Do the Intel based Macs finally destroy the myth that Intel are inferior to Motorola?

StormRider

Diamond Member
Mar 12, 2000
8,324
2
0
I grew up being an Atari guy. I had the Atari 800, 800XL, 130XE, and MegaST4.

I always thought the 6502 was superior to the Z80 and 8080.

Then I thought the 68K architecture was superior to the Intel x86 architecture. But Motorola couldn't keep up with Intel and the 68040 was basically the last in that line. We all explained it away by saying that Intel just had more resources available and if Motorola had the same resources then the 68K would be still alive today. However, at this time, I began to waver in my belief that the Intel x86 architecture was inferior. I began to have doubts.

Then the PowerPC came with a new RISC architecture. And it flourished. However, now we have Intel based Macs and from what I've read, the Intel based Macs are much faster than the PowerPC Macs when running native applications.

So, I think the Myth of the Inferiority of the Intel Architecture is finally dead.
 

Bobthelost

Diamond Member
Dec 1, 2005
4,360
0
0
It's not a myth, the PPC system is much more elegant than the 86 system. But for one reason or another they haven't scaled the speed up, and motorola failed consistently to provide chips on time, in bulk and basically screwed up.

The shift to the new yonah cores are a move to a new generation of CPUs, i'd be surprised if they were slower.
 

StormRider

Diamond Member
Mar 12, 2000
8,324
2
0
Originally posted by: Bobthelost
It's not a myth, the PPC system is much more elegant than the 86 system. But for one reason or another they haven't scaled the speed up, and motorola failed consistently to provide chips on time, in bulk and basically screwed up.

The shift to the new yonah cores are a move to a new generation of CPUs, i'd be surprised if they were slower.


But we heard the same thing about the 68K. The 68K was much more elegant than the x86. It had an orthogonal instruction set, non-segmented memory addressing, big-endian instead of little-endian (nice for reading assembly dumps) etc.

But in the end, Motorola couldn't keep up with Intel in terms of clock speed. In the end, the Intel architecture ran things faster and that's what really counts.

I see the same thing happened with the PowerPC architecture.

Once, I can explain it away. But twice in a row a supposedly inferior architecture beats out a superior architecture? Actually, it's 3 times in a row if you count the Itanium.
 

Bobthelost

Diamond Member
Dec 1, 2005
4,360
0
0
I'm doing a degree that will be focusing on this area, come back and ask me again in two years when i know what i'm talking about :D

But for now: Betamax and VHS.
 

DaveSimmons

Elite Member
Aug 12, 2001
40,730
670
126
Once, I can explain it away. But twice in a row a supposedly inferior architecture beats out a superior architecture? Actually, it's 3 times in a row if you count the Itanium.

Itanium ran into the problem of requiring a new code base (tools, OS build, etc.) when hundreds of millions of systems were already using x86. The AMD approach of just extending x86 one more time was much more practical.

For the other two, like you say the designs didn't scale in terms of clock speed, so they were more elegant but fatally flawed in the most important area.
 

0roo0roo

No Lifer
Sep 21, 2002
64,795
84
91
err, the myth has been pretty much dead for a while now. its why they switched after all
 

Goi

Diamond Member
Oct 10, 1999
6,766
7
91
Well, the current G5 powermacs are all using IBM architectured CPUs, not Motorola 68K.
 

omissible

Member
Aug 21, 2004
57
0
0
Intel sells twenty times as many chips as Motorola or IBM ever did. Therefore, Intel has an R&D budget many times the size of Motorola's or IBM's.

The 68k was better than x86. SPARC was better than x86. MIPS was better than x86. PowerPC was better than x86. In some cases, SPARC and PowerPC still are, but that won't last. With enough engineers and enough money, you can turn the tables.
 

Cerb

Elite Member
Aug 26, 2000
17,484
33
86
I think fanboys suck on both sides.
* The G3 sucked.
* The G4 sucked until it got fast DDR. By the time that happened, we had 3GHz P4s and 2.2GHz Athlons. Yes, it has nice vector performance, but few apps used it, and the SSE and SSE2 performance of Athlon64 and P4 chips is pretty good, now.
* The G5 is fine, but nothing special.
* Apple fanboys refused to admit they bought the things because of UI and style. These are the same kinds of people that try to make Lian-Li cases looks like good performers, too.

Maybe x86 isn't elegant. But it has scaled very well. It may not be perfect, but it has the benefit of having 'grown up' rather than being envisioned, and has a lot of little tweaks and hacks from decades of this that help make our Athlons and Pentiums perform well for their cost and transistor counts. Also, it doesn't hurt that the Alp^H^H^HAMD folks are still offering competition for Intel :).