For hotflashing two things are relevant:

- the EEPROM needs to be supported by Uniflash and the donor board
- the 'donor' board you flash on needs tp support Uniflash

Apart from that, any differences between boards is irrelvant.

If you are getting data verification errors, either the EEPROM is dead (not implausible!) or the flash is not (fully) supported by the board.

Have you been able to flash anything else on this board? Can you flash the board's own BIOS with Uniflash?

When you're doing a hot swap flash you generally don't press the chip into the socket fully - I've had issues similar to that where it was caused by the chip not quite making good enough contact.

Back in the Nforce days when hot swap BIOS flashing became popular, I discovered the /F switch which: "/F - stands for reprogramming by means of the system BIOS. Most contemporary BIOS's feature the procedure of FlashROM reprogramming. The key /F enables AwardFlash to reprogram FlashROM with the algorithms of the current BIOS version. If a mainboard peculiarities do not allow applying AwardFlash Writer algorithms, you should use the key /F." So you may be able to use Awardflash to do this if Uniflash can't.

This is why buying a TL866 or equal pays itself in just a hand ful of uses IMHO, and if you flash wrong bios image or brick your board you can more easily fix it 😀
sorry know that is not the answer but as a member here since 2015 thought you would have one by now....

Horun wrote on 2023-10-21, 02:13:

This is why buying a TL866 or equal pays itself in just a hand ful of uses IMHO, and if you flash wrong bios image or brick your board you can more easily fix it 😀
sorry know that is not the answer but as a member here since 2015 thought you would have one by now....

I wholeheartedly concur. That being said, if OP has only needed one once since 2015, it would probably be more cost effective to order a flashed-to-order chip on the unnameable auction site.

I can't find anything about a GA-5DX. Typo ?

Is the board that you are using to flash equipped with a jumper that needs to be set for 12V vs 5V VPP and if so, is it set to 5V ?

🤣 maybe a GA-5AX ? The only GA-5DX is GA-5DXSL I know of and a diff era ;p

Sorry guys, it is ga-586dx
As far as I can tell it is perfectly compatible with Uniflash but I did not use it to update the board.
I used it to hotflash a m577, but did used a regular awdflash program.
This time I used aflash and awdflah as well, but both hangs (aflash crash and reboots the system!).
I have the feeling that there is a problem with the way the P5A-b update the bios? Maybe it needs a particular version of aflash OR awdflash?

For the programmer, I do have a ch341a and was looking for something for ancient chips actually.

Got no specific experience of those boards or Uniflash, but if I'm to guess, using the /F parameter will force chip specific erase commands and page mode writing. You might want to try going back to the basics and erase the chip by writing 1 Mbit of 0xFF bytes, and then write the BIOS image byte by byte instead. Should work with any 29 series EEPROM.

The only reason for those functions is time saving when programming large batches of chips. When programming a single chip it doesn't really matter if it takes 10 seconds or a minute.

Also, if it's possible to manipulate the write/read timings in Uniflash, you might want to try to go a little bit slower.

Otherwise I agree that a standalone programmer is a must have for the retro computing enthusiast. My personal recommendation is the XGecu T48, the successor of the popular TL866 😀

dionb wrote on 2023-10-20, 21:43:

For hotflashing two things are relevant: […]
Show full quote

For hotflashing two things are relevant:

- the EEPROM needs to be supported by Uniflash and the donor board
- the 'donor' board you flash on needs tp support Uniflash

Apart from that, any differences between boards is irrelvant.

If you are getting data verification errors, either the EEPROM is dead (not implausible!) or the flash is not (fully) supported by the board.

Have you been able to flash anything else on this board? Can you flash the board's own BIOS with Uniflash?

Ok I tried flashing the ATMEL EEPROM with the bios of the same board I am using to flash, with AWDFLASH, the same program I use to flash the board.
So I am hotflashing but the same board with the same bios already loaded.
The version of AWDFLASH should definitely recognize the chip.
But I get "unknown flash type" message, even with /F or any other parameter.
Maybe the chip itself is corrupted?

Mamba wrote on 2023-10-21, 08:47:

Ok I tried flashing the ATMEL EEPROM with the bios of the same board I am using to flash, with AWDFLASH, the same program I use […]
Show full quote

dionb wrote on 2023-10-20, 21:43:

For hotflashing two things are relevant: […]
Show full quote

For hotflashing two things are relevant:

- the EEPROM needs to be supported by Uniflash and the donor board
- the 'donor' board you flash on needs tp support Uniflash

Apart from that, any differences between boards is irrelvant.

If you are getting data verification errors, either the EEPROM is dead (not implausible!) or the flash is not (fully) supported by the board.

Have you been able to flash anything else on this board? Can you flash the board's own BIOS with Uniflash?

Ok I tried flashing the ATMEL EEPROM with the bios of the same board I am using to flash, with AWDFLASH, the same program I use to flash the board.
So I am hotflashing but the same board with the same bios already loaded.
The version of AWDFLASH should definitely recognize the chip.
But I get "unknown flash type" message, even with /F or any other parameter.
Maybe the chip itself is corrupted?

Well, AWDFLASH also identifies the chip by device ID (1FD5). From your OP it seems to be in its order, matching the chip make/model in this list.

I'd guess the version of AWDFLASH you're trying doesn't support that chip after all.

Also, check that the original chip in the GA-586DX is a 5V one aswell. There are low voltage versions of 29 series EEPROMs, that I've seen in computer mobos a few times. If you try to access/program a 5V chip with say 3,3V, you might experience symptoms like what you describe.

kaputnik wrote on 2023-10-21, 09:08:

Also, check that the original chip in the GA-586DX is a 5V one aswell. There are low voltage versions of 29 series EEPROMs, that I've seen in computer mobos a few times. If you try to access/program a 5V chip with say 3,3V, you might experience symptoms like what you describe.

Both chips are 5V
The version of awdflash I used was exactly the one with atmel chip support

Definitely there is something wrong with the chip as the taylored awdflash version recognise it as 12v chip, not 5V.
AT29C10A is 5V only.

Guess I need to buy a compatible chip.
Any suggestion to be 100% safe?

Mamba wrote on 2023-10-21, 10:33:

Definitely there is something wrong with the chip as the taylored awdflash version recognise it as 12v chip, not 5V. AT29C10A is […]
Show full quote

Definitely there is something wrong with the chip as the taylored awdflash version recognise it as 12v chip, not 5V.
AT29C10A is 5V only.

Guess I need to buy a compatible chip.
Any suggestion to be 100% safe?

The only way the software could identify the chip is by device ID, and that seems correct in your OP, assumingy read with Uniflash. Then it's up to the software to choose programming parameters.

If it's possible to read and print the device ID with AWDFLASH, you could verify if it's read correctly or not. If it is, you can be almost completely certain the problem is with AWDFLASH's routines, not the chip itself.

As a rule of thumb, those old parallel EEPROMs are more or less immortal. They might lose data over time, but it's very uncommon that they go bad hardware wise.

kaputnik wrote on 2023-10-21, 10:50:

The only way the software could identify the chip is by device ID, and that seems correct in your OP, assumingy read with Unifl […]
Show full quote

Mamba wrote on 2023-10-21, 10:33:

Definitely there is something wrong with the chip as the taylored awdflash version recognise it as 12v chip, not 5V. AT29C10A is […]
Show full quote

Definitely there is something wrong with the chip as the taylored awdflash version recognise it as 12v chip, not 5V.
AT29C10A is 5V only.

Guess I need to buy a compatible chip.
Any suggestion to be 100% safe?

The only way the software could identify the chip is by device ID, and that seems correct in your OP, assumingy read with Uniflash. Then it's up to the software to choose programming parameters.

If it's possible to read and print the device ID with AWDFLASH, you could verify if it's read correctly or not. If it is, you can be almost completely certain the problem is with AWDFLASH's routines, not the chip itself.

As a rule of thumb, those old parallel EEPROMs are more or less immortal. They might lose data over time, but it's very uncommon that they go bad hardware wise.

Thanks

But I am not certain I understood your words.
The chip is not recognised by uniflash at all and by multiple awdflash versions.
The only version that recognise it (but not completely so does not complete the flash process) is 5.2B tailored for ATMEL AT29C010A /5V Flash ROM that is exactly the chip I am using (as said, it is written on it).
So, I am not sure how it can be a problem of the program.
Sorry probably my English is not good.
Can you explain?

Just ordered this
W27E010-70 DIP-32 1Mbit 128k Eeprom
I hope it will be ok

Mamba wrote on 2023-10-21, 11:59:

Thanks […]
Show full quote

kaputnik wrote on 2023-10-21, 10:50:

The only way the software could identify the chip is by device ID, and that seems correct in your OP, assumingy read with Unifl […]
Show full quote

Mamba wrote on 2023-10-21, 10:33:

Definitely there is something wrong with the chip as the taylored awdflash version recognise it as 12v chip, not 5V. AT29C10A is […]
Show full quote

Definitely there is something wrong with the chip as the taylored awdflash version recognise it as 12v chip, not 5V.
AT29C10A is 5V only.

Guess I need to buy a compatible chip.
Any suggestion to be 100% safe?

The only way the software could identify the chip is by device ID, and that seems correct in your OP, assumingy read with Uniflash. Then it's up to the software to choose programming parameters.

If it's possible to read and print the device ID with AWDFLASH, you could verify if it's read correctly or not. If it is, you can be almost completely certain the problem is with AWDFLASH's routines, not the chip itself.

As a rule of thumb, those old parallel EEPROMs are more or less immortal. They might lose data over time, but it's very uncommon that they go bad hardware wise.

Thanks

But I am not certain I understood your words.
The chip is not recognised by uniflash at all and by multiple awdflash versions.
The only version that recognise it (but not completely so does not complete the flash process) is 5.2B tailored for ATMEL AT29C010A /5V Flash ROM that is exactly the chip I am using (as said, it is written on it).
So, I am not sure how it can be a problem of the program.
Sorry probably my English is not good.
Can you explain?

Ah, sorry, was under the impression you got the device ID (1FD5) by reading it from the actual chip with Uniflash. If you can't read it correctly in any way, then yes, your chip (or something else in the HW chain) might be busted.

Mamba wrote on 2023-10-21, 14:04:

Just ordered this
W27E010-70 DIP-32 1Mbit 128k Eeprom
I hope it will be ok

That's a 27 series chip that needs 12V programming voltage, otherwise it should be pin compatible. Seen a few mobos where you can select different programming voltages with jumpers, but in my experience it's not really commonplace. Consult the manual for your board.

kaputnik wrote on 2023-10-22, 08:40:

Ah, sorry, was under the impression you got the device ID (1FD5) by reading it from the actual chip with Uniflash. If you can't […]
Show full quote

Mamba wrote on 2023-10-21, 11:59:

Thanks […]
Show full quote

kaputnik wrote on 2023-10-21, 10:50:

The only way the software could identify the chip is by device ID, and that seems correct in your OP, assumingy read with Uniflash. Then it's up to the software to choose programming parameters.

If it's possible to read and print the device ID with AWDFLASH, you could verify if it's read correctly or not. If it is, you can be almost completely certain the problem is with AWDFLASH's routines, not the chip itself.

As a rule of thumb, those old parallel EEPROMs are more or less immortal. They might lose data over time, but it's very uncommon that they go bad hardware wise.

Thanks

But I am not certain I understood your words.
The chip is not recognised by uniflash at all and by multiple awdflash versions.
The only version that recognise it (but not completely so does not complete the flash process) is 5.2B tailored for ATMEL AT29C010A /5V Flash ROM that is exactly the chip I am using (as said, it is written on it).
So, I am not sure how it can be a problem of the program.
Sorry probably my English is not good.
Can you explain?

Ah, sorry, was under the impression you got the device ID (1FD5) by reading it from the actual chip with Uniflash. If you can't read it correctly in any way, then yes, your chip (or something else in the HW chain) might be busted.

Mamba wrote on 2023-10-21, 14:04:

Just ordered this
W27E010-70 DIP-32 1Mbit 128k Eeprom
I hope it will be ok

That's a 27 series chip that needs 12V programming voltage, otherwise it should be pin compatible. Seen a few mobos where you can select different programming voltages with jumpers, but in my experience it's not really commonplace. Consult the manual for your board.

Damn…
I really do not know if my board support 5V or 12V chips.
There is nothing on the manual about voltages.
And I need to take into consideration that we are talking about two mainboards.
The bricked one (P5A-B) and the hotflash board (GA-586DX).

So basically I am walking in the dark with this… And they say hotflashing is straightforward…

kaputnik wrote on 2023-10-22, 08:40:

Ah, sorry, was under the impression you got the device ID (1FD5) by reading it from the actual chip with Uniflash. If you can't […]
Show full quote

Mamba wrote on 2023-10-21, 11:59:

Thanks […]
Show full quote

kaputnik wrote on 2023-10-21, 10:50:

The only way the software could identify the chip is by device ID, and that seems correct in your OP, assumingy read with Uniflash. Then it's up to the software to choose programming parameters.

If it's possible to read and print the device ID with AWDFLASH, you could verify if it's read correctly or not. If it is, you can be almost completely certain the problem is with AWDFLASH's routines, not the chip itself.

As a rule of thumb, those old parallel EEPROMs are more or less immortal. They might lose data over time, but it's very uncommon that they go bad hardware wise.

Thanks

But I am not certain I understood your words.
The chip is not recognised by uniflash at all and by multiple awdflash versions.
The only version that recognise it (but not completely so does not complete the flash process) is 5.2B tailored for ATMEL AT29C010A /5V Flash ROM that is exactly the chip I am using (as said, it is written on it).
So, I am not sure how it can be a problem of the program.
Sorry probably my English is not good.
Can you explain?

Ah, sorry, was under the impression you got the device ID (1FD5) by reading it from the actual chip with Uniflash. If you can't read it correctly in any way, then yes, your chip (or something else in the HW chain) might be busted.

Mamba wrote on 2023-10-21, 14:04:

Just ordered this
W27E010-70 DIP-32 1Mbit 128k Eeprom
I hope it will be ok

That's a 27 series chip that needs 12V programming voltage, otherwise it should be pin compatible. Seen a few mobos where you can select different programming voltages with jumpers, but in my experience it's not really commonplace. Consult the manual for your board.

That’s it…

I followed your suggestion and got a T48 programmer.
It signaled a bad pin on the chip.
So I assume the chip itself is broken, right?