Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [Chiphell]Bulldozer Sample Benchmark Leaks
New Posts  All Forums:Forum Nav:

[Chiphell]Bulldozer Sample Benchmark Leaks - Page 6  

post #51 of 285
I don't believe the cinebench results. If that 4.6 result is to be believed, BD's per-core performance has taken a massive drop VS PII. Assuming these results are legit, maybe BD will be like P4 and run at super high clocks to make up for it, or maybe these results are due to the chip being an engineering sample and having some sort of bug.
Edited by qwertymac93 - 6/9/11 at 2:56pm
z87
(18 items)
 
Media Rig
(18 items)
 
 
CPUMotherboardGraphicsRAM
AMD Phenom II 720 ASRock 870 Extreme 3 HIS 5750 (840core/1200mem) 2*2GB A-Data DDR3 1600 
Hard DriveHard DriveOptical DriveCooling
Seagate Barracuda ST1000DM003 Seagate Barracuda ST3000DM001 Lite-On Blue Ray WH14NS40 Arctic Cooling Freezer 64 PRO 
OSMonitorKeyboardPower
Windows 10 pro 64-bit Vizio M43-c1 Logitech K400 Corsair CX-430 
CaseMouseAudioAudio
NZXT Beta EVO Gigabyte M7700 Sony STR-DN1070 Definitive Technology BP8 Tower Speakers 
AudioAudio
Definitive Technology 2002 Series Center Speaker Definitive Technology ProMonitor 1000 Bookshelf... 
  hide details  
z87
(18 items)
 
Media Rig
(18 items)
 
 
CPUMotherboardGraphicsRAM
AMD Phenom II 720 ASRock 870 Extreme 3 HIS 5750 (840core/1200mem) 2*2GB A-Data DDR3 1600 
Hard DriveHard DriveOptical DriveCooling
Seagate Barracuda ST1000DM003 Seagate Barracuda ST3000DM001 Lite-On Blue Ray WH14NS40 Arctic Cooling Freezer 64 PRO 
OSMonitorKeyboardPower
Windows 10 pro 64-bit Vizio M43-c1 Logitech K400 Corsair CX-430 
CaseMouseAudioAudio
NZXT Beta EVO Gigabyte M7700 Sony STR-DN1070 Definitive Technology BP8 Tower Speakers 
AudioAudio
Definitive Technology 2002 Series Center Speaker Definitive Technology ProMonitor 1000 Bookshelf... 
  hide details  
post #52 of 285
Quote:
Originally Posted by qwertymac93 View Post
I don't believe the cinebench results. my AII scores ~3.4 @3ghz, if that 4.6 result is to be believed, BD's per-core performance has taken a massive drop VS PII. maybe it will be like P4 and run at super high clocks to make up for it, OR maybe these results are incorrect.
Maybe Cinebench doesn't play well with that "modules instead of cores" thingy? Could it be a reason?

I don't know, but yeah, it does look horrible.
Son of a Bridge
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @ 4.2ghz | 1.24v Asus P8P67 EVO [B3 Revision] Sapphire HD6870 1GB GDDR5 @ 1ghz Core 4GB Kingston HyperX DDR3 1600mhz 
Hard DriveOptical DriveOSMonitor
Hitachi Deskstar P7K500 500gb 16MB Cache SATA II SATA LG DVD burner Windows 7 Ultimate x64 Acer LCD 17" 1280x1024 
PowerCaseMouseMouse Pad
Corsair TX650 Modded regular case X7 X-760H 2000 DPI N@NO 
  hide details  
Son of a Bridge
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @ 4.2ghz | 1.24v Asus P8P67 EVO [B3 Revision] Sapphire HD6870 1GB GDDR5 @ 1ghz Core 4GB Kingston HyperX DDR3 1600mhz 
Hard DriveOptical DriveOSMonitor
Hitachi Deskstar P7K500 500gb 16MB Cache SATA II SATA LG DVD burner Windows 7 Ultimate x64 Acer LCD 17" 1280x1024 
PowerCaseMouseMouse Pad
Corsair TX650 Modded regular case X7 X-760H 2000 DPI N@NO 
  hide details  
post #53 of 285
They're probably accurate, unfortunately for AMD.
I didn't expect them to make a 3 generation leap (The Phenom II line is slower than the Yorkfields which came in 2008) and Intel went from there to Nehalem (way faster) to Westmere (again, way faster) and to Sandy Bridge (much faster yet).
Matching the SB architecture would have been a feat unto itself.
Silent PC
(15 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II X4 955 Gigabyte MA790X-UD4P EVGA GTX 460 1GB 6GB DDR2 @ 675mhz 
Hard DriveOptical DriveCoolingOS
WD Caviar Black 640GB+ WD Caviar 250GB LG DVD Burner Cooler Master Hyper 212+ EVO Windows 7 Professional x64 
MonitorKeyboardPowerCase
Westinghouse 22 inch ABS M1 Antec Earthwatts 500 Antec Designer 500 
MouseAudio
Logitech G5 v2 Asus XONAR DX with JVC HA-RX 700 
  hide details  
Silent PC
(15 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II X4 955 Gigabyte MA790X-UD4P EVGA GTX 460 1GB 6GB DDR2 @ 675mhz 
Hard DriveOptical DriveCoolingOS
WD Caviar Black 640GB+ WD Caviar 250GB LG DVD Burner Cooler Master Hyper 212+ EVO Windows 7 Professional x64 
MonitorKeyboardPowerCase
Westinghouse 22 inch ABS M1 Antec Earthwatts 500 Antec Designer 500 
MouseAudio
Logitech G5 v2 Asus XONAR DX with JVC HA-RX 700 
  hide details  
post #54 of 285
Quote:
Originally Posted by MacA View Post
This is not B0, simple as that. If that was B0, do you think AMD would wait until now to say "hey, the perfs suck, they suck so bad they're not even better than Deneb, let's delay BD!".

But you are right, a stepping is not magic, it's the same architecture slightly revised.
Er.. you do realize that's exactly what happened, right? Pretty much all the tech sites said mobo manufacturer sources told them B0 was underperforming, and that the new stepping fixed/would fix them.

(Example, from Tom's 990fx article I mentioned before) "Processors based on the B0 stepping are in the hands of motherboard manufacturers. But they’re all telling me that performance is nowhere near what they were expecting, and it’ll take another stepping to fix them."

The *rumors* were that the different cores weren't clocking up correctly under load. Whether that's true or not, I don't know. If half of the 8 cores aren't making it to 2.8ghz that would explain the performance problem. This is something that could (reasonably) be fixed with a new stepping.

Quote:
Originally Posted by MacA View Post
Real, maybe, but in no way representative of BD's final specs. AMD will not release something similar to Deneb after 5 years of dev! It's just obvious!
Yeah, I agree. Thinking this will be the production performance is questionable, indeed.
Why upgrade?
(14 items)
 
Why overclock?!
(15 items)
 
 
CPUMotherboardGraphicsRAM
Core i5 760 ASUS 55i Sabertooth EVGA 960 4GB 8gb 1600mhz Vengeance 
Hard DriveHard DriveOSMonitor
120gb Intel 320 (OS) 500gb Samsung 850 Evo Win 7 64 Viewsonic VX2250-Wm 
PowerCaseMouseAudio
Seasonic G 550w Gold Lancool PC-K9B Logitech G5 X-Fi Titanium/AT-H700 
CPUMotherboardGraphicsRAM
Core i5 4690k Asus Z97 Sabertooth mk2 EVGA 970 8gb Kingston 1833mhz 
Hard DriveHard DriveCoolingOS
128GB Samsung 830 500GB Samsung 840 Evo allota fans but quiet! Win 7 64 
MonitorKeyboardPowerCase
Dell U2311H Rev. A01 Mechanical w/ Cherry Blues NZXT Hale 650w NZXT Phantom - white 
MouseMouse PadAudio
Anker CG100 Steelseries 9HD Sennheiser HD555 
  hide details  
Why upgrade?
(14 items)
 
Why overclock?!
(15 items)
 
 
CPUMotherboardGraphicsRAM
Core i5 760 ASUS 55i Sabertooth EVGA 960 4GB 8gb 1600mhz Vengeance 
Hard DriveHard DriveOSMonitor
120gb Intel 320 (OS) 500gb Samsung 850 Evo Win 7 64 Viewsonic VX2250-Wm 
PowerCaseMouseAudio
Seasonic G 550w Gold Lancool PC-K9B Logitech G5 X-Fi Titanium/AT-H700 
CPUMotherboardGraphicsRAM
Core i5 4690k Asus Z97 Sabertooth mk2 EVGA 970 8gb Kingston 1833mhz 
Hard DriveHard DriveCoolingOS
128GB Samsung 830 500GB Samsung 840 Evo allota fans but quiet! Win 7 64 
MonitorKeyboardPowerCase
Dell U2311H Rev. A01 Mechanical w/ Cherry Blues NZXT Hale 650w NZXT Phantom - white 
MouseMouse PadAudio
Anker CG100 Steelseries 9HD Sennheiser HD555 
  hide details  
post #55 of 285
Quote:
Originally Posted by PanicProne View Post
Maybe Cinebench doesn't play well with that "modules instead of cores" thingy? Could it be a reason?

I don't know, but yeah, it does look horrible.
Cinebench wouldn't even see "modules", all it knows is that it has 8 cores to work with.
post #56 of 285
Quote:
Originally Posted by Iceman23 View Post
Cinebench wouldn't even see "modules", all it knows is that it has 8 cores to work with.
It definitely doesn't look good for AMD, then.
Son of a Bridge
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @ 4.2ghz | 1.24v Asus P8P67 EVO [B3 Revision] Sapphire HD6870 1GB GDDR5 @ 1ghz Core 4GB Kingston HyperX DDR3 1600mhz 
Hard DriveOptical DriveOSMonitor
Hitachi Deskstar P7K500 500gb 16MB Cache SATA II SATA LG DVD burner Windows 7 Ultimate x64 Acer LCD 17" 1280x1024 
PowerCaseMouseMouse Pad
Corsair TX650 Modded regular case X7 X-760H 2000 DPI N@NO 
  hide details  
Son of a Bridge
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @ 4.2ghz | 1.24v Asus P8P67 EVO [B3 Revision] Sapphire HD6870 1GB GDDR5 @ 1ghz Core 4GB Kingston HyperX DDR3 1600mhz 
Hard DriveOptical DriveOSMonitor
Hitachi Deskstar P7K500 500gb 16MB Cache SATA II SATA LG DVD burner Windows 7 Ultimate x64 Acer LCD 17" 1280x1024 
PowerCaseMouseMouse Pad
Corsair TX650 Modded regular case X7 X-760H 2000 DPI N@NO 
  hide details  
post #57 of 285
Quote:
Originally Posted by awaizy View Post
They're probably accurate, unfortunately for AMD.
I didn't expect them to make a 3 generation leap (The Phenom II line is slower than the Yorkfields which came in 2008) and Intel went from there to Nehalem (way faster) to Westmere (again, way faster) and to Sandy Bridge (much faster yet).
Matching the SB architecture would have been a feat unto itself.
So much this.

I've had this on my mind too.

BD has been in development for 5+ years now. It was supposed to be out in 2009 originally. If it was released on time, it would've came out to compete with 1st Gen core i7 series, and by the looks of it, it still would've been slower, but they could've improved from there.

They skipped 45nm BD and went straight to 32nm.

Now that SB is out, the gap seems that much wider.

People make fun of Intel's Tock - Tick process, but it's been working out well in increasing performance, and they keep making strides forward.
2010rig
(14 items)
 
Galaxy S3
(8 items)
 
 
CPUMotherboardGraphicsRAM
X5660 @ 4.5  ASUS P6X58D-E 980TI? 12GB OCZ Platinum - 7-7-7-21 
Hard DriveCoolingOSMonitor
1 80GB SSD x25m - 3TB F3 + F4 NH-D14 Windows 7 Ultimate LG 47LH55 
KeyboardPowerCaseMouse
Natural Wireless Keyboard Corsair 750HX CM 690 II Advanced MX 518 
CPUGraphicsRAMHard Drive
Snapdragon S4 Dual core 1500mhz Adreno 225 Samsung 2GB 16GB Onboard Flash 
OSMonitorPowerCase
Android 4.4.2 - CM11 4.8" AMOLED 1280x720 2100 mAh battery Otterbox Defender 
  hide details  
2010rig
(14 items)
 
Galaxy S3
(8 items)
 
 
CPUMotherboardGraphicsRAM
X5660 @ 4.5  ASUS P6X58D-E 980TI? 12GB OCZ Platinum - 7-7-7-21 
Hard DriveCoolingOSMonitor
1 80GB SSD x25m - 3TB F3 + F4 NH-D14 Windows 7 Ultimate LG 47LH55 
KeyboardPowerCaseMouse
Natural Wireless Keyboard Corsair 750HX CM 690 II Advanced MX 518 
CPUGraphicsRAMHard Drive
Snapdragon S4 Dual core 1500mhz Adreno 225 Samsung 2GB 16GB Onboard Flash 
OSMonitorPowerCase
Android 4.4.2 - CM11 4.8" AMOLED 1280x720 2100 mAh battery Otterbox Defender 
  hide details  
post #58 of 285
Quote:
Originally Posted by pursuinginsanity View Post
Er.. you do realize that's exactly what happened, right? Pretty much all the tech sites said mobo manufacturer sources told them B0 was underperforming, and that the new stepping fixed/would fix them.
Yeah sure, but the problem in the mobo manufacturers rumors case seems to be caused by some bug. Not something directly architecture related.
However in this case, BD is underperforming but not buggy, right? and that means the problem is architecture related, and I doubt it's the case when AMD has been working on it for 5 years.

Basicallt according to chiphell, AMD will release after 5 years something comparable to Deneb, and that is very unlikely.

If AMD could release just one benchmark I'd be happy. I'd be able to know if I can get a 2600K now or just wait for BD. But no nothing.
post #59 of 285
***Translated from one of the Forum Post!***
(poor English translation, see what you can make of it)

Price according to AMD official statement, I personally predict performance in the SNB under the bulldozer. Price positioning performance can often be seen.

We reviewed before release under the HD6970 is not enthusiasm and performance in the CHH guess to know, out of the final price of a basic performance evaluation without looking at the performance orientation can guess, because AMD is also a businessman. So I guess now, bulldozers are close to this location.

ES CPU test is currently estimated that the wheels will encounter a lot of trouble, is not very obvious defects, AMD will not be re-amended. The longer delay time for the market, AMD lost on the more.

------------------------------------

I'm not believing this chip can be that horrible, not for one second...

Then there is the question about what this forum poster is referring to, which is obviously a defect, and a large one, which is reported not to be fixed, and most likely will not be detected..

Make sense to me, AMD Put it off so they can make money on their other chips, then say it's not going to be released, and revise it, or not revise it, release it like Sandy Bridge, and then what may come, will come...

Tough to cypher through the broken english on this forum, but it appears everyone believes this test to be real and true... (though sad results)

I think what everyone should keep in mind is, this is a sample, any most likely will NOT be the finished product, but if it is, ouch, this would hurt AMD Pretty bad..

They were all big about it does 11, well this is looking like a 6, whoosh!
The Rock
(15 items)
 
  
The Rock
(15 items)
 
  
post #60 of 285
Quote:
Originally Posted by 2010rig View Post
So much this.

I've had this on my mind too.

BD has been in development for 5+ years now. It was supposed to be out in 2009 originally. If it was released on time, it would've came out to compete with 1st Gen core i7 series, and by the looks of it, it still would've been slower, but they could've improved from there.

They skipped 45nm BD and went straight to 32nm.

Now that SB is out, the gap seems that much wider.

People make fun of Intel's Tock - Tick process, but it's been working out well in increasing performance, and they keep making strides forward.
I never saw anyone mocking Intel's tick-tock schedule, but if someone actually did, then the joke is on them.

Intel can actually deliver great IPC increase with their Die shrinks, somewhere in the 10%-15% range (which is amazing, considering they are just improvements to current architectures).

Good thing LGA 1155 will have Ivy Bridge first, I might jump on that even though my 2600k won't be anywhere near obsolete in the next few years
Son of a Bridge
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @ 4.2ghz | 1.24v Asus P8P67 EVO [B3 Revision] Sapphire HD6870 1GB GDDR5 @ 1ghz Core 4GB Kingston HyperX DDR3 1600mhz 
Hard DriveOptical DriveOSMonitor
Hitachi Deskstar P7K500 500gb 16MB Cache SATA II SATA LG DVD burner Windows 7 Ultimate x64 Acer LCD 17" 1280x1024 
PowerCaseMouseMouse Pad
Corsair TX650 Modded regular case X7 X-760H 2000 DPI N@NO 
  hide details  
Son of a Bridge
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @ 4.2ghz | 1.24v Asus P8P67 EVO [B3 Revision] Sapphire HD6870 1GB GDDR5 @ 1ghz Core 4GB Kingston HyperX DDR3 1600mhz 
Hard DriveOptical DriveOSMonitor
Hitachi Deskstar P7K500 500gb 16MB Cache SATA II SATA LG DVD burner Windows 7 Ultimate x64 Acer LCD 17" 1280x1024 
PowerCaseMouseMouse Pad
Corsair TX650 Modded regular case X7 X-760H 2000 DPI N@NO 
  hide details  
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Rumors and Unconfirmed Articles
This thread is locked  
Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [Chiphell]Bulldozer Sample Benchmark Leaks