Vega 10 could launch in 1H 2017, possibly March 2017 (Fudzilla)

Toggle sidebar Toggle sidebar
S

Sweepr

Diamond Member
May 12, 2006
5,148
1,142
131
''March 2017 launch possible''

f443b242790e7bb3a77d56652c680eb3_L.jpg


Vega 10 is the successor to Polaris and if all goes well you will see the world’s first HBM 2.0 based GPU from the house of AMD – RTG in the first part of 1H 2016, possibly even in Q1 2017.

The timeframe doesn’t come as a great surprise as significant milestones have been reached. The second part of the puzzle is that Nvidia already said that its HBM 2.0 compute part, the GP100, will ship to a wider circle of customers in early 2017.

Fudzilla repeatedly claimed that there won’t be any HBM 2.0 capable GPUs in 2016 as our sources simply told us that that the memory won’t be ready for mass production. They seem to be right about this one.
Click to expand...

www.fudzilla.com/news/graphics/41034-vega-10-amd-hbm-2-can-launch-in-1h-2017
 
E

Erenhardt

Diamond Member
Dec 1, 2012
3,251
105
101
Quite a large window for a rumor.
1H2016 is gone now. Plenty of Launch opportunities until Q1 2017
 
psolord

psolord

Golden Member
Sep 16, 2009
1,928
1,194
136
Polaris 10 is the big chip right? There will be a Polaris 11 for the RX460 and the like?

So Vega 10 will also be the big chip right? For the Fury. Vega 11 will be the one for RX490?

What chip sizes are we reasonably expecting for both?

And what about reasonably expected memory configurations? 384bit GDDR5, 256bit GDDR5X, HBM2 for both, what?
 
3

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
psolord said:
Polaris 10 is the big chip right? There will be a Polaris 11 for the RX460 and the like?

So Vega 10 will also be the big chip right? For the Fury. Vega 11 will be the one for RX490?

What chip sizes are we reasonably expecting for both?

And what about reasonably expected memory configurations? 384bit GDDR5, 256bit GDDR5X, HBM2 for both, what?
Click to expand...

Actually for Vega the 10 is supposed to be the smaller one. They are named chronologically, not by size.
 
Glo.

Glo.

Diamond Member
Apr 25, 2015
5,711
4,559
136
What the hell then is RX 490 that would supposedly be released before the end of the year, if it must have wider than 256 bit memory bus, it cannot be Polaris 10, because it is full die?
 
StrangerGuy

StrangerGuy

Diamond Member
May 9, 2004
8,443
124
106
AMD can't be serious holding the fort with just Polaris 10 until March 2017 right? They won't even have an answer to GP104 let alone GP102.
 
D

dacostafilipe

Senior member
Oct 10, 2013
772
244
116
Did they not had a party for finalising Vega some days ago?

As software was already started because of Polaris, I don't think it should take that long :/
 
V

VulgarDisplay

Diamond Member
Apr 3, 2009
6,193
2
76
I could see them launching one of the Vega chips with gddr5x this year and then an hbm card next year.

Sent from my XT1575 using Tapatalk
 
Q

Qwertilot

Golden Member
Nov 28, 2013
1,604
257
126
StrangerGuy said:
AMD can't be serious holding the fort with just Polaris 10 until March 2017 right? They won't even have an answer to GP104 let alone GP102.
Click to expand...

I know people like to act as if they're locked into some sort of rather weird zero sum game with NV but they're really not.

They're just trying to turn a profit - any kind of profit/positive cash flow is good for them just now. Polaris they needed for the console stuff, so they did that and got a bunch of desk/laptop SKU's to try and sell more or less as a side effect.

The question is almost more what the purpose of Vega is going to be - how are they planning to turn a profit with it when big(ish) Pascal will be so well established by the time it launches? Hope they've got a plan.
 
P

plopke

Senior member
Jan 26, 2010
238
74
101
Little bit disappointing if it would only show in early 2017 , not sure if it matters since if NVIDIA sells a turd and charges you 500$ , some people will still buy it. I used to agree with the entire point of AMD has to have something out because ladidadida. In the end they have a card ready for the people buying price/performance, the enthusiastic people will buy whatever is best and the fanboys will buy whatever brand they favour. And not sure if AMD has still many fanboys to lose that might jump ship.

Is this good for the consumer NVIDIA prices of the 1070,etc not to be soul crunching , HELL NO but not much we can do about it :\.

Side notse:

-Making sense if it are only HBM2 designs but they always stated they can used GDDR5X, not sure why people think AMD has priority on HBM2 , unless stated in a contract pretty sure NVIDIA can bully the supply around with their cash.

-Isn't that quiet a old slide been roaming around for some time?

-And yes some "rumors" were indicating that they might push for a earlier launch but oh well , rumors are rumors.
 
T

trane

Member
May 26, 2016
92
1
11
I know it is taken for granted that Vega 10 is a large die, and Vega 11 is a larger die.

But is there any corroboration for this at all? All we know about for sure is that there was a Greenland chip, which is probably what Vega 10 became. With HBM2.

So, where do we get the idea that Vega 11 is a bigger die? As far as I can see, it's just hearsay.
 
K

krumme

Diamond Member
Oct 9, 2009
5,952
1,585
136
trane said:
I know it is taken for granted that Vega 10 is a large die, and Vega 11 is a larger die.

But is there any corroboration for this at all? All we know about for sure is that there was a Greenland chip, which is probably what Vega 10 became. With HBM2.

So, where do we get the idea that Vega 11 is a bigger die? As far as I can see, it's just hearsay.
Click to expand...
It doesnt make any sense to make two different dies for a market that is both very small and have plenty competing products from nv. 1080 titan 1080ti. Imo even vega doesnt make sense.

The only reason i can see is if ms ordered some special less larger size that could also be used? Or perhaps some professional market?

My guess is what we get is polaris and vega that sony and ms is using in their apu also.
 
J

jpiniero

Lifer
Oct 1, 2010
14,629
5,246
136
Vega is for HPC really. The smaller one is probably for the HPC APU while the bigger one is something to compete with GP100.

And by 'big' I mean probably 450-500ish mm2 for the bigger one.
 
J

jpiniero

Lifer
Oct 1, 2010
14,629
5,246
136
Worth noting that AMD removed "Scalability" from the Navi list of features.

digitaldurandal said:
Where did you get this information from?
Click to expand...

It's just a guess. The rumor that's been talked about is that the bigger Vega has 4096 cores, which with 1/2 DP and decent clock speed would be competitive with the Pascal Tesla.
 
poofyhairguy

poofyhairguy

Lifer
Nov 20, 2005
14,612
318
126
Qwertilot said:
The question is almost more what the purpose of Vega is going to be - how are they planning to turn a profit with it when big(ish) Pascal will be so well established by the time it launches? Hope they've got a plan.
Click to expand...

Meh, did Fiji ever really make them money? Seems like outside of the Nano that was a completely disappointing product line.

If I was AMD I would maybe avoid the high end too. Compete where you can.
 
P

Piroko

Senior member
Jan 10, 2013
905
79
91
1H 2017 is pretty much what was expected all along. That October rumor might be a dual Polaris card to bridge the 6 month gap.
 
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom