Forums | MacLife

Thx NightCougar_37, that's exactly what I was looking for.

So, I'll install the bigger fan, apply some greese, cross my fingers and try to hit 433.  If I make it, I would be very close to the limit, so it wouldn't be worth the effort of replacing the heat sink.  Thanks again.

I just clock chipped a Rev. D iMac tonight to 400Mhz. I ran Xbench before and after. If you look at the Xbench results online you should look at "Guest's Computer". Thats my  iMac. From what I've read it doesnt sound like anyone has gotten a rev. D. beyond 400Mhz. The few posts that I have read say that the iMac wont boot at 433Mhz. If you figure out a way, please post it here.

There are cooling suggestions on the iMac clock chipping site FYI... but from what I've read you dont need to add an additional fan to cool a rev. D running at 400Mhz. I have read that some people with 7200rpm drives have had problems.. but others did not (the success story was running a seagate drive).

I've read about special fans you can buy to make you computer quieter and cooler. That might be a good choice for the iMac... if one fits. I havent found a temperature monitoring software that works with the iMac in OS X...anyone know of one that works with these old iMacs?

Also, is there a way to get a rev D to take more than 512 megs of RAM?

Nightcougar,

We aren't using software to upgrade our iMacs... we are soldering resitors on our processor boards to obtain faster clock speeds.  Its documented here:

http://www.thelittlemacshop.com/imac/imacmod1.html

You can also see other people's results here:

http://forums.xlr8yourmac.com/cpureview.lasso

and select  "Apple iMac" as the Mac Model
and select  "Apple"      as the Card Manufacturer

This will show you entries made by other people who have
similar systems and have clock-chipped them.

This is a great way to see what other machines have been
modified to, and what the limits of stability are.


Here are some notes from my experience clock-chipping last night:

First I benchmarked the iMac before the procedure using XBench:

http://www.xbench.com/

I was really glad that I did this as a reference point. I later ran it afterward to see the difference. Be sure to deactivate the screensaver though and dont touch the computer while benchmarking (and quit all apps beforehand).

I taped my processor board to the back side of a dinner plate. I used strips of thin cardboard between the tape and the processor board. This is because tearing adhesive tape can cause static electricity (its visible if you are in a photographic darkroom). I used the back of the dinner plate as a surface because that allowed me to rotate the whole plate including my processor board quickly and solder from different angles. One false move and you can toast the whole board its nice to have the flexibilty to choose your angle.  Also, at one point, the lip of the plate captured one of the resistors preventing me from losing it.

I used just a normal cheap soldering iron and a pair of high quality tweazers. The iron had a sharpened point tip. I let the iron get very hot so that I didn't have to linger on the part... just a few seconds and the solder on one end of the resistor was melted and then another few and the heat going through the resistor melted the other end of the resistor's solder. At that point the resistor would slide across the surface of the processor board like butter on a pan. I usually could pick it up with the tweazers at that point.

Dont do this: On my first resistor I started out by using desoldering tape that I bought at Radio Shack. Its designed to lift up unwanted solder. It works well for some tasks but not this one. Dont do this... I realized that it was leaving microscopic hairs of copper on the processor board that were then getting stuck to the solder. It also didn't leave much solder behind for resoldering the resistor to the new jumper location. On the second resistor I just used the iron and heated the hell out of it until it slipped.

When I put it all back together the iMac wouldn't turn on... no chime or anything. I took the iMac apart again and discovered a few things. One was that the dust and filth had lodged itself in the RAM sockets and processor sockets. I blew that clean. That might have been my problem. But I decided to take a closer look at my soldering job. I used the lens from a pair of binoculars to study my soldering. I saw how uneven my soldering was. I then spent 20 minutes heating one of the resistors and carefully trying to position it perfectly. It would have been easier to do it right the first time. Word to the wise, dont be sloppy about this...because the solder eventually gets 'sucked' out of the pad for lack of a better term. Because of this sucking, I then used the thinnest silver solder I had from radio shack (less than a milimeter in diameter) and put a miniscule amount on the tip of the iron. Then I deposited that onto the resistor end that seemed to lack solder. I made sure there were no cold solder joints and then put the iMac back together. It booted right up (faster than before I might add) and the About box reported 400 Mhz. Hooray!

Xbench reported a %16 overall performance increase (not bad for free). Oddly, drive performance went down a tiny bit on the tests after the upgrade...but overall the mac feels faster.

Here is the comparison of before and after of my Rev. D iMac 333Mhz to 400Mhz upgrade.

http://ladd.dyndns.org/xbench/merge.xht … oc2=109397

Anyone else have thoughts on this?

I guess in the ideal scenario, you would have a well lit room, with a magnifing glass on an arm and a surface mount soldering iron and maybe an assistant... but I did it at 3 in the morning in a dark kitchen with a $5 soldering iron. It probably tripled the time needed to do it.

Last edited by Macmurph (2005-05-03 7:03 pm)

I know the process of what you guys are doing so I know its not software. What I was refering to was the settings for changing the voltage on the board.

Your L2 cache speed can be controlled and clocked on the fly via apps like CPUD. For those CPUs i'd try to keep the L2 cache speed under 250Mhz. Tho you'd be more likely safer with it under 240 or less. When I clocked my 300 to 375 it clocked the L2 to about 250. While it worked, it would not be stable everytime. So clocking it down to 187Mhz was the only way to keep it stable.

If you guys want an easier way to do all that resistor work, just use the rear window defogger method. Never have to ever use a soldering iron with it. Just pry the resistor off then apply the defogger paint.