Overclock.net › Forums › Industry News › Hardware News › [ExtremeTech] Analyzing Bulldozer: Why AMD’s chip is so disappointing
New Posts  All Forums:Forum Nav:

[ExtremeTech] Analyzing Bulldozer: Why AMD’s chip is so disappointing

post #1 of 292
Thread Starter 
Quote:
AMD’s Bulldozer is finally here, after years of development — and its performance is significantly worse than anyone expected. The situation is ugly enough that it may explain why so many executives left AMD over the past twelve months, and why the company was so tight-lipped about their departure. Bulldozer’s general performance has been widely covered; our goal here is to drill into why the CPU performs the way it does rather than covering it in a wide range of real world scenarios.

The first thing to understand about Bulldozer is that it leverages aspects of simultaneous multi-threading to combine the functions of what would normally be two discrete cores into a single package (AMD refers to this combination as a “module“). Each module contains what Windows identifies as two cores, but combining instruction scheduling and CPU resources has an impact on CPU scaling in multi-threaded tests when compared to the same programs running on “traditional” multi-core processors.
Source

Discuss.
Edited by 2010rig - 10/24/11 at 10:13pm
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  
Reply
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  
Reply
post #2 of 292
flame thread..i tell ya why it's disappointing..cuz it's slower than..NetBurst Powahhh! ::seriously..when it comes to web-browsing..it is..and there's this other weird thing..you have to mouse-over the browser to keep it going..
 
Thuban Powah!
(14 items)
 
 
CPUMotherboardGraphicsRAM
Xeon X5650 Asus P6X58D-E Sapphire HD 7950 G.Skill Trident-X 2600 CL10 
Hard DriveCoolingPowerCase
Samsung 840 pro Bong cooler / RASA waterblock Rosewill Capstone Gold 750 None 
MouseMouse PadAudio
cheap! Newegg box panel ibeats with onboard. 
CPUMotherboardGraphicsRAM
AMD Phenom II X6 1075T w/2 cores locked @ 4.2+ M4a89GTD-Pro XFX Radeon HD 5870 8gb Samsung ddr3 1600 11-11-11-28@ 8-8-8-24 
Hard DriveHard DriveOptical DriveCooling
OCZ Agility2 40gb WD Blue 500GB Lite-On RASA waterblock 
OSMonitorKeyboardPower
Windows & Linux Samsung 1080p 2 ms Dell Thermaltake TR2-800 
CaseMouse
Cooler Master Elite 330 (was) cheap 
  hide details  
Reply
 
Thuban Powah!
(14 items)
 
 
CPUMotherboardGraphicsRAM
Xeon X5650 Asus P6X58D-E Sapphire HD 7950 G.Skill Trident-X 2600 CL10 
Hard DriveCoolingPowerCase
Samsung 840 pro Bong cooler / RASA waterblock Rosewill Capstone Gold 750 None 
MouseMouse PadAudio
cheap! Newegg box panel ibeats with onboard. 
CPUMotherboardGraphicsRAM
AMD Phenom II X6 1075T w/2 cores locked @ 4.2+ M4a89GTD-Pro XFX Radeon HD 5870 8gb Samsung ddr3 1600 11-11-11-28@ 8-8-8-24 
Hard DriveHard DriveOptical DriveCooling
OCZ Agility2 40gb WD Blue 500GB Lite-On RASA waterblock 
OSMonitorKeyboardPower
Windows & Linux Samsung 1080p 2 ms Dell Thermaltake TR2-800 
CaseMouse
Cooler Master Elite 330 (was) cheap 
  hide details  
Reply
post #3 of 292
Thread Starter 
Quote:
Originally Posted by Schmuckley View Post
flame thread..i tell ya why it's disappointing..cuz it's slower than..NetBurst Powahhh! ::seriously..when it comes to web-browsing..it is..and there's this other weird thing..you have to mouse-over the browser to keep it going..
It doesn't have to be, it's actually a pretty good article. They even tested 4M/4C



Quote:
We checked the impact of Windows 7′s scheduler by measuring CPU performance in Maxwell Render 1.7 and Cinebench 11.5. Both programs allow the user to define a specific number of threads (four, in our case). The 4M/4C label means that all eight cores are active, 4M/4C means that all four modules are active, with one core operating per module, and 2M/4C denotes a dual-module/quad-core configuration. Both of these tests show a 4M/4C arrangement outperforming a 4M/8C system by roughly eight percent when four threads are used. This suggests that scheduler inefficiencies could indeed be hurting Bulldozer’s general performance in workloads that can’t take advantage of all eight cores.
This part was alarming....



Quote:
Disabling Turbo mode lets us accurately measure the IPC hit Bulldozer takes compared to Thuban. An 11% decrease is downright ugly considering how much Thuban already lagged Sandy Bridge. Statistically, we’d need to push the FX-8150 to around 5.5GHz to match Sandy Bridge’s 3.4GHz performance in this test.

Edited by 2010rig - 10/24/11 at 7:17pm
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  
Reply
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  
Reply
post #4 of 292
So you get more performance by having 4M/4C rather then full CPU 4M/8C?
Ishimura
(21 items)
 
Silent Knight
(13 items)
 
 
CPUMotherboardGraphicsRAM
Intel Core i7 3770K @ 4.6GHz ASRock Z77E-ITX eVGA GTX 1080 Ti Hybrid AMD Radeon R9 16GB DDR3-2400MHz  
Hard DriveHard DriveCoolingCooling
SanDisk Ultra II 960GB Toshiba X300 5TB Corsair H100i GTX eVGA Hybrid Water Cooler  
CoolingOSMonitorKeyboard
4x GentleTyphoon AP-15 Windows 10 Pro 64-Bit Philips Brilliance BDM4065UC 4K Razer BlackWidow Chroma  
PowerCaseMouseMouse Pad
eVGA SuperNOVA 750 G3 Define Nano S Logitech G502 Proteus Core PECHAM Gaming Mouse Pad XX-Large 
AudioAudioAudioAudio
Audioengine D1 DAC Mackie CR Series CR3 Audio-Technica ATH-M50 Sennheiser HD 598 
Audio
Sony XB950BT 
CPUMotherboardGraphicsRAM
AMD Phenom II X4 955 @ 4.2GHz ASUS M4A79XTD EVO AMD Radeon HD 7970 3GB @ 1200/1500 2x 4GB G.SKILL Ripjaws X DDR3-1600 
Hard DriveHard DriveHard DriveCooling
OCZ Agility 3 60GB WD Caviar Green 1.5TB 2 x Seagate Barracuda 2TB XSPC Raystorm 
CoolingCoolingOSPower
EK-FC7970 XSPC RS360 Windows 10 Pro 64-Bit Corsair TX750 
Case
NZXT Switch 810  
  hide details  
Reply
Ishimura
(21 items)
 
Silent Knight
(13 items)
 
 
CPUMotherboardGraphicsRAM
Intel Core i7 3770K @ 4.6GHz ASRock Z77E-ITX eVGA GTX 1080 Ti Hybrid AMD Radeon R9 16GB DDR3-2400MHz  
Hard DriveHard DriveCoolingCooling
SanDisk Ultra II 960GB Toshiba X300 5TB Corsair H100i GTX eVGA Hybrid Water Cooler  
CoolingOSMonitorKeyboard
4x GentleTyphoon AP-15 Windows 10 Pro 64-Bit Philips Brilliance BDM4065UC 4K Razer BlackWidow Chroma  
PowerCaseMouseMouse Pad
eVGA SuperNOVA 750 G3 Define Nano S Logitech G502 Proteus Core PECHAM Gaming Mouse Pad XX-Large 
AudioAudioAudioAudio
Audioengine D1 DAC Mackie CR Series CR3 Audio-Technica ATH-M50 Sennheiser HD 598 
Audio
Sony XB950BT 
CPUMotherboardGraphicsRAM
AMD Phenom II X4 955 @ 4.2GHz ASUS M4A79XTD EVO AMD Radeon HD 7970 3GB @ 1200/1500 2x 4GB G.SKILL Ripjaws X DDR3-1600 
Hard DriveHard DriveHard DriveCooling
OCZ Agility 3 60GB WD Caviar Green 1.5TB 2 x Seagate Barracuda 2TB XSPC Raystorm 
CoolingCoolingOSPower
EK-FC7970 XSPC RS360 Windows 10 Pro 64-Bit Corsair TX750 
Case
NZXT Switch 810  
  hide details  
Reply
post #5 of 292
Something straight up broke with BD, hard to say if it is software or the chip itself though. I think I'm leaning towards some wrong with chip...
Sig Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX 8350 ASUS Crosshair IV Formula Powercolor 7870 16GB G.SKILL Sniper 
Hard DriveOSMonitorPower
Plextor PX-M3S SSD Windows 7 Professional 64-bit 23" Acer 1920x1080 1000 Watt Ultra X3 
Case
Zalman GS1000 
  hide details  
Reply
Sig Rig
(13 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX 8350 ASUS Crosshair IV Formula Powercolor 7870 16GB G.SKILL Sniper 
Hard DriveOSMonitorPower
Plextor PX-M3S SSD Windows 7 Professional 64-bit 23" Acer 1920x1080 1000 Watt Ultra X3 
Case
Zalman GS1000 
  hide details  
Reply
post #6 of 292
Thread Starter 
Quote:
Originally Posted by ZealotKi11er View Post
So you get more performance by having 4M/4C rather then full CPU 4M/8C?
Basically.

I believe AMD needs to figure out an approach where apps that use 4 or less cores, Bulldozer will operate in the 4M / 8C mode. And in Multi-Threaded apps switch to 4M / 8C for those rendering sessions.

If they can figure this out on the fly, they'd be golden. ( Yeah, that's easier said than done. )

Or maybe people just need to do this manually, if you know you're rendering, then enable all 8 cores, otherwise, just run it 4M/4C. Either way, they need to improve IPC.
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  
Reply
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  
Reply
post #7 of 292
Quote:
Originally Posted by 2010rig View Post
Basically.

I believe AMD needs to figure out an approach where apps that use 4 or less cores, Bulldozer will operate in the 4M / 8C mode. And in Multi-Threaded apps switch to 4M / 8C for those rendering sessions.

If they can figure this out on the fly, they'd be golden. ( Yeah, that's easier said than done. )

Or maybe people just need to do this manually, if you know you're rendering, then enable all 8 cores, otherwise, just run it 4M/4C. Either way, they need to improve IPC.
I'm sure it could be done via a within-OS BIOS modifier. Set up profiles, one for 4M/8C, one for 4M/4C, then have those profiles attached to programs that use one more efficiently over the other.
T3k
(13 items)
 
The Tibaldi
(9 items)
 
WooJoo
(13 items)
 
CPUMotherboardGraphicsGraphics
Intel i5 3210 h77 HD4000 GeForce 640 LE 
RAMHard DriveOSOS
4gb of something 5400rpm slowness 13.04 Ubuntu x64  Windows 8 Pro 
Monitor
1920x1080 
CPUMotherboardGraphicsRAM
965 BE 4.0GHZ (250 x 15.5) @1.5V m2n32 sli deluxe - 2000mhz @ 1.3V Asus TOP 5850 1GB 950/1200 Kingston HyperX 5-5-5-15 
Hard DriveOptical DriveOSMonitor
WD Velociraptor 150GB, WD raptor 74gb TDK 880N DVDRW Windows 7 x64 Ultimate Westinghouse 22" lcd 
KeyboardPowerCaseMouse
saitek eclipse II Corsair 620W modular Cooler Master ATCS 840 black emprex 
  hide details  
Reply
T3k
(13 items)
 
The Tibaldi
(9 items)
 
WooJoo
(13 items)
 
CPUMotherboardGraphicsGraphics
Intel i5 3210 h77 HD4000 GeForce 640 LE 
RAMHard DriveOSOS
4gb of something 5400rpm slowness 13.04 Ubuntu x64  Windows 8 Pro 
Monitor
1920x1080 
CPUMotherboardGraphicsRAM
965 BE 4.0GHZ (250 x 15.5) @1.5V m2n32 sli deluxe - 2000mhz @ 1.3V Asus TOP 5850 1GB 950/1200 Kingston HyperX 5-5-5-15 
Hard DriveOptical DriveOSMonitor
WD Velociraptor 150GB, WD raptor 74gb TDK 880N DVDRW Windows 7 x64 Ultimate Westinghouse 22" lcd 
KeyboardPowerCaseMouse
saitek eclipse II Corsair 620W modular Cooler Master ATCS 840 black emprex 
  hide details  
Reply
post #8 of 292
Quote:
Originally Posted by rusky1 View Post
I'm sure it could be done via a within-OS BIOS modifier. Set up profiles, one for 4M/8C, one for 4M/4C, then have those profiles attached to programs that use one more efficiently over the other.
Interesting thought you have there.

All I know is I want BD to at least be a better dollar/performance for my games. If it could at least do that I might pick it up for the fun of it. I think Microsoft and AMD need a long chat personally.

Oh, anyone know what happened to the supposed temporary fix that was supposed to get BD a 40%-70% performance increase? I read that somewhere but forget where. Something about having to enable the patch every time you shutdown the computer.
Sovereign
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i5 3570K ASRock Z77 Pro4 EVGA GTX 470 Kingston 4x4GB DDR3 1600 
Hard DriveOptical DriveOSMonitor
128 GB SSD LITE-ON Black 24X DVD Writer Windows 7 PRO x64 ASUS 1920x1080 LED 23.6" 
KeyboardPowerCaseMouse
Logitech USB Media Keyboard CORSAIR CMPSU-750TX 750W Antec Nine Hundred Two Logitech 
Audio
Turtlebeach Earforce P11 
  hide details  
Reply
Sovereign
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i5 3570K ASRock Z77 Pro4 EVGA GTX 470 Kingston 4x4GB DDR3 1600 
Hard DriveOptical DriveOSMonitor
128 GB SSD LITE-ON Black 24X DVD Writer Windows 7 PRO x64 ASUS 1920x1080 LED 23.6" 
KeyboardPowerCaseMouse
Logitech USB Media Keyboard CORSAIR CMPSU-750TX 750W Antec Nine Hundred Two Logitech 
Audio
Turtlebeach Earforce P11 
  hide details  
Reply
post #9 of 292


I wonder what the 2600k set up was. My 2600k gets 7's in Cinebench at stock speeds. Most bulldozer benches show 2600k speeds way below what they should be... it's like they're gimping them to make a crappy chip look slightly better.
The SandBox
(14 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7-2600k (5.3GHz Max) ASUS Maximus IV GENE-Z GTX 980 Ti SC+ BP 16GB G.SKILL Sniper 1600MHz 1.25v 4x4GB 
Hard DriveCoolingOSMonitor
Samsung 850 Pro 256 Corsair H80 Windows 7 Ultimate 64bit Asus PA279Q 
PowerCase
SeaSonic X Series X650 Gold Silverstone FT03B 
  hide details  
Reply
The SandBox
(14 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7-2600k (5.3GHz Max) ASUS Maximus IV GENE-Z GTX 980 Ti SC+ BP 16GB G.SKILL Sniper 1600MHz 1.25v 4x4GB 
Hard DriveCoolingOSMonitor
Samsung 850 Pro 256 Corsair H80 Windows 7 Ultimate 64bit Asus PA279Q 
PowerCase
SeaSonic X Series X650 Gold Silverstone FT03B 
  hide details  
Reply
post #10 of 292
Interesting that 4M/4C are better, although to be honest I am not surprised.

AMD sure does have it's work cut out for it by the time they release Piledriver, but hey if Intel can surpass AMD then AMD can do the same. It just might take a little longer than what people originally thought.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
Overclock.net › Forums › Industry News › Hardware News › [ExtremeTech] Analyzing Bulldozer: Why AMD’s chip is so disappointing