ploaf, how did you open the bridges?

[NedClocker]

flounder43:
The problem is, that you need to connect some bridges and disconnect other bridges on the cpu to achieve different default multipliers. People have done that, but I do not think that it is very easy to do. Ploaf has done it on a regular duron, I think.

I usually just boost the voltage up, and overclock on the fsb.
--------------------------------------------------------------------------------
NedClocker :
I hope it's not a big problem. We'll see how easy it is. The real pain is that, as you work your way up, you may have to close one that you just opened and visa versa.
--------------------------------------------------------------------------------
flounder43:
Bingo! That is the problem. I am too scared to try it... Maybe you can get some tips on that from ploaf.

How about it, ploaf? Can you give me any tips on opening and re-closing bridges. I've got a 1GHz morgan. flounder already showed me where to find the bridge configuration.

[res0r9lm]

I would try an excto knife and a pair of the strongest reading glasses you can find to open the bridges and a pencil to close bridges temporaryly until you find your max stable speed.
Have you thought about using a Golden Socket ? would be alot easier

[Ploaf]

I did this on every amd chip I have :)
I used a dremel with a diamond bit to open the bridges. It mostly requires a steady hand. Others have tried to use an xacto knife, I think with some success, but the diamond bit made getting through the ceramic easier and it was quicker. A magnifying glass helps and it also helped that I had a cpu packing tray to hold the athlons/duron. If you use the diamond bit w/magnifying lense make a couple of practice passes somewhere where you won't hurt anything so that you can better judge how to hit the bridges. If it's an xacto knife I don't have many helpful hints.

I closed the appropriate bridges with the window defogger repair paint. So far it's held up on every chip that I tried it on although a conductive ink pen would have probably been easier. The defogger paint is hard to spread. I used the edge of a razor to apply it and had to redo it a few times, always cleaning the cpu after each failed attempt. On a couple of cpu's I opened all of the bridges, and I mean every last one of them and then reconnected as I needed playing with various multipliers/voltages, but it's easier to just do the minimum. Much less fuss if you know what the chip will do. Is that what you were looking for?

[NedClocker]

ploaf, yes, that's exactly what I was looking for. Thank you. An idea hit me a while ago. What do you think about this?

I have a mobo that supports multiplier and voltage changes. Epox 8KTA3+. I can just put the duron in it, figure out how high it'll go using the bios settings, and then pull it out and change the bridges.

What do you think?

Of course, I know that there's no guarantee that a chip will oc as high in one board as it will in another. But it should at least give me a better idea of a good starting point.

res0r9lm, yes, I have read about those. How much do they cost? And what do you think about my idea?

[Ploaf]

btw. if you have to reopen a bridge that you closed with defogger paint it's not that big of a deal. It will pretty much just scrape off. Just make sure that the bridge is definitely open. It's hard to see sometimes, because they are so small.

[Ploaf]

Originally posted by NedClocker
ploaf, yes, that's exactly what I was looking for. Thank you. An idea hit me a while ago. What do you think about this?

I have a mobo that supports multiplier and voltage changes. Epox 8KTA3+. I can just put the duron in it, figure out how high it'll go using the bios settings, and then pull it out and change the bridges.

What do you think?

Of course, I know that there's no guarantee that a chip will oc as high in one board as it will in another. But it should at least give me a better idea of a good starting point.

res0r9lm, yes, I have read about those. How much do they cost? And what do you think about my idea?

I like the idea of the Golden socket. Much easier, but then it's something else you have to buy. Either way. Cheaper than screwing up a CPU though ;)
Testing it in a mobo that supports multiplier changes is probably a good idea. I had a good idea of what my cpu's would do before I monkeyed with them. You can OC it to the max and then back it off a bit. I assume that you are going mostly for a multiplier OC and not messing with the fsb all that much or at all.
oh yeah, and don't cross the bridges ;)

[res0r9lm]

Originally posted by NedClocker
ploaf, yes, that's exactly what I was looking for. Thank you. An idea hit me a while ago. What do you think about this?

I have a mobo that supports multiplier and voltage changes. Epox 8KTA3+. I can just put the duron in it, figure out how high it'll go using the bios settings, and then pull it out and change the bridges.

What do you think?

Of course, I know that there's no guarantee that a chip will oc as high in one board as it will in another. But it should at least give me a better idea of a good starting point.

res0r9lm, yes, I have read about those. How much do they cost? And what do you think about my idea?
golden socket are around $50 kinda expensive might as well get a mb with multipier adjustment if you were to go that route becuase you would still have to take hsf and cpu off.
I like your idea kind slick.I'm glad I thought of it.:beer:

[NedClocker]

Originally posted by Ploaf


I like the idea of the Golden socket. Much easier, but then it's something else you have to buy. Either way. Cheaper than screwing up a CPU though ;)
Testing it in a mobo that supports multiplier changes is probably a good idea. I had a good idea of what my cpu's would do before I monkeyed with them. You can OC it to the max and then back it off a bit. I assume that you are going mostly for a multiplier OC and not messing with the fsb all that much or at all.
oh yeah, and don't cross the bridges ;)

I don't know. 1 gig durons are getting pretty cheap!
Anyway, yes, the multiplier setting. I am using the ECS K7S5A mobo. Your only options with it are fsb of 100 or 133, unless you install the UNofficial bios update. The words unofficial and bios, used together in the same sentence, do not give me a warm fuzzy feeling inside. :D

Do you suppose it would work at 133 fsb if I maxed out the voltage, from 1.75 to 1.85??? and left the multiplier alone?

[NedClocker]

Cool. I'm glad you did, too.:) Don't know what we'd do without us.

[Ploaf]

You know. I didn't think about that- the fsb on those being 100 instead of 133. If you can get that Morgan to 1330 that would be really good and it would be a much easier way to go instead of messing with the multiplier. Are they going that high? Mine is an 800@1000 and I dont have a morgan to try out. The only thing left would be the voltage and that should be simple enough since you could probably just connect all of the voltage bridges with a good hsf.

[res0r9lm]

the ECS K7S5A mobo. Your only options with it are fsb of 100 or 133, unless you install the UNofficial bios update. The words unofficial and bios, used together in the same sentence, do not give me a warm fuzzy feeling inside.
Might work Just give it a try and see what happens. I read on amdmb.com that people were having sucess running on 133 fsb

[NedClocker]

p and r, ok. maybe try that first. to hop it up from 1.75 v to 1.85 v, just connect one bridge. easier than popping cpus back and forth between boards. will definitely have mobo monitor running. have thermaltake volcano 5 hsf. don't no if it's good enough. it keeps it at 100 F right now, full load.

33% is a big jump, though.

[NedClocker]

Gentlemen, I regret to inform you that baby Morgan would not even post at 133 fsb, even with vcore of 1.85. Connecting bridges is easy. If it wasn't almost 1:00 AM, I would go back in there and change the multiplier to see if I could get 1.2 GHz. All I have to do is cut 2 and close 2 to try that. Voltage is still 1.85; I haven't changed it back. Temp is 100 F at full load; running folding. Yikes, I gotta get up at 7.

[res0r9lm]

Ha, thats funny my cpu will do 133 but board won't and you have a board that will but chip won't. go figure:burn:

[NedClocker]

Maybe it's the power supply. It's only 200 watts. But all I have in that box is a hard drive and a pci video card.

[res0r9lm]

you think that's small. my old hp k2-550 had a 145w psu needless to say didn't last 6 mos

[NedClocker]

Nope. Tried 300 watt ps. No post. :(