• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

NVIDIA GeForce RTX 5080 Founders Edition

When you put them in heaviest situations out there, the 5090 is nearly double the fps.
In any other situation, 5090 is a victim of a component that causes bottleneck.

View attachment 382418
View attachment 382419
+61% and +64% more performance at 4K RT is pretty good, but there is still a lot of performance left on the table since it has 2x more CUDA Cores and ~87% more Memory Bandwidth...
 
I got a question…
How likely are we to see 50 series super cards coming next year?
I'd say very. Only the 5080 is based around a fully enabled die as far as I know. All the rest of the lineup are cut-down chips. That's probably what the "shortage" is about. Nvidia is saving differently binned chips for the Super cards.
 
Not sure which wiki link you're looking at but this wiki page has a nice table outlining all the architectures and instruction sets.



R700 was when they started the small die strategy with the 4850/4870/4870X2. Evergreen was the successor and continued the same strategy but with larger dies (256mm > 334mm) with the 5850/5870/5970. Then came nothern islands which was VLIW4 for only the 69xx but the other models were rebrands of Evergreen. There was also a naming change, with 5870 effectively being replaced with 6970. All of these architectures used the Terascale instruction set with slight modifications depending on the iteration. They were all very successful too, especially the 5xxx generation.

Then Southern Islands came along with the HD 7xxx's and all used the GCN instruction set for almost the next decade. Heyday of GCN was the 7000 and 290x launch times when they were much faster and had a better/more forward looking arch than nv at the time. Polaris was released after a few years when GCN was long in the tooth and was a small die priced cheaply to tide them over till RDNA. They did release two more flagships though with HBM of which the last GCN, Radeon VII, was the pick of the HBM bunch and a swansong to the first arch AMD designed after taking over from ATI.
I remember despite their problem, i desperately wanted a pair of 6970s.

Worth noting the 5970 was a dual GPU card, where as the 6970 was a single GPU, with the dual GPU being the 6990.

This was peak GPU

Diamond_6970PE52G_Radeon_HD_6970_PCIE_1322054677_828829.jpg
 
I'd say very. Only the 5080 is based around a fully enabled die as far as I know. All the rest of the lineup are cut-down chips. That's probably what the "shortage" is about. Nvidia is saving differently binned chips for the Super cards.

I think it has more to do with B100 using the same node and being able to sell them for 30k each as fast as they can make them.
 
I think it has more to do with B100 using the same node and being able to sell them for 30k each as fast as they can make them.
Or maybe both.
 
I remember despite their problem, i desperately wanted a pair of 6970s.

Worth noting the 5970 was a dual GPU card, where as the 6970 was a single GPU, with the dual GPU being the 6990.

This was peak GPU

View attachment 382437

I was pretty torn between the GTX 580 and 6970 so much so I grabbed a 7970 on launch day. Truly the good old days.

I ended up going 580 but I remember them being pretty damn close in Crysis my whole reason for wanted to grab one for 1200p gaming...
 
I said half a 5090, not half the performance of a 5090.
Fair enough, my mistake. Though you have to admit, that's an easy mistake to make given that the general topic of discussion is the performance.
HALF (well, technically a little less than half)

In fact, the only thing that's not physically halved is the ROP count.
Ok, I see what your talking about now. And this goes along with the statement of this should not have been a 5080, but instead a 5070 or 5070ti.

A thought about that, perhaps there's a yield problem at TSMC?
 
Or maybe both.

I'm sure they always have plans just in case AMD catches them with their pants down and while I don't think they really care that much about gamers I do think Jensen does not want to lose to AMD even if it's just a perception thing. He likes to steal AMDs thunder every chance he gets lol.
 
Could that be L2 cache size not doubled hence the bottleneck?

Or someone could test in native 8K maybe the gap between 5080 and 5090 will widen.
I've always wondered how much performance the 4090 would have had if it had the full 96MB of L2 Cache... Because 72MB was just 12.5% more than the 64MB of the 4080/SUPER ! 96MB would have been 50% and we all know the 4090 has 60% more CUDA Cores.
Also the 5090 has 96MB of L2 Cache but the full GB202 die is 128MB, maybe that's why the performance is not scaling that well on both 4090 & 5090.
 
Does anyone know if someone has tested the RTX 5080 at Pcie 4? Is the performance drop still only 1% or is it lower/higher? I'm worried because of der8auer's review if there are issues running the FE model at Pcie 4 speeds.
 

Validation for the 24Gb GDDR7 in next-generation AI computing systems from major GPU customers will begin this year, with plans for commercialization early next year. - posted October 17, 2024
Again, ready for for the 5000 refresh or whatever, but they were not ready to go to go into these GPUs
 
A thought about that, perhaps there's a yield problem at TSMC?
You mean for TSMC 3nm node or for Blackwell dies (4nm) in general ?

Again, ready for for the 5000 refresh or whatever, but they were not ready to go to go into these GPUs
The refresh will also get more VRAM, like the 5080 Ti will probably get 3GB GDDR7 chips aka 24GB total. But I wonder if Nvidia will release a 5090 Ti then with 48GB GDDR7 to keep 2x more VRAM on x90 GPUs.
 
Paul did a nice summary on this GPU, definitely worth watching:
 
I'm sure they always have plans just in case AMD catches them with their pants down and while I don't think they really care that much about gamers I do think Jensen does not want to lose to AMD even if it's just a perception thing. He likes to steal AMDs thunder every chance he gets lol.
You always need at least a duopoly to maintain the illusion of choice. Coca-Cola and Pepsi, Chevy and Ford, Android and iOS, etc. Without competition, there's no need to improve, and without improvement, there's no incentive to buy the next best shiny new thing.
 
You always need at least a duopoly to maintain the illusion of choice. Coca-Cola and Pepsi, Chevy and Ford, Android and iOS, etc. Without competition, there's no need to improve, and without improvement, there's no incentive to buy the next best shiny new thing.

Keeps them out of legal trouble also lol.... I mean the US looked at Nvidia just because they have an insane lead in AI...
 
The sad thing is, we won't even know how bad the situation is.

Even if the cards reach the stores at +30 to +100% pre-scalped levels and just sit on the shelves, and Mindfactory sales show they aren't moving even at the Lovelace levels (which were at times outsold by AMD cards), Nvidia can just show the quarterly revenue as record high!

DeepSeek fiasco won't put a dent in Nvidia's revenue, as they have allegedly sold all the AI acceleration hardware they can build this year. And as for it being illegal to doctor the revenue numbers? Either dismissed, or laughably cheap to breach - and since Nvidia changed the sector name to "Gaming and AI PC", they can legally include all kinds of non-Gaming related stuff.

That's what you get when you expect a nice toy from a company that doesn't want to be associated with toy making.
 
I remember despite their problem, i desperately wanted a pair of 6970s.

Worth noting the 5970 was a dual GPU card, where as the 6970 was a single GPU, with the dual GPU being the 6990.

This was peak GPU

View attachment 382437
Man I got a 6990 engineering sample gifted to me from a friend who worked at AMD at the time and it was mega. Like the thing couldn't handle the heat from two GPU's on a screaming blower maxed out. And when a game supported CF it was time to max shit up on a multiple monitors.

Fun times.
 
Unsurprisingly the 5080 is very very.. mid. Not even really surprised. 4090 owners are rejoicing though for sure. This year is looking to be a badddd year for GPU's at this rate. 5070Ti and 5070 hopefully will bring everything back to norm.
 
Man I got a 6990 engineering sample gifted to me from a friend who worked at AMD at the time and it was mega. Like the thing couldn't handle the heat from two GPU's on a screaming blower maxed out. And when a game supported CF it was time to max shit up on a multiple monitors.

Fun times.
I remember my 6950, it was a great card and you could unlock it, if you knew how. :D

Battlefield - Bad Company 2 was a blast, damn, I miss those days, such a good time.

Unsurprisingly the 5080 is very very.. mid. Not even really surprised. 4090 owners are rejoicing though for sure. This year is looking to be a badddd year for GPU's at this rate. 5070Ti and 5070 hopefully will bring everything back to norm.

I don't think this normal thing exists.
 
I wonder if the game developers are kind of disappointed with these kinds of releases? I imagine they plan certain effects in games for hardware to catch up when they are on a long development - but if mainstream releases (ignore the RTX 5090) only bring 10 - 15% improvements in two years, it's basically the same, two years of waiting for anything to maybe happen?

Especially RayTracing - it is still so taxing in a lot of games that people just turn it on to check how it looks, and then comfortably game with it turned off...

Or what about VR hardware makers? It's mostly very taxing to run high FOV, high resolution on modern games, and various tricks like foveated rendering with eye tracking have their own issues - it has to run on higher frame rate, or you often look at a blurred out of focus area that has to be refreshed to full resolution... Another generation lost?
 
I don't think this normal thing exists.
Norm, as in, actually something to be excited about. In complete fairness I stopped caring about Titan / xx90 and xx80 cards a long time ago, so I wasn't super excited for the 5090 and 5080 in the first place, but I still didnt expect the 5080 to fall as short as it did. As long as the mid-range products this generation aren't as mid as the 5080 (which if our guesstimates are anything near accurate; they hopefully wont be), I think things may be okay. But we'll see I guess.

I wonder if the game developers are kind of disappointed with these kinds of releases? I imagine they plan certain effects in games for hardware to catch up when they are on a long development - but if mainstream releases (ignore the RTX 5090) only bring 10 - 15% improvements in two years, it's basically the same, two years of waiting for anything to maybe happen?
I mean to be fair that's why most big companies make games for consoles rather than PC (and usually port them later to PC, a good example is some of the sony exclusives) Some developers, like Valve back around the release of Source, target entry to mid range hardware and try to stretch the best they can out of that. Unfortunately we are getting way less developers doing that now it seems

I think the only people really getting hit hard by this are smaller studios that aren't necessarily AAA but not indie either.


I got a question…
How likely are we to see 50 series super cards coming next year?
You know, as much as I like the 40 supers and the 20 supers, I dont really want there to have to be product refreshes like that every single generation. Theres obviously a practical reason to do product refreshes, which is what the supers mostly are besides some tweaked performance. If they're gonna do 50 supers, they better look as awesome as the 40 supers and 20 supers did. If your gonna be a product refresh, atleast make the FE look good lol.

If I had to hazard a guess; its certainly possible, but I hope it doesn't happen unless somehow the 5070Ti and 5070 are just meck.
 
Last edited:
I was guessing one digit improvement. 12% @4K is not far off what I anticipated. Pretty mediocre but then again, considering the specs, really was not expecting a lot of this one. Also the 5090 is not so much of an improvement so where would all the other stack. It's like super versions or Ti version of the 4000 series though we already got those.
 
Back
Top