[First Look] DFI 680i LT NF-680i LT SLI (56K Warning~~~)

[virtualrain]

My board & Smartguardian reports temps just fine using the 521 BIOS.

Regards

Craig.

Same here. Use CoreTemp or Speedfan (which reports CoreTemp but you need to add 15-deg offset) for CPU temps.

Craig, I will try high-clock 2T tonight in the yellow slots and see what it can do. Last time I played with high-clock 2T, my Read bandwidth at 1120MHz 5-5-5 2T was lower than 4-3-3 1T at 800MHz.

[Jobeo]

Thats about all I can figure out, 3.04v right after clearing CMOS via jumper.

hrmm.... wow. thats not good. which BIOS were you using?

kinda scary to think about cause once ya boot with the overvolt like that its all over... nothing you can do to fix it real quick.

i haven't cleared the CMOS by jumper in a long time... i always use the reset/power button trick. hrmmmmmmmmmmm.

[xTr33m sammich!]

Happened while using Beta 4CTST521. I had been using it for several days, and had originally cleared CMOS via jumper after updating to it.

I'm a little put off by it, and lucky enough I guess to have Newegg on my side, they agreed to full refund for board and ram no restocking fee.

Now I just have to decide what I'm gonna do. Anyone know if theres an sli hack for the p35? otherwise I may have to sell a pair of 8600GT's.

[Jobeo]

I'm a little put off by it, and lucky enough I guess to have Newegg on my side, they agreed to full refund for board and ram no restocking fee.

hrmm... if you give up on the board the DFI 680i userbase will drop by about 25% ... lol j/k

i feel your pain. i would be very upset if this board just up and decided to overvolt my ram that way. so far it hasn't happened to me.

i hope they continue to put time into developing this BIOS ... although i fear they won't. i expect about 1 or 2 more updates before they move onto other things.

[virtualrain]

hrmm... if you give up on the board the DFI 680i userbase will drop by about 25% ... lol j/k

i feel your pain. i would be very upset if this board just up and decided to overvolt my ram that way. so far it hasn't happened to me.

i hope they continue to put time into developing this BIOS ... although i fear they won't. i expect about 1 or 2 more updates before they move onto other things.

With pics of the X38 board in circulation, I'm afraid we may get left in the dust also. Hopefully they at least get bugs like this fixed before moving on.

BTW, I also use the power/reset button to clear CMOS and it has worked flawlessly for me (many many times!).

[Jobeo]

With pics of the X38 board in circulation, I'm afraid we may get left in the dust also.

ya but as you pointed out a few posts ago who knows when DFI will release that. i mean we saw pics of the 680i board at the start of 2007 and look when it came out.

if they do get it out a few months from now i'd consider moving over to it... if it still supports my qx6700 that is. i like the idea of better stock cooling for the PWM and NB, things they should have done for the 680i to begin with. for now i'm satisfied which is good cause i doubt DFI's x38 solution will be released this year.

[virtualrain]

ya but as you pointed out a few posts ago who knows when DFI will release that. i mean we saw pics of the 680i board at the start of 2007 and look when it came out.

if they do get it out a few months from now i'd consider moving over to it... if it still supports my qx6700 that is. i like the idea of better stock cooling for the PWM and NB, things they should have done for the 680i to begin with. for now i'm satisfied which is good cause i doubt DFI's x38 solution will be released this year.

Jobeo, what is it going to take to get you to switch to water cooling!

Do it man!

[owcraftsman]

ya but as you pointed out a few posts ago who knows when DFI will release that. i mean we saw pics of the 680i board at the start of 2007 and look when it came out.

if they do get it out a few months from now i'd consider moving over to it... if it still supports my qx6700 that is. i like the idea of better stock cooling for the PWM and NB, things they should have done for the 680i to begin with. for now i'm satisfied which is good cause i doubt DFI's x38 solution will be released this year.

[Grinch]

http://www.xtremesystems.org/forums/...d.php?t=147017

[Jobeo]

Jobeo, what is it going to take to get you to switch to water cooling!

ya i prob should ... maybe if i knew what to buy. because its so new to me i can just imagine things going very wrong like my case being flooded with coolant or something.

i'm sure once i got it going i'd be very happy... i guess i'm stuck in the past i've cooled on air for so long ... =/

[virtualrain]

ya i prob should ... maybe if i knew what to buy. because its so new to me i can just imagine things going very wrong like my case being flooded with coolant or something.

i'm sure once i got it going i'd be very happy... i guess i'm stuck in the past i've cooled on air for so long ... =/

Ahhh... no worries. If you get inclined PM me or drop into the water cooling forum. Just be prepared to layout more $$$

[virtualrain]

It's been reported that you should use the yellow slots for increased memory bandwidth and the orange slots for higher clocks with 1T (Link - see post 30/31). So I tried running my Mushkin PC2-8500 2x2GB kit in both the orange and yellow slots to see if there was any truth to this... well, here are my results...



Comments:

• Strangely, I could not get the board to POST with the RAM running at 1T in the yellow slots.
• At 1100MHz running 2T command rate, there was no difference between the yellow and orange slots with my memory.
• Oddly, when switching RAM between slots, the board requires a CMOS clear to POST. Then after adjusting settings (loading from cmos reloaded profiles for example) and rebooting, it would hang on POST code #75 for 2 minutes before rebooting normally. I'm not sure why it does this, any thoughts? At any rate, be patient if it does this.
• As I reported earlier, for me, 1T is marginally faster at 800MHz than 2T at 1100MHz (at least according to Everest... Sisoft Sandra has 1T slightly slower).

Based on my results, I feel bad for anyone trying to get good clocks on four sticks! Yikes!

[virtualrain]

BTW, I've updated my Tips and Tricks Blog Article to represent the most current wisdom with this board.

It's also available via my sig.

[Borgod]

Based on my results, I feel bad for anyone trying to get good clocks on four sticks! Yikes!

Yep, it has me very worried for future upgrades.

[C-N]

It's been reported that you should use the yellow slots for increased memory bandwidth and the orange slots for higher clocks with 1t (Link - see post 30/31). So I tried running my Mushkin PC2-8500 2x2GB kit in both the orange and yellow slots to see if there was any truth to this

I read that over at TTR but I cant find any difference between the slots at all & never have with any of the BIOS's I have used.

I put my OCZ 8500 SLI RAM & 8800gts in last night & ran 456 x 8 4.4.3.8 1T sync pci-e 120mhz spp 300mhz graphics card 595 x 950 the memory bandwidth is 11950mb/s & the latency is mid 43ns & it run Orthos 8 hours + again.

My NB voltage was 1.57 & the temp never hit 50oC

I am a happy man this board is super fast & rock solid stable I love it.

Craig.

[AlphaHeX]

Bios 2007/05/21 has been moved to stable section on DFI webpage

Major Reasons of Change:
1.Fixed Core Multi-Processing function fail issue on Intel Core 2 Extreme QX6700 /X6800 CPUs.
2.For Intel Core 2 Quad ,Intel Core 2 Extreme & X6800 CPUs only.
3.Fixed CPU Bselect can’t set to Auto mode with E6750(FSB 1333MHz) CPU.
4.DRAM Voltage will be selected on 3.04V when doing the load last bootable setting.
5.Fixed CPU Fan speed control.

As you can see point 4 is reffering to the DRAM voltage problems.

I'm wondering why it has happend to xTr33m sammich! if he was using that version of BIOS. Maybe there is a difference between beta 05/21 and final 05/21?

[vincey2kr1]

Well guys picked up a Asus Pk5Deluxe and no more shutdowns during 8k quadprimes now at 3.6ghz 8*450 (q6600) after only an hour of playing and I'm not done tweaking.. No sli dammit though but at least they implemented 1t timings now .. REally think with a nice aftermarket PWM cooler the DFI should of done the same.

[virtualrain]

Bios 2007/05/21 has been moved to stable section on DFI webpage

Major Reasons of Change:
1.Fixed Core Multi-Processing function fail issue on Intel Core 2 Extreme QX6700 /X6800 CPUs.
2.For Intel Core 2 Quad ,Intel Core 2 Extreme & X6800 CPUs only.
3.Fixed CPU Bselect can’t set to Auto mode with E6750(FSB 1333MHz) CPU.
4.DRAM Voltage will be selected on 3.04V when doing the load last bootable setting.
5.Fixed CPU Fan speed control.

As you can see point 4 is reffering to the DRAM voltage problems.

I'm wondering why it has happend to xTr33m sammich! if he was using that version of BIOS. Maybe there is a difference between beta 05/21 and final 05/21?

An interesting change is that the original 2CTST file was noted as being for Q6600. The Beta 4CTST was noted as being for Quad Core and Extremes... very ambiguous if you owned a Q6600. Since the 2C version clearly stated it was for Q6600, I used that one (and it didn't affect my OC at all).

Now with it being published to the regular BIOS page, the 4CTST is noted as being for quad core and extreme processors while there is no mention of what processors the 2C version is for. Although I suppose it's obviously for dual-cores.

I guess I will have to try the 4C version now. Maybe it will improve my FSB wall issue. We shall see.

[virtualrain]

Well guys picked up a Asus Pk5Deluxe and no more shutdowns during 8k quadprimes now at 3.6ghz 8*450 (q6600) after only an hour of playing and I'm not done tweaking.. No sli dammit though but at least they implemented 1t timings now .. REally think with a nice aftermarket PWM cooler the DFI should of done the same.

Arg! Another blow to this board's community... leaving a lone few to soldier on!

[virtualrain]

More chaos with this board.

Earlier this week, I tried lowering my NB and RAM voltage to see if I could maintain my OC (400x9 Sync'd). No go. I couldn't lower either my NB or my RAM voltage even one notch and stay PRIME stable for more than an hour.

Last night I tried moving my memory from orange to yellow and then back to orange slots. I don't know what possessed me to try this, but before I went to bed, I lowered the NB voltage a few notches and the RAM voltage a notch and ran PRIME. Well much to my surprise, it was still running happily after 6+hrs this morning.

I just don't understand this board sometimes.

[Grinch]

Genie BIOS Settings:

Voltage Control:

CPU VID Control................ - 1.28125V
CPU VID Special Add............ - Auto
CPU Drive Strength............. - Weak
DRAM Drive Strength............ - Weak
DRAM Voltage Control........... - 2.01V
NB Voltage Control............. - 1.44V
SB Voltage Control............. - 1.56V
SB Dual Voltage................ - 1.52V
LDT Voltage Control............ - 1.22V
CPU VTT 1.2V Voltage........... - 1.21V
GTL REF Voltage Control........ - Disable


FSB & Frequency Seting:

FSB-Memory Clock Mode.......... - Linked
FSB-Memory Ratio............... - Sync Mode
FSB (QDR), MHz................. - 1600

=====

PCI Bus, Slot 1 MHz............ - 100
PCI Bus, Slot 2 MHz............ - 100
SPP <-> MCP Ref Clock, MHz..... - 200
LDT Frequency.................. - 5x
CPU BSelect.................... - Auto
CPU Clock Ratio Unlock......... - Enabled
CPU Clock Ratio................ - 8x


CPU Feature:

PPM Mode....................... - Native Mode
Limit CPUID MaxVal............. - Disabled
C1E Function................... - Disabled
Execute Disable Bit............ - Enabled
Virtualization Technology...... - Enabled
Core Multi-Processing.......... - Enabled

Memory Timing Setting:

Memory Timing Setting.......... - Expert
CAS Latency Control (TCL)...... - 4
RAS# to CAS# Delay (TRCD)...... - 4
Row Precharge Time (TRP)....... - 4
Min RAS# Active Time (TRAS).... - 12
Command Per Clock (CMD)........ - 2T

=====

Bank To Bank CMD (TRRD)........ - 3
Row Cycle Time (TRC)........... - 22
Write Recovery (TWR)........... - 5
Write -Read Commad............. - 9
DRAM Refresh Rate.............. - 7.8uS


not that this is a BIG overclock...just wanted to share where I am 100&#37; stable right now...here are my temps:

Cpu-Z Shot:

[owcraftsman]

Genie BIOS Settings:

Hey Grinch,
Thanks for sharing all the hard work. This is exaclty what I've been looking for here. I wish someone would start a data base thread.

This will be (almost) my set up when it gets here. I already received the mobo & e6600 but the ram (xms8500C5DF) % the GPU (EVGA 640-P2-N821-AR) got stollen from my front porch. Believe it or not UPS thinks there driver stole the FedX box who came late that day. They have placed a dummy box on my front porch with a micro cam in it to maybe catch him snooping around my porch. It has 40hour tape in it.
It's a real bummer cause Fedx won't pay up until they investigate which usually take 6 weeks before they payoff and zipzoomfly won't do anything until they do. Screwed up rebates and checking the hardware in case I need to use the 30day return/exchange/defectives with the suppliers. I usually order a complete system all at once for just that reason. I timed it so well it all came the same day. I don't think I'll do that again.

O well maybe it's a blessing I will seek a refund and rethink my choices for ram & gpu.

[Borgod]

Oh man that sucks

I get all my stuff delivered to my workplace now so I'm guaranteed to get it on delivery. Good luck with the investigation.

Also, I agree with the database thread

[Grinch]

Hey Grinch,
Thanks for sharing all the hard work. This is exaclty what I've been looking for here. I wish someone would start a data base thread.

This will be (almost) my set up when it gets here. I already received the mobo & e6600 but the ram (xms8500C5DF) % the GPU (EVGA 640-P2-N821-AR) got stollen from my front porch. Believe it or not UPS thinks there driver stole the FedX box who came late that day. They have placed a dummy box on my front porch with a micro cam in it to maybe catch him snooping around my porch. It has 40hour tape in it.
It's a real bummer cause Fedx won't pay up until they investigate which usually take 6 weeks before they payoff and zipzoomfly won't do anything until they do. Screwed up rebates and checking the hardware in case I need to use the 30day return/exchange/defectives with the suppliers. I usually order a complete system all at once for just that reason. I timed it so well it all came the same day. I don't think I'll do that again.

O well maybe it's a blessing I will seek a refund and rethink my choices for ram & gpu.

Sorry to hear about your stuff...are you going to watercool? You know how fun it would be to catch somebody in the act of stealing your stuff? I can say this..by the time the police arrived there would be only 1 story...MINE!

Oh man that sucks

I get all my stuff delivered to my workplace now so I'm guaranteed to get it on delivery. Good luck with the investigation.

Also, I agree with the database thread

We should have a database thread....

[Grinch]

wonder if nvidia monitor will ever be updated to read our dfi boards properly?