Intel: Steer Clear Of Our Patents (axios.com) 87
An anonymous reader writes: Intel posted a long blog post yesterday touting the success and evolution of its 40-year-old x86 microprocessor -- the one that powered the first IBM personal computer in 1978 and still powers the majority of PCs and laptops. But it wasn't just a stroll down memory lane. Intel ended the post with a reminder that it won't tolerate infringement on its portfolio of patents, including those surrounding x86. The company wrote, "Intel invests enormous resources to advance its dynamic x86 ISA, and therefore Intel must protect these investments with a strong patent portfolio and other intellectual property rights. [...] Intel carefully protects its x86 innovations, and we do not widely license others to use them. Over the past 30 years, Intel has vigilantly enforced its intellectual property rights against infringement by third-party microprocessors. [...] Only time will tell if new attempts to emulate Intel's x86 ISA will meet a different fate. Intel welcomes lawful competition, and we are confident that Intel's microprocessors, which have been specifically optimized to implement Intel's x86 ISA for almost four decades, will deliver amazing experiences, consistency across applications, and a full breadth of consumer offerings, full manageability and IT integration for the enterprise. However, we do not welcome unlawful infringement of our patents, and we fully expect other companies to continue to respect Intel's intellectual property rights. Also read: Intel Fires Warning Shot At Qualcomm and Microsoft Over Windows 10 ARM Emulation.
Like the AMD-64 instruction set? (Score:5, Interesting)
Re:Like the AMD-64 instruction set? (Score:4, Interesting)
Re: (Score:2)
i686 and x86-64 are not "a 40-year-old architecture".
Re: (Score:1)
Re: (Score:2)
The former also got SSE2 grafted onto it, which is newer than 20 years old and on which Windows 8.1 and 10 depend. Besides, application publishers are dropping support for the former.
Re: (Score:2)
Re:Like the AMD-64 instruction set? (Score:5, Interesting)
Intel never hauled AMD to court for AM64. At the time, Intel was hoping to break clean of x86 by introducing the Itanium, while AMD took the tack of extending the x86 instruction set to 64-bit, something that Intel desperately wanted to avoid. In short, Intel tried to shut AMD out of the market the innovative, rather than the legal way: it just happened that VLIW, or EPIC, was such a bust that even Linux hated it, while AMD scored a coup in the market.
After that, Intel tried coming up w/ their own 64-bit extension to the x86, but Microsoft, which by then had already sunk effort into making 64-bit versions of Windows XP based on AMD, made it clear to Intel that they were not gonna support 2 different x86 instruction sets. This was similar to what Microsoft had done in the past, when they forced AMD, Cyrix, Centaur and Winchip to agree on multimedia extensions. Once Intel got this message, they realized that the only clean way of doing this was doing a cross licensing agreement w/ AMD. There are a couple of instructions in the Intel-64 instruction set that are a tad different from AMD64, but otherwise, they are identical.
At any rate, the biggest thing Intel demonstrated was that having sheer expertise at fabs and manufacturing capacity beat the crap out of any inherent architectural superiority any competitor might have: that's how they felled every RISC rival that they had. Like the Alpha & the PA-RISC was way superior to them, but once they could pack 2 or more cores in a package, along w/ the Windows NT kernel being the unified basis of all Windows OSs, it was easy to catch up w/ them from a stance of price points. Intel doesn't have to sue anybody to preserve x86: worst case, they could simply start manufacturing Snapdragons or A10s or whatever, and horn in on the action
Re: (Score:2)
Intel did have battles w/ AMD in the past over x86, whenever AMD tried making 386s, 486s & the like. At that time, the CPUs that AMD made were similar to Intel's, but slower. The first time AMD tried anything new was when they acquired NexGen, but they were still way behind Intel in that they lacked an FPU. It was only after Dirk Meyer's Alpha team from DEC joined them that AMD really had a quality product in the Athlon, and on that one, Intel didn't sue them.
I also disagree about your second poi
Re: (Score:2)
And so, going down the same rabbit hole, for every claim on an unexpired patent, there'll be a work-around. Instead of fighting all of this, Intel could just take ARM licensing and make their own silicon and oh yeah, have to deal with competition.
If Microsoft can turn 180 in a decade on Linux, Intel can turn in five years to ARM. The Atom family has been cute, but does Intel have even a tiny fraction of the smartphone and tablet market? So how did that work for ya, Intel? Hmmmm?
I personally think that proce
Re: (Score:2)
Well, Intel had the StrongARM processor, which they got from DEC, rebranded as XScale, and then later sold to Marvell. Like you say, there's nothing to stop them from licensing it from ARM. Heck, since Apple & Qualcomm are both fabless, they can cut deals w/ them to put them on their 14nm process and give them some of the best performing energy saving silicon.
The argument here is not RISC vs CISC, and hasn't been for 2 decades. The problem is a simple running code written in the instruction set of
Re: (Score:2)
No one's analyzed the emulator code to my knowledge, to understand if it's indeed an x64/AMD instruction set.... but we know that there is a 32-bit variant, and so, whose x86 is it, anyway?
How one arrives at an instruction execution matters very much in terms of predictability, state machine, even mundane stuff like bus timings and core cohesiveness/thread control/state.
I don't think Microsoft will be the litigant... and I believe Intel will try to cow its OEMs. And I believe they'll fail, having held so ma
Re: (Score:2)
IIRC, Intel-AMD cross licensing dates back to the 1980s when major PC makers insisted that there be a second source for Intel CPUs. Otherwise they would design their hardware around a competitor's (e.g. Motorola) chips.
Re: (Score:2)
Re: (Score:2)
You're thinking 2000. The original cross licensing was around 1980 when Intel was a much less dominant company and failure was still a (remote) possibility. I don't recall the full details of why the cross licensing was continued into the 21st century. Antitrust concerns and customer pressure? By "customer" I mean the likes of Dell and Compaq, not consumers. No one cared much about consumers back then any more than they do today.
Re: (Score:2)
MIPS used patents to discourage clones of its processors.
https://en.wikipedia.org/wiki/... [wikipedia.org]
https://en.wikipedia.org/wiki/... [wikipedia.org]
Re: (Score:2)
I yield to no man in my ignorance of IP law, but I'm pretty sure you can't patent an instruction set. You can patent the implementation hardware, or at least a lot of it, but I doubt that restricts emulation. And I'm pretty sure that you can copyright microcode. Patents are only 20 years BTW. OTOH, copyrights last for all eternity (OK, OK ... maybe only a century or two).
Re: (Score:2)
There has been some activity on the front in the past with MIPS and the patents they used to prevent clones.
https://en.wikipedia.org/wiki/... [wikipedia.org]
Re: (Score:1)
Re: (Score:2)
Intel invests enormous resources to advance its dynamic x86 ISA
Looks like that got typo'd. Should have read:
Intel invests enormous resources to protect its cash cow x86
16 years... (Score:1)
The patents on x86 have surely expired by now.
SSE is still patented (Score:5, Informative)
Patents last 20 years after filing.* Most x86 programs nowadays rely on "i686" instructions introduced with the Pentium Pro (1995) and Pentium II (1997), whose patents have presumably expired just recently, and the Pentium III (1999), whose patents still subsist. Furthermore, many application developers have stopped building for i686 protected mode in favor of the newer x86-64 long mode.
* A few U.S. patents filed before mid-1995 and granted after mid-2000 still subsist because they're grandfathered into the pre-1995 rules.
Re:SSE is still patented (Score:5, Interesting)
Re: (Score:2)
Most x86 programs nowadays rely on "i686" instructions introduced with the Pentium Pro (1995) and Pentium II (1997)
Bullshit. Very few new instructions and not all that useful of ones came out with those CPU's.. and that remained true until SSE2.. not even MMX was useful enough for "most" programs use.
The x86 instruction set minus the SIMD stuff was pretty well fleshed out when the 386 hit, which is a couple generations and a full decade BEFORE the cpu's you are talking about.
Probably the most important single instruction added to the x86/x64 legacy in the past 10 years is POPCNT, which isnt Intels either.
Re: (Score:3)
Re:SSE is still patented (Score:4, Informative)
SSE essentially doubled the number of registers, and although none of the new registers are fully general purpose thats a huge win. The SIMD portion not so much (its a rare programmer that will use the SIMD extensions of their compiler,) and it didnt help that Intel screwed up their SSE SIMD adding all these "horizontal" operations that defeat the entire advantage of SIMD if you use them.
Those horizontal operations are convenient, but if you are doing high performance SIMD then the data within one of your wide registers is all going to be exactly the same kind and isnt very suitable for any sort of horizontal operations. A registers of high performance SSE SIMD contains X3:X2:X1:X0 (all the same component of a 4 different vectors), not W0:Z0:Y0:X0 (a complete 4-component vector.) The terminology in the SIMD world for these two views of the registers are Structure of Arrays (SoA) and Array of Structures (AoS).
There isnt a single GPU that offers horizontal operations in its instruction set like Intel gave to SSE, because it defeats the purpose and would destroy the performance.
Re: (Score:2)
The patents are probably almost expiring by now though. Remember the priority dates of Pentium 4 SSE2 patents released in Nov 2000 are probably around 1999 at the latest for example. SSE1 itself is just a subset of SSE2 with only single precision floating point instructions BTW. That itself was released with Pentium III in Jan 1999, meaning the priority date can't be later than 1998.
Re: (Score:2)
The minimum these days is SSE2 and what the Pentium 4 supported. I find many applications and libraries which require SSE2 whether they should need it or not.
Careful not to poke Microsoft (Score:5, Funny)
Yeah, get into a fight with MS - "I don't know why Windows performs so poorly on your newest, highest-margin chip. Maybe because we had to disable certain compiler options that infringed on your patents. Everything works full-speed on the AMD chips, though. Weird."
Re:Careful not to poke Microsoft (Score:4, Insightful)
The reason we will not see patent reform is that situations like you describe don't normally happen.
Big companies like MS and Intel can do patent licensing exchanges that cover both parties.
It's the little guys, who don't have the patent portfolio or deep pockets to go to war with, that get stomped on like bugs by the big players and preyed on by the trolls.
Re: (Score:2)
The Java Trap (Score:5, Interesting)
Then I guess we can now consider the x86 and x86-64 instruction sets subject to what Richard Stallman has referred to as the Java Trap [gnu.org]. A free program with proprietary dependencies is trapped, and Intel is asserting that the x86 and x86-64 instruction sets are proprietary.
Speak softly and carry a big stick... (Score:2)
Are intel's patents and legal team so pathetic that they need to resort to twitter-level insults?
(* penis penis peeeennniiisss)
Re: (Score:2)
Software patents aren't what they used to be (Score:4, Insightful)
If Intel was picking on little guys, maybe they'd curl up in the corner. Hard to see it in this case.
Honey Badger (Score:2)
China don't care. Neither do consumers, for that matter. Given the length of court cases, if I was an Intel competitor I'd be awfully tempted to do what many startups do: steal everything and hope the resulting lawsuit harmlessly drags out until after you've cashed out. With that in mind...what's the point of posting your legal policy in some summer intern's blog? (If I was Intel, I'd be working with my
Re: (Score:2)
Re: (Score:2)
China cared in the past. Why would this be any different?
https://en.wikipedia.org/wiki/... [wikipedia.org]
Déjà vu (Score:2)
Where have I seen this before? Oh yeah, Slashdot this morning.....
https://games.slashdot.org/sto... [slashdot.org]
Missing closing quote (Score:2)
warning: missing terminating " character
error: missing terminating " character
error: expected primary-expression before 'return'
Editors--that makes it hard to read, because I can't trust when Intel's quote ends so I don't know if it's Intel speaking or editorializing for any given sentence.
The x86 did not power the first IBM PC (Score:2, Informative)
The first IBM PC was release in 1981 with an 8088 processor and optional 8087 math co-processor. While I may be wrong on the date, I am sure of the CPU because I have one of the original system right here.
Re: (Score:3)
The 8088 is an x86 CPU, released in 1979. It's an 8086 (released in 1978) with an halved data bus (8-bit instead of 16-bit). Or maybe were you thinking of the IA-32, introduced with the 80386?
Re:The x86 did not power the first IBM PC (Score:5, Informative)
The first IBM PC was release in 1981 with an 8088 processor and optional 8087 math co-processor. While I may be wrong on the date, I am sure of the CPU because I have one of the original system right here.
It's semantics really. The 8088 was a cheaper version of the 8086.....it used the 16-bit x86 instruction set, but the less expensive 8-bit data bus.
https://en.wikipedia.org/wiki/... [wikipedia.org]
Cease and Desist (Score:1)
I hold the patent for duping articles on the front page
I like where this is going. (Score:5, Insightful)
What this really shows is that Intel is incredibly insecure because they are highly vulnerable. Intel doesn't really have a leg to stand on regarding patents for x86 so they are just lashing out and hoping to scare off people. They are reverting to their anti-competitive nature because they are now losing on both in the server market (due to AMD's Zen arch) and if Microsoft doesn't blow it, the commodity Desktop market could go to ARM. Intel has really earned this fate and I know they will break the law repeatedly to avoid it. They are getting their just deserts. :)
Re: (Score:2)
Intel's fab costs per product are probably rising at a terrifying speed if they can't get maximum volumes. ARM licensees and other fabless companies are able to spread the load with the rest of the industry.
That is true on the manufacturing side but the various ARM licensees also duplicate development so it is not Intel facing off against the ARM juggernaut; instead it is the Intel ox facing a flock of ARM chickens which duplicates the situation when Intel was facing off against the various RISC workstation and server processors except back then, development costs were lower for RISC developers which is not the case for ARM designs if they want to compete with Intel.
On the manufacturing side itself, Intel has
Re: (Score:2)
Zen is nice, but the reality is it's pretty much only on par with Intel parts from 2010.
lol
Re: Isn't emulation technically legal? (Score:3)
Game console emulation is a quagmire unto itself.
NES/SNES emulation is "legal" in that you're hard pressed to find a for-profit company coding and selling one, so the most they could do is send a C&D order to Sourceforge or Github to kill the downloads - trying to file a lawsuit against an OSS developer would be pretty fruitless. Nintendo is far more apt to go after ROM hosting sites.
PSX emulation (and many after) rely on a BIOS file that's copyrighted. Emulators can exist, but they're useless without t
Dupe (Score:3)
Translation: (Score:4, Informative)
"The market for new CPUs hasn't been so hot in the last few years, ARM processors are becoming more and more popular, and AMD is starting to bring stiff competition again, so we're going to become patent trolls now to make up for all that lost income. So beware!"
Re: (Score:2)
I'm not an AMD fanboy either, so you can take your strawman elsewhere. I didn't give AMD credit for anything other than the release of competitive chips (after many years of not being so competitive).
Looks like Intel is specifically worried about ARMs eating desktop market share:
https://www.theregister.co.uk/... [theregister.co.uk]
Re: (Score:1)
Obligatory:Intel CPU Backdoor Report (May 5 2017) (Score:4, Interesting)
The goal of this report is to make the existence of Intel CPU backdoors a common knowledge and provide information on backdoor removal.
What we know about Intel CPU backdoors so far:
TL;DR version
Your Intel CPU and Chipset is running a backdoor as we speak.
The backdoor hardware is inside the CPU/Bridge and the backdoor firmware (Intel Management Engine) is in the chipset flash memory.
30C3 Intel ME live hack:
[Video] 30C3: Persistent, Stealthy, Remote-controlled Dedicated Hardware Malware [youtube.com]
@21:43, keystrokes leaked from Intel ME above the OS, wireshark failed to detect packets.
[Quotes] Vortrag [events.ccc.de]:
"the ME provides a perfect environment for undetectable sensitive data leakage on behalf of the attacker".
"We can permanently monitor the keyboard buffer on both operating system targets."
Backdoor removal:
The backdoor firmware can be removed by following this guide [github.io] using the me_cleaner [github.com] script.
Removal requires a Raspberry Pi (with GPIO pins) and a SOIC clip.
Decoding Intel backdoors:
The situation is out of control and the Libreboot/Coreboot community is looking for BIOS/Firmware experts to help with the Intel ME decoding effort.
If you are skilled in these areas, download Intel ME firmwares from this collection [win-raid.com] and have a go at them, beware Intel is using a lot of counter measures to prevent their backdoors from being decoded (explained below).
Useful links:
The Intel ME subsystem can take over your machine, can't be audited [ycombinator.com]
REcon 2014 - Intel Management Engine Secrets [youtube.com]
Untrusting the CPU (33c3) [youtube.com]
Towards (reasonably) trustworthy x86 laptops [youtube.com]
30C3 To Protect And Infect - The militarization of the Internet [youtube.com]
30c3: To Protect And Infect Part 2 - Mass Surveillance Tools & Software [youtube.com]
1. Introduction, what is Intel ME
Short version, from Intel staff:
Re: What Intel CPUs lack Intel ME secondary processor? [intel.com]
Amy_Intel Feb 8, 2016 9:27 AM
The Management Engine (ME) is an isolated and protected coprocessor, embedded as a non-optional part in all current Intel chipsets, I even checked with the engineering department and they confirmed it.
Long version:
ME: Management Engine [libreboot.org]
The Intel Management Engine (ME) is a separate computing environment physically located in the MCH chip or PCH chip replacing ICH.
The ME consists of an individual processor core, code and data caches, a timer, and a secure internal bus to which additional devices are connected, including a cryptography engine, internal ROM and RAM, memory controllers, and a direct memory access (DMA) engine to access the host operating system's memory as well as to reserve a region of protected external memory to supplement the ME's limited internal RAM. The ME also has network access with its own MAC address through the Intel Gigabit Ethernet Controller integrated in the southbridge (ICH or
Re: (Score:2)
Intel vs...... the world? (Score:2)
Dupe much? (Score:2)
BeauHD literally just posted this story this morning.
https://games.slashdot.org/sto... [slashdot.org]
Intel DOES have a leg to stand on (Score:2)
First the usual IANAL.
But I am an electornics engineer (but not in Utah).
If you recall, Microsoft and Qualcom are using emulation to run 32bit apps. Therefore IA-32.
IA-32 is firmly intel patented. These patents are licensed to AMD under a variety of agreements, on in particular in 2001. The other side of the coin is that Intel Licenses a big chunck of AMD's AMD-64 ISA, which is not emulated by Microsoft-Qualcom.
Therefore, any claim over the use of IA-32 has to be done by intel (AMD would be more than happy
Re:Intel DOES have a leg to stand on (Score:4, Informative)
Apparently also hasn't been following Supreme Court patent jurisprudence. If Microsoft is emulating these instructions with software and a general purpose computer, there is a good chance that Microsoft's actions will be found non-infringing.
It may be a closer call if instructions were added to make the emulation easier, though.
Re: (Score:3)
IA-32 is firmly intel patented.
The IA-32 chip designs -- the way instructions get implemented on a chip are the Intel patent.
What you cannot patent, or what can be challenged - is the particular user interface.
For example, you cannot patent the notion of "ADD Instruction" with such and such hex code, that takes 3 registers as input, that is written like this [example example example]; the invention would be a specific machine implementation or specific logic circuit which processes this instruct
what is a set of instructions? (Score:3)
One may argue that at this point it is a simply of specification of interface interpreted by microcode, and interface is not covered by IP, at least every USA court of law rejected any such case. In case of instruction set what matters is its implementation on a microcode level, and of course Qualcomm would implemented it in a way completely independent from the way Intel does. IMHO, Intel would have a hard time proving it otherwise.
Translation (Score:2)
"Our company now employs more attorneys than engineers. Our product pipeline is pretty much tapped out. We got nothin'."
Network effect (Score:2)
Patent protection for x86 ISA show how the patent system is broken. x86 ISA is not innovative nowadays, and the number of transistors required to decode x86 instructions makes x86 ISA a technical liability
The real value of x86 ISA is compatibility, with a lot of programs built for it. That means patent here protect a network effect and not innovation.
Everyone is already staying away from Intel (Score:2)
Every new product has been using ARM, which offers convinient licensing not only of ABI but actual silicon designs. They still have server market, but ARM will pop up there too as soon as they are dumb enough to jack up prices. New performance war is anyway in GPUs and now TPUs. Don't care, LOL, bye!