DFI LP UT X58-T3EH8 - First bench, bios screens etc

[StimulatedBoy]

After extensive reading of the threads created by EVA2000 and AndreYang I finally got beyond basics. First SuperPi 32m run on new 24/7 settings; Vcore 1,475, Vtt 1,47 and Vdimm 1,65. Now some BF2 gaming.

Thx to Eva2000 and AndreYang for providing the solutions to a NOOB like me!

[PRJ]

Someone wrote earlier, that he had to many led lights on at power off, I have two yellow and the green power on.

Is this normal..?

[Sam_oslo]

Thanks for your efford nemesis. I too was wondering WTF went on about this overvoltage thing reported in Smart Guardian, so you are saying SG reports crap and the voltages set in bios are pretty close to actual? I sure hope so.. gotta measure it, where can I find the spots to measure on?

The strange correlation between VTT and Vdimm I noticed as well, for instance if I go too far up or down with Vdimm while leaving VTT the board will only post showing 4GB instead of 6. The range on my current 920 seems to be 0,25-0,3V difference between the 2.

And yeah I tried 1,55V all right, went as high as 1,65V with some CPUs, to no avail. And remember I am not talking about 2Ghz+ here just stupid 1600Mhz. Impossible with normal/low volts... if I feed it 2V Vdimm then sure it'll do it.. but I sure as hell ain't runnin 2V for 1600Mhz.

It's also not the uncore speed, I just upped the uncore to 16x at 210 BCLK and its still linX stable at 4,4Ghz, no change whatsoever. Changing the ram mult to 8x causes the system to freeze on boot

I'll maybe play with it some more tomorrow, and I wouldnt be complaining if I had trouble hitting 2Ghz+ ramspeed, but we are talking officially supported 1600Mhz here. Somehow I feel like an idiot running 4,4Ghz with HT rockstable but running the memory at 1200 something

And I ain't exactly new to OCing or to DFI either, to say the least... tried 5 different CPus so far and 3 different mem kits. Maybe I need to try a kit with XMP profile or something, cause the ones I tested were all CSX and they don't have XMP or anything.

I have been experiencing some strange MEM behavior (compeered to what we used to on CD2) at higher BCLK (over 200).

My sticks (OCZ3P1333LV6GK) are rated @1333MHZ CL7 1.65v, but they fail memtest @1266 CL7 1.65v. The strange part is that it will pass the memtest with same timing and even lower (6-6-6-24) with 1.6v if i turn the HT off. It has been passing memtest at several x8 speeds, with and without HT, opp to BCLK 200 with different timing/dividers and had no problems..

Something fishy happens over BCLK 200 with my sticks, they appear to be stable, and can pass SuerPI 32M, and stay stable a good while in memtest too, but they will fail after a while in memtes 3.8.

I'm still investigating this, but something (could be CPU or MB, or stick's capability) bugs the MEM at BCLK over 200 and HT:on, even with x6 multi.

[REBEL900]

Hey guys, read all the posts on this thred and over at i4memory, got a similar set-up, Core i7 920, DFI X58, Corsair 6GB DDR3 8 8 8 24 and these are my results. Please comment, critic, point things out as necessay.
Like to thank eva, jcool, The nemesis, and any one else i left out for thier valuable and very imformative info

[CmB]

Hey guys, read all the posts on this thred and over at i4memory, got a similar set-up, Core i7 920, DFI X58, Corsair 6GB DDR3 8 8 8 24 and these are my results. Please comment, critic, point things out as necessay.
Like to thank eva, jcool, The nemesis, and any one else i left out for thier valuable and very imformative info

You're on air with a true right mate, tossing it up whether to go water, most likely will. Nice results, i'll look more later I'm tired now.

[jcool]

Hey no need to thank me, all I ever do around here is complain Eva did really great work, even though I can't seem to copy his results, ram-wise that is...

Your screenshot looks perfectly fine, also very nice low Vcore if that is also your load Vcore. I only had 2 920's that could handle 3,8 at below 1,25V. WHat batch # is yours?

[REBEL900]

You're on air with a true right mate, tossing it up whether to go water, most likely will. Nice results, i'll look more later I'm tired now.

Thanks
Sorry i left out some info.. It's a water cooled CPU (apogee GTZ) on WinXP and the screen shot is with Prime running in the background for about 2 hours with OCCT just finishing an hour ago.

Hey no need to thank me, all I ever do around here is complain Eva did really great work, even though I can't seem to copy his results, ram-wise that is...

Your screenshot looks perfectly fine, also very nice low Vcore if that is also your load Vcore. I only had 2 920's that could handle 3,8 at below 1,25V. WHat batch # is yours?

Thanks jcool, please explain load Vcore, batch# is 3839A523

[Sam_oslo]

It have been doing some heavy MEM-testing, with different timing/dvider/volt and it seams that my RAM won't get totally stable at BCLK 211. I want to try some other sticks and I can get 6GB of either Corsair Dominator DHX+ 1600MHz CL8, or Kingston HyperX 1600MHz CL8 from my local retailer. But I'm open for other suggestions too, What kind av RAM would you guys recommend on DFI LP UT?

[eva2000]

Hey no need to thank me, all I ever do around here is complain Eva did really great work, even though I can't seem to copy his results, ram-wise that is...

My thread at http://i4memory.com/showthread.php?t=12855 and the tips post (sig) is more about the process than the results. Hence why you see more about oc'ing than stability tests from me. The process and what i experience when you play with volts, dividers, multipliers and memory is what I'm trying to share with others. They can take that process and behaviour and find their own acceptable overclocks and stable points since every cpu/mem/system is different

When I read reviews, I want to know the process and behavioural aspects more than the final clocks (which are important too)

It have been doing some heavy MEM-testing, with different timing/dvider/volt and it seams that my RAM won't get totally stable at BCLK 211. I want to try some other sticks and I can get 6GB of either Corsair Dominator DHX+ 1600MHz CL8, or Kingston HyperX 1600MHz CL8 from my local retailer. But I'm open for other suggestions too, What kind av RAM would you guys recommend on DFI LP UT?

Yeah i'm wondering how i7/x58 plat form does with various memory out there as well. Shame I can't get a set of every kit listed at http://i4memory.com/i7mem to try

[jcool]

My thread at http://i4memory.com/showthread.php?t=12855 and the tips post (sig) is more about the process than the results. Hence why you see more about oc'ing than stability tests from me. The process and what i experience when you play with volts, dividers, multipliers and memory is what I'm trying to share with others. They can take that process and behaviour and find their own acceptable overclocks and stable points since every cpu/mem/system is different

When I read reviews, I want to know the process and behavioural aspects more than the final clocks (which are important too)

Exactly, and I don't mean by copying that I just pasted your results
More that I can't get a decent Ram OC on my DFI even after having read up on a lot, including your guide.. I guess I either got a bad board, or simply the wrong memory for the DFI. They have always been very picky about rams, so I might need to try a "real" Tri-Kit next.

@Rebel900: Load Vcore is the Vcore that is displayed during stress testing

[StimulatedBoy]

Hi guys,

I have a couple of questions related to Memory LowGap and QPI Frequency. What is it and how do they work? And last, but not least, at what levels should I set them? The NOOB factor kicking in here...

Currently runnning 4,4GHz @ 200*22 with 1,47 CPU VID Con, 1,50 CPU VTT, 1,665 DRAM Bus voltage - Memory humming nicely along at 2000MHz at 7-8-7-21-1n. System temp as displayed by SG is 39 C for system and 30 C for chipset.

Thx for any input!

[slapmehard]

i can give you some input....dont use sg for temps.
core temp
http://www.alcpu.com/CoreTemp/
real temp
http://www.techpowerup.com/downloads...Temp_3.00.html

low gap info here
http://www.xtremesystems.org/Forums/...d.php?t=217711

[Sam_oslo]

I have been experimenting with MEM, an observing some strange results. After long and heavy MEM-testing, loose tRAS seams to play a bigger role (bigger than what we are used to on CD2) for MEM-stability at higher BCLK.

I have been testing with 8 instance av 50MB memtest 3.8, and have made these observations:

@4GHZ(211x19) HT:on:
@1266 7-7-7-20 1.65v fails (same with tRAS 21 to 24). I didn't test higher tRAS.
@1266 6-6-6-18 fails (same with tRAS 19 to 29), but tRAS 30 passes

@3.8GHZ (200x19) HT:on:
@1200 7-7-7-20 1.65v fails, but tRAS 21 passes
@1200 6-6-6-18 1.65v fails (same with tRAS 19 to 23), but tRAS 24 passes

Almost all fails accrued in 60% to 70% during the memtest.

Loose tRAS seams to be play a big role on MEM-stability at higher BCLK (at least on my CPU and RAM) and seams to have some effect on Vcore too. After i got it stabilized @1266 6-6-6-30 1.65v, i could prime "pass the kill zone" with 1.32v @4GZ(211x19) Ht:On, with is "several notch" lower than the previous Vcore requirement (1.34v).

Anybody else has observed similar behavior?

[jcool]

Interesting, will try higher Tras later. Maybe I can finally run a decent memclock then...

Edit: What did you run the subtimings at? Trfc and such? Auto?

[Xenozx]

OK i figured I would ask a question here based on my overclock, as it seems there are many knowledgeable i7 920 overclockers in this thread

I have an i7 920 (batch # 38B) currently clocked at

4.1ghz (196x21)


1.375v Vcore
Vdroop disabled
1.39v CPU VTT
1.80v PLL
1.65v RAM
all other voltages are defaults.

Whats weird is I am 100% prime 95 stable at this speed. if I move up even 1 BCLK to say 197, and run prime 95, i will get a BSOD. If I drop my multiplier to 19x, and change my BLCK to say 207 (same clock speed, just higher BCLK) I get a BSOD durring prime 95. What could cause that 1 BCLK to make such a difference? What would I need to do to overclock higher?

2.) also my 21x is on 24/7 in windows. If I run 4 core prime 95, it stays at 21x. If I run 8 thread prime and I watch CPU-Z the multiplier drops to 20x. What would cause that to happen?

I have

Intel i7 920 @ 4100mhz (196x21) @ 1.375v
DFI T3eH8 X58 w/ TR Ultra Extreme 120
6GB 2GBx3 OCZ 1560mhz @ 7-7-7-20
Tagan 1100w Quad SLI PSU
1 Visiontek 4870x2 (800/4000)
Soundblaster x-FI Fatality PCI-E
300GB Velociraptor + 150GB 10,000 rpm Raptor

Also, just to note, I can easily boot at 4.3ghz, but anywhere over 197 BCLK i am no longer prime 95 stable. I can still do everything else like super PI and games and stuff without fail. I have duplicated this across 2 chips (38a and 38b) and 2 boards (DFI x58 and asus P6T) so I almost thing its something setting wise I am doing wrong.

[Sam_oslo]

Interesting, will try higher Tras later. Maybe I can finally run a decent memclock then...

Edit: What did you run the subtimings at? Trfc and such? Auto?

Yes everything else on Auto, I ended up with this (for now):

@1266 6-6-6-30 1.65v


Did a bit Priming, with same MEM-settings, (for 30 min) just to see how it behaves, and so far so good.

@4GHZ (19x211) HT:on, 1.32v , 30 min primestable

[faster3200]

Sam oslo, thank you so much for your discovery. This pretty much confirms all of the issues I have been having are mem related. Thank you.

[eva2000]

Exactly, and I don't mean by copying that I just pasted your results
More that I can't get a decent Ram OC on my DFI even after having read up on a lot, including your guide.. I guess I either got a bad board, or simply the wrong memory for the DFI. They have always been very picky about rams, so I might need to try a "real" Tri-Kit next.

More likely just the cpu IMC and ram combination isn't the best maybe ? Only time will tell as more folks try different memory kits as to whether the ram spd/pcb design may effect memory clocks. I assume you already done the memory module musical chairs between dimm slots ?

I have been experimenting with MEM, an observing some strange results. After long and heavy MEM-testing, loose tRAS seams to play a bigger role (bigger than what we are used to on CD2) for MEM-stability at higher BCLK.

I have been testing with 8 instance av 50MB memtest 3.8, and have made these observations:

@4GHZ(211x19) HT:on:
@1266 7-7-7-20 1.65v fails (same with tRAS 21 to 24). I didn't test higher tRAS.
@1266 6-6-6-18 fails (same with tRAS 19 to 29), but tRAS 30 passes

@3.8GHZ (200x19) HT:on:
@1200 7-7-7-20 1.65v fails, but tRAS 21 passes
@1200 6-6-6-18 1.65v fails (same with tRAS 19 to 23), but tRAS 24 passes

Almost all fails accrued in 60% to 70% during the memtest.

Loose tRAS seams to be play a big role on MEM-stability at higher BCLK (at least on my CPU and RAM) and seams to have some effect on Vcore too. After i got it stabilized @1266 6-6-6-30 1.65v, i could prime "pass the kill zone" with 1.32v @4GZ(211x19) Ht:On, with is "several notch" lower than the previous Vcore requirement (1.34v).

Anybody else has observed similar behavior?

tRAS stability sometime is related to lack of vdimm. So if you push vdimm up with lower tRAS do you get more stability ?

OK i figured I would ask a question here based on my overclock, as it seems there are many knowledgeable i7 920 overclockers in this thread

I have an i7 920 (batch # 38B) currently clocked at

4.1ghz (196x21)
http://www.xenshardware.com/images/myi7overclock.JPG

1.375v Vcore
Vdroop disabled
1.39v CPU VTT
1.80v PLL
1.65v RAM
all other voltages are defaults.

Whats weird is I am 100% prime 95 stable at this speed. if I move up even 1 BCLK to say 197, and run prime 95, i will get a BSOD. If I drop my multiplier to 19x, and change my BLCK to say 207 (same clock speed, just higher BCLK) I get a BSOD durring prime 95. What could cause that 1 BCLK to make such a difference? What would I need to do to overclock higher?

2.) also my 21x is on 24/7 in windows. If I run 4 core prime 95, it stays at 21x. If I run 8 thread prime and I watch CPU-Z the multiplier drops to 20x. What would cause that to happen?

I have

Intel i7 920 @ 4100mhz (196x21) @ 1.375v
DFI T3eH8 X58 w/ TR Ultra Extreme 120
6GB 2GBx3 OCZ 1560mhz @ 7-7-7-20
Tagan 1100w Quad SLI PSU
1 Visiontek 4870x2 (800/4000)
Soundblaster x-FI Fatality PCI-E
300GB Velociraptor + 150GB 10,000 rpm Raptor

Also, just to note, I can easily boot at 4.3ghz, but anywhere over 197 BCLK i am no longer prime 95 stable. I can still do everything else like super PI and games and stuff without fail. I have duplicated this across 2 chips (38a and 38b) and 2 boards (DFI x58 and asus P6T) so I almost thing its something setting wise I am doing wrong.

tried bumping CPU VTT and/or VDIMM ?? there's a big difference between booting at a bclk vs super pi/games vs prime95

For air cooled you're probably heat limited so probably asking too much for >4.3ghz prime95 stability or even 4ghz for some i7 920s.


Moving my focus over to 3D benchmarks on DFI UT X58-T3EH8 I've flashed to latest 02/06 beta bios and swapped in my EVGA 9800GX2 stock hsf video card for some 3d benching. Alot more fun now that Rivatuner 2.22 and Voltage Factory app by AwardFabrik can do software GPU voltage adjustments

EVGA 9800GX2 clocks:

• Before default 1.15v GPU = 742/1917/2304 Max
• After 1.24v GPU = 786/1998/2304 so far

New personal bests for 9800GX2

3dmark06 = 26,807
3dmark05 = 36,027

Click 3dmark06 image to see full screen including Rivatuner reported GPU volts







System specs

• Intel Core i7 920 3836A756
• Thermalright Ultra 120 Extreme + LGA1366 Bolt Thru kit + 2x 120x38mm Scythe UltraKaze 2000rpm 87cfm fans in push/pull
• DFI LP UT X58-T3EH8 R.A51 - 02/06 beta bios
• EVGA 9800GX2 stock hsf (removed casing) + 181.22 WHQL Drivers
• 6GB (3x2GB) Corsair TR3X6G2000C7GT Dominator GT DDR3-2000Mhz 7-8-7-20 1.65v
• 750GB Samsung SATAII
• Pioneer DVD-RW
• Silverstone OP1200
• WinXP Pro SP3 full nLite updated hotfixes / Windows 7 beta dual boot

[REBEL900]

eva2000, how's that bios doing? I believe i'm on your previous beta bios 1/23

Thanks

[Sam_oslo]

...
tRAS stability sometime is related to lack of vdimm. So if you push vdimm up with lower tRAS do you get more stability ?
...

I have to admit, i have been a chicken, and never raised the Vdimm over 1.65v. I'm trying to set up a "safe" 24/7-setup, and don't want to be the one to find that the "Vdimm over 1.65v is safe with 0.5 different between VVT and Vdimm"-theory is false.

But i believe you are right, more Vdimm could stabilize tRAS, and other timings/dividers too, but it won't have much effect on my real-life applications (i think). I'm still testing the effect av loose tRAS, but i haven't noticed any different yet, so far. So I'll play safe until there is more reliables references on that theory.

[Xenozx]

Hey eva thanks for the response. My temps are in order. Even at the 4.1ghz, they never go over 70'c. I played arround a little bit, and upped the VTT to 1.45v and it seems to net me 1 BCLK extra for stability. Still no matter what I do I cant hit 200x21 (4.2ghz and exactly 1600mhz memory) which is where I would really like to be.

What I did find interesting last night, was if I disabled HT I could do this (primed for about an hour no problem)



temps never went over 80.

Whats even weirder, is I can 4 thread prime 95 above fully stable

If I go back to 197x21 with HT on and try to run just 4 thread prime I will eventually get a BSOD, but if I do 207x21 with HT off, I can prime 95 just fine?

Whats the deal with that.

outside of PLL, CPU VTT, and Vcore, is there any other voltage I should play with that might help me obtain that sweet spot of 200x21?

[eva2000]

probably need cpu vtt >1.5-1.6v some cpus just need more cpu vtt than others.

[Xenozx]

Is that safe for 24/7 opperation?

Is there any point in using CPU VTT Special add? is that just adding that much extra onto the value you select for CPU VTT?

[Slash1981]

Hi guys, i've been reading this thread for some time and i got my t3eh8 a couple of days ago and i got into some trouble. everything was ok at first but then i updated the bios with the 02 06 beta bios and i got a problem that i cannot enter bios screen. when i hit the DEL key during the POST all i get is the blinking curser on the top left hand side of the screen. everything else runs well i can use windows without a problem. then i got the 01 14 official bios and updated and nothing changed then i reloaded the 114 abs profile and nothing changed either.

i already emailed dfi but did not get an anwer yet, if you guyz could help it is much appreciated,

My rig:

DFI LP UT X58 T3eH8 bios 01 14 official
i7 920 batch# 3836A594
3x2 gb gskill 1600 mhz.
2xmsi4870x2 oc edition
creative sb xfi fat champ (PCI)
cpu is watercooled and fans blowing to pwm area and nb area
i am not using raid or anyother add on cards

thx in advance

[CmB]

Hi guys, i've been reading this thread for some time and i got my t3eh8 a couple of days ago and i got into some trouble. everything was ok at first but then i updated the bios with the 02 06 beta bios and i got a problem that i cannot enter bios screen. when i hit the DEL key during the POST all i get is the blinking curser on the top left hand side of the screen. everything else runs well i can use windows without a problem. then i got the 01 14 official bios and updated and nothing changed then i reloaded the 114 abs profile and nothing changed either.

i already emailed dfi but did not get an anwer yet, if you guyz could help it is much appreciated,

My rig:

DFI LP UT X58 T3eH8 bios 01 14 official
i7 920 batch# 3836A594
3x2 gb gskill 1600 mhz.
2xmsi4870x2 oc edition
creative sb xfi fat champ (PCI)
cpu is watercooled and fans blowing to pwm area and nb area
i am not using raid or anyother add on cards

thx in advance

Are you using a PS2 keyboard? USB keyboard don't always work in BIOS. I always use a PS2 keyboard, at least when oc'ing or working in BIOS, tho I have one USB KB that almost always gets me in bios. Kinda odd about the cursor tho, maybe you hosed yuor bios chip with the beta...did you flash while OC'ed? Best bet is to take all down to stock then flash. Just a suggestion.

--Chris