What point are you trying to make here? That poster has actual real world design experience on some of the most important chips ever made.
What are you trying to nitpick? That time has marched forward?
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Developer Delves Into Reasons Why Apple's M1 Chip is So Fast
- Thread starter MacRumors
- Start date
- Sort by reaction score
What are you trying to nitpick? That time has marched forward?
My point? That there are other factors at play other than RISC versus CISC. Of the four processors he claims to have worked on two are dead and one is almost dead. All three of which are RISC processors. The only one still standing, by a long shot, is the CISC design.
No, it is not a change of name. They're related but they're not the same. If you're claiming they are then your credibility is even worse than originally thought.
You clearly not only seem to have no clue about processor design, but you lack all respect for great engineering tasks as well. My advice: instead of trolling here, go to AnandTech, ArsTechnica or so and learn one or two things about technology. The x86-64 is a great improvement over the x86 and I draw my hat in respect to anyone having contributed to such a project. I'd gladly invite you to a beer or two, cmaier, and have a bit of insight into such a great piece of work.
However, I marvel at the M1 as well, these numbers are insane. I have a 2015 iMac, and the geekbench single-core rating is at 1100, the 2020 imac is at around 1200, but the mac mini is at 1700. It's insane. I also marvel at the Ryzen family. Gosh, that would be a lot of beers... ;-)
My next computer will probably be a mac mini for everything except games, for which I might build a zen 3, nvidia 3080/90 machine. Hehe, maybe I can put the mac mini inside the pc tower... squeeze it diagonally into two drive bays or so...
There is no need to bash x86 or any other architecture. Compete and learn from each other to improve all mankind.
Intel tried to kill StrongARM back in the early 2000's when they sold the former DEC division to an underpowered holding company. Little did they know then that that was a big mistake.
Your number don't seem right. Mac has been steady, if not increasing, for the past few years. https://www.statista.com/statistics/576473/united-states-quarterly-pc-shipment-share-apple/
LOL! Typical response on this site. Claim someone as no clue. I've heard all of this before when Apple switched from 68K to PPC. Then they abandoned a RISC design for a CISC design. Perhaps you can explain why?
Neither are celeron and Xeon. I didn’t say they were the same.
I also didn’t keep working on Mips, and I didn’t keep working on SPARC. The ones I did work on were the fastest chips in the world at the time.
It’s been explained to you many times. IBM was focussed on chips for its own workstations. Motorola was inept. Neither was able to deliver chips suitable for laptops. Not the fault of PowerPC architecture, but the fault of business decisions and circumstances. And 68k was a huge mess, but it’s actually less of a mess than x86. Anyone designing these chips today, no matter how good they are, are constrained by decisions made by other people decades ago. Even Intel understood this, when they tried Merced. Which didn’t work for the same reasons that the risc chips didn’t work - they didn’t run windows in a reasonable way. Intel had to resort to sticking a crappy x86 core in the corner of the chip. Because business considerations, not technology, got in the way.
Well, I guess its been you who questioned @cmaier’s expertise in the first place. So you of all people are guilty of what you are accusing others more than anyone else. cmaier, who, by way way, forgets more about CPU design every single day than you know about CPUs at all
ehm... yes, sorry, that was a bit harsh, I apologise... maybe you have a clue but I cannot recognize it from the posts I read.
The RISC vs CISC debate was not really meaningful then and it is even less meaningful now, imo. I have no real insight on why they switched, but very convincing reasons have been written here before:
- there was no path forward for the PowerPC, both IBM and Motorola had their focus elsewhere.
- PowerPC did not do well on the laptop level, which would become a major segment
And there is always the economies of scale. At that time, Intel looked much more trustworthy than any other option. It gave Apple stability, continuity and, most importantly, the time to slowly build up their own capacity in chip design, to never be dependent of such 'partners' anymore. The time has come, we see the fruits in the iPhones/iPads and now the M1.
The RISC vs CISC debate was not really meaningful then and it is even less meaningful now, imo. I have no real insight on why they switched, but very convincing reasons have been written here before:
- there was no path forward for the PowerPC, both IBM and Motorola had their focus elsewhere.
- PowerPC did not do well on the laptop level, which would become a major segment
And there is always the economies of scale. At that time, Intel looked much more trustworthy than any other option. It gave Apple stability, continuity and, most importantly, the time to slowly build up their own capacity in chip design, to never be dependent of such 'partners' anymore. The time has come, we see the fruits in the iPhones/iPads and now the M1.
Why couldn't IBM or Motorola deliver chips suitable for laptops? Wasn't the PPC a RISC processor?
I agree. Hence why I said "The performance benefit of M1 has less to do with CISC versus RISC...". IMO RISC is not the primary reason for M1s performance advantage. Other factors are more important.
If you are going to just pick a few words out of a sentence you can make it say anything you want.
Motorola couldn’t deliver chips that didn’t suck even for desktops. They were bad at chip design.
And IBM had no interest in doing so because, as I said before, they were focussed on chips for their own RS/6000 workstations, which made them a ton more money than selling chips to Apple.
Your entire argument is based on a flawed hypothesis. Just because an architecture is good doesn’t mean that a DESIGN will be good. Intel and AMD both use x86-64, but that doesn’t mean that their designs are equally successful.
I am not making it say anything. Did you or did you not write those words?
What factors? You just keep throwing these statements out without any details. What are the factors, why are they more important, and what background do you have that suggests that your opinion is meaningful?
I said “It’s a change of name. Same fundamental architecture. Like Xeon vs. Celeron.”
And I stand by my statement. “Power” is the new name for the architecture that was previously called “PowerPC.” Of course, it has evolved over the years, just like today’s x86-64 is not the same as it was 10 years ago. ”PowerPC” was the name of a series of chips which supported the PowerPC architecture (which is now called “Power”).
Here’s proof:
IBM Gives Away PowerPC; Goes Open Source
“With great power comes great responsibility.” – Uncle Ben Remember PowerPC? Nah, me neither. But now it can be yours. For free. Srsly. IBM has released the Power Architecture instruction set to t…
www.eejournal.com
POWER® family and PowerPC® architecture overview
A POWER® family or PowerPC® microprocessor contains a branch processor, a fixed-point processor, and a floating-point processor.
www.ibm.com
PowerPC
PowerPC (short for Performance Optimization With Enhanced RISC – Performance Computing, sometimes abbreviated as PPC) is a RISC architecture created by the 1991 Apple–IBM–Motorola alliance, known as AIM. PowerPC, as an evolving instruction set, has since 2006 been renamed Power ISA but lives on...
(And if you start googling, don’t get confused - POWER and Power are not the same thing, though POWER became PowerPC (with changes) and is now merged back into Power).