is there a way of pin modding the cpu ......
so we can make the mobo /bios belive that it is a 533 fsb celeron,
maybe that will unlock sumat else with more fsb .. ?
is there a way of pin modding the cpu ......
so we can make the mobo /bios belive that it is a 533 fsb celeron,
maybe that will unlock sumat else with more fsb .. ?
@ ibby :
Yes, doing the "wire mod on socket" between two pin holes, option : forcing MoBo to think that has inserted 133Fsb Cpu by using BSEL (bus_select, BCLK) pin connecting it to nearest Vss (GND) pin!: on Asus_CT-479 adapter by grounding/shorting a Bsel(0) - C16 pin (forcing it on L(ow)!) to some nearest Vss pin: B16; C15; D15,17
something like this ...
http://www.picturebio.com/images/laptoppics/image7.jpg
http://www.picturebio.com/images/laptoppics/image6.jpg
... and Cpu will boot already 33% OC-ed ...
@ lktuio : I also do not believe that the problem is Cpu, but rather something else /we don´t know for sure what yet/! Try disabling all stuff (other peripherrals) on your MoBo Bios except FD & IDE controller, find some old IDE disk & old PCI video card & try OC more & try if having any success booting higher than before, this "mine"win already made micro setup: http://users.volja.net/image/Files/ResQsys.htm ... report us the result ...
I've got that pinmod thing in yet I don't seem to notice any difference. Also it doesn't boot at the 133FSB straight off. I've tried setting the jumper to say its a 133FSB cpu with no luck. I've refitted the wire twice with no luck. Any ideas?
Surely something like this, though, could be done with a bios mod..
@Spajky,
There must be sumat holding us back.
there is no way all these cpus have a therotical max fsb at 180 (for example).
IT must be some thing else.
I know with the Zeons 1.6 LV one, if u did not do the pin mod then u would only get 2.4 gig ish max.
But once the pin mod had been done .... u could attain more 3.2 gig atleast.
I think the issue maybe adapter releated.
Has any one got a 479 dedicated mobo to test the cpu in ?
There could be:Originally Posted by ibby
- peripherrals clock problems (no propper dividers PCI/otherStuff kicking_in in Bios)
- software block in bios (cpu microcode) forced by Intel to Bios manufacturers
- hardware block on internal PLL (crippled by Intel design just like for powersaving)
- all the above :-(
IMHO, CT-479 is not a problem (works perfectly with FSB over 200MHz with P-M), but could also be there some pin preventing C_M more than 80% higher basic CPU clock ... There are a bunch of pins mentioned in datasheet, that control a lot of stuff! I have to compare those to a P-M datasheet , maybe there is some difference somewhere & so a place for another mod proposal ...
all ur suggestions are correct.
are all Celeron M processors multi locked ?
if they are is there away to "lower the multi", via the
socket ?
Does any one know about BIOS and see if they can change / make the cpus a Unknown cpu ?
Hae a look at the data sheet and keep us informed, even tho iam not using a celeron,
iam willing to go buy 1 next month, for trial and error.
well the reason they open the multi is so that mobos can downclock them to save power (i.e. speedstep) - if celerons have speedstep then they are downwards unlocked...
no, they are locked!
Some other facts to add to my previous modding ideas:
-All pin listings for CM & PM are same (checked!) except! older /april-2004/ PM7x5 one 30218905.pdf datasheet has a difference: C14 is on all others Bsel(1) while there RSVD & C16 (Bsel-0) is there TEST3 [rsvd + test unconnected, the last has to have option of pull-down by 1k resistor] WHAT the F*CK ??? Same datasheet number dated January-2005 has no this exeption!!! Something weird here ...
- BSEL pins (C14,16) have L-ow state @ 0V & H-igh state @ 0,7V /=same almost as GTLREF/...
Bsel(1) has same status as for CM/PM also for P4 (so no problem here), while the other pin Bsel(0) has a reverse logic status on P4 than on Dothan; so there on CT479 has to be at least one transistor for that for reversing logic state to a P4 MoBo & means same time that BSEL pins are connected on Asus adapter; the other one is just directly going thru to P4 MoBo !!!
- ITP_CLK(1:0) 2 Fsb clock pins (A15,16)[output signals] are not connected IF the MoBo has "debug port interposer" implemented; if not so are connected to a MoBo/adapter - this should be checked with Ohm-Meter. IMHO that term means (not sure), that MoBo has hardware self-Post check implemented in Bios (majority of them now have this), so those 2 pins should be not connected anywhere IMO on the CT adapter ...
- BSEL pins can be or not be connected!!! If the Cpu selects FSB (& Cpu clock), than are connected & sent to MoBo clock generator chip; if they are not, than means that MoBo itself independently controls the Fsb /& clock to a CPU/ : this IS the case with MoBo based on older i852 & i855 chipsets (and NOT the case with i9xx chipset based ones) !!!
This last could explain, why even a malfunctions Banias Celly /if not really lucky example/ could do 180Fsb_! on his Aopen i855 board !!!, while on his P4 & CT adapter did not wanna go higher than 166Fsb, because those Bsel pins are sent/connected via adapter to a MoBo & something is preventing going higher ...
Also as I found till now (not sure but anyway), also PM7x5 (100Fsb/400qp) have same "Fsb problem" as Cellys do regardless they have down unlocked multipliers ... so there must be something involved with those BSEL pins or Bios detecting Fsb IMHO ...
Since Bsel pins look like are not needed to be connected (see note on i855 chipset MoBo), there IMHO could be left unconnected on P4 ! socket (isolating there AD5&6 pins, setting them on H-igh so; datasheet states this is a "reserved" option for P4). To achieve same is another option: isolating just on P4 socket pin AD5 & on CT-479 socket pin by grounding/shorting a Bsel(0) - C16 pin (forcing it on L!) to some nearest Vss pin: B16; C15; D15,17. The result would be IMHO the same! The P4 MoBo would go IMO into a test mode (other possibilities enabled, maybe even-not really believable, unlocking the Celeron_M multiplier ...)
But isolating pins properly & REVERSABLE way in the socket in a delicate job; needs some practice opening socket cover with pin holes to get access to pin contacts to bent them a little & widening pin hole on the cover; best is (like I did) practicing before on an old socket of some death old MoBo before few times with some dead CPU too. All mine Tuallie slotket mods were done reversable like this: [ http://freeweb.siol.net/jerman55/HP/...TualModsoc.JPG ]
Momently, I have no further ideas ....
can any 1 try this mod ? PLEASE ......
Spajky @ could u post a diagram on which pins would need to be modded ?
ibby
@ ibby :
...you have all info for DL on Intel's Literature Center in datasheets for Dothan & P4 : http://appzone.intel.com/literature/index.asp for pin assignments ...
@ wwwww :
have you tried to lower the multiplier & raise more Fsb on your P_M?
Watercooled Celeron.....
i own a Cooler Master aquagate system
Recently bought a Celeron M 370 (c0) +asus CT-479
Running them on an Asus P4GD1 (Bios 1005...)
Should receive them tomorrow.....(and tell charlie...
Can i expect 2.6 Ghz (15x173...) ??
Trying to put the waterblock on Ce M die......
Do you think it can work.....
The link shows the specs on the aquagate.....Should work so....
http://www.extremeoverclocking.com/r...ALC-U01_3.html
Man is a wolf to Man. Then i wanna be the Tex Avery wolf......at least to enjoy the kiss of a woman.....
I have the same Celeron M 370 (c0) and with stock cooling i can get 2.4Ghz stable (160x15) at 1.425v.i own a Cooler Master aquagate system
Recently bought a Celeron M 370 (c0) +asus CT-479
Running them on an Asus P4GD1 (Bios 1005...)
Should receive them tomorrow.....(and tell charlie...
Can i expect 2.6 Ghz (15x173...) ??
Trying to put the waterblock on Ce M die......
Do you think it can work.....
The link shows the specs on the aquagate.....Should work so....
http://www.extremeoverclocking.com/r...ALC-U01_3.html
No matter what vcore i put it does not stay stable at more than 165x15 (on windows)
You should have bought the 350J (with 13x multi) and maybe you could boot it at 200x13 with the CoolerMaster Aquagate.
But i don't think the cooling is the problem, my 370 is only 55º at load and 37 at idle with stock cooling.
When you install it please tell us the temperature in idle and full just to check out the eficiency of the CoolerMaster Aquagate.
I think this might be the nail on the head, I read over the older Pentium M 90nm specs and it states a typ. 1.8V for VccA. The newer P-M 533FSB CPU's run 1.5V typ. for VccA.- modding the adapter: rising VccA [measuring it first, PLL supply voltage, max.120mA - see those my previouly linked posts for PM-dothan discussion] to 1,8V instead of 1,5V, since C_M supports both voltages. This (overvolting) would help internal Cpu generated_clocks to work better/at higher clocks, same as does overvolting pure core!
This last may be needed, because I found 2 datasheets for P_M, older ones states VccA 1,8V, newer one 1,5V & Asus IMHO took for all CPUs lower one not thinking much about extreme OC-ing!
Damn, if I would be so good in money earnings like I am with modding ideas, I´d be a millionaire ..
do you happen to know where to measure the VccA voltage? I could probably find out myself but if you know it'd save me the homework :p
On some of those VccA pins & trying to determine on the adapter from where it comes folowing the traces on PCB; I do not know where that voltage is sourced; I asked before for a largest close/makro detailed photo that somebody could make me of the adapter to study it better; haven´t got it yet ...
I can try to give you some detailed shots of my adapter, but my camera sucks very badly and very old (640x480 and it uses floppies!)
edit: Nevermind, tried it and I can't make out any details at all. Asus used a very dark PCB for this and it's difficult to see anything with the naked eye.
Last edited by Pongi; 07-28-2005 at 04:18 PM.
I promise to do some very high res over the weekend for you.
But they will be very big files
will be Ok at 2048x1536/16bit Jpg best quality saved (if I need more, that quality is more than enough if is (upper side only) adapter almost on whole pic surface to enlarge it later at will on my monitor ...
To Mike@Portugal
Well....i was hopping for 2.6 stable....
Haven't received it yet...Midday maybe
trying to make the water block fit and stay on the Ce M die....
screws are too short i guess...
Talking about the aquagate.....
Previous CPU=Prescott3.2@4.0 Ghz Vcore 1.6
Room temp 25°C/idle 37°C/Load 46°C (sensor directly on the waterblock...shows 55°C in Asus Probe
Room temp 17°C/idle 30°C/Load 39-40°C.....
I modified it..In the stock version, the exhaust gaz (warm) directly go in the tower....so i've added a fan right there (no fan in stock version...) and exhausts gaz go right through the Power case.....(Tagan 380W...
Also, 've cutted the front entrance for more fresh air and changed the front fan (more efficiency/less noise....)
I think that's all
Thanks for infos about your rig.....great result fron 1.5Ghz anyway.....
Will tell you when i'll have received it....can't wait any more....
Man is a wolf to Man. Then i wanna be the Tex Avery wolf......at least to enjoy the kiss of a woman.....
I can host them for you guys.
Send us a copy to ibby @ cmia.co.uk
(no spaces) ......
and make the title ct479.
oops
Last edited by Jericho; 07-31-2005 at 06:00 AM.
MSG me if link required.
If the server gets hammered i will remove them !
Last edited by Jericho; 08-01-2005 at 02:46 PM.
Thanks ..
then they dont have speedstep
yeah basically its like this
17x157
16x166
15x173
14x176
10x181
same as before the mod...
guys i know its a little off beat
but i have a pentim M banjis 1.5 and want to put 1.7v
into it.
whch pins do i mod, and any particulat wires i should use.
Thanks.
Ibby
Posting Permissions
Reply With Quote
Bookmarks