In 2011 AMD released the Bulldozer architecture, with a somewhat untraditional implementation of the "multicore" technology. Now, 4 years later, they are sued for false advertising, fraud and other "criminal activities". From TFA:
In claiming that its new Bulldozer CPU had "8-cores," which means it can perform eight calculations simultaneously, AMD allegedly tricked consumers into buying its Bulldozer processors by overstating the number of cores contained in the chips. Dickey alleges the Bulldozer chips functionally have only four cores—not eight, as advertised.
Related Stories
AMD agrees to cough up $35-a-chip payout over eight-core Bulldozer advertising fiasco
AMD has agreed to pay purchasers of its FX Bulldozer processors a total of $12.1m to settle a four-year false advertising lawsuit.
Considering the number of processors sold and assuming a 20 per cent take-up by eligible purchasers, that works out to $35 a chip, the preliminary agreement argues: a figure that is "significantly more than 50 per cent of the value of their certified claims had they prevailed at trial."
It's a good deal, the agreement [PDF] explains, because of the "risks and expenses that further litigation would pose in this case."
The chip giant advertised its processors as being the "first native 8-core desktop processor" and charged a premium for it. But a significant number of those purchasers were then surprised to find that the chip did not contain eight fully independent, fully featured processing units but rather four Bulldozer modules that each contain a pair of fully fledged instruction-executing CPU cores.
A final nail in the module coffin.
Previously: AMD Sued by Customer Over Misrepresentation of "Multicore"
When is a CPU core not a CPU core? It's now up to a jury of 12 to decide.
(Score: 2) by Whoever on Saturday November 07 2015, @03:16AM
Does that mean the my 3-core processor really has 1 1/2 cores?
(Score: 5, Funny) by M. Baranczak on Saturday November 07 2015, @03:22AM
(Score: 5, Interesting) by Hairyfeet on Saturday November 07 2015, @04:31AM
No it means this dumbass doesn't know anything about CPU arches. I sell AMD exclusively in the shop and have read extensively about how the BD/PD arch works. What you have is a PAIR of integer cores and the FPU can be ONE 256bit FPU (for AVX) OR, and here is the key point, it can be used as TWO 128bit FPUs.
So each module is 2 integers with 2 128bit FPUs that can be joined as a single 256bit if you require AVX. If the jobs you are doing is primarily multicore aware, which just FYI if it was a "fake core" the opposite would be true you will have FASTER performance with the AMD than you will if its a single core process, which Intel chips are better at...that's it, that is all the evidence you need to see its bullshit. Just look at these Linux benchmarks [phoronix.com] and you will see the more cores? The better it runs because these Linux programs can take advantage of more cores. If it were truly a "half core" then you simply wouldn't see those kinds of gains. There are also plenty of benchmarks that shows the FX-6350 is a good match for the Phenom II 1100T X6 [cpuboss.com] which according to their logic is a three core versus a 6 core...yet they are evenly matched? Bullshit.So this is just another case of "lawsuit lotto" by a dumbass that doesn't understand even the most basic of hardware design.
Oh and just FYI, anybody who wants to build a truly awesome PC on a budget? The FX8320E and the FX6300 are the sweet spots at the moment in pure CPU, very fast, great at multitasking, easily OCs if you desire, and dirt cheap. In the APU line the A10-7870K is pretty damned impressive, can play modern games like BF4, and can be had for $127 shipped. Finally if you want an INSANELY cheap HTPC? Grab the Athlon 5350. Quad cores at 2.06Ghz, nice Radeon HD8400 baked in, only uses a max 25w (and according to kill-a-watt they average less than 12w for most tasks) and if you use a Linux OS like OpenELEC you can build a DAMN nice HTPC for less than $150. THIS is why I have no trouble selling AMD exclusively, as they have really nice chips at truly crazy cheap prices.
And I put my money where my mouth is, I'm typing this on my FX-8320E with 16GB of RAM, R9 280 GPU, 3TB HDD with 240GB SSD, BD and DVD burners, Win 7 HP and a Rosewill Thor gamer case...cost? A hair under $700 after MIR. No way in hell you could get that kind of performance for that cost going Intel, no way, my games look jaw dropping purty and it just blows through transcodes and layering effects on my multitrack recordings like it was nothing. I had both the Phenom II X4 (currently used by my wife who is slaughtering a carrier in World Of Warships with it as we speak) and a Phenom II X6 (currently being used by my oldest in War Thunder to grind the American tank tree) which according to this lawsuit are both "superior" CPU due to the design of the FP and...yeah, just no. While they both make for great gaming CPUs the extra cores and higher clocks cut my transcodes and renders damned near in half, does that sound like the performance one would be seeing from "half a core" to you?
ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
(Score: 3, Interesting) by Snotnose on Saturday November 07 2015, @02:20PM
So this is just another case of "lawsuit lotto" by a dumbass that doesn't understand even the most basic of hardware design.
I think it's more like some dude is playing lawsuit lotto hoping he gets a dumbass judge and/or jury.
Interesting post hairyfeet. I've got an A10 myself and am pretty happy with it.
It was a once in a lifetime experience. Which means I'll never do it again.
(Score: 2) by bzipitidoo on Saturday November 07 2015, @03:04PM
I keep a foot in every camp-- in the PC world. (Don't use Macs, but do have a few ARM computers, a Beaglebone and a smartphone, and keep thinking about getting a Raspberry Pi.) When one of these hardware vendors finally releases an open source graphics driver for Linux (or FreeBSD, I'm not picky) with decent 3D acceleration, I'll make that my next computer. After all these years, we're still not quite there. Nouveau still does not have as good 3D acceleration as the proprietary Nvidia driver. On the AMD side, I hear Catalyst is a mess of a proprietary driver, and the Radeon/Mesa driver is only half the speed, and there are quite a few games that don't work on it at all. So I was interested when Intel stepped up their game with their integrated graphics. They used to be so bad that a PC with the previous generation of CPU and 1/3 the clock speed, but with NVidia graphics, would outperform the Intel integrated garbage. Intel's HD series is not glacially slow even at 2D, and can actually compete with old, low end Radeons and Nvidias. I have another PC with an i5-3317u and Intel HD4000 graphics, and it's not bad, a bit better than my aging, main PC with a Phenom II X4 945 and Radeon HD 5450.
One of the things I like a lot is low power consumption. That Intel PC uses 30W max, when something with intense 3D accelerated graphics usage is running, 20W when watching a video, and only 10W when text editing. What does AMD have these days that matches that? (My old AMD takes 116W max, 70W for text editing.) The APU stuff? That Athlon 5350 you mentioned? 25W, you say? Presumably easy to repurpose from home theater to desktop usage. And the Linux Radeon driver, when will it be able to do decently speedy 3D acceleration?
(Score: 2) by Hairyfeet on Sunday November 08 2015, @12:51AM
Here is a Phoronix review using Ubuntu [phoronix.com] and they liked it, and that is a year old, the drivers are even better now. AMD has been paying for extra FOSS devs to support their APUs and most of them run quite nice now, and you can score one of these chips just crazy cheap, we're talking $30 for the dual core Sempron and $54 for the top of the line Athlon 5350. If you want to build a really cheap HTPC that's Linux friendly? Well here ya go. I've been using them at the shop with both windows 8 (frankly the only place windows 8 works is as a 10 foot UI) and OpenELEC (Linux based XBMC/Kodi OS) and they seem to play quite nicely with both. I've also had customers use them as ULP office boxes and they are VERY happy with the performance, for all your basic tasks like web surfing, office work, watching vids? Its smooth and easy and is so low power its practically silent.
ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
(Score: 4, Interesting) by Whoever on Saturday November 07 2015, @03:19AM
What does this mean for people who have paid for licenses on a per-core basis?
(Score: 3, Interesting) by edIII on Saturday November 07 2015, @03:59AM
Wow. Is that an interesting idea or what? Plenty of stuff is licensed that way. If he can prove it only has 4 cores in a court of law, then anybody with licenses just got double what they needed, or a slam dunk lawsuit against AMD for the difference.
It seems logical to me that you would only pay for each literal core, not the extra virtual core from hyperthreading (or similar). If AMD promised a literal 8 cores, but then did some funny 'unusual implementation' where there weren't literally 8 processing cores then this gentleman has them by the short and curlies so to speak.
Ironic that in argument that is sure to be about technicalities, that we have nothing technical yet to argue over. Love to know what they mean specifically. From Wikipedia [wikipedia.org] it doesn't look like 8 cores to me.
If I'm reading the part in bold correct, it does indeed sound like there isn't really 8 cores.
Technically, lunchtime is at any moment. It's just a wave function.
(Score: 2) by frojack on Saturday November 07 2015, @04:36AM
In terms of hardware complexity and functionality, this module is equal to a dual-core processor in its integer power, and to a single-core processor in its floating-point power: for each two integer cores, there is one floating-point core. The floating-point cores are similar to a single core processor that has the SMT ability, which can create a dual-thread processor but with the power of one (each thread shares the resources of the module with the other thread) in terms of floating point performance.
So the upshot of that is if these processors were not used for gaming and complex numerical calculation, and reserved for the server market, there's a good chance no one would ever notice this floating point limitation.
Most of the work done in server situations is integer math, (well, most of it is just byte slinging hither and yon). Encryption may be some of the most taxing work in the server market.
But I have no idea how those processors were marketed.
No, you are mistaken. I've always had this sig.
(Score: 2) by Pino P on Saturday November 07 2015, @03:06PM
Most of the work done in server situations is integer math, (well, most of it is just byte slinging hither and yon).
Unless the server is, say, transcoding uploaded video to fifteen different formats for streaming to viewers. But perhaps a lot of that can be written in OpenCL and run on the integrated GPGPU. Does a Xeon even have an GPGPU?
(Score: 2) by frojack on Sunday November 08 2015, @04:49AM
Yes, but most streaming stuff isn't transcoded from one format to the other every time someone requests a stream.
You do it once, and save the file, then chuck what ever format they ask down the socket as fast as the requester can consume it.
Admittedly, you still have a transcoding task just to arrive at a copy for each format. And maybe these processors do that just fine, and maybe they don't, I donno.
No, you are mistaken. I've always had this sig.
(Score: 2) by Pino P on Sunday November 08 2015, @06:48PM
most streaming stuff isn't transcoded from one format to the other every time someone requests a stream.
If someone is sending a live stream that has few simultaneous viewers, the server might end up serving the transcoded stream at each detail level to one viewer or at most a handful. Even apart from live streaming, I'm told some adaptive streaming platforms do a real-time transcode for a few seconds rather than waiting for the next keyframe to switch detail levels when the Internet connection's throughput changes or when the user fast-forwards or rewinds.
Admittedly, you still have a transcoding task just to arrive at a copy for each format.
Even apart from live streaming, uploaders on big video sharing sites such as Dailymotion and YouTube initiate so many transcoding tasks that I shudder to think of how many must be running at once.
(Score: 5, Informative) by Hairyfeet on Sunday November 08 2015, @01:21AM
You are reading it wrong because you are ignoring this part, bold for highlight.."Two symmetrical 128-bit FMAC (fused multiply–add capability) floating-point pipelines per module that can be unified into one large 256-bit-wide unit if one of the integer cores dispatches AVX instruction and two symmetrical x87/MMX/SSE capable FPPs for backward compatibility with SSE2 non-optimized software."
So each core still has a FPU, it simply has a weaker 128bit FPU that can be combined into a single 256bit FPU if AVX instructions are required. The reason why they did this their engineers have spoken at length about, they believed multicore processing was the future (which it is) and would be upon us as quickly as 64bit computing was (which it wasn't) and so bet on having more cores versus having higher performance per core. If you are like me and are using plenty of multicore aware tasks like transcoding or effects layering? This kicks ass because having high single core performance would be slower than having multicores working on the task, while for someone that used nothing but single process programs it would be a better choice to go for a higher per core performance over having more cores.
So it isn't a "half core", it is simply a different approach to the same task.
ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
(Score: 3, Informative) by edIII on Sunday November 08 2015, @02:05AM
Thanks for the explanation
Technically, lunchtime is at any moment. It's just a wave function.
(Score: 4, Informative) by Runaway1956 on Saturday November 07 2015, @03:23AM
All of my Opterons have shown up in system diagnostics to have the same number of cores that were advertised. All of them have shown clock speeds as advertised. All of them have given benchmarks very similar to those advertised. All of my Opterons have performed as well or better than similarly advertised Intel chips.
Maybe my dual hex-core system only has six real cores, instead of the twelve indicated by the diagnostics. If that be true, then maybe AMD's cores are superior to similar Intel cores?
This looks a little silly at first glance. But, I'll be watching the thread - maybe I'll learn something.
I would be upset if my CPU's were advertised as capable of smoking the very best of Intel CPU's, and they completely failed to do so. But, that is not the case. Every AMD CPU I've owned has roughly matched it's advertised performance levels. (Sure, one runs a wee bit faster, and another runs a wee bit slower, but they are all close.)
ICE is having a Pretti Good season.
(Score: 3, Insightful) by Runaway1956 on Saturday November 07 2015, @03:37AM
I should have noted that monitoring programs actually appear to monitor all 12 of my cores. That is, I don't have six pairs of cores, with the activity of one core slaved to that of another. Each of my 12 cores shows increasing and decreasing usage independent of any other core. We can be reasonably sure that diagnostic program writers aren't in collusion with AMD to disguise the core usage on my computer.
ICE is having a Pretti Good season.
(Score: 3, Informative) by frojack on Saturday November 07 2015, @07:03AM
Monitoring programs?
You know how those really work? There is no set rule you know.
Microsoft [microsoft.com] says it just counts passes through their idle routine. Linux does it differently, watching for time when there no tasks waiting to be dispatched.
But none of those methods mean didly squat when processors can be throttled up or down by a simple voltage change.
Percentage of idle time means nothing when the processor is slowed to just a fraction of its clock speed, and a tiny little routine can use all the available cycles because the processor is running so slowly.
No, you are mistaken. I've always had this sig.
(Score: 2) by opinionated_science on Saturday November 07 2015, @02:02PM
run LINPACK. Very hard to hide maths capability if there is none...
(Score: 3, Informative) by Pino P on Saturday November 07 2015, @03:14PM
An Atom N450 CPU has 1 core and 2 threads in a similar setup, branded "Hyper-Threading Technology" by Intel. The monitoring programs shipped with Xubuntu (top, Xfce4 Task Manager, and Xfce4 CPU Graph Plugin) see it as two cores.
(Score: 1, Informative) by Anonymous Coward on Saturday November 07 2015, @03:56AM
I don't even have to RTFA to tell you what he's complaining about.
To put it simply, the bulldozer arch has 1 floating point unit for every 2 integer units. So an 8 core chip can run 8 integer threads simultaneously, but can only do 4 floating point threads simultaneously. If a thread mixes a lot of int and float work, then under pathological conditions scheduling will make it behave like a 4 core system.
(Score: 3, Funny) by maxwell demon on Saturday November 07 2015, @08:49AM
I guess then the original 8086 was a zero-core processor because it didn't have a floating point unit at all. You could buy an external one, the 8087.
The Tao of math: The numbers you can count are not the real numbers.
(Score: 5, Informative) by gman003 on Saturday November 07 2015, @04:05AM
Modern CPU cores ("modern" in this case meaning Pentium Pro / K5 and later) are complicated beasts. Within each core, you have an instruction decoder (which can decode multiple instructions per clock), a few buffers between decode and execute (for instruction reordering), and then a number of execution units, each of which can process only a subset of all possible instructions. Multiple execution units can process instructions simultaneously - originally this was so you could do a store/load while doing an FDIV, but now CPUs will have multiple copies of the same execution unit so you can do several at once. As an extreme example, a current Intel core can do four integer multiplies and four load/stores at once.
While all that horsepower was originally added to speed up a single execution thread, if you have a bad instruction mix (or just stall on memory reads) it makes more sense to allow multiple execution threads to share the core (this is separate from OS-level multithreading, which simply time-slices the CPU to allow different processes full access to the core). Intel was the first to do this (at least in x86 land*), starting in the Pentium 4 era. Their trademarked term for it is "HyperThreading" - they advertise each "core" as merely a core, and list twice the number of "threads" on chips with it enabled. Since they get about a 20-40% performance boost, that's reasonable.
AMD did not implement simultaneous multithreading until Bulldozer**. They really came at it from a different angle - instead of growing one core, they were trying to shrink their cores (so as to pack more onto one chip), and decided to have some execution units "shared" between two cores - the two-"core" component being named a "module". Each module has two decoders (capable of two instructions per clock), four integer units, two load units, two store units, and two floating-point SIMD units (which chain together for 256-bit SIMD). They also have two separate L1 caches and a shared L2 cache (the L3 cache is shared between all modules in one chip).
On the face of it, each Excavator"module" is about on part with a Haswell "core". Same decode rate, about the same execution rate. Excavator is significantly more limited in using its full resources for a single process, though it does take up slightly less die space (or would, if they were on the same process node, but AMD fell way behind on that... they're still putting out 28nm chips while Intel is at 14nm****).
I can't decide if this lawsuit is valid or not. AMD definitely chose their terminology with marketing in mind. The thing they call a "core" is a pretty pathetic core for a desktop CPU - it's about equal to a core in an Apple iPad chip. But on the other hand, it does make sense to call it a "core", given the history. You could just as well argue that Intel's "core" is really two cores. I'm pretty sure it's not enough to sue over, at any rate, but you'll have to find someone else to explain advertising law to you.
* This stuff was used long before on mainframes. Current IBM processors actually do four-way multithreading, running four threads on a single core at once. Oracle's SPARC does a different thing - each "core" processes eight threads, but in lockstep, processing one instruction from each in turn, as a way to reduce the hit from cache misses.
** Quick primer on CPU codenames: Intel throws a new uarch out every year, but they haven't really changed their general design since Nehalem in 2009, they've just been slowly growing them while working mainly on lowering idle power draw. I'm using Haswell as my reference point - there's the newer Broadwell and Skylake, but they haven't unveiled any really deep architectural details. AMD updates far less often, and their last big update was Bulldozer. They followed this with Piledriver, Steamroller and now Excavator. AMD is currently working on a ground-up new design, called Zen, expected sometime next year.
*** Oh, and both Intel and AMD have a low-power architecture as well. Intel has Silvermont, aimed at bottom-end netbooks and top-end smartphones, while AMD has Bobcat/Jaguar/Puma, which is designed more for the laptop market (and was also used in the PS4 and Xb1 game consoles... for some reason). AMD is also working on an ARM64 processor design, aimed at the server market, called "K12", and Intel has a third uarch designed for the embedded market.
**** Don't take the process node names too literally. They measure only the smallest feature - with FinFET in particular, transistors are more rectangular than square, and the longer dimension hasn't been shrinking as much, so Intel doesn't have nearly a 4x lead in transistor density.
(Score: 1, Informative) by Anonymous Coward on Saturday November 07 2015, @03:32AM
Really?
(Score: 5, Funny) by Runaway1956 on Saturday November 07 2015, @03:34AM
Well, yes - costumers would presumably work on Broadway, or in Hollywood. And, they probably use computers. So - a costumer has sued AMD because he could only design 4 costumes on his computer, instead of the 8 he expected.
ICE is having a Pretti Good season.
(Score: 2) by Pino P on Saturday November 07 2015, @03:39PM
costumers would presumably work on Broadway, or in Hollywood.
Someone used this same "customer"/"costumer" misspelling in a solution to this riddle [stackexchange.com]. But what does it say about me that I see "Broadway" and "Hollywood" and immediately think of codenames for the chipset in the Wii game console?
(Score: 0) by Anonymous Coward on Saturday November 07 2015, @05:44AM
(Score: 1, Insightful) by Anonymous Coward on Saturday November 07 2015, @03:45AM
Intel's Xeon has hyperthreading, which means that two threads can progress at times when they require different resources, but in the worst case, the throughput is the same as if there were no hyperthreading.
When you examine the output of /proc/cpuinfo, each hyperthread is counted as a separate processor, which of course is bogus. More marketing bullshit from Intel.
(Score: 2, Interesting) by Anonymous Coward on Saturday November 07 2015, @04:01AM
> but in the worst case, the throughput is the same as if there were no hyperthreading.
I bet I could put together a workload that ran slower with hyperthreading enabled, all you gotta do is thrash the shit out of the cache with aliasing (access memory addresses that are distinct but collide in the cache's hashing function so each thread is constantly evicting the other thread's cache). It is actually a real problem you have to watch out for when doing higher performance computing, happens by accident way more than you might expect.
(Score: 2, Insightful) by robind on Saturday November 07 2015, @09:18AM
To be fair you can always construct a workload that will break a given architecture.
(Score: 2) by sjames on Saturday November 07 2015, @10:02AM
Floating point intensive programs almost inevitably perform better if hyperthreading is disabled.
(Score: 1) by xav on Saturday November 07 2015, @12:25PM
> When you examine the output of /proc/cpuinfo, each hyperthread is counted as a separate processor, which of course is bogus. More marketing bullshit from Intel.
Watch out! /proc/cpuinfo is a marketing weapon in the hands of Intel!
Do you really think that Intel wrote the kernel code that generates /proc/cpuinfo and made in buggy in order to deceive the customer?
You tin-foil hat must be covering your eyes. Look better at /proc/cpuinfo, especially at those "core id" lines.
# fgrep -w -e id -e processor /proc/cpuinfo
(Score: 4, Informative) by Runaway1956 on Saturday November 07 2015, @02:49PM
No - I don't think they did that, but I remember they released a compiler that crippled code when it ran on an AMD processor. That is one of the reasons I don't do Intel.
* Intel Forced to Remove "Cripple AMD" Function from Compiler?
posted by Thom Holwerda on Sun 3rd Jan 2010 20:32 UTC
IconHere's something you probably don't know, but really should - especially if you're a programmer, and especially especially if you're using Intel's compiler. It's a fact that's not widely known, but Intel's compiler deliberately and knowingly cripples performance for non-Intel (AMD/VIA) processors.
Agner Fog details this particularly nasty examples of Intel's anticompetitive practices quite well. Intel's compiler can produce different versions of pieces of code, with each version being optimised for a specific processor and/or instruction set (SSE2, SSE3, etc.). The system detects which CPU it's running on and chooses the optimal code path accordingly; the CPU dispatcher, as it's called.
"However, the Intel CPU dispatcher does not only check which instruction set is supported by the CPU, it also checks the vendor ID string," Fog details, "If the vendor string says 'GenuineIntel' then it uses the optimal code path. If the CPU is not from Intel then, in most cases, it will run the slowest possible version of the code, even if the CPU is fully compatible with a better version."
It turns out that while this is known behaviour, few users of the Intel compiler actually seem to know about it. Intel does not advertise the compiler as being Intel-specific, so the company has no excuse for deliberately crippling performance on non-Intel machines.
"Many software developers think that the compiler is compatible with AMD processors, and in fact it is, but unbeknownst to the programmer it puts in a biased CPU dispatcher that chooses an inferior code path whenever it is running on a non-Intel processor," Fog writes, "If programmers knew this fact they would probably use another compiler. Who wants to sell a piece of software that doesn't work well on AMD processors?"
In fact, Fog points out that even benchmarking programs are affected by this, up to a point where benchmark results can differ greatly depending on how a processor identifies itself. Ars found out that by changing the CPUID of a VIA Nano processor to AuthenticAMD you could increase performance in PCMark 2005's memory subsystem test by 10% - changing it to GenuineIntel yields a 47.4% performance improvement! There's more on that here [print version - the regular one won't load for me].
In other words, this is a very serious problem. Luckily, though, it appears that the recent antitrust settlement between AMD and Intel will solve this problem for at least AMD users, as the agreement specifically states that Intel must fix its compiler, meaning they'll have to fix their CPU dispatcher.
The Federal Trade Commission is investigating Intel too, and it is also seeking a resolution of the compiler issue, but the FTC takes it all a step further than the Intel-AMD settlement. Since the latter only covers AMD, VIA could still be in trouble. Consequently, the FTC asks that Intel do a lot more than what's described in the AMD settlement:
Requiring that, with respect to those Intel customers that purchased from Intel a software compiler that had or has the design or effect of impairing the actual or apparent performance of microprocessors not manufactured by Intel ("Defective Compiler"), as described in the Complaint:
Intel provide them, at no additional charge, a substitute compiler that is not a Defective Compiler;
Intel compensate them for the cost of recompiling the software they had compiled on the Defective Compiler and of substituting, and distributing to their own customers, the recompiled software for software compiled on a Defective Compiler; and
Intel give public notice and warning, in a manner likely to be communicated to persons that have purchased software compiled on Defective Compilers purchased from Intel, of the possible need to replace that software.
Fog also offers up a number of workarounds, such as using GNU GCC, whose optimisations are similar to that of Intel's compiler, "but the Gnu function library (glibc) is inferior". You can also patch Intel's CPU dispatcher - Fog even provides a patch to do so in "Optimizing software in C++: An optimization guide for Windows, Linux and Mac platforms".
This is a particularly nasty kind of anticompetitive practice, as it really requires deep knowledge of matters in order to find it out. God knows how many benchmarks have been skewed in favour of Intel simply because people unknowingly used Intel's compiler in good faith. Intel's compiler is seen as the cream of the crop and delivers superior performance, but apparently only if you stick to GenuineIntel.
e p (11) 123 Comment(s)
http://www.osnews.com/story/22683/Intel_Forced_to_Remove_quot_Cripple_AMD_quot_Function_from_Compiler_ [osnews.com]
(Score: 0) by Anonymous Coward on Saturday November 07 2015, @03:15PM
I can find valid reasons to default to low levels of optimization on another product vs one that is produced by the compiler vendor and if it is documented this isn't inherently a bad thing. It is bad to cheat and lie, it is not bad to clearly state you will optimize conservatively for an architecture that was not produced in house, and it is good to clearly document that behavior.
This does not have to be evil, it could be stupidity.
I would not put it past Intel to try to slide this in and cover it up under a false flag, but I don't necessarily see anything wrong with not fully optimizing for execution on another chip *by default.*
(Score: 2) by zugedneb on Saturday November 07 2015, @06:51PM
I can find valid reasons...
There are none.
Nobody working with consumer products asks a compiler to do work for a specific _brand_.
They want to compile to a specific set of _instructions_.
It is not the job of the compiler writer to be paranoid about some vendor not implementing the instruction set correctly. The job is to make the compiler produce as requested.
old saying: "a troll is a window into the soul of humanity" + also: https://en.wikipedia.org/wiki/Operation_Ajax
(Score: 0) by Anonymous Coward on Saturday November 07 2015, @04:27AM
I'm drunk and finding this hilarious.
Everyone knows Bulldozer is worth half the cores and had known for years.
(Score: 2) by meisterister on Sunday November 08 2015, @12:48AM
My FX has 8 cores. That's final. "Core" is basically just a marketing term generally accepted as meaning "a chunk of execution resources that can independently form its own processor." This flexibility allows everything from an 8088 to an itanic to be called a "core."
What this person is saying here is that only his specific, myopic understanding of a core is what should be legally recognized.
(May or may not have been) Posted from my K6-2, Athlon XP, or Pentium I/II/III.