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

[eva2000]

nice... win7 is really nice

Hey guys got back to DFI since UT-X48-T3RS.

Beta BIOS of 0805 is out anybody tried?

BTW my PWM area temp is getting ultra high...72C @ idle...maybe RMA time?...

tips for lowering pwm temps here http://i4memory.com/102537-post3.html

yet to try 05/08 beta bios yet

[cstkl1]

nice... win7 is really nice


tips for lowering pwm temps here http://i4memory.com/102537-post3.html

yet to try 05/08 beta bios yet

not much difference atm.
seems to be the same for the rams

2180mhz CL7 stable.. thats mad.

[hadji_a]

A few of us are discussing same experience with win7 at http://i4memory.com/f80/windows-7-df...8-t3eh8-17336/, some findings are VID Special Add set to AUTO fixes it (although i didn't have problems with it set to disabled) and also that cpuz is incorrectly reporting lower multipliers when turbo 21x multi is enabled - as my benchmarks are all showing results reflective of 21x being in place even if cpuz fluctuates for me between 5x, 8x, 12x, 15x, 17x, 18x, 20x and 21x in windows 7

Proof that cpuz reported cpu multiplier fluctuations in Windows 7 is cpuz related as both realtemp 3.20RC4 and coretemp 0.99.5 reported right 21x multi all the time

Fluctuating

Thx for the heads up eva! I have taken a look at the thread over at i4memory and tried all of the suggestions but unfortunately I am stuck at 12. I have also given i7turbo tool a go that unclewebb (thx!) has suggested and the multi is not changing at all . My windows setup is an upgrade from a previous beta (this may be the culprit?). I have also tried the 5/8 bios and still no change on the multi issue. Overclock wise I am stable at 21x191 with ram (ocz blade) at 950 (7-7-7-18 4-way/4-way) same as the 4/28 bios.

[eva2000]

not much difference atm.
seems to be the same for the rams

2180mhz CL7 stable.. thats mad.

I'm at 2160 7-8-7-20 stable with 4/28 hehe

Thx for the heads up eva! I have taken a look at the thread over at i4memory and tried all of the suggestions but unfortunately I am stuck at 12. I have also given i7turbo tool a go that unclewebb (thx!) has suggested and the multi is not changing at all . My windows setup is an upgrade from a previous beta (this may be the culprit?). I have also tried the 5/8 bios and still no change on the multi issue. Oveclock wise I am stable at 21x191 with ram (ocz blade) at 950 (7-7-7-18 4-way/4-way) same as the 4/28 bios.

strange indeed.. - is this win7 32bit or 64bit ?

[hadji_a]

x64. The upgrade was from 7068.

[eva2000]

i'm using 7100 from MS TechNet download fresh install without problems - so maybe try fresh install ?

[hadji_a]

I will and report back.

[ACE76]

The info here is so good that I ordered this board last night...I also ordered a 6gb G.Skill memory kit that runs 8-8-8 timings...I got tired of waiting for the 7-8-7 kit to come back in stock. I was going to put my D0 Xenon CPu in my Asus Revolution board but I'm gonna try it on this instead...hopefully I can get some nice stable clocks out of this set-up.

[cstkl1]

The info here is so good that I ordered this board last night...I also ordered a 6gb G.Skill memory kit that runs 8-8-8 timings...I got tired of waiting for the 7-8-7 kit to come back in stock. I was going to put my D0 Xenon CPu in my Asus Revolution board but I'm gonna try it on this instead...hopefully I can get some nice stable clocks out of this set-up.

ah thats going to be interesting

eva suppose to test that asus mobo also..

waiting the results on that.

[drnip]

I suspect it's just cpuz misreporting - try with VID Special Add left at AUTO instead of disabled... no problems for me in terms of end result performance

Core Temp was reporting me at 12 mult as well. Not to mention my SuperPi runs slowed way down which could only lead me to believe I was at 12 mult. My boot screen on the other hand showed me at 21 mult. Weird thing was I finally got it to stick after I activated Windows 7 but soon screwed up again when I jacked the volts up to 1.4 and beyond. I finally gave up and went back to XP and Vista as this is too weird of a mystery.

[johnp1983]

Here I am with my i7 920 C0, the lanparty beast & Ocz Blade 2000Mhz CAS9.

Tried to go over 1850Mhz... but I've got some instability... crash and bsod.

Can someone post me some ram settings for 2000Mhz CAS9?

Bios 508.

Thanks!!

[Praz]

Here I am with my i7 920 C0, the lanparty beast & Ocz Blade 2000Mhz CAS9.

Tried to go over 1850Mhz... but I've got some instability... crash and bsod.

Can someone post me some ram settings for 2000Mhz CAS9?

Bios 508.

Thanks!!

Need a really strong IMC to run high memory clocks with non-Elpida chips.

[ACE76]

ah thats going to be interesting

eva suppose to test that asus mobo also..

waiting the results on that.

The Asus board is a very good board BUT, when DFI has everything going for them on the BIOS side and the board is working the way it should be, you can't beat that kind of tweaking fun.

[ci2kla]

Here I am with my i7 920 C0, the lanparty beast & Ocz Blade 2000Mhz CAS9.

Tried to go over 1850Mhz... but I've got some instability... crash and bsod.

Can someone post me some ram settings for 2000Mhz CAS9?

Bios 508.

Thanks!!

What VTT Voltage you're in?

[cstkl1]

u need to confirm ure vcore first with long runs of linx say 50 runs with lower ram frequency ..also u need to test it on cold boot , reboot etc.
atleast 3 times.

raising vtt will cause some instability with previously borderline vcore

only then u can confirm ure vdimm and vtt.
to confirm it best way is prime95 largefft custom 5.4gb

[johnp1983]

What VTT Voltage you're in?

Tried 1.25VTT with 1800Mhz CAS8-8-8-22 1.65v-mem

and it goes like heaven...

but a friend who have this main and Ocz Blade 2000 CAS7 say's that i need to set-up the VTT around 1.45 at least for the 2000Mhz.

I'm going to test it...

[cstkl1]

Tried 1.25VTT with 1800Mhz CAS8-8-8-22 1.65v-mem

and it goes like heaven...

but a friend who have this main and Ocz Blade 2000 CAS7 say's that i need to set-up the VTT around 1.45 at least for the 2000Mhz.

I'm going to test it...

vtt is mhz dependent
not latency

u need around 1.39 for 2000mhz.

[drnip]

1.39vtt for 2000mhz, that might be why I get post code 26 half the time. I figured it wasn't enough vtt as I have read that is one of the reasons to get a post code 26 but I didn't think it required that much. Good to know though. Now I can come back at it with a different angle.

Is this a standard whether it is a C0 or D0 stepping or will each CPU vary?

[cstkl1]

1.39vtt for 2000mhz, that might be why I get post code 26 half the time. I figured it wasn't enough vtt as I have read that is one of the reasons to get a post code 26 but I didn't think it required that much. Good to know though. Now I can come back at it with a different angle.

Is this a standard whether it is a C0 or D0 stepping or will each CPU vary?

both but here is the trick

u need to confirm ure vcore
i mean seriously no borderline thing

so far all my clocking on this board with 2000mhz above..
its very paticular on the vcore.
overvolting or slight undervolts can cause serious misinterpretation on what vtt and vdimm or even ioh/ich voltages..

get the vcore right first.

[drnip]

Good to know thanks.

[BLHealthy4life]

both but here is the trick

u need to confirm ure vcore
i mean seriously no borderline thing

so far all my clocking on this board with 2000mhz above..
its very paticular on the vcore.
overvolting or slight undervolts can cause serious misinterpretation on what vtt and vdimm or even ioh/ich voltages..

get the vcore right first.

VERY nice advice and it is very kind of you to post such helpful information.

Based on this thread, I'm going to get a DFI and dump this R2E.

I plan on spending a lot more time in this thread soon.

Thanks for posting guys!

BL

[cstkl1]

VERY nice advice and it is very kind of you to post such helpful information.

Based on this thread, I'm going to get a DFI and dump this R2E.

I plan on spending a lot more time in this thread soon.

Thanks for posting guys!

BL

no offence to this thread
but here is just a lot of posting's

for dfi i find scouring around i4memory and eva posting gives out a lot of hints pretty helpful

[drnip]

VERY nice advice and it is very kind of you to post such helpful information.

Based on this thread, I'm going to get a DFI and dump this R2E.

I plan on spending a lot more time in this thread soon.

Thanks for posting guys!

BL

Oh yeah most definitely dump the Rex 2 as it doesn't live up to the same expectation as the Rex did.

+1 on the i4memory as there are in depth threads on the DFI mobos as well as others.

[BLHealthy4life]

Would it be worthwhile to wait for the B3 revision for this board?

[eva2000]

Would it be worthwhile to wait for the B3 revision for this board?

no one knows when the B3 revision is coming and whether it will be a new model board for respective manufacturers or just revision of existing models....