Help needed...unlocking Higher m on 2100

[Camelot One]

I read everything I can find on unlocking the XP's, all seem to be focused on allowing 12.5 and below. My FSB is limited to 145 by the board, so a lower multiplier wouldn't do me any good. Can anyone tell me how to unlock the higher multipliers on a 2100? 13.5 and 14 in particular. I don't think the chip would do much more than 14 without a MB vmod.

[candjac]

Originally posted by Camelot One
I read everything I can find on unlocking the XP's, all seem to be focused on allowing 12.5 and below. My FSB is limited to 145 by the board, so a lower multiplier wouldn't do me any good. Can anyone tell me how to unlock the higher multipliers on a 2100? 13.5 and 14 in particular. I don't think the chip would do much more than 14 without a MB vmod.

There are several issues here, the first is the issue of closing the L1 bridges which allows circuits in the mobo/bios to over-ride the 5 default Multiplier signal circuits on the CPU.

The second issue seems to be that "some" mobo/bios are not able to change the HI setting of the 8X Bit Value circuit to LO, so users are unable to reset to the 5X thru 12.5X range. The workaround is to change that setting manually...but that is not what you want.

So to unlock, close the L1 bridges and see if your mobo/bios is able to reset default 13X to 13.5X/14X. Changing the 8X Bit Value is not involved so possibly mobo/bios might be able to reset the other Bit Values as needed. But IF "due to some mobo/bios incompatibility" this did/does not work, THEN there is only one other option...make manual changes to the L3/L4 bridge pairs which set the lower Bit Values.

You can find detailed info, diagrams, circuits, Algorithms, Table of bridge settings, about these settings at
http://www.beachlink.com/candjac/index.htm link to Palomino decoding article. We would also suggest leaving the L1 bridges "open" if manual mods are required, for the simple reason that if the mobo/bios are not able to affect changes then closed L1 bridges are of no use, and it just might be possible that closed L1 bridges might in some way cause interference with the manual mods...it has happened at least once on a Duron/MSI mobo.
John C.

[Camelot One]

Thank you

[Camelot One]

I found an article on unlocking the 2100, saying that it is different from the previous chips. This I know, but the article says that closing the L1 bridges without first sealing the lazer cuts with superglue would result in a default 14x multiplier. It says this as a "what not to do", because you will only get the 14x, not everything 12.5 and below. So I am not sure how much faith to put in it. Anyone know anything about it? If its true, and grounding the bridges to the exposed copper is whats needed, wouldn't all of the bridges then be grounding to the same piece of copper, and thus there would be no need to connect each induvidual bridge, just connect all of them with 1 glob of conductive ink?

[candjac]

Originally posted by Camelot One
I found an article on unlocking the 2100, saying that it is different from the previous chips. This I know, but the article says that closing the L1 bridges without first sealing the lazer cuts with superglue would result in a default 14x multiplier. It says this as a "what not to do", because you will only get the 14x, not everything 12.5 and below. So I am not sure how much faith to put in it. Anyone know anything about it? If its true, and grounding the bridges to the exposed copper is whats needed, wouldn't all of the bridges then be grounding to the same piece of copper, and thus there would be no need to connect each induvidual bridge, just connect all of them with 1 glob of conductive ink?

Not responding from experience but from Table of Settings in Mobile Athlon4/Palomino datasheet...and Multiplier Algorithms thus derived. All 5 signal busses grounded/LO should get 11X, not 14X. A 14X multiplier is supposed to result from Sum of HIs = 11X = 1X + 2X + 8X Bit Value signal circuits set HI. If the article is not precisely specific as to having actually tested circuits with multimeter we'd discount it. Did they actually get 14X with all L1s "proven" to be grounded??
John C.

[OPPAINTER]

All I can tell you is I managed to get it on a KX7 and a Duron 1300.
The L1 bridges are slightly cut on these chips so I went over them with a pencil and this is what I got.

OPP

[candjac]

Originally posted by OPPAINTER
All I can tell you is I managed to get it on a KX7 and a Duron 1300.
The L1 bridges are slightly cut on these chips so I went over them with a pencil and this is what I got.

OPP

This is getting complicated, thought we were talking about an XP 2100 which needs to insulate the trenches else grounds...now we have a Duron 1300 which we belive is ceramic, L1s closed/?reclosed with a pencil?, so no chance of sub-surface grounds??
Apples and oranges...no explanation between the two chips.

But can't fight wcpuid, so only Duron explanation is that the "1300" Duron has its Multiplier set for 5X which remaps to 14X. We know there is a 1400 Tbird like that...is there a Duron??? Or did the mobo/bios do it??
John C.

[OPPAINTER]

And this is when I set the multiplier to 10 by the way. :D

OPP

[candjac]

Originally posted by OPPAINTER
And this is when I set the multiplier to 10 by the way. :D
OPP

Ha! Now we feel bad and good at the same time.;-)

Bad because this 14X Duron situation is inexplicable.

Good because we can surely blame this on your mobo/bios, and in no way changes the "normal" Multiplier decoding which has worked well to date. And it also doesn't cause us to change earlier posts re the XP 2100.
John C.

[Camelot One]

Well I was all set to try just connecting the L1 bridges, but found I am fresh out of AS3. So it will be a few days before I can try.

[OPPAINTER]

Well the thing about the 1300 chip is that I think in order to unlock it you need to do the pin mod and close the L1s. I had a XP2100 chip that all I did to it was close the L1s on it and would give me 2 multipliers, 13 and 13.5. So when you just close the L1s without the pin mod or L3s you end up getting funky readings. I think the mobo and the chip worked together to get the higher multipliers, I don't just think the mobo screwed up all on it's own.

OPP

[candjac]

Oppainter,
Not to beat this thing to death, but believe the Duron/Tbird chips only have 4 bits, 4 Multiplier circuits. So there is no pair of L10 bridges to set the 8X Bit Value as on the XP Palominos. Therefore the 8X pin wire trick ought not to do anything. Notice that we say ought.;-)

Can you verify that the 1300 Duron indeed has no L10 bridges as we suspect...we know all the early Duron/Tbirds thru 12.5X only had the 4 L3/L4 bridge pairs, no L10s.
John C.
Edit:- Just looked at a 1400/100 Tbird pic in out files. No L10 bridges, Sum of HIs = only the 2X Bit Value, which "normally" remaps to 5X, but which we all know is further remapped internally on this 1400 Tbird to 14X. Shouldn't we expect similar from Duron??

[OPPAINTER]

Sorry I no longer have the chip. I may be able to find a picture on the web somewhere.

OPP

[candjac]

See edit above
John C.

[OPPAINTER]

Heres one

[candjac]

OK, seeing is believing, the Duron does indeed have a pair of L10 bridges, unlike our pic of a Tbird/1400. But now we might also be able to explain the 14X on the Duron.

to get 13X from XPs 5 bit code (and ?your Duron?) the Sum of HIs = 3X Less = 10X = only 2X and 8X Bit Values set HI. Now if you do the wire trick on the 8X bit ( which we thought was non-existant on Duron 1300) then only the 2X Bit Value remains HI which gets remapped to 5X and 14X same as in 1400Tbird... edited post above.

Duron 1300 at 14X mystery solved??
John C.
Edit:- We can see in the duron 1300 pic that the L10s are open/closed L to R which is the setting for the 8X Bit Value = HI...as we assumed above.

[ns_ripper]

Hi guys, I bought a duron 1300 yesterday and would like to unlock it. So I could be alble to solve the mistery for you guy's If you could help me with the unlocking part. I already tried with connecting the l1's. it gave as result that I had to turn on and off the pc several times before it posted and stuff like that :s

So I guess I got something more to do to unlock this baby. :(