I can think of two reasons for this - assuming ASUS did this intentionally.

1. To protect certain CPUs from being damaged by any Vcore too far out op specs.
2. To protect the Vcore voltage regulators on the ASUS board.

Since Prescotts had high power consumptions (> 130W), when Vcore (U) is incresed, the respective DC current (I) rises also, the Power (P= I x U) drawn from the onboard voltage regulators would probably excced the maximum values ASUS designed the boards for when Northwood was just released.
This would mean they implemented a power limit by setting a voltage barrier at 1.6V.
I therefore assume that 2. is the case because mainboard vendors don´t care about the lifespans of CPUS so much, but try to avoid damages to their own products.

majestyk wrote on 2021-01-31, 15:22:

I can think of two reasons for this - assuming ASUS did this intentionally. […]
Show full quote

I can think of two reasons for this - assuming ASUS did this intentionally.

1. To protect certain CPUs from being damaged by any Vcore too far out op specs.
2. To protect the Vcore voltage regulators on the ASUS board.

Since Prescotts had high power consumptions (> 130W), when Vcore (U) is incresed, the respective DC current (I) rises also, the Power (P= I x U) drawn from the onboard voltage regulators would probably excced the maximum values ASUS designed the boards for when Northwood was just released.
This would mean they implemented a power limit by setting a voltage barrier at 1.6V.
I therefore assume that 2. is the case because mainboard vendors don´t care about the lifespans of CPUS so much, but try to avoid damages to their own products.

Not a bad idea. Although then it begs the question - why didn't Asus just design a better VRM that could handle higher Vcore settings, like their competitors probably did? Why do they do so much to design such a nice looking BIOS and a feature-packed board, go out of their way to make the board a OCer-friendly board, yet skimp out on the VRM? Something isn't right.

Did they believe a max. overvolting 0f 0.2V would be sufficient for Prescotts?
On the other hand: This was the time, when the "capacitor plague" was raging everywhere. My own ASUS "P4C800 Deluxe" had issues with the caps in the VRM circuits after a few years.
Supposed they knew that there was a lot of trouble ahead because of failing capacitors and they knew that those caps would fail the sooner the more users would be allowed to overvolt (due to more power consumption -> more heat -> shorter cap life). And with Prescotts entering the market these effects were even stronger.
Then the BIOS limitzations could have been an emergency break.

majestyk wrote on 2021-01-31, 16:22:

Did they believe a max. overvolting 0f 0.2V would be sufficient for Prescotts? On the other hand: This was the time, when the "c […]
Show full quote

Did they believe a max. overvolting 0f 0.2V would be sufficient for Prescotts?
On the other hand: This was the time, when the "capacitor plague" was raging everywhere. My ASUS "P4C800 Deluxe" had issues in the caps in the VRM circuits after a few years.
Supposed they knew that there was a lot of trouble ahead because of failing capacitors and they knew that those caps would fail the sooner the more users would be allowed to overvolt (due to mure power cunsuption -> more heat -> shorter cap life). And with Prescotts entering the market these effects were even stronger.
Then the BIOS limitzations could have been an emergency break.

Hmm... unfortunate, in any case.

Then the remaining question is, are there any BIOSes for the P4C800-E Deluxe that ever had uncapped Vcore settings?

Best chance would be a version before they added the Prescott CPU-IDs.

majestyk wrote on 2021-01-31, 16:31:

Best chance would be a version before they added the Prescott CPU-IDs.

Noted.

Also, would it be possible to find the offending hex values that limit Vcore in newer BIOSes using the older BIOSes as a reference and remove that limitation? I'd think a software solution is way easier to back-track on than a hardware fix for the issue.