Overclock.net › Forums › Industry News › Hardware News › [xbitlabs] AMD: We Are On Track With Steamroller Micro-Architecture in 2013.
New Posts  All Forums:Forum Nav:

[xbitlabs] AMD: We Are On Track With Steamroller Micro-Architecture in 2013. - Page 35  

post #341 of 718
Quote:
Originally Posted by sdlvx View Post

Why are you comparing clock to clock on an architecture that's designed for high IPC and low frequency to low IPC high frequency? Does power7 blow because it needs high clockspeed too?

Clock speed doesn't matter. Marketing has been fooling people into think it matters when comparing architectures even before Pentium 4 Netburst. It pains me to have to have lived to see hardcore Intel guys go "AMD sucks they can't hit high frequency" to "AMD blows they need higher frequency" in a matter of about 10 years. There's always been level headed people on Intel and AMD teams that don't buy the BS, but there's always someone running around talking about how IPC or clockspeed or whatever is the holy grail of total performance of a CPU.

Hell, if AMD released Steamroller and it had a 5% IPC decrease, and a 25% clock increase, it'd still be a great upgrade in total performance provided power consumption didn't go wild.
He's comparing stock vs stock with turbo disabled. It's a somewhat valid comparison, just insignificant to this community.

There's a minority of synthetic benchmarks that show PhII and Piledriver being evenly matched.
Money
(17 items)
 
Z575
(9 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 7 1700 @ 3.8Ghz MSI B350 Pro Carbon GIGABYTE R9 390 8GB 1100/1650Mhz 16GB Team DDR4-3000 
Hard DriveHard DriveHard DriveCooling
PNY CS1311 240GB - OS 2 x 2TB WD Black Raid 0 Synology DS416slim 4 x 1TB (raid 5) Corsair H100 
OSOSMonitorKeyboard
Windows 10 Pro ElementaryOS Loki LG 29UM67P 29" 2560x1080 FreeSync HPE 87 w/ Browns + Ducky PBT caps 
PowerCaseMouseAudio
EVGA 600W B2 Lian Li PCA05N Steelseries Rival 100 FiiO G10K + Philips SN9500HiFi 
CPUMotherboardGraphicsRAM
AMD A6-3420M @ 2.2Ghz 1.1v Lenovo Torpedo A55 6720g2 8GB Kingston 1600 CL9 
Hard DriveOSMouse
Samsung 840 250GB ElementaryOS x86_64 Abyssus 3.5G 
  hide details  
Money
(17 items)
 
Z575
(9 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 7 1700 @ 3.8Ghz MSI B350 Pro Carbon GIGABYTE R9 390 8GB 1100/1650Mhz 16GB Team DDR4-3000 
Hard DriveHard DriveHard DriveCooling
PNY CS1311 240GB - OS 2 x 2TB WD Black Raid 0 Synology DS416slim 4 x 1TB (raid 5) Corsair H100 
OSOSMonitorKeyboard
Windows 10 Pro ElementaryOS Loki LG 29UM67P 29" 2560x1080 FreeSync HPE 87 w/ Browns + Ducky PBT caps 
PowerCaseMouseAudio
EVGA 600W B2 Lian Li PCA05N Steelseries Rival 100 FiiO G10K + Philips SN9500HiFi 
CPUMotherboardGraphicsRAM
AMD A6-3420M @ 2.2Ghz 1.1v Lenovo Torpedo A55 6720g2 8GB Kingston 1600 CL9 
Hard DriveOSMouse
Samsung 840 250GB ElementaryOS x86_64 Abyssus 3.5G 
  hide details  
post #342 of 718
@Brutuz I want to bring up a question for you to ponder.

If PD @ 4GHZ is matching an i5-750 @ 2.6 GHZ

Question: How high would PD need to clock just to match the IPC of the I5-750 when it's @ 4.0, a 53% overclock?

Keep in mind the average OC for PD is around 4.8 on air ( which I predicted way before PD released remember? ), which is a mere 20% overclock.

Answer: PD would need to clock to 6 GHZ just to match the i5-750's IPC @ 4GHZ, something PD isn't capable of on water, let alone air.

Let's bring Ivy Bridge @ 4.5 into the equation, how high would PD need to clock to match its IPC? Ponder that for a bit, actually, don't, since you can't tell the difference between the 2. tongue.gif

My point is, the 8350 just doesn't have as much overclocking headroom compared to Intel chips in general.

It's not like it can clock to 5.5 nor did it match Nehalem IPC, like you once "predicted", funny how back then you WERE discussing IPC & single thread...
Quote:
Originally Posted by Brutuz View Post


Xbit saying how BD apparently has a lead over 1155 keeps me a bit optimistic, as well as common sense and what chew* was saying about OCing, if my original expectations of Nehalem IPC are met, the 5.5Ghz on water mark would put an FX-8150 right at SB level for single threads, but far faster for 8 threads on integer stuff.

Cinebench is something that matters, especially compared to SuperPi and wPrime...It's actually realistic.

Also, what was the original point in this whole discussion? The mere fact that PD is matching the IPC of a 45nm chip from 2009 is not impressive by any means, since it needs to be clocked 1.3GHZ higher in order to do so, and the 8350 only has 20% - 25% overclocking headroom beyond that 4 GHZ stock clock.

If PD clocked to 6 GHZ+ easily, you may have an argument I could get on board with, but due to the limited overclocking headroom, I can not and will not. wink.gif

My thoughts on Steamroller are completely different, and I'm beyond optimistic on that front, but wanted to clarify why the clock for clock comparisons are important. I'm sure a lot of you will disagree, I'll get back to you tomorrow.
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 #343 of 718
Quote:
Originally Posted by 2010rig View Post

I'm comparing Single Thread performance, clock for clock. Shouldn't be that hard to grasp.
You got proof Piledriver is NOT behind Phenom II in single threaded performance at the same clocks?

Start by explaining this.


I think that 1100t is probably overclocked in that result and the 8320 is not... this seems a little more reasonable:

*deleted image*

Notice the 8320 has the same score, but the 1100t is much, much different than yours for the cinebench single thread.

3300 score / 3.3ghz = 1000 score per 1ghz so if the 1100t was at 4ghz it would get a score around 4000, which you showed in your benchmark.

lol, looked at other benchmarks, cpuboss is on crack, where they got that 1100t result I don't know... maybe they disabled turbo core or something.

I still think BD was a good switch, with one refresh (PD) they have managed to get more or less the same ipc as an almost fully tweaked phenom 2, doing so with less power per core, and the ability to overclock much higher (which probably has to do a lot with the decrease in die size rather than architecture lol, but still). Sometimes you gotta take a few steps back to take a bunch more forward. I think AMD's bulldozer, piledriver, steamroller etc.. should have been modeled after a golf swing lol, bulldozer was the backswing biggrin.gif.
Edited by Ultracarpet - 4/4/13 at 12:53am
The Second Coming
(12 items)
 
   
CPUMotherboardGraphicsRAM
AMD Ryzen™ 1700 Asrock Fatal1ty x370 Gaming K4 Zotac mini GTX 1080ti TridentZ 16GB (2x8GB) DDR4 
Hard DriveHard DriveOSMonitor
256GB M.2 SSD 1TB HDD Windows 10 Pro Acer XG270HU 
KeyboardPowerCaseMouse
Logitech G15 750W Thermaltake Toughpower Gold Rated Thermaltake View 27 w/ riser Logitech G40 
CPUMotherboardGraphicsRAM
i7-4790k MSI Z97M Gaming Gtx 1070 Gskill 8gig 1600mhz (2x4) @2133 mhz 
Hard DriveCoolingOSKeyboard
1TB WD Caviar Black Corsair H75 Windows 8.1 Logitech G15 
PowerCaseMouseAudio
Thermaltake Toughpower 750w Gold  Fractal Design Define Mini Logitech G400 Sennheiser PC 350 (hero mod) 
CPUMotherboardGraphicsRAM
fx-8350 @5ghz 990fxa-ud3 rev 1.1 PowerColor 290x @1150/1500mhz Gskill 8gig 1600mhz (2x4) @2033 mhz 
Hard DriveCoolingOSPower
1TB WD Caviar Black rs360 raystorm kit Windows 8 Pro 700 Watt Corsair 
CaseMouse
HAF 912  Logitech G400 
  hide details  
The Second Coming
(12 items)
 
   
CPUMotherboardGraphicsRAM
AMD Ryzen™ 1700 Asrock Fatal1ty x370 Gaming K4 Zotac mini GTX 1080ti TridentZ 16GB (2x8GB) DDR4 
Hard DriveHard DriveOSMonitor
256GB M.2 SSD 1TB HDD Windows 10 Pro Acer XG270HU 
KeyboardPowerCaseMouse
Logitech G15 750W Thermaltake Toughpower Gold Rated Thermaltake View 27 w/ riser Logitech G40 
CPUMotherboardGraphicsRAM
i7-4790k MSI Z97M Gaming Gtx 1070 Gskill 8gig 1600mhz (2x4) @2133 mhz 
Hard DriveCoolingOSKeyboard
1TB WD Caviar Black Corsair H75 Windows 8.1 Logitech G15 
PowerCaseMouseAudio
Thermaltake Toughpower 750w Gold  Fractal Design Define Mini Logitech G400 Sennheiser PC 350 (hero mod) 
CPUMotherboardGraphicsRAM
fx-8350 @5ghz 990fxa-ud3 rev 1.1 PowerColor 290x @1150/1500mhz Gskill 8gig 1600mhz (2x4) @2033 mhz 
Hard DriveCoolingOSPower
1TB WD Caviar Black rs360 raystorm kit Windows 8 Pro 700 Watt Corsair 
CaseMouse
HAF 912  Logitech G400 
  hide details  
post #344 of 718
@2010rig I absolutely agree than BD and PD is disappointing, but the µarch itself has great potential. That's why I'm so optimistic about SR, since it's meant to correct most of the flaws that prevents PD from performing as well as it should. The problem I see is that FX will be 1 year behind the APUs. So Kaveri will have SR this year, but FX won't until 2014.
Mini ITX
(15 items)
 
AMD FX
(15 items)
 
 
CPUMotherboardGraphicsRAM
Intel i5 4670K 4.6GHz @ 1.22v MSI Z87I Sapphire Radeon 7950 825/1250 @ 1.00v Crucial Ballistix Elite 1600MHz 
Hard DriveHard DriveCoolingOS
Plextor M5P Xtreme 128GB Western Digital Blue 1TB Corsair H60 (2013 Edition) Microsoft Windows 7 Home Premium x64 
MonitorKeyboardPowerCase
HP 2311x Logitech K200 Silverstone Strider Gold ST55F-G Fractal Design Node 304 
MouseMouse PadAudio
Zalman M350 Zalman P500 Sennheiser HD428s 
CPUMotherboardGraphicsRAM
FX-6300 @ 4.5GHz Gigabyte GA-970A-UD3 HIS Radeon 7770 Ghz Edition G.Skill 1333MHz 2x4GB 
Hard DriveOptical DriveCoolingOS
Crucial m4 64GB Sony DVD-RW Xigmatek Gaia sd1283 Windows 7 Home Premium x64 
MonitorKeyboardPowerCase
HP 2311x Logitech Rosewill Hive 550 Lian-Li PC-K58 
MouseMouse PadAudio
Logitech G400 Razer Kabuto Creative T40 
  hide details  
Mini ITX
(15 items)
 
AMD FX
(15 items)
 
 
CPUMotherboardGraphicsRAM
Intel i5 4670K 4.6GHz @ 1.22v MSI Z87I Sapphire Radeon 7950 825/1250 @ 1.00v Crucial Ballistix Elite 1600MHz 
Hard DriveHard DriveCoolingOS
Plextor M5P Xtreme 128GB Western Digital Blue 1TB Corsair H60 (2013 Edition) Microsoft Windows 7 Home Premium x64 
MonitorKeyboardPowerCase
HP 2311x Logitech K200 Silverstone Strider Gold ST55F-G Fractal Design Node 304 
MouseMouse PadAudio
Zalman M350 Zalman P500 Sennheiser HD428s 
CPUMotherboardGraphicsRAM
FX-6300 @ 4.5GHz Gigabyte GA-970A-UD3 HIS Radeon 7770 Ghz Edition G.Skill 1333MHz 2x4GB 
Hard DriveOptical DriveCoolingOS
Crucial m4 64GB Sony DVD-RW Xigmatek Gaia sd1283 Windows 7 Home Premium x64 
MonitorKeyboardPowerCase
HP 2311x Logitech Rosewill Hive 550 Lian-Li PC-K58 
MouseMouse PadAudio
Logitech G400 Razer Kabuto Creative T40 
  hide details  
post #345 of 718
Quote:
Originally Posted by S.M. View Post

It's not difficult to grasp, it's just irrelevant to the conversation of single threaded throughput between two different architectures.

Your benchmark is inconclusive. Both chips would turbo: 3.7Ghz and 4.0Ghz respectively.

I don't understand. Since when a synthetic benchmark became the be all end all when it comes to evaluating single thread ipc? Let's take a look at a game that is infamous for disliking bulldozer architecture and that sees only 2 threads-the phenom should maul every bulldozer when clocked near each other.



Using a crappy mobo that doesn't even support FX, mind you. When turbo is calculated 6100 should be at best 200mhz ahead. Yet the bulldozer clearly beats the phenom.
Mastodon Ryzen
(12 items)
 
HP Z220
(8 items)
 
 
CPUMotherboardGraphicsRAM
R7 1800X Asus Crosshair VI Hero Sapphire RX Vega 64 reference Gskill TridentZ 
Hard DriveOptical DriveCoolingOS
Pny SSD 240GB Crucial MX100 CM Nepton 280L Win 10 
MonitorPowerCaseMouse
Acer Predator XG270HU Freesync XFX 750W Pro HAF XM Logitech G502 
CPUMotherboardGraphicsCooling
i7 3770 HP Quadro K2000 HP 
OSPowerCaseMouse
Win 7  HP 400W HP CMT RAT 7 
  hide details  
Mastodon Ryzen
(12 items)
 
HP Z220
(8 items)
 
 
CPUMotherboardGraphicsRAM
R7 1800X Asus Crosshair VI Hero Sapphire RX Vega 64 reference Gskill TridentZ 
Hard DriveOptical DriveCoolingOS
Pny SSD 240GB Crucial MX100 CM Nepton 280L Win 10 
MonitorPowerCaseMouse
Acer Predator XG270HU Freesync XFX 750W Pro HAF XM Logitech G502 
CPUMotherboardGraphicsCooling
i7 3770 HP Quadro K2000 HP 
OSPowerCaseMouse
Win 7  HP 400W HP CMT RAT 7 
  hide details  
post #346 of 718
Quote:
Originally Posted by 2010rig View Post

@Brutuz I want to bring up a question for you to ponder.

If PD @ 4GHZ is matching an i5-750 @ 2.6 GHZ

Question: How high would PD need to clock just to match the IPC of the I5-750 when it's @ 4.0, a 53% overclock?

Keep in mind the average OC for PD is around 4.8 on air ( which I predicted way before PD released remember? ), which is a mere 20% overclock.

Answer: PD would need to clock to 6 GHZ just to match the i5-750's IPC @ 4GHZ, something PD isn't capable of on water, let alone air.

Let's bring Ivy Bridge @ 4.5 into the equation, how high would PD need to clock to match its IPC? Ponder that for a bit, actually, don't, since you can't tell the difference between the 2. tongue.gif

My point is, the 8350 just doesn't have as much overclocking headroom compared to Intel chips in general.

It's not like it can clock to 5.5 nor did it match Nehalem IPC, like you once "predicted", funny how back then you WERE discussing IPC & single thread...
Also, what was the original point in this whole discussion? The mere fact that PD is matching the IPC of a 45nm chip from 2009 is not impressive by any means, since it needs to be clocked 1.3GHZ higher in order to do so, and the 8350 only has 20% - 25% overclocking headroom beyond that 4 GHZ stock clock.

If PD clocked to 6 GHZ+ easily, you may have an argument I could get on board with, but due to the limited overclocking headroom, I can not and will not. wink.gif

My thoughts on Steamroller are completely different, and I'm beyond optimistic on that front, but wanted to clarify why the clock for clock comparisons are important. I'm sure a lot of you will disagree, I'll get back to you tomorrow.
Your overclock comparisons are unfair. You're comparing the highest clocked chip from one with a downclocked chip from the other.

3.0Ghz to 4.0Ghz is 33% (i7-870~)
4.0Ghz to 4.8Ghz is 20% (FX-8350)
3.5Ghz to 4.8Ghz is 37% (i5-3770k)
3.5Ghz to 4.8Ghz is 37% (FX-8320)
3.4Ghz to 4.8Ghz is 41% (i5-3570k)
2.66Ghz to 4.0Ghz is 53% (i7-750)

I think I'm being pretty fair saying Ivy's average overclock on air is 4.8Ghz, grabbed from here.

I think we can agree Bulldozer has the pedigree to clock high.

AMD's approach makes sense to me. Create a smaller, scale-able and modular architecture that matches the old one, get 8 of them on silicon, then from there, any improvements they make to the new architecture is multiplied heavily. I'm excited to see how dedicated decode effects multi-core scaling.
Money
(17 items)
 
Z575
(9 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 7 1700 @ 3.8Ghz MSI B350 Pro Carbon GIGABYTE R9 390 8GB 1100/1650Mhz 16GB Team DDR4-3000 
Hard DriveHard DriveHard DriveCooling
PNY CS1311 240GB - OS 2 x 2TB WD Black Raid 0 Synology DS416slim 4 x 1TB (raid 5) Corsair H100 
OSOSMonitorKeyboard
Windows 10 Pro ElementaryOS Loki LG 29UM67P 29" 2560x1080 FreeSync HPE 87 w/ Browns + Ducky PBT caps 
PowerCaseMouseAudio
EVGA 600W B2 Lian Li PCA05N Steelseries Rival 100 FiiO G10K + Philips SN9500HiFi 
CPUMotherboardGraphicsRAM
AMD A6-3420M @ 2.2Ghz 1.1v Lenovo Torpedo A55 6720g2 8GB Kingston 1600 CL9 
Hard DriveOSMouse
Samsung 840 250GB ElementaryOS x86_64 Abyssus 3.5G 
  hide details  
Money
(17 items)
 
Z575
(9 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 7 1700 @ 3.8Ghz MSI B350 Pro Carbon GIGABYTE R9 390 8GB 1100/1650Mhz 16GB Team DDR4-3000 
Hard DriveHard DriveHard DriveCooling
PNY CS1311 240GB - OS 2 x 2TB WD Black Raid 0 Synology DS416slim 4 x 1TB (raid 5) Corsair H100 
OSOSMonitorKeyboard
Windows 10 Pro ElementaryOS Loki LG 29UM67P 29" 2560x1080 FreeSync HPE 87 w/ Browns + Ducky PBT caps 
PowerCaseMouseAudio
EVGA 600W B2 Lian Li PCA05N Steelseries Rival 100 FiiO G10K + Philips SN9500HiFi 
CPUMotherboardGraphicsRAM
AMD A6-3420M @ 2.2Ghz 1.1v Lenovo Torpedo A55 6720g2 8GB Kingston 1600 CL9 
Hard DriveOSMouse
Samsung 840 250GB ElementaryOS x86_64 Abyssus 3.5G 
  hide details  
post #347 of 718
Quote:
Originally Posted by Kuivamaa View Post

I don't understand. Since when a synthetic benchmark became the be all end all when it comes to evaluating single thread ipc? Let's take a look at a game that is infamous for disliking bulldozer architecture and that sees only 2 threads-the phenom should maul every bulldozer when clocked near each other.



Using a crappy mobo that doesn't even support FX, mind you. When turbo is calculated 6100 should be at best 200mhz ahead. Yet the bulldozer clearly beats the phenom.

Source: http://www.computerforum.com/211066-phenom-ii-vs-fx.html

960t @ 3.3GHz

960t @ 4.0GHz


fx6100 @ 3.3GHz

fx6100 @ 4.0GHz

fx6100 @ 4.5GHz
Mini ITX
(15 items)
 
AMD FX
(15 items)
 
 
CPUMotherboardGraphicsRAM
Intel i5 4670K 4.6GHz @ 1.22v MSI Z87I Sapphire Radeon 7950 825/1250 @ 1.00v Crucial Ballistix Elite 1600MHz 
Hard DriveHard DriveCoolingOS
Plextor M5P Xtreme 128GB Western Digital Blue 1TB Corsair H60 (2013 Edition) Microsoft Windows 7 Home Premium x64 
MonitorKeyboardPowerCase
HP 2311x Logitech K200 Silverstone Strider Gold ST55F-G Fractal Design Node 304 
MouseMouse PadAudio
Zalman M350 Zalman P500 Sennheiser HD428s 
CPUMotherboardGraphicsRAM
FX-6300 @ 4.5GHz Gigabyte GA-970A-UD3 HIS Radeon 7770 Ghz Edition G.Skill 1333MHz 2x4GB 
Hard DriveOptical DriveCoolingOS
Crucial m4 64GB Sony DVD-RW Xigmatek Gaia sd1283 Windows 7 Home Premium x64 
MonitorKeyboardPowerCase
HP 2311x Logitech Rosewill Hive 550 Lian-Li PC-K58 
MouseMouse PadAudio
Logitech G400 Razer Kabuto Creative T40 
  hide details  
Mini ITX
(15 items)
 
AMD FX
(15 items)
 
 
CPUMotherboardGraphicsRAM
Intel i5 4670K 4.6GHz @ 1.22v MSI Z87I Sapphire Radeon 7950 825/1250 @ 1.00v Crucial Ballistix Elite 1600MHz 
Hard DriveHard DriveCoolingOS
Plextor M5P Xtreme 128GB Western Digital Blue 1TB Corsair H60 (2013 Edition) Microsoft Windows 7 Home Premium x64 
MonitorKeyboardPowerCase
HP 2311x Logitech K200 Silverstone Strider Gold ST55F-G Fractal Design Node 304 
MouseMouse PadAudio
Zalman M350 Zalman P500 Sennheiser HD428s 
CPUMotherboardGraphicsRAM
FX-6300 @ 4.5GHz Gigabyte GA-970A-UD3 HIS Radeon 7770 Ghz Edition G.Skill 1333MHz 2x4GB 
Hard DriveOptical DriveCoolingOS
Crucial m4 64GB Sony DVD-RW Xigmatek Gaia sd1283 Windows 7 Home Premium x64 
MonitorKeyboardPowerCase
HP 2311x Logitech Rosewill Hive 550 Lian-Li PC-K58 
MouseMouse PadAudio
Logitech G400 Razer Kabuto Creative T40 
  hide details  
post #348 of 718
Quote:
Originally Posted by boot318 View Post

30% over BD people.

... What's your point? Of course it is 30% over Bulldozer. That still means 20% faster than Phenom II per clock.
   
AGP bencher
(14 items)
 
CPUMotherboardGraphicsRAM
Ryzen R7 1700 Gigabyte GA-AX370-Gaming 5 Sapphire HD 6950 2GiB 2x8GB KFA2 HOF DDR4-3600 
Hard DriveHard DriveHard DriveHard Drive
Crucial MX100 256GB Seagate 600 Series 240GB Seagate 7200.14 2TB Samsung F3 1TB 
CoolingCoolingCoolingCooling
EKWB Supreme HF XSPC Rasa GPU EK XT360 EK 4.0 
OSMonitorMonitorKeyboard
W10 Pro LG IPS235 LG E2250V KUL ES-87 
PowerCaseMouseAudio
SF Leadex II 650W Lian Li PC-A05NB Logitech G9 Xonar DX 
AudioAudio
SMSL SA-S3+Technics CB-250 Sennheiser HD555 
CPUMotherboardRAMHard Drive
AMD A10-5700 Gigabyte F2A75M-HD2 G.SKILL Ares 2133 CL9 Hitachi 5K750 
Hard DriveCoolingOSMonitor
Momentus .7 200GB Noctua NH-L9a Server 2012 R2 Standard AUO B156HW01 
PowerCaseOther
PicoPSU-80-WI-25V AIO Aluminium Handmade TP-Link Archer Something Something Wi-Fi AC 
CPUCPUCPUMotherboard
Core2Duo E6400 Core2Quad Q6600 Pentium Dual Core E5200 AsRock 4COREDUAL-SATA2 R2.0 
GraphicsRAMHard DriveOptical Drive
A dumpload of ancient AGP cards Kingston Value DDR2-667 CL4 2T @CL3 1T Seagate 160GB 7200.10 LG IDE DVD-ROM 
CoolingCoolingOSMonitor
Ghettomade CPU waterblock 49cc 2stroke engine copper radiator WinXP SP2 32bit ProView 17" 
PowerCase
Tacens Radix V 550W Ghetto aluminium bench 
  hide details  
   
AGP bencher
(14 items)
 
CPUMotherboardGraphicsRAM
Ryzen R7 1700 Gigabyte GA-AX370-Gaming 5 Sapphire HD 6950 2GiB 2x8GB KFA2 HOF DDR4-3600 
Hard DriveHard DriveHard DriveHard Drive
Crucial MX100 256GB Seagate 600 Series 240GB Seagate 7200.14 2TB Samsung F3 1TB 
CoolingCoolingCoolingCooling
EKWB Supreme HF XSPC Rasa GPU EK XT360 EK 4.0 
OSMonitorMonitorKeyboard
W10 Pro LG IPS235 LG E2250V KUL ES-87 
PowerCaseMouseAudio
SF Leadex II 650W Lian Li PC-A05NB Logitech G9 Xonar DX 
AudioAudio
SMSL SA-S3+Technics CB-250 Sennheiser HD555 
CPUMotherboardRAMHard Drive
AMD A10-5700 Gigabyte F2A75M-HD2 G.SKILL Ares 2133 CL9 Hitachi 5K750 
Hard DriveCoolingOSMonitor
Momentus .7 200GB Noctua NH-L9a Server 2012 R2 Standard AUO B156HW01 
PowerCaseOther
PicoPSU-80-WI-25V AIO Aluminium Handmade TP-Link Archer Something Something Wi-Fi AC 
CPUCPUCPUMotherboard
Core2Duo E6400 Core2Quad Q6600 Pentium Dual Core E5200 AsRock 4COREDUAL-SATA2 R2.0 
GraphicsRAMHard DriveOptical Drive
A dumpload of ancient AGP cards Kingston Value DDR2-667 CL4 2T @CL3 1T Seagate 160GB 7200.10 LG IDE DVD-ROM 
CoolingCoolingOSMonitor
Ghettomade CPU waterblock 49cc 2stroke engine copper radiator WinXP SP2 32bit ProView 17" 
PowerCase
Tacens Radix V 550W Ghetto aluminium bench 
  hide details  
post #349 of 718
the problem i see with steamroller is after a very long time AMD is going to be moving to a bulk process from SOI. the 28nm bulk process is going to be one of the key factors in determining how good or bad the steamroller products turn out. AMD has needed SOI all these years to get competitive transistor performance compared to Intel at the same node. If AMD can keep similar clocks on steamroller as piledriver within the same TDP then they would have done a good job with all the IPC improvements and improved decode bandwidth which will improve single thread and multi thread preformance. Also AMD does not need to match Intel for single thread performance. They need to just make sure the gap does not exceed 20%. AMD's multithreaded scaling is much better than Intel's Hyperthreading which scales at 20 - 25%. Piledriver is aleady scaling at 50 - 60% and Steamroller should increase that even more. AMD needs single thread IPC in the nehalem range to really pose a challenge to Intel when it comes to multithread performance.
Fragbox
(14 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core i5 2400 DH67BL AMD Radeon HD 6900 Series XFX HD 6950 2GB 
RAMRAMHard DriveOptical Drive
Corsair Corsair DDR3 1333 Mhz 2 GB Corsair DDR3 1333 Mhz 2 GB Western Digital Caviar Green SONY DVD-RW AD-7260S 
OSMonitorKeyboardPower
Windows 7 Professional 32 bit BENQ G2420HD Logitech K200 Seasonic VX550 psu 
CaseMouse
ANTEC 200 V2 Logitech mouse 
  hide details  
Fragbox
(14 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core i5 2400 DH67BL AMD Radeon HD 6900 Series XFX HD 6950 2GB 
RAMRAMHard DriveOptical Drive
Corsair Corsair DDR3 1333 Mhz 2 GB Corsair DDR3 1333 Mhz 2 GB Western Digital Caviar Green SONY DVD-RW AD-7260S 
OSMonitorKeyboardPower
Windows 7 Professional 32 bit BENQ G2420HD Logitech K200 Seasonic VX550 psu 
CaseMouse
ANTEC 200 V2 Logitech mouse 
  hide details  
post #350 of 718
Quote:
Originally Posted by raghu78 View Post

the problem i see with steamroller is after a very long time AMD is going to be moving to a bulk process from SOI. the 28nm bulk process is going to be one of the key factors in determining how good or bad the steamroller products turn out. AMD has needed SOI all these years to get competitive transistor performance compared to Intel at the same node.

GF is the only fab who still uses SOI for high performance VLSI chips. TSMC dropped it years ago, in favor of bulk. So did Intel.

In moving from SOI to bulk, they also have a much better portability for their designs, meaning they are much easier to manufacture in other fabs that already manufacture bulk silicon. And IIRC, they were using TSMC to build their Kabini and Temash APUs, so in theory they could source production of Steamroller chips if so was requiered, with relatively minor modifications to the plans (compared to adapting the design from SOI to bulk).
   
AGP bencher
(14 items)
 
CPUMotherboardGraphicsRAM
Ryzen R7 1700 Gigabyte GA-AX370-Gaming 5 Sapphire HD 6950 2GiB 2x8GB KFA2 HOF DDR4-3600 
Hard DriveHard DriveHard DriveHard Drive
Crucial MX100 256GB Seagate 600 Series 240GB Seagate 7200.14 2TB Samsung F3 1TB 
CoolingCoolingCoolingCooling
EKWB Supreme HF XSPC Rasa GPU EK XT360 EK 4.0 
OSMonitorMonitorKeyboard
W10 Pro LG IPS235 LG E2250V KUL ES-87 
PowerCaseMouseAudio
SF Leadex II 650W Lian Li PC-A05NB Logitech G9 Xonar DX 
AudioAudio
SMSL SA-S3+Technics CB-250 Sennheiser HD555 
CPUMotherboardRAMHard Drive
AMD A10-5700 Gigabyte F2A75M-HD2 G.SKILL Ares 2133 CL9 Hitachi 5K750 
Hard DriveCoolingOSMonitor
Momentus .7 200GB Noctua NH-L9a Server 2012 R2 Standard AUO B156HW01 
PowerCaseOther
PicoPSU-80-WI-25V AIO Aluminium Handmade TP-Link Archer Something Something Wi-Fi AC 
CPUCPUCPUMotherboard
Core2Duo E6400 Core2Quad Q6600 Pentium Dual Core E5200 AsRock 4COREDUAL-SATA2 R2.0 
GraphicsRAMHard DriveOptical Drive
A dumpload of ancient AGP cards Kingston Value DDR2-667 CL4 2T @CL3 1T Seagate 160GB 7200.10 LG IDE DVD-ROM 
CoolingCoolingOSMonitor
Ghettomade CPU waterblock 49cc 2stroke engine copper radiator WinXP SP2 32bit ProView 17" 
PowerCase
Tacens Radix V 550W Ghetto aluminium bench 
  hide details  
   
AGP bencher
(14 items)
 
CPUMotherboardGraphicsRAM
Ryzen R7 1700 Gigabyte GA-AX370-Gaming 5 Sapphire HD 6950 2GiB 2x8GB KFA2 HOF DDR4-3600 
Hard DriveHard DriveHard DriveHard Drive
Crucial MX100 256GB Seagate 600 Series 240GB Seagate 7200.14 2TB Samsung F3 1TB 
CoolingCoolingCoolingCooling
EKWB Supreme HF XSPC Rasa GPU EK XT360 EK 4.0 
OSMonitorMonitorKeyboard
W10 Pro LG IPS235 LG E2250V KUL ES-87 
PowerCaseMouseAudio
SF Leadex II 650W Lian Li PC-A05NB Logitech G9 Xonar DX 
AudioAudio
SMSL SA-S3+Technics CB-250 Sennheiser HD555 
CPUMotherboardRAMHard Drive
AMD A10-5700 Gigabyte F2A75M-HD2 G.SKILL Ares 2133 CL9 Hitachi 5K750 
Hard DriveCoolingOSMonitor
Momentus .7 200GB Noctua NH-L9a Server 2012 R2 Standard AUO B156HW01 
PowerCaseOther
PicoPSU-80-WI-25V AIO Aluminium Handmade TP-Link Archer Something Something Wi-Fi AC 
CPUCPUCPUMotherboard
Core2Duo E6400 Core2Quad Q6600 Pentium Dual Core E5200 AsRock 4COREDUAL-SATA2 R2.0 
GraphicsRAMHard DriveOptical Drive
A dumpload of ancient AGP cards Kingston Value DDR2-667 CL4 2T @CL3 1T Seagate 160GB 7200.10 LG IDE DVD-ROM 
CoolingCoolingOSMonitor
Ghettomade CPU waterblock 49cc 2stroke engine copper radiator WinXP SP2 32bit ProView 17" 
PowerCase
Tacens Radix V 550W Ghetto aluminium bench 
  hide details  
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
This thread is locked  
Overclock.net › Forums › Industry News › Hardware News › [xbitlabs] AMD: We Are On Track With Steamroller Micro-Architecture in 2013.