http://www.hardocp.com/ne...g_sample_benchmarks63/
 
Authenticity, who knows but it's interesting!
 
 
 
 

If that is real, then that is most disappointing indeed.

In before Seronx and Lehpron posts/rundown..possibly more interesting than the actual benchmarks:)
 
 

IBL +2(In Before Lehpron)
 
Maybe we will get some action from the new AMD management for the non-factory CPU's....

4 cores 8 threads in that one bench? o.O

seronx and I are very different people, for one he already made up his mind.  My interest in AMD is not in the FX parts, only in the APU's: Either Intel Haswell vs. AMD Kaveri, no doubt Intel will be better on CPU-intensive, I'm only curious about the IGP.  I'm planning on Maxwell/Sea Islands graphics card, which means I'll be stuck on IGP for a while and it needs to be decent enough from my GTX260 and as RMA backup.  AMD is so far lucky with Llano and Trinity, but if they drop the ball on Kaveri, I'm done.  The rumor that Intel plans to put 64MB of integrated flash memory as an L4 dedicated to the Haswell IGP is intriguing; unless they drop the ball too, I'll be a sad panda.
 
As for the leak, why does it say 4C/8T?  I read that for some reason some programs react to AMD's cores like SMT rather than physical cores which has lead to the suspicion that their FX8000's are really just quads with HT, but I don't know enough about the situation to be so certain.  A relatively underperforming 8-core doesn't mean there are actually less cores, Intel bothered to make 1.8GHz Xeon 8-core in LGA2011 which no doubt any i7 would beat; which ironically beats this AMD ES processor.
 
Another aspect of curiousity, Cinebench is purely CPU-intensive, so it scales with frequency and cores 100% within the same architecture; just like 3Dmark06's CPU test is also fully CPU-intensive.  So far, the reports for AMD's "8-core" are of a stock 4GHz part with Turbo to 4.2GHz.  So if this leak shows a 3.32GHz ES part scoring 5.73, then the retail FX8350 should actually score 7.0.  In the sense of absolute performance, yes it is pitiful given clock-for clock against Intel hasn't changed in six years.  But the only thing AMD has going for them is retail price in the end, it will always be cheaper than the performance difference.

Interesting read, definitely going to be an interesting period ahead of us!

lehpron
seronx and I are very different people, for one he already made up his mind.  My interest in AMD is not in the FX parts, only in the APU's: Either Intel Haswell vs. AMD Kaveri, no doubt Intel will be better on CPU-intensive, I'm only curious about the IGP.  I'm planning on Maxwell/Sea Islands graphics card, which means I'll be stuck on IGP for a while and it needs to be decent enough from my GTX260 and as RMA backup.  AMD is so far lucky with Llano and Trinity, but if they drop the ball on Kaveri, I'm done.  The rumor that Intel plans to put 64MB of integrated flash memory as an L4 dedicated to the Haswell IGP is intriguing; unless they drop the ball too, I'll be a sad panda.

Steamroller is mainly a CPU enhancement* rather than a GPU enhancement.  Kaveri has:
4 SR Cores vs 4 PD Cores
8 GCN Clusters vs 6 VLIW4 Clusters
Dual Channel DDR4 vs Dual Channel DDR3
FM2+ vs FM2

lehpron
As for the leak, why does it say 4C/8T?

Windows 7 patches

lehpron
I read that for some reason some programs react to AMD's cores like SMT rather than physical cores which has lead to the suspicion that their FX8000's are really just quads with HT, but I don't know enough about the situation to be so certain.

Windows 7 patches

lehpron
A relatively underperforming 8-core doesn't mean there are actually less cores, Intel bothered to make 1.8GHz Xeon 8-core in LGA2011 which no doubt any i7 would beat; which ironically beats this AMD ES processor.

I'm not sure if they ever updated Cinebench R11.5 but currently R11.5 uses AVX codepaths for Intel and SSE2 codepaths for AMD.  This wasn't a problem with Phenom II but is a huge problem for Bulldozer-derived architectures.

lehpron
Another aspect of curiousity, Cinebench is purely CPU-intensive, so it scales with frequency and cores 100% within the same architecture; just like 3Dmark06's CPU test is also fully CPU-intensive.  So far, the reports for AMD's "8-core" are of a stock 4GHz part with Turbo to 4.2GHz.  So if this leak shows a 3.32GHz ES part scoring 5.73, then the retail FX8350 should actually score 7.0.  In the sense of absolute performance, yes it is pitiful given clock-for clock against Intel hasn't changed in six years.  But the only thing AMD has going for them is retail price in the end, it will always be cheaper than the performance difference.

It falls in line with the Trinity improvements both use improved Komodo cores.
 
*Kaveri improvements to the core:
Improved Front-end with the ability to dispatch to both cores simultaneous. (Decouples dispatch from the VMT front-end)
Improved Floating-point unit with the ability to execute 4 AVX2** FMA3(Intel)/FMA4(AMD) ops per cycle(2 FP+2 Int)
The cache design was also changed but the only changes that are available is L2 size for Kaveri.  Which is 2.4 MB per L2.
 
**Please note: http://software.intel.com...tref_avx2_overview.htm
AVX2 -> 256-bit Floating-point and 256-bit Integer
 
Ivy Bridge iGPU:
24 EUs * 4 32-bit * 2 FMA * ~1.3 GHz => ~249.6 GFlops
 
Haswell iGPU:
40 EUs * 4 32-bit * 2 FMA * ~1.5+ GHz => ~480 GFlops w/ 64 MB L4
 
Trinity iGPU:
384 ALUs * 1 32-bit * 2 FMA * ~0.76 GHz => ~583.68 GFlops w/ 256-bit bus connected to two independent 64-bit DDR3 channels
^ 7660D
 
Kaveri iGPU:
512 ALUs * 1 32-bit * 2 FMA * ~0.76 GHz => ~778.24 GFlops w/ 256-bit bus connected to two independent 64-bit DDR4 channels
^ 8660D
 
CPU wise: (Only going to do Kaveri and Haswell)
Kaveri -> 4 Cores * 4 FP units * 4 32-bit * 2 FMA * ~4 GHz => ~512 GFlops
Haswell -> 4 Cores * 2 FP Units * 8 32-bit * 2 FMA * ~4 GHz => ~512 GFlops
^-- The design choices are really which one is efficient:
Do you want to finish your work faster but consume more power? (AMD) or vice versa (Intel)
 
Edit: Here is Trinity and Ivy Bridge in comparison
Trinity -> 4 Cores * 2 FP units * 4 32-bit * 2 FMA * ~4 GHz => ~256 GFlops
Ivy Bridge -> 4 Cores * 2 FP units * 8 32-bit * ~4 GHz => ~256 GFlops
^--- Design choices:
Does your workloads support FMA or your applications benefit from FMA? (AMD) or vice versa (Intel)
^^--- Remember both Steamroller and Haswell will support FMA. FMA3 Non-destructive and FMA4 Non-destructive.

I have bad fealing about this and I'm sorry if some people pass like with Bulldozer again. 

The CPUZ screenshot bugs me, why does it say "Zambezi", as if it is just a Bulldozer ES passed off as a Piledriver ES just to still up negative hype?  Afterall, anyone that hates on AMD is least likely to investigate the claims and just buys into it as if it was expected. 

lehpron


The CPUZ screenshot bugs me, why does it say "Zambezi", as if it is just a Bulldozer ES passed off as a Piledriver ES just to still up negative hype?  Afterall, anyone that hates on AMD is least likely to investigate the claims and just buys into it as if it was expected. 

Probably because it is.

I think normal people can't hate AMD. AMD is not terrorist or dictator or killer. But they currently hinder CPU progress. AMD or Intel or NVIDIA only crazy man can hate. If AMD make something wrong Intel not need to afraid because their CPU work on little higher temps.
Stronger AMD better price for customers both Intel and AMD. Situation on market is change always is good criticize when something is not write. How someone can stay blind in situation where if they make wonderfull CPU they will be on zero like Phenom X6. After 3 years.
Where is 3 year. Missing, freezing in time. 2% better but be better 2%. Progress is everything, if one stop they can do everything in future.

Not to mention those benchmarks are apperently from 2008 going by those screen shots.
Yea I dont think those are piledriver CPUs they are testing there. 

starsmine


Not to mention those benchmarks are apperently from 2008 going by those screen shots.
Yea I dont think those are piledriver CPUs they are testing there. 

 
Somebody is s**t stirring

lehpron
The CPUZ screenshot bugs me, why does it say "Zambezi", as if it is just a Bulldozer ES passed off as a Piledriver ES just to still up negative hype?  Afterall, anyone that hates on AMD is least likely to investigate the claims and just buys into it as if it was expected.

Piledriver is the fix of Bulldozer.  Unlike Agena -> Deneb AMD isn't going from one node to another node we are on the same node.  Orochi Rev B =/= Orochi Rev C but CPU-Z doesn't know the difference.
 
Orochi Rev B3 and C0 both started production in Q1 2012 based on that AMD should launch with C1 or C2 in early October.
 
Piledriver:
Up to 20% more memory throughput
Up to 20% more vector throughput
Up to 20% more scalar throughput
Up to 25% less power consumption
About a 10% increase in speed with interconnect
etc. (In total a 16-24% improvement around the board)
 
This particular engineer sample is the FX-8300 which will replace the FX-8100 in the OEM market.  3.3 GHz from 2.8 GHz is pretty good improvement.
http://www.coolaler.com/showthread.php/292963
^-- If I haven't posted it before here is the real source.
 


 
The two C0s

Rumors.
http://www.guru3d.com/news/amd-to-skip-piledriver-and-place-hopes-on-steamroller/

^seems like a good idea.

Well if they can get steamroller out before Haswell, Ill agree with that looking like a good idea. 

I'll remain an intel user, to be sure,  but I want AMD to be strong ...competition is good for all of us...
 

That's it than next almost one year for desktop... only what...???
i7-3970X in October. I say one year because I expect one Haswell delay for about 3 months. Because improvment, enough powerfull CPU on market... 
I'll remain Intel user too... only blame myself because didn't saw AMD way and few months before SB show up I buy X4 (later and second board for AM3) better I choose X58 3Way Classified E-760 and i7-950/960... I would compensate money difference two times from now.
But better ever than never and now I choose parts I follow i7 with K natural not extreme and E-ATX/ FTW.
 
 

That cinebench score is so disappointing.

Clo, more then likely its not piledriver, but the codename and date of the screen shot clue us into such a conclusion. 

http://www.obr-hardware.com/2012/08/amd-fx2-piledriver-is-here-last-amd.html  
 
What is your comment on this writing before few days.
People prepear easy to pay i7-3770k future brother like 3930K at least.

Oh screw that -.- are they not going to make server chips anymore either? I think if such is true that will be a huge mistake for AMD.

AMD Piledriver will be the last processor to use the AMx socket.  After that it will be EOL'ed and the next platform will either be FMx or GC3x.
 
FMx -> APUs, CPUs, Dual-channel
GC33? -> bAPUs, CPUs, Quad-channel, 2P Overclocking
GC36? -> sbAPUs, CPUs, Octo-channel, No overclocking
 
FCH -> Fusion Controller Hub
SCH -> Server Controller Hub(supposedly has an interconnect between other SCHs more SAS/more SATA/etc)
 
Hypertransport is moved to being only a CPU interconnect
4 Cache Coherent HT Links
 
New power management system >0w to <350w safe overclocking.  Turbo Core 4.0?
New ISEs
 
^stuff I can find about Steamroller
---