NVIDIA Confirms Issues Affecting Early Production Run of GeForce RTX 2080 Ti Graphics Cards

[bug]

While I'm not willing to call Steve a lair, one has to question the why NVidia themselves has not given details themselves. It's not like they have much of anything to lose.

Probably because it doesn't really matter?
If was some faulty component in there. It's like if we were 100% sure it was the VRAM, we'd all start cancelling the millions of GB we have ordered from Micron.
The parties that need to know are probably informed already. And we will know when we compare the original cards with future batches. But no, some of us need to know here and now otherwise the sky is falling. As usual.

 

[lexluthermiester]

Probably because it doesn't really matter?

Good point. They did say that it was limited to early production cards. The problem may have already been solved.

 

[Captain_Tom]

Nice "Premium" Founders Edition. LMAO.

Sometimes I wonder if Nvidiots will ever stop shoveling Huang's mountain of shit...

 

[EarthDog]

There are places on the web, bug and lexy, that offer less toxic and informative conversations than, for example, the post above mine.

 

[Captain_Tom]

There are places on the web, bug and lexy, that offer less toxic and informative conversations than, for example, the post above mine.

Alright seriously, I am really curious here - What's wrong with what I said?

Nvidia is charging a massive mark-up and a a monopoly on the early release for their "founders editions." Those early and premium cards are apparently the least reliable. What nuanced comments should one expect? Enlighten me.

 

[R-T-B]

The implication that people who bought them expecting a working product are "nvidiots," mainly.

We also aren't certain the entire founders edition series, as a whole, is affected. You act like that is a factual statement.

 

[cucker tarlson]

Just for context, the "RTX 2080 Ti problem" consists of at least two distinct types of driver problems causing BSODs, and at least once distinct problem manifesting itself as artifacts.

and a hardware problem that causes them to artifact and die

 

[rules of entertainment]

says in your article that nvidia has comfirmed widespread rtx 2080 ti failures. yet i fail to find any evidence of nvidia saying failures are widespread . they confirmed some customer have experienced issues from limited test escapes but last time i checked that does not mean widespread issues. seems this guy should be working for cnn with his fake news and shoddy research, the tech media sites are becoming a bit of a joke

 

[lexluthermiester]

says in your article that nvidia has comfirmed widespread rtx 2080 ti failures.

You need to learn context;

NVIDIA, via a blog post on its forums, has confirmed widespread reports of failures affecting their flagship RTX 2080 Ti graphics card.

@rules of entertainment
Read what was written, not what you want to read.

 

[bug]

says in your article that nvidia has comfirmed widespread rtx 2080 ti failures. yet i fail to find any evidence of nvidia saying failures are widespread . they confirmed some customer have experienced issues from limited test escapes but last time i checked that does not mean widespread issues. seems this guy should be working for cnn with his fake news and shoddy research, the tech media sites are becoming a bit of a joke

If only we could have a magic thing where you could look up info on stuff... we could find things like: https://www.digitaltrends.com/computing/nvidia-rtx-2080-ti-founders-issues/ (i.e. widespread = below 4%)

 

[rules of entertainment]

If only we could have a magic thing where you could look up info on stuff... we could find things like: https://www.digitaltrends.com/computing/nvidia-rtx-2080-ti-founders-issues/ (i.e. widespread = below 4%)

in that article you linked it said due to small sample size take it with a grain of salt. also they are using anecdotal evidence which means less then nothing. in the article it is confirmed by overclockers uk that they have seen less 2080 ti faults then 1080 ti faults so hardly seems widespread does it. still yet to see any evidence the failures are widespread all your posting is more opinion peices and guess work with anecdotal evidence. am i wrong for asking for evidence that something is as bad as the cry babies on the internet say it is or should i just follow blindly .

 

[lexluthermiester]

in that article you linked it said due to small sample size take it with a grain of salt. also they are using anecdotal evidence which means less then nothing. in the article it is confirmed by overclockers uk that they have seen less 2080 ti faults then 1080 ti faults so hardly seems widespread does it. still yet to see any evidence the failures are widespread all your posting is more opinion peices and guess work with anecdotal evidence. am i wrong for asking for evidence that something is as bad as the cry babies on the internet say it is or should i just follow blindly .

It seems like you're trying to spread FUD.

 

[bug]

in that article you linked it said due to small sample size take it with a grain of salt. also they are using anecdotal evidence which means less then nothing. in the article it is confirmed by overclockers uk that they have seen less 2080 ti faults then 1080 ti faults so hardly seems widespread does it. still yet to see any evidence the failures are widespread all your posting is more opinion peices and guess work with anecdotal evidence. am i wrong for asking for evidence that something is as bad as the cry babies on the internet say it is or should i just follow blindly .

I wasn't objecting to your asking for numbers (I'd like to see those myself). I was objecting to your using the "widespread" term in the absence of numbers.

 

[EarthDog]

It seems like you're trying to spread FUD.

Quite the opposite. The FUD is using the term widespread when we dont know if it is or not.

I think you both are misunderstanding his point??

 

[lexluthermiester]

Quite the opposite. The FUD is using the term widespread when we dont know if it is or not.

I think you both are misunderstanding his point??

Naturally, I disagree. Their very first post called out the article author for stating something that was never stated. So either they failed to play attention to the vocabulary used in the article, or they selectively(and incorrectly) interpreted the statements of the article to create a non-existent narrative.

Their second post was much the same but with a bit of distracting nonsense thrown in for good measure. Neither of the posts from said user offered anything of value in the context of the information offered by the article. To me, it seems like FUD offered by someone seemingly attempting to degrade NVidia over a problem that is effectively very minor.

I was objecting to your using the "widespread" term in the absence of numbers.

And I was objecting to the misquoting of the article.

 

[EarthDog]

Nope. YHPM.

 

[lexluthermiester]

1) multi-monitor and 2) certain G-Sync displays, which should be reproducible on all Turing cards

That's the thing though, the discussion of the problems in this article are limited in scope to the 2080ti. I have a 2080 and dual displays have have not had a single problem.

Unfortunately there are too much misinformation around, partly due to anyone with any problem with RTX-cards throwing their hat into the ring.

Exactly this!

 

[EarthDog]

I think I have the perfect storm... 2080ti fe, 2 high hz monitors (acer predator 144hz 1440p ips xbu27 something or other, and another 27" acer 75hz ips...

Still good so far.

 

[Xzibit]

Looks like its not limited to Micron variants.

HardOCP - RTX 2080 Ti FE Escapes Testing by Dying After 8 Hours

 

[efikkan]

That's the thing though, the discussion of the problems in this article are limited in scope to the 2080ti. I have a 2080 and dual displays have have not had a single problem.

I don't have a Turing card myself, but the two distinct BSOD problems should be reliably reproducible on all Turing cards. Don't mistake this for claiming any dual monitor setup will cause BSODs, just that the specific ones which do, does so across all Turing cards.

A good portion of returned cards have probably been suffering from driver bugs rather than actual defective hardware. Still, there is no evidence of any widespread hardware issues. Only time will tell, we never know if every card is a ticking time bomb.

 

[Dr_b_]

Also, waiting gives drivers time to mature a bit

That can be a lengthier process and take even more time

 

[efikkan]

I usually wait 2-3 months for new platforms to mature, but not more, otherwise I would wait forever. With CPU families I actually pay more attention to specific motherboards, and usually don't pick one until I've seen solid reviews.

 

[bug]

I usually wait 2-3 months for new platforms to mature, but not more, otherwise I would wait forever. With CPU families I actually pay more attention to specific motherboards, and usually don't pick one until I've seen solid reviews.

Agreed. Waiting 6 months or more for companies to put out a finished product and then buying it, it just sends the wrong signals.

 

[Captain_Tom]

The implication that people who bought them expecting a working product are "nvidiots," mainly.

We also aren't certain the entire founders edition series, as a whole, is affected. You act like that is a factual statement.

That's not what I was inferring - I am not inferring people are dumb or fanboys for expecting a working product.

I was merely asking: Why are so many people willing to buy launch cards from a company that lied about how much VRAM was in their cards (Multiple times actually), approved VRM designs that literally lit on fire (EVGA 1080's), and obsolete their previous architectures ASAP with "Gamesworks" packages. And this is just in the past 4 years.

It blows my mind I am often accused of being a "fanboy" for asking these common sense questions. I don't get why so many put up with such low quality products.

 

[EarthDog]

Captain, not sure you have your ducks in a row....

1. They did not lie about the amount of VRAM. There is 4GB there. Just 512MB is slower. Any performance issues because of this were not common. Hell, you can use that slow 512MB w/o issue if it wasn't swapping out constantly.
2. EVGA 1080's had a cooling issue that had nothing to do with NVIDIA. IIRC, it was also on their CUSTOM CARD, not using NVIDIA's specs, but a more robust VRM implementation. The issue was with the AIB and how it cooled some parts, not NVIDIA.
3. Gameworks? Wha? I don't believe they are/were 'obsoleting' their previous archiectures. There have been plenty of quality reviews showing this isn't happening.

I'm certain you'll have things to say about these things, but, it is what it is...