Overclock.net › Forums › Intel › Intel Motherboards › Are all Intel boards much more costly than AM2 boards?
New Posts  All Forums:Forum Nav:

Are all Intel boards much more costly than AM2 boards? - Page 4

post #31 of 39
Quote:
Originally Posted by MGX1016 View Post
Cause he uses SuperPi times which almost everyone knows is all for Intel..
Well then why the P4 at 3Ghz is slower in SuperPI than the Athlon 64 at 3Ghz? Because the Athlon 64 architecture is much superior over it in the same way the Core 2 architecture is much superior than the Athlon 64.

The only thing that help the Athlon stay in the race is the on-die memory controller and HT bus. On pure CPU arithmetic the Core 2 is much faster than the Athlon 64.

That is the reason why the Athlon 64 dosen't gain much by overclocking the FSB, it's already capping the limit of its calculation power. The Core 2 gain a lot with boosting the FSB because the bottleneck is the I/O. on bench that stress the I/O more than the CPU crunching power the Athlon come very close like in the bench you posted.

But in situation where the I/O is not a factor like SuperPi, the dominance of the Core 2 architecture is shown. If a bench would fit in the cache of both CPU and would only test the CPU without even going to the ram once the Athlon 64 would get pulverized.

It's the same reason that for certain task the IBM/Motorola G5 CPU at 2Ghz smokes the x86 CPU's and yet for other tasks the deficient FSB of the G5 make it be worst than a P4.
Cerebro
(13 items)
 
  
CPUMotherboardGraphicsRAM
C2D E6420 3.4Ghz (8x425FSB 1:1) Evga 680i SLI NF68-T1 - P31 Asus Geforce 8800GTX OCZ Titanium Alpha VX2 + Crucial Ballistix PC8500 
Hard DriveOptical DriveOSMonitor
7 Hard Drives - All 7200RPM External LG DVD-Writer 16x via Firewire Windows XP SP2 Compaq V1000 21" CRT 
KeyboardPowerCaseMouse
Logitech G15 System - Enermax 565W / Hard Drives - Enermax 350W Thermaltake Mozart TX Logitech MX1000 
  hide details  
Reply
Cerebro
(13 items)
 
  
CPUMotherboardGraphicsRAM
C2D E6420 3.4Ghz (8x425FSB 1:1) Evga 680i SLI NF68-T1 - P31 Asus Geforce 8800GTX OCZ Titanium Alpha VX2 + Crucial Ballistix PC8500 
Hard DriveOptical DriveOSMonitor
7 Hard Drives - All 7200RPM External LG DVD-Writer 16x via Firewire Windows XP SP2 Compaq V1000 21" CRT 
KeyboardPowerCaseMouse
Logitech G15 System - Enermax 565W / Hard Drives - Enermax 350W Thermaltake Mozart TX Logitech MX1000 
  hide details  
Reply
post #32 of 39
Quote:
Originally Posted by whe3ls View Post
with a time demo you also have to factor in gfx ram cpu and clocks there are a alot of variables
Thats why we run in in software rendering mode wise guy... No 3d acceleration whatsoever. And the data is so small for an old game like that the FSB and memory won't be stressed to capacity anyway. Thats the closest I can find to a purely synthetic number crunching bench.
Cerebro
(13 items)
 
  
CPUMotherboardGraphicsRAM
C2D E6420 3.4Ghz (8x425FSB 1:1) Evga 680i SLI NF68-T1 - P31 Asus Geforce 8800GTX OCZ Titanium Alpha VX2 + Crucial Ballistix PC8500 
Hard DriveOptical DriveOSMonitor
7 Hard Drives - All 7200RPM External LG DVD-Writer 16x via Firewire Windows XP SP2 Compaq V1000 21" CRT 
KeyboardPowerCaseMouse
Logitech G15 System - Enermax 565W / Hard Drives - Enermax 350W Thermaltake Mozart TX Logitech MX1000 
  hide details  
Reply
Cerebro
(13 items)
 
  
CPUMotherboardGraphicsRAM
C2D E6420 3.4Ghz (8x425FSB 1:1) Evga 680i SLI NF68-T1 - P31 Asus Geforce 8800GTX OCZ Titanium Alpha VX2 + Crucial Ballistix PC8500 
Hard DriveOptical DriveOSMonitor
7 Hard Drives - All 7200RPM External LG DVD-Writer 16x via Firewire Windows XP SP2 Compaq V1000 21" CRT 
KeyboardPowerCaseMouse
Logitech G15 System - Enermax 565W / Hard Drives - Enermax 350W Thermaltake Mozart TX Logitech MX1000 
  hide details  
Reply
post #33 of 39
Quote:
Originally Posted by MGX1016 View Post
The C2D is not 50% faster...

CPU is at 2.87
Everything is is sig just click the Chasing a Conroe link..

Set your PC to that and run CPU-z + SuperPi 1M.
I'll do that later and post screen.. My rig is off.. As I said it's now 7AM almost.
Alright will do that.




2.87 x 18.533 = 53.18971 exactly in line with the previous times I posted. The ratio Hz:Time is exactly in line with the other Core 2 CPU's whatever the FSB is at. This prove what I said earlier.

I am sure that my 15.250 sec score at 3.4Ghz fit too. Lets calculate it shall we? 3.4 x 15.250 = 51.85 still within the margin of error I would say. That is because the gain per Mhz in Super Pi is linear. It's linear because it almost exclusively give the CPU raw power and is not affected by other bottlenecks.
Cerebro
(13 items)
 
  
CPUMotherboardGraphicsRAM
C2D E6420 3.4Ghz (8x425FSB 1:1) Evga 680i SLI NF68-T1 - P31 Asus Geforce 8800GTX OCZ Titanium Alpha VX2 + Crucial Ballistix PC8500 
Hard DriveOptical DriveOSMonitor
7 Hard Drives - All 7200RPM External LG DVD-Writer 16x via Firewire Windows XP SP2 Compaq V1000 21" CRT 
KeyboardPowerCaseMouse
Logitech G15 System - Enermax 565W / Hard Drives - Enermax 350W Thermaltake Mozart TX Logitech MX1000 
  hide details  
Reply
Cerebro
(13 items)
 
  
CPUMotherboardGraphicsRAM
C2D E6420 3.4Ghz (8x425FSB 1:1) Evga 680i SLI NF68-T1 - P31 Asus Geforce 8800GTX OCZ Titanium Alpha VX2 + Crucial Ballistix PC8500 
Hard DriveOptical DriveOSMonitor
7 Hard Drives - All 7200RPM External LG DVD-Writer 16x via Firewire Windows XP SP2 Compaq V1000 21" CRT 
KeyboardPowerCaseMouse
Logitech G15 System - Enermax 565W / Hard Drives - Enermax 350W Thermaltake Mozart TX Logitech MX1000 
  hide details  
Reply
post #34 of 39
Beautiful thread, just beautiful!!

I was always an avid AMD lover, had one sind i had the choice of building my own rig. My new rig tho, as can be seen from the sig. is a sexy e6420 coz i heard C2Ds OWN X2s severly at similar clocks.

I like the SuperPi results and the theory seems to fit so far, but the idea of a Q2 benchie has really got my attention!

Comon guys, get the results out there, i wanna see a x2 @2.87 vs. c2d @ 2.87. Give the x2 some bonus coz of lacking cache maybe, or run it against an e4300 or summin with less cache, but the results could be really nice to see!

gogogo

edit: spelling
yes yes, fx-62, same thing as a x2 really, just naturally beefed up, right?
Happy Sunshine
(13 items)
 
  
CPUMotherboardGraphicsRAM
E6420 @ 3.5 Ghz (500x7) 1.42v Asus P5B-Deluxe Gainward 8800 GTS @ 650/2010 Crucial Ballistix DDR2-6400-C4@ 1000MHz CL-5 
OSMonitorPowerCase
WinXP 32bit NEC 1970GX Targan 580W PSU Thermaltake Xaser3 
  hide details  
Reply
Happy Sunshine
(13 items)
 
  
CPUMotherboardGraphicsRAM
E6420 @ 3.5 Ghz (500x7) 1.42v Asus P5B-Deluxe Gainward 8800 GTS @ 650/2010 Crucial Ballistix DDR2-6400-C4@ 1000MHz CL-5 
OSMonitorPowerCase
WinXP 32bit NEC 1970GX Targan 580W PSU Thermaltake Xaser3 
  hide details  
Reply
post #35 of 39
Quote:
Originally Posted by MGX1016 View Post
Well from some review sites I've read just random google searches.. I can't seem to find a good review of the e21 compared to the X2
Here's one that compares the Pentium E2160 Vs X2 3800+

Quote:
Originally Posted by MGX1016 View Post
SuperPi relies heavily on the Cache and has always worked better for Intel.
SuperPi sees benefits in Intel systems when there is more cache available, though this is simply down to the fact that Core 2's architecture utilise an FSB interface between the CPU and main system RAM, which is actually a hinderance. K8 CPU's have less need for high levels of cache as they are connected directly to the RAM via the HTT and do not require a dedicated MCH.

Core 2's superior performance in SuperPi is purely down to the fact that its architecture is superior, end of. Though bear in mind that although K8 CPU's perform in a disappointing manner in SuperPi, it doesn't mean that they see the same level of lower performance in other applications. As a general rule of thumb, K8 CPU's at 3.0Ghz perform roughly on par with a Core 2 at 2.4Ghz, in most mainstream applications. Core 2 is more akin to 20% faster in most applications, clock per clock, but its most certainly not 50% to 70% faster.

Quote:
Originally Posted by MGX1016
DDR3? Meh..
I'm thinking the same thing about DDR3 myself. I'll be making a thread tomorrow with an explanation as to why...
post #36 of 39
Thread Starter 
Quote:
Originally Posted by Cyant View Post
Alright will do that.




2.87 x 18.533 = 53.18971 exactly in line with the previous times I posted. The ratio Hz:Time is exactly in line with the other Core 2 CPU's whatever the FSB is at. This prove what I said earlier.

I am sure that my 15.250 sec score at 3.4Ghz fit too. Lets calculate it shall we? 3.4 x 15.250 = 51.85 still within the margin of error I would say. That is because the gain per Mhz in Super Pi is linear. It's linear because it almost exclusively give the CPU raw power and is not affected by other bottlenecks.

Okay I'll get back to you with this in a few.. Just a question if you can run that again but with your RAM at DDR2 800. 5-5-5-15-2T-19
How does that affect the score?
Workoholic REborn
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 2600K P8Z68-V PRO NVIDIA GeForce GTX 1080 Ti 12GB 1x4GB+1x8GB 
Hard DriveHard DriveOptical DriveCooling
Intel 520 Series WD Black ASUS DVD+RW Sunbeam Twister 120 
OSMonitorMonitorKeyboard
Windows 7 Ultimate LG OLED65C7P Oculus Rift Logitech K400 
PowerCaseMouse
Corsair 620 Modular Lian Li V1020B G9x 
  hide details  
Reply
Workoholic REborn
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 2600K P8Z68-V PRO NVIDIA GeForce GTX 1080 Ti 12GB 1x4GB+1x8GB 
Hard DriveHard DriveOptical DriveCooling
Intel 520 Series WD Black ASUS DVD+RW Sunbeam Twister 120 
OSMonitorMonitorKeyboard
Windows 7 Ultimate LG OLED65C7P Oculus Rift Logitech K400 
PowerCaseMouse
Corsair 620 Modular Lian Li V1020B G9x 
  hide details  
Reply
post #37 of 39
Thread Starter 
AMD ..

Loses..
Workoholic REborn
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 2600K P8Z68-V PRO NVIDIA GeForce GTX 1080 Ti 12GB 1x4GB+1x8GB 
Hard DriveHard DriveOptical DriveCooling
Intel 520 Series WD Black ASUS DVD+RW Sunbeam Twister 120 
OSMonitorMonitorKeyboard
Windows 7 Ultimate LG OLED65C7P Oculus Rift Logitech K400 
PowerCaseMouse
Corsair 620 Modular Lian Li V1020B G9x 
  hide details  
Reply
Workoholic REborn
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 2600K P8Z68-V PRO NVIDIA GeForce GTX 1080 Ti 12GB 1x4GB+1x8GB 
Hard DriveHard DriveOptical DriveCooling
Intel 520 Series WD Black ASUS DVD+RW Sunbeam Twister 120 
OSMonitorMonitorKeyboard
Windows 7 Ultimate LG OLED65C7P Oculus Rift Logitech K400 
PowerCaseMouse
Corsair 620 Modular Lian Li V1020B G9x 
  hide details  
Reply
post #38 of 39
Quote:
Originally Posted by MGX1016 View Post
Okay I'll get back to you with this in a few.. Just a question if you can run that again but with your RAM at DDR2 800. 5-5-5-15-2T-19
How does that affect the score?


4-4-4-8 1T is a whole 2.69% faster in SuperPi than 5-5-5-15 2T

This prove that SuperPI is barely affected by memory I/O and rely on CPU efficiency at number crunching power instead.

The deficiencies of the old Athlon 64 architecture are offset in many tests where the memory bandwidth is equally important as the number crunching power of the CPU. On the other side the Core 2 is handicaped by the I/O in many applications.

By raising the available I/O bandwidth on the Athlon there won't be much gains because the CPU architecture is already at it's top.

A 2.4 Ghz Athlon 64 X2 with a HT of 1000 vs a 2.4Ghz athlon 64 X2 with a HT of 2000 won't have much difference in applications even if they require a lot of bandwidth.

That is why the Barcelona is important for AMD. They have the best FSB technology but the CPU power is now trailing behind. The barcelona will change this by making the CPU core much more efficient which coupled with the HT bus will have great performance.

By raising the FSB on the Core 2 Architecture you net a lot of gain because the CPU power is strangled by the inneficient I/O speed. So by releasing higher FSB CPU's there is still gains to be made.

A 2.4 Ghz Core 2 CPU with a FSB of 1066 vs a 2.4 Ghz Core with a FSB 1600 will be much slower in applications requiring a lot of bandwidth.
Cerebro
(13 items)
 
  
CPUMotherboardGraphicsRAM
C2D E6420 3.4Ghz (8x425FSB 1:1) Evga 680i SLI NF68-T1 - P31 Asus Geforce 8800GTX OCZ Titanium Alpha VX2 + Crucial Ballistix PC8500 
Hard DriveOptical DriveOSMonitor
7 Hard Drives - All 7200RPM External LG DVD-Writer 16x via Firewire Windows XP SP2 Compaq V1000 21" CRT 
KeyboardPowerCaseMouse
Logitech G15 System - Enermax 565W / Hard Drives - Enermax 350W Thermaltake Mozart TX Logitech MX1000 
  hide details  
Reply
Cerebro
(13 items)
 
  
CPUMotherboardGraphicsRAM
C2D E6420 3.4Ghz (8x425FSB 1:1) Evga 680i SLI NF68-T1 - P31 Asus Geforce 8800GTX OCZ Titanium Alpha VX2 + Crucial Ballistix PC8500 
Hard DriveOptical DriveOSMonitor
7 Hard Drives - All 7200RPM External LG DVD-Writer 16x via Firewire Windows XP SP2 Compaq V1000 21" CRT 
KeyboardPowerCaseMouse
Logitech G15 System - Enermax 565W / Hard Drives - Enermax 350W Thermaltake Mozart TX Logitech MX1000 
  hide details  
Reply
post #39 of 39
Thread Starter 
Quote:
Originally Posted by Cyant View Post


4-4-4-8 1T is a whole 2.69% faster in SuperPi than 5-5-5-15 2T

This prove that SuperPI is barely affected by memory I/O and rely on CPU efficiency at number crunching power instead.

The deficiencies of the old Athlon 64 architecture are offset in many tests where the memory bandwidth is equally important as the number crunching power of the CPU. On the other side the Core 2 is handicaped by the I/O in many applications.

By raising the available I/O bandwidth on the Athlon there won't be much gains because the CPU architecture is already at it's top.

A 2.4 Ghz Athlon 64 X2 with a HT of 1000 vs a 2.4Ghz athlon 64 X2 with a HT of 2000 won't have much difference in applications even if they require a lot of bandwidth.

That is why the Barcelona is important for AMD. They have the best FSB technology but the CPU power is now trailing behind. The barcelona will change this by making the CPU core much more efficient which coupled with the HT bus will have great performance.

By raising the FSB on the Core 2 Architecture you net a lot of gain because the CPU power is strangled by the inneficient I/O speed. So by releasing higher FSB CPU's there is still gains to be made.

A 2.4 Ghz Core 2 CPU with a FSB of 1066 vs a 2.4 Ghz Core with a FSB 1600 will be much slower in applications requiring a lot of bandwidth.
Interesting.. In all honsety though... I have used some C2D systems and real world.. I don't see a difference.. Mostly due to the 880GTS...
Workoholic REborn
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 2600K P8Z68-V PRO NVIDIA GeForce GTX 1080 Ti 12GB 1x4GB+1x8GB 
Hard DriveHard DriveOptical DriveCooling
Intel 520 Series WD Black ASUS DVD+RW Sunbeam Twister 120 
OSMonitorMonitorKeyboard
Windows 7 Ultimate LG OLED65C7P Oculus Rift Logitech K400 
PowerCaseMouse
Corsair 620 Modular Lian Li V1020B G9x 
  hide details  
Reply
Workoholic REborn
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i7 2600K P8Z68-V PRO NVIDIA GeForce GTX 1080 Ti 12GB 1x4GB+1x8GB 
Hard DriveHard DriveOptical DriveCooling
Intel 520 Series WD Black ASUS DVD+RW Sunbeam Twister 120 
OSMonitorMonitorKeyboard
Windows 7 Ultimate LG OLED65C7P Oculus Rift Logitech K400 
PowerCaseMouse
Corsair 620 Modular Lian Li V1020B G9x 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel Motherboards
Overclock.net › Forums › Intel › Intel Motherboards › Are all Intel boards much more costly than AM2 boards?