Overclock.net › Forums › Intel › Intel CPUs › Help Getting P4 650 Stable
New Posts  All Forums:Forum Nav:

Help Getting P4 650 Stable - Page 5

post #41 of 48
Quote:
Originally Posted by MaFi0s0
Its higher frequency of course is gonna be faster, I have researched it and posted it in this thread:
http://www.overclock.net/intel-memor...ps-tested.html

When I set timings to 1:1 my computer ran applications like a dog.
No, higher frequency does not equate to faster. If your fsb is 800 and your ram is 900 there are (other than for ECC) unused cycles and this does NOT equate to faster.

Now:
Quote:
Originally Posted by MaFi0s0
I tested what effect the RAM frequency would have in the stress test, and discovered it could well be part of the bottleneck along with CPU.
I ran each test twice and under the same conditions:
550+Megs of RAM free
256 meg Heapsize
fps_max 300
steam.exe set to low priority
hl2.exe set to high priority
hyper threading off (turning it off improves your frame rate btw)
EIST off
TM2 off
C1E off

My relative system specs are: FSB 211x18(3.8GHz) Corsair 2x512MB Twin 8000UL RAM @ Dual channel. (this RAM is capable of running at 1Ghz)
I will start with my results under stock frequency wich at the time was 3.33x211 (702MHz)
154.17 fps average
153.98 fps average
Now with just a slightly higher freq 844Mhz (5-5-5-15)
158.48 fps average
158.55 fps average

Keep in mind these figures also affected the FPS's lows too, not just the highs.

Now to confirm that the RAM did indeed make a difrence, I ran the test with the lowest freq I could get, wich was 316Mhz (4-3-3-9)
135.31 fps average
134.41 fps average
I reiterate. This is NOT scientific proof. I want scientific data proving this phenomina. If it were true then it would be out there all over the web. Do you think that if this were the case there would not be data proving it? On X-Bit labs etc? This is one person testing their system and scientific proof is replicated. I tested and proved no such replication. Read to my previous post on this thread.

R
Roped In
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2600k P8P67 Pro Rev (3.1) @ B3 XFX Radeon HD 6950 XXX 16GB G.SKILL Ripjaws 9-11-9-28 
Hard DriveOptical DriveOSMonitor
2 X Callisto Deluxe 25nm 60GB Plextor PX-B910SA 4x Blu-ray DVD-RW Win7 64 2 x Samsung 275T+ 
KeyboardPowerCaseMouse
Logitech G15 Corsair HX10000 Antec 1200 Logitech G7 
Mouse Pad
Splatter Game Pad 
  hide details  
Reply
Roped In
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2600k P8P67 Pro Rev (3.1) @ B3 XFX Radeon HD 6950 XXX 16GB G.SKILL Ripjaws 9-11-9-28 
Hard DriveOptical DriveOSMonitor
2 X Callisto Deluxe 25nm 60GB Plextor PX-B910SA 4x Blu-ray DVD-RW Win7 64 2 x Samsung 275T+ 
KeyboardPowerCaseMouse
Logitech G15 Corsair HX10000 Antec 1200 Logitech G7 
Mouse Pad
Splatter Game Pad 
  hide details  
Reply
post #42 of 48
When I had my 660 & 650 I had the vcore as high as 1.5 vcore with no problem, the 660 had clocked to 4.4ghz and the 650 to 4.1ghz stable and the temps never went past 46c under load with the typhoon.
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
FX-57 lapped DFI Lanparty NF4 SLI DR 2x7800GT OC 2x512 PNY DDR600 
Hard DriveOSMonitorKeyboard
4x36gb Raptors raid-0 2000 pro HP P1100 20 inch Micro wireless 
PowerCaseMouse
Antic 550 True Control TT Armor Ge optical 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardGraphicsRAM
FX-57 lapped DFI Lanparty NF4 SLI DR 2x7800GT OC 2x512 PNY DDR600 
Hard DriveOSMonitorKeyboard
4x36gb Raptors raid-0 2000 pro HP P1100 20 inch Micro wireless 
PowerCaseMouse
Antic 550 True Control TT Armor Ge optical 
  hide details  
Reply
post #43 of 48
Thread Starter 
Thanks for the info jrabb1920.
My System
(13 items)
 
  
CPUMotherboardOSMonitor
P4 650 ASUS P5WD2-P XP Pro Sp2 Samsung 19" LCD 
PowerCase
400W Antec Antec Performance TX640B 
  hide details  
Reply
My System
(13 items)
 
  
CPUMotherboardOSMonitor
P4 650 ASUS P5WD2-P XP Pro Sp2 Samsung 19" LCD 
PowerCase
400W Antec Antec Performance TX640B 
  hide details  
Reply
post #44 of 48
why doesnt he just get GOOD cooling, and really REALLY push the FSB
then he could run divider to closer speeds(5:6),
or even try 1:1 when he gets close to the RAM's defaults speed
Finaly Dual Core
(13 items)
 
  
CPUMotherboardGraphicsRAM
Pentium D 805 4Gz ABIT AW9D-MAX Saphire x1950 Pro 512 4x512 Buffalo Firestix 
Hard DriveOptical DriveOSMonitor
160 SATAII 2x30 IDE RAID Samsung DVD-RW e / XP Pro / Ubuntu 7.04 Viewsonic 19" Pro 
KeyboardPowerCaseMouse
Generic 500 generic 20amp 12v My Sexy Baby Inland Laser Mouse 
Mouse Pad
My Desk 
  hide details  
Reply
Finaly Dual Core
(13 items)
 
  
CPUMotherboardGraphicsRAM
Pentium D 805 4Gz ABIT AW9D-MAX Saphire x1950 Pro 512 4x512 Buffalo Firestix 
Hard DriveOptical DriveOSMonitor
160 SATAII 2x30 IDE RAID Samsung DVD-RW e / XP Pro / Ubuntu 7.04 Viewsonic 19" Pro 
KeyboardPowerCaseMouse
Generic 500 generic 20amp 12v My Sexy Baby Inland Laser Mouse 
Mouse Pad
My Desk 
  hide details  
Reply
post #45 of 48
Quote:
Originally Posted by Ropey
No, higher frequency does not equate to faster. If your fsb is 800 and your ram is 900 there are (other than for ECC) unused cycles and this does NOT equate to faster.

Now:


I reiterate. This is NOT scientific proof. I want scientific data proving this phenomina. If it were true then it would be out there all over the web. Do you think that if this were the case there would not be data proving it? On X-Bit labs etc? This is one person testing their system and scientific proof is replicated. I tested and proved no such replication. Read to my previous post on this thread.

R
True it was not scientific proof but I thought I would show you atleast an example.
Did you mean phenomina or rumour? I would like to hear your views on why you believe higher freq is not faster.
post #46 of 48
Quote:
Originally Posted by MaFi0s0
True it was not scientific proof but I thought I would show you atleast an example.
Did you mean phenomina or rumour? I would like to hear your views on why you believe higher freq is not faster.
All you have to do is read this entire thread carefully and you will see my views.

Basically this is not addition of velocities as it is a two way transfer.

Now a stream of data sent at 800MHz to 1000Mhz (for theories sake) returns a stream of data at 1000MHz returned and dropped to 800MHz will bring equality, not increase and this has been proven scientifically with DDR.

Until there is scientific proven data showing this is not the case I will not believe it to be fact.

R
Roped In
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2600k P8P67 Pro Rev (3.1) @ B3 XFX Radeon HD 6950 XXX 16GB G.SKILL Ripjaws 9-11-9-28 
Hard DriveOptical DriveOSMonitor
2 X Callisto Deluxe 25nm 60GB Plextor PX-B910SA 4x Blu-ray DVD-RW Win7 64 2 x Samsung 275T+ 
KeyboardPowerCaseMouse
Logitech G15 Corsair HX10000 Antec 1200 Logitech G7 
Mouse Pad
Splatter Game Pad 
  hide details  
Reply
Roped In
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2600k P8P67 Pro Rev (3.1) @ B3 XFX Radeon HD 6950 XXX 16GB G.SKILL Ripjaws 9-11-9-28 
Hard DriveOptical DriveOSMonitor
2 X Callisto Deluxe 25nm 60GB Plextor PX-B910SA 4x Blu-ray DVD-RW Win7 64 2 x Samsung 275T+ 
KeyboardPowerCaseMouse
Logitech G15 Corsair HX10000 Antec 1200 Logitech G7 
Mouse Pad
Splatter Game Pad 
  hide details  
Reply
post #47 of 48
If there wasn't ANY benefit from running a 1:1 ratio then there wouldn't be no need for DDR2 533, DDR2 667, DDR2 800, DDR2 1000 and DDR2 1066, everything would've stopped at DDR2 400, right?

So, with the upcoming DDR3 being @ 1333mhz, I wonder why would develop what seems to not have an impact in performance if there's no need for more than DDR2 400?

RAM will be always limited, it may be (which I don't think) that a DDR400 will run 800mhz in Dual Channel but does it go from 3.2GB/s to 6.4GB/s also? I don't think so, it will stay with the same throughput of 3.2GB/s.


CPU : RAM
960mhz : 960mhz (480mhz x2)

Have I seen a performance increase from 1:1 to 3:5 and 1:2, I can say that I have, that's why I'm not using the 1:1, if my 660 would have an IMC, maybe it wouldn't need that but being that the MemHub is out-side the chip, it needs a faster way to communicate through the MemHub and it does when it's upto 960mhz which is more than 6.4GB/s systems runs faster, not screaming fast but fast enough to notice, tangibile? not really, just enough.
post #48 of 48
Quote:
Originally Posted by Nasgul
If there wasn't ANY benefit from running a 1:1 ratio then there wouldn't be no need for DDR2 533, DDR2 667, DDR2 800, DDR2 1000 and DDR2 1066, everything would've stopped at DDR2 400, right?
Of course not. These speeds are for the overclocking enthusiast and will also be for future supported speeds. When you are overclocking the Conroe top monster @ 1333 FSB then you will need those speeds for 1:1. The technology is often on a faster applied track than current needs dictate. As well, when you are running a 1200FSB on water or 1300 on phase you will need those speeds for 1:1. These speeds are for the enthusiast.

Quote:
Originally Posted by Nasgul
tangibile? not really, just enough.
Can you show an Everest Benchmark of your system 1:1 and 1:2 please? Not gaming benchmarks but actual memory subsystem benchmarks. Not Sisoft Sandra as that is not a real subsystem test and is more of an algorithmic based test that has bypassed it's usefullness.

If you need a copy to do the test give me a pm.

R
Roped In
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2600k P8P67 Pro Rev (3.1) @ B3 XFX Radeon HD 6950 XXX 16GB G.SKILL Ripjaws 9-11-9-28 
Hard DriveOptical DriveOSMonitor
2 X Callisto Deluxe 25nm 60GB Plextor PX-B910SA 4x Blu-ray DVD-RW Win7 64 2 x Samsung 275T+ 
KeyboardPowerCaseMouse
Logitech G15 Corsair HX10000 Antec 1200 Logitech G7 
Mouse Pad
Splatter Game Pad 
  hide details  
Reply
Roped In
(13 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 2600k P8P67 Pro Rev (3.1) @ B3 XFX Radeon HD 6950 XXX 16GB G.SKILL Ripjaws 9-11-9-28 
Hard DriveOptical DriveOSMonitor
2 X Callisto Deluxe 25nm 60GB Plextor PX-B910SA 4x Blu-ray DVD-RW Win7 64 2 x Samsung 275T+ 
KeyboardPowerCaseMouse
Logitech G15 Corsair HX10000 Antec 1200 Logitech G7 
Mouse Pad
Splatter Game Pad 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › Help Getting P4 650 Stable