X48 Rampage Formula Preview.

[Alien Grey]

Glad to see you got the microcode sorted in the end.
You may find that slightly less vcore is required with the new microcode as apparently GTL improvements for QX9650 C0 was one of the fixes. I found this
Anyway with regards to the MSR issues I know that there are some people who believe that errors are expected.
Well I have done this test today on a DELL at work.

Both pass with flying colours.
If Dell can do this, why can't ASUS?
John

Because they don't want to. You should buy a new ASUS motherboard and maybe* than they are going to fix some BIOS issues for you.


* There's no guarantee that they are going to fix them and there's no guarantee that you ever going to see a BIOS with out issues.


@SoLoR

Did you build this R3 from scratch?

I mean did you also replace the Option ROM again in your last R3 build because I see a major improvement to mount my HDD in Ubuntu. With R2 it took a few seconds before the HDD was mounted.

[SoLoR]

Did you build this R3 from scratch?

I mean did you also replace the Option ROM again in your last R3 build because I see a major improvement to mount my HDD in Ubuntu. With R2 it took a few seconds before the HDD was mounted.

no R3 is R2+microcodes.

[Alien Grey]

no R3 is R2+microcodes.

I've also found another major improvement. The SMI latency dropped by 4 us with this BIOS. Going from 511 us to 507 us.

[Alien Grey]

I think that I discovered the most important fix that was ever done in the ASUS Rampage Formula BIOS history. I still can't believe it and I want someone else that also has this issue to try it out and confirm it.

Halleluja!! Praise the Lord! Loadline Calibration is fixed with 1002B-R3.

[SoLoR]

I think that I discovered the most important fix that was ever done in the ASUS Rampage Formula BIOS history. I still can't believe it and I want someone else that also has this issue to try it out and confirm it.

Halleluja!! Praise the Lord! Loadline Calibration is fixed with 1002B-R3.

I used to have this issue, however only with certain combination of settings (and this was totaly random)... at the end i found certain combination that worked fine for my overclock... anyway i didnt pay any attention to this since then and to check it i would again need to go play with overclock settings and do testings and im not sure if im up for it... its summer and its hot, dont want to torture my computer more then its neccessery

[Alien Grey]

I used to have this issue, however only with certain combination of settings (and this was totaly random)... at the end i found certain combination that worked fine for my overclock... anyway i didnt pay any attention to this since then and to check it i would again need to go play with overclock settings and do testings and im not sure if im up for it... its summer and its hot, dont want to torture my computer more then its neccessery

It's working but it doesn't mean that I can use it. There's still the 0x124 BSOD issue and now it's only a matter of time before one gets triggered if I enable LLC. But someone should try it out that doesn't have this 0x124 BSOD issue with his board.

[Alien Grey]

BTW have you had any luck on your MSR quest Alien?
John

Maybe there's a way to find out if it's necessary to have the MSR Fails fixed in the BIOS. There's a utility, called Read & Write Utility, that should make it possible to fix the MSR in Windows. This utility can be downloaded here: http://jacky5488.myweb.hinet.net/

[JohnZS]

That is a great find Alien
The problem is I do not have a clue how to use the program. I can find the MSR section and display data, however I am not sure of what it means and what needs to be changed.
Perhaps now you have located the tool and posted it maybe some of the experts here and chip in some advice?

SoLoR please could you PM me details of how you are able to extract the new Option ROM as I can only seem to extract option ROMs from the DX58SO2 and fail miserably with the Z68 based BIOS
I think I need a new tool
Thanks
John

[Spyrus]

I have LLC ON all the time from early bioses. 0x124 seems not to be related to LLC at least for my case, unless this has been broken in latest bioses.
But funny thing is that lately all my 0x124 will only happen minutes after boot and not after more than 4 hours of playing.

[SoLoR]

SoLoR please could you PM me details of how you are able to extract the new Option ROM as I can only seem to extract option ROMs from the DX58SO2 and fail miserably with the Z68 based BIOS
I think I need a new tool

i did not try to extract any option rom from efi bioses for a while, because there is always someone that does it before me so i dont need to... but i was doing this with phoenixtool...

[Alien Grey]

I have LLC ON all the time from early bioses. 0x124 seems not to be related to LLC at least for my case, unless this has been broken in latest bioses.
But funny thing is that lately all my 0x124 will only happen minutes after boot and not after more than 4 hours of playing.

I know that it doesn't have to do anything with LLC. The problem is that if I enable LLC I always run in to a 0x124 BSOD no matter if it's at full load or Idle. With LLC disabled I can at least have it stable running and the 0x124 BSOD onlu happens under full load with Prime95 or LinX. I also know that it isn't the CPU that is faulty. I'm almost 100% sure that it's a BIOS bug because before updating the Microcode it was almost impossible to boot in to Windows 7 with LLC enabled.

How long are these Microcodes already available?

I think long enough to say that ASUS doesn't care if you've got issues with there board.

[Alien Grey]

@Spyrus

Use BIOS 1002b-R3 and try with DRAM CLK Skew on Channel A and B Delay 50ps. This seems to stop the 0x124 BSOD on my board with CPU Clock Skew Delay 200ps and NB Clock Skew Delay 100ps with LLC Enabled. I need some more testing to see if it's really stable.

[Alien Grey]

I now know for sure that it's the BIOS causing these BSOD and instabilities. I tried to run LinX and it didn't BSOD but Windows gave me a warning if I wanted to stop running LinX to prevent damage, data corruption. I cancelled it and gues what happened...

Nothing at all. LinX didn't stop or crash and everything stayed stable. I think that now is the time to fix those MSR fails in BIOS 0001 for us ASUS.

[Commander_HK47]

This is an older site and may not fully be applicable to newer AIM bios roms(or other bios Mnf.), but i am finding it very useful to read through the BIOS workshop guide. In general it is covering a lot of the stuff we would want to adjust in our bios for this board.

Even bits about how might be able to get around AIMBCP save corruption.

http://www.rebelshavenforum.com/sis-...ubb=forum;f=52



What version of AMIBCP/ MMtool is being used?

I have 3.47, and 3.26 respectively.

I know that there are newer versions out, ie 4.06 and beyond. Among other tools needed for version 8 BIOSes.


To outline what we all are looking to resolve:

Update Microcode cpu patch to latest revision(SoloR and Alien Gray did this)

Update Intel OptionROM to latest stable version(SoloR did this? version: 11.xx)

Adjust bit values that cause IBI Test Suit MSR failures(unresolved)

Add adjustable ODT item/values(SHOW), or at minimum edit per ROM build(unresolved)

What else?

[Alien Grey]

This is what it should be.

1. Intel BIOS Implementation Test Suite MSR Fails.
2. Individual DRAM CLK Skew (A1-A2-B1-B2).
3. Adjustable MCH Read ODT (8T-9T-10T-11T-12T).
4. Adjustable DRAM Read ODT (3T-4T-5T).*
5. Adjustable DRAM Write ODT (6T-7T-8T).*

* I don't know if it's really necessary to have these available in the BIOS.

[leo27]

That's a good find Commander_HK47, so it looks like it could be possible to add in the options we want assuming we know the assembly code necessary to perform the function.

I find it ridiculous we have to program the BIOS ourselves to get it stable

So who's up for learning BIOS coding??

[jaytee]

I just wanted to thank you guys for this thread, especially Alien Grey, SoLoR, JohnZS, Neo_rtr & greg.m! I've been following this for at least the past year. Thanks to you guys I have a fast, stable, quiet machine with a respectable 24/7 overclock. I have learned an awful lot here! Thank you, guys!

[Alien Grey]

That's a good find Commander_HK47, so it looks like it could be possible to add in the options we want assuming we know the assembly code necessary to perform the function.

I find it ridiculous we have to program the BIOS ourselves to get it stable

So who's up for learning BIOS coding??

In all the time that I've been posting here in this thread, I don't think that I've ever seen someone posting from ASUS. If ASUS wants me to buy one of their boards for my next build they are going to have to do a lot better than what they are doing now. It's time that they realize that I'm not a fool and that I won't help them sell their boards by posting on the forums if it's to much asked to do something back when it's necessary.

So, is it ridiculous that we've got to fix the BIOS our selves? - No, it isn't... It's hilarious.

It's so hilarious that even the moderator from the ASUS forum thinks that maybe it's better to try your luck with an other company in the future.

I just wanted to thank you guys for this thread, especially Alien Grey, SoLoR, JohnZS, Neo_rtr & greg.m! I've been following this for at least the past year. Thanks to you guys I have a fast, stable, quiet machine with a respectable 24/7 overclock. I have learned an awful lot here! Thank you, guys!

You're welcome.

[Commander_HK47]

That's a good find Commander_HK47, so it looks like it could be possible to add in the options we want assuming we know the assembly code necessary to perform the function.

I find it ridiculous we have to program the BIOS ourselves to get it stable

So who's up for learning BIOS coding??

Well that guide mainly pertains to Award BIOSes, AIM bios are a bit different. They are a bit more complex, non modular, and less open ended.
Seems like Adding option ROMS and patching CPU micro codes is the easy/trivial stuff for AIM BIOSes when compared to everything else there after. It becomes that much more difficult as you have to recompile(first you have to decompile it...)to add new features it seems.


Here is a post that contains a link to a write up that covers AIM BIOS reverse engineering circa 2005:

http://www.rebelshavenforum.com/sis-...;f=52;t=000144

Direct link to the doc:
http://sites.google.com/site/pinczak...-engineering-1

Seems like one would need to know x86 assembly language, and Hardware codding, ect to be able to tackle these problems.


I am still looking around for more current information on this matter.
I currently have a hobby level interest in learning about this stuff and bettering my self with it But by no means am i a programer

[Commander_HK47]

In all the time that I've been posting here in this thread, I don't think that I've ever seen someone posting from ASUS. If ASUS wants me to buy one of their boards for my next build they are going to have to do a lot better than what they are doing now. It's time that they realize that I'm not a fool and that I won't help them sell their boards by posting on the forums if it's to much asked to do something back when it's necessary.

So, is it ridiculous that we've got to fix the BIOS our selves? - No, it isn't... It's hilarious.

It's so hilarious that even the administrator from the ASUS forum thinks that maybe it's better to try your luck with an other company in the future.

I wont be buying "ASS out of US" board come socket R build time... In terms of my over all experience with it, i would say that its been ok, but for the price i payed for the first one i got at lunch Hindsight seems like i could have done better. I sure feel like we got burned in terms of Asus dragging their feet on providing updates to fix things like this, and then using product EOL as an excuses to wipe their hands all together.

[BlueMax1916]

There was a time when ASUS fixed BIOSes of boards that were two years out of production already. Not anymore it ssems. I bought an ASUS P5A mainboard in 1998 and ASUS provided an ACPI fix for Windows XP with a new BIOS revision in 2002. EOL was in 2000. This was somewhat unexpected by me but very welcome

Seems that company is not that dedicated to their own products anymore as they were in the past.

[Alien Grey]

I think it's better to leave Loadline Calibration disabled. If I would want it to run stable with Loadline Calibration enabled it's going to need more voltage across the board to keep it stable and question is if it would still be possible to use the same clock skews. It's what I think the reason why it's so difficult to keep it stable running.

My conclusion is that Loadline Calibration disabled improves stability. Enabling Loadline Calibration results in kernel panics and it never did that with Loadline calibration Disabled at 100% load. Maybe when I've got the time to do it I'll try to find stability with Loadline Calibration to actually see the differences between enabled and disabled.

[Alien Grey]

@JohnZS

Download the Intel® 64 and IA-32 Architectures Software Developer's Manual
Combined Volumes 3A and 3B: System Programming Guide, Parts 1 and 2

Look at Appendix Model-Specific Registers (MSR). Chapters B.1,B.2 and B.3. You'll find the MSR's that fail consistency and you'll also notice that at least one of them has to do with the L2 cache. This explains the 0x124 BSOD's and it also means that ASUS is responsible for the issues with the BIOS. These MSR fails should be fixed and it's ASUS job to do that.

[Antinomy]

Do you have a reference board where BITS doesn't fail on MSR tests? Then I could easily take the microcode patches out from there. As for your requests the easy way is this: http://www.rebelshavenforum.com/sis-...132;p=1#000006
Add a new menu (which is real to do) and then write an optional ROM that will set registers the same way you do with your tools (MCHbaredit analogs from what I've seen) but integrated in BIOS. So it'll apply settings after POST but before Windows. The hard way is to disassemble the main BIOS module and write what you want there. I'm not a programmer so I can't help with this one definitely.

[Alien Grey]

Do you have a reference board where BITS doesn't fail on MSR tests? Then I could easily take the microcode patches out from there. As for your requests the easy way is this: http://www.rebelshavenforum.com/sis-...132;p=1#000006
Add a new menu (which is real to do) and then write an optional ROM that will set registers the same way you do with your tools (MCHbaredit analogs from what I've seen) but integrated in BIOS. So it'll apply settings after POST but before Windows. The hard way is to disassemble the main BIOS module and write what you want there. I'm not a programmer so I can't help with this one definitely.

Maybe the Intel DX48BT2 if it's possible to edit the BIOS file?

Adding a new menu and apply settings before OS load should be fine.