Testers needed for new TH7II bios

[arman68]

I need some testers before releasing a new bios for the TH7II and TH7II-RAID.

The new bios is based on the official Abit BF, includes the usual tunings and updated HPT bios (v2.34), but more importantly it contains a new mod: a bios extension in the shape of an integrated FDISK utility (press CTRL-F when the logo appears to activate it).

For the moment I can send you a version which only has the new HPT bios + the BiosFDisk extension (ie: no tuning), as I am at work and only have access to CBROM on Windows XP. I will do a 'tuned' version this weekend.

I you want it, PM me.

WARNING: by flashing this bios there is a very high risk of rendering your bios chip (and computer) unusable. Additionally when testing the BiosFDisk extension there is the possibility of wiping out the contents of your hard drives!

Only test this bios if you have a bios savior AND spare hard drives. I also recommend you disconnect your existing hard drives. I have not been able to test it as I accidentely damaged my spare bios chip :-(

If everything works, I would like to get some feedback on the BiosFDisk extension. Eg: does it work only on HD connected to IDE 1 & 2 or also on HD connected to the HPT controller? Does it work on RAIDed disks? Etc...

All credits for the BiosFDisk extension go to smKen. I did not write it, he did

 

[ozzie_no_good]

hhmm cant wait till the bios is tested i cant test it have broken my bios ones and had te wait 3 weeks for new one and thats a long time

 

[phantom punisher]

excelent work arman68. this is indeed a cool mod. hope it goes flawlessly for you

 

[arman68]

I have not had any volunteer for testing. Anyone?

BTW, BiosFDsik is now translated from chinese to english.

 

[ozzie_no_good]

no i am to scared to screw my bios to heaven for the second time hehe

 

[Justin2050]

I have not had any volunteer for testing. Anyone?

BTW, BiosFDsik is now translated from chinese to english.

Well I didnt exactly "volunteer" but I took your bf+ bios and added the highpoint 2.34 bios to it of course updated the drivers as well and its ran very well for me so far and my hard drive benchmarks on atto were substanially higher on many lines of the test and went from a top score of like 42000 to 46000. I am now wondering if hyper threading could some how be extracted from the microcode of another chipset bios and then inserted into the th7II bios in the same way the highpoint 2.34 is. Although I currently have a 1.8a that wouldnt support HT even if the bios mod worked, knowing it worked though would give me a reason to get a new processor for my board, any input?

 

[TechTones]

How come when I go into a TH7-II bios with modbin6, some selections I can't find in the bios. For instance I would like to change the rdram multiplier from a default of Auto, to 400 so 1066
rdram will always boot up without the need for PC800 anymore.
The thing is I can't find it in modbin6.

Also, some of choices are in garbled letters or perhaps Chinese?
How do you know what they are? So how do you know how to change them?

Also, many of the choices are marked in "XXXXX" in either blue, white or black. Are the mobo makers trying to hide these choices from people? How do you find out what they are and how to change them?

 

[arman68]

How come when I go into a TH7-II bios with modbin6, some selections I can't find in the bios. For instance I would like to change the rdram multiplier from a default of Auto, to 400 so 1066
rdram will always boot up without the need for PC800 anymore.
The thing is I can't find it in modbin6.

Strange, it is there for me, and I can change the default. Note that you can still change the default by going into the bios setup screen, that's what I do. I always leave it to Auto when using modbin, but change the default manually after flashing the bios.

Also, some of choices are in garbled letters or perhaps Chinese?
How do you know what they are? So how do you know how to change them?

I never thought about that. My advice: don't touch it.

Also, many of the choices are marked in "XXXXX" in either blue, white or black. Are the mobo makers trying to hide these choices from people? How do you find out what they are and how to change them?

I do not think those are hidden options, but rather data referenced by the *real* options. Same here: don't touch it.

Well I didnt exactly "volunteer" but I took your bf+ bios and added the highpoint 2.34 bios to it of course updated the drivers as well and its ran very well for me so far and my hard drive benchmarks on atto were substanially higher on many lines of the test and went from a top score of like 42000 to 46000.

Thanks, but it does not help me much. There should not be any problem with replacing the HPT bios. The thing that worries me is the BiosFDisk extension. PM me if you can recover from a bad flash.

I am now wondering if hyper threading could some how be extracted from the microcode of another chipset bios and then inserted into the th7II bios in the same way the highpoint 2.34 is. Although I currently have a 1.8a that wouldnt support HT even if the bios mod worked, knowing it worked though would give me a reason to get a new processor for my board, any input?

As you say, enabling hyperthreading might be as simple as updating the microcode. I doubt it though, as all Xeon bioses I have worked with have an option within the main bios itself to enable/disable hyperthreading. We will probably need to pressure Abit to get this one done.

This is what I think of hyperthreading (I quote myself from this thread http://forum.oc-forums.com/vb/showthread.php?s=&threadid=124816&highlight=hyperthreading )

As part of my work I benchmark and test diverse applications on multiple servers, and I have not observed much benefit to be gained from hyperthreading. In many cases applications performance become worse and hyperthreading even causes a few to crash. I view it a bit like SSE2: until some applications are written specifically with hyperthreading optimizations there is nothing to be gained from it; and even then, I think it is certainly not suitable for the desktop as they are too many apps that would have problem with it turned on.

 

[TechTones]

OK I found out I was using a very old version of modbin

I see the choices there now but do NOT see the choices for "Auto, 300, or 400" selections for the rdram. The default is Auto. We need a default of 400 for it never to need pc800 again I think.

What do you think?

 

[arman68]

OK I found out I was using a very old version of modbin

I see the choices there now but do NOT see the choices for "Auto, 300, or 400" selections for the rdram. The default is Auto. We need a default of 400 for it never to need pc800 again I think.

What do you think?

I would leave it at Auto for the simple reason that if someone was using a CPU with a 133MHz FSB (like I do) but they are using PC800 RDRAM which cannot be oc'ed to 1066MHz they might have problems.

It is easy to change the CMOS default afterwards - I am wrong? I have not used PC1066 yet, so there may be something I am missing.

One rule I always follow, for drivers and BIOSes, is: the default settings need to be safe and not overclocked.

 

[TechTones]

The TH7-II WILL NOT boot up first time with pc1066. The board or the bios doesn't recognize 1066 rdram. Mine wouldn't boot up either with it. You need to use pc800 to boot it up, change the default from Auto to 400, shut down, put the 1066 and then boot up.

If you overclock too high and mess up, the bios reverts back to "auto", and need to take out the 1066 and reboot with 800 and start over.

You didn't know this??

 

[Jirnsum]

TechTones, isn't that only the case with Kingston 1066? I read that that is the only kind of 1066 RDRAM that won't boot at 3x

 

[TechTones]

I have Samsung 1066 too and it didn't boot with it. So no.

 

[TechTones]

I can run my Kingston 1066 at 3x no problem once the board is booted. Maybe it would still boot the first time if the default was 300. I never tried.

 

[Jirnsum]

Ok, thanks for clearing that up. I am trying to find where I read it though

 

[TechTones]

Yes there was an 850E board that just wouldn't allow a change from 4x to 3x with Kingston.

My Kingston uses Toshiba chips. Since then Toshiba does not make rdram anymore. Now Kingston uses Samsuey only and they had some problems with it a few weeks ago. So the SPD might have been reprogrammed.

 

[Jirnsum]

ok, thanx. I couldn't find the post. And I guess my boss expects me to get some work done today

 

[TechTones]

 

[waja]

I would leave it at Auto for the simple reason that if someone was using a CPU with a 133MHz FSB (like I do) but they are using PC800 RDRAM which cannot be oc'ed to 1066MHz they might have problems.

It is easy to change the CMOS default afterwards - I am wrong? I have not used PC1066 yet, so there may be something I am missing.

One rule I always follow, for drivers and BIOSes, is: the default settings need to be safe and not overclocked.

arman68,
There is something wrong with the TH7-II BIOS in that it will not get past a c1 error if you reset the BIOS and then try to boot with PC1066 memory. At this point, you have to install PC800 memory, boot up and then change the clock from AUTO to 400. (I believe, but do not know for sure, that the TH7-II BIOS is not able to properly read the PC1066 SPD.) After setting the BIOS to 400, there is no problem booting with PC1066 memory. Since I and a few others no longer have PC800 memory available, we are stuck using the BIOS we have.
So what Techtones would like is a test version of your BIOS with 400 as the default. If that works out, it would be a great benefit.
As far as someone having a problem with 400 as the default, I think that it is highly unlikely. The default FSB is 100MHz on a TH7-II. And I would believe that most if not all of those willing to take the risk of flashing your BIOS know that they should always check all settings after flashing a new BIOS. (Well, I could be wrong after see some of the posts here (not this thread, of course). )

Thanks,

 

[Jirnsum]

Just remembered that I wanted to say I agree with Arman68 on the Auto setting. I can see Techtone's point, but in taking away the pc1066 boot problems you are bound to create pc800 boot problems. Looks like the best thing to do would be creating two BIOSes

Edit: ooops. Just read Waja's post, and realized I am at fault here. One could always reset the CMOS with pc800, which would revert to 100 FSB with 400 memspeed. Thanks for clearing that up Waja