• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

PPC740 G3 CPU on a machine that came with a 603e

mg.man

Well-known member
This is with 512K of cache installed
Ah. I knew it wouldn't be as good as backside L2, but figured better than nothing. Given the lack of backside L2, perhaps it's mean to compare to a 'real' G3/300... what does it bench compared to the original 603e/275Mhz?
 

Paralel

Well-known member
The only thing I can say for sure is that the one attempt with a Blackbird CPU daughterboard was a failure.
 

Paralel

Well-known member
the situation where they destroyed the mold for the connector?

No, here on 68kMLA, we tried to transplant a 740 onto an upgrade daughtercard that had a 603e on it. But it never worked. As far as I remember we were never sure what the problem was on why it wouldn't work. It was just dead silent, like the CPU was dead.
 

dosdude1

Well-known member
No, here on 68kMLA, we tried to transplant a 740 onto an upgrade daughtercard that had a 603e on it. But it never worked. As far as I remember we were never sure what the problem was on why it wouldn't work. It was just dead silent, like the CPU was dead.
Was the chip reballed before being installed? These chips (all PPC chips, not just the 740s) have these weird not-quite-solder 0.89MM balls applied, which I have yet to figure out how to actually make solder correctly. Whenever I attempt to, they just don't make proper contact with the pads. I've tried pre-tinning the pads before soldering, and that doesn't help either. I've given up at this point, and whenever I get chips with those original balls on them, I just remove them all and reball it with standard solder balls, which works every time.
 

Paralel

Well-known member
Was the chip reballed before being installed? These chips (all PPC chips, not just the 740s) have these weird not-quite-solder 0.89MM balls applied, which I have yet to figure out how to actually make solder correctly. Whenever I attempt to, they just don't make proper contact with the pads. I've tried pre-tinning the pads before soldering, and that doesn't help either. I've given up at this point, and whenever I get chips with those original balls on them, I just remove them all and reball it with standard solder balls, which works every time.

Yep, it was reballed.
 

3lectr1cPPC

Well-known member
Not sure where that thread would be, I've read through nearly every thread on the PowerBook subforums... hacks and development?
 

Paralel

Well-known member
Yep, that's the thread. Bolle still has that board? He must still have the processor then as well since I left it to him to conclude either putting the board back to stock or to try and work out why it wasn't working.

The processor was an MPC740ARX266LH, so it should have been a drop-in replacement. Never did figure out why it just acted stupid.
 

Paralel

Well-known member
I guess all we can do is ask @Bolle to get in contact with @dosdude1 to see if between the two of them they can figure out what went wrong. In the end if they can get it working, it goes back to @CC_333 as agreed. If nothing can be done, I leave it up to CC to decide, but I would imagine he would prefer to get the daughterboard back. Whoever ends up with the processor, being CC if there is success, or if not it can go to whoever thinks they can do something with it. I have no use for it, so it would just end up in my spare part pile to collect dust.
 

dosdude1

Well-known member
I guess I should update this thread with info about this upgrade being done on PowerBooks... Short story is, it doesn't work. I tried doing the same 740 swap on a PowerBook 3400C, and while it did chime, it did nothing more than that. No video, and no boot. Re-installing the original 603 restored functionality. I can only assume at this point that there is some check in the system firmware that needs to be patched, similar to what needs to be done when doing a 7448 upgrade on some of the later machines. Unfortunately, in the case of these old-world ROM machines, the BootROM is not editable, as they utilize mask ROMs to store the ROM content. I'm not sure if it'd be possible to replace the 4 ROMs with standard EEPROMs, but at this point I don't even have a dump of the machine's ROM to look at. If I had a dump available, I could most likely figure out the issue and patch it (assuming a check in the BootROM is indeed the cause) relatively easily.
 

3lectr1cPPC

Well-known member
That page is about a decade out of date from the looks, PPC emulators are out now and do require ROM images so there is a way. I’ll do some more research tomorrow on it and see if I can ROM dump my 3400 (well, later today, it’s nearly 1AM for me right now…)
 
Top