5 minute overclocking

a much better suggestion would be to desolder that component and not risk tearing the pad off the board.

Those components are not hard to desolder. Takes 3-5 seconds. You can even clean one side, and them tombstone the resistor on the other pad so you don't lose it.

Breaking this component pretty much destroys your ability to return the computer to it's original condition if there is a problem.

its not very hard to desolder something and it doesnt take to long expesialy with those new super fast heating up irons. those things are great.

If your finances are too tight for a bottom-end Ratty Shack soldering iron, you probably shouldn't be risking your G4 for a few megahertz. Cromm only knows where those little bits of conductive resistor shrapnel are going to go. How the heck do you cut an SMT resistor in half with needle-nose pliers, anyway?

There's a lot to be said for doing a job properly, with skill and care. It's safer, more likely to succeed, and just plain looks better.

If it were ever necessary to quickly overclock a G4 'in the field' with no proper tools around, then your method is good to consider. But I don't think that's a real-world scenario.

This method is a one-way hack. There's no going back, unless you do some soldering, which you should have been doing in the first place. It wouldn't take much for this to lift or tear a trace, then it's game over. I most strongly recommend against anyone attempting this.

BTW, is this something you've actually done yourself?

Not all folks are confident enough to swing an iron (or two) around such tiny SMT bits. The snipping process, while crude, does work and even if it rips the pads off the board it's not like we're talking an irreplacable artifact here. What do L3 cacheless CPU modules sell for nowadays? Maybe $50? So big deal if the whole thing is destroyed. And tearing that pad off wouldn't destroy the whole board anyhow.

This hack is for anyone who wants a quick and cheap speed jump without messing about with irons.

Hey, that said, I just remembered my own experience doing this same thing, only I used an iron. One iron. And guess what? I ripped one of the *&&&%$%@# pads off anyway. What a dumbass! I should have spent the extra few minutes or so it would have taken me to find my misplaced second iron.

Upshot was with that one jumper position destroyed I couldn't try the next speed jump up from 800. snif, snif :cry:

So kids, if you want to get anything faster than 800 out of your 733, buy 2 grounded irons ($7 each at Ratshack IIRC) and use 'em. Otherwise just smash the resistor and enjoy your 66MHz speed bump.

There's a lot to be said for doing a job properly, . . . and just plain looks better.

Heh heh heh, right. Bottom of the CPU module . . . looks better . . . hee hee hee

dan k

Why would you need 2 irons? 1 worked fine last time I checked, or is this another technique?

Correction: An improperly used soldering iron can ruin a computer just as much as cutting a resistor.

Not all folks are confident enough to swing an iron (or two) around such tiny SMT bits. The snipping process, while crude, does work and even if it rips the pads off the board it's not like we're talking an irreplacable artifact here. What do L3 cacheless CPU modules sell for nowadays? Maybe $50? So big deal if the whole thing is destroyed. And tearing that pad off wouldn't destroy the whole board anyhow.

Your logic baffles me. If this part costs you $50 to replace because you screwed up when you riped out that resistor with pliers, isn't it justified to get even a crude radioshack soldering iron for 1/3 the price of the replacement part?

Why would you need 2 irons? 1 worked fine last time I checked, or is this another technique?

Being able to heat and lift both ends of the resistor simultaneously greatly reduces the amount of futzing you have to do to get the part off the board. No, you don't need the second iron, but having it helps alot.

Your logic baffles me. If this part costs you $50 to replace because you screwed up when you riped out that resistor with pliers, isn't it justified to get even a crude radioshack soldering iron for 1/3 the price of the replacement part?

LOL, we're not talking logic here, because as I think we all agree it makes loads more sense to spend the dough for the right tools and do the job correctly.

However, even ripping the pads off the CPU isn't likely to kill the module, it'd just prevent you from turning the speed back down in the unlikely event the CPU wouldn't run at 800MHz. If that were the case, then yes, the CPU would be ruined (not really, but we'll say so for this discussion ...)

dan k

we're not bitchen, we're just proving a point that its better to not cut it because if you cut it you'll have to buy a whole new processor to fix it.

A temper tantrum in the middle of a discussion on the merits of brute force vs the right way of doing something does not help your point.

Calm down, Timmy boy. Telling people to DIE! could result in a one-way ticket to Bannedville, USA (if it were up to me, at least...and it ain't )

Your intentions for posting the hack were good, and it was a novel idea, but the rest of us are right to point out the risks and recommend alternatives.

Ya see, that's the whole joy of having an internet community like this. You get to express an idea, I get to say I don't agree, everyone else gets to examine both sides and choose their own path.

This post has me laughing my a$$ off. I just find smashing a resistor with a pair of pliers in order to overclock the machine by 67MHz to be funny as hell. Thats like saying you want a better flowing exhaust on your car so you just cut it off. Sure it works, but its pretty red neck, and has unwanted side effects. Using a soldering iron is not rocket science. You turn it on, put the tip on the shiny metal looking stuff, and poof... it liquifies. Its really quite amazing. Saying that this is for people who are "afraid to use a soldering iron" are most likely far more afraid of smashing a SMT resistor, sending its pieces flying all over and most likely damaging the pads it was soldered too. There is no way your going to just neatly "cut" a ceramic incased resistor in half. Oh, and as for your last post, I have done more then my fair share of overclocking Macs. From old 68k machines to modifying apple G4 cpu cards, as well as 3rd party upgrades (I think I am the only person to post how to mod the core voltage on an OWC G4 AGP upgrade). When people who have been doing this for ages, or work with/on electronics for a living (I know Bob has, as well as myself, not sure on others) say that what you suggest is a bad idea, it most likely means that it is. We are not trying to attack you personally, but I feel that its quite important to inform people that they can damage their computer. Not only can they damage the board, but if the computer is not stable after the overclock, they are screwed because they cant easily set it back.

A soldering iron can ruin a computer just as much as cutting a resistor. And yes I have done this.

you haven't done it right then. I used to do this all day long. takes about 3 seconds with 1 iron, or half a second with 2 irons.

Being able to heat and lift both ends of the resistor simultaneously greatly reduces the amount of futzing you have to do to get the part off the board. No, you don't need the second iron, but having it helps alot.

All you have to do is rapidly pass the tip of the iron between the two end of the resistor holding just long enough for the solder to melt. The component will heat up enough on about the 3rd pass that both ends are still liquid and it will easily slide off.

It's even easier if you have a tip that can reach both sides of the component at once, but those are a bit more specialized.

Well until you do (or if you are afarid) to to this hack then don't argue!

Timmy, posts like this and the all caps screaming above are innapropriate for AppleFritter. Behave like an adult, or find some place else to rant.

Consider yourself warned.