Initial Performance RTX 2070 Max-Q edition leaks - faster than Vega 64 and GTX 1080?

Published by

Click here to post a comment for Initial Performance RTX 2070 Max-Q edition leaks - faster than Vega 64 and GTX 1080? on our message forum
https://forums.guru3d.com/data/avatars/m/243/243702.jpg
Was DLSS involved? That's the question.
https://forums.guru3d.com/data/avatars/m/209/209146.jpg
Hmm not bad, was looking for the 2070 but from the 1070 and 1070 Max-Q I suppose it would be positioned higher than the mobile version. (Well yeah whatever the Max-Q does it's still a mobile chip.) The Vega tests are curious too, Vega 56 and then Vega Frontier (Slower clocked, 1440 Mhz isn't it?) for the Vega 64 then? Although for Final Fantasy 15 NVIDIA should be a pretty strong performer and I don't think DLSS would be involved where I guess 1920x1080 was the tested resolution, it kinda requires 3840x2160 does it not and then scaling from 2560x1440 or something like that. EDIT: No there's three Vega entries with funnily enough the Vega and Vega frontier right on top of one another and no numbers other than two "Vega" and one "Vega Frontier" hmm then again Vega does cap around 1500 - 1550 Mhz for clock speeds unless memory is also brought up and it's still 4 - 6% or so gains all in all so it's not like it's going to completely change it's performance target or anything. (Well I suppose a benchmark run with one of the tested GPU's until you land within a similar score would narrow down how this was tested.) Also Max-Q, guess it's supposed to be Max Quality or some such and silly names isn't anything new for hardware. (Military grade...)
https://forums.guru3d.com/data/avatars/m/226/226864.jpg
It makes sense that they perform faster than the Pascals, although I'd like to know what the prices will be compared to Pascal based models.
JonasBeckman:

Well yeah whatever the Max-Q does
Going by what Nvidia did with their Pascal Max-Q variants, Max-Q just is basically an underclocked and undervolted chip that throttles quicker and has a certain requirement for the quietness of the cooling system used for the card. Basically, a Max-Q is quite a bit slower than the regular variant with the same number but should run cooler/quieter due to it (no surprise there) and usually comes with a premium price for the hip "Max-Q" label. It's pretty smart by Nvidia, si nce they can take the chips that can't run at full clocks reliably for the regular cards and can still sell them labelled Max-Q, often at a premium price.
data/avatar/default/avatar34.webp
JonasBeckman:

Hmm not bad, was looking for the 2070 but from the 1070 and 1070 Max-Q I suppose it would be positioned higher than the mobile version. (Well yeah whatever the Max-Q does it's still a mobile chip.) The Vega tests are curious too, Vega 56 and then Vega Frontier (Slower clocked, 1440 Mhz isn't it?) for the Vega 64 then? Although for Final Fantasy 15 NVIDIA should be a pretty strong performer and I don't think DLSS would be involved where I guess 1920x1080 was the tested resolution, it kinda requires 3840x2160 does it not and then scaling from 2560x1440 or something like that.
Just look yourself on this doubtful results (OC or not we don't know) but run was at 4k res. http://benchmark.finalfantasyxv.com/result/?i=18155881045c209834edbc9&Resolution=3840x2160&Quality=High and here is mine on underclocked and undervolted mobile gtx 1080 https://i.imgur.com/O3hjIx4.jpg
JonasBeckman:

EDIT: No there's three Vega entries with funnily enough the Vega and Vega frontier right on top of one another and no numbers other than two "Vega" and one "Vega Frontier" hmm then again Vega does cap around 1500 - 1550 Mhz for clock speeds unless memory is also brought up and it's still 4 - 6% or so gains all in all so it's not like it's going to completely change it's performance target or anything. (Well I suppose a benchmark run with one of the tested GPU's until you land within a similar score would narrow down how this was tested.) Also Max-Q, guess it's supposed to be Max Quality or some such and silly names isn't anything new for hardware. (Military grade...)
It mean max-quiet https://www.geforce.com/whats-new/articles/nvidia-geforce-gtx-max-q-laptops https://www.notebookcheck.net/Nvidia-Max-Q-limits-fan-noise-to-40-dBA-when-gaming-so-why-are-we-recording-louder-results.258636.0.html
https://forums.guru3d.com/data/avatars/m/271/271560.jpg
yes, "Max-Q" is undervolted and underclocked. it is still the best equation for battery life on mobile. i have a full, fat 1070 in my laptop, with a modest tweak. it eats through battery life like a starving man at a buffet. i chose not to get "Max-Q"...but now with the RTX (afaik) only Alienware will offer full, fat 20xx OR Max-Q. everyone else is on board with the compromise.
https://forums.guru3d.com/data/avatars/m/198/198862.jpg
When did Final Fantasy XV start to represent general performance of the GPU's? Tittle is filled with nvidia features.
https://forums.guru3d.com/data/avatars/m/226/226864.jpg
tunejunky:

yes, "Max-Q" is undervolted and underclocked. it is still the best equation for battery life on mobile. i have a full, fat 1070 in my laptop, with a modest tweak. it eats through battery life like a starving man at a buffet. i chose not to get "Max-Q"...but now with the RTX (afaik) only Alienware will offer full, fat 20xx OR Max-Q. everyone else is on board with the compromise.
I'd assume that many high-end Clevo-based desktop replacement Notebook models will also come out with full 20xx chips on MXM add-in cards.