AMD Radeon Instinct MI100 "Arcturus" Hits the Radar, We Have its BIOS

[TheGuruStud]

Why was the Radeon 7 a failure in your eyes?

B/c it wasn't 3 hundred dollars and they couldn't afford it...and ignoring that it wasn't actually a gaming card.

 

[Ahhzz]

Whole lot of trolling going on, guys. Reply bans in place, and points incoming for you "special" ones. Keep it on topic, keep it civil.

 

[cucker tarlson]

TDP limit is set to a surprisingly low 200 W, especially considering this is a 128 CU / 8,192 shader count design

hpc ampere is doa

edit:

This suggests that the GPU will tick at up to 1334 MHz

oh yeah,that's what it is.

v100s was 1600mhz on 12nm at massive 815nm.ampere on 7nm euv ? probably dies not need remotely close to 8000sp to beat this given the clocks and asic hardware

 

[Kohl Baas]

This is very interesting. I'm quite curious if this is an HPC/AI/ML only architecture, then what is RDNA2? Is it Arcturus with it's 3D units using the extra HPC power to emulate/calculate some kind of hybrid software/hardware RT running specifically on "unused in 3D" parts of the RDNA2 architecture?

I know it's comming and I know that consoles probably will get RDNA2 for RT, but AMD must have at least an enthussiast card and if we have luck a performance/high performace card at the launch of the next gens to leverage some of the new graphics of the next-gen multiplatform titles. Perhaps Cyberpunk 2077?

 

[cucker tarlson]

what is RDNA2?

gaming specific.
that's actually very good they're developing specific architectures for hpc and gaming.
vega 14nm/7nm was a jack of all trades,master of none.people tend to want a gaming card or a hpc card,not hpc and gaming.

AMD must have at least an enthussiast card and if we have luck a performance/high performace card at the launch of the

high performance is a vague term.
I wouldn't call my 2070 super a high performance card,it's okay,kinda upper mid range and that's all.It's fine for 1440p.For now.
meanwhile 5700xt is high performance for amd despite it's 2070 non super.

 

[Cheeseball]

This is very interesting. I'm quite curious if this is an HPC/AI/ML only architecture, then what is RDNA2? Is it Arcturus with it's 3D units using the extra HPC power to emulate/calculate some kind of hybrid software/hardware RT running specifically on "unused in 3D" parts of the RDNA2 architecture?

You almost got the idea there. It seems that RDNA and RDNA2 are aiming for being the consumer-level/basic workstation-level (RX 5700 = Radeon Pro W5700) architecture, while the Instincts (which are GCN-based, Polaris/Vega) are for Deep Learning and HPC.

 

[gamefoo21]

So much hate from people that don't understand what this is for.

There's also likely a version with actual display out. V20 powers the MI50 and MI60 that have their display functionality cut. I also really hope that AMD decides to turn on the degenerated rendering hardware finally.

What'll be interesting is if AMD decided to add in AI cores to boost the same specific work loads that Tensor cores do.

You almost got the idea there. It seems that RDNA and RDNA2 are aiming for being the consumer-level/basic workstation-level (RX 5700 = Radeon Pro W5700) architecture, while the Instincts (which are GCN-based, Polaris/Vega) are for Deep Learning and HPC.

The Radeon VII is the people's champion of video and image processing right now.

 

[Cheeseball]

The Radeon VII is the people's champion of video and image processing right now.

This would be true for prosumers (professionals who need a consumer product for working at home), but the Radeon VII has been discontinued. It is basically a MI50 and MI60 without ECC, optimization and certification for workstation use (if you're using the Pro driver).

Also, the Radeon VII was NOT open to corporate pricing (e.g. bulk orders, support, etc.) and you would be recommended to go with the MI50 or MI60 at the time. It really was a niche halo card aimed at the consumer-market while AMD was getting Navi ready.

 

[gamefoo21]

This would be true for prosumers (professionals who need a consumer product for working at home), but the Radeon VII has been discontinued. It is basically a MI50 and MI60 without ECC, optimization and certification for workstation use (if you're using the Pro driver).

Also, the Radeon VII was NOT open to corporate pricing (e.g. bulk orders, support, etc.) and you would be recommended to go with the MI50 or MI60 at the time. It really was a niche halo card aimed at the consumer-market while AMD was getting Navi ready.

People's champion aka what anyone who walks in off the street can get.

I can buy a lot of Radeon VIIs for the price of a single V100.

Guess which will do Premiere faster? How many 16x slots do I get to fill 6, 7, 8?

I'll watercool the Radeons, and absolutely destroy the V100 and likely work on multiple projects at the same time.

Not all of us are named Steve and get NV corporate gear at solid prices or basically free. *snicker*

 

[Cheeseball]

I can buy a lot of Radeon VIIs for the price of a single V100.

Guess which will do Premiere faster? How many 16x slots do I get to fill 6, 7, 8?

I'll watercool the Radeons, and absolutely destroy the V100 and likely work on multiple projects at the same time.

Not all of us are named Steve and get NV corporate gear at solid prices or basically free. *snicker*

?????

The V100 is nowhere near a consumer product. I don't know why you keep comparing the Radeon VII to the Tesla V100. It's practically $680 versus ~$13,000 (NVLINK). Both of these aren't even workstation cards, where one is a normal consumer card and the other is a data center GPU not meant to be used in a workstation.

And where are you seeing that Premiere is faster on a Radeon VII? Unless you're running a NVIDIA card in OpenCL on Windows (which is counter-productive and not recommended), the Radeon VII is not the champion you think it is:

It is damn good in DaVinci Resolve and when using certain plugins.

Also, if you're testing the Radeon VII on a Mac, you should be using Metal in Premiere Pro as OpenCL has been deprecated in the new OSX. This is where I think you would see a big performance increase (matching the 2080 Ti).

 

[Totally]

GCN 3 and 4 were GFX8 so I don't see why it can't be GCN6

No reason, for it to be so either. Just going by what you just stated point out that there is no correlation between the codename and uarch.

 

[T4C Fantasy]

No reason, for it to be so either. Just going by what you just stated point out that there is no correlation between the codename and uarch.

GCN3 is Volcanic Islands and Pirate Islands, GCN4 is Polaris, GCN5 is Vega, GCN6 makes sense to be Arcturus. It has elements of Vega in it but its still different.

GCN6 is just speculation coined by me.... so its just speculation (unofficial).. but i know it makes sense either way.

 

[DeathtoGnomes]

There should be anything coming that is still on GCN.

GCN3 is Volcanic Islands and Pirate Islands, GCN4 is Polaris, GCN5 is Vega, GCN6 makes sense to be Arcturus. It has elements of Vega in it but its still different.

GCN6 is just speculation coined by me.... so its just speculation (unofficial).. but i know it makes sense either way.

 

[T4C Fantasy]

There should be anything coming that is still on GCN.

Definitely shouldn't but it's gfx9 which suggests it's still gcn.

 

[T1beriu]

GCN6 makes sense to be Arcturus

Arcturus is not an architecture name, it's just a silicon name. We were told this by multiple times by the AMD engineer that spilled the beans on "Arcturus" codename.

 

[FordGT90Concept]

Definitely shouldn't but it's gfx9 which suggests it's still gcn.

It makes sense. Radeon Instinct is a compute product and GCN is undeniably a compute architecture. RDNA sacrifices raw compute power for gaming performance per watt.

 

[T4C Fantasy]

Arcturus is not an architecture name, it's just a silicon name. We were told this by multiple times by the AMD engineer that spilled the beans on "Arcturus" codename.

Yes it's a codename like Vega, architecture to Vega is gcn5.

 

[T1beriu]

Yes it's a codename like Vega, architecture to Vega is gcn5.

That's exactly what it's not. The AMD engineer and I was saying that Arcturus is like Vega 10, Polaris 20, Lexa, Fiji, Ellesmere, the codename of a silicon die. We won't see Arcturus 10, Arcturus 11, Arcturus 12. Please see bridgman's comment again.

Architecture	Architecture Codename	Silicon codename
GCN 5 (GFX9XX)	Vega	Vega 10, Vega 12
GCN 5 (GFX9XX)	Vega	Vega 20
RDNA (GFX101X)	Navi	Navi 10, Navi 12, Navi 14
GCN 5 (GFX9XX)	Vega	ARCTURUS
RDNA 2 (GFX103X)	Navi	Navi 21, Navi 22, Navi 23

https://twitter.com/x/status/1205301982631624704

 

[T4C Fantasy]

That's exactly what it's not. The AMD engineer and I was saying that Arcturus is like Vega 10, Polaris 20, Lexa, Fiji, Ellesmere, the codename of a silicon die. We won't see Arcturus 10, Arcturus 11, Arcturus 12. Please see bridgman's comment again.

Architecture	Architecture Codename	Silicon codename
GCN 5 (GFX9XX)	Vega	Vega 10, Vega 12
GCN 5 (GFX9XX)	Vega	Vega 20
RDNA (GFX101X)	Navi	Navi 10, Navi 12, Navi 14
GCN 5 (GFX9XX)	Vega	ARCTURUS
RDNA 2 (GFX103X)	Navi	Navi 21, Navi 22, Navi 23

https://twitter.com/x/status/1205301982631624704

It's probably going to have a different codename for it's architecture codename than Vega while having the same instruction set with added stuff, CLRX 1.4.2 GCN6, Fiji was GFX803 CLRX 1.2, Ellesmere GFX803 CLRX 1.2.

AMD is not exactly organized...

The codename is probably MI100 tbh
The prototype chip says MI100 XL which is usually in the place of where it would say Polaris 10 for example.

Edit: Also Raven is GFX9 and not Vega architecture
It's like Polaris with early Navi encoder and decoder engine

VegaM (Polaris 22) is GFX804 CLRX 1.2
Polaris with HBM

AMD is confusing trust me I work on a lot of their stuff (for gpudb) So GCN6 is coined.. pure speculation but I have reason to think it's not GCN5.
also I wont be surprised if it is GCN5 either, this is just my speculation from the chaos that is AMD... that is all.

 

[Manoa]

accelerate what ? if he don't have cuda what the point ?

 

[eidairaman1]

accelerate what ? if he don't have cuda what the point ?

Cuda is proprietary, it's not needed.

 

[notb]

accelerate what ? if he don't have cuda what the point ?

In large HPC projects, which AMD cards will be used for mostly (probably), CUDA is not that beneficial. Software is low-level, highly optimized, built from scratch and usually developed by full-time programmers.

Where CUDA shines the most are smaller projects, ad-hocs, daily computation problems - especially when code is written not by full-time developers but scientists, engineers or analysts.

So as long as these clusters are used for weather models and so on - nothing bad happens.
But if this was a cluster for end users (i.e. private cloud), lack of CUDA would surely hurt.

Cuda is proprietary, it's not needed.

CUDA is easy to use and very popular. This is not a matter of what's needed, but what's the most beneficial.

 

[Totally]

GCN3 is Volcanic Islands and Pirate Islands, GCN4 is Polaris, GCN5 is Vega, GCN6 makes sense to be Arcturus. It has elements of Vega in it but its still different.

GCN6 is just speculation coined by me.... so its just speculation (unofficial).. but i know it makes sense either way.

I suppose but you say that now, but watch the speculation starting here somehow makes it's way over semi accurate then back to over to TPU as a news article.

 

[medi01]

For gamers, AMD is planning a new line of Navi 20-series chips leveraging 7 nm EUV

Is it known for a fact, that it would be 7nm EUV?

 

[Adam Krazispeed]

HBM2 ??? NOooooo, price will be too high once again

this is for server or AI / compute / Machine Learning????

this is "ARCTURUS" SERVER/COMPUTE ACCELERATOR..
****THIS DOESN'T EVEN HAVE 3D CAPABILITIES****


BIG NAVI IS COMING!!!!!! but WILL NOT BE AS LARGE AS THIS "ARCTURUS" cGPU... aka(COMPUTE GPU) or ML/AI/COMPUTE

Arcturus is not an architecture name, it's just a silicon name. We were told this by multiple times by the AMD engineer that spilled the beans on "Arcturus" codename.

ACTUALLY....TBASA...lol

Arcturus is a red giant star in the Northern Hemisphere of Earth's sky and the brightest star in the constellation Boötes (the herdsman). Arcturus is also among the brightest stars that can be seen from Earth. Astronomers say Arcturus will end up as a white dwarf at the end of its life

this may be the last of GCN in other words.... At the end of its life, is AMD throwing a Riddle at us??? using this as the name of the silicon>>>

idk