they didn't go the cheap route? some AIBs took the cheap route and everyone knows about it now so they can't hide it now they will have to recall all the cards or do a cheap firmware update
 
chrcoluk said:
Well you obviously want to keep the price you paid for it, so I would wait a month or two for new production runs, then RMA the card at that point.
Click to expand...

Yes I plan to wait the full RMA period and hope that drivers fix it until then. Hopefully by the time I return the card in mid October, things stabilise for these cards and a quick replacement would also be possible.
 
John11 said:
how can the AIBs not be ready for production so how come Asus was ready and is looking like they didn't go the cheap route? some AIBs took the cheap route and everyone knows about it now so they can't hide it now they will have to recall all the cards or do a cheap firmware update
Click to expand...

I think they'll use their get out of jail free card in that the boost clocks aren't guaranteed. However them putting out a BIOS/fix that reduces them therefore stabilising the card is going to be a bitter pill to swallow for the affected owners. If i had an affected card I'd be angling for a refund to get an FE or a known card that isn't skimping on parts. (Well, personally i'd wait to see what AMD have stashed away... :D)
 
chrcoluk said:
Well you obviously want to keep the price you paid for it, so I would wait a month or two for new production runs, then RMA the card at that point.
Click to expand...

Disagree. You'll like get an RMA now as you have a malfunctioning card with no solution. If you wait, the AIBs will no doubt resolve this with VBIOS updates to reduce clock speeds, at which point you won't be able to return it as they'll offer you the update to fix the problems.

I'd get shot of it now. I'm actually glad I didn't get an early card now and will wait this complete **** show of a launch out for several months.
 
Just about every aib cheaps out on components to some degree, this time it might well have bitten them in the ass.
 
this is looking like it's going to a be a big **** show from the AIBs looking at that JaysTweCents video some of the cards don't even have the special components on the back the Asus tuf gameing oc i have seen pictures of has 6 of the damn special components on the back so wtf was the rest of the AIBs thinking of :mad:
 
Status
Not open for further replies.
Back
Top Bottom