Nvidia GeForce RTX 40x0 rumors and speculation

NVIDIA's official specs for the 4070Ti Super listed it as having 48MB L2. Is it possible this is a mistake, same as when they listed 36MB for the 4070 Super? I'm inclined to say it is a mistake since the 4070Ti did sport 48MB and they may have copy pasted the specs they thought weren't changing.

Now I can't find mention of the cache config on NVIDIA's website. Seems they may have removed it.
Based on reviews I've seen the 4070 Ti Super does indeed have 64MB L2. (y)

Also the MSI Ventus 3x 4070Ti Super is possibly messed up or something according to Techspot. Up to 5% slower than other cards.
 
Last edited:

Even with the disclaimer, the results seem more or less corroborated with other outlets. And it's just.....bizarre. I honestly cannot understand the performance here. Just looking at specs, this thing shouldn't be that far off a 4080, but instead is quite a bit behind in almost all cases, and manages to only be a little bit faster than the regular 4070Ti.

Could it really be that shader-limited in almost every situation?
 

Looks like 4070ti is still getting bios updates to fix performance, even a new one this morning. Doesn't sound like it was ready to launch. I'd hold off on buying/seeing where performance lands.
Problem is only with the Ventus 3x? Techspot results sure do look fishy.

One explanation of the poor scaling vs 4070Ti is that Ada's huge L2 works really well. So adding a bunch of memory bandwidth doesn't make much difference. AD103/AD104 might be almost entirely compute limited as a result.
 
Looks like 4070ti is still getting bios updates to fix performance, even a new one this morning. Doesn't sound like it was ready to launch. I'd hold off on buying/seeing where performance lands.
Is that affecting only that model? Are reviews showing those models lower than other similarly clocked models from other brands?
 
Is that affecting only that model? Are reviews showing those models lower than other similarly clocked models from other brands?

Don't know. Digital Foundry holding back their review for the reason with the bios.


Hardware Unboxed used the MSI Ventus 3X for all of their test results. Seems to be the case for any of the reviews I've seen. I don't think BIOS udpates for GPUs are super common. Seems a bit of a disaster launch.
 
Anybody know what's wrong with it?

Four different VBIOS versions before the one that was released this morning.


Who knows if they'll ever actually increase performance in a meaningful way. Sounds like they're churning through bios versions trying to make it faster, but I don't think anyone (in the press at least) knows exactly what the issues they're trying to fix are besides general performance.
 
That was not the point. With a 30W lower power limit the 4070 Ti super will always be limited. At least it is 10% faster than a 4070 Ti and closed a cap in the lineup.
How is it not the point? GPU's aren't just infinitely limited by power. 285w is a whole lot of power to put into a GPU of this level(which would be more roughly like what the 3060Ti was last generation), so the idea that it's somehow heavily power-limited just doesn't sound right at all.
 

Even with the disclaimer, the results seem more or less corroborated with other outlets. And it's just.....bizarre. I honestly cannot understand the performance here. Just looking at specs, this thing shouldn't be that far off a 4080, but instead is quite a bit behind in almost all cases, and manages to only be a little bit faster than the regular 4070Ti.

Could it really be that shader-limited in almost every situation?
Techspot / HUB did the MSI Ventus and concludes that it is just 3% faster than regular 4070 Ti at 1080p and 1440p, 6% faster at 4k, and just 4% in raytracing.

Then there's Computerbase looking at the Asus Super TUF and saying between 7 and 10% faster at 1080p to 4k, and 6-14% in raytracing.
Tom's Hardware saying Super TUF 6% at 1080p, 7% at 1440p and, 10% at 4k, and 5-10% in raytracing.

Both of these cards use reference clocks.

Is it mostly the Ventus that is responsible for HUB's low tally?
 
Techspot / HUB did the MSI Ventus and concludes that it is just 3% faster than regular 4070 Ti at 1080p and 1440p, 6% faster at 4k, and just 4% in raytracing.

Then there's Computerbase looking at the Asus Super TUF and saying between 7 and 10% faster at 1080p to 4k, and 6-14% in raytracing.
Tom's Hardware saying Super TUF 6% at 1080p, 7% at 1440p and, 10% at 4k, and 5-10% in raytracing.

Both of these cards use reference clocks.

Is it mostly the Ventus that is responsible for HUB's low tally?
Maybe, or just the small-ish selection of games chosen.

Either way, all of these results are lackluster.
 
I'm not sure what results you're expecting? It's 10% more SMs with the same power limit. There's also going to be a lot of tests, especially at 1080p, that were already at diminishing returns with the 4070ti even due to likely stalling outside of the GPU.

This was always likely going to be the situation -

4070s - largest performance gain

4070ti s - VRAM

4080s - price drop
 
It's not as if it's a new generation so don't think anyone is expecting huge performance increases.
 
It'd be interesting to see it tested in a maxed out Ratchet & Clank. That game runs into clear vram limits on the 4070ti at max settings.
Computerbase has R&C numbers.
8.6% diff at 1080p with Raytracing, skyrocketing to 44.5% at 4K. It looks like DLSS Quality is used at both resolutions.

So yeah, the extra 4GB can certainly make a huge difference under the right circumstances.
 
Back
Top