ASUS mobo beep codes?
in [Home Built PC]
|
From: kony on 31 Aug 2006 17:14 On Thu, 31 Aug 2006 09:28:28 GMT, "beenthere" <Waiting(a)Home.com> wrote: > >"Ian R" <sorry(a)nospamthanks.com> wrote in message >news:zaSdnVf6o5hpOWvZRVnygw(a)bt.com... >> Hi >> >> I'd be very grateful if someone has any info on the meaning of POST beep >> codes for an ASUS A8N-SLI Premium. >> >SNIPPED (read the OP) > >Beep codes are a product of the BIOS. >When your machine first boots, check the black screen, >(bottom right, for the BIOS manufacturer). >That tells you what BIOS you`ve got. >You need the beep code set for that BIOS. How do you suppose he'll see that BIOS identification on the screen while his system only beeps instead of posting? Generally, there is either a sticker on the EEPROM or you can download the bios update from manufacturer's website to see what it is. Asus has often used slightly modified Award Bios, but that particular board I don't know for certain. > >The BIOS delivers beeps to your on board speaker, >Not your Audio card. Not always, some boards dont' even have a speaker, and some Asus boards have/had a jumper where you select to have the voice routed to an onboard speaker OR to the audio output jack (though if it goes to the tiny buzzer speaker, you can barely hear what the voice is saying. It brings up a funny story, one time I was overclocking and it wouldn't post, and I hear the voice over the buzzer speaker. I could've sworn the voice said "It's impossible", kinda like my motherboard was suggesting to me that there was no way in hell I'd be able to run the CPU that far overclocked. I finally jumpered it to the rear audio output and then I can make out that it said "CPU failure" or something like that, so it was telling me what I already knew (that it just wasn't going to overclock to that speed without more voltage), but not so sarcastically as it seemed. >The Audio card is controlled by your Operating system. >HTH. Many semi-modern boards (including some from the P3 era) did output to the integrated sound, audio out jack. I happen to have one here still, an MSI 6368, that also produces that final beep at the end of POST right before booting the OS.
From: kony on 31 Aug 2006 17:18 On Thu, 31 Aug 2006 11:44:42 GMT, nospam(a)needed.com (Paul) wrote: >There is a BIOS setting called "Speech IC Reporter" and it is >enabled by default in the BIOS. If it was disabled, I don't know if >it is possible to guarantee it would be reenabled after the >CMOS is reset. That would likely require a working processor, >able to complete at least part of the POST sequence, to enable >it again. No, the Speech IC Reporter doesn't necessarily require a working processor, it is controlled independantly by another monitoring chip... or at least, it can be, because I had an Asus board that did (A7V333, IIRC), although it had a different chip, the OP's board appears to have an ITE something-or-other Super I/O. >And in any case, I'm not sure that the disable setting, >could disable an error message indicating the CPU had failed, >as at least a couple of error messages are generated based on >a timer expiring, and there is no (easy) way to disable those >messages from coming out. On same board I was referring to above, the speech reporter could be disabled entirely, nothing would trigger it AFAIK- I know the CPU failure wouldn't if it was jumpered to disabled, but this was with a jumper, not a matter of loading up stored BIOS settings so it might depend on which method the board supports.
From: Paul on 31 Aug 2006 21:48
In article <1akef295q76lctgt9nul9u643pv9v5km30(a)4ax.com>, spam(a)spam.com wrote: > On Thu, 31 Aug 2006 11:44:42 GMT, nospam(a)needed.com (Paul) > wrote: > > > >There is a BIOS setting called "Speech IC Reporter" and it is > >enabled by default in the BIOS. If it was disabled, I don't know if > >it is possible to guarantee it would be reenabled after the > >CMOS is reset. That would likely require a working processor, > >able to complete at least part of the POST sequence, to enable > >it again. > > No, the Speech IC Reporter doesn't necessarily require a > working processor, it is controlled independantly by another > monitoring chip... or at least, it can be, because I had an > Asus board that did (A7V333, IIRC), although it had a > different chip, the OP's board appears to have an ITE > something-or-other Super I/O. > > > > >And in any case, I'm not sure that the disable setting, > >could disable an error message indicating the CPU had failed, > >as at least a couple of error messages are generated based on > >a timer expiring, and there is no (easy) way to disable those > >messages from coming out. > > On same board I was referring to above, the speech reporter > could be disabled entirely, nothing would trigger it AFAIK- > I know the CPU failure wouldn't if it was jumpered to > disabled, but this was with a jumper, not a matter of > loading up stored BIOS settings so it might depend on which > method the board supports. The source code for the Winbond speech chip is available. "No CPU installed" message is from a hard wired signal. "System failed CPU test" and "System failed memory test" use the timer inside the Winbond chip. The rest of the messages in the table, can only be triggered by the system processor writing the appropriate event code into the Winbond chip. If the BIOS setting is disabled, it is easy to see that everything except the three cases above will be disabled. What I don't see, is how the "No CPU installed" message can be stopped, as the signal just comes from the CPU socket and goes to the Winbond chip. The "System failed CPU test" runs off the timer, and if the CPU is failing, I don't see a hardware mechanism to stop the message from coming out from the Winbond chip. (A jumper that interrupts the audio signal itself would certainly do the job, but a number of boards no longer use jumpers for the Winbond speech chip. And the Winbond chip pinout doesn't have a "CS" or a disable pin on it.) The "System failed memory test" requires the timer to be reloaded, so the CPU has to be running and the CPU plays a part in that message being emitted. The reason a timer is used, is in case the CPU crashes while the memory test is running. If the CPU finishes the memory test, the CPU clears the timer in the Winbond chip, before it can be triggered. If the CPU doesn't finish the memory test, and the timer trips, the timer sends the "System failed memory test" voice sample. For all other error messages, the error conditions arise as part of the POST testing. The CPU sends the appropriate command to the Winbond chip, when it runs into an error condition later in the POST. For example, on a motherboard here, it takes 30 seconds before I hear "No keyboard connected", when the keyboard is not connected to the computer. The source code for the Winbond speech chip, is in the Winbond_Voice_Editor package, available for download from Asus (about 17MB). I think there are a couple of text files in there, with snippets of code. One file contains the sequencer code executed by the Winbond chip. The 8 pin DIP serial EEPROM, contains both the sequencer code and the voice samples, and the Winbond chip relies on the serial EEPROM for all of its needs. Paul |