Nvidia Ampere Discussion [2020-05-14]
- Thread starter Man from Atlantis
- Start date
-
- Tags
- nvidia
That supposed 3090 is huge. (later in comments he corrects himself saying $1400)
Bondrewd
Veteran
I sure hope so.
:^)
So... Either it's a monster and they don't f... around, or it's a nv30 remake
And back with the Ultra name again
I like VRAM amount is an option at the mid and higher tiers.
https://wccftech.com/nvidia-rtx-300...or-799-rtx-3070-for-599-and-rtx-3060-for-399/
https://wccftech.com/nvidia-rtx-300...or-799-rtx-3070-for-599-and-rtx-3060-for-399/
Last edited:
If the rear fan spins fast enough maybe it'll provide sufficient thrust to keep the tail-end from sagging.
I suppose that's the benefit of having the triple slot mounting plate and what looks to be like a fairly rigid construction. In any case, I'm glad I have one of those 90deg rotated Silverstone cases so the vid card sits vertically.
I suppose that's the benefit of having the triple slot mounting plate and what looks to be like a fairly rigid construction. In any case, I'm glad I have one of those 90deg rotated Silverstone cases so the vid card sits vertically.
Frenetic Pony
Veteran
Well you're in luck because that seems to be every mid tier GPU from Nvidia and AMD this year. It's just how it works out compared to consoles really. I wonder how much ram you'll need for the 8k textures in the Crysis remake, full 24gb, more? Doesn't seem like they've changed the engine around for high end SSD streaming so, could be.
The cooler, apart from being quite large and a 3-slot (which AFAIK is also a first for single-GPU reference cards), still seems odd to me. Especially the back side, where you cannot see any details (HS structure) through the fan blades, which you can for the front fan.
Qualcomm ported the SD845 from 10nm LPP to 7nm DUV. Transistor density went up 50% but efficiency for the GPU only ~20%:
https://www.anandtech.com/show/14072/the-samsung-galaxy-s10plus-review/10
The efficiency gain comes from a 50% wider GPU with a 27% lower clock rate. Samsung claims that 8nm LPP is 10% denser and 10% more efficient. Samsung is producing 10nm LPP chips since three and 8nm LPP since two years. Everybody should know the quality of the process...
Bondrewd
Veteran
Check 765G vs Kirin 990
Panino Manino
Newcomer
Something interesting.
Jensen, always the egocentric, said at the GeForce2 launch that the card was a "major step toward achieving" the goal of “Pixar-level animation in real-time”. Some people at Pixar really didn't liked heading this:
"Do you really believe that their toy is a million times faster than one of the cpus on our Ultra Sparc servers? What’s the chance that we wouldn’t put one of these babies on every desk in the building? They cost a couple of hundred bucks, right? Why hasn’t NVIDIA tried to give us a carton of these things? — think of the publicity milage [sic] they could get out of it!"
Besides getting mad at Jensen, they're explained how long they predicted a GeForce would be able to offer the quality they had achieved in real time, 20 years.
"At Moore’s Law-like rates (a factor of 10 in 5 years), even if the hardware they have today is 80 times more powerful than what we use now, it will take them 20 years before they can do the frames we do today in real time. "
What it's interesting is that the GeForce2 was launched in September 2000, exactly twenty years ago.
The cards may be expensive, but what about the jump in actual power and visual fidelity? Will Nvidia reach that "goal"?
I wonder if Jensen remembers that.
Jensen, always the egocentric, said at the GeForce2 launch that the card was a "major step toward achieving" the goal of “Pixar-level animation in real-time”. Some people at Pixar really didn't liked heading this:
"Do you really believe that their toy is a million times faster than one of the cpus on our Ultra Sparc servers? What’s the chance that we wouldn’t put one of these babies on every desk in the building? They cost a couple of hundred bucks, right? Why hasn’t NVIDIA tried to give us a carton of these things? — think of the publicity milage [sic] they could get out of it!"
Besides getting mad at Jensen, they're explained how long they predicted a GeForce would be able to offer the quality they had achieved in real time, 20 years.
"At Moore’s Law-like rates (a factor of 10 in 5 years), even if the hardware they have today is 80 times more powerful than what we use now, it will take them 20 years before they can do the frames we do today in real time. "
What it's interesting is that the GeForce2 was launched in September 2000, exactly twenty years ago.
The cards may be expensive, but what about the jump in actual power and visual fidelity? Will Nvidia reach that "goal"?
I wonder if Jensen remembers that.