[AMD Processors] The future CPUs

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Flash831

Member
Aug 10, 2015
60
3
71
In 2012, AMD released their Piledriver architecture, which is what is in all currently-for-sale AMD socket AM3+ CPUs (4, 6 and 8 core, 32nm).

In 2013, AMD released a refresh to Piledriver called Steamroller, which is what is in their current FM2+ APUs (4 cores only, 28nm).
Steamroller is not a refresh, it's a new architecture.
Otherwise Piledriver is also just a "refresh" of the original Bulldozer.

Piledriver (AM3+, 32nm) has slowly been reduced in price as Intel has released new CPUs, but has not been refreshed.
Piledriver has recieved a number of smaller revisions during the years to offer better performance and less leakage. The last one was in 2014, "King Vishera" (according to The Stilt).
FX-8370 and FX-8370E is of the latest revision.
 

mrmt

Diamond Member
Aug 18, 2012
3,974
0
76
Not lately. Intel's track record has actually been very crappy. 14 nm delays, Broadwell delays, 10 nm Cannonlake delays, Skylake shortage leading to actual delays, Haswell Refresh and KabyLake stopgap solutions introduced in panic, ... the list goes on.

You are taking 2-3 years of Intel deviating from their plan of record while AMD has an entire decade of delays/cancelations to talk about? You must be joking, really. For starters the two companies aren't in the same league.
 

YBS1

Golden Member
May 14, 2000
1,945
129
106
Unfortunately, we don't even have a vague idea of what performance will be like.

If "40% higher IPC than current AMD Excavator core" is true, Zen will still be slower than Sandy Bridge per clock. I'm hoping it's more than 40%.

Sandy Bridge level performance won't necessarily be a bad thing though, especially given 8 cores / 16 threads. It would definitely make AMD a viable option again depending upon what kind of workload you are wanting it for. I know I'd personally take a hypothetical 8 core SB-E over a 6700K any day of the week.
 

IlllI

Diamond Member
Feb 12, 2002
4,927
10
81
Sandy Bridge level performance won't necessarily be a bad thing though, especially given 8 cores / 16 threads. It would definitely make AMD a viable option again depending upon what kind of workload you are wanting it for. I know I'd personally take a hypothetical 8 core SB-E over a 6700K any day of the week.

I hope it is a little higher than that. I'm in no hurry to upgrade, but around 2017 they better have something that can do better than sandy bridge performance. otherwise i'll just buy intel again, and i suspect most would as well.
 

Yuriman

Diamond Member
Jun 25, 2004
5,530
141
106
Sandy Bridge level performance won't necessarily be a bad thing though, especially given 8 cores / 16 threads. It would definitely make AMD a viable option again depending upon what kind of workload you are wanting it for. I know I'd personally take a hypothetical 8 core SB-E over a 6700K any day of the week.

While true, it puts them squarely back where they were when Piledriver came out: More cores, lower IPC.

Some food for thought:


















Interestingly, the gap between Skylake and Sandy Bridge looks to be very roughly as large as Sandy vs Piledriver.
 
Last edited:

Fjodor2001

Diamond Member
Feb 6, 2010
3,989
440
126
Sure thing Mr. Skylake is H2 2016.

Looked in the mirror?

I said Skylake in 2015H2 to 2016. Turns out I was quite correct, given the delays. Also it was a rough estimate, since leaked schedules kept shifting back and forth. Should I bring up your "there will be no APU in the PS4" prediction again? Or one of the many other of your failed predictions? To what end?

It looks like one more troll has entered the arena to derail yet another thread...
 

YBS1

Golden Member
May 14, 2000
1,945
129
106
While true, it puts them squarely back where they were when Piledriver came out: More cores, lower IPC.

Some food for thought:
















Now imagine an 8 core version of the 2600K in those charts instead of a 4 core one. I'm not saying it would make the most idea gaming cpu but it would destroy the 6700K in most of those benches. This of course is assuming it is at Sandy Bridge levels of performance. It would still be in a difficult price niche though, just as the 6700K is, I'd rather spend just a tiny bit more and get the 5820K.
 

MiddleOfTheRoad

Golden Member
Aug 6, 2014
1,123
5
0
Sorry, but Intel track record is *much* better than AMD. I don't think you really want to discuss this, but if you do, I would be happy to bring every single AMD roadmap since 2008, then we can compare against Intel's and see which company has the biggest deviation.

Yeah, right.... I've seriously lost track of how many paper launches/delays Intel is guilty of. They are just as bad at missing dates to anyone with actual objectivity. Actually, if you add up all the paper launches and delays, Intel is probably worse than AMD.

A growing trend from Intel:
http://www.engadget.com/2015/07/16/intel-skylake-chips-delayed/

Broadwell delayed:
http://arstechnica.com/gadgets/2015...ore-broadwell-cpus-and-the-iris-pro-6200-gpu/

Skylake delayed:
http://www.financialstrend.com/inte...h-the-launch-of-skylake-to-q4-2015-22252.html

Devil's Canyon delayed:
http://wccftech.com/intel-paper-lau...e-september-desktop-broadwell-pushed-q2-2015/

10nm delayed:
http://techreport.com/news/28642/intel-delays-10-nm-process-third-14-nm-cpu-to-follow-skylake

Here's a classic -- the paper launch of the Pentium III 1 Ghz:
http://windowsitpro.com/windows-server/1ghz-processor-explosion
 
Last edited:

Yuriman

Diamond Member
Jun 25, 2004
5,530
141
106
Now imagine an 8 core version of the 2600K in those charts instead of a 4 core one. I'm not saying it would make the most idea gaming cpu but it would destroy the 6700K in most of those benches. This of course is assuming it is at Sandy Bridge levels of performance. It would still be in a difficult price niche though, just as the 6700K is, I'd rather spend just a tiny bit more and get the 5820K.

True, but the 8350 performed just fine against a 2600K in multithreaded things, aside from the 2600K's absurd overclocking potential. The issue was that Intel's chips were in the range of 40%(ish) faster in single-threaded stuff, and murdered AMD's chips in anything that couldn't fully utilize all of those threads. Yes, the software landscape has changed, but I suspect we're going to be talking about 8 threads vs 16, rather than 4 vs 8. It strikes me as highly unlikely that AMD will price their 8c/16t Zen chip against Intel's quads, if they can avoid it, but they may not have a choice.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
True, but the 8350 performed just fine against a 2600K in multithreaded things, aside from the 2600K's absurd overclocking potential. The issue was that Intel's chips were in the range of 40%(ish) faster in single-threaded stuff, and murdered AMD's chips in anything that couldn't fully utilize all of those threads. Yes, the software landscape has changed, but I suspect we're going to be talking about 8 threads vs 16, rather than 4 vs 8. It strikes me as highly unlikely that AMD will price their 8c/16t Zen chip against Intel's quads, if they can avoid it, but they may not have a choice.

ZEN will allow AMD to only use Quad Core 8 Threads APUs against the mainstream Intel platforms like Socket 1151.
Those 8 Core 16 Threads CPUs will compete against Intel HEDT platform.
The added feature this time around will be the common socket AM4 for both APUs and CPUs.
 
Last edited:

YBS1

Golden Member
May 14, 2000
1,945
129
106
It strikes me as highly unlikely that AMD will price their 8c/16t Zen chip against Intel's quads, if they can avoid it, but they may not have a choice.

That's why I mentioned the 5820K, I don't think they will have much choice. I think at the ~$200-350 price points it will end up being a choice of if you want exceptional multi-threaded performance or best IPC. Intel has really put the HEDT platform in a good price bracket to drive adaption with the 5820K, I'm assuming they will keep doing something similar going forward.
 

mrmt

Diamond Member
Aug 18, 2012
3,974
0
76
Yeah, right.... I've seriously lost track of how many paper launches/delays Intel is guilty of. They are just as bad at missing dates to anyone with actual objectivity. Actually, if you add up all the paper launches and delays, Intel is probably worse than AMD.

Do you want to compare a delay to launch a product (Intel) with getting out of entire market segments (AMD)? So much for your actual objectivity at work here.
 

Yuriman

Diamond Member
Jun 25, 2004
5,530
141
106
ZEN will allow AMD to only use Quad Core 8 Threads APUs against the mainstream Intel platforms like Socket 1151.
Those 8 Core 16 Threads CPUs will compete against Intel HEAD platform.
The added feature this time around will be the common socket AM4 for both APUs and CPUs.


If 4c/8t Zen is only 40% faster than Excavator...



... then it will not compete with Skylake, much less Kaby Lake or Cannonlake. AMD will need more than 40% over their current uarch to compete with the same number of cores/threads. Because of this, I'm hoping the 40% figure is incorrect.



Here we're seeing a Skylake i5 as being more than twice as fast as Kaveri. 40% more is simply not enough.
 
Last edited:

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
If 4c/8t Zen is only 40% faster than Excavator...



... then it will not compete with Skylake, much less Kaby Lake or Cannonlake. AMD will need more than 40% over their current uarch to compete with the same number of cores/threads. Because of this, I'm hoping the 40% figure is incorrect.



Here we're seeing a Skylake i5 as being more than twice as fast as Kaveri. 40% more is simply not enough.

This is about 40% higher IPC over Excavator, its not single thread or MT CPU performance.
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,989
440
126
That's why I mentioned the 5820K, I don't think they will have much choice. I think at the ~$200-350 price points it will end up being a choice of if you want exceptional multi-threaded performance or best IPC. Intel has really put the HEDT platform in a good price bracket to drive adaption with the 5820K, I'm assuming they will keep doing something similar going forward.

The 8 core Intel HEDT 5960X is $1000.
 

Yuriman

Diamond Member
Jun 25, 2004
5,530
141
106
This is about 40% higher IPC over Excavator, its not single thread or MT CPU performance.

Are we then expecting Zen to have far, far higher clocks? I thought we were talking about quad vs quad.
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,989
440
126
... then it will not compete with Skylake, much less Kaby Lake or Cannonlake. AMD will need more than 40% over their current uarch to compete with the same number of cores/threads. Because of this, I'm hoping the 40% figure is incorrect.
There is no 8 core Skylake-E yet, much less KabyLake or CannonLake. And KabyLake will be the same uArch as Skylake anyway.

Zen will compete against 8 core Broadwell-E. Intel price their 8 cores CPUs at $1000, see 5960X. If AMD Zen can deliver 8 cores for $400-500 or so, it'll be a great option (even if at around IB performance level per core). Much better than what Intel has in store.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
Are we then expecting Zen to have far, far higher clocks? I thought we were talking about quad vs quad.

I have no idea at what clocks ZEN will come, all im saying is that its another thing to say 40% higher IPC and another thing to compare ST/MT CPU performance.

Also games are very cache depended some times and from what we know ZEN has a completely different Cache architecture(more like Intel's) than what AMD is using today.
 

Yuriman

Diamond Member
Jun 25, 2004
5,530
141
106
I have no idea at what clocks ZEN will come, all im saying is that its another thing to say 40% higher IPC and another thing to compare ST/MT CPU performance.

Also games are very cache depended some times and from what we know ZEN has a completely different Cache architecture(more like Intel's) than what AMD is using today.

IPC and ST-performance (taking clockspeed into account) are nearly synonymous. And all I'm saying is, it needs to be a LOT more than 40% more ST/MT performance.
 

Sweepr

Diamond Member
May 12, 2006
5,148
1,143
131
Depending on clocks, SB/IB level IPC 8C/16T would still lose to Broadwell-E 'Core i7 6820K' 3.4-3.6GHz (base) at the majority of common desktop benchmarks and games. That is, assuming AMD can deliver this kind of performance at up to 95W.

And their APUs won't be facing Broadwell cores but Kabylake (same/similar to Skylake?) or Cannonlake cores, depending on when they arrive in 2017.
 

YBS1

Golden Member
May 14, 2000
1,945
129
106
Yeah, but Zen will be 8 cores. So the 5960X is what you should be comparing to.

LOL...no. Zen would have to be an amazing increase in per core performance to be able to compete against the 5960X. That was the whole point of the exchange above, assuming it performs at Sandy Bridge levels, the 8 core will probably end up slotting in with the top mainstream intel chips and low end HEDT chips. The performance and market demand dictates price, not the amount of cores.
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,989
440
126
LOL...no. Zen would have to be an amazing increase in per core performance to be able to compete against the 5960X. That was the whole point of the exchange above, assuming it performs at Sandy Bridge levels, the 8 core will probably end up slotting in with the top mainstream intel chips and low end HEDT chips. The performance and market demand dictates price, not the amount of cores.

Seems people are expecting it to be in IB/Haswell performance range per core. Since it'll be competing against Broadwell-E, Zen should be pretty close. At least far closer to Intel 8 core HEDT than 6 core HEDT.
 
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/    |