Tuesday, October 10th 2017

AMD "Navi" GPU by Q3-2018: Report

AMD is reportedly accelerating launch of its first GPU architecture built on the 7 nanometer process, codenamed "Navi." Graphics cards based on the first implementation of "Navi" could launch as early as by Q3-2018 (between July and September). Besides IPC increments with its core number-crunching machinery, "Navi" will introduce a slew of memory and GPU virtualization technologies.

AMD will take its multi-chip module (MCM) approach of building high-performance GPUs a step further, by placing multiple GPU dies with their HBM stacks on a single package. The company could leverage its InfinityFabric as a high-bandwidth interconnect between the GPU dies (dubbed "GPU module"), with an I/O controller die interfacing the MCM with the host machine. With multi-GPU on the decline for games, it remains to be seen how those multiple GPU modules are visible to the operating system. In the run up to "Navi," AMD could give its current "Vega" architecture a refresh on a refined 14 nm+ process, to increase clock speeds.
Source: TweakTown
Add your own comment

74 Comments on AMD "Navi" GPU by Q3-2018: Report

#26
Evildead666
BoosnieDo not expect this to be available to the general public.
If the cores will be accessible as indipendent CUs, those cards will be immediately loaded on a 747 parked in the lot outside the fab.
Unfortunately, I beleive that all the next AMD GPU's are going to be good miners.....
GorbazTheDragonNo it's not, different CPU architectures often have different CU counts which is what leads to different IPC...
Comparing Nvidia Shaders Units to AMD Shader Units is not possible.
They are completely different things.
Posted on Reply
#27
Mirkoskji
evernessinceComparing CUs between architectures is like comparing GHz between architectures, worthless.
yes, but comparing die size has a certain importance
Posted on Reply
#29
Frick
Fishfaced Nincompoop
XzibitI'd teach him not to give in to peer pressure and not cave in to children making fun of him.
Very tangential, but this is easy to say and very hard to do. If you are older than 20 you have literally no idea what it's like for kids these days.
Posted on Reply
#30
Xzibit
FrickVery tangential, but this is easy to say and very hard to do. If you are older than 20 you have literally no idea what it's like for kids these days.
More reason to be active when they come to you with a situation such as this.
Sure doesn't help things when your nephew comes to you after possible bullying or shaming and as the uncle you basically side with them. Speaks poorly and gives a stand off approach to a situation. You will have less of an idea what goes on with them.
Posted on Reply
#31
Tsukiyomi91
Volta is not even out yet & AMD is already afraid of it, so they are releasing Navi. Hope this is not another flop like Polaris & Vega. What's gonna be funny is despite using 7nm lithography, the whole die size will still be bigger than Pascal or the upcoming Volta. Expect early shortages like Vega since it's also going to be using HBM v2. Highly doubt they'll be opting GDDR6 to keep the prices low.
Posted on Reply
#32
Assimilator
Intel's 10nm, which is comparable to their competitors' 7nm, is supposed to be released by the end of THIS year. Considering GloFo's troubles with 14nm, and that they've just announced "12nm" which is merely an improved 14nm, I honestly doubt we'll see 7nm from them next year in any way, shape or form (unless, of course, they decide to become even more creative with node designations and "7nm" is just 14nm++).
Posted on Reply
#33
ShurikN
Tsukiyomi91What's gonna be funny is despite using 7nm lithography, the whole die size will still be bigger than Pascal or the upcoming Volta. Expect early shortages like Vega since it's also going to be using HBM v2. Highly doubt they'll be opting GDDR6 to keep the prices low.
Always nice to have a psychic here on TPU....
Posted on Reply
#34
ZoneDymo
Tsukiyomi91Volta is not even out yet & AMD is already afraid of it, so they are releasing Navi. Hope this is not another flop like Polaris & Vega. What's gonna be funny is despite using 7nm lithography, the whole die size will still be bigger than Pascal or the upcoming Volta. Expect early shortages like Vega since it's also going to be using HBM v2. Highly doubt they'll be opting GDDR6 to keep the prices low.
That makes so little sense.... might as well claim Nvidia is afraid of Vega which is why they are releasing Volta in the near future...

And flop flop flop, pretty sure Polaris sold like hotcakes.
Vega is a bit of a flop due to being shipped with higher voltage demands then it actually needs and low yield making it more of a paper launch, but as a product its quite fine.
Posted on Reply
#36
jigar2speed
HoodI know, it puzzles me why all Ryzen CPUs are being heavily discounted at retail - their latest and greatest ever - with no successor in sight? Doesn't make a lot of sense. Please enlighten me as to why Ryzen is losing popularity so quickly. I offered to build a Ryzen system for my nephew for his birthday this November. A week later he tells me he wants an Intel system, apparently because his friends laughed at him when he told them he was getting a Ryzen rig. Has AMD lost their mojo this quickly? What gives? What do teenage gamers know about it that we don't?
You know how funny your comment sounds and much funnier your interpretation. You are not good with propaganda spreading, you are just bad at it and just give up with your AMD hate already.
Posted on Reply
#37
Lycanwolfen
Lol I love AMD. They the #1 machine that comes into my shop with problems. I love them because they make me money fixing them. HP, Gateway, Acer, Dell, Asus, Etc. I used to own AMD. I had 5 of them long time ago tweaking having funs fixing issues all the time. Then I noticed when I switch all to Intel machines My problems went away. Also my power bills dropped by 100 dollars a month. I have all Intel now some older than 15 years so Ya I perfer stable and allways works than having issues and allways having problems.
Posted on Reply
#38
R0H1T
AssimilatorIntel's 10nm, which is comparable to their competitors' 7nm, is supposed to be released by the end of THIS year. Considering GloFo's troubles with 14nm, and that they've just announced "12nm" which is merely an improved 14nm, I honestly doubt we'll see 7nm from them next year in any way, shape or form (unless, of course, they decide to become even more creative with node designations and "7nm" is just 14nm++).
You mean 2018, 10nm isn't going into (mass?) production this year & I mean full launch not paper announcements.
Posted on Reply
#39
cucker tarlson
I said:
cucker tarlson(...) until they do something to come up with a power efficient architecture that can achieve higher clocks without pulling ridiculous amounts of power.
You said:
evernessinceActually Vega is quite power efficient. If you look at GamersNexus, they did a video where they compared Vega to the Fury X and Clock for Clock Vega is significantly more power efficient.
this is exactly what I said was wrong with AMD's cards. They need clockspeed to compete, and they're failing to achieve them without going way overboard with power consumption. You can't expect dx12 features to make up for such a huge disparity between pascal and vega clockspeeds, that's why you've got V64 competing (and usually losing to) GTX1080. If that continues into mcm nvidia vs mcm amd, they'll still fall behind geforce cards.


The way I predict things will happen is nvidia will wait out the 1st gen of AMD mcm GPUs with a big Volta chip (I'm talking 500-600mm2 2080Ti just like GM200), then roll out their own mcm GPUs in 2020. Games will need to be optimized for such innovations, I think the green team will opt for the strategy to give you more consistent performance with a single GPU and see how MCM will work out for AMD.
Posted on Reply
#40
ShurikN
LycanwolfenLol I love AMD. They the #1 machine that comes into my shop with problems. I love them because they make me money fixing them. HP, Gateway, Acer, Dell, Asus, Etc. I used to own AMD. I had 5 of them long time ago tweaking having funs fixing issues all the time. Then I noticed when I switch all to Intel machines My problems went away. Also my power bills dropped by 100 dollars a month. I have all Intel now some older than 15 years so Ya I perfer stable and allways works than having issues and allways having problems.
Damn, I got 10% dumber just by reading this...
Posted on Reply
#41
cucker tarlson
LycanwolfenLol I love AMD. They the #1 machine that comes into my shop with problems. I love them because they make me money fixing them. HP, Gateway, Acer, Dell, Asus, Etc. I used to own AMD. I had 5 of them long time ago tweaking having funs fixing issues all the time. Then I noticed when I switch all to Intel machines My problems went away. Also my power bills dropped by 100 dollars a month. I have all Intel now some older than 15 years so Ya I perfer stable and allways works than having issues and allways having problems.
Posted on Reply
#42
Lycanwolfen
Oh I'm so sorry I cannot type perfect for you. I did it in a hurry because I'm busy with clients. I personally do not have the time to spell check everything I type in a forum.
Posted on Reply
#43
Basard
Pap1erI somehow believe that by utilising infinity fabric on GPUs ( I mean larger heat dissipation area) we may see smaller GPU coolers or at least they won't be as noisy as it use to be nowadays. What do you think?
Not that you were asking me, but... I think they will just get bigger like always.
Posted on Reply
#44
cucker tarlson
LycanwolfenOh I'm so sorry I cannot type perfect for you. I did it in a hurry because I'm busy with clients. I personally do not have the time to spell check everything I type in a forum.
let me guess, amd rigs again ?
Pap1erI somehow believe that by utilising infinity fabric on GPUs ( I mean larger heat dissipation area) we may see smaller GPU coolers or at least they won't be as noisy as it use to be nowadays. What do you think?
I think that will still depend on their power draw.
And the diagram shows 2x as much dram stacks to cool.
Posted on Reply
#45
TheinsanegamerN
cucker tarlsonNo way this is coming out by Q3 2018. Maybe a working ES by Q3 2018. Maybe. AMD pulling the wool over people's eyes again, that's been their strategy for quite some time. But I guess they need something new to get their fanbase hyped again after Vega failed hard. Infinityfabric,HBM2 and all their fancy features will always fall short until they do something to come up with a power efficient architecture that can achieve higher clocks without pulling ridiculous amounts of power.

btw this is a nvidia slide ;)
Nah, see the GPU will totally be out by Q3 2018.

And by out, we mean that ES chips will be out. A Fanboy Edition model with 0 driver support will launch in Q4 2018, and the normal people model in Q2 2019. See? Techincally available in Q3 2018 :D

As a bonus, AMD will finally be on pascal's level!:p
Posted on Reply
#46
phanbuey
Pap1erI somehow believe that by utilising infinity fabric on GPUs ( I mean larger heat dissipation area) we may see smaller GPU coolers or at least they won't be as noisy as it use to be nowadays. What do you think?
I think the chips will be easier to cool due to the increased surface area, but if anything coolers will have as hard / harder time keeping them cool, since they will still need to dissipate the heat (if not even more heat, now that the overall chip will have increased thermal headroom).
Posted on Reply
#47
OneMoar
There is Always Moar
hello terrible performance scaling
has amd not learned that nobody is going to optimize for there stupid module approach
Posted on Reply
#48
gamerman
again...
so usual amd politics again...

nice,great graphics lines and blaa blaa then junk coming out aka R9 290 , fury x and of coz vega

and... come on! Q3/2018! august 2018!!?? .. then nvidia have out for sure volta #2

and we all know amd's timing release, paper and real one... not trust here.

amd looks dead.


ohyaa... when nvidia volta release, amd have to use liquid nitrogen for cooling and even that not help anymore..
so amd MUST get something NEW at least now bcoz last 2 amd 's gpu's are same junk.
all need watercool and effiency is terrible!
Posted on Reply
#49
Frick
Fishfaced Nincompoop
Tsukiyomi91Volta is not even out yet & AMD is already afraid of it, so they are releasing Navi. Hope this is not another flop like Polaris & Vega. What's gonna be funny is despite using 7nm lithography, the whole die size will still be bigger than Pascal or the upcoming Volta. Expect early shortages like Vega since it's also going to be using HBM v2. Highly doubt they'll be opting GDDR6 to keep the prices low.
Even disregarding the mining thing, I wouldn't call Polaris a flop.
Posted on Reply
#50
f22a4bandit
HoodI know, it puzzles me why all Ryzen CPUs are being heavily discounted at retail - their latest and greatest ever - with no successor in sight? Doesn't make a lot of sense. Please enlighten me as to why Ryzen is losing popularity so quickly. I offered to build a Ryzen system for my nephew for his birthday this November. A week later he tells me he wants an Intel system, apparently because his friends laughed at him when he told them he was getting a Ryzen rig. Has AMD lost their mojo this quickly? What gives? What do teenage gamers know about it that we don't?
Zen+ is supposedly right around the corner for early 2018. That might explain some of the discounts we've seen lately. Also likely to better position themselves against Coffee Lake.

It sounds like your nephew's friends simply go for Intel because they've been touted as the best for years without any competition from AMD. I've spoken with friends who only build Intel machines because they're the best for gaming which is pretty much the same reasoning.
Posted on Reply
Add your own comment
Dec 27th, 2024 03:43 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts