Guide Thread

Printable View

Show 100 post(s) from this thread on one page

12-18-2007, 04:07 PM
Praz

Quote:

Originally Posted by japanese

but does it or not has the LLC like loadline calibration? or tell me how much is the vdrop

Yes it does.
12-18-2007, 04:09 PM
Praz

Quote:

Originally Posted by johnp1983

The strange things is that if i leave the strap menu to AUTO and go from 266fsb*9 to 355*9 (meaning that the memory will go to 1066Mhz) the system start successfully without touching any NB-SB-VTT voice.

On auto the strap/divider is changing when FSB is scaled.
12-18-2007, 04:13 PM
lazy

hey guys im havin some problem trying to get my Q6600 stable here.
does this cpu really require more then 1.45 for it to be stable 400x9?
12-18-2007, 04:15 PM
japanese

no it doesnt mine with cpu vid 1.2850 i do it stable with 1.432
12-18-2007, 05:03 PM
AKHandyman

Got my X38 up and running real COOL like :cool: ... just now really getting into this mobo and so far I like what I see ... this is my test bench and I just put a little ole AMD 60mm fan on NB heatsink and a 70mm aluminum on the backplate heatsink so that the air blows through and over the RAM :cool: temps are WAY LOW ... :p: (notice window open and temp outside right now is -10F;) )

http://img206.imageshack.us/img206/6...mfandg8.th.jpg http://img204.imageshack.us/img204/2...empsex3.th.jpg

As you can see I am just starting out on my overclock adventure, but hope to achieve some decent/stable OCs so that I can do some gaming in the near future ... :D
12-18-2007, 05:09 PM
SNiiPE_DoGG

nice :) wait till you boost up those voltages on the chipset, you'll wish you had an 80mm (once you close the window that is)
12-18-2007, 05:18 PM
SNiiPE_DoGG

Quote:

Originally Posted by lazy

hey guys im havin some problem trying to get my Q6600 stable here.
does this cpu really require more then 1.45 for it to be stable 400x9?

be sure your cpu vtt is 1.6 in bios and set the the GTL ref to 110/109 for the cpu cores, use these charts to find the gtl for your NB: http://www.edgeofstability.com/artic...fault_gtl.html
12-18-2007, 06:22 PM
lazy

so the vtt needs to be that high even for a 400x9 setup and not a 450x8?
figured since its less fsb i can run it with less vtt voltage
12-18-2007, 07:38 PM
OnDborder

AKHandyman, what is that heatpipe used for, coming out from under the mb, on the right side of the photo?
I like your cooling system(window open with a foot of snow):p:
12-18-2007, 07:49 PM
Ace-a-Rue

Quote:

Originally Posted by Aivas47a

Great info Ace, thanks! :up: (FWIW, Maximus Extreme requires about the same NB and VTT voltages as you are using for quad core stability at 450+ fsb.)

What are you using for gtl reference settings at 450 fsb on the yorkfield?

right now i am using:

CPU GTL 1/3 REF VOLT = 122
CPU GTL 0/2 REF VOLT = 122
NB GTL REF VOLT = 124
12-18-2007, 07:51 PM
Ace-a-Rue

Quote:

Originally Posted by zsamz_

nice work ace
so you likin the board as much as me?

it's growing on me!:D ;)
12-18-2007, 08:02 PM
Ace-a-Rue

Quote:

Originally Posted by BeastNotro

Nice OC Ace-a-Rue and it appears it takes a lot to get 4.0-4.5ghz out of a
QX9650. I am just hoping the lower Yorks QX9450 or QX9550 can do 4.0 really well with low volts on this mobo.

i found up to 4 ghz (1000 mhz increase), i only needed 1.28v for actual vcore (1.15v is default set by this board)...the last 400-500 mhz required a large increase in vcore.
12-18-2007, 08:16 PM
AKHandyman

Quote:

Originally Posted by SNiiPE_DoGG

nice :) wait till you boost up those voltages on the chipset, you'll wish you had an 80mm (once you close the window that is)

Ha ... I have another nicer fan on the way, but this 'll do until then, and Sniipe, what do you recommend I start out with on the volts for my E6850? I was hoping there was a BIOs template for this board w/E6850 ... I've done some easy OC'ing, but this BIOS does have a lot more than I am used to ... :p:
12-18-2007, 08:21 PM
Dark79

I just can't get anything over 400x9 stable. Not even OCCT stable. Here are my current settings for 400x9 which is both OCCT and Prime stable:

Code:

Thermal Management Control Disabled PPM (EIST) Mode Disabled Limit CPUID MaxVal Disabled CIE Function Disabled Execute Disable Bit Disabled Virtualization Technology Disabled Core Multi-Processing Enabled Exist Setup Shutdown Mode 2 CLOCK VC0 divider AUTO CPU Clock Ratio 9x Target CPU Clock 3600MHz CPU Clock 400FSB Boot Up Clock AUTO DRAM Speed 266MHz/667MHz Target DRAM Speed DDR2-1000MHz PCIE Clock 100MHz Voltage Settings CPU VID Control 1.425v CPU VID Special Add AUTO DRAM Voltage Control 2.05v SB Core/CPU PLL Voltage 1.640v NB Core Voltage 1.479v CPU VTT Voltage 1.530v VCore Droop Control Disabled Clockgen Voltage Control 3.45v GTL+ Buffers Strength Strong Host Slew Rate Weak GTL REF Voltage Control Enabled x CPU GTL1/3 REF Volt 106 x CPU GTL 0/2 REF Volt 106 x North Bridge GTL REF Volt 107 DRAM Timing Enhance Data transmitting FAST Enhance Addressing FAST T2 Dispatch Disabled Channel 1 CLK Crossing Setting More Aggressive Channel 2 CLK Crossing Setting More Aggressive CH1CH2 Common CLK Crossing Setting More Aggressive CAS Latency Time (tCL) 5 RAS# to CAS# Delay (tRCD) 5 RAS# Precharge (tRP) 5 Precharge Delay (tRAS) 12 All Precharge to Act AUTO REF to ACT Delay (tRFC) AUTO Performance Level 7 Read delay phase adjust SEE BELOW MCH ODT Latency AUTO Write to PRE Delay (tWR) AUTO Rank Write to Read (tWTR) AUTO ACT to ACT Delay (tRRD) AUTO Read to Write Delay (tRDWR) AUTO Ranks Write to Write (tWRWR) AUTO Ranks Read to Read (tRDRD) AUTO Ranks Write to Read (tWRRD) AUTO Read CAS# Precharge (tRTP) AUTO ALL PRE to Refresh AUTO PCIE Slot Config 1X 1X CPU Spread Spectrum Disabled PCIE Spread Spectrum Disabled SATA Spread Spectrum Disabled read delay phase adjust press enter Channel 1 Phase 0 Pull-In AUTO Channel 1 Phase 1 Pull-In AUTO Channel 1 Phase 2 Pull-In AUTO Channel 1 Phase 3 Pull-In AUTO Channel 1 Phase 4 Pull-In AUTO Channel 2 Phase 0 Pull-In AUTO Channel 2 Phase 1 Pull-In AUTO Channel 2 Phase 2 Pull-In AUTO Channel 2 Phase 3 Pull-In AUTO Channel 2 Phase 4 Pull-In AUTO

I've tried 412x9 for 3.7GHz, but it always fails. I've upped all the voltages and still I can't get it stable. :(

I'll have 3DMark scores for the HD2900Pro at these settings later.
12-18-2007, 08:22 PM
AKHandyman

Quote:

Originally Posted by OnDborder

AKHandyman, what is that heatpipe used for, coming out from under the mb, on the right side of the photo?
I like your cooling system(window open with a foot of snow):p:

That is the backplate that comes with the Thermalright IFX-14 ... it is actually sandwiched between the back of the mobo and the backplate bracket under where the CPU sits ... I guess it helps cool the CPU as well ... so far I haven't seen any temps on either core go over 50C (I am actually doing prime right now with a tweak to find some stable voltage for the NB ... :) )
12-18-2007, 08:29 PM
AKHandyman

Quote:

Originally Posted by Dark79

I just can't get anything over 400x9 stable. Not even OCCT stable. Here are my current settings for 400x9 which is both OCCT and Prime stable ... I've tried 412x9 for 3.7GHz, but it always fails. I've upped all the voltages and still I can't get it stable. :(

What brand of RAM are you using? You know, I had a similar problem with my DFI Inf 975 board where I couldn't go past 400x9 without a good prime run making it crash ... I just kind of let it go as it is my everyday 24/7 rig, but this baby I want to really mess with and see what I can do ... :D
12-18-2007, 08:48 PM
Ace-a-Rue

Quote:

Originally Posted by Dark79

I just can't get anything over 400x9 stable. Not even OCCT stable. Here are my current settings for 400x9 which is both OCCT and Prime stable:

I've tried 412x9 for 3.7GHz, but it always fails. I've upped all the voltages and still I can't get it stable. :(

I'll have 3DMark scores for the HD2900Pro at these settings later.

you are fustrating yourself over WHAT!?...being OCCT or P95 stable!...if you are not folding or playing heavy cpu demanding games, you are wasting electrical energy and time...it proves nothing to me...use the wPrime bench...if you hack that you are fairly stable....if anything happens after proving wPrime is valid then it is a memory timing issue.

stable is an elusive word...each of us have different definitions!...i say stable is: what ever you do and your OS doesn't freeze; get BSOD; programs close out on their own; random reboots or games drop back to your desktop.
12-18-2007, 09:17 PM
zanzabar

dose the default fsb changer actually work on this board, ide be interested in just getting a higher default clock and working from there, like taking an e6850 and bumping it from 333-400 default and then getting started

and is there any way to lock the strap on the board or could that default fsb limit them
12-18-2007, 09:24 PM
Ace-a-Rue

Quote:

Originally Posted by zanzabar

dose the default fsb changer actually work on this board, ide be interested in just getting a higher default clock and working from there, like taking an e6850 and bumping it from 333-400 default and then getting started

and is there any way to lock the strap on the board or could that default fsb limit them

if you mean only increasing your FSB and nothing else, "Yes!"
12-18-2007, 09:43 PM
Aivas47a

Quote:

Originally Posted by Ace-a-Rue

right now i am using:

CPU GTL 1/3 REF VOLT = 122
CPU GTL 0/2 REF VOLT = 122
NB GTL REF VOLT = 124

Thanks! Looking forward to trying this board out.
12-18-2007, 10:39 PM
lazy

Quote:

Originally Posted by Dark79

I just can't get anything over 400x9 stable. Not even OCCT stable. Here are my current settings for 400x9 which is both OCCT and Prime stable:

Code:

Thermal Management Control Disabled PPM (EIST) Mode Disabled Limit CPUID MaxVal Disabled CIE Function Disabled Execute Disable Bit Disabled Virtualization Technology Disabled Core Multi-Processing Enabled Exist Setup Shutdown Mode 2 CLOCK VC0 divider AUTO CPU Clock Ratio 9x Target CPU Clock 3600MHz CPU Clock 400FSB Boot Up Clock AUTO DRAM Speed 266MHz/667MHz Target DRAM Speed DDR2-1000MHz PCIE Clock 100MHz Voltage Settings CPU VID Control 1.425v CPU VID Special Add AUTO DRAM Voltage Control 2.05v SB Core/CPU PLL Voltage 1.640v NB Core Voltage 1.479v CPU VTT Voltage 1.530v VCore Droop Control Disabled Clockgen Voltage Control 3.45v GTL+ Buffers Strength Strong Host Slew Rate Weak GTL REF Voltage Control Enabled x CPU GTL1/3 REF Volt 106 x CPU GTL 0/2 REF Volt 106 x North Bridge GTL REF Volt 107 DRAM Timing Enhance Data transmitting FAST Enhance Addressing FAST T2 Dispatch Disabled Channel 1 CLK Crossing Setting More Aggressive Channel 2 CLK Crossing Setting More Aggressive CH1CH2 Common CLK Crossing Setting More Aggressive CAS Latency Time (tCL) 5 RAS# to CAS# Delay (tRCD) 5 RAS# Precharge (tRP) 5 Precharge Delay (tRAS) 12 All Precharge to Act AUTO REF to ACT Delay (tRFC) AUTO Performance Level 7 Read delay phase adjust SEE BELOW MCH ODT Latency AUTO Write to PRE Delay (tWR) AUTO Rank Write to Read (tWTR) AUTO ACT to ACT Delay (tRRD) AUTO Read to Write Delay (tRDWR) AUTO Ranks Write to Write (tWRWR) AUTO Ranks Read to Read (tRDRD) AUTO Ranks Write to Read (tWRRD) AUTO Read CAS# Precharge (tRTP) AUTO ALL PRE to Refresh AUTO PCIE Slot Config 1X 1X CPU Spread Spectrum Disabled PCIE Spread Spectrum Disabled SATA Spread Spectrum Disabled read delay phase adjust press enter Channel 1 Phase 0 Pull-In AUTO Channel 1 Phase 1 Pull-In AUTO Channel 1 Phase 2 Pull-In AUTO Channel 1 Phase 3 Pull-In AUTO Channel 1 Phase 4 Pull-In AUTO Channel 2 Phase 0 Pull-In AUTO Channel 2 Phase 1 Pull-In AUTO Channel 2 Phase 2 Pull-In AUTO Channel 2 Phase 3 Pull-In AUTO Channel 2 Phase 4 Pull-In AUTO

I've tried 412x9 for 3.7GHz, but it always fails. I've upped all the voltages and still I can't get it stable. :(

I'll have 3DMark scores for the HD2900Pro at these settings later.

im having the same problem with mine also, although ive tried up to 1.47 volts for 400x9 setup

And yes i will be running wcg on this when im not personally on it so i need something stable..

I just got a pos Q6600 it pisses me off
12-18-2007, 10:48 PM
AKHandyman

Well, after tweaking a bit with some of the voltages, and running a bit of Prime95 for a few hours, I think I have at least a good starting point ... thanks guys ... hope to post a few screenies of solid/stable work ... :D
12-18-2007, 10:58 PM
zsamz_

here is a bios template for e6850
i'm sure volts can be reduced but that for later
can do either do 1000mhz or 1200mhz 5-5-5-15

cant go higher than 500fsb need more time n new bios

Code:

Thermal Management Control Disabled PPM (EIST) Mode Disabled Limit CPUID MaxVal Disabled CIE Function Disabled Execute Disable Bit Disabled Virtualization Technology Disabled Core Multi-Processing Enabled Exist Setup Shutdown Mode 2 CLOCK VC0 divider AUTO CPU Clock Ratio 8x Target CPU Clock 4000MHz CPU Clock 500FSB Boot Up Clock AUTO DRAM Speed 333MHz/667MHz Target DRAM Speed DDR2-1000MHz PCIE Clock 100MHz Voltage Settings CPU VID Control 1.531v CPU VID Special Add AUTO DRAM Voltage Control 2.3v less dependin on memory SB Core/CPU PLL Voltage 2.0v NB Core Voltage 1.6v CPU VTT Voltage 1.5360v VCore Droop Control Disabled Clockgen Voltage Control 3.45v GTL+ Buffers Strength Strong Host Slew Rate Weak GTL REF Voltage Control Enabled x CPU GTL1/3 REF Volt 105 x CPU GTL 0/2 REF Volt 105 x North Bridge GTL REF Volt 111 DRAM Timing Enhance Data transmitting FAST Enhance Addressing FAST T2 Dispatch Disabled Channel 1 CLK Crossing Setting More Aggressive Channel 2 CLK Crossing Setting More Aggressive CH1CH2 Common CLK Crossing Setting More Aggressive CAS Latency Time (tCL) 4 RAS# to CAS# Delay (tRCD) 4 RAS# Precharge (tRP) 4 Precharge Delay (tRAS) 12 All Precharge to Act AUTO REF to ACT Delay (tRFC) AUTO Performance Level 8 Read delay phase adjust SEE BELOW MCH ODT Latency AUTO Write to PRE Delay (tWR) AUTO Rank Write to Read (tWTR) AUTO ACT to ACT Delay (tRRD) AUTO Read to Write Delay (tRDWR) AUTO Ranks Write to Write (tWRWR) AUTO Ranks Read to Read (tRDRD) AUTO Ranks Write to Read (tWRRD) AUTO Read CAS# Precharge (tRTP) AUTO ALL PRE to Refresh AUTO PCIE Slot Config 1X 1X CPU Spread Spectrum Disabled PCIE Spread Spectrum Disabled SATA Spread Spectrum Disabled read delay phase adjust press enter Channel 1 Phase 0 Pull-In AUTO Channel 1 Phase 1 Pull-In AUTO Channel 1 Phase 2 Pull-In AUTO Channel 1 Phase 3 Pull-In AUTO Channel 1 Phase 4 Pull-In AUTO Channel 2 Phase 0 Pull-In AUTO Channel 2 Phase 1 Pull-In AUTO Channel 2 Phase 2 Pull-In AUTO Channel 2 Phase 3 Pull-In AUTO Channel 2 Phase 4 Pull-In AUTO

gl
when i get more time more tweakin to come:D
12-18-2007, 11:44 PM
AKHandyman

Here is my first memory bench ... not too bad for a beginner ... but I know I have a long way to go before I can become an Xtreme Overclocker ... :p:

http://img178.imageshack.us/img178/8...enchsq1.th.jpg
12-19-2007, 12:15 AM
johnp1983

Quote:

Originally Posted by zsamz_

set gtl settings 68% to 70% of voltage settings
on auto it will be set @ 67% its too low
cpu vtt must be set slightly higher than cpu v
improper gtl setting will give you many problems
gimme your voltage settings i'll give you gtl settings
btw i'm also new to dfi lol

Hello zsamz, maybe the problem can be found in the gtl settings. I leave it on "disabled".
I was trying two type of config: one with the vcore at 1.42 that I use for 425*8, and another with the vcore at 1.47 for the 450*8.

At this point I don't know how to set the memory strap because everytime i move it the mobo doesn't start on the next boot... even if i try a 267*9 with 200\800 (as example).

By the way, can you give me some trick about the NB e SB voltages? What is the maximum vcore that I can give to them? Are they involved with my faulty boot?

Thank you for your help! :)

Quote:

Originally Posted by Praz

On auto the strap/divider is changing when FSB is scaled.

Mmmh, so what can I do? Maybe I can try first with the settings reported zsamz?

Bye e thank you! :)
12-19-2007, 01:35 AM
zsamz_

Quote:

Originally Posted by johnp1983

Hello zsamz, maybe the problem can be found in the gtl settings. I leave it on "disabled".
I was trying two type of config: one with the vcore at 1.42 that I use for 425*8, and another with the vcore at 1.47 for the 450*8.

At this point I don't know how to set the memory strap because everytime i move it the mobo doesn't start on the next boot... even if i try a 267*9 with 200\800 (as example).

By the way, can you give me some trick about the NB e SB voltages? What is the maximum vcore that I can give to them? Are they involved with my faulty boot?

Thank you for your help! :)

Mmmh, so what can I do? Maybe I can try first with the settings reported zsamz?

Bye e thank you! :)

cpu vid 1.41 : cpu vid 1.465
sb core pll1,7 : sb core pll 1.7v
cpuvtt 1,42 : cpuvtt 1.47
nb core 1.5 : nb core 1.55
gtl cpu 98 : gtl cpu 100
gtl cpu 98 : gtl cpu 100
nb gtl 103 : nb gtl 106

try that gl
12-19-2007, 02:23 AM
johnp1983

God bless you zsamz_

The system boot correctly.

I've tried first a clearcmos by moving the jp12 jumper as it's explained on the manual,

then I tried your first config. (cpu vid 1.41)

selecting strap dram to 266/667

and as i said some words ago the system boot correctly not giving me anymore the long constantly beep (dram error) on the post.

Now I'm gonna do some work on the vid to see if I can be stable by lowering them. Then I move on to reach some fsb good frequencies.

Thank you alot zsamz_, your help has been critical to solve my problem. :bows: :rocker:

Have a nice day! :)
12-19-2007, 03:06 AM
eva2000

Quote:

Originally Posted by Praz

TRD Phase Control alters the Performance Level setting. If you set a PL of six with the 10 phases set to auto the PL stays at six. If you enable the 10 phases the performance level will be 5. TRD phase adjustment is a way to get a Performance Level in-between what can be set with the PL setting. To maximize the performance increase the same phase on both channels should be enabled. Depending on stability though this is not always possible.

I found the last 3 Phase pull-in settings on each channel don't do much at all for memory performance or latencies when enabled. Only the first 2 phase pull-ins can alter performance. Maybe because my subtimings are already tight ??

Read delay phases all AUTO

http://fileshosts.com/intel/DFI/DFI_...ndwidth_tn.png

http://fileshosts.com/intel/DFI/DFI_...3m15s406ms.png

Read delay phases
- last 3 Phase pull-in settings enabled
- 1st and 2nd Phase pull-in settings AUTO

http://fileshosts.com/intel/DFI/DFI_..._bandwidth.png

http://fileshosts.com/intel/DFI/DFI_...3m15s953ms.png
12-19-2007, 05:02 AM
CapFTP

eva, when I tested on P35, the most heavy effects were on FAST-FAST and the 3 MORE Aggressive settings on DRAM TImings.

the delay phases just had a small impact and morever a more serious impact on stabilty...i'll check again on X38 ...
12-19-2007, 05:06 AM
Praz

Quote:

Originally Posted by eva2000

I found the last 3 Phase pull-in settings on each channel don't do much at all for memory performance or latencies when enabled. Only the first 2 phase pull-ins can alter performance. Maybe because my subtimings are already tight ??

I'm thinking maybe a bug? At least with the BIOS version I currently have loaded. This is what I've noticed with MemSet. With PL and all phase delays set on auto and the correct configuration the BIOS will set the following. All phases of one channel to 4 and all phases except one or two to 4 on the other channel with the remaining set to 5. Memset will show a PL of 4 for one channel and 5 for the other which means the PL is being set to 5.

However, if i manually configure these settings Memset will show 5 for both channels as if the manually set phases are being ignored. I haven't tries leaving PL set to auto and manually setting the phases yet.
12-19-2007, 05:19 AM
eva2000

Quote:

Originally Posted by CapFTP

eva, when I tested on P35, the most heavy effects were on FAST-FAST and the 3 MORE Aggressive settings on DRAM TImings.

the delay phases just had a small impact and morever a more serious impact on stabilty...i'll check again on X38 ...

thanks looking forward to your results :)

Quote:

Originally Posted by Praz

I'm thinking maybe a bug? At least with the BIOS version I currently have loaded. This is what I've noticed with MemSet. With PL and all phase delays set on auto and the correct configuration the BIOS will set the following. All phases of one channel to 4 and all phases except one or two to 4 on the other channel with the remaining set to 5. Memset will show a PL of 4 for one channel and 5 for the other which means the PL is being set to 5.

However, if i manually configure these settings Memset will show 5 for both channels as if the manually set phases are being ignored. I haven't tries leaving PL set to auto and manually setting the phases yet.

with 11/28 and PL = 5 set, read delay phase adjust when set to AUTO all = 5 for both channels and memset 3.4 final shows PL = 5. When last 3 phase pull-ins enabled they show 4 instead of 5 for each.
12-19-2007, 05:55 AM
Praz

Quote:

Originally Posted by eva2000

with 11/28 and PL = 5 set, read delay phase adjust when set to AUTO all = 5 for both channels and memset 3.4 final shows PL = 5. When last 3 phase pull-ins enabled they show 4 instead of 5 for each.

I'll flash back to the 11/28 and try the same settings.
12-19-2007, 06:15 AM
CapFTP

Praz, only 2 bios available 11/28 and 12/18 right ?

I am currently running with 11/28 as bios has been shipped....any reason to change to newer ?
:)
12-19-2007, 07:08 AM
BeastNotro

Quote:

Originally Posted by sofarfrome

I do not believe there will be a QX9450 or a QX9550. The X means Xtreme with an unlocked multi. The QX's will be QX9650 and QX9770.

Ah yes .. i always use the X in reference to Quad-Core when i post as oppose to the Dual-Core when i post, some ppl think Q6850 means dual-core, but infact it is a Quad-Core so i use QX to mean 100% quad-core when i am posting.

Yes i know there will be no Extreme lower Yorks made.
12-19-2007, 07:15 AM
BeastNotro

Quote:

Originally Posted by Ace-a-Rue

you are fustrating yourself over WHAT!?...being OCCT or P95 stable!...if you are not folding or playing heavy cpu demanding games, you are wasting electrical energy and time...it proves nothing to me...use the wPrime bench...if you hack that you are fairly stable....if anything happens after proving wPrime is valid then it is a memory timing issue.

stable is an elusive word...each of us have different definitions!...i say stable is: what ever you do and your OS doesn't freeze; get BSOD; programs close out on their own; random reboots or games drop back to your desktop.

Well not quite true there about not being P95 or Orthos prime stable as it must be stable for at least 12-24hrs.

I have builds run P95 for say 10 hours and fail... i say ok fuk it i will run it... and yeppers... bam games froze up, and even at times when surfing the net the damn thing would freeze up and running 3dmark 06 PASSED with flying colors.....and so did 3dmark 05... but 3dmark 03 Locked her up.

Well the thing was .... the CPU needed more voltage just a :banana::banana::banana::banana: hair more and then i ran prime 95 for 18hrs before i decided she was good enough.... guess what i am still running strong on that build... the one i am using now... :)

I have had many ppl in the past :banana::banana::banana::banana::banana: about P95 not being stable, well guys all i can say is if it can't do at least 12-24hrs... basically 12-18hrs stable.. then it is no good in my book for 100% stability.

Now if your just a surfer on the net and play the odd game.... GAWD why bother overclocking at all.... ???
12-19-2007, 08:15 AM
Praz

Quote:

Originally Posted by BeastNotro

I have had many ppl in the past :banana::banana::banana::banana::banana: about P95 not being stable, well guys all i can say is if it can't do at least 12-24hrs... basically 12-18hrs stable.. then it is no good in my book for 100% stability.

Now if your just a surfer on the net and play the odd game.... GAWD why bother overclocking at all.... ???

Ace is just trying to help you not kill your hardware. I run Prime as much or more then most do testing these boards. Just a few observations. Even passing Prime for 24 hours is no guarantee that some other program will not have stability issues. But more importantly, is what Prime does. Nothing in the consumer market loads the cores as much as Prime with the exception of something like a folding program. None of these boards from any manufacturer are designed to be constantly hammered like this. Believe it or not they will die.

This a reason why quadcores are so difficult to get Prime stable for an extended period of time. The components are being stressed beyond their design parameters. The longer all four cores stay loaded at 100% the more heat there is generated. Ignore what software temperatures are. It's localized hot spots affecting both the board and components. This heat causes electron migration which results in degraded and/or failed components.

We are trying to prove stability with a program that was originally used for that purpose when components didn't even require active cooling to operate. Constantly beating on today's hardware with Prime is similar to running your car motor at 10,000 rpm for several hours to prove to yourself that it's worthy of the 100 mile trip you're getting ready to make.

In the end it's a choice the user has to make. But from experience, repeated, extended runs of Prime will do nothing for the longevity of hardware.
12-19-2007, 09:19 AM
BeastNotro

No doubt running any program for long instances 24/7 full load will sooner or later damage componets this includes running any Folding program for years and years and years.. heck even months on end will limit the life somewhat of componets.

I am just stating run prime 95 or orthos prime (is what i run) on the Stress CPU and Ram setting which is the default setting for Orthos for at least 12-18hours anyone i know who has done so including myself have never had any issues after that.

Yes sure any program or game can have issues, but that does not always mean it is a OC issue as even CRYSIS is buggy for me and many others cause it is a GAME issue. The problem is when a NEWBIE has such issues they automaticlaly assume it is a hardware issue or a overclocking issue.

I only had one game that had a OC issue after priming for 18hrs and that was HL2 cause of Valve and their stupid update to the steam program at that time about 1-2 years ago frigged it up for so many who were OCing their PC's, but Valve fixed that issue in a hurry cause of the Violent posts in their forums and across the internet about this OC throttling they made happen.

So i say run prime 95 in which the latest version allows for 2-4 cores run without running 2 or 4 instances of the program to prime all 4 cores.

Orthos on the other hand you will need to run 2 instances of it in order to run all 4 cores 100%, that is for now until a updated version comes out, but may i remind people it was Orthos who came out with a Dual prime program NOT Prime 95 makers and it only took 4 years for prime 95 makers to come out with a new release that supported 2-4 cores right out of the Box sort of speak.

Orthos prime is PRIME 95 coding, but the developer made a GUI for it is all and then updated it to run 2 cores at a time when dual-cores came out, but he has been lazy ... :) in not updating it to run 4 cores out of the Box sort of speak.... sooner or later i am sure he will.

Now if you have a weak componet in your build.. example a weak PSU or Video card then priming is not for you, neiter is OCing or even running a Folding program.. cause all you will be doing is hindering those weak componets to fail even faster.

Weak meaning.... examples a $65.00 500w PSU or a X1300 video card or a $90.00 low-end motherboard, but you have a Q6600 and 2GB's of memory.... don't prime that thing as your going to hurt your weak componets in which if they get damaged during the process they can/could take something else out with it.
12-19-2007, 10:41 AM
AKHandyman

I was just wondering if the instant e-mail notification is working ... I have subscribed to this thread, but I never received any notices about the additional posts when I accessed my email this morning ... :( Anyone know? I've checked all my settings and everything seems to be in order ... :)
12-19-2007, 10:48 AM
Praz

I've never received email notification from day 1. Think it's broke.
12-19-2007, 10:50 AM
Ace-a-Rue

Quote:

Originally Posted by AKHandyman

I was just wondering if the instant e-mail notification is working ... I have subscribed to this thread, but I never received any notices about the additional posts when I accessed my email this morning ... :( Anyone know? I've checked all my settings and everything seems to be in order ... :)

it was broken for a long time and now it works very sporadically!
12-19-2007, 11:02 AM
BeastNotro

Its probably like other sites (broke on purpose) to get you to come back here everyday for another Bonus in google page rankings.... :)
12-19-2007, 11:52 AM
shawn1998

So, I read mention of a new bios, from Dec 18th?
12-19-2007, 12:15 PM
AKHandyman

Quote:

Originally Posted by shawn1998

So, I read mention of a new bios, from Dec 18th?

Yeah, is that true? If so, what has been done to it?
12-19-2007, 12:18 PM
AKHandyman

Quote:

Originally Posted by Ace-a-Rue

it was broken for a long time and now it works very sporadically!

I was wondering about that ... thought it was, but you never know ... maybe because there are so many threads and posts and such ... it probably overwhelms the servers ... :eek:
12-19-2007, 12:40 PM
Brother Esau

No .Bin.file for this board yet hey?
12-19-2007, 01:00 PM
shawn1998

Well, found the bios on the DFI website and flashed. One thing it solved, on the negative side, is that my Areca 1210 card will no longer work in the 2nd pci-e 16X slot, under any circumstances. I can get it to boot once in awhile, but only at 1X. Before, on the 11/28 bios, as long as I had both floppy power connectors unplugged, it would work at 8X. Now, with both unplugged, it only works at 1X. So, I benchmarked the card in 4X and 8X and found absolutely no difference in performance, so it really doesn't matter. I stuck it back into the top 4X slot and all is well (with both floppy connectord plugged in).
12-19-2007, 01:05 PM
BeastNotro

DFI's website guys go there for bios, don't be so lazy.. :)

Heres the link to the Bios page for this mobo.

http://us.dfi.com.tw/Support/Downloa...FLAG=A&SITE=US

Major Reasons of Change:
Improve Graphic Cards compatibility issue
12-19-2007, 01:16 PM
shawn1998

The reason I asked, before heading over to the website, is that usually there are posts about it's release, feedback and a link on the first page. There was none of this yet, so I thought it might be a "private" release of sorts.
12-19-2007, 01:22 PM
C-N

Notifications work fine for me although sometimes they stop & you have to unsubscribe & re-subscribe to the effected threads.

CN :)
12-19-2007, 01:23 PM
Brother Esau

1 Attachment(s)

Anybody that would Like to get a Extra Pre Programmed Bios Chip for the D.F.I LanParty LT X38 to have in case of Bios Corruption or Bios Flash gone wrong the doors are now open for Orders:up: Hurry up they move FAST:sonic: & I only have 15 left:)
12-19-2007, 01:30 PM
BeastNotro

How much and where and how i/we can get one if need to be?
12-19-2007, 01:32 PM
sofarfrome

Quote:

Originally Posted by BeastNotro

Well not quite true there about not being P95 or Orthos prime stable as it must be stable for at least 12-24hrs.

I have builds run P95 for say 10 hours and fail... i say ok fuk it i will run it... and yeppers... bam games froze up, and even at times when surfing the net the damn thing would freeze up and running 3dmark 06 PASSED with flying colors.....and so did 3dmark 05... but 3dmark 03 Locked her up.

Well the thing was .... the CPU needed more voltage just a :banana::banana::banana::banana: hair more and then i ran prime 95 for 18hrs before i decided she was good enough.... guess what i am still running strong on that build... the one i am using now... :)

I have had many ppl in the past :banana::banana::banana::banana::banana: about P95 not being stable, well guys all i can say is if it can't do at least 12-24hrs... basically 12-18hrs stable.. then it is no good in my book for 100% stability.

Now if your just a surfer on the net and play the odd game.... GAWD why bother overclocking at all.... ???

My 2-cents...I have had systems with high OCs that pass all of the available benchmarks including all 3D benchmarks and P95 for 24 hours. Load up UT2004 and play, system locks up within 10 minutes, or less. Bump vcore a smidge and all is well. I will no longer go for ridiculous P95 times. I will run all 3D benchmarks first to get a clue on what the OC will be then run prime on small ffts for about an hour. If it passes then it is off to the most stressful game I can get my hands on. I guess I should go out and get crysis or the new UT for testing.
12-19-2007, 01:33 PM
Brother Esau

Bios Chip sales you must talk in PM please they have busted my balls once already:D because I had it advertising in my siggy new rules coming down on forum lately:)
12-19-2007, 01:56 PM
Praz

Quote:

Originally Posted by sofarfrome

My 2-cents...I have had systems with high OCs that pass all of the available benchmarks including all 3D benchmarks and P95 for 24 hours. Load up UT2004 and play, system locks up within 10 minutes, or less. Bump vcore a smidge and all is well. I will no longer go for ridiculous P95 times.

Exactly. Hours and hours of running anything only proves that particular program is stable. A quick check of stability is looping 3dMark for a couple of hours with Prime or OCCT running in the background.
12-19-2007, 01:57 PM
Ace-a-Rue

Quote:

Originally Posted by shawn1998

Well, found the bios on the DFI website and flashed. One thing it solved, on the negative side, is that my Areca 1210 card will no longer work in the 2nd pci-e 16X slot, under any circumstances. I can get it to boot once in awhile, but only at 1X. Before, on the 11/28 bios, as long as I had both floppy power connectors unplugged, it would work at 8X. Now, with both unplugged, it only works at 1X. So, I benchmarked the card in 4X and 8X and found absolutely no difference in performance, so it really doesn't matter. I stuck it back into the top 4X slot and all is well (with both floppy connectord plugged in).

SHAWN,

have you tried putting the areca card in slot 1 and your VC in slot 2?...i did that with the original bios without any aux. 5/12v power and it worked just great..8x bandwidth for the areca card and 16x for the VC...posted and booted fine and i was able to access the areca bios during post regimen.

have you tried the above with the newest bios?

i found the areca card will get higher burst speed at 8x bandwidth over the 4x slot.
12-19-2007, 02:08 PM
CapFTP

anyone has tried such settings ?

FSB=400
333/1066 (5:8) that means DDR2-1280
200/667 (3:5) that means DDR2-1333

?

my mobo hangs up with this settings on C1 and whistle...:D:D
12-19-2007, 02:16 PM
Brother Esau

@ ACE......Hows you're ARECA Raid Controller working in this motherboard please:)
12-19-2007, 02:24 PM
zanzabar

esau can test xfire in the 16x adn 4x slot i was thinking of getting an areca and wanted to see numbers with pci-e2 and a pci-e1 4x
12-19-2007, 02:27 PM
Brother Esau

PCI-E 2.0 does not do much in terms of performance because the cards can't utilize the Bandwidth headroom right now not the way they will in up coming GFX Cards releases:D

You'll get about 800mb Burst in X4
12-19-2007, 03:08 PM
shawn1998

Quote:

Originally Posted by Ace-a-Rue

SHAWN,

have you tried putting the areca card in slot 1 and your VC in slot 2?...i did that with the original bios without any aux. 5/12v power and it worked just great..8x bandwidth for the areca card and 16x for the VC...posted and booted fine and i was able to access the areca bios during post regimen.

have you tried the above with the newest bios?

i found the areca card will get higher burst speed at 8x bandwidth over the 4x slot.

I did that with the original bios as well, it worked. I have not tried with the newest bios, however, so I couldn't tell you if it still works or not. Maybe if I am bored later tonight I'll try it out.

I have tested with it in 4X and compared the results with the results from *x and there was no change at all. I plan on adding the 4th Raptor this weekend, so I will bench it in 4X and see what happens. If what Brother Esau says is true, which I believe it is, and I can see up to 800mb burst speed in 4X, then I don't honestly think 8X will be any better than 4X for the 4 Raptors I will be running. Who knows, though, I try it out and let you know. Thanks for the info guys!
12-19-2007, 03:12 PM
Praz

Code:

CPU Feature Page Thermal Management Control................Enabled PPM (EIST) Mode...........................Enabled Limit CPUID MaxVal........................Disabled CIE Function..............................Auto Execute Disable Bit.......................Enabled Virtualization Technology.................Enabled Core Multi-Processing.....................Enabled Main BIOS Page Exist Setup Shutdown......................Mode 2 Shutdown After AC Loss....................Disabled O. C. Fail Retry Counter..................0 CLOCK VC0 Divider.........................Auto CPU Clock Ratio...........................10x CPU Clock.................................266 MHz Boot Up Clock.............................Auto DRAM Speed................................Auto PCIE Clock................................100MHz PCIE Slot Config..........................1X 1X CPU Spread Spectrum.......................Disabled PCIE Spread Spectrum......................Disabled SATA Spread Spectrum......................Disabled Voltage Setting Page CPU VID Control...........................Auto CPU VID Special Add.......................Auto DRAM Voltage Control......................1.904V SB Core/CPU PLL Voltage...................1.510V NB Core Voltage...........................1.265V CPU VTT Voltage...........................1.211V VCore Droop Control.......................Enabled Clockgen Voltage Control..................3.45V GTL+ Buffers Strength.....................Strong Host Slew Rate............................Weak GTL REF Voltage Control...................Disabled CPU GTL1/3 REF Volt.......................110 CPU GTL 0/2 REF Volt......................110 North Bridge GTL REF Volt ................110 DRAM Timing Page Enhance Data Transmitting.................Auto Enhance Addressing........................Auto T2 Dispatch...............................Auto Clock Setting Fine Delay..................Listed Below CAS Latency Time (tCL)....................Auto RAS# to CAS# Delay (tRCD).................Auto RAS# Precharge (tRP)......................Auto Precharge Delay (tRAS)....................Auto All Precharge to Act......................Auto REF to ACT Delay (tRFC)...................Auto Performance Level.........................Auto Read Delay Phase Adjust...................Listed Below MCH ODT Latency...........................Auto Write to PRE Delay (tWR)..................Auto Rank Write to Read (tWTR).................Auto ACT to ACT Delay (tRRD)...................Auto Read to Write Delay (tRDWR)...............Auto Ranks Write to Write (tWRWR)..............Auto Ranks Read to Read (tRDRD)................Auto Ranks Write to Read (tWRRD)...............Auto Read CAS# Precharge (tRTP)................Auto ALL PRE to Refresh........................Auto Read Delay Phase Adjust Page Channel 1 Phase 0 Pull-In.................Auto Channel 1 Phase 1 Pull-In.................Auto Channel 1 Phase 2 Pull-In.................Auto Channel 1 Phase 3 Pull-In.................Auto Channel 1 Phase 4 Pull-In.................Auto Channel 2 Phase 0 Pull-In.................Auto Channel 2 Phase 1 Pull-In.................Auto Channel 2 Phase 2 Pull-In.................Auto Channel 2 Phase 3 Pull-In.................Auto Channel 2 Phase 4 Pull-In.................Auto Clock Setting Fine Delay Page Ch1 Clock Crossing Setting................Auto DIMM 1 Clock fine delay...................Current[14] DIMM 2 Clock fine delay...................Current[5] DIMM 1 Control fine delay.................Current[9] DIMM 2 Control fine delay.................Current[7] Ch 1 Command fine delay...................Current[9] Ch2 Clock Crossing Setting................Auto DIMM 3 Clock fine delay...................Current[13] DIMM 4 Clock fine delay...................Current[3] DIMM 3 Control fine delay.................Current[2] DIMM 4 Control fine delay.................Current[4] Ch 2 Command fine delay...................Current[10] Ch1Ch2 CommonClock Setting................Auto Ch1 RDCAS GNT-Chip Delay..................Auto Ch1 WRCAS GNT-Chip Delay..................Auto Ch1 Command to CS Delay...................Auto Ch2 RDCAS GNT-Chip Delay..................Auto Ch2 WRCAS GNT-Chip Delay..................Auto Ch2 Command to CS Delay...................Auto
12-19-2007, 03:23 PM
eva2000
Clock Fine Delay At Work

Still very early into my testing of how memory behave on DFI LP LT X38-T2R with 11/28 bios. It basically seems like a more matured version of what is implemented on DFI LP UT P35-T2R :)

Playing with 2:3 divider (266/800) and 2x 1GB Crucial Ballistix PC2-8500, I found myself hitting memory clock wall at several points while doing my routine Memtest86+ v1.70 testing.

Only salvation to break through that memory clock wall was fine tuning Clock Fine Delay values for DIMM 1 & 3. To my surprise I broke through that 651Mhz wall and ended up at 666Mhz 5-5-5-9 at 2.49v being single Super Pi 32M stable!
- 633mhz 5-5-5-9 at 2.27v was okay
- 640mhz 5-5-5-9 at 2.31v was okay
- 651mhz 5-5-5-9 regardless of vdimm used kept freezing in test #5 loop testing - clock fine delay was all set to Current.
- 651mhz 5-5-5-9 at 2.38v errored out in memtest, but elimated freezing by increasing dimm 1 + 3 clock fine delay values from Current (2) to manually set 3.
- 660mhz 5-5-5-9 at 2.45v was okay in memtest now by further increasing dimm 1 + 3 clock fine delay values from manually set 3 to 5.
- 666mhz 5-5-5-9 at 2.49v would error out in test #5 loop on 2nd pass with a few errors everytime, no voltage adjustments helped.
- 666mhz 5-5-5-9 at 2.49v was okay when i further increased dimm 1 + 3 clock fine delay values from manually set 5 to 6 or 7.
- 670mhz 5-5-5-9 at 2.49v would error out in test #5 loop on 2nd pass with a few errors everytime, voltage adjustments didn't help much to stablise it. Dimm 1 + 3 clock fine delay values manually set at 7 helped reduce it to a few errors in memtest86+ v1.70 test #5 loop.
So here's where I end up at so far 666Mhz 5-5-5-9 at 2.49v! Pretty awesome for initially having hit a brick wall at 651Mhz 5-5-5-9!

http://fileshosts.com/intel/DFI/DFI_...cpuz_valid.png

http://fileshosts.com/intel/DFI/DFI_...3m20s110ms.png

http://fileshosts.com/intel/DFI/DFI_..._bandwidth.png

DFI LP LT X38-T2R Bios Settings

Quote:

PC Health Status
Adjust CPU Temp: +7C

CPU Feature
- Thermal Management Control: Disabled
- PPM(EIST) Mode: Disabled
- Limit CPUID MaxVal: Disabled
- CIE Function: Disabled
- Execute Disable Bit: Disabled
- Virtualization Technology: Disabled
- Core Multi-Processing: Enabled

Exist Setup Shutdown: Mode 2
Shutdown after AC Loss: Disabled
CLOCK VC0 divider: AUTO
CPU Clock Ratio Unlock: Enabled
CPU Clock Ratio: 8x
- Target CPU Clock: 3552
CPU Clock: 444
Boot Up Clock: AUTO
DRAM Speed: 266/800
- Target DRAM Speed: 1335
PCIE Clock: 100mhz
PCIE Slot Config: 1X 1X

CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled
SATA Spread Spectrum: Disabled

Voltage Settings
CPU VID Control: 1.2875
CPU VID Special Add: AUTO
DRAM Voltage Control: 2.49
SB Core/CPU PLL Voltage: 1.51
NB Core Voltage: 1.643
CPU VTT Voltage: 1.377
Vcore Droop Control: Enabled
Clockgen Voltage Control: 3.45v
GTL+ Buffers Strength: Strong
Host Slew Rate: Weak
GTL REF Voltage Control: Disable
x CPU GTL1/3 REF Volt: 110
x CPU GTL 0/2 REF Volt: 110
x North Bridge GTL REF Volt: 110

DRAM Timing
- Enhance Data transmitting: FAST
- Enhance Addressing: FAST
- T2 Dispatch: Enabled

Clock Setting Fine Delay
Ch1 Clock Crossing Setting: More Aggressive
- DIMM 1 Clock fine delay: 7 (manually increased from 2 allowed me to pass 651Mhz mem clock wall in memtest86+ v1.70)
- DIMM 2 Clock fine delay: Current 7
- Ch 1 Command fine delay: Current 11
- Ch 1 Control fine delay: Current 8

Ch2 Clock Crossing Setting: More Aggressive
- DIMM 3 Clock fine delay: 7 (manually increased from 2 allowed me to pass 651Mhz mem clock wall in memtest86+ v1.70)
- DIMM 4 Clock fine delay: Current 6
- Ch 2 Command fine delay: Current 11
- Ch 2 Control fine delay: Current 6

Ch1Ch2 CommonClock Setting: More Aggressive

Ch1 RDCAS GNT-Chip Delay: Auto
Ch1 WRCAS GNT-Chip Delay: Auto
Ch1 Command to CS Delay: Auto

Ch2 RDCAS GNT-Chip Delay: Auto
Ch2 WRCAS GNT-Chip Delay: Auto
Ch2 Command to CS Delay: Auto (where cpuz sees 1T or 2T SETTING)

CAS Latency Time (tCL): 5
RAS# to CAS# Delay (tRCD): 5
RAS# Precharge (tRP): 5
Precharge Delay (tRAS): 9
All Precharge to Act: 4
REF to ACT Delay (tRFC): 30
Performance LVL (Read Delay) (tRD): 6

Read delay phase adjust: Enter

Ch1 Read delay phase (4~0)
- Channel 1 Phase 0 Pull-In: Auto
- Channel 1 Phase 1 Pull-In: Auto
- Channel 1 Phase 2 Pull-In: Auto
- Channel 1 Phase 3 Pull-In: Auto
- Channel 1 Phase 4 Pull-In: Auto

Ch2 Read delay phase (4~0)
- Channel 2 Phase 0 Pull-In: Auto
- Channel 2 Phase 1 Pull-In: Auto
- Channel 2 Phase 2 Pull-In: Auto
- Channel 2 Phase 3 Pull-In: Auto
- Channel 2 Phase 4 Pull-In: Auto

MCH ODT Latency: AUTO
Write to PRE Delay (tWR): 14
Rank Write to Read (tWTR): 11
ACT to ACT Delay (tRRD): 3
Read to Write Delay (tRDWR): 8
Ranks Write to Write (tWRWR): 4
Ranks Read to Read (tRDRD): 5
Ranks Write to Read (tWRRD): 4
Read CAS# Precharge (tRTP): 3
ALL PRE to Refresh: 4
12-19-2007, 03:24 PM
CERO

So how is this board compared to the P35 , or it is to early to make any conclusions?
12-19-2007, 04:25 PM
AKHandyman

I have 2 questions ... EVA ... how did you get the Everest beta version1227? I have the 1226 version, yet I can't find the one you have ...:confused:
Two, does anyone know why in the BIOS under Advanced Chipset Features > PCI-E compliancy mode ... it shows the version to be 1.0a ? I thought it was supposed to be 2.0 ... anyone notice this? :confused:
12-19-2007, 05:09 PM
AKHandyman

EVA - re your post #562 ... you say you did that with Crucial Ballistix PC2 8500, yet your CPU-Z shows OCZ PC2 6400 ... I have 4GB OCZ PC2 6400 and a E6850 ... if I use those settings, I should see something very similar to yours, right?
12-19-2007, 05:10 PM
Ace-a-Rue

Quote:

Originally Posted by Brother Esau

@ ACE......Hows you're ARECA Raid Controller working in this motherboard please:)

working great!...no problems while in slot #1...it won't work in slot #2; meaning, it hangs on the initial post screen...haven't upgraded to the newer bios...i might try later on.
12-19-2007, 05:18 PM
Brother Esau

So its working then/ whats the Burst rate that you are getting?
12-19-2007, 05:39 PM
Ace-a-Rue

hey BRO...does DFI boards have a serious problem with a lot of bios chips dying on flash???
12-19-2007, 05:40 PM
Ace-a-Rue

Quote:

Originally Posted by Brother Esau

So its working then/ whats the Burst rate that you are getting?

the last time i checked was like 660 mb/sec for 2 drives in raid 0
12-19-2007, 05:48 PM
Praz

How about this thread gets back on topic.
12-19-2007, 05:53 PM
Brother Esau

Quote:

Originally Posted by Ace-a-Rue

hey BRO...does DFI boards have a serious problem with a lot of bios chips dying on flash???

Not that I am aware of ACE but sheet happens you know? I just had a Flash go bad on me not long ago my Floppy drive threw a rod in the middle of the Flash and it was Good Night Irene:D

Sometimes too the Flash doesn't go right for what ever reason even if what you did was flawless thats happened to me a couple of times the Past year maybe bad chips who the hell knows:shrug:

But the Point is that its better to spend a couple of bucks on insurance so you don't get caught with you're pants down as RMA for Board takes 10-14 days vs a few minutes to plop a new chip in.........:D
12-19-2007, 07:35 PM
AKHandyman

Quote:

Originally Posted by Praz

How about this thread gets back on topic.

I agree, I've asked for a little clarification on a few things and haven't yet gotten an answer ... hope we can get back to what this thread is all about ... :p:
12-19-2007, 08:52 PM
pumbertot

just a Q and a quick answer wanted. would you guys that had the P35 LP recommend selling and getting this board or is it not worth it yet?
12-19-2007, 09:19 PM
emoners
Quote:
Originally Posted by eva2000

Clock Fine Delay At Work

Still very early into my testing of how memory behave on DFI LP LT X38-T2R with 11/28 bios. It basically seems like a more matured version of what is implemented on DFI LP UT P35-T2R :)

Playing with 2:3 divider (266/800) and 2x 1GB Crucial Ballistix PC2-8500, I found myself hitting memory clock wall at several points while doing my routine Memtest86+ v1.70 testing.

Only salvation to break through that memory clock wall was fine tuning Clock Fine Delay values for DIMM 1 & 3. To my surprise I broke through that 651Mhz wall and ended up at 666Mhz 5-5-5-9 at 2.49v being single Super Pi 32M stable!

633mhz 5-5-5-9 at 2.27v was okay
640mhz 5-5-5-9 at 2.31v was okay
651mhz 5-5-5-9 regardless of vdimm used kept freezing in test #5 loop testing - clock fine delay was all set to Current.
651mhz 5-5-5-9 at 2.38v errored out in memtest, but elimated freezing by increasing dimm 1 + 3 clock fine delay values from Current (2) to manually set 3.
660mhz 5-5-5-9 at 2.45v was okay in memtest now by further increasing dimm 1 + 3 clock fine delay values from manually set 3 to 5.
666mhz 5-5-5-9 at 2.49v would error out in test #5 loop on 2nd pass with a few errors everytime, no voltage adjustments helped.
666mhz 5-5-5-9 at 2.49v was okay when i further increased dimm 1 + 3 clock fine delay values from manually set 5 to 6 or 7.
670mhz 5-5-5-9 at 2.49v would error out in test #5 loop on 2nd pass with a few errors everytime, voltage adjustments didn't help much to stablise it. Dimm 1 + 3 clock fine delay values manually set at 7 helped reduce it to a few errors in memtest86+ v1.70 test #5 loop.

So here's where I end up at so far 666Mhz 5-5-5-9 at 2.49v! Pretty awesome for initially having hit a brick wall at 651Mhz 5-5-5-9!

http://fileshosts.com/intel/DFI/DFI_...cpuz_valid.png

http://fileshosts.com/intel/DFI/DFI_...3m20s110ms.png

http://fileshosts.com/intel/DFI/DFI_..._bandwidth.png
nice one Eva!:up: is it ok to push NB to 1.6v?
12-19-2007, 09:34 PM
AKHandyman

I am liking what I see so far, but I am far from being an expert, but right now I have it purring along pretty good ... it's been stable for most of going 4 hours at (9 x 423 MHz)= 3.8 GHz. Memory is stable as well ... I know this board has a lot more in it and I am learning as I go along ... and having some fun ... :up:

http://img529.imageshack.us/img529/7...at38sd2.th.jpg
12-19-2007, 09:36 PM
eva2000

Quote:

Originally Posted by AKHandyman

I have 2 questions ... EVA ... how did you get the Everest beta version1227? I have the 1226 version, yet I can't find the one you have ...:confused:
Two, does anyone know why in the BIOS under Advanced Chipset Features > PCI-E compliancy mode ... it shows the version to be 1.0a ? I thought it was supposed to be 2.0 ... anyone notice this? :confused:

From everest support forums :) No idea on PCI-E version mode

Quote:

Originally Posted by AKHandyman

EVA - re your post #562 ... you say you did that with Crucial Ballistix PC2 8500, yet your CPU-Z shows OCZ PC2 6400 ... I have 4GB OCZ PC2 6400 and a E6850 ... if I use those settings, I should see something very similar to yours, right?

re-read screenie no OCZ listed in 666mhz 5-5-5-9 screenie. As to settings they might differ as no set of memory modules or board NB memory controller are exactly the same. Really folks shouldn't focus on the resulting end settings but the process itself that is involved in getting to them :)

Quote:

Originally Posted by emoners

nice one Eva!:up: is it ok to push NB to 1.6v?

for benching only.. for everyday it's easy to drop NB volts needed by loosening your memory sub timings and additional tight tweaks i.e. clock crossing and clockcommon settings drop down from More Aggressive
12-19-2007, 09:59 PM
AKHandyman

Quote:

Originally Posted by eva2000

.. re-read screenie no OCZ listed in 666mhz 5-5-5-9 screenie. As to settings they might differ as no set of memory modules or board NB memory controller are exactly the same...

The CPU-Z validation page you used shows at the bottom left hand side in slot 1 and slot 2 OCZ Technology ... that is what I was referencing ... but what you're saying is that the 666MHz 5-5-5-9 is from the Ballistix RAM, right? I was just trying to clarify what RAM did what ... that's all :shrug:
12-19-2007, 10:01 PM
Brother Esau

Anybody running the New Cards in Cross Fire yet?
12-19-2007, 10:04 PM
eva2000

Quote:

Originally Posted by AKHandyman

The CPU-Z validation page you used shows at the bottom left hand side in slot 1 and slot 2 OCZ Technology ... that is what I was referencing ... but what you're saying is that the 666MHz 5-5-5-9 is from the Ballistix RAM, right? I was just trying to clarify what RAM did what ... that's all :shrug:

huh? i see both slots saying crucial
12-19-2007, 10:08 PM
Brother Esau

ooooh you have some cards don't you Eva:)
12-19-2007, 10:26 PM
eva2000

Quote:

Originally Posted by Brother Esau

ooooh you have some cards don't you Eva:)

Yeah will test later but they're in my Blitz Extreme rig for now while i figure out DFI X38's memory oc'ing behaviour first :)
12-19-2007, 10:28 PM
Brother Esau

Thanks George;) I can appreciate that:up:
12-20-2007, 01:24 AM
CapFTP

Quote:

Originally Posted by Brother Esau

Anybody running the New Cards in Cross Fire yet?

2x3850 crossfire, yep:)
12-20-2007, 03:29 AM
eva2000
Again clock fine delay and controller fine delay tweaking allowed me to stabilize my 2x 1GB Crucial Ballistix PC2-8500 @585Mhz 4-4-4-5 PL 5 at 2.51v.
- 540mhz 4-4-4-5 at 2.23v = okay
- 550mhz 4-4-4-5 at 2.31v = okay
- 565mhz 4-4-4-5 at 2.38v = okay
- 575mhz 4-4-4-5 at 2.45v = okay
- 580mhz 4-4-4-5 at 2.47v = okay
- 585mhz 4-4-4-5 at 2.49v = 4 errors in 1st pass of test #5 memtest86+ v1.70 loop
- 585mhz 4-4-4-5 at 2.51v = okay (manually set clock fine delay from current 3 to 4 and control fine delay for DIMM 1 + 3 changed from current 10 to manually set 11)
http://fileshosts.com/intel/DFI/DFI_...cpuz_valid.png

Single 32M
http://fileshosts.com/intel/DFI/DFI_...3m29s875ms.png

Dual 32M
http://fileshosts.com/intel/DFI/DFI_...2m_halfway.png

http://fileshosts.com/intel/DFI/DFI_...dual32m_tn.png

http://fileshosts.com/intel/DFI/DFI_..._bandwidth.png

DFI LP LT X38-T2R Bios Settings

Quote:

PC Health Status
Adjust CPU Temp: +7C

CPU Feature
- Thermal Management Control: Disabled
- PPM(EIST) Mode: Disabled
- Limit CPUID MaxVal: Disabled
- CIE Function: Disabled
- Execute Disable Bit: Disabled
- Virtualization Technology: Disabled
- Core Multi-Processing: Enabled

Exist Setup Shutdown: Mode 2
Shutdown after AC Loss: Disabled
CLOCK VC0 divider: AUTO
CPU Clock Ratio Unlock: Enabled
CPU Clock Ratio: 9x
- Target CPU Clock: 3512
CPU Clock: 390
Boot Up Clock: AUTO
DRAM Speed: 266/800
- Target DRAM Speed: 1172
PCIE Clock: 100mhz
PCIE Slot Config: 1X 1X

CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled
SATA Spread Spectrum: Disabled

Voltage Settings
CPU VID Control: 1.2875
CPU VID Special Add: AUTO
DRAM Voltage Control: 2.51
SB Core/CPU PLL Voltage: 1.51
NB Core Voltage: 1.555
CPU VTT Voltage: 1.377
Vcore Droop Control: Enabled
Clockgen Voltage Control: 3.45v
GTL+ Buffers Strength: Strong
Host Slew Rate: Weak
GTL REF Voltage Control: Disable
x CPU GTL1/3 REF Volt: 110
x CPU GTL 0/2 REF Volt: 110
x North Bridge GTL REF Volt: 110

DRAM Timing
- Enhance Data transmitting: FAST
- Enhance Addressing: FAST
- T2 Dispatch: Enabled

Clock Setting Fine Delay
Ch1 Clock Crossing Setting: More Aggressive
- DIMM 1 Clock fine delay: 4 (manually increased from 3)
- DIMM 2 Clock fine delay: 6 (manually set from current 6)
- Ch 1 Command fine delay: 11 (manually increased from current 10)
- Ch 1 Control fine delay: 7 (manually set from current 7)

Ch2 Clock Crossing Setting: More Aggressive
- DIMM 3 Clock fine delay: 4 (manually increased from 3)
- DIMM 4 Clock fine delay: 6 (manually set from current 6)
- Ch 2 Command fine delay: 11 (manually increased from current 10)
- Ch 2 Control fine delay: 5 (manually set from current 5)

Ch1Ch2 CommonClock Setting: More Aggressive

Ch1 RDCAS GNT-Chip Delay: Auto
Ch1 WRCAS GNT-Chip Delay: Auto
Ch1 Command to CS Delay: Auto

Ch2 RDCAS GNT-Chip Delay: Auto
Ch2 WRCAS GNT-Chip Delay: Auto
Ch2 Command to CS Delay: Auto (where cpuz sees 1T or 2T SETTING)

CAS Latency Time (tCL): 4
RAS# to CAS# Delay (tRCD): 4
RAS# Precharge (tRP): 4
Precharge Delay (tRAS): 5
All Precharge to Act: 4
REF to ACT Delay (tRFC): 30
Performance LVL (Read Delay) (tRD): 5

Read delay phase adjust: Enter

Ch1 Read delay phase (4~0)
- Channel 1 Phase 0 Pull-In: Auto
- Channel 1 Phase 1 Pull-In: Auto
- Channel 1 Phase 2 Pull-In: Auto
- Channel 1 Phase 3 Pull-In: Auto
- Channel 1 Phase 4 Pull-In: Auto

Ch2 Read delay phase (4~0)
- Channel 2 Phase 0 Pull-In: Auto
- Channel 2 Phase 1 Pull-In: Auto
- Channel 2 Phase 2 Pull-In: Auto
- Channel 2 Phase 3 Pull-In: Auto
- Channel 2 Phase 4 Pull-In: Auto

MCH ODT Latency: AUTO
Write to PRE Delay (tWR): 14
Rank Write to Read (tWTR): 11
ACT to ACT Delay (tRRD): 3
Read to Write Delay (tRDWR): 8
Ranks Write to Write (tWRWR): 4
Ranks Read to Read (tRDRD): 5
Ranks Write to Read (tWRRD): 4
Read CAS# Precharge (tRTP): 3
ALL PRE to Refresh: 4
12-20-2007, 04:11 AM
M3kk

Someone with this board + Enermax Liberty ? I`m curious if this incompatible was resolved .. Thanks.
12-20-2007, 04:29 AM
eva2000

Quote:

Originally Posted by M3kk

Someone with this board + Enermax Liberty ? I`m curious if this incompatible was resolved .. Thanks.

don't think it has been resolved.. get a new psu ;)
12-20-2007, 04:31 AM
Praz

Quote:

Originally Posted by M3kk

Someone with this board + Enermax Liberty ? I`m curious if this incompatible was resolved .. Thanks.

Don't know about the Liberty but the Galaxy appears to have issues.

http://csd.dficlub.org/forum/showthread.php?t=5005
12-20-2007, 05:18 AM
M3kk

Well i have 4 psu s now, but i want to put the liberty in the case with this board or with a p35 one, but the p35 have issues with liberty, and i dont know yet about this :(.

Op1000 and zeus1200 wont fit in my case :D.

Praz, yes i know about p35 and liberty, but maybe was resolved with this board..
I`m curious if that adapter will resolve this problem, would be nice :).
12-20-2007, 05:24 AM
eva2000

OP850 or OCZ 1KW ProXStream both will fit in a case :)
12-20-2007, 05:54 AM
Praz

Quote:

Originally Posted by M3kk

Praz, yes i know about p35 and liberty, but maybe was resolved with this board:)

Wouldn't bet on it. The power planes are almost identical. There is mod that will allow most of the power supplies to work with different degrees of success although it is not user doable. I will check if the X38 has been altered in this fashion.
12-20-2007, 07:06 AM
lazy

so i have been unsucessful priming my cpu for more then 10mins
Do the Q6600 G0's need that much voltages or am i doing something wrong?

FSB = 400
Multi = 9
Vcore = 1.48
VVT = 1.6
NB = 1.5
SB = 1.74

GTL set according to the link provided to be a few pages back
12-20-2007, 07:12 AM
Praz

Quote:

Originally Posted by lazy

so i have been unsucessful priming my cpu for more then 10mins
Do the Q6600 G0's need that much voltages or am i doing something wrong?

FSB = 400
Multi = 9
Vcore = 1.48
VVT = 1.6
NB = 1.5
SB = 1.74

GTL set according to the link provided to be a few pages back

That GTL Reference link is just that, a reference. Board to board variations may result in the settings being off 5 or so points in either direction. Also the tables show the values needed for a 67% setting. Most quad-cores, when being overclocked, prefer a setting closer to 70%.
12-20-2007, 08:53 AM
M3kk

Quote:

Originally Posted by eva2000

OP850 or OCZ 1KW ProXStream both will fit in a case :)

In my case not :D. The Liberty 500 fits with mm, so if i want to put there a psu bigger than the liberty with 1 cm, then wont fit :).
12-20-2007, 08:54 AM
M3kk

Response from enermax :

Quote:

Im afraid there is no solution. The adapter is for the Galaxy series only.

There is no fix that we can apply to this PSU that will fix this issue as it is related to DFI’s motherboard.

The only choice you have is to change PSU or Motherboard L

:(
12-20-2007, 08:58 AM
eva2000
Ah that suxs then.. new case for Xmas ? :D

Clock Fine Delay At Work Continued....

Swapped out one set of 2x 1GB Crucial Ballistix PC2-8500 for another set of 2x 1GB Crucial Ballistix PC2-8500 since I'm re-testing all my memory on DFI LP LT X38-T2R to find the best matched pair for this board :)

Seems similar process to gain stability occurs with this set too - meaning fine tuning clock fine delay DIMM 1 + 3 as well as controller clock fine delay allows this memory to fly as high if not slightly better than the first set I tested.
- 633mhz 5-5-5-9 at 2.27v was okay (with lower NB 1.566v volts)
- 640mhz 5-5-5-9 at 2.27v was okay (with lower NB 1.566v volts)
- 651mhz 5-5-5-9 at 2.31v Froze at 88% 1st pass of test #5 loop (with lower NB 1.566v volts)
- 651mhz 5-5-5-9 at 2.31v was okay now by increasing dimm 1 + 3 clock fine delay values from Current (2) to manually set 3. (with lower NB 1.566v volts)
- 660mhz 5-5-5-9 at 2.42v was okay - dimm 1 + 3 clock fine delay values still manually set to 3. (with lower NB 1.566v volts)
- 666mhz 5-5-5-9 at 2.47v would error out in test #5 loop - went to check bios clock fine delays that are set automatically (Current values) and they all automatically shifted down from the values that were stable at 660mhz 5-5-5-9 at 2.42v). So applied same theory I discovered with DFI LP UT P35-T2R, I manually adjusted the remaining Current auto values @666Mhz to the ones matching @660Mhz 5-5-5-9 stable clocks except the Dimm 1+ 3 clock fine delays which manually set to 6.
- 666mhz 5-5-5-9 at 2.47v with clock fine delay values of 6-6-10-7 / 6-6-10-5 for each channel respectively allowed me to loop Memtest86+ v1.70 test #5 for 10 passes before 2 errors occured. Still not quite there but better than before. NB volts still at 1.566v
- 666mhz 5-5-5-9 at 2.49v was okay with NB still 1.566v
- 669mhz 5-5-5-9 at 2.47v with NB raised from 1.566v to 1.617v and manually set clock fine delay values to 6-6-10-7 / 6-6-10-5. Ended up with 3 errors after the 4th pass of memtest86+ v1.70 test #5 looping.
- 669mhz 5-5-5-9 at 2.49v with NB = 1.617v. Decided to try manually setting clock fine delay values to match 1st set of memory's 666Mhz 5-5-5-9 32M stable settings which are 7-7-11-8 / 7-6-11-6. Resulted in 4 errors after 2nd pass of test #5 looping.
- 669mhz 5-5-5-9 at 2.51v with NB = 1.630v and CPU VTT = 1.387V raised from 1.377v. Clock fine delay set at 6-6-10-7 / 6-6-10-5.
http://fileshosts.com/intel/DFI/DFI_...cpuz_valid.png

http://fileshosts.com/intel/DFI/DFI_...3m16s157ms.png

http://fileshosts.com/intel/DFI/DFI_..._bandwidth.png

DFI LP LT X38-T2R Bios Settings

Quote:

PC Health Status
Adjust CPU Temp: +7C

CPU Feature
- Thermal Management Control: Disabled
- PPM(EIST) Mode: Disabled
- Limit CPUID MaxVal: Disabled
- CIE Function: Disabled
- Execute Disable Bit: Disabled
- Virtualization Technology: Disabled
- Core Multi-Processing: Enabled

Exist Setup Shutdown: Mode 2
Shutdown after AC Loss: Disabled
CLOCK VC0 divider: AUTO
CPU Clock Ratio Unlock: Enabled
CPU Clock Ratio: 8x
- Target CPU Clock: 3571
CPU Clock: 446
Boot Up Clock: AUTO
DRAM Speed: 266/800
- Target DRAM Speed: 1341
PCIE Clock: 100mhz
PCIE Slot Config: 1X 1X

CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled
SATA Spread Spectrum: Disabled

Voltage Settings
CPU VID Control: 1.2875
CPU VID Special Add: AUTO
DRAM Voltage Control: 2.51
SB Core/CPU PLL Voltage: 1.51
NB Core Voltage: 1.630
CPU VTT Voltage: 1.387
Vcore Droop Control: Enabled
Clockgen Voltage Control: 3.45v
GTL+ Buffers Strength: Strong
Host Slew Rate: Weak
GTL REF Voltage Control: Disable
x CPU GTL1/3 REF Volt: 110
x CPU GTL 0/2 REF Volt: 110
x North Bridge GTL REF Volt: 110

DRAM Timing
- Enhance Data transmitting: FAST
- Enhance Addressing: FAST
- T2 Dispatch: Enabled

Clock Setting Fine Delay
Ch1 Clock Crossing Setting: More Aggressive
- DIMM 1 Clock fine delay: 6
- DIMM 2 Clock fine delay: 6
- Ch 1 Command fine delay: 10
- Ch 1 Control fine delay: 7

Ch2 Clock Crossing Setting: More Aggressive
- DIMM 3 Clock fine delay: 6
- DIMM 4 Clock fine delay: 6
- Ch 2 Command fine delay: 10
- Ch 2 Control fine delay: 5

Ch1Ch2 CommonClock Setting: More Aggressive

Ch1 RDCAS GNT-Chip Delay: Auto
Ch1 WRCAS GNT-Chip Delay: Auto
Ch1 Command to CS Delay: Auto

Ch2 RDCAS GNT-Chip Delay: Auto
Ch2 WRCAS GNT-Chip Delay: Auto
Ch2 Command to CS Delay: Auto (where cpuz sees 1T or 2T SETTING)

CAS Latency Time (tCL): 5
RAS# to CAS# Delay (tRCD): 5
RAS# Precharge (tRP): 5
Precharge Delay (tRAS): 9
All Precharge to Act: 4
REF to ACT Delay (tRFC): 30
Performance LVL (Read Delay) (tRD): 6

Read delay phase adjust: Enter

Ch1 Read delay phase (4~0)
- Channel 1 Phase 0 Pull-In: Auto
- Channel 1 Phase 1 Pull-In: Auto
- Channel 1 Phase 2 Pull-In: Auto
- Channel 1 Phase 3 Pull-In: Auto
- Channel 1 Phase 4 Pull-In: Auto

Ch2 Read delay phase (4~0)
- Channel 2 Phase 0 Pull-In: Auto
- Channel 2 Phase 1 Pull-In: Auto
- Channel 2 Phase 2 Pull-In: Auto
- Channel 2 Phase 3 Pull-In: Auto
- Channel 2 Phase 4 Pull-In: Auto

MCH ODT Latency: AUTO
Write to PRE Delay (tWR): 14
Rank Write to Read (tWTR): 11
ACT to ACT Delay (tRRD): 3
Read to Write Delay (tRDWR): 8
Ranks Write to Write (tWRWR): 4
Ranks Read to Read (tRDRD): 5
Ranks Write to Read (tWRRD): 4
Read CAS# Precharge (tRTP): 3
ALL PRE to Refresh: 4
12-20-2007, 09:13 AM
lazy

Quote:

Originally Posted by Praz

That GTL Reference link is just that, a reference. Board to board variations may result in the settings being off 5 or so points in either direction. Also the tables show the values needed for a 67% setting. Most quad-cores, when being overclocked, prefer a setting closer to 70%.

then what would you recommend?
12-20-2007, 09:14 AM
M3kk

No man :) I dont use case, i just want to put this board or the p35 one in the case for 24/7 .
I have a modified Cheftec case, with dual rad at the top, because of that wont fit other psu :). http://mekk.homeip.net/cpgallery/thu...s.php?album=12

Eva, can you try max 4-4-4-12 ?
Do you have teamgorup or cellshock kits? Or praz, you know if the "issues" with this rams are resolved?
12-20-2007, 09:27 AM
Praz

Quote:

Originally Posted by lazy

then what would you recommend?

At this point I don't have a recommendation. Haven't had time to explore GTL values yet. I don't known at this point if the settings can even be carried over from the P35. Will probably get into it this weekend if I switch over to water. With air cooling I can't get to the MOSFETs to measure the voltages
12-20-2007, 09:57 AM
eva2000

Quote:

Originally Posted by M3kk

No man :) I dont use case, i just want to put this board or the p35 one in the case for 24/7 .
I have a modified Cheftec case, with dual rad at the top, because of that wont fit other psu :). http://mekk.homeip.net/cpgallery/thu...s.php?album=12

Eva, can you try max 4-4-4-12 ?
Do you have teamgorup or cellshock kits? Or praz, you know if the "issues" with this rams are resolved?

The set of crucial pc2-8500 i tried before this did 585mhz 4-4-4-5 at 2.51v http://www.xtremesystems.org/forums/...&postcount=587

Have yet to test this set of crucial pc2-8500 for 4-4-4-5 yet.

Yeah i have old school Team 6400C3 Micron D9GMH does 570-575mhz at 2.4-2.45v on P35 chipset. No cellshock DDR2 though, Cellshock DDR3 kit coming soon though for DDR3 boards :)
12-20-2007, 10:20 AM
Praz

CellShock 8000 on this board acts the same as on the P35. After messing with the SPD it seems like a lost cause. I'm almost willing to bet money it's a PCB issue.

Show 100 post(s) from this thread on one page

All times are GMT -8. The time now is 04:35 PM.

XtremeSystems