x86 vs Power PC in depth; the real issue is heat not benchmarks

“This article is concerned with the technical differences between [x86 and PowerPC] not the market differences,” writes Nicholas Blachford for OS News.

“The x86 family of CPUs began life in 1978 as the 8086, an extension to the 8 bit 8080 CPU. It was a 16bit CISC (Complex instruction Set Computing) processor. In the following year the 8088 was introduced which was used in the original IBM PC. It is this computer which lead to todays PCs which are still compatible with the 8086 instruction set from 1978,” Blachford begins. “The PowerPC family began life with the PowerPC 601 in 1993, the result of a collaboration started in 1991 between Apple, IBM and Motorola. The family was designed to be a low cost RISC (Reduced Instruction Set Computing) CPU, it was based on the existing IBM POWER CPU used in the RS/6000 workstations so it would have an existing software base.”

“x86 has the advantage of a massive market place and the domination of Microsoft. There is plenty of low cost hardware and tons of software to run on it, the same cannot be said for any other CPU architecture. RISC may be technically better but it is held in a niche by market forces which prefer the lower cost and plentiful software for x86. Market forces do not work on technical grounds and rarely chose the best solution. Could that be about to change? There are changes afoot and these could have an unpredictable effect on the market,” Blachford explains.

“Computers are now so fast it’s getting difficult to tell the difference between CPUs even if their clock speeds are a GHz apart. What’s the point of upgrading your computer if you’re not going to notice any difference? How many people really need a computer that’s even over 1GHz? If your computer feels slow at that speed it’s because the OS has not been optimised for responsiveness, it’s not the fault of the CPU – just ask anyone using BeOS or MorphOS. There have of course always been people who can use as much power as they can get their hands on but their numbers are small and getting smaller. Notably Apple’s software division has invested in exactly these sorts of applications,” writes Blachford.

Blachford concludes, “What is going to be a hurdle for x86 systems is heat. x86 CPUs already get hot and require considerable cooling but this is getting worse and eventually it will hit a wall. A report by the publishers of Microprocessor Report indicated that Intel is expected to start hitting the heat wall in 2004. x86 CPUs generate a great deal of heat because they are pushed to give maximum performance but because of their inefficient instruction set this takes a lot of energy. In order to compete with one another AMD and Intel will need to keep upping their clock rates and running their chips at the limit, their chips are going to get hotter and hotter.”

“You may not think heat is important but once you put a number of computers together heat becomes a real problem as does the cost of electricity. The x86’s cost advantage becomes irrelevant when the cooling system costs many times the cost of the computers,” writes Blachford. “RISC CPUs like the 970 are at a distinct advantage here as they give competitive performance at significantly lower power consumption, they don’t need to be pushed to their limit to perform. Once they get a die shrink into the next process generation power consumption for the existing performance will go down. This strategy looks set to continue in the next generation POWER5.”

“The POWER5 (of which there will be a “consumer version” [read: G6]) will include Simultaneous Multi-Threading which effectively doubles the performance of the processor unlike Intel’s Hyper Threading which only boosted the performance by 20% (although this looks set to improve). IBM are also adding hardware acceleration of common functions such as communications and virtual memory acceleration onto the CPU. Despite these the number of transistors is not expected to grow by any significant measure so both manufacturing cost and heat dissipation will go down,” writes Blachford.

There is significant depth and much more to read in the full article here.

31 Comments

  1. Which is all very encouraging, and I can’t wait to get my hands on a G5. However, let me tell you a story…

    I got into work today an done of my colleagues was trying to get photos from her digital camera onto her Win2K PC and having all sorts of driver problems. She asked for my help so I suggested she try plugging it into the sole Mac in our office, an ageing 233MHz G3 iMac running OSX. Guess what? Yup, the memory card appeared immediately on the desktop and we copied the photos from there. Problem solved, everyone’s happy.

    It’s not how fast the hardware or software is, it’s what it enables you to do.

  2. Massive heatsinks? Try again. They are not “massive” in the Power Mac G5. All you need to do is compare a P4 with a G4 or G5 to see the power comsumption differences. x86 is old tech now hitting the wall. Apple Mac is positioned perfectly (no thanks to Motorola) now and will lead “Wintel” for the foreseeable future. In fact, “Wintel” will continue to fall behind.

  3. Edd, that is true. However, Windows 2K was never meant to be a multimedia OS. It’s a workstation OS. XP is better for that, and yes it does recognize my memory card readers and cams.

    And speed does matter. The macs at work do what I want it to do, but they are SLOW.

  4. Chomper, the 9 fans are for optimized zoned cooling. They are not the 9 full-size full-power fans that are required to cool a typical 2.5GHz and up PC. I know, the PC I have at work is LOUD and it’s loud all the time. The Mac fans aren’t as loud and they aren’t needed at all times.

  5. Overall a very good report.

    It is interesting to note the author does not give Motorola’s 88000 chip set as part of what was used as a basis for the PowerPC. The reason the PowerPC was capable of adding more functional units (such as AltiVec) was because the internal bus of the PowerPC was based upon the internal bus of the 88000 which was explicitly designed to include a bus which would allow the easy addition of more specific functional units. (Motorola was thinking about adding in a graphics processing unit, among other things, into the CPU in those days.)

    I also disagree with the author’s premise that CISC is inherently more power hungry and produces more heat than RISC. The original POWER chip sets came in 5 chip and 7 chip sets. Yes, the CPU was not on one chip. This was primarily due to huge power and heat of the original POWER chip sets. Sure, the POWER chip sets stomped anything from Intel, Sun, Mips, Motorola, HP, or anyone else, but they paid the price in power requirements and heat output.

    It took IBM over two additional years to figure out how to make a single chip version. It took another couple of years to figure out how to get the POWER functionality (combined with the 88000 features) into a desktop single chip CPU.

    However, my overall impression of the article is that it is fairly accurate and well balanced.

  6. I think it’s going a little far to say that Chomper is an A$$. Now, if you were to say he was a DUMBA$$, then I could go along with that.

    I must say, though, I wouldn’t laugh nearly as much were his comments not included. Ignorance is bliss – and very funny!

  7. Chomper:

    Until people have got their hands of the merchandise, this is all clearly speculation. However, it would seem to me that the size of the heatsinks is principally to increase the dissipative surface area and hence allow for a reduction in airflow. This would tie in with the use of 9 small, low rev. fans (rather than a couple of whizzing whoppas). The overarching purpose of all this seems to be noise reduction (yay) – which, if the G5 does indeed run at just 35 dbA (normal r.t.), seems to have been achieved.

    Cheers.

    Brother Mugga

    PS: I’m not saying the 970s don’t run hot (they clearly do), but rather that the size of the heatsinks is not commensurate to the temperature of the chips and should not be taken as a direct indication of power dissipation.

  8. Chomper, it uses 9 fans, it doesn’t need them except to keep it quiet… which is the same problem on most x86 computers… of course, you have all sorts of extra crap in a wintel box that isn’t designed to do anything but get in your way when you mistakenly pry open the case.

    That said, the heat sinks in the G5 aren’t that big, they are well placed and should be QUIET (which is, honestly, necessary, not an aesthetic issue) while the heatsinks on the new HP P4 desktops that we have here in my office are using 3 fans, all of which are LOUD, and heatsinks that are at least three, and quite possibly four times the size of those in the G5. So who has the real issue here? You, I would presume, since even though those computers are at my company, I have a relatively quiet MDD DP G4 and a PowerBook… ah, the life…

    heat is an issue, as anyone with a 12″ PB will tell you, but I know I read somewhere about the heat of the 970 (G5) being less than the high end G4s because of a design correction… now if I could find that link, I’d be really happy…

  9. PC’s are just as guilty of cooling fans. Some PC’s must have a fan attached to the CPU or they will not work not to mention the video card. And I know many PC’s with VERY loud fans. You tell me when a group of PC people banded together like the Mac Fanboys as one seems to call them to complain about something. And they fix it. You just dont get that sense of community from the PC world. It’s too big that it seems cold just like the very people that use them.

  10. why is it mac fanboys always have to attack people if they disagree? I use a Mac at work and favor macs, no problem there. I just hate seeing propaganda.

    I agree with some points and glad that some people are willing to have an intelligent convo. Of course there are others who aren’t as bright.

    Grow up…

Reader Feedback

This site uses Akismet to reduce spam. Learn how your comment data is processed.