Samsung and Global Foundries will produce AMD's Next Gen Greenland GPU and Zen CPU

Page 12 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Feb 2, 2009
12,997
283
126
Well actually AMD always had a second source, they always made their GPUs on TSMC.
 

guskline

Diamond Member
Apr 17, 2006
5,340
65
126
I agree that having Samsung in the mix is good for AMD. Being tied only to GloFo isn't.

I think the question is "Did the GloFo relationship with AMD require AMD to produce all new gpus with GloFo?"

If so the Samsung revelation probably indicates that GloFo and Samsung worked out a deal that actually benefits AMD.
 
Last edited:
Apr 22, 2012
20,395
0
106
-However AMD is not tied down by the WSA as much as before, since they e.g. can start producing the chips at Samsung and then later move production to GF.
Its doesn't work that way. They use different tools etc. So you cant just transfer it from one to the other. Its not a copy exact setup. You need chip redesign to move it.
 

monstercameron

Diamond Member
Feb 12, 2013
3,829
0
0
I can't really see where all these conclusions of yours are coming from. Here are my two cents:

- Apple indeed got a second-source for their chips, but that second source is Samsung, not Globalfoundries, because their first source is still (and will be for the foreseeable future) TSMC. Globalfoundries got a total of 0 Apple orders allocated to them.

- That Samsung might be manufacturing chips for AMD is much more likely of a settlement between Samsung and Globalfoundries than a sudden relief from the stringent conditions on the WSA. The trend on the WSA is for it to become more and more stringent, not lax: It comprised part of the CPU, then all CPUs, then finally all CPUs and part of the GPU line up.

- Globalfoundries wasn't shy to make AMD swallow a bill to backport a low volume part like Jaguar to their 28nm process.

- Nothing was reported at AMD that tilted the WSA balance towards them, quite the opposite, the shrinking volumes gives even more bargaining power to Globalfoundries.
That seems counter intuitive why would amds shrinking volume not give them better leverage when the whole wsa thing has been about making sure amd had enough manufacturing capacity.
 
Apr 22, 2012
20,395
0
106
That seems counter intuitive why would amds shrinking volume not give them better leverage when the whole wsa thing has been about making sure amd had enough manufacturing capacity.
Why would lower volume give AMD anything in relation to the WSA as leverage? It means AMD is being weaker and weaker in the negotiation and got less to say about it.
 

mrmt

Diamond Member
Aug 18, 2012
3,976
0
76
That seems counter intuitive why would amds shrinking volume not give them better leverage when the whole wsa thing has been about making sure amd had enough manufacturing capacity.
No, the WSA was about giving Globalfoundries a guaranteed stream of revenue from AMD for the duration of the agreement.

And not fulfilling this commitment triggers the take-or-pay clause, which means that by the end of the year AMD must chose between paying the charge with cash they do not have or offering more commitments outside the initial scope of the contract in order to waive the take-or-pay charge. Do you think that AMD backported Jaguar to Globalfoundries because it was great business? Were AMD able to fulfill its commitments it wouldn't have to worry about it.
 

mrmt

Diamond Member
Aug 18, 2012
3,976
0
76
We have no idea how this settlement is arranged, we can only speculate. But the main point is that AMD now can have their chips produced at both Samsung and GF. And Samsung already has the process up and running. So in that respect AMD is not tied down by the WSA as much as before. AMD's chip launches will not be delayed by GF potentially now having their 14 nm process ready in time.
I'm not saying that it's negative for AMD, being able to manufacture chips in any place other than Globalfoundries is certainly good news no matter how this is get done.

There's a difference between being able to manufacture at Globalfoundries foundry partner through Globalfoundries and being able to manufacture chips wherever it suits you, what I'm saying is that given the history between the two companies the deal is likely to be the former, not the latter, and that the WSA can be even more stringent than before.

Why is this important? Globalfoundries already said that they won't go with Samsung for 10nm, so this will be a one off deal if the WSA is still in full force. The agreement can even be more stringent than before, especially if the orders are being channeled through Globalfoundries (and that would make business sense, as the orders would count towards the WSA quota).

OTOH what you are speculating is that AMD is somehow free of the WSA, but there is nothing pointing out for that and AMD didn't tell investors that too. And in the past whenever WSA conditions went for the worst they had to describe the changes to investors, nothing of that happened in the last few months.
 
Feb 6, 2011
1,862
255
136
Actually, there were some tweaks to L1 and L2 cache in Carrizo that flew under my radar. The L1 data cache doubled in size without reduction in efficiency while the l2 size was cut in half. The l2 is faster in Carrizo than it is in Kaveri . . .
.
Yes the sizes changed but nothing else did, amd never stated the 1mb l2 latency for ex so we only have some very old numbers. But the whole mostly write through thus mostly inclusive cache policy the wcc and the poor latencies never got fixed. Davr kanter on rwt has said that as he understands it the cmt design creates massive layout issues for the l2 and thats a big reason for its poor performance. He also said the actual l2 arrays are very fast, so hopefully we are goign to see a ~10 cycle l2 for Zen.
 

Tuna-Fish

Senior member
Mar 4, 2011
979
242
136
Its doesn't work that way. They use different tools etc. So you cant just transfer it from one to the other. You need chip redesign to move it.
This is incorrect. The process and tools used at GloFo are exactly the same as those at Samsung. If you could transport masks, you could use the same masks at both facilities to produce the same chips. (Since you can´t, you just have to pay extra for duplicating your masks.)

Its not a copy exact setup.
Copy exact refers to the way the manufacturing is set up at the GloFo facility, it has no impact on the process itself, other than how lack of using it makes it harder to replicate good yields.
 
Apr 27, 2000
12,764
1,563
126
Yes the sizes changed but nothing else did, amd never stated the 1mb l2 latency for ex so we only have some very old numbers. But the whole mostly write through thus mostly inclusive cache policy the wcc and the poor latencies never got fixed.
Right, but having more l1 is pretty nice. If you have software that stays mostly in cache during operation, you may see a nice performance boost moving from SR to XV. The penalty for inclusive cache architecture is most certainly still there.

Davr kanter on rwt has said that as he understands it the cmt design creates massive layout issues for the l2 and thats a big reason for its poor performance. He also said the actual l2 arrays are very fast, so hopefully we are goign to see a ~10 cycle l2 for Zen.
If Zen has 10 cycle l2 then that will be awesome. AMD hasn't had that in forever.
 

prtskg

Senior member
Oct 26, 2015
229
24
71
This is incorrect. The process and tools used at GloFo are exactly the same as those at Samsung. If you could transport masks, you could use the same masks at both facilities to produce the same chips. (Since you can´t, you just have to pay extra for duplicating your masks.)

Copy exact refers to the way the manufacturing is set up at the GloFo facility, it has no impact on the process itself, other than how lack of using it makes it harder to replicate good yields.
Yes you are right. There was an article in english.etnews.com about porting between samsung and GF foundries. The article also EXPECTED that 14nm deal was not just between GF and Samsung but AMD being included in in the deal too. Maybe the article was right as this benefits all 3 of them. AMD will be able to bring its products earlier and there shouldn't be problems of volume at least after Q3 when both the foundries will be in full swing. GF will lose some volume from AMD but will gain from others as it'll not be too late to 14nm and Samsung will gain new customers.
 
Jul 5, 2015
133
0
0
Surely the console wafers by themselves should be enough to fulfil the WSA obligations, and it's not like the silicon has to be on the cutting edge.
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,395
0
76

dark zero

Platinum Member
Jun 2, 2015
2,516
8
91
If that so... expect Samsung helping AMD also on the HBM2 production.

Hoping thag HBM1 enters as standard too since it might help nVIDIA and AMD a lot.
 
May 2, 2008
190
102
116
www.chip-architect.com
If that so... expect Samsung helping AMD also on the HBM2 production.

Hoping thag HBM1 enters as standard too since it might help nVIDIA and AMD a lot.
HBM1 and HBM2 are part of the same standard. (JEDEC Standard no. 235A)

There are parts defined with 1.0Gbps, 1.6Gbps and 2.0Gbs. SKHynix and
Samsung are both producing the 2.0Gbps 8GByte versions (Samsung will
demonstrate a 2.4Gbps version next month with 307.2 GByte/s. at ISSCC)
 
Last edited:

el etro

Golden Member
Jul 21, 2013
1,581
0
81
HBM1 and HBM2 are part of the same standard. (JEDEC Standard no. 235A)

There are parts defined with 1.0Gbps, 1.6Gbps and 2.0Gbs. SKHynix and
Samsung are both producing the 2.0Gbps 8GByte versions (Samsung will
demonstrate a 2.4Gbps version next month with 307.2 GByte/s. at ISSCC)
Hey Hans, do you know if we will see a Zen briefing at ISSCC 2016?
 
Apr 27, 2000
12,764
1,563
126
Surely the console wafers by themselves should be enough to fulfil the WSA obligations, and it's not like the silicon has to be on the cutting edge.
Unknown. The volume of sales surely exceeds that of desktop Kaveri and FX processors, and AMD has been moving production of their chips away from TSMC and towards GF. There's still the question of which wafers GF needs/wants AMD to take as a part of the WSA. If AMD is still required to take wafers from fabs only tooled for 32nm SOI (for example), then what do they do? I'm not sure it's that restrictive, though, and there's always chipsets . . .
 

Dresdenboy

Golden Member
Jul 28, 2003
1,730
0
136
citavia.blog.de
That would be odd, because it doesn't match shipments of PS4/Xbox One. If anything 2014 was a better year. Also backing up financials. So they had to have an explosion in sales in Q4 to make this possible.

Now we have:
Sony said:
Tokyo, January 5, 2016 – Sony Computer Entertainment Inc. (SCE) today announced that the PlayStation®4 (PS4™) computer entertainment system has sold through more than 5.7 million units*1 during the 2015 holiday season*2. PS4 has now cumulatively sold through more than 35.9 million units globally as of January 3, 2016.
Now it looks like the same number of units sold according to your chart.
 
Apr 22, 2012
20,395
0
106
Yes, this number is also down from 2014 holidays that accounted for 6.4 million.

But it shouldn't be a surprise since the financials state it as well :)
 
May 2, 2008
190
102
116
www.chip-architect.com
Now we have:

Now it looks like the same number of units sold according to your chart.
According to official Sony numbers:

2013: _4.2 million Playstation 4 sold
2014: 14.3 million Playstation 4 sold (total 18.5)
2015: 17.4 million Playstation 4 sold (total 35.9)

Holiday Season 2014 - PS4 sold: 4.1 million
Holiday Season 2015 - PS4 sold: 5.7 million


http://bgr.com/2014/01/07/playstation-4-sales-2013/
http://www.cnet.com/news/playstation-4-sales-hit-18-5-million-in-2014/
 
Last edited:
Mar 10, 2006
11,719
122
126
Apple's A10 can not be exclusively produced by TSMC, which slashed its
capex by 30% due to lost opportunities in the mobile sector... To late now.
10nm is the next opportunity, the next race.
Do you still stand by this?

We estimate our foundry market segment share of 16, 14-nanometer node increases from about 40% in 2015 to above 70% in 2016 exceeding the previous prediction we made in mid-2014.
http://seekingalpha.com/article/381...-results-earnings-call-transcript?part=single
 


ASK THE COMMUNITY

TRENDING THREADS