Hi everyone, reviews said that 3090 achieve 280% more fps than 3080 with 8K resolution.
Do you think then that 3090 will have 280% more performance for 8Kx? If that it’s true then the 3090 has a better cost/performance.
Ok I’m confused…but it doesn’t take much these days…
4 Likes
I doubt it will be quite that much. 8K res has twice as many pixels as 2 4K res panels. I think a large part of this difference is due to the required video ram. The 8KX will need less screen buffer ram (the amount will vary due to your super sampling settings). Depending on the game, you could see a major difference in speed, which could probably be alleviated by reducing the graphics quality settings to reduce texture size and geometry.
I think the better choice would be to wait for the 20GB 3080, which should be a lot cheaper and won’t run out of ram, like the 10GB version will.
6 Likes
First off, gaming at 8K resolution has nothing to do with the Pimax 8KX. The 8KX is just a marketing term for a 4K display in each eye. Reason the 3090 is much faster at 8K gaming is the high VRAM but basically no one can afford 8K gaming monitors so it’s really dumb of Nvidia to market the card that way.
But the 3090 is roughly 15 to 20% faster than the 3080 which makes quite a difference in VR, where a bit of extra framerate is the difference between smooth gaming performance or lots of reprojection. Basically I believe with a 3090, virtually all VR games may be able to run at a smooth 75Hz and graphics maxed out without any problem.
The VRAM would be very helpful in Skyrim VR when you mod it to use 8K textures as that would use more.
10 Likes
This. Listen to this man.
I still got an ASUS RTX 3080 TUF OC Gaming card though. That’s 6% faster than the FE. I’ll see how that performs. Cannot justify ~10% performance increase to the 3090 for 1000USD extra.
Once it’s here.
And the 8KX.
Yeah.
Good night.
P.S. I hope you guys can get your cards soon!
4 Likes
Hmm but sure I would want to blow money on this personally. I mean you could save that money for the 4000 series 
3 Likes
Edit:
The capacitor"issue" was a false alert, and fixed by software without downclocking.
Don’t believe YouTubers I guess , hahaha
4 Likes
No one goes to gamer nexus for the benchmarks, we all go to look at his hair.
11 Likes
I wonder what shampoo he uses. Could learn from it.
6 Likes
which is funny because if you look at the Codemasters interview on Dirt 5 by Digital Foundry , it seems that they wen to great lengths to support features like 120Hz on Televisions.
They think this niche market worth developing for , but completely ignore adding VR support into the game. WTF?
6 Likes
The video you mentioned was actually a surprise for me as I would not expect someone in the corporate game dev to be so enthusiastic about the optimizing their engine 
. I have never played Dirt (or any other race sim apart from Forza), but I am definitely going to watch this game to see how it will fare.
This should be used as a “motivation video” for people from ED, Asobo, Frontier, etc.
3 Likes
Except its not a “sim” Thank god. I have enough of those.
So, the capacitor alert was a false alert. On the stability issue, to summarize:
All cards were affected, Asus TUF as well.
The big caps on the backs of the cards are high quality 470 spcaps, NOT poscaps.
A mix of spcaps and mlcc is generally better than going all mlcc.
The stability issues were not capacitor induced, but a regulation issue fixed by today’s drivers
People cancelled their working orders on Zotac and Gigabyte for nothing.
Jayz2c might get sued (?) Because he portrayed it as true, Igor just suspected AND is safe in Jerryland, Make ready the popcorn.
2 Likes
I agree, while the “TDP” of a 3080 is 320 W against a 3090 with 350 W. Which is a difference of about 10%. This can mean: even as the 3090 card has much more processor units, the card will slow down while it will overheat.
In the end: nothing to win for a very high price.
Are You sure about that? If that’s the case ASUS cards just went from best bet to “worst”?
Nobody knows but an Nvidia PDN engineer. I hate seeing so many people call poscaps cheap too. Tantalum caps are not cheap!
As a general statement it may be true. In my experience designing boards in sensitive applications multiple bypass capacitor values are used in parallel, because low value caps respond better to higher frequencies. Placing them in parallel lowers ESR, raises capacitance, and lowers inductance. By using different values there is better response over frequency. Poscaps can fit a whole hell of a lot more capacitance per package size than MLCCs, and have excellent frequency response, just not quite as good
I don’t think even the best educated engineer can make a quality statement on the matter because we just don’t know the types noise that board sees.
6 Likes
Seems way to early to be claiming POSCAP vs MLCC was a false alert. Most of the indepth testing shows clear correlation and also shows it is only part of the issue. Obviously downclocking - or rather not clocking higher than they should have in the first place - will resolve issues. Driver stability will also help. ASUS Tuf was never affected out of the box in any of the benchmarks I saw - and I tried to see as many as I could. It also was more resilient to increased frequencies under the launch day conditions. But yes - you overclock it and it is gonna drop. Would be nice to see how much thermals play in all of this. Anyway - in a year we will have a much clearer idea of what is better value for money.
2 Likes
A lot of people reported crashing issues with the TUF on Reddit as well. Just because you didn’t see it doesn’t make it false.
1 Like
Agreed. I can only go off the information I find and assess whether I think it is legit. Also - like I said in another post - doesn’t matter which card you get - silicon lottery - there will always be lemons. Though I am looking for these reports and I can’t find them. Can you link please?
3 Likes
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.