[Info-vax] Problem with old firmware on zx2000 system board

[abrsvc]

On Friday, March 18, 2022 at 9:08:58 AM UTC-4, valer... at gmail.com wrote:
> Hello! Here is a brief description of what happened during the reanimation of my old zx2000 machine. Really need advice on what to do next )) 
> 
> I have 3 zx2000 system boards at hand: 
> - the original one, worked without problems for many years and eventually died (it still can run EFI and fails with machine checks when loading OS; HP diagnostic utility finds a number of damages); 
> - 2 ebayed replacements, both are of older hardware revisions and with outdated firmware. Every attempt to boot ends with a swf hang. I tried several single core Itanium chips, including a 1.0Ghz Deerfield (the original system was equipped with it), 1.4Ghz and 1.5Ghz Madisons (not specified in the zx2000 docs however the original machine was OK with both), and finally an old 900Mhz McKinley - however, with minor differences the result is the same: ia64 firmware hangs and after the timeout BMC reports a firmware error. 
> 
> I assume that the probability of having 2 physically damaged replacements from different sellers is close to 0 and what I see is really a firmware hanging. What could have caused this? Besides a newer Itanium chip, it can be a scsi (excluded trying different versions of Lsi adapter and even by unplugging it at all), VGA (not fully excluded bacause I have a limited number of AGP VGAs at hand, but the old system is OK even with newer Radeons), and a priori anything else (it seems paranoid, but I have changed the pw supply, fans and the front panel card). 
> 
> I guess there's something I don't know. Why I was so successful with the original system board and everything ends in failure with new ones? 
> 
> There is a surgical solution: solder off the AMI chip from the old board and attach it to the new one. A softer version of that is to use the ISP interface, of course, undocumented. Are there any other ways? 
> 
> I would be grateful for any advice )) 
> 
> Regards, 
> Valery

This may sound a little strange, but I would look at the power supply.  Check to see that the voltages are what they should be.  If you have (or know someone that has) an oscilloscope, you should check the power rails.  Excessive noise (AC on the DC line) can cause all kinds of strange behavior.

Not sure where you are located, but I have a working larger version (ZX6000) available for short money.  This unit is located in Ma. USA

Dan