Overclock.net › Forums › Industry News › Hardware News › [TechReport] Intel gives peek into Nehalem bag of tricks
New Posts  All Forums:Forum Nav:

[TechReport] Intel gives peek into Nehalem bag of tricks - Page 4

post #31 of 55
This is very entertaining and informative
Cube
(18 items)
 
  
CPUMotherboardGraphicsGraphics
i7 6700k MSI Z170A GAMING 7 ASUS STRIX 1070 ASUS STRIX 1070 
RAMHard DriveHard DriveHard Drive
EVGA 32GB PC4 3000 SAMSUNG EVO 840 256GB 2 TB Hitachi 2 TB Hitachi 
Optical DriveCoolingOSMonitor
BDR R/W Corsair H115i WIN 10 PRO 27" DELL S2716DG 144HZ Gsync 
KeyboardPowerCaseAudio
Razer Black Widow Ultimate Corsair RM 850X Corsair Carbide AIR 540 YAMAHA R-S202 
AudioAudio
BIC F12 AMERICA 12" DAYTON B652 AIR 
  hide details  
Reply
Cube
(18 items)
 
  
CPUMotherboardGraphicsGraphics
i7 6700k MSI Z170A GAMING 7 ASUS STRIX 1070 ASUS STRIX 1070 
RAMHard DriveHard DriveHard Drive
EVGA 32GB PC4 3000 SAMSUNG EVO 840 256GB 2 TB Hitachi 2 TB Hitachi 
Optical DriveCoolingOSMonitor
BDR R/W Corsair H115i WIN 10 PRO 27" DELL S2716DG 144HZ Gsync 
KeyboardPowerCaseAudio
Razer Black Widow Ultimate Corsair RM 850X Corsair Carbide AIR 540 YAMAHA R-S202 
AudioAudio
BIC F12 AMERICA 12" DAYTON B652 AIR 
  hide details  
Reply
post #32 of 55
Quote:
Originally Posted by pauldovi View Post
I disagree with this.

First of all, Nehalem is designed for application improvements from servers to notebooks. The QPI and monolithic die is as much for multi-socket bandwidth as it is for their integrated GPU project.

You did not answer my questions about where you assessment of 20-30% improvement in clock-for-clock performance and speeds starting at 3Ghz came from.

You raise a valid point that the performance improvements, clock for clock in a single L2 thread is minimal. However you seem to suggest that multithreading performance is only desired and beneficial to the server market?

What I find interesting is your assessment of Barcelona's supurior performance in 4 socket server market, because it isn't true:








Now please entertain us with a source for the 20-30% IPC inprovement and the 3.0Ghz+ launch clocks.
I usually agree with you Pauldovi but i have read a number of multi-CPU server performance assessments concerning the Barcelona core and i will tell you Barcelona is a great server chip. That's what it was designed for (as this is where most of AMD's CPU business is, not desktops, another reason Phenom hit the wall so hard, it was a server chip on a different socket.) When considering its drastically less power consumption (plus the fact AMD motherboards use as much and sometimes more then 90w less power, although i don't know how many server boards use AMD chipsets) especially on idle and its lower price point the Barcelona chip is an ideal server processor. Especially since it is common knowledge AMD servers scale better (which is why Opterons are used in super computers, not Xeons.) Also when i look at your graphs i see a lower clocked Barcelona taking on higher clocked Xeons and defeating Xeons on the same clock.
Lee XT
(17 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX-6300 Asus M5A97 SAPPHIRE Radeon HD 7850 AMD 4GB DDR3 1333MHZ 
RAMRAMRAMHard Drive
AMD 4GB DDR3 1333MHZ AMD 4GB DDR3 1333MHZ AMD 4GB DDR3 1333MHZ OCZ Vertex 4 256GB 
CoolingOSMonitorKeyboard
Corsair H80 Windows 8.1 Pro MCE Dell P2414H WHXV7  Microsoft Generic 
PowerCaseMouseMouse Pad
Ultra 600W Limited Edition NZXT Black Steel Razer Deathadder Razer Goliath 
Audio
Realtek HD Audio 
  hide details  
Reply
Lee XT
(17 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX-6300 Asus M5A97 SAPPHIRE Radeon HD 7850 AMD 4GB DDR3 1333MHZ 
RAMRAMRAMHard Drive
AMD 4GB DDR3 1333MHZ AMD 4GB DDR3 1333MHZ AMD 4GB DDR3 1333MHZ OCZ Vertex 4 256GB 
CoolingOSMonitorKeyboard
Corsair H80 Windows 8.1 Pro MCE Dell P2414H WHXV7  Microsoft Generic 
PowerCaseMouseMouse Pad
Ultra 600W Limited Edition NZXT Black Steel Razer Deathadder Razer Goliath 
Audio
Realtek HD Audio 
  hide details  
Reply
post #33 of 55
Quote:
Originally Posted by pauldovi View Post
I disagree with this.

First of all, Nehalem is designed for application improvements from servers to notebooks. The QPI and monolithic die is as much for multi-socket bandwidth as it is for their integrated GPU project.

You did not answer my questions about where you assessment of 20-30% improvement in clock-for-clock performance and speeds starting at 3Ghz came from.

You raise a valid point that the performance improvements, clock for clock in a single L2 thread is minimal. However you seem to suggest that multithreading performance is only desired and beneficial to the server market?

What I find interesting is your assessment of Barcelona's supurior performance in 4 socket server market, because it isn't true:








Now please entertain us with a source for the 20-30% IPC inprovement and the 3.0Ghz+ launch clocks.
Pauldovi are you F-ing kidding me? or do you not even read the benchmarks you post??? look at that quad barcelona setup, AMD's 2.3ghz is beating Intel's 2.4ghz, and coming DAMNED close to their 2.9 ghz

And your claims that Nehalem will perform 40% better in single threaded apps is complete rubbish as darkcloud has already proven

The fact is, while the Nehalems will perform much better than the kent/yorkfield processors in multi-threaded and multi-socket server apps, they dont perform much, if at all better in the things most people on this website care about... Games.

And with the Phenom only trailing the current Core2Quads by about 10% clock-for-clock, I don't think it's too long of a shot that the 45nm Phenoms might be able to surpass both the C2Qs, and even the Nehalems in single threaded apps clock-for-clock.

However since AMD wont be implementing High-k on their 45nm chips, Intel will probably have a clock ceiling lead on AMD, which might mean we are going back into another Netburst/K8 era.

(except this time the Netbursts wont suck as hard)
Edited by HugeDink - 6/17/08 at 9:53pm
Von Strauss
(13 items)
 
  
CPUMotherboardGraphicsRAM
Athlon 5000x2 BE @ 3.2 ASUS M2R-32 MVP Sapphire HD4850 700/1100 2x1gb G.Skill HZs 
Hard DriveOptical DriveOSMonitor
250gb WD IDE lol LG DVD-RAM w/ Lightscribe Vista Ultimate 32-bit Sceptre, 1280x1024, 5ms 
PowerCase
Corsair TX 650W Enermax Chakra w/ 250mm Side Panel Fan 
  hide details  
Reply
Von Strauss
(13 items)
 
  
CPUMotherboardGraphicsRAM
Athlon 5000x2 BE @ 3.2 ASUS M2R-32 MVP Sapphire HD4850 700/1100 2x1gb G.Skill HZs 
Hard DriveOptical DriveOSMonitor
250gb WD IDE lol LG DVD-RAM w/ Lightscribe Vista Ultimate 32-bit Sceptre, 1280x1024, 5ms 
PowerCase
Corsair TX 650W Enermax Chakra w/ 250mm Side Panel Fan 
  hide details  
Reply
post #34 of 55
Quote:
Originally Posted by darkcloud89 View Post
I wouldn't be so sure about this. I commonly see that people think that AMD has no answer to Nehalem until Bulldozer. Really this isn't correct because Nehalem is actually Intel's answer to Barcelona. How do I make this conclusion? Barcelona and Nehalem are pretty similar chips.

AMD's transition from K8 to Barcelona to K10.5/Shanghai is very similar to Intel's transition from Netburst to Conroe to Nehalem. Conroe brought improved single-threaded performance and IPC (as well as efficiency) compared to Netburst, and the integration of the memory controller should give Nehalem scalability. AMD, on the other hand, released Barcelona and took care of the scalability improvements first by improving cHT links and a monolithic quad core design. And if they stay on track, Shanghai should bring improved IPC, power requirements/thermals (transition to HKMG 45nm), and single threaded performance. I guess you could say both companies were headed towards the same destination, but took different paths to get there.

If AMD can deliver the IPC improvements with the transition to 45nm in Shanghai, then that should suffice as Nehalem's competitor. Not only would K10.5 be faster per clock, but HKMG should allow it to clock much higher than it does now. Phenom was a ridiculously complex die to manufacture at 65nm and I think the process is holding it back more than most people think, and the 45nm transition if executed timely may open some eyes. The performance advantage with Nehalem will not be as big with single-threaded performance as it will be in multithreaded. AMD was already very good when it came to multithreaded performance, so really they just need to improve single-threaded.

EDIT: I see my post from the other thread was already quoted
Totaly, 100% disagree, and I will put money on Shanghi not holding a candle to Nehalem.

Shanghi is to Barcelona as Intels 45nm chips are to it's 65nm chips, meaning minor IPC improvements.

Intel went from a 3 issue core with a 31 state pipeline on its last netburst chips, to a 4 issue core with a 14 stage pipeline on the Core 2. AMD is not doing anything so massive. NetBurst to Conroe is not even close to a valid analogy for the switch to Shanghi.

I am certain that the performance gap between Intel and AMD is about to widen, not close. Nehalem will beat Shanghi by more, probably much more than Core 2 beats Barcelona, even in AMDs best case scenario.

K10 was dead before it was even released. The 3 issue core almost guaranteed that.

As for servers, AMD has been able to hold its own as far as performance, but the main reason for that has been the HT bus and integrated memory controler, meaning much better multi-socket scalability. AMD is about to lose this advantage/monopoly.
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
post #35 of 55
Quote:
Originally Posted by pauldovi View Post
First of all, Nehalem is designed for application improvements from servers to notebooks. The QPI and monolithic die is as much for multi-socket bandwidth as it is for their integrated GPU project.
I bet I can find something similar in the Nehalem press kit. All this means is they want a scalable architecture. I thought I already said that.

Quote:
Originally Posted by pauldovi
You did not answer my questions about where you assessment of 20-30% improvement in clock-for-clock performance and speeds starting at 3Ghz came from.
I never said that it would have a 20-30%, I was hypothetically speaking if they can get a 20-30% improvement. However, I do realize that a 30% IPC improvement is very unlikely at this point and putting too much effort into it would be a waste of time when there should be some significant resources put into Bulldozer right now. However, the 20% is a much more realistic and reasonable expectation, and is in line with what is rumored:
Quote:
Well-informed sources have confirmed that Shanghai K10.5, a 45nm version of AMD’s native quad core, and it will be about ten to twenty percent faster than the current K10 at 65nm. This is a clock per clock comparison and this is just as much juice as AMD needed to run faster than Core 2 Duo and Quad.
Source
Now this does fail to account for the fact that not only would performance-per-clock increase, but that this would come at the same time as a die shrink, so you get the double whammy of IPC improvements and clock speed improvements. If executed correctly, this will be a pretty significant jump. As for the estimated clock speed, the 3GHz was a personal guess based on the results in the 9850BE Low Voltage Overclocking thread at techpowerup.

Quote:
Originally Posted by pauldovi
You raise a valid point that the performance improvements, clock for clock in a single L2 thread is minimal. However you seem to suggest that multithreading performance is only desired and beneficial to the server market?
It's not that improved multithreaded performance is ONLY beneficial to servers, but that they will see MORE benefit. Judging from all the hype I see generated by Nehalem, people are expecting it to be a revolutionary desktop processor. Most desktop applications are still single threaded, and these simply will not see that great of a benefit from the move to Nehalem.

Quote:
Originally Posted by pauldovi
What I find interesting is your assessment of Barcelona's supurior performance in 4 socket server market, because it isn't true:
What these graphs tell me is that Intel needs to rely on increasing clock speed (2.93GHz) in order to compete with the 2.3GHz Opteron. The results of the rendering benchmarks are to be expected because they tend to be more reliant on clock speed which obviously the Xeon has, but that isn't to say that AMD won't get there eventually. Intel knows they can only increase clock speed to stay competitive up to a certain point. Obviously they felt threatened by AMD's scalability or they wouldn't see the need to go about creating a more scalable architecture themselves.
post #36 of 55
Quote:
Originally Posted by HugeDink View Post
Pauldovi are you F-ing kidding me? or do you not even read the benchmarks you post??? look at that quad barcelona setup, AMD's 2.3ghz is beating Intel's 2.4ghz, and coming DAMNED close to their 2.9 ghz

And your claims that Nehalem will perform 40% better in single threaded apps is complete rubbish as darkcloud has already proven

The fact is, while the Nehalems will perform much better than the kent/yorkfield processors in multi-threaded and multi-socket server apps, they dont perform much, if at all better in the things most people on this website care about... Games.

And with the Phenom only trailing the current Core2Quads by about 10% clock-for-clock, I don't think it's too long of a shot that the 45nm Phenoms might be able to surpass both the C2Qs, and even the Nehalems in single threaded apps clock-for-clock.

However since AMD wont be implementing High-k on their 45nm chips, Intel will probably have a clock ceiling lead on AMD, which might mean we are going back into another Netburst/K8 era.

(except this time the Netbursts wont suck as hard)
If AMD has a 2.9Ghz K10 that would be a problem for Intel, but they don't.

I want to see solid number about your IPC improvements from AMD.
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
System
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2500k ASRock P67 Extreme4 Gen 3 AMD 7970 16GB DDR3 
Hard DriveOptical DriveOSMonitor
Intel 520 256GB SATA DVD Burner Windows 7 64 bit Deal U2410 
KeyboardPowerMouse
Adesso Mechanical Silverstone OP650 Logitech G700 
  hide details  
Reply
post #37 of 55
Quote:
Originally Posted by Blameless View Post
Intel went from a 3 issue core with a 31 state pipeline on its last netburst chips, to a 4 issue core with a 14 stage pipeline on the Core 2. AMD is not doing anything so massive.
I see what you're saying, and I know the IPC changes aren't going to be massive from Barcelona to Shanghai, but they are significant. But the thing is, for single threaded performance on a per-clock basis, the current Phenom is not that far behind the C2Q. A 15-20% per-clock performance improvement would comfortably put Phenom on par with C2Q. And for single threads, if it's on par with C2Q, that isn't going to change a whole lot with Nehalem. So effectively it would be a cores/GHz race for performance at this point.
Edited by darkcloud89 - 6/17/08 at 10:34pm
post #38 of 55
Quote:
Originally Posted by pauldovi View Post
If AMD has a 2.9Ghz K10 that would be a problem for Intel, but they don't.
It doesn't matter, because it doesn't change the fact that Companies choose AMD because Intel needs an expensive 120W 2.93ghz processor with a power hungry northbridge to compete with a cheap, 75w 2.3ghz processor from AMD.

Power usage and heat dissipation are HUGE concerns for big companies with big servers.
Von Strauss
(13 items)
 
  
CPUMotherboardGraphicsRAM
Athlon 5000x2 BE @ 3.2 ASUS M2R-32 MVP Sapphire HD4850 700/1100 2x1gb G.Skill HZs 
Hard DriveOptical DriveOSMonitor
250gb WD IDE lol LG DVD-RAM w/ Lightscribe Vista Ultimate 32-bit Sceptre, 1280x1024, 5ms 
PowerCase
Corsair TX 650W Enermax Chakra w/ 250mm Side Panel Fan 
  hide details  
Reply
Von Strauss
(13 items)
 
  
CPUMotherboardGraphicsRAM
Athlon 5000x2 BE @ 3.2 ASUS M2R-32 MVP Sapphire HD4850 700/1100 2x1gb G.Skill HZs 
Hard DriveOptical DriveOSMonitor
250gb WD IDE lol LG DVD-RAM w/ Lightscribe Vista Ultimate 32-bit Sceptre, 1280x1024, 5ms 
PowerCase
Corsair TX 650W Enermax Chakra w/ 250mm Side Panel Fan 
  hide details  
Reply
post #39 of 55
Quote:
Originally Posted by darkcloud89 View Post
I see what you're saying, and I know the IPC changes aren't going to be massive from Barcelona to Shanghai, but they are significant. But the thing is, for single threaded performance on a per-clock basis, the current Phenom is not that far behind the C2Q. A 15-20% per-clock performance improvement would comfortably put Phenom on par with C2Q. And for single threads, if it's on par with C2Q, that isn't going to change a whole lot with Nehalem. So effectively it would be a cores/GHz race for performance at this point until they start integrating GPUs into the processors. When this happens, I think AMD will have an advantage because they already have both a strong processor and a strong GPU, while Intel can only claim one of those.
And they will have experience with integrating north bridges onto chips as well as doing that with multiple dies from Puma. Might not think it but experience gos a long way in chip design i guarantee it. (Also Laurabee uses a bunch of x86 processors, so they will have to use their old integrated graphics on a CPU+GPU chip, which obviously won't yield great results. Assuming they don't develop a GPU architecture specifically to be Nahalem's sidekick.)
Lee XT
(17 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX-6300 Asus M5A97 SAPPHIRE Radeon HD 7850 AMD 4GB DDR3 1333MHZ 
RAMRAMRAMHard Drive
AMD 4GB DDR3 1333MHZ AMD 4GB DDR3 1333MHZ AMD 4GB DDR3 1333MHZ OCZ Vertex 4 256GB 
CoolingOSMonitorKeyboard
Corsair H80 Windows 8.1 Pro MCE Dell P2414H WHXV7  Microsoft Generic 
PowerCaseMouseMouse Pad
Ultra 600W Limited Edition NZXT Black Steel Razer Deathadder Razer Goliath 
Audio
Realtek HD Audio 
  hide details  
Reply
Lee XT
(17 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX-6300 Asus M5A97 SAPPHIRE Radeon HD 7850 AMD 4GB DDR3 1333MHZ 
RAMRAMRAMHard Drive
AMD 4GB DDR3 1333MHZ AMD 4GB DDR3 1333MHZ AMD 4GB DDR3 1333MHZ OCZ Vertex 4 256GB 
CoolingOSMonitorKeyboard
Corsair H80 Windows 8.1 Pro MCE Dell P2414H WHXV7  Microsoft Generic 
PowerCaseMouseMouse Pad
Ultra 600W Limited Edition NZXT Black Steel Razer Deathadder Razer Goliath 
Audio
Realtek HD Audio 
  hide details  
Reply
post #40 of 55
Quote:
Originally Posted by darkcloud89 View Post
I see what you're saying, and I know the IPC changes aren't going to be massive from Barcelona to Shanghai, but they are significant. But the thing is, for single threaded performance on a per-clock basis, the current Phenom is not that far behind the C2Q. A 15-20% per-clock performance improvement would comfortably put Phenom on par with C2Q. And for single threads, if it's on par with C2Q, that isn't going to change a whole lot with Nehalem. So effectively it would be a cores/GHz race for performance at this point.
I think a 20% boost in IPC is a touch optimistic, and even then, it would only bring AMD to aproximate parity with the 65nm Core 2s, the 45nm chips are another 5-10% faster, clock for clock.

Assuming Intel makes some IPC improvements with Nehalem, that are not related to the on-die memory controler, that could be another 5-10%.

The on-die memory controler itself will be a huge boost in real-world IPC. Intel is going to cut out 25-40% of it's memory latency with that alone. The K8 was essentially a K7 with a handful of minor tweaks, and an on-die memory controler. I distinctly remember the change I went from a 2.2GHz Barton to a 2.2GHz Clawhammer (I bought the 3400+ a few days after it came out in 2004), and the Clawhammer annihilated the barton in every single test I threw at it, with all other components, besides motherboard (and I had a fast K7 motherboard), being identical.

If Nehalem was just another Core 2 revision, I would probably agree that Shanghi may have a fighting chance, but Nehalem is much more than that.
Edited by Blameless - 6/17/08 at 10:42pm
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
Overclock.net › Forums › Industry News › Hardware News › [TechReport] Intel gives peek into Nehalem bag of tricks