Robert - RE: Power PC vs. Intel.
When IBM/Apple and tag-along Motorola announced the PowerPC in 1991, it's forecasted specs were quite a bit more powerful than the 486 which Intel was making at that time.
However, by the time IBM/MOTO/APPLE finally delivered a POWERPC equipped Macintosh (late 1993/early 1994), Intel was producing its SECOND generation Pentium (0.6 micron process, 90 MHz and 100 MHz speeds).
The PowerPC 601 came out of the shoot at 80 - 100 MHz, maybe a little higher. Macintoshes outperformed comparable PCs at that time by perhaps 10%, maybe 15%. Heavily integer oriented software ran as fast or faster on a PC, but the PowerPC MACs had much superior floating point capability.
MAC magazines always published Adobe Photoshop speed tests (heavily floating point oriented) which ALWAYS showed the MAC faster than the PC. However, for most applications, the MAC wasn't appreciably faster than a PC
So, a 10% - 15% speed factor for a 30% or more HIGHER cost (MACS cost more, especially back then) was not a compelling argument for anyone to switch from a PC to a MAC. MAC afficionados, a very loyal group, loved the PowerPC, and they tended to buy the new machines. Result - no gain in market share.
In the mean time, software development for the WINTEL platforms was snowballing, with many Apple defectors switching to Wintel in order to survive. This loss of underlying software support (and mindset among the MAC crowd) has contributed to the PowerPC/Apple's woes.
Since 1994/1995, Intel has accelerated the clock speed of the Pentiums and introduced a streamlined architecture in the advanced Pentium Pro with greatly improved floating point capability. All these introductions launched Intel into the catbird seat of performance. IBM was struggling to get the POWERPC 601 speeded up, and had to go to 603, 603e, 604 and 604E, as well as spending a lot of effort on a 620 which has never seen the light of day.
Basically, Intel took on IBM + MOTOROLA + APPLE in CPU chip architecture, design, and wafer fab process development and beat the three of them single handedly. No Mean Feat!
This victory is by no means insignificant, as Apple's dwindling market share illustrates, and IBM is not even manufacturing a reasonable machine based on the PowerP*ssy chip. And, IBM through in the towel on OS/2 for the PowerP*ssy chip last year.
Paul |
