Rumor: NVIDIA GeForce Ampere to be fabbed at 10nm, all cards RTX ?

Venix

2020-03-13 08:51

True or not does not really matter , the cards will be judged based on their performance /price/ consumption ... If they are on 7 nm or 10nm .....hell if it delivers will not matter if it is on 1mm 😛, but for the love of god Nvidia i hope the 3060 will get at least 8gb ram .... On the 1060 6gb where fine 2060 questionable but ...ok on a 3060 6gb will not allow the card to age that well

#5769105

Astyanax

2020-03-13 09:20

nevcairiel:

And where is the proof that any twitter account is any specific person?

Why are you owed that information? Leakers can't leak if you can trace them.

#5769110

UnrealGaming

2020-03-13 09:30

THE CURSED 10nm https://cdn.frankerfacez.com/emoticon/214681/1 AMD might have a chance to catch up after all

#5769131

Netherwind

2020-03-13 10:45

..those expecting a leap in rasterizing are gonna be let down

We've been let down the last few iterations anyway so one more time around wouldn't be a surprise. I expect the usual 20-25%. Let's just hope we don't get another leap in pricing, that last one was sour AF.

#5769154

nevcairiel

2020-03-13 12:18

Astyanax:

Why are you owed that information? Leakers can't leak if you can trace them.

Leakers also can't be trusted if they could be any random troll with no mention of sources or evidence, or even a confirmed history of leaks. (and "same person with new account" is not proof of anything. How do you know thats the same person? Coming back to my original question.)

#5769180

asturur

2020-03-13 13:43

Khronikos:

Well if this is true then Nvidia does not feel competition in the sector. 10nm would probably be a skip for me with a 1080ti. I don't need to spend money anyway. PS5 is coming out soonish.

How does 10nm compare to a 1080ti and with a console? I mean AMD is at 7nm and no AMD card is a current upgrade over a 1080ti, also big navi is still 7nm and could be an upgrade of some sort. Why people give so much importance to nm? AMD 8 core CPU at 7nm are as good as intel 8 core at 14nm.

#5769405

Astyanax

2020-03-14 10:50

Hyperus:

Generally speaking, whenever I play BFV with RTX on ultra, I see a 100% 3D GPU Usage in my task manager, suggesting that the RT cores really are not the limit here.

the actual bottleneck in raytracing is the availability of Int processing.

#5769427

pharma

2020-03-14 12:39

Could you get the stated performance increase (70-75%) in Indiana's Big Red 200 if Ampere were in fact at 10nm instead of 7nm?

It is also mentioned that Big Red 200 gained an additional 2 petaflops of performance even though it uses a smaller number of GPUs than the Volta V100 based design. The reason for going with a smaller number of next-generation GPUs is simply because they offer 70-75% better performance than existing parts and by that, we are comparing it with Volta-based Tesla V100 GPUs as no Tesla GPUs based on the Turing GPU architecture, aside from the Tesla T4, exist. ... In a previous interview, NVIDIA's CEO, Jensen Huang, had confirmed that the majority of the orders for their next-generation 7nm GPU will be handled by TSMC while a small portion will be sent to Samsung for production.

[SPOILER]https://wccftech.com/nvidia-next-gen-ampere-gpu-75-percent-faster-existing-gpus/[/SPOILER]

#5769442

Andrew LB

2020-03-14 14:47

The leather jacket was pretty clear that Ampre would be made using TSMC 7nm.

#5769459

Astyanax

2020-03-14 16:03

Hyperus:

An integer is a whole number, there is no fraction there, it lacks accuracy to a high degree and is generally not all that used in graphics processing, simply due to accuracy constraints.

DXR (and RT cores) are implemented as (fixed) Integer math, not floating point, only the resulting shading of the post calculated scene is float point. Pascal can't do int and fp in parallel which is where their performance tanks, Volta can which is why it can do RTX without RT cores, Volta's general purpose Int cores have to handle both BVH and pre-calculation (like pascals) but the concurrent execution permits it without significant overhead.

#5769583

Venix

2020-03-15 00:31

asturur:

AMD 8 core CPU at 7nm are as good as intel 8 core at 14nm.

sure if you take out completely the power consumption !

#5769634

Astyanax

2020-03-15 05:44

Hyperus:

Tensor cores are the other thing that sets Volta apart, we can do vector math in 3D space with matrices.

The tensor cores are not used at all in DXR on Volta, and are not exposed to the system outside of CUDA and NVAPI Denoising.

#5769764

nizzen

2020-03-15 16:56

Venix:

sure if you take out completely the power consumption !

When did someone actually care about power consumption? I have 2x 1200w psu's in my 2. computer from the 2x Evga 780ti classified (1500mhz) days + 3900k @ 5200mhz 😀 Drew 1700w from the wall 😀

#5769807

Aura89

2020-03-15 20:12

nizzen:

When did someone actually care about power consumption? I have 2x 1200w psu's in my 2. computer from the 2x Evga 780ti classified (1500mhz) days + 3900k @ 5200mhz 😀 Drew 1700w from the wall 😀

Lots of people care about power consumption. Just because you don't, and you don't have to, doesn't mean anyone doesn't. You're not everyone, don't act like it.

#5769849

Venix

2020-03-15 23:40

nizzen:

When did someone actually care about power consumption? I have 2x 1200w psu's in my 2. computer from the 2x Evga 780ti classified (1500mhz) days + 3900k @ 5200mhz 😀 Drew 1700w from the wall 😀

the question is when they did not ? let's see the most famous cases of power consumption leading to bad perception , pentium4 (Prescott ) ,fermi 480 , Rx290x , Fx 9590 ,Vega 64 should i keep going ? power performance per watt is one of the dominant metrics in the industry and a major criteria if your product is good . Except if i missed all the years of people calling products space heaters

#5769917

Astyanax

2020-03-16 08:08

Hyperus:

They do not need to be exposed, we are talking about a driver emulation here, NV drivers can do with the hardware whatever they want.

We're talking about you not knowing what you're talking about, tensor cores are fixed function matrix multiplication, they cannot do anything else.

#5770053

TalentX

2020-03-16 15:37

How ironic is this "KittyCorgi" ? https://meme.xyz/uploads/posts/t/l-37084-things-that-dont-matter-opionions-from-people-with-anime-profile-pictures.jpg

#5771086

pharma

2020-03-19 15:42

Super XP:

This guy is not a guy, it's a girl. Apparently she's been right several times before or quite close to what actually comes out. But stuff change from time to time.

She is guessing and is know for modifying her postings. Just another looking for twitter fame.

#5771087

pharma

2020-03-19 15:43

Super XP:

Where would Nvidia get 7nm from?

From the horse's mouth.

#5771088

Denial

2020-03-19 15:45

Super XP:

Where would Nvidia get 7nm from? It wouldn't be from TMSC as they are already fully booked especially with current ZEN2 and upcoming RDNA2 plus next generation gaming consoles.

This logic makes no sense to me. Someone said this earlier in the thread, maybe it was another thread, but who cares if TSMC is fully booked now? These contracts are worked out a year in advance, in some cases multiple years. AMD has a 4 year contract with global foundries - Nvidia/AMD/Qualcomm/etc all make similar contracts. How do you know Nvidia didn't book this two years ago and it part of the reason why TSMC is "fully booked"? The silicon itself is directly linked to the node - this isn't like 6-8 months before they launch they go "hey TSMC do you have any wafers left for us?!?!?" Even if you ignore the flawed logic there.. why is Nvidia's CEO telling investors/media people that they are primarily fabbing on TSMC? And yeah, I'm with Pharma - this person isn't like Evan Blass of phone leaks, where they have been 85%+ consistent for years. This person to me is nobody. They don't have a history. They've already changed a few of their posts and deleted others. Who knows - maybe the chips will come out and they'll be 100% correct but as of right now their posts are just speculation.