The only logical buyer if Apple didn’t purchase them is either Motorola or IBM if they were to only be competitive in the embedded systems market. If they were never purchased given the rise of ARM processors they could have actually started making those for the mobile phone market instead and probably would have been competitive to Qualcomm .
They shouldnt have sold. They should be competing with AMD & Intel & Apple.
You also have the wrong market segment. This was supposed to be a 25w chip with crazy high i/o. Dual 10Gbit, plenty of pcie & sata... back in 2006... at low power, all integrated... this chip was supposed to be a dara pushing monster.
Maybe they could have scaled down but the intent was obviously different, to deliver performant extremely well integrared embedded & bigger cpus.
> You also have the wrong market segment. This was supposed to be a 25w chip with crazy high i/o. Dual 10Gbit, plenty of pcie & sata... back in 2006... at low power, all integrated... this chip was supposed to be a dara pushing monster.
Most of that stuff was a result of their very impressive integrated northbridge, but they never publicly demonstrated any CPU that could actually drive all that bandwidth.
Supposedly they had plans to eventually ship a 16 core processor that would really use all of the I/O bandwidth, but they never manufactured anything more than a 7w, dual core part. The Amiga X1000 only featured a single 1 GigE port.
So it’s not like they sold when they had a revolutionary monster ready to dominate the market sitting in warehouses. They had long-term plans, but they may have been starting to doubt the feasibility of executing them, or being relevant by the time they’d scaled the architecture up to sixteen cores, by the time Apple came along.
I think of Ryzen as the first modern chip that has a multipurpose serdes, but here's PA Semi kicking out this [1] on their first & only chip:
> The device also integrates a flexible I/O subsystem that supports eight PCI Express controllers, two 10-Gigabit Ethernet controllers and four Gigabit Ethernet controllers that share 24 configurable SERDES lanes.
It's interesting that you highlight the cpu as such a limiting factor:
> they never publicly demonstrated any CPU that could actually drive all that bandwidth.
Agreed that the plans to scale up cores were very interesting & promising. But the other ultra-notable features of this ultra-integrated badass SoC monster (the one that shipped) was that it had a colossal crossbar powered DMA engine.
In many applications the cpu could serve as control plane, & the rest of the chip could act like a data-plane, shipping data around. The northbridge can happily pipe gobs of ethernet over to your SATA system, semi-autonomously.
At the time that usually took special carefully tuned code to pull off, I imagine. I like to think today we'd be able to use this DMA engine far more readily- io_uring for async io, all manners of dma-buf linux capabilities for managing device-side memory, plenty of dma engines & offloads in various kernel drivers. Direct device to device communication has continued to be underscores as a huge way to u leash computing, and wow, PA Semi dedicated gobs of die-space to the task way back then.
Such a huge huge pity such a promising new competitor just vanished.
> They had long-term plans, but they may have been starting to doubt the feasibility of executing them,
They indeed ran hotter than expected & core performance was not nearly what was hoped for. According to various LKML posts of people who had dev boxes. Still, so impressed. Tons of brilliant innovation. A little more runway to get off the ground before being torn apart & swallowed by a titan would have been great.
> I think of Ryzen as the first modern chip that has a multipurpose serdes,
Do you mean specifically that Ryzen was the first CPU to offer that kind of IO flexibility? Because Intel desktop chipsets were doing that kind of thing long before Ryzen hit the scene. Here's are chipsets for Haswell and Skylake:
The mobile phone market wouldn't have saved them. Qualcomm killed off everyone who wasn't Apple with their licensing schemes. The CPU tech was largely irrelevant because what really mattered was the wireless patents. Being acquired by Apple was probably the best outcome, and we have PA Semi to partially thank for the rise of ARM processors as we consider it today.
The PowerBooks were so far behind in performance at that point that Apple would have had to figure out something else if they hadn’t gone to Intel.
If not Intel switch, Macs would not have had the popularity surge — that was driven by the idea that, worst case, you could still run Windows full time if you had to. It might not have mattered as much once the iPhone was out, but it probably would have increased the pressure for non-Xcode dev tools for the iPhone.
G4 Powerbooks were very usable and are still highly collectable today. I have two G4s myself.
Yes, Intel kit was somewhat quicker, in some tests, some of the time, but bear in mind the comparison at the time: Intel's state of the art then was the Pentium 4.
The P4 had a nickname of being a fanheater, because it ran so hot and had such a low IPC that its performance even at high clock speeds was poor.
AMD was just bringing in Sledghammer, the Opteron/Athlon64 range, but they were not laptop chips either. So that meant that they were up against the G5 as a desktop chip, which was respectable.
The things you are missing are:
* Apple saw that laptops were more important than desktops. POWER was strong and still is, but as a server chip, not in portables.
* Apple was sold on the promise of future Intel chips, mainly the Core 2 series, and nothing Intel had at that time...
* ... or that AMD had at that time.
* There is another big advantage that is forgotten now that Apple saw then.
PowerPC was Apple's old platform. The pre-OS X OS, Classic, ran on PowerPC. OS X could run Classic in a VM on PowerPC. This was good for long-time Apple users.
But if it switched to Intel, notably the future 64-bit Intel chips that didn't even exist yet, then it would lose the ability to run Classic in a VM... but it would gain the ability to run x86 OSes natively in VMs. In other words it could run Windows in a VM at near-full-speed, giving the ability to run Windows apps, which looking to the future was far more important than the ability to run old PowerPC Classic MacOS apps.
It was 100% not an issue of "Apple compared Intel to PowerPC and Intel was quicker." At the time of the comparison, Intel was slower and hotter.
> Apple was sold on the promise of future Intel chips, mainly the Core 2 series, and nothing Intel had at that time...
Yup. It's worth remembering that at the time of the switch, Apple downgraded from 64 bit Power PC chips to Intel's initial version of Core which was only 32 bit.
They didn't get back to a 64 bit implementation until Core 2 launched.
>The P4 had a nickname of being a fanheater, because it ran so hot and had such a low IPC that its performance even at high clock speeds was poor.
The Pentium IV ran so hot that the infamous demo back in the day was of people cooking pasta on it.
> It's worth remembering that at the time of the switch, Apple downgraded from 64 bit Power PC chips to Intel's initial version of Core which was only 32 bit.
Oh my word! That's an excellent point which I did not notice even at the time.
Reports from back in the day said that Apple strongly considered adopting this chip for their laptops before ultimately switching to Intel.
>The Register is reporting that PA Semi had a close relationship with Apple in the months prior to the switch, and that the two companies were working together to gauge the feasibility of running native PowerPC-coded software on 2GHz dual-core chips that PA Semi claims run at only a third of the 21 to 25 watts consumed by Intel's Core Duo models. Sources who spoke to El Reg say that executives were virtually positive that they'd win the contract, and that CEO Dan Dobberpuhl was understandably "furious" when he found out PA Semi had been passed over.
Had Apple adopted a chip from PA Semi, PA Semi could not have been acquihired for the same price for certain, which could have significantly been worse for the Apple ARM chips. In hindsight it's hard to see that it could have worked out better for Apple in the long term, but being passed over by Apple was bad news for PASemi.
The PA6T was absurdly better perf/watt than what Freescale was offering at the time, and IBM had already signaled it was was no longer interested in making any Power chip that wasn't designed specifically for its own mainframes.
IBM made POWER for its own RISC workstations and servers: originally the RT/PC, later RS/6000, now IBM p Series.
Not mainframes, which have a different CPU architecture altogether.
Part of Apple's insight was that laptops were at that time (i.e. 2006-2008) becoming more important than desktops. It had G5 desktops but only G4 laptops.
IBM said it wasn't going to be doing any laptop chips, and that was a factor pushing Apple away from PowerPC. PA Semi could have been an alternative path, though.
I know there are thousands of unsung engineers that drive the progress of technology, but it is amazing how a few folks seem to have this outsized influence. In no way I am saying that they alone are responsible, but it does make me curious about the actual topology of innovation.
The semiconductor world is very small by tech standards, and so there are correspondingly many fewer people at the management or upper IC levels that can coordinate large, consequential decisions and even fewer that become well known. Even ignoring attribution biases, it should be unsurprising that there are a few folks that seem to have outsized influence.
But I will say that there are many more people who are not well known outside of the semiconductor world, but are minor legends within it. Unfortunately, the development processes for hardware tend to go through enough hands to strip attribution and most hardware people tend not to talk in public about their accomplishments. Doesn't help that some of the main companies have been stagnant and absorbed in inner and outer turf wars for nontechnical reasons. It's not a great environment for stars to shine.
On top of that, most software people I've seen seem uninterested in understanding and dissecting the hardware enough to appreciate what they've done. If that's the case for software people, I have no hope for anyone else.
I remember quite well hopping for a PowerBook based on what PA Semi was rumored to be making.
I was devastated when Steve dropped the Intel bomb. The PC industry was destined to standardize on the most mediocre architecture, brute forced into a high performance chip due to sheer economies of scale.
DEC and Sun where no more and even Apple had to give in if it wanted a different fate.
Thankfully, the mighty cell phone saved us all from boredom and we're living interesting times in CPU land again.
34 comments
[ 4.8 ms ] story [ 94.6 ms ] threadYou also have the wrong market segment. This was supposed to be a 25w chip with crazy high i/o. Dual 10Gbit, plenty of pcie & sata... back in 2006... at low power, all integrated... this chip was supposed to be a dara pushing monster.
Maybe they could have scaled down but the intent was obviously different, to deliver performant extremely well integrared embedded & bigger cpus.
Most of that stuff was a result of their very impressive integrated northbridge, but they never publicly demonstrated any CPU that could actually drive all that bandwidth.
Supposedly they had plans to eventually ship a 16 core processor that would really use all of the I/O bandwidth, but they never manufactured anything more than a 7w, dual core part. The Amiga X1000 only featured a single 1 GigE port.
So it’s not like they sold when they had a revolutionary monster ready to dominate the market sitting in warehouses. They had long-term plans, but they may have been starting to doubt the feasibility of executing them, or being relevant by the time they’d scaled the architecture up to sixteen cores, by the time Apple came along.
> The device also integrates a flexible I/O subsystem that supports eight PCI Express controllers, two 10-Gigabit Ethernet controllers and four Gigabit Ethernet controllers that share 24 configurable SERDES lanes.
It's interesting that you highlight the cpu as such a limiting factor:
> they never publicly demonstrated any CPU that could actually drive all that bandwidth.
Agreed that the plans to scale up cores were very interesting & promising. But the other ultra-notable features of this ultra-integrated badass SoC monster (the one that shipped) was that it had a colossal crossbar powered DMA engine.
In many applications the cpu could serve as control plane, & the rest of the chip could act like a data-plane, shipping data around. The northbridge can happily pipe gobs of ethernet over to your SATA system, semi-autonomously.
At the time that usually took special carefully tuned code to pull off, I imagine. I like to think today we'd be able to use this DMA engine far more readily- io_uring for async io, all manners of dma-buf linux capabilities for managing device-side memory, plenty of dma engines & offloads in various kernel drivers. Direct device to device communication has continued to be underscores as a huge way to u leash computing, and wow, PA Semi dedicated gobs of die-space to the task way back then.
Such a huge huge pity such a promising new competitor just vanished.
> They had long-term plans, but they may have been starting to doubt the feasibility of executing them,
They indeed ran hotter than expected & core performance was not nearly what was hoped for. According to various LKML posts of people who had dev boxes. Still, so impressed. Tons of brilliant innovation. A little more runway to get off the ground before being torn apart & swallowed by a titan would have been great.
[1] http://vita.mil-embedded.com/news/p-processor-ever-designed/
Do you mean specifically that Ryzen was the first CPU to offer that kind of IO flexibility? Because Intel desktop chipsets were doing that kind of thing long before Ryzen hit the scene. Here's are chipsets for Haswell and Skylake:
Haswell: https://images.anandtech.com/doci/7963/FlexIO.png
Skylake: https://images.anandtech.com/doci/9582/122%20-%20HSIO%20H.jp...
would they have bought them and stayed ppc the whole time?
would macs have gotten as popular (without bootcamp/intel compatibility) as they are today?
If not Intel switch, Macs would not have had the popularity surge — that was driven by the idea that, worst case, you could still run Windows full time if you had to. It might not have mattered as much once the iPhone was out, but it probably would have increased the pressure for non-Xcode dev tools for the iPhone.
G4 Powerbooks were very usable and are still highly collectable today. I have two G4s myself.
Yes, Intel kit was somewhat quicker, in some tests, some of the time, but bear in mind the comparison at the time: Intel's state of the art then was the Pentium 4.
The P4 had a nickname of being a fanheater, because it ran so hot and had such a low IPC that its performance even at high clock speeds was poor.
AMD was just bringing in Sledghammer, the Opteron/Athlon64 range, but they were not laptop chips either. So that meant that they were up against the G5 as a desktop chip, which was respectable.
The things you are missing are:
* Apple saw that laptops were more important than desktops. POWER was strong and still is, but as a server chip, not in portables.
* Apple was sold on the promise of future Intel chips, mainly the Core 2 series, and nothing Intel had at that time...
* ... or that AMD had at that time.
* There is another big advantage that is forgotten now that Apple saw then.
PowerPC was Apple's old platform. The pre-OS X OS, Classic, ran on PowerPC. OS X could run Classic in a VM on PowerPC. This was good for long-time Apple users.
But if it switched to Intel, notably the future 64-bit Intel chips that didn't even exist yet, then it would lose the ability to run Classic in a VM... but it would gain the ability to run x86 OSes natively in VMs. In other words it could run Windows in a VM at near-full-speed, giving the ability to run Windows apps, which looking to the future was far more important than the ability to run old PowerPC Classic MacOS apps.
It was 100% not an issue of "Apple compared Intel to PowerPC and Intel was quicker." At the time of the comparison, Intel was slower and hotter.
Yup. It's worth remembering that at the time of the switch, Apple downgraded from 64 bit Power PC chips to Intel's initial version of Core which was only 32 bit.
They didn't get back to a 64 bit implementation until Core 2 launched.
>The P4 had a nickname of being a fanheater, because it ran so hot and had such a low IPC that its performance even at high clock speeds was poor.
The Pentium IV ran so hot that the infamous demo back in the day was of people cooking pasta on it.
https://www.youtube.com/watch?v=sG3wceb05Sg
Oh my word! That's an excellent point which I did not notice even at the time.
Good catch, and a great video reference, too!
>The Register is reporting that PA Semi had a close relationship with Apple in the months prior to the switch, and that the two companies were working together to gauge the feasibility of running native PowerPC-coded software on 2GHz dual-core chips that PA Semi claims run at only a third of the 21 to 25 watts consumed by Intel's Core Duo models. Sources who spoke to El Reg say that executives were virtually positive that they'd win the contract, and that CEO Dan Dobberpuhl was understandably "furious" when he found out PA Semi had been passed over.
https://www.engadget.com/2006-05-22-apple-considered-small-c...
(The Reg is my day job but this is long before my time there.)
The PA6T was absurdly better perf/watt than what Freescale was offering at the time, and IBM had already signaled it was was no longer interested in making any Power chip that wasn't designed specifically for its own mainframes.
IBM made POWER for its own RISC workstations and servers: originally the RT/PC, later RS/6000, now IBM p Series.
Not mainframes, which have a different CPU architecture altogether.
Part of Apple's insight was that laptops were at that time (i.e. 2006-2008) becoming more important than desktops. It had G5 desktops but only G4 laptops.
IBM said it wasn't going to be doing any laptop chips, and that was a factor pushing Apple away from PowerPC. PA Semi could have been an alternative path, though.
https://www.youtube.com/watch?v=JQgLYm2YYfg
I know there are thousands of unsung engineers that drive the progress of technology, but it is amazing how a few folks seem to have this outsized influence. In no way I am saying that they alone are responsible, but it does make me curious about the actual topology of innovation.
But I will say that there are many more people who are not well known outside of the semiconductor world, but are minor legends within it. Unfortunately, the development processes for hardware tend to go through enough hands to strip attribution and most hardware people tend not to talk in public about their accomplishments. Doesn't help that some of the main companies have been stagnant and absorbed in inner and outer turf wars for nontechnical reasons. It's not a great environment for stars to shine.
On top of that, most software people I've seen seem uninterested in understanding and dissecting the hardware enough to appreciate what they've done. If that's the case for software people, I have no hope for anyone else.
If anyone has the content, there's at least one very curious HNer here :)
I was devastated when Steve dropped the Intel bomb. The PC industry was destined to standardize on the most mediocre architecture, brute forced into a high performance chip due to sheer economies of scale.
DEC and Sun where no more and even Apple had to give in if it wanted a different fate.
Thankfully, the mighty cell phone saved us all from boredom and we're living interesting times in CPU land again.
It's a different word (/ˈhoʊpɪŋ/) from "hopping" (/ˈhɒpɪŋ/) with a different meaning.
:-)
I loved PowerPC, and was very hopeful for PA Semi. They took it in a different, but still interesting direction.