Thursday, September 27th 2018

AMD "Navi" GPU Architecture Successor Codenamed "Arcturus"?

Arcturus is the fourth brightest star in the night sky, and could be the a new GPU architecture by AMD succeeding "Navi," according to a Phoronix report. The codename of Navi-successor has long eluded AMD's roadmap slides. The name "Arcturis" surfaced on Phoronix community forums, from a post by an AMD Linux liaison who is a member there. The codename is also supported by the fact that AMD is naming its GPU architectures after the brightest stars in the sky (albeit in a descending order of their brightness). Polaris is the brightest, followed by Vega, Navi, and Arcturus.

AMD last referenced the Navi-successor on a roadmap slide during its 2017 Financial Analyst Day presentation by Mark Papermaster. That slide mentioned "Vega" to be built on two silicon fabrication processes, 14 nm and "14 nm+." We know now that AMD intends to build a better-endowed "Vega" chip on 7 nm, which could be the world's first 7 nm GPU. "Navi" is slated to be built on 7 nm as the process becomes more prevalent in the industry. The same slide mentions Navi-successor as being built on "7 nm+," which going by convention, could refer to an even more advanced process than 7 nm. Unfortunately, even in 2017, when the industry was a touch more optimistic about 7 nm, AMD expected the Navi-successor to only come out by 2020. We're not holding our breath.
Source: Phoronix
Add your own comment

43 Comments on AMD "Navi" GPU Architecture Successor Codenamed "Arcturus"?

#26
efikkan
R0H1T
Yes & that was the retailers, in some cases the wholesaler, being greedy. In this case however if RTX models are still selling above their "suggested MRP" then it boils down to 2 more plausible scenarios ~ they don't have enough margins on RTX (% wise or perhaps due to lower volume) &/or they're having problems clearing the inventories of the erstwhile Pascal lineup. The latter was mentioned by more than one reputable source IIRC, so no this pricing fiasco is totally on Nvidia this time around & miners can't take the blame for this!
If retailers choose to overprice Turing while selling out Pascal, the retailers are to blame.

theoneandonlymrk
Yet you blamed Amd for +£100 msrp a few posts ago, Nvidias set their prices too high to start with.
No, that was AMD falsely claiming the MSRP of Vega 64 was $499, when it was actually $599 with $100 worth of game coupons included.
They did change this within a few weeks after massive protests, but people have of course long forgotten.
Posted on Reply
#27
SIGSEGV
efikkan
If retailers choose to overprice Turing while selling out Pascal, the retailers are to blame.


No, that was AMD falsely claiming the MSRP of Vega 64 was $499, when it was actually $599 with $100 worth of game coupons included.
They did change this within a few weeks after massive protests, but people have of course long forgotten.
lol. sorry i couldn't resist to laugh
Posted on Reply
#28
ppn
Yeah 7nm is the real deal. 50M transistors /mm2. Current 14 nm with 25 Mtr/mm2 is simply unacceptable anymore.
Posted on Reply
#29
TheinsanegamerN
ppn
Yeah 7nm is the real deal. 50M transistors /mm2. Current 14 nm with 25 Mtr/mm2 is simply unacceptable anymore.
I dont know, my vega 64 still runs pretty well, and the 2080ti is a performance king, just way too expensive.

7nm is gonna cost a LOT too when it comes out, so itll take a few years for prices to fall, assuming AMD ever competes again.
Posted on Reply
#30
efikkan
TheinsanegamerN
7nm is gonna cost a LOT too when it comes out, so itll take a few years for prices to fall, assuming AMD ever competes again.
Yes, the production volume will be fairly low for 7 nm in the beginning, until they manage to do major improvements to the process, either through the use of EUV or other means.
Posted on Reply
#31
cucker tarlson
The situation in Poland is kinda funny now, it almost seems like retailers overpriced 1080Ti's after 2080 got bad press.
Posted on Reply
#32
londiste
cucker tarlson
The situation in Poland is kinda funny now, it almost seems like retailers overpriced 1080Ti's after 2080 got bad press.
1080Ti's were never really cheap. There is a surge of miner cards to the used market but today almost all of these appear to be blower models.
Posted on Reply
#33
DeathtoGnomes
NC37
It's all fun and games till some crazy engineer names one betelgeuse...
Space:1999 ?
Posted on Reply
#34
Vya Domus
HTC
Unless AMD comes up with a way to have Zen like chips to form a single GPU, i don't see AMD catching up to nVidia in performance any time soon.
They can catch up and overtake Nvidia, there's no question about it. Is it worth it ? Probably not. Not even Zen was the best AMD could have made but rather it was a very cost effective design that was simply good enough for all intents and purposes. AMD got burned too many times investing tons of cash in high end consumers GPUs.
Posted on Reply
#35
efikkan
Vya Domus
They can catch up and overtake Nvidia, there's no question about it. Is it worth it ? Probably not.
I just love this level of ignorance, simply priceless. Once again, the story about AMD secretly winning by holding back.

On a serious note; their GPU department is burning money right now, if they could make a better GPU they would.
Posted on Reply
#36
R0H1T
efikkan
I just love this level of ignorance, simply priceless. Once again, the story about AMD secretly winning by holding back.

On a serious note; their GPU department is burning money right now, if they could make a better GPU they would.
Yeah they're not going to overtake Nvidia anytime soon, but they're not so far behind that it'll take eons for AMD to overtake Nvidia's GPU. Unlike x86 or ARM CPU's a breakthrough in GPU tech is much more plausible & more likely. GPU uarch ~ as complex as they are ~ aren't comparable to what AMD & Intel (or ARM) do in their free time.

Burning money, so they're making a loss, this based on what exactly?
Posted on Reply
#37
Th3pwn3r
TheinsanegamerN
I dont know, my vega 64 still runs pretty well, and the 2080ti is a performance king, just way too expensive.

7nm is gonna cost a LOT too when it comes out, so itll take a few years for prices to fall, assuming AMD ever competes again.
AMD can easily compete with a card a little better than a 1080ti and better pricing than the RTX series. Freesync is AMD's weapon X. The only reason I would go with Nvidia is only for 4K gaming, aside from that I think a 1080ti is the best performer until you have to pay for Gsync. That's where I tell people to go with AMD for a whole lot less money.
Posted on Reply
#38
Divide Overflow
hat
... turn 360 degrees and walk away.


If you turn 360 degrees, you won't be walking away.
Posted on Reply
#39
TheinsanegamerN
Divide Overflow
If you turn 360 degrees, you won't be walking away.
......michael jackson moonwalks backwards. Cmon man. This is easy stuff.

Vya Domus
They can catch up and overtake Nvidia, there's no question about it. Is it worth it ? Probably not. Not even Zen was the best AMD could have made but rather it was a very cost effective design that was simply good enough for all intents and purposes. AMD got burned too many times investing tons of cash in high end consumers GPUs.
I'm fine with them not going uber high end. Polaris was good, we were just begging for a 3072 core part with a 384 bit bus for SOMETHING faster then a 290x.

Vega 56 fit that mold well, as the middle-high end gray area sells well and has fat margins, it was just a year too late.
Posted on Reply
#40
SIGSEGV
Vya Domus
They can catch up and overtake Nvidia, there's no question about it. Is it worth it ? Probably not. Not even Zen was the best AMD could have made but rather it was a very cost effective design that was simply good enough for all intents and purposes. AMD got burned too many times investing tons of cash in high end consumers GPUs.
For me, zen is simply brilliant design and clear winner in terms of price/perf. But, when we talk about GPU, it's a different story. It's not easy to compete with nvidia. They have a lot of resources in both R&D and Marketing. Nvidia can easily tackle AMD in every way. When AMD has powerful GPU to offer then they will somehow cripple its performance by taking control of game development in some games (i believe they can easily control most of AAA games development). It's also happened in the scientific world with their CUDA proprietary. They have lots of resources to back up and support their CUDA environments.
Posted on Reply
#41
TheinsanegamerN
SIGSEGV
For me, zen is simply brilliant design and clear winner in terms of price/perf. But, when we talk about GPU, it's a different story. It's not easy to compete with nvidia. They have a lot of resources in both R&D and Marketing. Nvidia can easily tackle AMD in every way. When AMD has powerful GPU to offer then they will somehow cripple its performance by taking control of game development in some games (i believe they can easily control most of AAA games development). It's also happened in the scientific world with their CUDA proprietary. They have lots of resources to back up and support their CUDA environments.
You say that, but PC games blatantly favoring nvidia are not as common as they used to be. The consoles running AMD hardware has had an effect.
Posted on Reply
#42
theoneandonlymrk
SIGSEGV
For me, zen is simply brilliant design and clear winner in terms of price/perf. But, when we talk about GPU, it's a different story. It's not easy to compete with nvidia. They have a lot of resources in both R&D and Marketing. Nvidia can easily tackle AMD in every way. When AMD has powerful GPU to offer then they will somehow cripple its performance by taking control of game development in some games (i believe they can easily control most of AAA games development). It's also happened in the scientific world with their CUDA proprietary. They have lots of resources to back up and support their CUDA environments.
It is and arcturus could be, they are said to be and admitted they wanted to bring the zen ideology to graphics, personally I think they will at some point around 7nm but I am thinking lately ,if Ms knows that rays are coming Amd would be told and they know Ai is a thing so why not go all in on chiplets.
Modular with two gpu an Nnapi chip and something for rays(chip) on one package.

Who knows , it's a good pr spoiler, traditional style now RTX is out.
Posted on Reply
#43
ypsylon
To be honest instead meaningless code-names I would actually prefer some meaningful availability of Navi, but oh wait at best it's 2 years away... And this Arcutrus (Charon Relay, yay for shameless Mass Effect plug ;) ) project that's like 5-6 years away? What's the point of even making this a news in any capacity.

In 2 years time we can have WWIII on our hands and people getting all excited about code-name of a VGA line that may not even materialize in 5-6 years time frame? :wtf: I missed some memo somewhere. In a word of immortal Anthony McAuliffe: Nuts!
Posted on Reply
Add your own comment