Man looks good to me if it's true. Makes me wonder if AMD will go in a similar direction getting rid of the blower style reference cooler. Makes me wonder what the difference in temp would be compared to the AIB partner cards such as MSI and EVGA etc. Since there was always a huge difference in temp compared to the blower style cooler to the 2 and 3 fan coolers that AIB partners used.

RavenMaster:

Looks nice but why would they go with a dual fan design when the Titan V and Quadro don't require two fans to cool those cards adequately?

1) quieter 2) cooler 3) SFF cases (unlike the Titan and Quadro...i mean you can, but who does that with those cards?)

so who is going to sell their 1080ti for one of these ?

@igorfiuza lol, look up how many benches almost always show better performance when cards are in a x8 slot (vs x16). tested a few cards (50/ti/60/70/80) on 2 rigs and its always a tick faster on x8. and you might need to read up how much BW a 3.0 slot has at 8x (and what cards are actually using).. @Evildead666 dust part is irrelevant on any rig with forced-air-cooling. almost all dust is dead skin (particles) which you wont be able to remove (unless using hospital/bio lab grade air filter). so ANY airflow (even with just one fan) will introduce dust, one reason i chuckle when ppl talk about only buying cases that have dust filter/want the rig to be "positive" on air pressure, as its not preventing dust to get inside, just makes the particles finer and prolongs the time i can use it before dust builds up... another reason i prefer a silent (read low rpm setup), as less airflow means less dust (entering rig).

fry178:

@igorfiuza lol, look up how many benches almost always show better performance when cards are in a x8 slot (vs x16).

Can you cite this? I can't see why that would be the case, and at least according to benches done by gamersnexus it isn't.

Multiple sites have tested, especially with lots of ppl having newer cards and older boards. Iirc hilbert did it as well, but think he was doing it to look at pcx 2.0 bottlenecking big cards. Didn't really believed it until i tried it myself, and cards were consistently getting higher results for 3d mark (06 and 11). Haven't messed with games, the synthetic benches are just more consistent (scene/gpu load etc).

fry178:

Multiple sites have tested, especially with lots of ppl having newer cards and older boards. Iirc hilbert did it as well, but think he was doing it to look at pcx 2.0 bottlenecking big cards. Didn't really believed it until i tried it myself, and cards were consistently getting higher results for 3d mark (06 and 11). Haven't messed with games, the synthetic benches are just more consistent (scene/gpu load etc).

Not exactly on topic, but I would think that, litteraly overclocking on the razor's edge, dropping the PCIe link to 8x would lower the power used by that link, and that extra power could be used by the GPU. That would only be in extreme overclocking cases though, and power limited cases.

Just saw the prices. Nvidia is the new Apple.