X48 Rampage Formula Preview.

[greg.m]

You are not the only one thinking of buying other brands.

My next purchase will be a new Graphics Card and it isn't going to be a ASUS one.

I've been very happy with my previous ASUS products and never complained about anything. This board has been nothing but trouble for me. It started when i first build this system with a faulty LCD Poster connector on the board. It took me three days and a lot of frustrations before i found the problem. Two pins in the connector made contact with eachother.
My second problem was the Thermal Sensor Cables. Why the hell didn't they include them in the box. They only cost 3 Dollars and it wouldn't have been a problem for me to pay an extra 3 Euros to have them. In the manual you can read that you can buy them optional. They aren't available in Europe. It looks like someone from the ASUS sales forgot to send them to Europe.
I've got a nice Swiftech watercooling system to cool my CPU, NB and Graphics Card. I've bought it to have lower temperatures and less noise from the fans. The fan speeds on this board have been a frustration for me from day one. For months i've been asking ASUS to do something about it.
Why can't they just give us lower fan speed settings in the BIOS? I don't need that high speed because of my watercooling and i want less noise on my ears. Was it a mistake from me to use three 120mm fans in my case?

I think it's time that ASUS start fixing the problems. Maybe they don't care if we don't buy their products anymore at least not now. Two generations of motherboards further they will see that they made a mistake because the enthusiasts will be using an other brand and the guys that can use a little bit support won't have no choice than to buy that brand to.

Hi Alien Grey!
I allways check your threads in asus site. Im glad now that youve started writing here with the other capable guys. Maybe now we are gonna see some results with all problems WE have.
I would like to ask whats your watercooling system specs.

[mikeyakame]

I would add to your letter to fix the backlight issue of LCD Poster.
If I set from BIOS the backlight OFF, it stays OFF untill S3 resume. From that point it is always ON.

Example:

- backlight in BIOS is OFF
- starting a cold boot
- the light is ON until the last POST step
- as soon the LCD Poster is displaying time, the light goes OFF (as programmed in BIOS).
- I order Vista to put PC in S3 mode (S3 is forced in BIOS)
- the PC goes OFF except the power on led that is blinking
- I wake up the PC touching a keyboard key
- after some seconds I have my PC completely on and functional
- the light of LCD poster is ON (not correct).

I can duplicate in my system this bug 100% of times.

Heh thats part of the S3 corruption "bug"

Backlight staying on means the skewing is completely fubared to the point something going to be corrupted soon! Thats the only time I've ever had it not work properly. I wish this "backlit bug" was a real indicator of errors but it's not it just so happens that one of the critical segments (maybe POST_SEG) that handles it ends up with invalid data and result is light doesn't go off. It's bad news no matter how u look at it or why.

My rule is if it ever stays on after INIT ROM or anytime during Windows, either if outside OS I will immediately power it off or within OS shutdown and if that fails power off. Rather lose data than corrupt it. The former is easier to fix.

I speak from corrupting OS/wiping out half my registry experience here ;D I've done it more times than i'd like to ever admit! I've actually done it so many times that I've realized OS partition backups are much easier than full reinstalls I don't mind it when it's all fault of my own it just sucks when it isn't!

I learnt about the dangers of Performance Level phase pull-ins that way! I think the worst one was when I've corrupted the boot manager for Windows! THat was pretty bad!

[Alien Grey]

Some guy at ASUS Rampage Formula forum says that the CPU 3 Pin Fan issue is already fixed from BIOS 0410.

Can anyone with a 3 Pin CPU Fan confirm this?

The fan on the CPU Fan connector on my board is always at the same speed, idle or load.

[SoLoR]

Since ASUS is not updating ich9r and jmicron roms, i decided to find whatever versions i could extract and where latest and made custom bios that you can download here: http://rapidshare.com/files/185584480/c0701.ROM

it has Intel(R) Matrix Storage Manager ROM v8.5.0.1030 from EVGA x58 mobo (yes ICH10R ROM is working fine on ICH9R) and JMicron(R) JMB368 v1.06.78 from some DFI mobo, probably it doesnt matter that much i was just messing around trying to learn some new things i know that at least IMSS ROM v 8.6.0.1007 is out in some intel mobo bioeses but they are impossible to extract, so if someone knows for any normal award/ami bios that is using newest IMSS rom please tell me!

[Alien Grey]

Hi Alien Grey!
I allways check your threads in asus site. Im glad now that youve started writing here with the other capable guys. Maybe now we are gonna see some results with all problems WE have.
I would like to ask whats your watercooling system specs.

Hi greg.m

Nice to see someone else found the way to XS.

I'm using a Swiftech H20-220 Apex Ultima Plus. I'm very happy with it. I never have to worry anymore about the temperatures.

[Alien Grey]

I've actually done it so many times that I've realized OS partition backups are much easier than full reinstalls

Always keep a Norton Ghost Backup image from your OS partition, in case you f****d up your OS. It takes 10 minutes to make one and another 10 minutes to restore it.

[mikeyakame]

Since ASUS is not updating ich9r and jmicron roms, i decided to find whatever versions i could extract and where latest and made custom bios that you can download here: http://rapidshare.com/files/185584480/c0701.ROM

it has Intel(R) Matrix Storage Manager ROM v8.5.0.1030 from EVGA x58 mobo (yes ICH10R ROM is working fine on ICH9R) and JMicron(R) JMB368 v1.06.78 from some DFI mobo, probably it doesnt matter that much i was just messing around trying to learn some new things i know that at least IMSS ROM v 8.6.0.1007 is out in some intel mobo bioeses but they are impossible to extract, so if someone knows for any normal award/ami bios that is using newest IMSS rom please tell me!

JMicron one shouldnt be a problem, but Intel MSM ROM might cause instabilities on raid arrays that already exist. 7.5.x are for ICH9R, 8.x are for ICH10R. I considered doing this but after reading Intels documentation about it, I retracted my ideas.

Cool though mate, nice work

[RealTelstar]

Some guy at ASUS Rampage Formula forum says that the CPU 3 Pin Fan issue is already fixed from BIOS 0410.

Can anyone with a 3 Pin CPU Fan confirm this?

The fan on the CPU Fan connector on my board is always at the same speed, idle or load.

Yes, i think to remember so.
I have a double connector to my ultra kaze and keep it always at 100%, but i remember the bios would allow % change.

[Alien Grey]

If it's true that they already fixed that i find it very annoying that the ASUS Tech Support can't say that to me.

The fan profiles don't do their job as should be.
With the CPU Fan Control enabled in the BIOS and Performance Mode profile selected the fan should start spinning faster according to the load of the CPU, no matter what the CPU temperature is.
That's what Performance Mode profile should do and that's what it did on my P5WD2 Premium.

With the fan profile Optimal the fan should start spinning faster as the CPU temperature rises.

This is something that should be fixed by the ASUS BIOS Engineers.

Some guy on the ASUS Rampage Formula forum is using a 4 Pin CPU Fan and says that the fan Profiles in the BIOS don't work like they should be. The fan only starts spinning faster when his CPU reaches 60° Celsius. That's not how it should be.

[mikeyakame]

Figured out the AI Clock Twister bug and can replicate it and know how to fix it. Weird.

Bug:

AI Clock twister incorrectly setting DRAM skews at high FSB, cmos register storing menu item index select value not updating?

Fix:

On changing DRAM skew values manually forcing cmos register update by changing AI Clock twister setting to another and then back to previous consecutively.

This resets my DRAM skews to the correct values for the AI Clock Twister setting and thus don't have to keep adding advance skew to fix it up!

[SoLoR]

JMicron one shouldnt be a problem, but Intel MSM ROM might cause instabilities on raid arrays that already exist. 7.5.x are for ICH9R, 8.x are for ICH10R. I considered doing this but after reading Intels documentation about it, I retracted my ideas.

Cool though mate, nice work

well im using raid0 and it seems to work fine, ill report if there will be any instabilitys, also i finaly found method of integrating slic in to it without that y "invalid boot block" error in ezflash (i dont want to force dos flash, with something that is giving so dangerous errors).

so more updates: http://rapidshare.com/files/185925112/C0701_SLIC.ROM

this is with asus oem SLIC, updatable from ezflash without errors, you just need key+certificate and vista is activated. Before someone attacks me, yes i DO have original vista, but after i changed 2 components (cpu+mobo), it didnt want to activate over internet anymore and since i cba with phone activations i finaly found "safe" solution

edit: also some wierd bug i noticed couple of days back, FSB VTT went on MAX for no aperant reason, it didnt happened again after i changed it back tho. some wierd glitch i guess, but can fry your cpu....

[mikeyakame]

Hehe yeah thats why editing 1B module can be risky unless you know what you are doing SLIC data is stored inside 1B module in ACPI SLIC table. Sop you just need to be carefully when screwing around with a tool that patches it, because you don't know if it screws up any index pointers for bios functions, etc.

EVGA X58 is using Award bios right? So you just extracted the PCI option for Intel MSM with module extraction tool cbrom or something like that if i recall then replaced one inside RF rom with MMTool

I need to find some time to figure out where the Rampage II Extreme / P6T store the Nvidia SLi certificate too. I would assume its kept inside 1B module.

[SoLoR]

yes, used cbrom for award bioses and replaced them with mmtool in rampage bios. Also there are 2 different sizes of imsm rom, one is 64k (most mobos are using also rampage) and other is 79k (some award mobos are using) and i have no clue what is different, same versions and everything with different size... for adding slic i was using http://drop.io/slictool with SSV3 method inserted "manualy" with mmtool, that automatic module insertation is giving errors with ezflash. And if someone wants extracted imsm 8.5.0.1030 rom is HERE.

[Zucker2k]

Figured out the AI Clock Twister bug and can replicate it and know how to fix it. Weird.

Bug:

AI Clock twister incorrectly setting DRAM skews at high FSB, cmos register storing menu item index select value not updating?

Fix:

On changing DRAM skew values manually forcing cmos register update by changing AI Clock twister setting to another and then back to previous consecutively.

This resets my DRAM skews to the correct values for the AI Clock Twister setting and thus don't have to keep adding advance skew to fix it up!

Nice find, Mike. What bios are you running? I wonder if this affects all bioses.

[mikeyakame]

0701 mate. By consecutively I mean just ie. I use moderate, so i set it to light, set the dram clock skew i want then change it back to moderate and save settings.

If I dont do this it does all kinds of wacky things, like I advanced 50ps more than previous and it set to Fine clock delay from 5 to 0. Instead of 5->4. 0 is shown by Fine clock delay for slot as missing. My skews went from 3/2/3/2 to 7/6/7/6 between boots sometimes without touching a setting. This seems to fix that so far at least. Now my 4/2/4/2 is 4/2/3/2 with Normal on CHA, 50ps advance CHB which is correcto.

It might happen at low FSB dont know though, I've pretty much always used above 460+ since I've had the board. Only few times have I used less.

To me it seems like if you change the DRAM clock skew without changing AI clock twister setting to another it is "twisting" the delay skews more than it is supposed to.

[mikeyakame]

yes, used cbrom for award bioses and replaced them with mmtool in rampage bios. Also there are 2 different sizes of imsm rom, one is 64k (most mobos are using also rampage) and other is 79k (some award mobos are using) and i have no clue what is different, same versions and everything with different size... for adding slic i was using http://drop.io/slictool with SSV3 method inserted "manualy" with mmtool, that automatic module insertation is giving errors with ezflash. And if someone wants extracted imsm 8.5.0.1030 rom is HERE.

Did you check the one the R2E is using?

Nevermind.
Heres 8.0.0.1038 from Rampage II Extreme bios. It's exactly 65536bytes.

Becareful though mate, I just replaced the 7.5.x with 8.0.0.1038 in a test bios and MMTool doesn't look like its writing module 80 correctly, Image Information. Its 40 bytes larger and all over the place.

Heres a binary compare.

[ 1 ] D:\INCOMING\Bios.Stuff\ii-orig.bin 20/01/2009, 1:48:35 am
[ 2 ] D:\INCOMING\Bios.Stuff\ii-update.bin 20/01/2009, 1:48:21 am
------------------------------------------------------------------------------------------------------------------------
binSmartModebinMode
00000000 [ 1 ] * 09 00 60 04 01 34 2E 31 34 10 00 10 02 01 01 03 ..`..4.14.......
00000000 [ 2 ] * 07 00 10 02 01 01 03 14 00 20 08 01 E0 FF 00 00 ......... ..àÿ..

00000010 [ 1 ] * 1D 00 20 08 01 E0 FF 00 00 50 FF 00 00 2A 00 2F .. ..àÿ..Pÿ..*./
00000010 [ 2 ] * 50 FF 00 00 21 00 2F 08 01 E8 FF 00 00 EC FF 00 Pÿ..!./..èÿ..ìÿ.

00000020 [ 1 ] * 08 01 E8 FF 00 00 EC FF 00 00 37 00 30 08 01 00 ..èÿ..ìÿ..7.0...
00000020 [ 2 ] * 00 2E 00 30 08 01 00 00 02 00 00 00 01 00 45 00 ...0..........E.

00000030 [ 1 ] * 00 02 00 00 00 01 00 4E 00 40 06 03 04 10 FF FF .......N.@....ÿÿ
00000030 [ 2 ] * 40 06 03 00 00 00 80 1B 00 01 10 FF FF FF FF 04 @.....€....ÿÿÿÿ.

00000040 [ 1 ] * FF FF 00 00 00 80 1B 00 01 10 FF FF FF FF FF FF ÿÿ...€....ÿÿÿÿÿÿ
00000040 [ 2 ] * 10 FF FF FF FF 5E 00 50 0A 02 01 FF 00 00 00 00 .ÿÿÿÿ^.P...ÿ....

00000050 [ 1 ] * 50 0A 02 01 FF 00 00 00 00 10 00 00 00 01 FF 00 P...ÿ.........ÿ.
00000050 [ 2 ] * 10 00 00 00 01 FF 00 20 1C 00 F0 DF 00 00 FF FF .....ÿ. ..ðß..ÿÿ

00000060 [ 1 ] * 20 1C 00 F0 DF 00 00 FF ..ðß..ÿ
00000060 [ 2 ] * 60 04 01 34 2E 31 34 FF FF 51 12 02 00 00 00 00 `..4.14ÿÿQ......

00000070 [ 1 ] *
00000070 [ 2 ] * 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 01 ................

00000080 [ 1 ] *
00000080 [ 2 ] * 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................

------------------------------------------------------------------------------------------------------------------------

So if this isn't written properly chances are something else might not be either, you need to do a binary compare for the rom images and check the original against the changed one. You gotta be really careful when doing this kind of thing as there are a lot of static index's and address pointers in a bios so you don't know what you might break by replacing, changing or patching modules.

Other problem is X48 AMI bios are slightly different than P35/P45 AMI bios. I haven't figured out what it exactly is though but it's enough to completely break AMIBCP when loading it, and after debugging it (and yet to analyze 400mb segment dump properly) all I know is AMIBCP is moving a weird memory address offset to EAX register and seg faulting when its trying to access memory address that isn't allocated to it. It doesn't do this on any P45 bios.

[Rickpatbrown]

While we're complaining about stuff . . . Is it possible to enable a system shut down based on CPU temp. I know that you can set it to shut down based on the NB and the temperature cables (SOLD SEPERATELY!), but I wasn't able to find a setteing for the CPU. This would be very useful for us watercooling guys in the inevitable case of a pump failure.

[mikeyakame]

While we're complaining about stuff . . . Is it possible to enable a system shut down based on CPU temp. I know that you can set it to shut down based on the NB and the temperature cables (SOLD SEPERATELY!), but I wasn't able to find a setteing for the CPU. This would be very useful for us watercooling guys in the inevitable case of a pump failure.

CPU will automatically power off if TJMAX distance is reached. My Q6600 G0 chip seems to cut voltage at about 85-90c on any core.

There is CPU TM which we disable also, but that is for throttling when temps get too high and can cause instability.

[Brama]

Heh thats part of the S3 corruption "bug"

Backlight staying on means the skewing is completely fubared to the point something going to be corrupted soon!

I am not so convinced...

Probably you are right, if this happens after a cold boot but in the case back light goes ON after S3 (instead to be OFF as programmed) the bug is present from BIOS 219, with or without overclock.

[mikeyakame]

The Rampage II Extreme has a fix for a similar sounding problem in one of its newer bios'. Who wants to guess whether this is a global or localized issue Sounds like an AMI Core8 issue to me, the S3 problem I mean.

[Alien Grey]

I guess it's both. I would say it's a global issue but the ASUS Tech Support is going to say that it's a localized issue because no one else reports a problems with that. So the question is who is right?
I think the ASUS Tech Support because of the fact that i don't use it and i've got no problems with it.

It's a serious problem and the least you would expect is that the ASUS BIOS Engineers fix it.

[mikeyakame]

http://dlcdnet.asus.com/pub/ASUS/mb/...0_61026520.zip

Updated Soundmax drivers v6.10.2.6520 dated 10/07/2008

Blackhawk are 6.10.2.6480 dated x/03/2008

These are the drivers Asus supplies for P6T6 revolution X58 board which uses ADI chipset for sound, so luckily they supplied a nice newer updated driver to go along with it!

Enjoy. I;ve installed so don't worry they work as they should. Contains driver for all the ADI chipsets anyhow.

Blackhawk has been given a face lift which it needed like Bush needed that awesome shoe

[Rickpatbrown]

CPU will automatically power off if TJMAX distance is reached. My Q6600 G0 chip seems to cut voltage at about 85-90c on any core.

There is CPU TM which we disable also, but that is for throttling when temps get too high and can cause instability.

Will this harm the CPU at all? Sorry for being off topic, but I have no idea what you guys are talking about

[Zucker2k]

Will this harm the CPU at all? Sorry for being off topic, but I have no idea what you guys are talking about

It's simple, it's diificult to kill a modern cpu through heat, as it is rigged to shut down before the cores burn out.

[Alien Grey]

This is a copy from my post at the ASUS Rampage Formula forum.

I received an email from the dutch ASUS Tech Support today asking me what can be done to fix the issues that we are experiencing with the ASUS Rampage Formula BIOS.

Tomorrow i'm going to answer them back and send them all the information they need to fix the issues.

All the issues that we've been discussing here are going to be in my answer to them. Mikeyakame from XS is going to write a note for me with all the issues that they have found with the BIOS.

For now it looks that all the hard work, research and hours of stress testing are finally going to pay off.