In 2006, for the work i was doing, the intel powered imacs machines were a godsendįor me it felt like the first REALLY capable multi-thread multi-cpu computer. Im sure alot of G5 users electric bills were pretty high in that time period! but even it sucks way too much AC $$ juice for my liking, and was somewhat unstable i cant imagine what the quads would have pulled from 2005-2008 i have the single CPU 1.8ghz pci-x G5 now, but i didnt even own a G5 untill i purchased this one cheap around the time snow leopard arrived. Well i never used a quad G5 so there you go. Seriously though, the G5 doesn't do much more that crash and burn. Though I do believe I owe someone instructions on setting up a late model iBook G4. If I get some time I will post the OF commands required to boot OS 9 on the G5 so that you can have your own giant paperweight. Take all this with a grain of salt though as a lot of it is based on my knowledge and opinion. So in theory, this could be worked on and developed using the G4 and non-mac SATA cards. It had no value to anyone at the time to create these extensions. Once SATA really came around, OS 9 was pretty much irrelevant to the hardware manufactures. It hasn't been done because it was far easier to have IDE busses emulate SCSI for boot and such. This can be done now by writing an system extension/driver to accomplish this, but that has not happened.
#G5 MAC OS DRIVER#
So essentially proper full SATA driver and support will need to be added to the system. On the G5 only AHCI is supported and not IDE. One hardware issue is the SATA ports/bus. It can't even boot far enough to finish booting. Can you boot far enough to do anything? Not even close. It was extremely hard to get it to boot as it was not possible to get to open firmware or reset the nvram from the keyboard. When I worked on the possibility of OS 9 on the G5, my only G5 at the time almost became a brick. It's also very dangerous if you don't want a large paper weight. OS 9 recognizes some hardware incorrectly so it really messes with stuff. The interrupt controllers actually don't seem to hold it, but how the hardware is configured in the G5 certain hardware beyond some interrupt controllers would only disable if the interrupt controller is disabled. Video had to be greatly modified, and almost all interrupt controllers. Very important hardware had to be disabled such as the memory controller (not to be confused with the MMU on the CPU), PCI buss, power management, all sensors, USB, FW. To get this happening about 80% of the hardware has to be disabled or removed from the device tree (mostly removed). To even get some output will require the hardware to be setup and running. Now here is the reason for my thoughts above.Īs of right now it is possible to boot a G5 up to the point of it starting the nanokernel at which point it fails so early that getting debug output is basically not an option.
#G5 MAC OS DRIVERS#
By the time we get to even attempting G5, I do believe the knowledge will be there and will just be time needed to write correct drivers for the hardware. From there, I actually don't think it will be extremely difficult to get the rest of the system working, but there would still be significant work on the nanokernel and 68K emulator. That will be the first step to getting it all working on a G5. We are working at it as it will absolutely be needed to even get things like the mini and other partially working systems (Xserve among others). Knowledge of it's internal workings are theoretical at best right now. The trampoline (second stage boot loader that sets up a lot of hardware) is the first major road block. So here is where the G5 sits and will sit for some time. The later models that ditched the PCI bus will have some more difficulty. Once the all G4 hardware is supported then it's possible to move on to the G5.Īs far as the G5 though, the earlier once are more likely to get working. Right now it's easier to use the G4's to gain this knowledge as all of them are of the same basic hardware (this is a very extreme generalization). Not only that, but man hours are expensive monetarily. More knowledge of the inside of OS 9 is needed and we are gaining this, it just takes time and a lot of it.
![g5 mac os g5 mac os](https://i.ebayimg.com/images/g/7IgAAOSwq4lglbLT/s-l300.jpg)
This would take a huge amount of time and at this point not enough is known about certain internal workings that would get this working. The requirements (usually time and money) are the biggest barriers.