legion said:
The "idea that Apple holds off" should have started with even the announcement of the G5 servers. They should not announce them until the servers are good to go.
Doesn't that about sum it up? Everyone's assuming there are hardware problems. Delay does not equal problem. It's nice to see you look wider than the crowd
🙂
Apple's PR created this dilemma. I'm sure the hardware developers are quite unhappy about the situation and the degree of pre-announcement. It's all very Osbourne.
For those questioning the competency of Apple's hardware teams, go design and test something simple yourself, like an IDE interface or and ARM-based SBC. Once you have designed, tested and qualified it, you'll better understand the basic process, the types of faults that can occur, etc. Apple has a design process in place that works very well. Remember, it's far more complex than any PC desktop manufacturer because PC manufacturers have chipsets/core logic designed by experts. Apple has to design their own boards, their own supporting chipsets, their own case designs, etc.
Got a problem with your supporting chipset? Maybe if register X is set to such and such, and there is data traveling to X device while data is coming from Y device during a cache miss on the processor then there is a 1 in 100,000,000 chance that something unintended will occur. With a 2GHz processor that could happen every few seconds! Go round with a chip mod, get samples (6 weeks), build new samples, requalify, repeat.
These add a lot to cost and to time-to-market, but they also add that 'value' that makes people want an Apple in the first place.
