• 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.

NVIDIA ‘Ampere’ 8nm Graphics Cards

You know what would be funny, or rather not for potential purchasers that if they just came out before Wednesday and said there won't be any reviews until after Thursday :D Nvidias mentality is you have to purchase blind without seeing any results, wonder what people think then.
I wouldn't think anything, because I'd order one anyway then cancel if necessary. No drama.
 
Yep although the rumoured Titan is sopossed to use a full fat GA102 and have 48Gigs of ram so it's ticked both of those boxes which I believe Kaapstad was interested in.

The last two Titans have been based very closely on NVidias pro cards (Titan V and RTX Titan) so it is possible in the future when GDDR6X 2gb chips are available that we could see a Titan with 48gb of it.
 
KLNSAIQ.png


lg6i16E.jpg


7OCH9PN.jpg


https://www.techpowerup.com/272126/evga-geforce-rtx-3090-xc3-and-ftw3-pictured-in-the-flesh
 
Whats the average % increase in clocks these days on a decent overclock? 2-5%?

my old 2080ti could go some the ram especially from 7000mhz stock all the way upto 8350mhz which is a good 18% more, core would only reach around the 2ghz mark even with big offset applied, from around 1905mhz stock under normal boost conditions
 
Always found my 2080ti quite unstable on clocks. Wouldn't get any artifacts on testing, but would cause general instability issues. Possibly a little sensitive to memory clocks.
 
No thats what Nvidia claims. Might be true. Might not be,.

Surely if it’s a ‘driver issue’ the reviewers will mention that e.g. they were issued new drivers which increased performance before the NDA lifted. I know plenty have mentioned that happening before with both Nvidia and AMD, sometimes just hours before release.

If no new drivers were issued or the drivers were not issued late then I can’t see how it could be a driver issue. Given that most said in their unboxing videos last week they have already ran the cards I think the driver being late isn’t a factor here.

The alternative is that nvidia were telling the truth... not everything is a conspiracy you know. Let’s wait until the NDA lifts and find out instead of all this nonsense speculation. :rolleyes:
 
Surely if it’s a ‘driver issue’ the reviewers will mention that e.g. they were issued new drivers which increased performance before the NDA lifted. I know plenty have mentioned that happening before with both Nvidia and AMD, sometimes just hours before release.

If no new drivers were issued or the drivers were not issued late then I can’t see how it could be a driver issue. Given that most said in their unboxing videos last week they have already ran the cards I think the driver being late isn’t a factor here.

The alternative is that nvidia were telling the truth... not everything is a conspiracy you know. Let’s wait until the NDA lifts and find out instead of all this nonsense speculation. :rolleyes:

Indeed its not always a conspiracy but some reviewers had had there cards for weeks now but they didnt get the driver to do anything with them until last week. So if you are going to send out cards weeks before launch and and NDA on reviews then shipping delays dont become the issue. The fact that reviews didnt get their "first" driver until late last week is what slows up reviews.

But like you say, it will all come out after the NDA is lifted.
 
Whats the average % increase in clocks these days on a decent overclock? 2-5%?

Do you mean as in how much of a difference does the overclocks make to game perfromance?

Stock 2080ti boosted to around 1860.

Good overclocked ones went to 2100 so a 12.9% overclock on the gpu plus the memory would go to 8000/

In games that then translated to between 6% and 17% more fps depending on the game

https://www.thefpsreview.com/2020/02/10/nvidia-geforce-rtx-2080-ti-fe-overclocking/5/

So if a 3090 boots to 2100 but 2080s only boost to 1900 then thats a an extra 10.5% overclocks which if the cards work the same as the 2080ti might equate to 5-15% better performance in games.
 
I bought an O11 Dynamic XL just for the 3090.. and changed to AIO - thus far £700 worth of changes, and the card isn't even out yet lol

bFqwGTu.jpg
 
Indeed its not always a conspiracy but some reviewers had had there cards for weeks now but they didnt get the driver to do anything with them until last week. So if you are going to send out cards weeks before launch and and NDA on reviews then shipping delays dont become the issue. The fact that reviews didnt get their "first" driver until late last week is what slows up reviews.

But like you say, it will all come out after the NDA is lifted.


It’s also pretty normal to only get the driver a few days before the NDA lifts, that’s how you control external leaks.

They will have had the driver for the best part of a week before the first NDA, that’s plenty to get a round of benchmarks down on it and shoot a video/write an article. That said it doesn’t mean they haven’t had new drivers issued in the mean time.

It also preferable that driver issues are fixed before the release so I don’t know why people would try and spin this negatively.

More in-depth analysis always takes more time and comes out after the fact and AIB reviews.
 
Back
Top Bottom