NVIDIA Files for "Hopper" and "Aerial" Trademarks

[btarunr]

In a confirmation that a future NVIDIA graphics architecture will be codenamed "Hopper," the company has trademarked the term with the US-PTO. The trademark application was filed as recently as December 4, and closely follows that of "Aerial," another trademark, which is an SDK for a GPU-accelerated 5G vRANs (virtual radio-access networks). Named after eminent computing scientist Grace Hopper, the new graphics architecture by NVIDIA reportedly sees one of the first GPU die MCMs (package with multiple GPU dies). It reportedly succeeds "Ampere," NVIDIA's next graphics architecture.



View at TechPowerUp Main Site

 

[Xzibit]

Better headline would have been Nvidia to name chips after Disney characters Hopper from A Bug's Life & Aerial (Ariel) from The Little Mermaid.

 

[xkm1948]

Better headline would have been Nvidia to name chips after Disney characters Hopper from A Bug's Life & Aerial (Ariel) from The Little Mermaid.

Hopper (Grace Hopper) is the name of yet another mathematician being applied to architecture. She has been involved in groundbreaking projects, such as the Mark I super-computer (at the time) at Harvard University, and revolutionized computer science with the first high-level programming language COBOL.

 

[fancucker]

Will RDNA2/5nm be able to compete on an equal footing with 7nm+ Ampere? Seems very unlikely just from the process details. That's 30% free performance just from the node for nvidia. They could dedicate a greater portion of the die to tensor/RT and still ensure greater raster perf

 

[Xaled]

Scientists should not again accept their names being used by the greediest company in the world

 

[Berfs1]

aight fam RTX 4000 is gonna be named Einstein

 

[cucker tarlson]

Scientists should not again accept their names being used by the greediest company in the world

yes especially those who are long dead

 

[Xaled]

yes especially hose who are long dead

Their descendants, successors, etc.
I am not saying that Nvidia shouldn't be greedy, but just asking them to be honest, as honest as Mr Crap, just name them:
Money,
Dollars,
Wealth,
More money,
More dollars etc

 

[fancucker]

Scientists should not again accept their names being used by the greediest company in the world

Their infamy means their names are part of the public domain. Do you expect their successors to just whip up lawsuits against nvidia? If anything, it extends the legacy of those scientists and helps promote their work

 

[Space Lynx]

we hop all your telemetry to ALL YOUR BASE ARE BELONG TO US servers

can't wait for 6700 XT ^^

 

[fancucker]

Their descendants, successors, etc.
I am not saying that Nvidia shouldn't be greedy, but just asking them to be honest, as honest as Mr Crap, just name them:
Money,
Dollars,
Wealth,
More money,
More dollars etc

Ultimately the administration of these corps is accountable to the board/shareholders, not angry consumers who want special (and financially unrealistic) favours. AMD for instance can't placate fans with irrational and suicidal decisions because of their financial situation. Sure nvidia is greedy, but they're doing their job brilliantly.

 

[Xaled]

Their infamy means their names are part of the public domain. Do you expect their successors to just whip up lawsuits against nvidia? If anything, it extends the legacy of those scientists and helps promote their work

Promoting doesn't happen by using their names in a commercial product just for the purpose of promoting this product and the producer firm
Does using Einstein's name as a VR porn product promote his work? Just no

 

[fancucker]

aight fam RTX 4000 is gonna be named Einstein

Einstein practically redefined all macroscopic physics, dethroning Newton. His name should be reserved for the biggest paradigm shift in graphical computing history. Quantum GPU?

Promoting doesn't happen by using their names in a commercial product just for the purpose of promoting this product and the producer firm
Does using Einstein's name as a VR porn product promote his work? Just no

Distasteful Yes, but what do you expect their kin to do?

 

[fancucker]

All knowledge is built upon other knowledge over thousands of years. Einstein and Newton are equals, because Einstein would have been nothing without the pre-requisites in play.

You like my new sig @W1zzard nom nom nom (it's in response to Elon Musk saying this could all be a computer simulation)

Poincare, Planck, Lorentz and countless others had access to the information and legacy of Newton. None of them managed to produce a theory which is still being reconfirmed today. What Einstein did is one of the great triumphs of humanity.

 

[Noztra]

That's 30% free performance just from the node for nvidia. They could dedicate a greater portion of the die to tensor/RT and still ensure greater raster perf

So you think Samsung will be able to get 30% of "free" performance just by a node shrink?

This is from TSMC;
"TSMC claims their 7nm process will deliver a 20% performance improvement and a 40% reduction in power consumption "

And in reality TSMC perhaps only had around 15% more performance compared to 12/14nm, so there is no-way that Samsung will be able to get 30% more performance just by a node shrink.

 

[fancucker]

So you think Samsung will be able to get 30% of "free" performance just by a node shrink?

This is from TSMC;
"TSMC claims their 7nm process will deliver a 20% performance improvement and a 40% reduction in power consumption "

And in reality TSMC perhaps only had around 15% more performance compared to 12/14nm, so there is no-way that Samsung will be able to get 30% more performance just by a node shrink.

Read it again. Raster performance and not process. You really don't think going down to 7nm won't bring about 30?

 

[cucker tarlson]

Read it again. Raster performance and not process. You really don't think going down to 7nm won't bring about 30?

If going down to 7nm euv doesnt bring about 30% ppw improvement that'd be pretty sad.
We dont know tho.its possibile.

 

[Noztra]

Read it again. Raster performance and not process. You really don't think going down to 7nm won't bring about 30?

My mistake.

But I don't think it will bring 30% just because of moving to 7nm.

AMD moving their GPU from 12/14nm to 7nm and switching to RDNA didn't bring 30% more performance over Vega and that was a more mature node than Samsungs.

 

[R0H1T]

Without Newton, Einstein would have had nothing.

Yeah there's absolutely no way to prove that.

 

[R0H1T]

Of course but you're saying that Einstein would've been had nothing without Newton, that's like saying America wouldn't have been discovered without Columbus

 

[R0H1T]

I'm not sure what you're talking about, Gravity as a force or concept was known well before Newton. If it wasn't Newton we would've had someone else come up with some/many of his theories even if decades or centuries later. Einstein was no doubt at the right place & the right time, but his own success wasn't just contingent on Newton.

 

[BakerMan1971]

Here's me thinking Hopper was going to be an upside down mounted Stranger Things themed card
I'll get my coat...

 

[Turmania]

Debating about a name?!? no one here got a job?

 

[renz496]

My mistake.

But I don't think it will bring 30% just because of moving to 7nm.

AMD moving their GPU from 12/14nm to 7nm and switching to RDNA didn't bring 30% more performance over Vega and that was a more mature node than Samsungs.

for nvidia that 30% performance increase might not that unrealistic. remember they can get performance increase even without node shrink.

 

[fancucker]

If going down to 7nm euv doesnt bring about 30% ppw improvement that'd be pretty sad.
We dont know tho.its possibile.

My mistake.

But I don't think it will bring 30% just because of moving to 7nm.

AMD moving their GPU from 12/14nm to 7nm and switching to RDNA didn't bring 30% more performance over Vega and that was a more mature node than Samsungs.

Radeon VII is 331mm sq. vs 486 for Vega 64, still not an apples to apples architectural comparison but still better than RDNA. It achieved 30% improvement with 1.56x density at TSMC 7nm.

Assuming that Samsungs 7nm+ is within range for 7nm TSMC (not +), its not a bad bet. My guess is that if it equals or exceeds TSMC's regular 7nm they're going to ease the brakes on competition with AMD and allocate greater die area to RT/tensor components. This is a good window to not only cement the technology but build an unassailable lead over AMD, increase its adoption and make it more justifiable. Even quash AMD's initial RT response. Never a better time for nvidia to do this.

AMD needs to knock RDNA2 out of the park at 5nm. One of their greatest ever otherwise the market will suffer