wiretap wrote on 2021-02-04, 18:30:

@pshipkov -- that may be the case with incompatibility for some odd reason. The chip does get warm, but no post. As far as the little board, it is pretty sturdy. I pulled jumpers off and put them on with one hand, just holding my thumb on the motherboard connector side. It does lean a little bit, but it doesn't fall of the board and has a good amount of friction-fit.

@Mamba -- I didn't test that, since this mod adds the 4th row of jumpers which specifically takes advantage of the 9250 clock gen. The 4th row would not map to anything on the old one.

Yes I can understand.
So it is not usable even if no pin has been jumpered on the 4th row?

I don't really know what the point would be. The purpose is for 133MHz access via that 4th row. Otherwise just use the existing 3 rows on the motherboard.

wiretap wrote on 2021-02-04, 20:27:

I don't really know what the point would be. The purpose is for 133MHz access via that 4th row. Otherwise just use the existing 3 rows on the motherboard.

Sorry,
I have to elaborate.
Can't wait to buy one from you, so the test on the rest of the rows could be enough for me...

How did you replace that chip? Hot air gun?

I used a hot air pencil.

Update: I've received feedback from 3 people who built this little adapter board and say that it works on their board, although one person says they have issues with AGP not working at 133MHz FSB. I suspect this is due to the inherent AGP frequency overclock, so maybe a different graphics card would work that is more tolerant to AGP overclocking. (PCI graphics work fine for them)

It is also recommended to perform the motherboard modifications listed in the Tipperlinne link in the first post -- for overclock stability reasons to control voltage better.

Please share your experiences of what hardware you're using if you perform this mod, so other people can see what works. Also list your motherboard rev and spec. (D02, D04, etc)

wiretap wrote on 2021-02-02, 03:47:

I'll be testing once I get the ICS 9250CF-08's and get that soldered in. […]
Show full quote

I'll be testing once I get the ICS 9250CF-08's and get that soldered in.

So this little thing just plugs into the existing jumpers, then transposes them and adds the 4th needed jumper for 133MHz FSB. The R99 pad obviously connects to the motherboard's R99 resistor with a jumper wire.

Would this work with an XG-DLS board? There was a thread a while back that detailed a modification similar to this or same as this, clock gen is the ICS9250BF-08

Asus XG-DLS POST issues [solved, now working at 133MHz FSB!]

wiretap wrote on 2021-03-16, 11:27:

Please share your experiences of what hardware you're using if you perform this mod, so other people can see what works. Also list your motherboard rev and spec. (D02, D04, etc)

Some of my experience with p2b-d 1.06 d03 among other things in a big boring post. In short - works well at 1575/150/37.5. There are some videos at the end showing the stuff in action.

BTW, not sure if there is d04 stepping, d03 seems to be the last one.

m1919 wrote on 2021-05-02, 04:46:

wiretap wrote on 2021-02-02, 03:47:

I'll be testing once I get the ICS 9250CF-08's and get that soldered in. […]
Show full quote

I'll be testing once I get the ICS 9250CF-08's and get that soldered in.

So this little thing just plugs into the existing jumpers, then transposes them and adds the 4th needed jumper for 133MHz FSB. The R99 pad obviously connects to the motherboard's R99 resistor with a jumper wire.

Would this work with an XG-DLS board? There was a thread a while back that detailed a modification similar to this or same as this, clock gen is the ICS9250BF-08

Asus XG-DLS POST issues [solved, now working at 133MHz FSB!]

Those boards are so rare, i can't imagine the risk / reward is there.

I spent almost a decade getting my xg-dls. If anything happened to it that was avoidable, I'd never forgive myself.

luckybob wrote on 2021-05-03, 11:16:

m1919 wrote on 2021-05-02, 04:46:

wiretap wrote on 2021-02-02, 03:47:

I'll be testing once I get the ICS 9250CF-08's and get that soldered in. […]
Show full quote

I'll be testing once I get the ICS 9250CF-08's and get that soldered in.

So this little thing just plugs into the existing jumpers, then transposes them and adds the 4th needed jumper for 133MHz FSB. The R99 pad obviously connects to the motherboard's R99 resistor with a jumper wire.

Would this work with an XG-DLS board? There was a thread a while back that detailed a modification similar to this or same as this, clock gen is the ICS9250BF-08

Asus XG-DLS POST issues [solved, now working at 133MHz FSB!]

Those boards are so rare, i can't imagine the risk / reward is there.

I spent almost a decade getting my xg-dls. If anything happened to it that was avoidable, I'd never forgive myself.

I have two of them, would be fun seeing how those 900 Mhz 2MB cache Xeons perform with 133 MHz FSB

XG-DLS will work at 133 MHz FSB with only one CPU sadly, could not get it to POST with two CPU's at 133 MHz no matter what. Either of them ran happily at 133 Mhz all day long. And these were 700/2M ones that ran perfectly at 933 MHz.

Paadam wrote on 2021-05-03, 17:34:

XG-DLS will work at 133 MHz FSB with only one CPU sadly, could not get it to POST with two CPU's at 133 MHz no matter what. Either of them ran happily at 133 Mhz all day long. And these were 700/2M ones that ran perfectly at 933 MHz.

Hmmm. I have a Freeway board I might as well pull out for this then, been a while since it last ran, 🤣.

Is this an appropriate resistor and header pins to use for this PCB:
https://www.digikey.ca/en/products/detail/sam … 8J103CS/3965393
https://www.digikey.ca/en/products/detail/sam … -07-G-T/2685893

Yep, those will work. You also need the 3x3 2.54mm female pin socket as well.

You can always use single row header strips and cut them too. That's what I did.