4 ms·
There is a downside to this - and one which could cause Intel to fall off a wall. Their tick-tock method of introducing a new product and then half-way through
by excuse-me 15y ago
There is a downside to this - and one which could cause Intel to fall off a wall.
Their tick-tock method of introducing a new product and then half-way through it's life switching it to a small design rule to get a speed and cost benefit as you introduce the next bigger chip is really great - so long as you can make the design rules smaller AND the world wants bigger more powerful versions of the same chip
But it prevented them stepping back and designing a crude little 30K transistor RISC processor from a blank sheet - which later goes onto rule the world.
- kryptiskt 15y ago> But it prevented them stepping back and designing a crude little 30K transistor RISC processor from a blank sheet - which later goes onto rule the world. It's unfair to tick-tock to blame it for their failure to preempt ARM, it is far more recent than that. It's not that Intel hasn't tried to wean away folks from x86 with novel and innovative blank-sheet architectures. They tried with ipx432, i860 and Itanic, but all were expensive and well-deserved failures. It certainly seems like Intel engineers need the constraints of the x86 ISA to keep their wild impulses in check.
- fidotron 15y agoIt's not recent at all. Ironically the ARM exists at all because Intel wouldn't let anyone license the 286 design, and at one point Intel had a license to make StrongARMs, which they willingly gave up. Intel are simply an example of brute force in action.
- jasonwatkinspdx 15y agoIt's irrational to expect a business to get every decision about the future correct.
- InclinedPlane 15y agoThe irony is that Itanium is now a decent processor. But when it came out it was a trial of a new type of architecture (VLIW) and it made a lot of questionable tradeoffs for the time. Also, consider their many other big projects which have succeeded, such as Coppermine, Core2, etc.
- wmf 15y agoIt has been argued that ARM profit margins are not big enough to pay for Intel's R&D, so ruling the world would require some serious strategic changes and restructuring.