After IBM-compatible clones took over from Apple, Atari and ZX Spectrum, we just got used to that a PC is an X86 with MS Windows and Office on it. Around a decade ago Apple fought back with OSX on which Windows 7 (launched in 2009) was the first real answer. Meanwhile Apple switched to Intel, since IBM was not fast enough with the development of the POWER-processor – a huge operation, which seemed a one-time-only step for Apple at the time. SemiAccurate now speaks of Intel being replaced by ARM on Apple’s laptops.

A few weeks ago I asked Computer Science students if they knew ARM. Not even 1% had heard of it, but lots more knew there was a Samsung-chip in their smartphone. So what’s going on without us knowing it?

I’ll try to describe the market for a few key-years and then try to put the big names in it. There is a lot going on between i.e. Nvidia, Samsung, Texas Instruments and Imagination Technologies in the ARM-market, but I’ll leave that out of the story. Also not mentioned are the game-consoles and servers, but they did have big influences on the home-PC market.

In the picture at the right you see an idea of how fast the markets would have grown from a 2006 perspective. (Click on it for the full report). You see that the explosive growth of smartphones was not expected; the other detail is that the cloud also was not foreseen here.

After reading you understand why Nvidia focuses so much on HPC and mobile.

After IBM-compatible clones took over from Apple, Atari and ZX Spectrum, we just got used to that a PC is an X86 with MS Windows and Office on it. Around a decade ago Apple fought back with OSX on which Windows 7 (launched in 2009) was the first real answer. Meanwhile Apple switched to Intel, since IBM was not fast enough with the development of the POWER-processor – a huge operation, which seemed a one-time-only step for Apple at the time. SemiAccurate now speaks of Intel being replaced by ARM on Apple’s laptops.

A few weeks ago I asked Computer Science students if they knew ARM. Not even 1% had heard of it, but lots more knew there was a Samsung-chip in their smartphone. So what’s going on without us knowing it?

I’ll try to describe the market for a few key-years and then try to put the big names in it. There is a lot going on between i.e. Nvidia, Samsung, Texas Instruments and Imagination Technologies in the ARM-market, but I’ll leave that out of the story. Also not mentioned are the game-consoles and servers, but they did have big influences on the home-PC market.

In the picture at the right you see an idea of how fast the markets would have grown from a 2006 perspective. (Click on it for the full report). You see that the explosive growth of smartphones was not expected; the other detail is that the cloud also was not foreseen here.

After reading you understand why Nvidia focuses so much on HPC and mobile.

2000: a new millennium

X86 is ruling the PC-industry for a decade now. Apple Macintosh was slowly washed out, and Windows Alpha was being abandoned. AMD has the fastest processors on the market, but Intel’s market share is bigger. We still had the 3Dfx Voodoo, which was partly bought by Nvidia soon. The latest games needed the latest CPUs and GPUs to be playable and the latest Windows needed more power too to run smoothly – the waiting time after pressing the F1-key in Windows ME made clear that you needed a new PC or not. This sets the coming years: AMD vs Intel and ATI vs Nvidia to have the fastest hardware, giving new potential every year.

2006: changes initialise

For years not really much happened: AMD had the faster (and hotter) CPUs, Intel’s market share was still bigger. The competition between these two where fought on all fronts, as you can read all over the web. NVIDIA and ATI had alike battles, but their processors are getting so complex that there is no way possible competitors could in between with a surprise-product (as Intel found out with Larrabee two years later).

Intel introduces the Core-processor, which beats AMD after years. New Apple MACs run on Intel processors, abandoning IBM. AMD buys ATI, which seems to be a big mistake in the beginning (as combining the two companies was not easy), but as we see now with their Fusion it is the best choice ever.

2007: a simple market

	CPU		GPU
	Low-power	Desktop	onboard	Card
Intel	Atom	Core	GMA	–
AMD	–	Athlon/Phenom	Radeon	Radeon
Nvidia	–	–	GeForce M	GeForce

You see a different market than we have now: the idea of low-power computers just starts to enter the X86 camps.

2008: the market loses it’s drivers

The iPhone and GPGPU were released a year before, starting the big revolutions we now encounter. Nvidia started CUDA, ATI released Stream (after Close-To-Metal). Intel Atom was introduced to answer demands for netbooks. At the end of the year the first Android-phone introduced by HTC, just in time to compete with the iPhone.

What changed here is that for most ordinary users a budget-PC was fast enough, putting pressure on PC-sales and profits. Along with that you see an increasing demand for low-power devices. For GPUs there was always the gamer’s high-end cards, and (onboard) graphics for normal users, but the latest games worked actually pretty well on older GPUs. The way DirectX 11 got adopted makes it even more clear that the latest GPUs were more a nice-to-have than mandatory. Microsoft Windows’ “hardware selling magic” to boost PC-sales also lost its speed, since Microsoft was more and more forced to listen to their customer’s demands after Vista, instead of helping out their hardware-partners.

Both ATI and NVIDIA saw this coming and started GPGPU (ATI was actually first in GPGPU with their CTM). The fusion of AMD and ATI took a lot of energy and set them in a lower gear for years, as Fusion-processors were announced in 2006 already. Read this blog about GPGPU in February 2007 to get an idea.

2010: still X86 everywhere

Sales of ARM rise, as tablets and smartphones replace more and more functionality of the PC. Intel tries to enter the smartphone-market with Moorestown and focuses on a promising netbook-market. Linux sees a future on those netbooks, but Microsoft fights back hard to get Windows on all netbooks.

There is much debate how big the space for smartphones, tablets and netbooks really are. Do most people really just want to check their mail and surf the web?

2011: a very, very, very broad market

	CPU		SoCs		GPU
	Netbook	Desktop	Hybrid X86	Smartphone	Smartphone	onboard	Card
Intel	Atom	Core	Sandy Bridge	Atom Z-series	–	GMA	–
AMD	–	Athlon/Phenom	Fusion	–	–	Radeon	Radeon
Nvidia	–	–	–	–	Tegra	GeForce M	GeForce
ARM	–	–	–	X	X	–	–

Here you see where which company currently is. “ARM is not 64bit/fast enough and does not have enough cores”, while the 2GHz dual-cores and quad-cores are already announced, 64 bit is designed.

2012: ARM bigger than X86

Sales of X86 and Smartphones per year (2011 and 2012 by Gartner)

At the ARM-side we now have Windows on ARM, ARM-based iBooks(?), Apple iOS, Google Android and Linaro (Linux on ARM, including Ubuntu). Android is probably still the biggest smartphone OS, competing with iOS for tablets. Now .NET really falls in place, since most software is easy to port now (Apple, beat that!).

If you read “Why AMD Should Buy ARM Now” you see that ARM is seen as dangerous for Intel and even more for AMD. We have seen that in only 3 years time loads and loads of software can be created for new platforms, so the software goes where the money is.

	CPU		SoCs		GPU
	Netbook	Desktop	Hybrid X86	Smartphone	Smartphone	onboard (?)	Card (?)
Intel	Atom	Core	Ivy Bridge	Atom Z-series	–	GMA	–
AMD	Fusion	Fusion	Fusion	–	–	Radeon?	Radeon?
Nvidia	–	–	–	Denver (ARM)	Tegra	GeForce M	GeForce
ARM	X	X	– no need	X	X	– no need	– no need

See where ARM is. Onboard and GPU-cards become unnecessary, pushing Nvidia out of the X86 consumer-world, going to ARM and HPC.

A last word

The big reason why X86 did not succeed to stop the rise of ARM is not the power and specifications – which Intel has shown to be able to provide. It is the freedom. With the rise of Asia and the demands for flexibility by Google and Apple, there was no interest to buy products, and ARM was exactly the thing all wanted. It’s like having the choice between ordering 20 cakes and buying the recipe.

You see that Nvidia’s move to ARM was not a bad choice. I don’t want to drop to conclusions now, but I’m curious which conclusions you make from this data.2000: a new millennium

X86 is ruling the PC-industry for a decade now. Apple Macintosh was slowly washed out, and Windows Alpha was being abandoned. AMD has the fastest processors on the market, but Intel’s market share is bigger. We still had the 3Dfx Voodoo, which was partly bought by Nvidia soon. The latest games needed the latest CPUs and GPUs to be playable and the latest Windows needed more power too to run smoothly – the waiting time after pressing the F1-key in Windows ME made clear that you needed a new PC or not. This sets the coming years: AMD vs Intel and ATI vs Nvidia to have the fastest hardware, giving new potential every year.

2006: changes initialise

For years not really much happened: AMD had the faster (and hotter) CPUs, Intel’s market share was still bigger. The competition between these two where fought on all fronts, as you can read all over the web. NVIDIA and ATI had alike battles, but their processors are getting so complex that there is no way possible competitors could in between with a surprise-product (as Intel found out with Larrabee two years later).

Intel introduces the Core-processor, which beats AMD after years. New Apple MACs run on Intel processors, abandoning IBM. AMD buys ATI, which seems to be a big mistake in the beginning (as combining the two companies was not easy), but as we see now with their Fusion it is the best choice ever.

2007: a simple market

	CPU		GPU
	Low-power	Desktop	onboard	Card
Intel	Atom	Core	GMA	–
AMD	–	Athlon/Phenom	Radeon	Radeon
Nvidia	–	–	GeForce M	GeForce

You see a different market than we have now: the idea of low-power computers just starts to enter the X86 camps.

2008: the market loses it’s drivers

The iPhone and GPGPU were released a year before, starting the big revolutions we now encounter. Nvidia started CUDA, ATI released Stream (after Close-To-Metal). Intel Atom was introduced to answer demands for netbooks. At the end of the year the first Android-phone introduced by HTC, just in time to compete with the iPhone.

What changed here is that for most ordinary users a budget-PC was fast enough, putting pressure on PC-sales and profits. Along with that you see an increasing demand for low-power devices. For GPUs there was always the gamer’s high-end cards, and (onboard) graphics for normal users, but the latest games worked actually pretty well on older GPUs. The way DirectX 11 got adopted makes it even more clear that the latest GPUs were more a nice-to-have than mandatory. Microsoft Windows’ “hardware selling magic” to boost PC-sales also lost its speed, since Microsoft was more and more forced to listen to their customer’s demands after Vista, instead of helping out their hardware-partners.

Both ATI and NVIDIA saw this coming and started GPGPU (ATI was actually first in GPGPU with their CTM). The fusion of AMD and ATI took a lot of energy and set them in a lower gear for years, as Fusion-processors were announced in 2006 already. Read this blog about GPGPU in February 2007 to get an idea.

2010: still X86 everywhere

Sales of ARM rise, as tablets and smartphones replace more and more functionality of the PC. Intel tries to enter the smartphone-market with Moorestown and focuses on a promising netbook-market. Linux sees a future on those netbooks, but Microsoft fights back hard to get Windows on all netbooks.

There is much debate how big the space for smartphones, tablets and netbooks really are. Do most people really just want to check their mail and surf the web?

2011: a very, very, very broad market

	CPU		SoCs		GPU
	Netbook	Desktop	Hybrid X86	Smartphone	Smartphone	onboard	Card
Intel	Atom	Core	Sandy Bridge	Atom Z-series	–	GMA	–
AMD	–	Athlon/Phenom	Fusion	–	–	Radeon	Radeon
Nvidia	–	–	–	–	Tegra	GeForce M	GeForce
ARM	–	–	–	X	X	–	–

Here you see where which company currently is. “ARM is not 64bit/fast enough and does not have enough cores”, while the 2GHz dual-cores and quad-cores are already announced, 64 bit is designed.

2012: ARM bigger than X86

Sales of X86 and Smartphones per year (2011 and 2012 by Gartner)

At the ARM-side we now have Windows on ARM, ARM-based iBooks(?), Apple iOS, Google Android and Linaro (Linux on ARM, including Ubuntu). Android is probably still the biggest smartphone OS, competing with iOS for tablets. Now .NET really falls in place, since most software is easy to port now (Apple, beat that!).

If you read “Why AMD Should Buy ARM Now” you see that ARM is seen as dangerous for Intel and even more for AMD. We have seen that in only 3 years time loads and loads of software can be created for new platforms, so the software goes where the money is.

	CPU		SoCs		GPU
	Netbook	Desktop	Hybrid X86	Smartphone	Smartphone	onboard (?)	Card (?)
Intel	Atom	Core	Ivy Bridge	Atom Z-series	–	GMA	–
AMD	Fusion	Fusion	Fusion	–	–	Radeon?	Radeon?
Nvidia	–	–	–	Denver (ARM)	Tegra	GeForce M	GeForce
ARM	X	X	– no need	X	X	– no need	– no need

See where ARM is. Onboard and GPU-cards become unnecessary, pushing Nvidia out of the X86 consumer-world, leaving ARM and HPC.

A last word

The big reason why X86 did not succeed to stop the rise of ARM is not the power and specifications – which Intel has shown to be able to provide. It is the freedom. With the rise of Asia and the demands for flexibility by Google and Apple, there was no interest to buy products, and ARM was exactly the thing all wanted. It’s like having the choice between ordering 20 cakes and buying the recipe.

You see that Nvidia’s move to ARM was not a bad choice. I don’t want to drop to conclusions now, but I’m curious which conclusions you make from this data.