Short version: not worth it.
---
So I'm the proud owner of a Powerbook G4 12 inch. I bought it mostly-mint and I love it. I surprisingly am able to actually use it for daily work which is impressive considering it's over 10 years old! I'm typing this thread on it. Anyway, while I'm fairly happy with the thermal output on it, I wondered if I could do better by applying new thermal paste. I've heard some accounts of 8c drop in temps and some others with no drop at all. So I decided to give it a go.
First the specs:
12 inch Powerbook G4 Aluminum
1.5ghz 1.25gb ram SSD
Arctic Silver MX-4 thermal paste
I measured temps before and after and took some pictures. I ran both before and after tests in a room with 73F ambient temp using a Flash youtube video playing in full screen. I let it go for 20 minutes then checked the cpu and gpu temps using Temperature Monitor.
Before: cpu: 61.5c -- gpu: 58.5c
I followed the guide and removed the heatsink. Several things stood out to me. First the thick thermal pads and then the exposed heatpipes
![]()
^The heatsink covers 3 chips. From the top: the g4, a very large Agere chip (southbridge?), and the nvidia gpu. The bottom two have a thick reusuable thermal pad. The G4 had something else on it though. My best guess is some sort of phase change material. It was a pain to remove.
![]()
^Here's a shot with the pads removed. You can clearly see the exposed heatpipes. This is less than ideal as it allows massive air pockets between the thermal pad and the heatsink.
![]()
^ here's a shot of the chips without the heatsink. The g4 is on top, then an Agere chip (aka big-chip), and last the nvidia gpu. It looks like the vram is stacked on top of the gpu in a package-on-package config.
![]()
^a shiny and clean G4!
So my original plan was to apply thermal paste on all 3 chips. However after opening up the heatsink I had to change my plan for several reasons. First the exposed heatpipes would mean I would need to use a lot of thermal paste to remove the air bubbles. Second, the nvidia gpu was partially covered by the laptop's internal frame. If I had to reapply thermal paste it would be very difficult to clean up the gpu without having to remove the entire frame. And most importantly, the frame forced a gap between the heatsink and the big-chip and gpu. This isn't an issue for a thermal pad since it's thick enough to easily bridge the two. But a thermal paste is only supposed to be a thin spread. It's not meant to fill 1.5 mm gaps!
The G4 made direct chip to heatsink contact, so I used thermal paste on there. But I refitted the big-chip and gpu thermal pads. I closed everything back up and ran my test again.
Before: cpu: 61.5c -- gpu: 58.5c
After: CPU: 61.0c -- GPU: 58.2
I suppose they are lower. But they're so close they might as well be the same.
All in all I'm a bit disappointed by the cooling system of the G4 Powerbook 12 inch. 1.5 mm thermal pads and exposed heatpipes that create huge air pockets! But even then it's a huge step up from my 15 inch tibook.
While it was cool to take a peak inside the machine, there was virtually no movement in temps and you run a risk of causing serious damage to the computer. I wouldn't recommend doing this.
---
So I'm the proud owner of a Powerbook G4 12 inch. I bought it mostly-mint and I love it. I surprisingly am able to actually use it for daily work which is impressive considering it's over 10 years old! I'm typing this thread on it. Anyway, while I'm fairly happy with the thermal output on it, I wondered if I could do better by applying new thermal paste. I've heard some accounts of 8c drop in temps and some others with no drop at all. So I decided to give it a go.
First the specs:
12 inch Powerbook G4 Aluminum
1.5ghz 1.25gb ram SSD
Arctic Silver MX-4 thermal paste
I measured temps before and after and took some pictures. I ran both before and after tests in a room with 73F ambient temp using a Flash youtube video playing in full screen. I let it go for 20 minutes then checked the cpu and gpu temps using Temperature Monitor.
Before: cpu: 61.5c -- gpu: 58.5c
I followed the guide and removed the heatsink. Several things stood out to me. First the thick thermal pads and then the exposed heatpipes
^The heatsink covers 3 chips. From the top: the g4, a very large Agere chip (southbridge?), and the nvidia gpu. The bottom two have a thick reusuable thermal pad. The G4 had something else on it though. My best guess is some sort of phase change material. It was a pain to remove.
^Here's a shot with the pads removed. You can clearly see the exposed heatpipes. This is less than ideal as it allows massive air pockets between the thermal pad and the heatsink.
^ here's a shot of the chips without the heatsink. The g4 is on top, then an Agere chip (aka big-chip), and last the nvidia gpu. It looks like the vram is stacked on top of the gpu in a package-on-package config.
^a shiny and clean G4!
So my original plan was to apply thermal paste on all 3 chips. However after opening up the heatsink I had to change my plan for several reasons. First the exposed heatpipes would mean I would need to use a lot of thermal paste to remove the air bubbles. Second, the nvidia gpu was partially covered by the laptop's internal frame. If I had to reapply thermal paste it would be very difficult to clean up the gpu without having to remove the entire frame. And most importantly, the frame forced a gap between the heatsink and the big-chip and gpu. This isn't an issue for a thermal pad since it's thick enough to easily bridge the two. But a thermal paste is only supposed to be a thin spread. It's not meant to fill 1.5 mm gaps!
The G4 made direct chip to heatsink contact, so I used thermal paste on there. But I refitted the big-chip and gpu thermal pads. I closed everything back up and ran my test again.
Before: cpu: 61.5c -- gpu: 58.5c
After: CPU: 61.0c -- GPU: 58.2
I suppose they are lower. But they're so close they might as well be the same.
All in all I'm a bit disappointed by the cooling system of the G4 Powerbook 12 inch. 1.5 mm thermal pads and exposed heatpipes that create huge air pockets! But even then it's a huge step up from my 15 inch tibook.
While it was cool to take a peak inside the machine, there was virtually no movement in temps and you run a risk of causing serious damage to the computer. I wouldn't recommend doing this.
