Official GIGA P35-S3/DS3/DS3R/DS3P/DS4/DQ6 bios/info thread

[rge]

I don't know if it works with DQ6 but I know for a fact that the F3 BIOS for DS4 allows you to manually adjust the performance level. In that vesrion the tRD value corresponds to Memsets performance level. So I actually had to downgrade the stock BIOS to get the performance I wanted. Hope this helps.

Very helpful, thanks! Just tried it, tRD adjusts performance in DQ6 as well, somehow missed that.

Does anyone know besides secondary timings what else turbo or extreme adjust versus standard? I have all timings on manual except phase, and with bios F6 set on turbo or extreme instead of standard, I still get improvements in everest measured read (from 9350 to 9500) copy (from 7627 to 8425) and latency (from 54.4 to 52.9), and spi (from 13.04 to 12.95), despite all memset visible settings being same and all everest memory timings listed under chipset same.

[Capt Proton]

That was the problem! I had also been having an issue with my music & videos (only the audio) not playing smoothly when doing ANY activity (even scrolling in a web page). After I found out about the x16 -> x1 issue, I investigated, and sure enough, I was running at x1.

How do you determine how many lanes the Vid Card is getting?

[SySTaXer]

How do you determine how many lanes the Vid Card is getting?

CPU-Z "Mainboard" tab under "Graphic Interface." Or nVIDIA control panel-->Help-->System Information.

[miahallen]

CPU-Z "Mainboard" tab under "Graphic Interface." Or nVIDIA control panel-->Help-->System Information.

Correct, I used the CPU-Z method.

BTW, I stated that I boosted the MCH voltage as part of my fix. But I was just reading the posts on the last page and saw people having issues with too much MCH voltage as well. So for the record, I wasnted to be more specific. I'm currently running my Q6600 G0 at 400x8 with 4x1GB of Crucial Ballistic PC2-6400 with a 1:1 ratio and 4-4-4-12 timings.

When I was having the PCIe x1 problem, I was running the following settings:
PCIe clock - auto

voltages:
PCIe - normal
MCH - normal
RAM - +0.2V
CPU - 1.325V

To solve the PCIe x1 issue I adjusted my settings as follows:
PCIe clock - 110MHz

voltages:
PCIe - +0.05V
MCH - +0.05V
RAM - +0.2V
CPU - 1.325V

I'm not sure why the others needed so much extra MCH voltage, mine seems to be perfectly stable with these settings (they must be pushing 500MHz+ FSB speeds or something).

[pokerezzet]

Vdrop problem seems almost gone with new F7a beta bios for the DQ6. 1,5375 in bios, 1,53 real.. Very nice news for us DQ6 owners.

[Krohling]

But, was the vdrop reduced or the sensor was calibrated?
Heh... who knows.

I'l try it later...

[Capt Proton]

CPU-Z "Mainboard" tab under "Graphic Interface." Or nVIDIA control panel-->Help-->System Information.

Thanks.

[mackintire]

I m updating my computer again.

This time I m installling a P35-DS3P rev2.0 with a QX6850.


Any recomendations on RAM

I was looking at using Corsair Dominator TWIN2X2048-8500C5D 2x1024

I was also looking at 4GB kits.

Do you have any advice?

Also this means my QX6700 is now up for sale.....priced to sell.... Find me on IM and leave a message.

[miahallen]

New G.Skill PC2-8000 2x2GB kits OCing pretty well.
Also check out this thread to see some OCing results.

[r4st4m4n]

I m updating my computer again.

This time I m installling a P35-DS3P rev2.0 with a QX6850.


Any recomendations on RAM

I was looking at using Corsair Dominator TWIN2X2048-8500C5D 2x1024

I was also looking at 4GB kits.

Do you have any advice?

Also this means my QX6700 is now up for sale.....priced to sell.... Find me on IM and leave a message.

Crucial Ballistix 8500 => cheap, micron D9 chips guarantee and very good performers: http://www.xtremesystems.org/forums/...d.php?t=143009

I fyou like the *bling, you can take them in tracers

Anyone having a link to the bios F6b for P35-DS3 ?

On station-drivers the link actually point to the P35C-DS3 and on softpedia in point to the F2 bios :/

EDIT: I found it: http://www.jzelectronic.de/jz2/index.php

[SySTaXer]

Anyone knows the difference between rev 1.0 and rev 1.1 P35-DQ6 boards ?

What has been changed/improved in rev 1.1 board ?

[Krohling]

Anyone knows the difference between rev 1.0 and rev 1.1 P35-DQ6 boards ?

What has been changed/improved in rev 1.1 board ?

Looks like only the Silentpipe was replaced by a smaller cooling system for chipset and mosftets, to widen the area for bigger CPU HSFs.

I don't know if the back panel parallel and serial ports were removed. Some newewer Gigabyte P35 revisions have more USB ports in the back panel.

[SySTaXer]

Looks like only the Silentpipe was replaced by a smaller cooling system for chipset and mosftets, to widen the area for bigger CPU HSFs.

I don't know if the back panel parallel and serial ports were removed. Some newewer Gigabyte P35 revisions have more USB ports in the back panel.

Thank you for the information. More backpanel USB ports is a plus in my book cause I use many USB devices and the expansion slots area of my mainboard is pretty crowded already. Adding the USB port bracket is not going to look pretty. Though my flat panel gives me two more USB ports.

Do you own the rev 1.1 board? If the answer is yes, can you post a rather high resolution picture of it?

Here are the pictures of both revisions I got from gigabyte's website but they are pretty small. I am assuming since the rev 1.1 board has its own webpage and BIOS files they might have also changed some electronic components, the clock generator IC for example. The clock generator IC on the rev 1.0 boards (ICS 9LPRS587EGLF if I am not mistaken) is not supported in the latest version of SetFSB as far as I know maybe the rev 1.1 board is supported. High-res picture helps with tracking down those kind of changes.

[mackintire]

Version 2.0 of the P35 Gigabyte boards has auto reset fuses on the USB ports. That should diminish some of the RMA returns.

[BuddhaTB]

Does anyone know of any sites still selling the GA-P35-DQ6?

[Richard Dower]

Version 2.0 of the P35 Gigabyte boards has auto reset fuses on the USB ports. That should diminish some of the RMA returns.

Huh?...what the hell are those?

[SySTaXer]

Correct, I used the CPU-Z method.

BTW, I stated that I boosted the MCH voltage as part of my fix. But I was just reading the posts on the last page and saw people having issues with too much MCH voltage as well. So for the record, I wasnted to be more specific. I'm currently running my Q6600 G0 at 400x8 with 4x1GB of Crucial Ballistic PC2-6400 with a 1:1 ratio and 4-4-4-12 timings.

When I was having the PCIe x1 problem, I was running the following settings:
PCIe clock - auto

voltages:
PCIe - normal
MCH - normal
RAM - +0.2V
CPU - 1.325V

To solve the PCIe x1 issue I adjusted my settings as follows:
PCIe clock - 110MHz

voltages:
PCIe - +0.05V
MCH - +0.05V
RAM - +0.2V
CPU - 1.325V

I'm not sure why the others needed so much extra MCH voltage, mine seems to be perfectly stable with these settings (they must be pushing 500MHz+ FSB speeds or something).

Thanks for sharing your settings. I was this ----> <----- close to replacing my P35-DQ6 with another board but everything is working perfectly fine now.

[SaleGseK]

I update my p35-ds3r with the new bios F7d waiting for a vdrop reduction, but that wasn't the case, same vdrop as before (F4-F6a-F6) only a few new things, nothing really importan imo.

[dominick32]

UPDATE
Issue still remains with the new f3i bios.
Currently in vista at 1740 mhz....

[Flinch]

DS3R and the latest F7b bios now shows actual voltages

[mackintire]

When USB devices die, they usually go out in style. Typically it destroys the device (usually a flash drive) and the USB port.
Asus often uses a drop resistor on their USB ports that makes using the port for high powered device more like playing roulette. Fair odds it will work, but not always.

With unprotected ports it gets more interesting, with the user often ending up shorting the entire board out.

I have 2 dead motherboards killed by the same genius with the same flash drive here where I work.

With auto resetting fuses on the USB ports. You can't destroy the port or the board with a shorted USB flash drive.


BuddhaTB: I have a brand new GA-P35-DQ6 Rev 1.0 here unopened if you want it. Toss me a resonable bid and its yours. Private message me for details.

[miahallen]

Thanks for sharing your settings. I was this ----> <----- close to replacing my P35-DQ6 with another board but everything is working perfectly fine now.

No problem, glad I could help

[dnottis]

Why does every bios for the DS3R say - fix PS2 compatibility issues.


F7d 2007/10/26 Support Intel 45nm Yorkfield CPU

Flashed - no diff, still droop.

[rge]

Vdrop problem seems almost gone with new F7a beta bios for the DQ6. 1,5375 in bios, 1,53 real.. Very nice news for us DQ6 owners.

I just updated to F7a on DQ6, getting the same decrease in vdroop.

On F6 at 1.475 in bios, CPUZ reads 1.44 (load). On that voltage, 3.9 stable, 3.93 would reliably fail orthos in 5-8 mins , 3.969 would fail to boot (reboot into auto).

On F7a at 1.475 in bios, CPUZ reads 1.472 (load), 3.93 orthos stable at least 4 hours (I shut off did not fail), 3.969 failed orthos after 2 mins. At 1.4875 in bios, still reads 1.472 CPUZ load, 1.488 idle.

Vdroop seemed little variable depending on the voltage set at, but ranged from .003 to .016 in F7a versus .03 to .035 in F6

[swaaye]

I read somewhere that vdroop can't be fixed by the BIOS.