I see it losing in several benchmarks, so I would not agree that it is definitively faster than piledriver. Jaguar wins on cinebench and x264, loses on 3dmark and superpi and wprime. So no, not necessarily always faster...
Jaguar IPC is higher than PileDrive at low frequencies of 1.5-2GHz. Also Kabini doesn’t suffer from the CMT, a quad core 1.5GHz Kabini A4-5000 is faster than Quad core Piledriver at 1.6GHz A8-4550M.
But, I don’t expect a 2GHz Kabini to win over a 2.5GHz (3.5GHz turbo) Piledriver A10-5750M
I see it losing in several benchmarks, so I would not agree that it is definitively faster than piledriver. Jaguar wins on cinebench and x264, loses on 3dmark and superpi and wprime. So no, not necessarily always faster...
This is very interesting. As shown in the PS4 thread*, 4 jaguar cores have about the same performance than dual sandy bridge cores with hyperthreading (aka an i3). And 8 jaguar cores have about the same performance than quad sandy bridge cores with hyperthreading (aka an i7).
Now, if steamroller is twice more powerful than jaguar, this means that 4 steamroller cores will have about the same performance than quad sandy bridge cores with hyperthreading (aka an i7) or about the same performance than octo piledriver cores (aka an FX-8). :awe:
This would explain the new server roadmaps.
The 8C piledriver Opterons 3300 are replaced by 4C steamroller Berlin. And the Warsaw piledriver coming only in 12C/16C configurations. with the old configurations dropped.
I think it is safe to speculate that Warsaw chips will be finally replaced by future 6C/8C Berlin APU/CPU or what was named.
This twice performance of the new steamroller would explain also why the desktop roadmaps for 2013 only consider 2C/4C kaveri, with the early rumoured 6C being dropped. The 4C kaveri would be competing with a sandy bridge i7 (in CPU of course, kaveri graphics will be much more advanced).
*
http://forums.anandtech.com/showpost.php?p=35021461&postcount=1120
http://forums.anandtech.com/showpost.php?p=35028201&postcount=1193
They say that Berlin will have "almost" 8 x the Gflops/watt throughput
of a 16 cores 3.2g opteron wich has a theorical peak of 204.8 gflops
at 140W TDP = 1.46 Gflops/Watt , hence they are targeting about
1.46 x 8 = 11.68 Gflops/Watt , that is 1168 Gflops for a 100W Kaveri.
Since they said "almost" 8x let s round it to 1100Gflops.
Using 512 SPs at 1 GHz for the GPU they can reach 1000Gflops ,
the rest is to be provided by the FPU of the CPU , something
that will be between 50 and 100Gflops.
I see it losing in several benchmarks, so I would not agree that it is definitively faster than piledriver. Jaguar wins on cinebench and x264, loses on 3dmark and superpi and wprime. So no, not necessarily always faster...
yes...
Jaguar 2 cores vs 1 module is faster for most MT software at the same clock, but it doesn't matter because Jaguar can't work at the same clock...
it's like.. Pentium 4 vs Pentium 3 I guess.
for 35W and higher I think PD is simply better.
Well we don't know really know how far Jaguar can actually clock since currently Kabini is currently only go up to 2Ghz at 25W TDP.
That is strange if you think about it.
8core consoles. AMD will get what they wanted and needed. Mutlicore ultilization in games forced on game devs by next gen console design. Going back to 2/4core CPU now makes no sense, that is if games will truly be multithreaded.
But I can see how they may be frustrated or even disappointed by games optimizations (even GE titles!), and don't want to deal with those problems again.
I also find that part difficult to swallow, going back to a 4 core setup after the consoles would appear to make a strong case for 8.
They say that Berlin will have "almost" 8 x the Gflops/watt throughput
of a 16 cores 3.2g opteron wich has a theorical peak of 204.8 gflops
at 140W TDP = 1.46 Gflops/Watt , hence they are targeting about
1.46 x 8 = 11.68 Gflops/Watt , that is 1168 Gflops for a 100W Kaveri.
Since they said "almost" 8x let s round it to 1100Gflops.
Using 512 SPs at 1 GHz for the GPU they can reach 1000Gflops ,
the rest is to be provided by the FPU of the CPU , something
that will be between 50 and 100Gflops.
Testing performed by AMD Performance Labs. Calculated compute performance or Theoretical Maximum GFLOPS score for 2013 Kaveri (4C, 8CU) 100w APU, use standard formula of (CPU Cores x freq x 8 FLOPS) + (GPU Cores x freq x 2 FLOPS). The calculated GFLOPS for the 2013 Kaveri (4C, 8CU) 100w APU was 1050. GFLOPs scores for 2011 A-Series “Llano” was 580 and the 2013 A-Series “Trinity” was 819. Scores rounded to the nearest whole number.
(CPU Cores x freq x 8 FLOPS)
I'd imagine if you scale Jaguar cores up to higher clock speeds, it'd perform a lot like a Phenom II would. If I'm not mistaken, the original Bobcat cores were based a lot off a stripped down K10 design with "90% the IPC of a K8" as the target goal. Jaguar nicely improves on that, but now that I think about it more (because what I'm saying probably contradicts what I've posted in the past), I don't think it'd make it a more competitive chip [at the high end] than AMD's Bulldozer line.
yes a SR as detailed by hotchips, not the die shot floating around.
you are mistaken, bobcat is a complete new design, nothing in it shares anything with K8. jaguar continues the deviation by implementing cache hierarchies that AMD have never implemented before.
K8 was used as a benchmark of performance and im guessing that's because it too only had 64bit ALU's so it was most comparable.
I know folks say it was a complete new design but just how much of a "complete new design" is truly "new"?
Would it make sense, both engineering-wise as well as project management-wise, to build a bobcat core from the ground up by paying zero heed to the decades of lessons learned and accessible know-how that had gone into every preceding microarchitecture?
I find it hard to believe, but agree it is not impossible, that bobcat was created in the vacuum of taking liberties with existing circuit blocks and experience.
I know folks say it was a complete new design but just how much of a "complete new design" is truly "new"?
Would it make sense, both engineering-wise as well as project management-wise, to build a bobcat core from the ground up by paying zero heed to the decades of lessons learned and accessible know-how that had gone into every preceding microarchitecture?
I find it hard to believe, but agree it is not impossible, that bobcat was created in the vacuum of taking liberties with existing circuit blocks and experience.