HTT is present on ALL northwood cores (i don't know anybody else), but it's disabled.
probably pin A31 you mean ... well ... some people have tried to do everything with that pin, without result, so it's not only the pin.
the point is that HTT is enabled by default so the difficulty is to prevent disabling.

if you are volunteer to check all the other pins ... be my guest ... that leaves "only" 477 pins ...

the card for older mobo's is only in case that the cpu does support HTT and the mobo doesn't, so this makes no advantage because many mobo's support it and we want HTT on a non-HTT cpu. (also been investigated) ...

vrforums: is there really no back-up at all??? (or is it something bigger??? is it possible to post the old thread somewhere else?)

the document with P4 pins
http://www.xhw.dk/download/29864305.pdf

[RiStaR]

I believe it's something to do with pin A31 still... somehow with the multiple reboots it caused the CPU to function differently... (I recall someone also concurred that Intel is able to change things with CPUs by doing complicated processes through voltages to certain pins) I must have gotten a fluke in the right sequence and accidentally unlocked it.

The reason why the HTT adapter that powerleap came up with didn't work is probably just this... I'd say prophet is half there with the NVRAM theory as well. It does seem probable to explain why the powerleap adapter doesn't change the status of A31 for the CPU...

nice to hear you again!
have you heared something about prophet? i didn't "see" him since the beginning of the war ... i think he was one of the best people on this thread ...
what do you mean about about the nvram?

i have a thought again ... isn't it possible the HTT is controlled by two pins? the know a31 to disable and enable HTT. and then one more to control that pin, so the function of the pin is to say whether HTT may be used or not ...
it can explain your HTT-case as well ... since HTT is enabled by default ... so "burning" that other pin would be enough ...

edit: is there any more news about the rom inside the cpu?

[RiStaR]

NVRAM = rom in the CPU

As for the 2 CPU reported in DMI reporter or 3dmarks...

if you've got a Gigabyte board... you've got a good chance of that showing up :p

ha, damn, that's thrue ... mixed it up ...

but i found something very strange ...
i looked at the asus-site if there would be a new bios for my mobo (p4s533) ... but there wasn't; but this is not the point.
"officially" (don't know if it does or not in reallity ...) it does not support HTT ... a few months ago i mailed to asus with the question if it would do in the future; the answer was 'no, because of chipset-restricition". i never really believed it ... now i have "some evidence": asus has another mobo in that serie, the p4s533-x ... and guess what ... it support some things mine doesn't: sdram, usb2.0 and HTT.
both of the mobo's are build on the very same chip, sis645dx ...
is there anyone who can explain this?
according to me, it is all in the bios
(something i've said before)

edit: on sis' website, there is some info about this chip ... guess what ... it supports HTT officially, so why doesn't this mobo?
http://www.sis.com/products/chipsets...ium4/645dx.htm

[RiStaR]

"Third party" chipsets often give a lot of doubt with regard to HTT... one of the most problematic is the SiS 655... the A0 and the B0 steppings... the A0 has support in some way but with bugs but the B0 has full support... so if it's buggy do you still consider it HTT support since it does work some of the time? (Subjective) I'll say... either way... for HTT to work, I'm sure most of you should already know by now that it needs 3 main things.

1. CPU must support HTT officially
2. BIOS must support HTT
3. Chipset must support HTT

Obviously our problems are linked to no.1 which is the hardest to crack...

[QUOTE]Originally posted by RiStaR
[B]I believe it's something to do with pin A31 still... somehow with the multiple reboots it caused the CPU to function differently...

Ya, that makes sence, I once had a CB-radio, at that time, here in Denmark only 23 channels was allowed, so it was modifyed by the distributor, but if I turned it on and off very fast a few times, it had the original 40 channels, I guess the many resets confused the logic in it, kinda what happend to your CPU.

Originally posted by Granny
since HTT is enabled by default ... so "burning" that other pin would be enough ...

So removing that other pin could also be an option ... interesting ... and doable

Originally posted by Crispy
So removing that other pin could also be an option ... interesting ... and doable

No it's not, before the crash people had suggested trying to isolate A31, and to no avail. Removing it would be no different, except you won't be able to get it back and you could end up killin your CPU (we stil don't know EXACTLY what A31 does). I wouldn't recommend removing that pin. unless ya got a few bucks you want to buy a new proc with.

that's why i'm talking about another pin ... a controller pin or so. maybe (probably?) not only for HTT, but much more. could it be that one pin directs others?
i don't which pin it would be or so.

something else: i discovered that there are two pins to control the FSB (in the 533Mhz-cpu): it allows four combinations, one for 100Mhz (400), another one for 133Mhz (533), and then two combinations which are "reserved"; is it possible that these combinations are the 100&133Mhz with HTT?
just a thought ...
i was wondering what it could be, it was the HTT-option or the 66&200Mhz-busspeed that were possible to me ...

OK,

Just to clarify for Granny, as I understand where he is coming from with his reply to my earlier post.

From what I have read, cards to enable HTT on non-HTT Boards puts a signal on pin A31 that is missing on the non-HTT motherboards.
So, I was thinking that the A31 pin was disconected from the circuit, which stopped the signal going through the circuit and that stopped hyperthreading.

Ristar: So your processor didnt come with HTT enabled, it was enabled by accident?
If thats so, than my idea about intel burning jumpers on the die is wrong.

Mayby a string is stored in the NVRam and by continually resetting i the CPU, you corrupted the NVRam forcing a rewrite and the cleared NVRam didnt disable HTT.

EDIT: Looking at the data, and considering you said you reset it continually, mayby holding RESET# high for more than 10ms resets the chip, as the data says "RESET# must not be kept asserted for more than 10 ms while PWRGOOD is asserted."
Or holding RESET high could kill the chip.

NOTE: By high I mean asserted. By data I mean datasheet.

Cyas

[RiStaR]

Originally posted by Alaskan DJ
OK,

Just to clarify for Granny, as I understand where he is coming from with his reply to my earlier post.

From what I have read, cards to enable HTT on non-HTT Boards puts a signal on pin A31 that is missing on the non-HTT motherboards.
So, I was thinking that the A31 pin was disconected from the circuit, which stopped the signal going through the circuit and that stopped hyperthreading.

Ristar: So your processor didnt come with HTT enabled, it was enabled by accident?
If thats so, than my idea about intel burning jumpers on the die is wrong.

Mayby a string is stored in the NVRam and by continually resetting i the CPU, you corrupted the NVRam forcing a rewrite and the cleared NVRam didnt disable HTT.

Cyas

The CPU did not come with HTT... and it still does not run with HTT... It was turned on by accident

So HTT is no longer enabled? Was it only enabled for one power on, or multiple?


EDIT from above (incase you dont read it): Looking at the data, and considering you said you reset it continually, mayby holding RESET# high for more than 10ms resets the chip, as the data says "RESET# must not be kept asserted for more than 10 ms while PWRGOOD is asserted."
Or holding RESET high could kill the chip.
By high I mean asserted. By data I mean datasheet.

[RiStaR]

Originally posted by Alaskan DJ
So HTT is no longer enabled? Was it only enabled for one power on, or multiple?


EDIT from above (incase you dont read it): Looking at the data, and considering you said you reset it continually, mayby holding RESET# high for more than 10ms resets the chip, as the data says "RESET# must not be kept asserted for more than 10 ms while PWRGOOD is asserted."
Or holding RESET high could kill the chip.
By high I mean asserted. By data I mean datasheet.

It worked for two soft reboots before i shut it down to investigate