Xtreme SUPERCOMPUTER
Nov 1 - Nov 8 Join Now!
Athlon64 3700+ KACAE 0605APAW @ 3455MHz 314x11 1.92v/Vapochill || Core 2 Duo E8500 Q807 @ 6060MHz 638x9.5 1.95v LN2 @ -120'c || Athlon64 FX-55 CABCE 0516WPMW @ 3916MHz 261x15 1.802v/LN2 @ -40c || DFI LP UT CFX3200-DR || DFI LP UT NF4 SLI-DR || DFI LP UT NF4 Ultra D || Sapphire X1950XT || 2x256MB Kingston HyperX BH-5 @ 290MHz 2-2-2-5 3.94v || 2x256MB G.Skill TCCD @ 350MHz 3-4-4-8 3.1v || 2x256MB Kingston HyperX BH-5 @ 294MHz 2-2-2-5 3.94v
chew*, I'm sorry but I have to dissagree with you on that. :o My nforce5 board has drive strength tweaking, and it does way more than what you suggest.
Setting different drive strengths (wrong ones) vs setting drive strengths (good ones) can make a HUGE difference in stability on DDR2. I've noticed after moving to DDR3 that it's not as important anymore, but I've certainly noticed at least 5-10 Mhz gains/losses by tweaking them. Is that significant? No.
Was going from 1000 4-4-4-5 to 1035 4-4-4-5 on my Promos sticks significant? Yes. I also tweaked drive strengths for my 5600+ to reach higher IMC capability, with that kit I could do 1050 CL5 @ -.-v, after tweaking drive strengths I was able to hit much higher.
I believe nforce4 (specifically Abit and DFI) had this as well.
Actually, you do that with every single post you make.
Last edited by BeepBeep2; 10-20-2010 at 04:37 PM.
Smile
heatware chew*
I've got no strings to hold me down.
To make me fret, or make me frown.
I had strings but now I'm free.
There are no strings on me
funny im sitting here looking at drive strength settings here too on my NF4 ultra. so your wrong.
and nice to see if you say anything to anyone on here they report your sig to mods, i have had my sig on here for months and nothing said, i object to a post and its reported.
sad bully boy tactics if you ask me. you say it to there face, and they say it behind your back.
figures, funny how its all the ones that have techreaction accounts that are objecting.
pure gang up, bully tactics. very sad
thanks for posting to both slappa and enjoy. i personally think it was informative and achieved what it was meant for.
sorry to see the thread get torn away from it's original intent though
on a side note perhaps we could all make something constructive or useful from the rest of this thread. paticle, you want more drive strength information
and you seem to like charts and such so why not have a seperate thread that we all could assist in gathering that info. perhaps like this
bench spi 32m
cpu speed capped at 4000mhz + or -
nb speed at 2800mhz + or -
ht no limits
memory no limit
listing of memory settings that pertain to drive strengths
2 screen shots of cpuz (cpu) (mem)
those who wished to share would participate, those who did not would not.
just my 2 cents worth.
again, thanks slappa and enjoy for your effort.
http://www.nvidia.com/content/nforce...erclocking.pdf
Look closely at page 26.
Changing any one of those values (even from 1.5 to 1.25 or 1.0 to .75) was the difference between complete stability and getting a BSOD upon windows bootup.
The fact that I (,myself, you can call me sir) have a TechReaction account means nothing in my opinion. I only have one "blog" over there anyway, a mini-review of SuperTalent Speed Series DDR3-2200 with CAS 6 scaling. (One stick seen at left.) Maybe nobody pays enough attention to you until you start flamewars or something. I know that everything I've said in this thread would have been the same whether I'd had a TR account or not. I designed the logo in my sig, just because I've got an article over there. I'll have a few others soon, specifically a full review of my SuperTalent 2000C7 sticks, (with Thuban,) showing whether they do rated timings as well as overclockability vs my 2200 C8's. Max CL5,6,7, and 8. Just because my review is featured on TechReaction and it's viewers (as well as XS) doesn't mean I'm biased towards them.
Why don't you talk positively and attempt to help others in a positive manner? It makes what you contribute to the forum seem quite a bit better. I know I'm guilty of spreading faux information, but hey, I was, and still are a kid. I mature more and more every day, I can even tell by looking back at my previous posts since 2009. Some of the things I have said before I wouldn't even have thought about saying now.
Does that make me any lesser than anyone else on this forum? No, no it does not.
Last edited by BeepBeep2; 10-21-2010 at 02:29 PM.
Smile
990FXA-UD3 | FX8350@4.7Ghz | Asus HD7870 | 2x 4GB Crucial Ballistix Tracer 2050Mhz 8-8-8-22 | AX850W |SSD Vertex3 Max IOPS 120GB | Auzentech Forte + TAPCO-S5
EK Supreme Full-Gold | XSPC RX240 + EX120 | MCP35x | 3x Koolance Blue Led @PWM | Tygon Black 1/2 | Bitspower Compression | @ FM CM690 I
I noticed with Hypers (DDR3 in general) it doesnt matter as much, I played with them alot on my board but I never really gain anything. I gain about 2 Mhz max I'm sure there are Mhz to gain/lose, I tried to play with them for my 2200 CL8 mini-review, I'd find highest stable on auto and then play with drive strengths and try to stay stable at the same frequency. I didn't really get anywhere with it.
Just run auto and you should be okay.
DDR2 is a different story.
Last edited by BeepBeep2; 10-21-2010 at 05:51 PM.
Smile
from my knowledge barrell rider i think when going for high mhz your better off starting with lower drive strengths but to be honest i have no hard data to back it up, its just what i have noticed.. hopefully someone who knows more about what they actually do can chime in
LEO!!!! amd phenom II x6 1100T | gigabyte 990fxa-ud3 . . 2x2gb g.skill 2133c8 | 128gb g.skill falcon ssd sapphire ati 5850 | x-fi xtrememusic. . . samsung f4 2tb | samsung dvdrw . . corsair tx850w | windows 7 64-bit. ddc3.25 xspc restop | ek ltx | mc-tdx | BIP . . lycosa-g9-z2300 | 26" 1920x1200 lcd .
Actually it's best to run the lowest drive strengths possible while keeping stability, as drive strengths are how sharp the instruction edges are. Basically drive strengths work like an analog audio signal. The weaker the signal, the more noise there is an the more another source bleeds into that channel. (Kinda like a radio) ...The stronger the signal, the less noise there is but the more amps are pushed into a small speaker. The weaker the signal, the more the ram will clock Mhz wise, but you must worry about instability at (all) clocks.
I've found that with DDR2. DDR3 seems different and a bit more useless.
See here:
http://www.xtremesystems.org/forums/...d.php?t=140870
The higher drive strengths are, the more the IMC and DIMM's are stressed, resulting in lower clocks.
Colder temps also cause drive strengths to become stronger for some odd reason.
This is why it helps with coldbugs on K8 like chew* stated. (Also ram clocking when subzero, I believe chew* has vaguely mentioned in the past that Drive Strength only helps when cold )
Drive strengths become so strong it results in a coldbug.
Some chips do not like having tons of current being pushed into them.
I've noticed my DDR2 DIMM's running way hotter with higher drive strengths than lower ones. My ProMOS needs higher drive strengths, (unstable with lower) and my Elpida based OCZ LV DDR2 (I will probably sell within the next few weeks) works much better with low drive strengths.
To those with Hypers (especially MNH-E), I'd run lower drive strengths if I were you. Mine seem to work fine even down 50% to 1.0x from 1.50x. I lose a maybe a few Mhz, but hey...maybe they will stick around longer in the end.
Hope that helps
Last edited by BeepBeep2; 10-21-2010 at 06:44 PM.
Smile
Bookmarks