• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

1333 bsel mod for 800 FSB chips?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
benbaked

benbaked

Folding/SETI/Rosetta Team Member
​
Joined
Oct 20, 2005
Location
WA
Hi:

I did the 1066 BSEL mod on my Conroe-L Celeron last night with my Biostar GF7050V-M7 board and it booted right up at 3.2 GHz...so I wanted to try it one step further and push the thing to 4 GHz with the 1333 BSEL mod.

Checking online I found this thread over at XS with a diagram of how to do it. However, this was intended for an E4300, I'm unsure if this is the correct mod for a Conroe-L.

Regardless, I tried it, and the system would not boot. Perhaps 4 GHz is too much for my Celeron without a vcore bump, which my board won't do beyond very small amounts (106% of normal max :()..so I'll probably have to try for one of those vcore pad mods too.

Can anyone confirm if this is the correct BSEL mod to increase the FSB of a Celeron to 1333?

Thanks in advance. :thup:
 
3.2ghz seems like a no brainer but after that, its not easy/or is luck on the cpu.
 
vss=ground
vcc=hot,voltage of the cpu as set by the VID.
vtt, cant find any info on that,need to do more reading. i assume its a another HOT voltage point on the cpu for the NB/FSB.

*Edit*
ok so from that doc you posted NB4, VTT= miscellaneous voltage supply. also expanded on VCC per that doc.
 
Evilsizer said:
vss=ground
vcc=hot
vtt, cant find any info on that,need to do more reading. i assume its a another HOT voltage point on the cpu for the NB/FSB.
Click to expand...

I would imagine its the same as the CPU VTT setting in your bios. Certain CPUs need it upped (quads for instance)

VTT usually satands for some kind of voltage termination. Beyond that you got me.
 
Neur0mancer said:
I would imagine its the same as the CPU VTT setting in your bios. Certain CPUs need it upped (quads for instance)

VTT usually satands for some kind of voltage termination. Beyond that you got me.
Click to expand...

as it makes sense, see table 25, 9 of 9,on page 74. the description of the other VTT pins, gives it what we both talked about. cpuvtt also helps dualcores as well, its the CPUGTL that really helps quads over dual cores. on CPUGTL see the description for it on page 69,table 25, 4 of 9
 
btw... OP what conroe-L celly do you have? i dont think any of them have a 12x multi?


as per the correct way to do the 1333 fsb mod, there are many ways to skin the cat...

You just have to understad what your doing... for BSEL 0 1 2 respectivly the default values for your 200 fsb cpu is L H L ... to get 333 fsb you need to make the pins appear to be L L H (low low high, or 0 0 1) to the motherboard... so you need to make BSEL 1 go from H to L and BSEL2 from L to H

You can make a H pin L by connecting the H to an L pin (or ground) and you can make a L pin appear to be H by isolating it (putting a piece of tape over it (used this process on my E4300 to pad mod for lower voltage, simply put a piece of tape over the Low pad and it made it aappear to be a High pad to the mobo, thus feeding it in my case a lower voltage)

So how i would do the 333 FSB mod would be connect BSEL 2 to 1 with conductive ink to make BSEL 1 a low pin, then place a piece of tape over BSEL 2 to make it appear to be a high pin. (take a look at the pic i attached)

Same goes for voltage, but guidance on pad modding for more voltage cant be done untill you give us the default VID of ur chip.
 

Attachments

  • s775-p4-prescott-3_6-bottom.jpg
    s775-p4-prescott-3_6-bottom.jpg
    48 KB · Views: 6,213
Last edited:
nd4spdbh2 said:
btw... OP what conroe-L celly do you have? i dont think any of them have a 12x multi?
Click to expand...

according to his sig its a Celeron-L 460, though it just says C460... following the other models listed, 420=8x,430=9x,440=10, leaving 450=11x and 460=12x, looks like someone got a ES celeron-L cpu. :eek:
 
Hi, sorry, I've been farting with trying to keep my quad cool. I should have some time to fart with this Celly today, getting windoze installed to get the VID. :) Yeah it's an extra special celly. :sn:

So if I understand correctly, you're saying to draw a line between BSEL1 and BSEL2, and then cover BSEL2, no other traces or covered pins, and that should fool it into 1333 FSB? I'll try it out.

BTW, thanks for that intel document link, very cool. :thup:
 
benbaked said:
BTW, thanks for that intel document link, very cool. :thup:
Click to expand...
only true ocing nerds read intel whitepapers! :)

yes ben what NB4 said is correct.... whats it is doing is routing the H signal to bsel2 from bsel1,then by puting the tape on bsel1 its making the board think its a L signal.
 
Evilsizer said:
only true ocing nerds read intel whitepapers! :)

yes ben what NB4 said is correct.... whats it is doing is routing the H signal to bsel2 from bsel1,then by puting the tape on bsel1 its making the board think its a L signal.
Click to expand...

haha ya white papers do come in handy for this stuff...

Nextly, you got me confused when you said BEN (which is my real name) then said my ocf nick... which is nD4spdbh btw :beer:

and Lastly you got the theory right on the bsel modding but the numbers partly wrong.... like you said and i said earlier you connect BSEL 1 and 2 so you essientally ground out BSEL1 and make it appear LOW to the mobo. then you place a piece of tape over BSEL 2 making it appear high. So yes benbaked you got the process correct. what you might try to do (if you can do it) is lower the multi down to like 8 or 9 then do the 333fsb pad mod to see if it works without needing to do the voltage pad mod... once you get the fsb pad mod working then you can go onto overvolting the sucker!

lol rough day evilizer?
 
ND4,
long day plus tring to help in 2 threads about bsel mods, getting confused QUICK! hopefully i have the bsel mod in the other thread straight now..
 
Last edited:
Good news guys! First of all I got Windows installed, however CoreTemp wouldn't give me the VID. Checking CPU-Z with default settings showed 1.25v. Looking at the list of voltage mods at this website the closest I could find is 1.2875v; so I went gung-ho and did the 1.6v voltage mod in addition to nd4spdbh's 333 bsel mod. She posted at 3.98 GHz with 1.53v, but would not boot Windows, just a strange error screen after POST.

I went into the BIOS and gave the vcore the max 106% that it would take, rebooted and she got into Windows 2000! :cool:

Is it stable? I don't know yet. But it is certainly running warm, considering the voltage and that it's running on a stock heatsink. :eek:

celeron4604ghzuq2.jpg


I'm happy that it's working, especially on this cheapo $31 open box motherboard from newegg, and on stock cooling. Time to load up Prime 95. I'll let you know how it goes.
 
HOLLY VOLTAGE BATMAN!!!!!

and YEAH BOI my 333fsb mod worked :p


as per voltage mods i figured a way that you can do it that you do not need to know the stock vid.

let me do some painting on a pic or 2

(btw how did you do the voltage mod? connect all the vid pins and cover vid 1 to make it 1.6v?)
 
Last edited:
nd4spdbh2 said:
(btw how did you do the voltage mod? connect all the vid pins and cover vid 1 to make it 1.6v?)
Click to expand...

69485298lo9.jpg


I connected the pads along the red line, and then covered the green square.

EDIT: Yeah, the stock cooler definitely will not cut it for these runs. As-is the load temps will probably kill it, if the high vcore doesn't strike it down first. :D
 
Last edited:
Well it was good while it lasted, but I had to undo the mods. This little Biostar GF7050V-M7 board has apparently died on me, and I have no other boards that will work with the pad mods.

It ran for a little while earlier in the week, but then started randomly shutting off, I initially had some trouble booting it back up, it got worse, and then shortly before it quit POSTing on me it began giving me a "MCP73 VGA BIOS" message that said something about "engineering release - not for public" and followed by a complete hard lock.

I removed all pad-mods, and at that point it simply refused to boot, and for a moment I thought I had killed the Celly. I placed my E6600 and a known good stick of RAM into the Biostar, and it still refused to boot, only spinning fans, confirming that the board looks to be toasted. I wonder if supplying 1.76Vcore was a contributing factor? :D

Fortunately the Celeron 460 boots up fine in my GA-965GM-S2, so the chip apparently survived my experiments last week. Tough little soldier.
 
Last edited:
haha tough lil soldier is right, just brings back the image of a cely 430 runing w/o a hs doing spi 1m runs haha


GETS ANOTHER BOARD AND CLOCK THE SNOT OUTTA THAT SUCKER!!!
 
Amazing. I tried putting it together one more time, and this time the Biostar booted up. So much for jumping to conclusions. :shrug: It's humming along at 3.2 GHz with no pad-mods, on the stock cooler. I'll leave it like this until I get a new cooler for my quad, and then I'll place that SI-128 onto this Celly and give 4 GHz another try.
 
Last edited:
You must log in or register to reply here.

Similar threads

T
New config. advice
2
Replies
30
Views
717
shalena20
S
A
AMD FX-6300 what is the correct voltage for 3.5 GHz ?
Replies
3
Views
1K
EarthDog
EarthDog
T
is overclocking possible on the SuperMicro X8DT-L X58 dualsocket motherboard?
Replies
2
Views
522
ssjwizard
ssjwizard
mackerel
FEATURED Zen 4 APUs anyone?
Replies
19
Views
782
dejo
dejo
CYRIX
Wrong RAM memory recognition
Replies
3
Views
600
freeagent
freeagent
Back