speed vs timing

Willy8382 · 04-14-12, 05:16 AM

Is there some form of rough math that one can do to tell if one ram setting is faster than another. say like 1333@ 7 7 7 vs 1600@ 9 9 9 so you dont have to run bench marks all the time.

redduc900 · 04-14-12, 11:22 AM

There is no formula for calculating transfer rate between different frequencies and timings, although booting to Memtest or Memtest86+ can give you a theoretical peak transfer rate measured in MB/s, which can be used to quickly gauge any DRAM related changes.

trents · 04-14-12, 08:05 PM

The other complication here is that it will vary with the application somewhat. Frequency and timing both affect "speed". It might be better to refer to frequency as bandwidth rather than speed.

Willy8382 · 04-15-12, 06:05 AM

Thanks for the replys. Thought i could get a short cut but I guess its just trail and error.
Part of the fun anyways.

WARDOZER9 · 04-15-12, 10:03 AM

This is something I am trying to figure out for a budget gamer build I'm attempting to put together.

freeagent · 04-15-12, 10:52 AM

Its pretty much trial and error at first.. Dont forget to write your results for each setting you try. It can get confuseing after awhile, and you could find yourself repeating old mistakes

Once you get the hang of it its not so bad. Thats when you start playing with sub timeings and other things you normally wouldnt

Keep your windows disk (or usb) handy

bluezero5 · 04-15-12, 11:14 AM

keep in mind trail and error will be for the application you are testing only.

Application A and Application can have VERY different results.

In General

1, RAM SPEED
2, Primary TIMING
3, Command Rate
4, Secondary TIMING (I don't know of anyone that touch this)

04-14-12, 05:16 AM	Thread Starter #1
Willy8382 Registered Join Date: Apr 2011	speed vs timing Is there some form of rough math that one can do to tell if one ram setting is faster than another. say like 1333@ 7 7 7 vs 1600@ 9 9 9 so you dont have to run bench marks all the time. __________________ Case-HAF 912 With 4*120mm Fans Mobo-MSI 890FXA-GD65 Cpu-AMD 955 BE @4.0 Cpunb@2600 Heatsink-Venomous X RT 24c idle 45c loaded Ram-Gskill Ripjaws @1600 8-8-8-20-27-1T Gpu-Sapphire 5850 Extreme@850/1200 Psu-Seasonic 750(X series) Hdd-WD 750Gb Black
	QUOTE Thanks

04-14-12, 11:22 AM	#2
redduc900 Super Moderator Join Date: Dec 2000 Location: WCG@home in Portland	There is no formula for calculating transfer rate between different frequencies and timings, although booting to Memtest or Memtest86+ can give you a theoretical peak transfer rate measured in MB/s, which can be used to quickly gauge any DRAM related changes. __________________ "Taking Your Posting to the Next Level" "Unofficial Posting Guidelines" "Forum FAQs, Guidelines, and Rules" "Signatures: Content, Size and readability"
	QUOTE Thanks

04-14-12, 08:05 PM	#3
trents Senior Member Join Date: Dec 2008 Location: Centralia, WA	The other complication here is that it will vary with the application somewhat. Frequency and timing both affect "speed". It might be better to refer to frequency as bandwidth rather than speed. __________________ Custom built desktop #1: Phenom II X4 960T Zosma unlocked to 6 cores/4.0 ghz/Noctua D14 CPU cooler/GSkill Ripjaws 1600 2x4gb/Asus M5A97 motherboard/Coolermaster HAF 912/case/OCZ 750W PSU/Nvidia 8800 GTS 640mb GPU/WD Caviar Black 500 GB hard drive Custom built desktop #2: PII X4 955 4.1 ghz, Asus M5A97 motherboard, Coolermaster Hyper 212 plus, Nvidia Geforce 220 video card, Enermax 465W PSU, 2x2gb PNY 1600 ram. Toshiba SatelliteL775 laptop/17.3" display, AMD 3400M APU, 8 gb of 1333 Jesus said, "I am the way and the truth and the life. No one comes to the Father but through me". John 14:6
	QUOTE Thanks

04-15-12, 06:05 AM	Thread Starter #4
Willy8382 Registered Join Date: Apr 2011	Thanks for the replys. Thought i could get a short cut but I guess its just trail and error. Part of the fun anyways. __________________ Case-HAF 912 With 4*120mm Fans Mobo-MSI 890FXA-GD65 Cpu-AMD 955 BE @4.0 Cpunb@2600 Heatsink-Venomous X RT 24c idle 45c loaded Ram-Gskill Ripjaws @1600 8-8-8-20-27-1T Gpu-Sapphire 5850 Extreme@850/1200 Psu-Seasonic 750(X series) Hdd-WD 750Gb Black
	QUOTE Thanks

04-15-12, 10:03 AM	#5
WARDOZER9 Registered Join Date: Oct 2007	This is something I am trying to figure out for a budget gamer build I'm attempting to put together.
	QUOTE Thanks

speed vs timing

Internet Services

Website Development

Tech Hardware

04-15-12, 10:52 AM	#6
freeagent Member Join Date: Sep 2004 Location: Winnipeg!	Its pretty much trial and error at first.. Dont forget to write your results for each setting you try. It can get confuseing after awhile, and you could find yourself repeating old mistakes Once you get the hang of it its not so bad. Thats when you start playing with sub timeings and other things you normally wouldnt Keep your windows disk (or usb) handy __________________ Xeon X5690 E.S. @ 4412Mhz 1.4v/Corsair H100/3x 38x120 Panaflo Asus Rampage III Formula @ 163/3595/3268 6x2GB STT+G.Skill (MGH-E+BBSE) @ 817Mhz 7-9-7 1T 1.65v Asus GTX580 Matrix Platinum @ 900/1052 1.1v OCZ RevoDrive 120GB\|Intel 520 120GB\|WD 1TB Black Auzentech Prelude\|TT SP850M\|ZM-MFC2\|CM Stacker STC-01
	QUOTE Thanks

04-15-12, 11:14 AM	#7
bluezero5 Winner, Rig-o'-the-Quarter Fourth Quarter 2012 Join Date: Mar 2012 Location: Tokyo	keep in mind trail and error will be for the application you are testing only. Application A and Application can have VERY different results. In General 1, RAM SPEED 2, Primary TIMING 3, Command Rate 4, Secondary TIMING (I don't know of anyone that touch this) __________________ CM Storm Trooper Chassis - ASUS Rampage 4 Extreme Core i7 3930k 5.0Ghz (1.52Vcore) select1: GTX680 3x SLI @1280Mhz select2: HD7970 GE @ 1300Mhz 32Gb Trident 2400 DDR3 10-10-12-30 T1 Motto: I didn't go water-cooling for an air-cooling Vcore!! My Rig's build log ::: Fans 101: Going Scientific ::: Water Cooling FAQ
	QUOTE Thanks

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search