It just isn't moving in a direction that makes it very applicable to the average desktop PC I'm aware that the Power architecture is still around, but aside from a few systems running AmigaOS, MorphOS, Linux, etc it is dead on the desktop. Quote: You are always starting threads like this that only require a simple internet search for the answer. OK I'm bored The problem for RISC is everything else was far from equal, it was a huge advantage for x Why did PowerPC fail? In the early s, we had worked out how to do out of order, superscalar, branch-predicting processors.
Designers realised there was TONS of instruction-level parallelism to extract Now how to do that in 5 million features? RISC was one solution to the problem.
RISC also solved the problem of compilers being stupid, because the entire thing was stupid, there was nothing for a compiler to be stupid in. It couldn't make mistakes as there were no mistakes to make, there was only one way to do an operation. The same transistors which did decoding and microcode stayed about the same count, but the available real estate on the die increased exponentially.
Dual ALUs became common. Pipelined FPUs would fit in too. It didn't really matter what the front end was doing, or what instructions or registers it was remapping. Because RISC was all about trimming the fat off the front end, it didn't matter either. It itself had become all clunky and decodey and microcodey. It was just another instruction set.
Didn't even need a heatsink and gave you all the performance of a fairly decent Pentium. They all either used lots of power and were fast, or used no power at all and were reasonably fast. Turns out the second category was quite popular as mobile devices took off. PowerPC had nothing there. So what PowerPC applications still live on? Supercomputers, a Blue Gene model holds the number 3, 5, 8, and 9 spots out of the top 10 supercomputers.
All using 1. Quote: RISC also solved the problem of compilers being stupid, because the entire thing was stupid, there was nothing for a compiler to be stupid in. Each core is eight-way hardware multithreaded and can be dynamically and automatically partitioned to have either one, two, four or all eight threads active.
I was actually contacted by email based on a comp. At least Paul DeMone had chip design experience, and he recognized that he was not a good fit for a Senior Architect position. The amount of effort put into PowerPC design seems to have been relatively limited.
While some of this may be attributed to economic necessity even with a healthy Apple and substantial embedded use, PPC sales volume was not huge , I suspect that Motorola-internal factors may have contributed to such. Motorola's concern for the embedded market may also have discouraged rapid microarchitectural change somewhat since many embedded developers want longer lifetimes on highly-compatible parts. The need for the embedded market to provide a substantial portion of production volume may also have discouraged development of microarchitectures more suited to personal computers and workstations.
The power-aware embedded focus may have helped Apple for laptops and small form factor computers. It might be noted that PowerPC was a successor in some markets for 68k and 88k. I have no idea whether Motorola's management of those two architectures should have been considered evidence of substantial internal issues, but in ignorant hindsight it seems betting on Motorola even with IBM as a second source may not have been a good plan for Apple.
Apple probably did not help volume with its anti-clone attitude. While PowerPC is not the best-designed ISA even for the trade-offs of the early s and x86 volume advantages would have eventually pushed it out of the personal computer market, I suspect with better yet economically reasonable implementations Apple might have stayed with PowerPC for a few more years than it did.
One might even speculate that if Intel had not improved its laptop chips as soon as it did, x86 would not look quite as attractive to Apple as early as it did.
Apple's laptop market share and perhaps even Transmeta's effort may have influenced how aggressively Intel pursued this market. Growing laptop volume and power-constrained performance would have pressed Intel to greater energy efficiency anyway, but perceived competition may have increased the pressure.
Hat Monster wrote: RISC also solved the problem of compilers being stupid, because the entire thing was stupid, there was nothing for a compiler to be stupid in.
This is either really good news or really bad news, depending on your perspective. The good news is, hey, free processor! Both are completely realized CPUs with real software support, both are free to use, and both are under the aegis of the Linux Foundation , meaning no one person or company controls their destiny.
On the other hand, this seems like a sad and humiliating end to a once-proud processor family. Oh, how the mighty have fallen. How did that ever turn out…? For a while, PowerPC was a big hit. Motorola shifted its sights downward from desktop processors to embedded chips. IBM back when it was selling microprocessors did the same thing, producing the PowerPC , , and related low-end devices.
Slowly, the PowerPC market bifurcated, with high-end silicon driving exotic IBM iron and low-end chips in various embedded systems, but with nothing much in between. Alas, it was not to be.
Licensing your processor IP was all the rage in the s and early s, and it was the surest route to mass adoption. Unofficially, IBM wanted to keep the ranks of its licensees small in order to reduce internecine competition. ARM, on the other hand, would seemingly welcome anyone who could fog a mirror and write a check.
In all, nearly two dozen companies paid for PowerPC licenses, which seems like a lot, but only a handful of those — Applied Micro, Cisco, Sony, STMicroelectronics, and Toshiba — were well known or produced chips in any appreciable volume.
The smartphone market started using ARM right from the beginning, and now ARM has become the dominant supplier in that market. A budding market provides opportunity for the superior technology to take foothold, which is exactly what ARM did. Because this architecture is widely used, developers already have experience with it and are much more likely to pick it up. The tooling required for making a new line of chips costs billions, so you have to churn out a lot of chips to offset the costs.
And it shows, because Intel has been having issues manufacturing anything smaller than 7nm while TSMC has already announced 4nm. Growth of the market is the enemy of economies of scale advantages. A larger market means that your fixed-to-variable cost ratio will go down. The differential has shrunk, and the fixed costs which were once choking ARM are less of a problem now.
With the advent of smartphones, the market for CPUs have indeed doubled in size. But my point stands, a larger market means that economies of scale barriers dissipate and more specialized and niche firms now have the chance to tackle the incumbent. A larger market has created room for these new firms to rise and create new economies of scale advantages of their own. Finally, Apple is using Rosetta 2 to help early adopter customers use x86 applications on the new Macs, preventing an early death like PowerPC on Windows.
All of these things combined means Apple silicon has a much better chance of succeeding than PowerPC did. Some sections of this post are near verbatim from this book.
Those sections have been marked with an inline citation. AMD marketshare expanding. Operating system market share. Like my writing about PowerPC? Here are just a few pieces that discuss different aspects of this era:. But like the PowerPC, Intel is struggling to keep up with the demand from a tough client.
So now, apparently, Apple looks to its in-house ARM chipsets to give it that long-wanted vertical integration, 35 years in the making. Good thing it has the money to throw around this time. Find this one an interesting read? Share it with a pal! Ernie Smith is the editor of Tedium, and an active internet snarker. Between his many internet side projects, he finds time to hang out with his wife Cat, who's funnier than he is.
Find me on: Twitter Like this? Well, you should read more of our stuff. Get more issues in your inbox. What is Tedium? Support us on Patreon Share your ideas! Disclosure: From time to time, we may use affiliate links in our content—but only when it makes sense. All rights reserved. Please, try the fish. Like what you're reading? About Archives Sponsor Us.
By Ernie Smith Jun 16, Tweet Share Subscribe. Consider subscribing to our freaking newsletter.
0コメント