I appreciate articles like these because they're useful for people who don't know the background and details and aren't interested in a deep dive, but it'd be nice if the author spent a little more energy on details and correctness.
For instance, 20 bits of address space support 1 megabyte. The 640K limit in the IBM PC was due to the IBM PC's design. Also, Apple would've considered the 8086 over the 8088, considering the need for memory throughput to have a responsive, fully bitmapped display.
The addressing of the 68000 wasn't critical for the Mac like the article says; the Mac's designers took shortcuts that made 4 megs the limit for the first generation 68000 because it was easy and they had lots of space, although it really wasn't the limit because the Mac Portable can have up to 9 megabytes with the original 68000 with 24 bits of address.
The Pentium didn't compete with the m68040. The Pentium came out after the PowerPC 601. The 80486, which made it to 100 MHz from Intel (higher from other vendors), competed with the m68040.
I think some steps were missed between the power-hungry and heat-generating PowerPC and the MacBook Air. I don't remember any history showing the Air as something that Apple was trying to, but couldn't, make. The issue was that the G4 was good for its time, but there was never a proper G4 successor because the G5 was too power hungry and too hot. If there was some MacBook Air project that never got off the ground during the PowerPC years, I'd love to hear about it.
The comment about performance-per-watt of x86 versus PowerPC was only true of the G5, it's worth noting.
"Superscalar Architecture" is in the section of the article talking about, "What made Intel x86 CPUs so much better?", even though Motorola, Intel and PowerPC had been superscalar since the '90s. Intel definitely improved their superscalar implementations, but they certainly weren't unique ("to get superscalar architecture working effectively" is not how I'd put it).
The author wrote, "I had to know what actually caused Intel’s x86 architecture to be so far ahead of its competition.", but never mentions AMD anywhere in the article. If AMD hadn't seriously outperformed Intel and forced Intel to actually compete, and if AMD hadn't forced the creation of 64 bit x86, then Intel certainly would not have been an option for Intel. The Pentium 4 had the same heat and power issues as the PowerPC G5. Intel ended up cancelling the Pentium 5 and instead going ahead with a much modernized Pentium III core for their Core release which made up the initial release of Intel Macs because AMD were eating their lunch.
There are some other things, but they're mostly minor. It's a well written collection of information.
> Intel ended up cancelling the Pentium 5 and instead going ahead with a much modernized Pentium III core for their Core release which made up the initial release of Intel Macs because AMD were eating their lunch.
Intel based the Core on Pentium M, because Pentium 4 sucked. AMD happened to be competitive at that point of time, but that doesn't mean that without AMD, Intel would be blind to the fact that Pentium 4 didn't achieve Intel's own expectations - the NetBurst's ability to scale up frequency (they targeted 10 GHz) and keep wattage/heat in check simply did not work out.
"The Pentium M represented a new and radical departure for Intel, as it was not a low-power version of the desktop-oriented Pentium 4, but instead a heavily modified version of the Pentium III Tualatin design (itself based on the Pentium II core design,"
Yes, the MacBook Air was much more about being able to completely remove the DVD drive, SSD storage able to replace a traditional hard disk, and the battery formed of space efficient flat cells rather than a series of cylinders.
For instance, 20 bits of address space support 1 megabyte. The 640K limit in the IBM PC was due to the IBM PC's design. Also, Apple would've considered the 8086 over the 8088, considering the need for memory throughput to have a responsive, fully bitmapped display.
The addressing of the 68000 wasn't critical for the Mac like the article says; the Mac's designers took shortcuts that made 4 megs the limit for the first generation 68000 because it was easy and they had lots of space, although it really wasn't the limit because the Mac Portable can have up to 9 megabytes with the original 68000 with 24 bits of address.
The Pentium didn't compete with the m68040. The Pentium came out after the PowerPC 601. The 80486, which made it to 100 MHz from Intel (higher from other vendors), competed with the m68040.
I think some steps were missed between the power-hungry and heat-generating PowerPC and the MacBook Air. I don't remember any history showing the Air as something that Apple was trying to, but couldn't, make. The issue was that the G4 was good for its time, but there was never a proper G4 successor because the G5 was too power hungry and too hot. If there was some MacBook Air project that never got off the ground during the PowerPC years, I'd love to hear about it.
The comment about performance-per-watt of x86 versus PowerPC was only true of the G5, it's worth noting.
"Superscalar Architecture" is in the section of the article talking about, "What made Intel x86 CPUs so much better?", even though Motorola, Intel and PowerPC had been superscalar since the '90s. Intel definitely improved their superscalar implementations, but they certainly weren't unique ("to get superscalar architecture working effectively" is not how I'd put it).
The author wrote, "I had to know what actually caused Intel’s x86 architecture to be so far ahead of its competition.", but never mentions AMD anywhere in the article. If AMD hadn't seriously outperformed Intel and forced Intel to actually compete, and if AMD hadn't forced the creation of 64 bit x86, then Intel certainly would not have been an option for Intel. The Pentium 4 had the same heat and power issues as the PowerPC G5. Intel ended up cancelling the Pentium 5 and instead going ahead with a much modernized Pentium III core for their Core release which made up the initial release of Intel Macs because AMD were eating their lunch.
There are some other things, but they're mostly minor. It's a well written collection of information.