Overclock.net › Forums › Components › Memory › DDR3 Drive Strength and other obscure timings....
New Posts  All Forums:Forum Nav:

DDR3 Drive Strength and other obscure timings....

post #1 of 10
Thread Starter 
So after looking a ton for other people that have fiddled with these settings and finding nothing.... I have took it upon myself to test.

For all my testing I will be using MaxxMem to find the latencies and copy/read/write speeds.

I will be using AMD Overdrive to ONLY show the timings and drive strengths. All my changes are being made in the bios.

The timing on the ram is easy to see as well. That is as low as I can go with my ram.

My ram was sold to me as G.Skill Ripjaw DDR3 1600 CL9. I have 4 of these sticks occupying all slots.

Ok so for my first test I went for an all or nothing approach:

This is the original (AUTO) drive strengths:

1535963

This is with all the drive settings maxed out:

1535970

This is with all the drive settings set to lowest:

1535974

Things to note that the benchmarks don't show:

With the higher strengths the maxxmem results vary a decent amount each run. Never quite getting the same score (might be due to not long enough waiting after boot).

With the lower strengths the maxxmem results are very stable, giving almost exactly the same results each time.
BUT with the lower strengths windows and specifically ASUS Probe act odd and malfunction occasionally giving off readings.

So far my conclusion is that higher drive strength = better stability.....

Let me know what you think, I will be testing some of the other oddball settings shortly.
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
post #2 of 10
How many tests were ran each time?
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
post #3 of 10
Thread Starter 
I ran 4-6 runs each time, capturing the best run.

I will be fiddling with the 'Processor On-die Termination' and 'DRAM Termination' values next.

Found this article on the subject: http://www.techarp.com/showfreebog.aspx?lang=0&bogno=434

Seems to suggest lower ohms is better....


Also this is for the CPU On-Die settings: http://www.techarp.com/showFreeBOG.aspx?lang=0&bogno=437
Edited by rush2049 - 1/4/12 at 4:57pm
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
post #4 of 10
Quote:
Originally Posted by rush2049 View Post

I ran 4-6 runs each time, capturing the best run.

Provide all the numbers... this will provide average and deviation.
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
post #5 of 10
Thread Starter 
Quote:
Originally Posted by DuckieHo View Post

Quote:
Originally Posted by rush2049 View Post

I ran 4-6 runs each time, capturing the best run.

Provide all the numbers... this will provide average and deviation.

I will be re-running the test 5 times for each and making an excel document....

I hope this thread will be informative enough for future overclockers looking for info on this obscure stuff.....

From what I have been reading drive strength adjusts the amperage used when changing voltage, effectively it changes how fast the ram can go from a high to low state and vice-versa. The strengths effect different ram chips and setups differently, but most will prefer a higher drive strength with more ram modules...... (This was gathered from reading about it in relation to servers)



Also from what i understand about the Termination values (measured in ohms). The lower these are the better signal ripple/reflextion it will help to reduce. Especially when using multiple modules. I have yet to find any accepted values for these settings, other than to put them as low as possible and adjust upwards for stability.
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
post #6 of 10
Thread Starter 
Here is some tabulated results.
It still doesn't show that my system is overall more stable at the high drive strengths......
But you can see there is a performance boost with the lower drive strengths.

1536331
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
post #7 of 10
Thread Starter 
There is no option in the crosshair IV bios to change the DRAM Termination value. I will have to fiddle with that in windows....
And my CPU Termination was already at its lowest setting (60 ohms), so I left that as is.

I tested what the Refresh Rate setting does instead.
1536654
1536658

So basically for that setting higher is better......

I also tried changing the Row Refresh Cycle Time to a lower value than 110 ns, but 90 ns would not post. I will try researching some more on the other settings.....
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
post #8 of 10
Drive strengths merely set buffer output impedance on the associated signal lines. This has an impact on waveform integrity, the end result generally interacts with skew between the DQ and DQS lines for logic sampling (these lines toggle the most) and is not going to impact peak bandwidth at a given clock frequency. If logic is seen high when it should be low then the software will simply generate an error or crash.

As we are dealing with waveform integrity, the effects of transmission line impedance and any associated gate capacitance, a higher or lower setting may achieve better stability - there is no law that says "higher = better" when it comes to settings like these. The output impedance of the drivers needs to be as closely matched to the line impedance/input impedance as possible - that's all. Any effects you are seeing in benchmark scores are not due to the drive strength settings (drift is mainly due to cache and open page hits). Bottom line is these settings impact stability, but not the maximum scores of a benchmark at a given clock frequency where all drive settings are stable.

-Raja
Edited by Raja@ASUS - 1/5/12 at 3:31am
post #9 of 10
Thread Starter 
Yes that is what I was understanding about it. A higher strength would equate to the waveform achieving an acceptable '1' state for the ram faster. This could work against a possible skew you might have in your wave or it might just correct for a slight shift due to any number of factors. It would also work for achieving the '0' state faster as well.

I know that higher does not equal better, but in my case it seems to have an effect on stability. And the more stable I can get the ram running the more I can push them.

I re-ran the tests a few times across different boots and every time I see similar numbers to the above with the different strengths. So the stability of the ram is having an effect on the benchmark. While these settings do not increase the performance at all it increases stability. That is the kind of thing I set out to find.

I thank you for taking the time to write such a detailed post. There is no information anywhere about these settings as detailed as that.
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
 
Tragbar
(14 items)
 
 
CPUMotherboardGraphicsRAM
Ryzen 1800X Asus Crosshair VI Hero Nvidia Geforce Titan X F4-3600C15D-16GTZ 
Hard DriveHard DriveHard DriveHard Drive
Combined: 6.13 TB of space Samsung SSD 960 EVO 500GB Crucial_CT1050MX300SSD1 M4-CT128M4SSD2 
CoolingCoolingCoolingCooling
EK-Supremacy EVO White Edition PrimoChill PrimoFlex Advanced LRT EK-RES X4 250 (R2.0) Reservoir EK-CoolStream XE 360 
CoolingCoolingCoolingOS
EK-Vardar F4-120ER (2200rpm)  EK-Ekoolant EVO Liquid Coolant EK-XTOP Revo D5 PWM Pump Windows 10 Pro 
MonitorKeyboardPowerCase
Asus VG278H Ducky YOTG Keyboard Corsair AX1200 Corsair Obsidian 800D 
MouseAudioAudioAudio
Logitech G700s Schiit Gungnir Multibit Schiit Mjolnir 2 Schiit LISST 
Audio
Audeze LCD 2 - pre fazor 
CPUMotherboardGraphicsRAM
intel i7 4770K ASRock Z87E-ITX Nvidia Geforce GTX Titan X GSkill F3-2400C10D-16GTX R 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Silverstone 12.7mm Slot Load Blu-Ray Burner Noctua NH-L12 Windows 8 Pro 
MonitorCaseAudioAudio
LG PA75U Slim LED Projector Silverstone SG08 Schiit Bifrost Schiit Asgard 2 
AudioOther
AKG Q701 Headphones Pelican 1510 Green Case  
  hide details  
Reply
post #10 of 10
Quote:
Originally Posted by rush2049 View Post

Yes that is what I was understanding about it. A higher strength would equate to the waveform achieving an acceptable '1' state for the ram faster. This could work against a possible skew you might have in your wave or it might just correct for a slight shift due to any number of factors. It would also work for achieving the '0' state faster as well.


Reading you post i think I should clarify:

Slew rate is directly proportional to the rail voltage. I would not confuse that too much with drive strength, they do impact slew rate behaviour but should be seen as impacting the waveform integrity (ringing artefacts and such) leading to a cleaner signal eye. Not so much making "faster" ones and zeros.


A higher drive strength does not mean a faster one or zero if the resulting waveform contains large amounts of ringing due to an impedance mismatch...

-Raja
Edited by Raja@ASUS - 1/5/12 at 6:59pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Memory
Overclock.net › Forums › Components › Memory › DDR3 Drive Strength and other obscure timings....