Overclock.net › Forums › AMD › AMD Memory › So I adjusted my memory timings...
New Posts  All Forums:Forum Nav:

So I adjusted my memory timings...

post #1 of 17
6/17/12 at 7:35am
Thread Starter 
The memory score jumped from 7.3 to 7.8!

WHAT?

I didn't know it was possible to get a 7.8 with a 4 GB kit that I bought like two years ago.

What's the point of overclocking the memory clock? I went from 1.3 GHz to 1.9 GHz and I jumped from 7.3 to 7.4. lol >.>
Edited by tahayassen - 6/17/12 at 7:38am
Main Rig
(6 items)
  		  
CPUMotherboardGraphicsRAM
960T ASUS M4A785TD-V EVO Galaxy GTX 560 Ti GC G.SKILL 
Hard DrivePower
Western Digital XFX Core Edition 550W 
View all
  hide details  
Reply
Main Rig
(6 items)
  		  
CPUMotherboardGraphicsRAM
960T ASUS M4A785TD-V EVO Galaxy GTX 560 Ti GC G.SKILL 
Hard DrivePower
Western Digital XFX Core Edition 550W 
View all
  hide details  
Reply
post #2 of 17
6/17/12 at 11:21am
The WEI score is nearly useless. It is intended to help uninformed, non-technical users determine whether or not the software they want to buy will run well on their computer.

http://en.wikipedia.org/wiki/Windows_Experience_Index
Kota's Rig
(8 items)
  		  
CPUMotherboardGraphicsRAM
Intel C2Q X3380 @ 4.0 GHz Foxconn BLACKOPS X48 XFX 7870 Double D Black Edition 2 GB 8 GB G.Skill RipjawsX DDR3-2133 
Hard DriveCoolingPowerCase
Samsung 840 250GB Corsair Hydro H50 Push/Pull Corsair HX620 Silver AeroCool Masstige (Modded) 
View all
  hide details  
Reply
Kota's Rig
(8 items)
  		  
CPUMotherboardGraphicsRAM
Intel C2Q X3380 @ 4.0 GHz Foxconn BLACKOPS X48 XFX 7870 Double D Black Edition 2 GB 8 GB G.Skill RipjawsX DDR3-2133 
Hard DriveCoolingPowerCase
Samsung 840 250GB Corsair Hydro H50 Push/Pull Corsair HX620 Silver AeroCool Masstige (Modded) 
View all
  hide details  
Reply
post #3 of 17
6/17/12 at 11:54am
Quote:
Originally Posted by Kota View Post

The WEI score is useless.

Fixed.

Sent from my GT-I9100 using Tapatalk 2
Citra ♥
(15 items)
 
Big Mac
(10 items)
  		 
CPUMotherboardGraphicsRAM
[Intel] Core i7 2600K [Asus] P8Z68-V Pro [Gainward] Geforce 9800 GTX+ [Corsair] Vengeance 16GB @ 1600 
Hard DriveHard DriveCoolingOS
[Kingston] HyperX 120GB SSD [Intel] 520 series 60GB [Corsair] H80 [Canonical] Ubuntu 12.10 
MonitorKeyboardPowerCase
[BenQ] XL2420T [Corsair] Vengeance K60 Performance [OCZ] ZX 850W [CM] Storm Enforcer 
MouseAudioOther
[Razer] Mamba 4G 2012 [Beats] urBeats [NZXT] 24/8/6 pin extensions 
View all
CPUMotherboardGraphicsRAM
[Intel] Core i5-3210M [Apple] Proprietary [Intel] HD 4000 [Hynix] H5TC2G83CFR DDR3L 8GB 
Hard DriveOSMonitorKeyboard
[Samsung] 128GB SSD [Apple] OS X 10.8.2 ML 13" 2560x1600 Retina [Apple] Backlit Keyboard 
CaseMouse Pad
[Apple] Unibody [Apple] Trackpad 
View all
  hide details  
Reply
Citra ♥
(15 items)
 
Big Mac
(10 items)
  		 
CPUMotherboardGraphicsRAM
[Intel] Core i7 2600K [Asus] P8Z68-V Pro [Gainward] Geforce 9800 GTX+ [Corsair] Vengeance 16GB @ 1600 
Hard DriveHard DriveCoolingOS
[Kingston] HyperX 120GB SSD [Intel] 520 series 60GB [Corsair] H80 [Canonical] Ubuntu 12.10 
MonitorKeyboardPowerCase
[BenQ] XL2420T [Corsair] Vengeance K60 Performance [OCZ] ZX 850W [CM] Storm Enforcer 
MouseAudioOther
[Razer] Mamba 4G 2012 [Beats] urBeats [NZXT] 24/8/6 pin extensions 
View all
CPUMotherboardGraphicsRAM
[Intel] Core i5-3210M [Apple] Proprietary [Intel] HD 4000 [Hynix] H5TC2G83CFR DDR3L 8GB 
Hard DriveOSMonitorKeyboard
[Samsung] 128GB SSD [Apple] OS X 10.8.2 ML 13" 2560x1600 Retina [Apple] Backlit Keyboard 
CaseMouse Pad
[Apple] Unibody [Apple] Trackpad 
View all
  hide details  
Reply
post #4 of 17
6/17/12 at 11:58am
WEI isn't an accurate test for rating your RAM. Pure speed is almost always better.

If you actually got that kind of increase from adjusting your timings (it's probably just WEI being stupid):
1) Your timings were way off
2) You adjusted them to the point where they aren't stable. Make sure you run memtest for 3-4 full loops. I've seen people get very tight timings that seem like their stable, but they are silently corrupting data.
Edited by Murlocke - 6/17/12 at 12:01pm
The Leviathan
(21 items)
 
43TB unRAID Server
(17 items)
 
40TB unRAID Server
(18 items)
 
CPUMotherboardGraphicsRAM
Intel i7 3770k (4.5GHz) ASUS P8Z77-V Deluxe 6GB EVGA GeForce GTX Titan (1164/6500) 16GB G.Skill Sniper (DDR3 1866) 
Hard DriveHard DriveOptical DriveCooling
256GB Crucial M4 2TB Western Digital Black (WD2002FAEX) LG WH12LS39 3x 120mm Noiseblocker PL-2  
CoolingOSMonitorKeyboard
Noctua NH-D14 Windows 7 Ultimate x64 55" Sony HX929 Ducky Shine DK9008S 
PowerCaseMouseMouse Pad
Corsair AX1200 Corsair Carbide 500R Logitech G700 SteelPad S&S Solo 
AudioAudioAudioAudio
Denon 4311CI (Receiver) 2x Klipsch RF-82 (Front Speakers) 4x Klipsch RS-52 (Surround Speakers) Klipsch RC-62 (Center Speaker) 
Audio
SVS 20-39PCi (Subwoofer) 
View all
CPUMotherboardRAMHard Drive
Intel i3-2120 (3.3GHz) Supermicro MBD-X9SCM-F-O 8GB Kingston (DDR3 1333) 14x 3TB Western Digital Greens (WD30EZRX) 
Hard DriveCoolingCoolingOS
1TB Western Digital Black (WD1002FAEX) 3x Noctua NF-P12 2x Noctua NF-R8 unRAID 5.0 
PowerCaseOther
Corsair AX850 Norco 4224 3x Supermicro AOC-SAS2LP-MV8  
View all
CPUMotherboardRAMHard Drive
Intel i3-2120 (3.3GHz) Supermicro MBD-X9SCM-F-O 8GB Kingston (DDR3 1333) 5x 3TB Western Digital Greens (WD30EZRX) 
Hard DriveHard DriveCoolingCooling
12x 2TB Western Digital Greens (WD20EARS) 1TB Western Digital Black (WD1002FAEX) 3x Noctua NF-P12 2x Noctua NF-R8 
OSPowerCaseOther
unRAID 5.0 Corsair AX850 Norco 4224 3x Supermicro AOC-SAS2LP-MV8  
View all
  hide details  
Reply
The Leviathan
(21 items)
 
43TB unRAID Server
(17 items)
 
40TB unRAID Server
(18 items)
 
CPUMotherboardGraphicsRAM
Intel i7 3770k (4.5GHz) ASUS P8Z77-V Deluxe 6GB EVGA GeForce GTX Titan (1164/6500) 16GB G.Skill Sniper (DDR3 1866) 
Hard DriveHard DriveOptical DriveCooling
256GB Crucial M4 2TB Western Digital Black (WD2002FAEX) LG WH12LS39 3x 120mm Noiseblocker PL-2  
CoolingOSMonitorKeyboard
Noctua NH-D14 Windows 7 Ultimate x64 55" Sony HX929 Ducky Shine DK9008S 
PowerCaseMouseMouse Pad
Corsair AX1200 Corsair Carbide 500R Logitech G700 SteelPad S&S Solo 
AudioAudioAudioAudio
Denon 4311CI (Receiver) 2x Klipsch RF-82 (Front Speakers) 4x Klipsch RS-52 (Surround Speakers) Klipsch RC-62 (Center Speaker) 
Audio
SVS 20-39PCi (Subwoofer) 
View all
CPUMotherboardRAMHard Drive
Intel i3-2120 (3.3GHz) Supermicro MBD-X9SCM-F-O 8GB Kingston (DDR3 1333) 14x 3TB Western Digital Greens (WD30EZRX) 
Hard DriveCoolingCoolingOS
1TB Western Digital Black (WD1002FAEX) 3x Noctua NF-P12 2x Noctua NF-R8 unRAID 5.0 
PowerCaseOther
Corsair AX850 Norco 4224 3x Supermicro AOC-SAS2LP-MV8  
View all
CPUMotherboardRAMHard Drive
Intel i3-2120 (3.3GHz) Supermicro MBD-X9SCM-F-O 8GB Kingston (DDR3 1333) 5x 3TB Western Digital Greens (WD30EZRX) 
Hard DriveHard DriveCoolingCooling
12x 2TB Western Digital Greens (WD20EARS) 1TB Western Digital Black (WD1002FAEX) 3x Noctua NF-P12 2x Noctua NF-R8 
OSPowerCaseOther
unRAID 5.0 Corsair AX850 Norco 4224 3x Supermicro AOC-SAS2LP-MV8  
View all
  hide details  
Reply
post #5 of 17
6/17/12 at 12:58pm
Thread Starter 
Should I do:

1.9 GHz 9-9-9-24-33 (timings were on auto)
or
1.6 GHz 6-8-5-16-25

Also, I'll be sure to run memtest.
Main Rig
(6 items)
  		  
CPUMotherboardGraphicsRAM
960T ASUS M4A785TD-V EVO Galaxy GTX 560 Ti GC G.SKILL 
Hard DrivePower
Western Digital XFX Core Edition 550W 
View all
  hide details  
Reply
Main Rig
(6 items)
  		  
CPUMotherboardGraphicsRAM
960T ASUS M4A785TD-V EVO Galaxy GTX 560 Ti GC G.SKILL 
Hard DrivePower
Western Digital XFX Core Edition 550W 
View all
  hide details  
Reply
post #6 of 17
6/17/12 at 3:26pm
  • xd_1771
  • Mess with the goat, get the horns
  • Forum Moderator

Timings can affect bandwidth on AMD systems but that doesn't mean that such large performance increases are necessarily reflected.  For one WEI is not very accurate, random occurences can affect test scores.  For the best results, try something like MaxxMem to compare bandwidth.  The results on 1600Mhz 6-8-6 are probably fairly close to 1900Mhz 9-9-9 but there may be variances in bandwidth vs. latency.  I guess go with whatever provides the best of the result that matters to you.

Hexagonal
(16 items)
 
Voyager [XT875]
(11 items)
  		 
CPUMotherboardGraphicsRAM
AMD Phenom II x6 1055T [4GHz, 300x13.5, 1.42V] Gigabyte GA-990XA-UD3 AM3+ [NB 3ghz 1.35v] MSI GeForce GTX 560 Ti Twin Frozr II 1GB 8GB [4x2GB] G.Skill ECO DDR3 [1600 6-8-6-20-24-1T] 
Hard DriveHard DriveHard DriveCooling
80GB Intel X25M SSD WD 1TB Caviar Black (WD1001FALS) WD 1TB Caviar Black (WD1002FAEX) Noctua NH-D14 
CoolingOSMonitorPower
Scythe GT-15 Gentle Typhoon 1500RPM fans (2x, o... Windows 8 Pro 64-bit ASUS PA238QR 23.6" 1920x1080 IPS LED monitor [x2] Corsair HX520 Modular PSU [520W] 
CaseAudioAudio
Antec P182 SE ASUS Xonar DG 5.1 Headphone Amp Audio Technica ATH-d40fs 
View all
CPUMotherboardGraphicsRAM
ARM Cortex A9 dual core [1.2Ghz]  TI OMAP 4430 PowerVR SGX 540 1GB low power DDR2 
Hard DriveHard DriveOSMonitor
16GB on board storage 32GB Class 10 MicroSDHC card Android 4.0.4 "Ice Cream Sandwich" (LiquidSmoot... 4.3" 960x540 TFT capacitive multi-touch screen 
KeyboardPowerCase
Swiftkey 3 Battery Motorola DROID BIONIC 
View all
  hide details  
Reply
Hexagonal
(16 items)
 
Voyager [XT875]
(11 items)
  		 
CPUMotherboardGraphicsRAM
AMD Phenom II x6 1055T [4GHz, 300x13.5, 1.42V] Gigabyte GA-990XA-UD3 AM3+ [NB 3ghz 1.35v] MSI GeForce GTX 560 Ti Twin Frozr II 1GB 8GB [4x2GB] G.Skill ECO DDR3 [1600 6-8-6-20-24-1T] 
Hard DriveHard DriveHard DriveCooling
80GB Intel X25M SSD WD 1TB Caviar Black (WD1001FALS) WD 1TB Caviar Black (WD1002FAEX) Noctua NH-D14 
CoolingOSMonitorPower
Scythe GT-15 Gentle Typhoon 1500RPM fans (2x, o... Windows 8 Pro 64-bit ASUS PA238QR 23.6" 1920x1080 IPS LED monitor [x2] Corsair HX520 Modular PSU [520W] 
CaseAudioAudio
Antec P182 SE ASUS Xonar DG 5.1 Headphone Amp Audio Technica ATH-d40fs 
View all
CPUMotherboardGraphicsRAM
ARM Cortex A9 dual core [1.2Ghz]  TI OMAP 4430 PowerVR SGX 540 1GB low power DDR2 
Hard DriveHard DriveOSMonitor
16GB on board storage 32GB Class 10 MicroSDHC card Android 4.0.4 "Ice Cream Sandwich" (LiquidSmoot... 4.3" 960x540 TFT capacitive multi-touch screen 
KeyboardPowerCase
Swiftkey 3 Battery Motorola DROID BIONIC 
View all
  hide details  
Reply
post #7 of 17
6/17/12 at 7:20pm
Thread Starter 
Quote:
Originally Posted by xd_1771 View Post

Timings can affect bandwidth on AMD systems but that doesn't mean that such large performance increases are necessarily reflected.  For one WEI is not very accurate, random occurences can affect test scores.  For the best results, try something like MaxxMem to compare bandwidth.  The results on 1600Mhz 6-8-6 are probably fairly close to 1900Mhz 9-9-9 but there may be variances in bandwidth vs. latency.  I guess go with whatever provides the best of the result that matters to you.

After some stability testing, 1600 MHz resulted in: 6-8-7-16-27. I'm going to benchmark this, switch to 1900 MHz and see if I can lower the timings and then benchmark again. I'll compare the results and post back here.
Main Rig
(6 items)
  		  
CPUMotherboardGraphicsRAM
960T ASUS M4A785TD-V EVO Galaxy GTX 560 Ti GC G.SKILL 
Hard DrivePower
Western Digital XFX Core Edition 550W 
View all
  hide details  
Reply
Main Rig
(6 items)
  		  
CPUMotherboardGraphicsRAM
960T ASUS M4A785TD-V EVO Galaxy GTX 560 Ti GC G.SKILL 
Hard DrivePower
Western Digital XFX Core Edition 550W 
View all
  hide details  
Reply
post #8 of 17
6/17/12 at 7:33pm
i followed the link for your ram and this what i found . . .

http://www.overclock.net/products/g-skill-ripjaw-x-4gb-2-x-2gb-7-8-7-24-1-5v

a lot of info there. you've got a good set there.
Cheap Grade
(14 items)
 
Second Intel Rig
(16 items)
  		 
CPUMotherboardGraphicsRAM
1600 4.0 / 1.4 880 7770 4 / 1600 
Hard DriveCoolingOSMonitor
96 / 250 120 7 64 19 / 40 1080 
PowerCase
550 430 
View all
CPUMotherboardGraphicsRAM
2700 4.5/ 1.28 77 7970 8 / 1600 
Hard DriveCoolingOSMonitor
500 120 (2) 7 64 1 
PowerCase
700 912 
View all
  hide details  
Reply
Cheap Grade
(14 items)
 
Second Intel Rig
(16 items)
  		 
CPUMotherboardGraphicsRAM
1600 4.0 / 1.4 880 7770 4 / 1600 
Hard DriveCoolingOSMonitor
96 / 250 120 7 64 19 / 40 1080 
PowerCase
550 430 
View all
CPUMotherboardGraphicsRAM
2700 4.5/ 1.28 77 7970 8 / 1600 
Hard DriveCoolingOSMonitor
500 120 (2) 7 64 1 
PowerCase
700 912 
View all
  hide details  
Reply
post #9 of 17
6/23/12 at 5:35am
download MaxxMEM2. its a great RAM benchmarking software.
My Black and Red Rig
(20 items)
  		  
CPUMotherboardGraphicsRAM
Intel 3930K 4.5GHz @ 1.34V Asus rampage extreme IV Shaphire 6950 unlocked to 6970 Corsair Dominators GT 16GB 4GBX4 OCed @ 2400MHz... 
Hard DriveHard DriveOptical DriveCooling
Intel SSD 330 series 180GB WD Black 1TB 64Mb Cache + Blue 500GB 16mb Cache... LG DVD EK-Supreme HF - EN (Nickel) 
CoolingCoolingCoolingCooling
EK-FB KIT RE4 - Acetal Laing D5 Vario 12V DC Pump (MCP 655)  EK-BAY SPIN Reservoir - Plexi EK-CoolStream RAD XT (240) 
CoolingOSMonitorKeyboard
EK-CoolStream RAD XTX (120) Windows 7 64-bit LG W2261 22inch 1920X1080@60Hz Razer Lycosa 
PowerCase
Cooler Master Silent Pro Gold 1000W Thermaltake Level 10 GT 
View all
  hide details  
Reply
My Black and Red Rig
(20 items)
  		  
CPUMotherboardGraphicsRAM
Intel 3930K 4.5GHz @ 1.34V Asus rampage extreme IV Shaphire 6950 unlocked to 6970 Corsair Dominators GT 16GB 4GBX4 OCed @ 2400MHz... 
Hard DriveHard DriveOptical DriveCooling
Intel SSD 330 series 180GB WD Black 1TB 64Mb Cache + Blue 500GB 16mb Cache... LG DVD EK-Supreme HF - EN (Nickel) 
CoolingCoolingCoolingCooling
EK-FB KIT RE4 - Acetal Laing D5 Vario 12V DC Pump (MCP 655)  EK-BAY SPIN Reservoir - Plexi EK-CoolStream RAD XT (240) 
CoolingOSMonitorKeyboard
EK-CoolStream RAD XTX (120) Windows 7 64-bit LG W2261 22inch 1920X1080@60Hz Razer Lycosa 
PowerCase
Cooler Master Silent Pro Gold 1000W Thermaltake Level 10 GT 
View all
  hide details  
Reply
post #10 of 17
6/23/12 at 9:38am
Quote:
Originally Posted by tahayassen View Post

After some stability testing, 1600 MHz resulted in: 6-8-7-16-27. I'm going to benchmark this, switch to 1900 MHz and see if I can lower the timings and then benchmark again. I'll compare the results and post back here.

FYI- Usually you are limited in reducing latency as the frequency increases because one clock cycle in real time decreases as frequency increases. So you may or may not be able to lower the latency when you increase the frequency but the "real time" of the latency settings will still be reduced slightly. Often you actually need to increase the latency to run the RAM stable at a higher frequency.

It's worth noting that synthetic benches show theoretical performance changes which can be much different than how actual applications perform with OC'd RAM.
Edited by AMD4ME - 6/23/12 at 1:42pm
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD Memory
Overclock.net › Forums › AMD › AMD Memory › So I adjusted my memory timings...