AMD Excavator on 28 nm, some production moving back to GF?

Fjodor2001

Diamond Member
Feb 6, 2010
3,938
408
126
http://www.extremetech.com/computin...-production-is-moving-back-to-globalfoundries

AMD leak confirms that Excavator APU [actually Carrizo APU using Excavator CPU cores] will be 28nm, and that some production is moving back to GlobalFoundries
[...]
A reader tipped us off to a forum post over at SemiAccurate where someone noticed LinkedIn data suggesting that Carrizo, AMD’s first APU with an Excavator-class CPU, is built on 28nm.




Just leaks/rumors currently though, as indicated by the text in the article.
 
Last edited:

Vesku

Diamond Member
Aug 25, 2005
3,743
28
86
Staying on 28nm doesn't give them much to play with unless Excavator is going to be a decent amount larger than SteamrollerB in allocated die space.
 

inf64

Diamond Member
Mar 11, 2011
3,765
4,223
136
Same socket as Kaveri. The author switched the words for Carrizo and EX but it doesn't matter that much^^.

Since 20nm is such a mess no wonder they are sticking with process that had Kaveri as "trial run". It should be solid enough by the time Carrizo is ready for market. Better hope that Carrizo is indeed Kaveri on steroids since it will be going against very tough competition.
 

erunion

Senior member
Jan 20, 2013
765
0
0
Staying on 28nm doesn't give them much to play with unless Excavator is going to be a decent amount larger than SteamrollerB in allocated die space.

20nm was an impossibility. AMDs roadmap was for a new arch every year. Obviously AMD didn't plan on having 20nm ready one year after 28nm.

Plus, when is GFs 20nm going be ready? If EXC was tied to 20nm we wouldn't see it until 2016-2017.
 

MisterMac

Senior member
Sep 16, 2011
777
0
0
Well that doesn't really bode well for Carrizo bringing much to the table.


Now that we're in the tinfoil hat zone - do you think Intel might have known about this and one of weighed in reasons that Broadwell atleast on desktop seems to have been "chosen" to be slipped?

Why bother rush desktop\mobile broadwell out - if Carrizo is going to be GF 28nm?

OEMs certainly can't seem to twist intel's arm at all.
 

NTMBK

Lifer
Nov 14, 2011
10,269
5,134
136
Christ. Stuck on 28nm, stuck with dual channel DDR3. The only possible way that Carrizo could excite me was if it had a nice fat ESRAM/EDRAM cache to speed up the graphics.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
Excavator will most probable use the new High Density Cell Libraries. If the claim about 30% area and power reduction is true then Carrizo APUs should be fine.



Also to note, that Excavator is suppose to bring the highest performance, Steamroller was going for more parallelism.
 

bullzz

Senior member
Jul 12, 2013
405
23
81
when is this supposed to be released? will be interestng to see how these stack up against broadwell i3s
 

el etro

Golden Member
Jul 21, 2013
1,581
14
81
Will be good to see XV like Maxwell bring awesome efficiency on the same 28nm node. Bad will be see AMD still stuck on a so overpast process in 2015.

Also to note, that Excavator is suppose to bring the highest performance, Steamroller was going for more parallelism.

This was part on pre-Bulldozer era, no? Did the plans shifted these years?
 

jpiniero

Lifer
Oct 1, 2010
14,842
5,457
136
Kinda offtopic, but I wonder when people are going to start getting in trouble/fired for posting stuff like this on LinkedIn.
 

Abwx

Lifer
Apr 2, 2011
11,172
3,868
136
http://www.extremetech.com/computin...-production-is-moving-back-to-globalfoundries





Just leaks/rumors currently though, as indicated by the text in the article.

Thoses slides suggest that it could well be taped out at 28nm
for debugging purposes and then ported to 20nm once the process
is ready , this would allow devellopement of the APU without waiting
for GF to get its next node ready , this would save considerable
amount of time while the cost would be negligible overall, tests waffers
costs for very low quantities is about 200 000$/waffer at 28nm,
it would cost them 1200-2000$ per die , the cost of the mask is
millions but several iterations are needed anyway so if they can
get a functional one at 28nm it would be all benefit for the next node.
 

Sweepr

Diamond Member
May 12, 2006
5,148
1,142
131
''Excavator is suppose to bring the highest performance''

Funny, it almost seems like I heard that before.
Just a month after we've discovered A10-7850K barely matches A10-6800K's CPU performance @ stock and here comes the ''next year's stuff will fix everything'' talk.
Outperforming 95W Kaveri with a 65W chip @ same 28nm node would already be quite a feat, yet it looks like some people (not talking about a specific user) expect it to demolish Kaveri CPU-wise and still pull out Maxwell-like perf/watt gains. Not impossible but looking back a few years, I'd keep my expectations in check.
 
Last edited:

norseamd

Lifer
Dec 13, 2013
13,990
180
106
would like a 8 core excavator fx on the old 28 mm process at 125 or 140 watts. would like to see what it can do.

what is the difference between am3+ and fm2+. can a fm2+ support a fx processor.

thinking about all the talent they hired. after excavator could we see full cores come back and no more modules. a cpu with 8 full cores and better ipc and such over excavator would be interesting.

what will these guys make

raja koduri
jim keller
mark papermaster
 

lakedude

Platinum Member
Mar 14, 2009
2,627
371
126
I know Intel is way ahead with 14nm right around the corner but I don't know if that matters as much as it did in days past, especially for desktops. Intel's 32nm chips are still pretty competitive and it seems like they overclocked better/easier. Going smaller just concentrates the heat into a smaller space.

AMD's bigger problem is architecture, not process size.

Now mobile is another matter...
 
Last edited:

bullzz

Senior member
Jul 12, 2013
405
23
81
@Abwx - I dont think so. why would they tape out on 28nm just for debugging when they can debug in an emulator which is easier, faster and cheaper
we dont know where GF is with their 20nm processes. I think it could be more for testing the process than the chip
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |