LYN's DDR2 Discussion, Come in and share
|
|
 Dec 13 2006, 12:22 AM
Return to original view | Post
#1
|
 
Elite
6,799 posts Joined: Jan 2003  
|
is this any good??
i have clocked till 1000 2.4v 5-5-5-15 but this is the best performance bandwidth that i got Kingston Hyperx PC6400 DDR2 SLI Low latency version 2gb kit 920mhz 4-4-4-12 timing 2.3v Buffered  This post has been edited by cstkl1: Dec 14 2006, 10:03 AM |
|
|
|
|
|
Dec 15 2006, 04:43 PM
Return to original view | Post
#2
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(yehlai @ Dec 14 2006, 11:17 PM) U mean Kingston Hyperx 667?? its stock timing is lower than xms... can i see ure sandra benchmark for that ram of urs??1000Mhz can use 4-4-4-12, can try ?  just wondering |
|
|
Dec 15 2006, 09:27 PM
Return to original view | Post
#3
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(soulfly @ Dec 6 2006, 12:55 PM) If you use common sense, obviously you would want lower latency for better performance. It's not the matter of which one you should use, but rather which one you want to use. i actually tested it on my ramson 920 4-4-4-12 sandra benches at 9800 on 1k 5-5-5-15 sandra benches at 9500 |
|
|
May 26 2007, 03:08 AM
Return to original view | Post
#4
|
|
Elite
6,799 posts Joined: Jan 2003
|
eh repeat anybody with nvidia mobo.. go 1T BRO.. .feel the difference.
at 2t u cant beat an intel chipsets but on 1t different story. so find the highest vdimm u dare.. 2.3-2.4 relatively safe with external active cooling and overclock away. currently reverted back to windows and running 2x1gb pc2-6400 CL3 at 1t difference?? biggest gain can be seen in aquamark. ultra overclocked at only 400x9 with those rams.. hit 200k..kekeke...damn... |
|
|
Feb 8 2008, 06:51 PM
Return to original view | Post
#5
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(king99 @ Feb 8 2008, 04:32 PM) Er..Why when I set OCCT to MAX priority the OCCT said 60% CPU testing but in the test manager said it is running 100 % priority just means finishing cycles..Boy can't do anything else when using OCCT at MAX Priority . stressing is percentage of core.. do the hardcore test.. get TAT.. watch ure temps fly.. |
|
|
Feb 9 2008, 06:05 PM
Return to original view | Post
#6
|
|
Elite
6,799 posts Joined: Jan 2003
|
i am still impresed by OCZ reaper PC2-8500 CL5
this rams are seriously almost as good as my team xtreem pc2-6400 Cl3's and team xtreem PC2-8500 Strongest Ai SKEw tRD 8 ( max on this board ) although i am a fan of 1:1 but since the latency is the same for 1200 CL5 and 1000 CL4 at 500 fsb tRD 8. y not just increase the bandwidth eh.. t fluctuates at 2.176 - 2.160vdimm.. awesome rams.. Take note sp32m was to show stability of pll/nb and vdimm .. memtest windows just to show stress indication of a section of ram during sp32m trial.. full test will be when i run it on memtest86 overnight.. [attachmentid=395652][attachmentid=395656][attachmentid=395660] 86% bandwidth efficiency highest i ever got.. these D9GMH's are seriously underestimated.. This post has been edited by cstkl1: Feb 9 2008, 07:01 PM |
|
|
|
|
|
Feb 9 2008, 07:22 PM
Return to original view | Post
#7
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(sniper69 @ Feb 9 2008, 07:16 PM) nope... yup and a few more things , there's no way you can check via software, all you have to do is open the heatspreader if any...  like charateristics most D9GMH can do 800 CL3-3-3-8 within 2.2-2.3v example.. others cant at this vdimm.. small small stuff like this u will know whether is a D9GMH or a D9GKx.. but i usually get confused with D9GKX and promos.. they are quite close.. so have to open the heatspreaders.. |
|
|
Feb 12 2008, 02:15 AM
Return to original view | Post
#8
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(alive88 @ Feb 12 2008, 01:50 AM) guys,im using hardware as in my siggy. hmm hate to post to show example againb4 i can OC my ballistix till ddr1200 wif 2.2vdimm,but now cant exceed more than ddr1160,eventhough i add some more vdimm and loose the timing but ure answer is at this answer of mine http://forum.lowyat.net/index.php?showtopi...post&p=15601628 on the bus 400.. ure trd's are wrong.. or ure NB voltage is wrong for that new Bus ure running which is higher than the previous my bet.. This post has been edited by cstkl1: Feb 12 2008, 02:15 AM |
|
|
Feb 12 2008, 02:29 AM
Return to original view | Post
#9
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(alive88 @ Feb 12 2008, 02:26 AM) first i suspect that,coz ive read ur post.. oops ure using amd.. hmm didnt notice that..im using the same setting,but this time cant..ive tried at my friend's rig,also cant then different u dont have trd forget that link doesnt apply to ure issue.. This post has been edited by cstkl1: Feb 12 2008, 02:30 AM |
|
|
Feb 12 2008, 02:31 AM
Return to original view | Post
#10
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(alive88 @ Feb 12 2008, 02:30 AM) ive tried to increase the nb voltage..also cant..is my ram faulty? noure ram has hit a limit on vdimm because of either cooling or pcb... did u know rams are not meant to be run above 60C heck make that 50.. for ure low latency crucial rams.. suicidal to run those in malaysia..even at 2.2 with those pcb u should have active cooling on them heck look at the reapers they are using with the same PCB but heatpipe... I told this before in DA's thread when he was selling them..guess nobody listened...figures.. This post has been edited by cstkl1: Feb 12 2008, 02:34 AM |
|
|
Feb 12 2008, 02:36 AM
Return to original view | Post
#11
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(alive88 @ Feb 12 2008, 02:34 AM) im using allngap ram cooler bro,plus direct pensonic stand fan and airconditioned room and other times..??ive touched the heatspreader..cold all the time?? remember amd with mem controller really stresses those rams more than intel chipset mem controller.. likes comparing getting punched by a small kid and mike tyson punching u. i wouldnt run those above rated specs. even on latencies and mhz on amd setup. Added on February 12, 2008, 2:38 amactually ure question would be better answered in the amd/am2 thread.. most of those users aint plug and play.. This post has been edited by cstkl1: Feb 12 2008, 02:38 AM |
|
|
Feb 12 2008, 02:46 AM
Return to original view | Post
#12
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(alive88 @ Feb 12 2008, 02:41 AM) what bother me the most is that i can run ddr1200 b4,wif the very same hardware and setting..but now suddenly cant sorry dude.. cant help ya.. been out of touch with amd /am2 for nearly 2 years approaching.. |
|
|
Feb 14 2008, 12:57 PM
Return to original view | Post
#13
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(Jibek @ Feb 13 2008, 02:50 PM) which 1 is better bro.,ocz 4gb/800 sli edition or ocz 4gb/800 reaper hpc..tq both same chipset...better cooling debatable.. overclocking junkies will say .. heck bare better.. so they will go for the platinum, titanium version ppl who likes the reaper part will say something to to justify that heatpipe. |
|
|
|
|
|
Apr 1 2008, 10:04 PM
Return to original view | Post
#14
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(tomatos @ Mar 29 2008, 04:19 PM) 4*1GB and 2*2GB would perform the same at same speed. However overclocking 4 sticks would need higher voltage for a lower latency. Not to forget the nb voltage too. me thinkest thou art drunk..If you don't mind pumping very high voltages and risk spoiling your ram, 4 sticks of D9GMH would be great at high voltage. |
|
|
Apr 22 2008, 06:38 PM
Return to original view | Post
#15
|
|
Elite
6,799 posts Joined: Jan 2003
|
Importance of Bandwidth and Latency
Settings - OCCT Stable 5 hours Mix Infinity + Memtest 5 passes Proc : E8500 Ram : Team Xtreem PC2-8500 CL5 4x1gb Mobo : DFI LP X48 T2R LT Settings 1. 500x8 : 4ghz@ 1.272v , 1000mhz 4-4-4-8 tRD 7 @2.16v , Strap 333/667 (1:1) 2. 400x9 : 3.8ghz@ 1.192v, 1000mhz 4-4-4-8 tRD 5 @2.3v , Strap 266/667 (5:4) 3. 400x9 : 3.8ghz@ 1.192v, 960mhz 4-4-4-8 tRD 6 @2.2v , Strap 333/800 (6:5) 4. 400x9 : 3.8ghz@ 1.192v, 800mhz 3-3-3-8 tRD 6 @ 2.2v , Strap 333/667 (1:1) 5. 500x7.5 : 3.8ghz@1.175v, 1000mhz 4-4-4-8 tRD 7@2.16v , Strap 333/667 Lowest Trd Calculation A) 500FSB @ 333/667 (1:1) CL4 7-4/1>2.75/1 3>2.75 = tRD 7 B) 400FSB @ 266/667 (5:4) CL4 5-4*4/5>2*4/5 25/5-16/25>8/5 9/5>8/5 =tRD 5 C) 400FSB @ 333/800 (6:5) CL4 6-4*5/6>2*5/6 36/6-20/6>10/6 16/6>10/6 =tRD 6 D) 400FSB @ 333/667 (1:1) CL3 6-3/1>2/1 =tRD 6 Everest Ultimate 4.5v 1330 Mem Based Scores Mem Read [attachmentid=450880] Mem Write [attachmentid=450882] Mem Copy [attachmentid=450883] Mem Latency [attachmentid=450884] CPU PhotoWorxx [attachmentid=450886] Total Proc Based Scores » Click to show Spoiler - click again to hide... « This post has been edited by cstkl1: Apr 22 2008, 06:46 PM |
|
|
Apr 22 2008, 08:06 PM
Return to original view | Post
#16
|
|
Elite
6,799 posts Joined: Jan 2003
|
eh... mr. rlhc17..
me suspect ure trd is very high.. kekekek looking at ure nb voltage.. run those 3 settings for 400fsb that i did above 1000mhz CL4 TRD 5 960mhz CL4 TRD6 800mhz CL3 TRD 6 This post has been edited by cstkl1: Apr 22 2008, 08:07 PM |
|
|
Apr 24 2008, 12:03 AM
Return to original view | Post
#17
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(rlhc17 @ Apr 23 2008, 09:50 PM) Kingmax Ram with 2.2vdimm DDR2 1000mhz CL4 yo rhlc7 at those straps lowest trd is still 6 and not 5 try 266/667... kekek and if u can do trd 5 on those at 2.2 i will say thumbs bro.. kekeke ure nb voltage says it all.. |
|
|
Apr 24 2008, 01:36 AM
Return to original view | Post
#18
|
|
Elite
6,799 posts Joined: Jan 2003
|
actually i like this board except a bit pissed with DFI frankie right now
hasnt answered my questions on the voltage issue y a lot of voltages range missing example 1.272v, 1.268v its like a jump from 1.26 to 1.28... for example Added on April 24, 2008, 1:42 amoh yeah raymond havent tried other procs yet but i find the key overclocking that should be done is find the NB GTL and vtt first then only the cpu gtl ( seems to be usually +20/25/30 of the nb gtl for cpu 1/3) with cpu 0/2 being -5/-10 off the cpu 1/3.. this is what so far all the GTL ranges showed for my proc.. but i can tell u the setting i use for 4ghz if one setting also i change now will lead to instability .. kekeke also hmm found that the cpu gtl usually is the same for all the vtt.. was hoping to ask irenic to lend me his proc. kekeke.. wanna try my theory on this dfi gtl's no offense to kaiser but this board really aint for the faint hearted.. but its awesome just like the p35 UT to find the absolute stable setting for everything.. hey have u tried eva2000 mem thingy thats he has been talking about.. to increase the performance he's been messing with the channel phase.. interesting.. thats the only setting i left at auto current.. This post has been edited by cstkl1: Apr 24 2008, 01:49 AM |
|
|
Apr 26 2008, 12:45 AM
Return to original view | Post
#19
|
|
Elite
6,799 posts Joined: Jan 2003
|
bumped the cpu voltage a bit
tweak the settings and this is the best setting WMDT 5 pass enchance chache and OCCT 5 hours result awesome latencies.. [attachmentid=454132] |
|
|
May 3 2008, 05:41 PM
Return to original view | Post
#20
|
|
Elite
6,799 posts Joined: Jan 2003
|
QUOTE(likito @ May 3 2008, 05:23 PM) i dont whether got posted it or not ! just wanna share .. good stuff..Crosair DDR2 ram IC listing » Click to show Spoiler - click again to hide... « source should be a sticky.. somewhere. since a lot of ppl buy this rams..  |
|
Topic ClosedOptions
|
| Change to: |  0.0491sec
 0.42
 7 queries
 GZIP Disabled
Time is now: 28th November 2025 - 04:14 AM |
Quote