Slashdot Mirror


AMD Sued Over Allegedly Misleading Bulldozer Core Count

An anonymous reader writes: A class action suit accuses AMD of misleading buyers about the number of cores in its Bulldozer-based CPUs. The complaint claims that the chips effectively had only four cores, while AMD claims there are eight. According to Ars: "AMD's multi-core Bulldozer chips use a unique design that combines the functions of what would normally be two discrete cores into a single package, which the company calls a module. Each module is identified as two separate cores in Windows, but the cores share a single floating point unit and instruction and execution resources. This is different from Intel's cores, which feature independent FPUs. The suit claims that Bulldozer's design means its cores cannot work independently, and as a result, cannot perform eight instructions simultaneously and independently. This, the claim continues, results in performance degradation, and average consumers in the market for a CPU lack the technical expertise to understand the design of AMD's processors and trust the company to give accurate specifications regarding its CPUs."

184 of 311 comments (clear)

  1. So AMD called their Hyperthreading a CPU core? by JoeyRox · · Score: 1, Troll

    Intel has a lot to learn from its smaller rival's marketing department :)

    1. Re:So AMD called their Hyperthreading a CPU core? by Entrope · · Score: 5, Informative

      AMD's CPU architecture has a similar purpose as hyperthreading -- to share hardware resources between what looks to the OS like independent cores -- but the tradeoff is different. Intel's hyperthreading approach only works to cover memory latency, because the hyperthreads share so many physical resources (I think basically everything except register files and hyperthreading-related state). AMD's is somewhat different in that each "module" has two independent integer ALUs, register files, and L1 data caches. The module has one L1 instruction cache, one L2 data cache, one FPU, and one instruction fetch/decode unit.

      But AMD has always been pretty up-front about this architecture. There is maybe a cause of action against resellers who package the AMD chips into systems and do gloss over which aspects each "core" shares with another core, but AMD publicly presented the core-vs-module distinction well before the chips were released.

    2. Re:So AMD called their Hyperthreading a CPU core? by Anonymous Coward · · Score: 1

      So it's sort of a lite, 2:1 version of Sun's cool threads chips from way back?
      Can we measure the number of instructions executed per hardware thread like we could in those Sparc chips?

      That was the only way to know how much work the proc was doing.

      BTW - When has a "core" meant a FPU was included? Don't we have some old precedence there, that they might not? :)

    3. Re:So AMD called their Hyperthreading a CPU core? by JoeMerchant · · Score: 1

      I haven't used a benchmark to do practical work, ever.

    4. Re:So AMD called their Hyperthreading a CPU core? by jedidiah · · Score: 2

      Full disclosure was made.

      It's more like the difference between 1000 and 1024 but actually spelling out which one you are using rather than leaving it for the end user to make an assumption that benefits the seller.

      --
      A Pirate and a Puritan look the same on a balance sheet.
    5. Re:So AMD called their Hyperthreading a CPU core? by gbnewby · · Score: 4, Interesting

      Concur. The design of the bulldozer modules was abundantly clear. The fact that the "cores" share an FPU was clearly disclosed and part of any diagram of the parts. The shared FPUs played a huge role in assessing bulldozer-based CPUs for high performance computing workloads. For HPC, the usual benchmark is HPL (a.k.a., high performance LINPACK), which is a measure of double precision floating point performance for a particular matrix operation called DGEMM. The fact that the FPU was doing double-duty for two "cores" on a module meant that the peak theoretical performance was limited by the number of FPUs in a CPU, not the number of cores or modules or anything else.

      As others have noted, hyperthreading via Intel can have exactly the same impact: the threads share various components, including the FPU.

      Another aspect that can have a major impact on performance is the number of memory channels, and how things like cache coherency is handled. Among other things, AMD's hyper-transport exhibits different scalability characteristics depending on the number of sockets. In a four or eight-socket configuration, latency due to cache coherency operations can have a big impact on performance.
          - gbn

    6. Re:So AMD called their Hyperthreading a CPU core? by Anonymous Coward · · Score: 1

      Intel has been the marketing master for a long time. They lead performance rankings by the simple matter of their own widely used compiler inserting a runtime check for Intel hardware into any generated binary. Background: The x86 architecture has been extended over the years with new instructions, the presence of these instructions can be queried from the CPU, allowing programs to execute optimized code for new CPUs while still supporting older hardware. The check inserted by the Intel compiler pretends that any non Intel hardware only has the basic x86 ( or x86_64 ) instructions.

    7. Re:So AMD called their Hyperthreading a CPU core? by TheDarkMaster · · Score: 1

      The problem is in the details. When you buy a I7 Intel is selling you a processor with 4 cores and hyperthreading. When you buy a bulldozer from AMD it is selling it as 8 cores but it is actually a 4 cores with something similar to hyperthreading. AMD is deceiving the consumer because her design have not really 8 cores, is a 4 cores with hyperthreading.

      --
      Religion: The greatest weapon of mass destruction of all time
    8. Re:So AMD called their Hyperthreading a CPU core? by TheRaven64 · · Score: 1

      It's not quite so clear cut. Both vendors' chips are superscalar and so have multiple pipelines. Intel chips have around 5 independent integer pipelines, plus the AVX ones. Very few workloads will saturate them all. This was the main motivation for IBM's SMT: if you have two threads then you can get a lot closer to saturating the execution units than with one. Of course, you're still likely to suffer a bit because of contention on a few units, which is why Hyperthreading often isn't an overall performance win on Intel.

      AMD implementations add more integer pipelines than any single-threaded workload is likely to be able to use, so they're more likely to get a speedup, but it's still not guaranteed if both workloads are very heavily biased towards a the same few pipelines.

      --
      I am TheRaven on Soylent News
    9. Re:So AMD called their Hyperthreading a CPU core? by TheRaven64 · · Score: 1

      There are two things wrong with this. The first is that no one does comparative benchmarks using Intel's compiler. Intel does, but they don't benchmark AMD chips, so you can't use them for Intel. The second is that a number of other x86 vendors have advertised features in CPUID and then either not implemented them at all, or implemented them in slow microcoded paths. Before Intel's compiler did the vendor ID checking, they were criticised for using features that were fast on their chips and very slow on others.

      --
      I am TheRaven on Soylent News
    10. Re:So AMD called their Hyperthreading a CPU core? by supremebob · · Score: 1

      It really wasn't AMD who was misleading customers, though. Retailers were the ones who were describing the PC's with these processors as something along the lines "blazing fast 8 core gaming monsters" when in reality they were being outperformed by most Core i5's that were available at the time.

    11. Re:So AMD called their Hyperthreading a CPU core? by tepples · · Score: 1

      Where was AMD dishonest that a Bulldozer contains four modules?

    12. Re:So AMD called their Hyperthreading a CPU core? by viperidaenz · · Score: 1

      Intel cores with hyperthreading have two sets of registers, two ALUs and one FPU. They don't pretend it's two cores, nor market it as such

    13. Re:So AMD called their Hyperthreading a CPU core? by wwalker · · Score: 1

      > As others have noted, hyperthreading via Intel can have exactly the same impact: the threads share various components, including the FPU.

      Well, then AMD should advertise its CPUs the same way Intel does: 4 cores, 8 threads. When I needed a CPU that can run 8 processes in parallel at full speed, I looked at Intel offerings and AMD. Compared the number of cores ( thought they were real cores in case of AMD), and the prices, and concluded that AMD was better, because it offered better "number of cores" * "frequency" / price. So I bought the AMD processor. And now I feel I was mislead, or lied to by AMD. And I liked them up until this point. No, I didn't look at "diagrams of the parts", or consult floating point benchmarks, since my particular need wasn't heavy on the floating point calculations (but it uses them somewhat). It just didn't occur to me, I thought I could trust AMD. I thought when it said "8 cores" on the box, it was a processor capable of running 8 processes in parallel at full speed. Was it so wrong of me?!

    14. Re:So AMD called their Hyperthreading a CPU core? by Anonymous Coward · · Score: 2, Insightful

      Was it so wrong of me?!

      Yes

      What answer did you want here. The only people this actually affects, know all about it already.

      Think of it this way. Those trucks with the dualies on them? They have six tires right? Why? More traction and more wight capacity. Even if they only have two axles like the standard four tire variants. So what would be better, a dualie or a truck with three axles so you can still carry that weight? Well you get the dualie. You dont really need another axle you just need capacity and traction. The other axle duplicates parts that you don't need duplicated. Do you really want to maintain extra drive shafts, bearings, differentials?

      It is the same thing with the AMD processor. You do not need extra FPU units, because most people will never benefit from them. So whats better, a hotter, more power hungry, more expensive chip, or one without the parts you don't need saving you money?

      Anyways, they NEVER HID THIS. It is your job to do the research if you need something in particular. Depending on the advertising material for the computer manufacturer or what the dumb kid at best buy tells you makes this 100% completely your bad. If you did research on benchmarks they show you what speed it gets, so are you not getting what you looked up? If it accomplishes it as fast as the benchmarks show, then what are you bitching about? Do you really care how it does this? Fuck I don't care if the processor contains an evil demon that is being forced to accomplish all of this via black magic, it's not like I can fix it if it breaks either way. As long as it accomplishes what it claims to in the amount of time that is similar to the benchmarks, why exactly do you care?

    15. Re:So AMD called their Hyperthreading a CPU core? by viperidaenz · · Score: 1

      They're not upfront about it now

      Have a look at their website
      The FX series is "The industry's only 8-core desktop processor" and "The industry's first and only native 8-core desktop processor for unmatched multitasking and pure core performance with "Bulldozer" architecture"

      http://www.amd.com/en-us/produ...

      Trying searching their website for information on Bulldozer. There's bugger-all.

    16. Re:So AMD called their Hyperthreading a CPU core? by viperidaenz · · Score: 1

      They get those quotes from AMD.com
      "AMD FX Processors unlock maximum, unrestrained processing performance for extreme responsiveness you can see and feel."
      "The industry's first and only native 8-core desktop processor for unmatched multitasking and pure core performance with "Bulldozer" architecture."

    17. Re:So AMD called their Hyperthreading a CPU core? by viperidaenz · · Score: 1

      They share instruction, instruction fetch, branch prediction, instruction decode and dispatch.

      Not to the same extent as hyperthreading, but a cache miss in one core can stall the pipeline of another.

    18. Re:So AMD called their Hyperthreading a CPU core? by tepples · · Score: 2

      Then perhaps AMD, Intel, Microsoft, and Oracle need to sit down and negotiate what constitutes a core for the purpose of licensing proprietary software.

    19. Re:So AMD called their Hyperthreading a CPU core? by Anonymous Coward · · Score: 1

      > I thought when it said "8 cores" on the box, it was a processor capable of running 8 processes in parallel at full speed.

      Define full speed. Unless they perform all their processing in their cache, your computer memory is still a shared resource, and they will slow each other down because there's only so much memory that can be accessed in a unit of time.

    20. Re:So AMD called their Hyperthreading a CPU core? by Anonymous Coward · · Score: 3, Informative

      Intel's cores since Nehalem have nothing of the sort. There's one set of ALUs, one FPU, one front-end, one each of L1i and L1d caches, and one peak of the memory hierarchy per core. However the front-end can decode two threads concurrently, tracking two sets of ISA regs at the same time, and sharing all resources between the two. But there are still ony one set of ALUs, and one FPU, per core.

      Intel gets more oomph per core this way. AMD merely took the other road, but their process tech was behind and their R&D hamstrung by budget cuts, so the things that would've made Bulldozer properly competitive only materialized in Piledriver, and then AMD was firmly in the "cheapo" tier already.

      The "pretend multicore" is a throwback to the first hyperthreaded Netburst family chips, where it was offered precisely as a substitute for AMD's genuine single-chip "SMP in a socket" multicore.

    21. Re:So AMD called their Hyperthreading a CPU core? by hairyfeet · · Score: 3, Informative

      Sigh, does nobody know how the BD/PD arch works here? It CAN share the FPU but absolutely does NOT "have to". Allow me to quote from Wikipedia: "."Two symmetrical 128-bit FMAC (fused multiplyâ"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 DOES have an FPU, they simply used a simpler 128bit FPU that CAN be joined together to make a single 256bit FPU for use with AVX instructions and ya know what? The majority of users will NEVER NOTICE as every program and game Joe Average runs will run great on an AMD chip. I myself run an FX-8320E which I upgraded from a Phenom II X6 (which according to the lawsuit is a "full" core compared to mine) and the FX blows through transcodes and multieffects layering a good 40% faster than the P II X6. And if his BS was true how does he explain the fact that the 6350 and 1100T are equal in performance despite the fact that according to him its a triple core versus a hexacore?

      --
      ACs don't waste your time replying, your posts are never seen by me.
    22. Re:So AMD called their Hyperthreading a CPU core? by petermgreen · · Score: 1

      Meanwhile Intel leads people to believe that "i7s" are the best processors and then slaps that same brand on 4-6 core desktop processors, 2-4 core mobile processors (clocked a bit lower than the desktop ones) and 2 core ultra-mobile processors (clocked much lower than the regular mobile ones) . By doing this they milead customers into thinking they are not sacrificing performance by buying a small/light machine.

      AIUI advertising is about misleading customers as much as possible without stepping over the legal lines. Afacit unless they promote hard numbers or direct comparisions that are demonstratablly wrong they generally get away with it :(. Unfortunately I don't think vauge headlines like "maximum performance" or references to 8 cores (this very discussion on /. shows there is no agreement over what constitutes a "core") are enough.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    23. Re:So AMD called their Hyperthreading a CPU core? by Anonymous Coward · · Score: 1

      You purchased a CPU based on the number of cores instead of the performance or efficiency? Do you purchase a car based on the number of cylinders or actual performance metrics?

    24. Re:So AMD called their Hyperthreading a CPU core? by gweihir · · Score: 2

      It is not hyperthreading at all. It is two full integer CPUs sharing one FPU. Hyperthreading gives you one additional brain-damaged CPU at maybe 40% the main speed. For integer loads, AMD gives you the performance of 8 full cores.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    25. Re:So AMD called their Hyperthreading a CPU core? by gweihir · · Score: 1

      I fully agree. AMD never glossed over anything here and gave all details that matter from the start. This lawsuit has no merit. They are probably hoping that AMD will give them money because it could still take a long time to find that this has no merit (the law not only being blind, but also stupid and slow as a snail).

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    26. Re:So AMD called their Hyperthreading a CPU core? by gweihir · · Score: 1

      It it not. It is 8 integer cores and 4 FPUs. And that is the performance you see. And that is what AMD gave in information.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    27. Re:So AMD called their Hyperthreading a CPU core? by viperidaenz · · Score: 1

      The lawsuit argues that the definition of a core is a set of execution resources inside a CPU that can operate independently from each other.
      Sharing an instruction pipeline does not make it independent, as one core is can only run if the other is also running. A pipeline stall in one core due to a cache miss will stall the other one too.

    28. Re:So AMD called their Hyperthreading a CPU core? by gstoddart · · Score: 2

      So, I actually have an AMD FX 8320E eight core as the processor on my personal desktop.

      This is what AMD says:
      "The industry's only 8-core desktop processor", or "The industry's first and only native 8-core desktop processor for unmatched multitasking and pure core performance with "Bulldozer" architecture".

      Now, I bought it knowing there was likely some behind the scenes tricks, and because I don't strictly need a high-level of sustained CPU intensive tasks. For me it was as much about letting multiple programs run without too much contention.

      However, AMD isn't exactly jumping up and down to point out how this isn't quite 8 full cores.

      So, me I bought it knowing that it wasn't likely to be a full 8-core CPU like the 8-core Xeons we have at work in some of the servers. But I think it's a little much to claim it's purely retailers making this claim. AMDs own marketing material isn't suggesting anything different.

      For my purposes, having bought it knowing it wasn't "really" 8 cores, I'm still happy with the performance. But if you bought something thinking you could have sustained 8-core performance for specific tasks, I can see you'd be unhappy.

      --
      Lost at C:>. Found at C.
    29. Re:So AMD called their Hyperthreading a CPU core? by mattventura · · Score: 1

      It was wrong to buy a CPU without looking at benchmarks first. AMD's IPC is far worse than Intel's, to the point where in many workloads, a 4-core Intel will beat an 8-core AMD even if they didn't have shared FPUs. Remember when AMD had better IPC, and would even rub it in peoples' faces by naming their CPUs after the equivalent Intel Mhz? Well, now it's the opposite.

  2. Extremetech sums it up pretty easily... : by Anonymous Coward · · Score: 5, Informative

    http://www.extremetech.com/extreme/217672-analysis-amd-lawsuit-over-false-bulldozer-chip-marketing-is-without-merit

  3. Re:i5, same thing? by Anonymous Coward · · Score: 2

    An i5 has 4 physical cores without HT, dummy.

  4. Pretty Laughable by TooManyNames · · Score: 5, Insightful

    Why is this even reported? This suit isn't going to go anywhere (unless AMD's lawyers are extremely incompetent, and the judge is extremely incapable of understanding basics about computer architecture and ISAs).

    The AMD cores shared an FPU, sure, but sharing a resource doesn't mean that cores cannot execute simultaneously. The AMD cores still have independent integer-based execution units (instruction registers, register files, ALUs, branch counters, etc.), after all, and are fully capable of executing integer instructions simultaneously (which accounts for the vast majority of instructions under typical loading).

    --
    "Is not a sentence" is not a sentence. Well damn.
    1. Re:Pretty Laughable by Ramze · · Score: 3, Interesting

      I agree. Also, if I understand correctly, that shared FPU can be used either by one CPU core as a 256 bit FPU, or by both simultaneously as 2 independent 128 bit FPUs.

      So, shared, but not likely to be a bottleneck.

      Also, since when do cores have to include all the "extras" ? I recall when 486's had a math co-processor and there were no mmx instructions or other such multimedia or physics sets. This guy is going to have a really tough time explaining how exactly AMD's architecture doesn't provide exactly the number of cores listed -- even if the architecture has its limitations due to sharing resources.

    2. Re:Pretty Laughable by beelsebob · · Score: 2, Informative

      They share *everything* except for an ALU. When 90% of the functionality of the core is not duplicated, it becomes pretty damn difficult to assert that it's actually 2 cores. Instead, it's 1 core with 2 ALUs.

      There are only 4 instruction fetch units on the chip - that suggests 4 cores
      There are only 4 instruction decode units on the chip - that suggests 4 cores
      There are only 4 L1 caches on the chip - that suggests 4 cores
      There are only 4 floating point units on the chip - that suggests 4 cores
      The only thing there are 8 of, is ALUs, but an ALU is not in and of itself, a core.

      The claim that it is 2 cores would be like intel claiming that their chips have twice as many cores, because they have multiple vector floating point units per core.

    3. Re:Pretty Laughable by wwalker · · Score: 1

      Really?! I thought the definition of a "core" was a unit that can run a process independently. Not "integer only" process. Then AMD should advertise it's CPUs as "8 integer cores, but only 4 floating point cores". You know, so that when you buy a CPU expecting 8 fully independent cores, as I did, because it says "8 cores" on the box, you won't be unpleasantly surprised that it can't actually run 8 processes in parallel. I liked AMD up until now, but I think it's going to be my last AMD processor. Sad. :(

    4. Re:Pretty Laughable by whit3 · · Score: 1

      ... I thought the definition of a "core" was a unit that can run a process independently. Not "integer only" process. Then AMD should advertise it's CPUs as "8 integer cores, but only 4 floating point cores".

      Cores don't run processes independently, however. They contend for memory, disk, I/O... You might be expecting something unrealistic, here.

      ...when you buy a CPU expecting 8 fully independent cores, as I did, because it says "8 cores" on the box, you won't be unpleasantly surprised that it can't actually run 8 processes in parallel

      That's why AMD will prevail in this suit. The cores DO, in parallel, run eight processes. Few, if any, situations in a typical computer's workload are FPU-bound, and having eight queues for four FPUs is certainly an improvement over four-core (four queues and four FPUs) systems. That's because there are eight cores.

      A good optimizing compiler that can handle this kind of system might do some novel things, though. A pre-compiled benchmark won't necessarily do justice to this architecture, and that means the trial situation could be a tangled mess. Lawyers for the complainant probably like that.

    5. Re:Pretty Laughable by Impy+the+Impiuos+Imp · · Score: 1

      > I recall when 486's had a math co-processor and there were no mmx instructions or other such multimedia or physics sets.

      This. Floating point was done the hard way with a software library back in the old days, or with a dedicated co-processor if you had the money for it. Multi-core does not need floating point to be considered such.

      --
      (-1: Post disagrees with my already-settled worldview) is not a valid mod option.
    6. Re:Pretty Laughable by jimbo · · Score: 2

      Well, I imagine bigger issues are the single instruction decoder and shared L1 cache. However, as mentioned before here, AMD was completely open about this architecture and this suit should go nowhere.

    7. Re:Pretty Laughable by dshk · · Score: 1

      That is correct, I measured double precision (64 bit) floating point performance on AMD Opterons, and the FPU is not a bottleneck, even in an artificial extreme FPU loading benchmark.

      As far as I know the shared FPU can be divided not only into two 128 but parts, but it can be further divided into four 64 bit parts, therefore a single Bulldozer core can execute two 64 bit floating point calculation at once. This suit is complete bullshit.

      They can as well sue Intel because their CPU has shared third level cache, and they have only a single, shared memory controller.

    8. Re:Pretty Laughable by Anonymous Coward · · Score: 2, Informative

      8 L1 caches and 4 L2 caches.

      4 FPU(SIMD) units. Each unit can break down from 1 256bit FPU unit into 8 32bit FPUs, 4 64bit FPUs, 2 64bit FPUs plus 1 128bit FPU, or any combination you can think of. It's a modular "FPU" that change itself depending on current usage.

    9. Re:Pretty Laughable by gweihir · · Score: 1

      And than you run it single core vs. 8 core for integer benchmarks and get, surprise!, 8x the performance. So your point is?

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    10. Re:Pretty Laughable by davros74 · · Score: 1

      I recall purchasing (actually it was my parents doing the purchasing) back in the day for a dedicated 80387 chip. Mainly so our computer could run Falcon 3.0 in High-Fidelity mode. heh.

      From the OS point of view, module, vs core, vs HT, it doesn't matter. The OS will see each bulldozer or HT core as a "single" core. For some of our HPC machines ($20K), we turn HT off because those extra "cores" confuse the benchmarking/load balancer software because half the cores "aren't real". Also the HT cores share the cache, so effectively jobs run with reduced cache or increased misses (see numastat). Turbo Boost makes it even harder. Have to shop around for the CPUs with the most non-HT cores, which can maintain the highest mulitipliers under full load. Multi-threaded isn't always better if it means slowing each core down by 200-600MHz. So many other things like L1 cache, memory bandwidth, ALUs, etc. that has a bigger impact in the real world than FPU.

      If you depend on FPU a lot, you probably know enough about computer architecture to also know what kind of CPU resources you need for your worksets. The vast masses don't really need FPU and it's the easiest thing to share due to size and lack of need for majority of today's type of computing.

    11. Re:Pretty Laughable by beelsebob · · Score: 1

      Except that you don't - because the lack of cache, instruction fetch, and instruction decode units to keep up with the work load prevents this.

  5. Re:Damnit by Gaygirlie · · Score: 4, Insightful

    The sooner AMD goes down the toilet, the sooner someone competent takes over, the sooner Intel and Nvidia get competitive again.
    At the moment, consumers and society as a whole are suffering from this beating of a dead horse. We need its death accelerated to the whole industry back on its toes again.

    Why do you think AMD going bust would magically mean someone new stepping up to the plate? There are effing ginormous obstacles for a start-up to come and compete in the x86-scene, so much so that it's nearly impossible, and if Intel got a monopoly on the market even for a brief moment the situation would become even worse!

    No, the better option is that AMD gets their shit together, never giving Intel full monopoly on the x86-market even for a bit.

  6. Re:The AMD chip by Anonymous Coward · · Score: 1

    At least their marketing is honest: a bulldozer is a very slow vehicle!

  7. Re:i5, same thing? by Anonymous Coward · · Score: 3, Informative

    i5 is just branding, it doesn't describe physical features. Broadwell desktop i5 has 4 cores no threads, mobile has 2 cores as 4 threads,

    https://en.wikipedia.org/wiki/Broadwell_(microarchitecture)#Mobile_processors

  8. Why did they buy based on "cores"? by rbrander · · Score: 4, Insightful

    I was shopping for VCRs about 20 years back and asked the Future Shop guy how much better it was for having (quoting from the card beside the VCR) a "19 micron tape head". Turns out they ALL had 19 micron tape-heads (whatever the hell that *meant*) as it was the spec for a VCR tape head, at the time, at least. It was just another bit of science-y sounding technobabble to put on the card.
    Buying based on core count is like buying for the 19-micron thing; it's either a fast machine for your purposes or not. Absolutely the only way to tell that for sure is a test. The only thing that was ever useful with, say, "megahertz" was that it had for a decade or so there a correlation with the performance you'd get in real use. I've never found "cores" to have anything of the sort.

    1. Re:Why did they buy based on "cores"? by UnknowingFool · · Score: 1

      The state of consumer CPUs today is that measuring by a single thing like gigahertz or number of cores is not useful. Even from generation to generation of AMD chips, you can't use those to gauge different chips as newer gen 6 core AMD might outperform an older 8 core AMD. There are however a number of CPU benchmarks that can gauge relative performance and the buyer should beware. That being said, for most consumers, they don't even come close to use all the potential of modern CPU.

      --
      Well, there's spam egg sausage and spam, that's not got much spam in it.
    2. Re:Why did they buy based on "cores"? by Ramze · · Score: 1

      Cores mattered when most systems had only 1 core and upgrading to systems with 2 cores or 2 cores with hyperthreading (4 virtual cores) made a huge difference in performance. That was a long time ago, though. The multiple cores meant less latency -- especially for multimedia tasks as the CPU didn't have to stop what it was doing to schedule time for *random background service* or other apps.

      Nowadays, it's as you say -- market droid speak. CPU makers hit a wall with Mhz and needed a new marketing ploy, so they went with cores. Now, even Intel is shying away from that and just going with the cryptic corei3, i5, i7 nonsense -- barely differentiating between the different generations of those..

      number of cores over 4 doesn't really matter unless you're doing virtualization, heavy multimedia processing, or just running tons of programs and services simultaneously that are heavy CPU intensive. I can't imagine a 16 or 32 core machine offering any improvements for my daily usage -- at least not unless a killer app like realistic AI becomes the norm.

    3. Re:Why did they buy based on "cores"? by jetkust · · Score: 1

      The amount of cores is extremely important when you working with multiple processes or multiple threads. And the reason isn't terribly complicated. Though it's possible that a single core processor could be faster than a multi-core processor, in the real world benchmarks of multi-core processors of similar product lines are always faster. In other words, double the cores, and roughly double the performance. Same reason video cards are fast. The amount of cores matters. There's a reason pretty much all "gaming" machines have at least quad core. It's not a superstition.

    4. Re:Why did they buy based on "cores"? by dotancohen · · Score: 3, Insightful

      Wrong.

      99% of software out there is still single threaded. that makes a old pentium 4 single core running at 4GHZ will beat the hell out of a i7 3ghz 8 core processor brutally if that i7 has hyperthreading turned on essentially cutting the processor speeds in 1/2 to emulate more cores.

      To make this clear: last year I wrote Python-based system to interface with Intuit Quickbooks, which the client was running on a 3.4 Ghz dual-core something. He asked what he could do to improve performance as we really were pushing the limit of that poor machine, I told him to get the fastest clock speed processor that he could as Quickbooks is single-threaded, even when in Multi-User mode or whatever they called it. So the expert at the computer shop convinces the client to buy an octocore 2.8 Ghz something (numbers pulled out of ass for example only) which absolutely tanked performance.

      Tools are tools and the best tool for somebody else's job might not be the best tool for _your_ job. Processors or any other complex component cannot be judged on a single merit alone. See how consumers buy their cameras based on megapixel count for example.

      --
      It is dangerous to be right when the government is wrong.
    5. Re:Why did they buy based on "cores"? by Lord+Apathy · · Score: 2

      99% of software out there is still single threaded. that makes a old pentium 4 single core running at 4GHZ will beat the hell out of a i7 3ghz 8 core processor brutally if that i7 has hyperthreading turned on essentially cutting the processor speeds in 1/2 to emulate more cores.

      Truth. If you have been reading the design and theoretical papers about 20 years ago it was widely known the mhz race was going to come to an end in the 3-4 ghz range. An that is pretty much what has happened. I've not seen a real cpu in the field faster than 4 ghz. Most servers that I work with run at 2-3ghz. Sure you see some "gimmicks" that make cpu's run at 5+ ghz but would you truly put one of those 220W beasts in your desktop?

      We have his the limits imposed by the laws of physics. The gate technology just can't switch any faster. It's impossible. We may see some other material like graphine or something in the future but we won't see the performance increase in CPU's that we saw in the late '90s and early 20's.

      Fortunately, modern CPU's are extremely powerful. Not many people realize how powerful a 3.2 ghz quad core cpu is, or even a dual core. Fifteen years ago I made a prediction that 1.5 ghz dual core will be plenty for normal office work. So far my theory has proven to be true. In the office area there was a ramp up to the extremely fast quad core cpu's for general office work. But lately that trend has reversed. Around the offices that I have been in new hardware tends to consist of dual core 1.5 to 2.x ghz CPU's.

      I guess what I'm saying people is get used to what you have because its not going to get much better or faster. Baring any radical advancement in technology like affordable quantum computers, 4 ghz is about it.

      --

      Supporting World Peace Through Nuclear Pacification

    6. Re:Why did they buy based on "cores"? by Anonymous Coward · · Score: 2, Informative

      That's entirely wrong. Pentium 4 have much, much, much worse IPC than Core i7, and so get a lot more done per clock cycle. Have you *used* a 2003-era Pentium 4 lately? Sure, single-threaded performance hasn't improved as much in the last 10 years as it did in the previous 10, but it has still improved a lot.

      Also, hyperthreading does not do anything like "cutting the processor speeds in half". If you are really running single-threaded code, the other thread is just in an idle loop, which uses no CPU resources and has next to no effect on performance. Even if you are running many-threaded code, it still helps a little since it hides memory latency and can make better use of resources if, for example, one thread is doing all floating-point math and the other all integer.

    7. Re:Why did they buy based on "cores"? by wwalker · · Score: 2

      Why did *I* buy based on the number of cores?! Because I needed a CPU that can run as many processes in parallel as possible. Why else would you need multiple cores? No, it's not a web server, and not for bitcoin mining. 8 cores means I should be able run 8 processes in parallel, for 8 times speed-up over 1 core.

      Yes, doing a testing would've been ideal, but you can't return a CPU, as far as I know. Plus, I thought I can trust AMD to sell an 8-core processor when, you know, it says "8 cores" right on the box. Turned out (after I bought the CPU) I can run 8 processes in parallel, but they were about 30% slower each compared to running just 1. Actually, running 4 processes was still fast, but once you went over 4, it started to slow down. For a while I wondered why, thinking it's probably scaling down the CPU frequency because of threat of overheating (max TDP), but now it all makes sense. My program doesn't use floating point values a lot, but it is used throughout the computations here and there.

      I liked AMD over Intel up until now, but I think this is going to be my last AMD processor. Yes, Intel is more expensive, but they don't play shenanigans with redefining what a "core" is.

    8. Re:Why did they buy based on "cores"? by nierd · · Score: 2

      I'm glad you still run programs via a single threaded OS and only ever have one program running on the machine at one time. 99.999% of the computers out there running x86 software use an OS that takes advantage of multiple processor cores to run the vast amount of threads and programs running on the computers at the same time. Typically OS level (1), Antivirus (2), Program (3), Network stack (4), etc. Just saying that's a very minor point - right now I have 140 processes and 2286 threads running concurrently on my machine. Having more cores allows these threads and processes to not block each other when needing resources. Yeah if you have a single threaded program that uses a ton of computing time having something that clocks a bit higher is nice - but spreading your OS load over more cores is never a bad thing and thinking that it's useless is really just wrong if you do anything more with your machine than a single task.

    9. Re:Why did they buy based on "cores"? by jetkust · · Score: 1

      All you have to do is browse the internet with multiple tabs open and multiple cores matter.

    10. Re:Why did they buy based on "cores"? by aaaaaaargh! · · Score: 1

      When you do audio processing, for instance, the number of real cores makes a huge difference. If you use an Intel chip.

    11. Re:Why did they buy based on "cores"? by aliquis · · Score: 1

      In my experience what people ask for is to beat X amount of frames per second in CS:GO.

      Which become the benchmark. And it's not all too good with more than four cores I've got the impression of.

      But gaming benchmarks, especially of the game you want to play, will totally help in deciding what to get for that game =P

    12. Re:Why did they buy based on "cores"? by MightyYar · · Score: 1

      I've never found "cores" to have anything of the sort.

      It depends on the task. Once clock rate is tapped out, all you can do is run more cores. I have some very parallel tasks and some not-very-parallel tasks. The parallel tasks see almost linearly proportional improvement with core count. I've even rented time on Google and Amazon's clouds just to get more cores.

      --
      W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
    13. Re:Why did they buy based on "cores"? by jetkust · · Score: 1

      Though "next to nothing" is an exaggeration, What has happened to Slashdot? The amount of cores obviously makes a difference and they vote up nonsense about how caring about amount of cores means THAT IS ALL YOU CARE ABOUT. Who ever said that in this entire story??????. WHO EVER SAID THAT A RASPBERRY PI 2 IS FASTER THAN EVERY COMPUTER EVER MADE THAT IS NOT 4 CORE??? ............. NOBODY. Just stop.

    14. Re:Why did they buy based on "cores"? by 91degrees · · Score: 1

      Turns out they ALL had 19 micron tape-heads (whatever the hell that *meant*) as it was the spec for a VCR tape head, at the time, at least. It was just another bit of science-y sounding technobabble to put on the card.

      First episode of "Mad Men" has something similar to this, with an "it's toasted" slogan.

      For the nerds who are wondering (this is a site for nerds after all), and don't already know (this is a site for nerds after all), the tape head is essentially a C-shaped electromagnet with a 19 micron gap between the prongs. It needs to be 19 microns wide to work with VHS tapes.

    15. Re:Why did they buy based on "cores"? by dotancohen · · Score: 1

      Right. However, Quickbooks is the de facto standard in the industry and the client was not about to be a revolutionary for no personal gain. People who work in the financial industry are very, very conservative (and stingy with their money).

      --
      It is dangerous to be right when the government is wrong.
    16. Re:Why did they buy based on "cores"? by UnknowingFool · · Score: 1

      we need 4-8 core 5ghz REAL sustainable speed, not "turbo" or "boosted" marketing bullshit. My SSD is already waiting for the slow ass processor and chipsets and there is no 5ghz real processor speeds in sight for at least another decade.

      And what do you need 4-8 cores to do? For most consumers, reading email and looking at Facebook doesn't require 4-8 cores. That was my point.

      And with how utterly shitty todays programmers are, we will not see much improvements in moving everything to multi-threaded programming for at least another decade as well.

      Because most consumers and programs don't need it. If you do things like video encoding, you might need it.

      --
      Well, there's spam egg sausage and spam, that's not got much spam in it.
    17. Re:Why did they buy based on "cores"? by gstoddart · · Score: 1

      LOL ... I remember somewhere around Y2K+-2 my brother was looking for a new VCR.

      One of them cost quite a bit more, but had "Home Theater Ready" stamped on it.

      When asked what that actually meant, the sales guy (after giving some really hand-waving explanations) eventually realized my brother knew a fair bit about VCRs. Eventually he said "It's got the same footprint as the rest of the components so it stacks nicely".

      There was absolutely no discernible technology which made it "Home Theater Ready". But the unit had been sized to stack nicely with other home theater components.

      You should always assume marketing is lying to you. Because they probably are.

      --
      Lost at C:>. Found at C.
  9. Meh, nothing is going to come of it. by nashv · · Score: 1

    A judge is likely to ask : Were there 8 cores ? If the answer is yes, which it seems to be , then AMD is in the clear.

    No multi-core CPU box ever came with a statement that all 8 cores would be capable of processing instructions in parallel at the same time. It does however mean, that AMDs 8-core is significantly worse than Intel's 8-core.

    --
    Entia non sunt multiplicanda praeter necessitatem.
    1. Re:Meh, nothing is going to come of it. by unixisc · · Score: 1

      In fact, the people suing will have a tough time convincing a judge. "But, your honor, there are only 4 FPU units." Is the judge gonna adjudicate whether 2 cores sharing an FPU unit merges them into 1?

    2. Re:Meh, nothing is going to come of it. by gnupun · · Score: 1

      Well two cores are not supposed to share things like integer and floating point units. So those two cores could be considered as one core by the court, and rightly so.

      Anyone know if each so-called core has separate L1, L2 caches or not? TFA is very light on details.

    3. Re:Meh, nothing is going to come of it. by hjf · · Score: 1

      according to another comment here: yes. unlike intel core i series, which share cache.

    4. Re:Meh, nothing is going to come of it. by swalve · · Score: 1

      Way back when dual core processors came out, the implication was that the processor was two independent processors in one package. That logic follows onto 4 and 8 core machines. The example being that Intel doesn't try to pass off hyperthreading as separate cores.

    5. Re:Meh, nothing is going to come of it. by viperidaenz · · Score: 1

      The two cores in a bulldozer module share L1 instruction cache and the entire instruction pipeline, down to the dispatch unit. It's like hyper-threading with dedicated ALU's for each thread.

    6. Re:Meh, nothing is going to come of it. by nashv · · Score: 1

      There are 8 physical cores on the chip by the definition of a CPU core :
      "A central processing unit (CPU) is the electronic circuitry within a computer that carries out the instructions of a computer program by performing the basic arithmetic, logical, control and input/output (I/O) operations specified by the instructions."

      The problem is that each core in a pair cannot do it without compromising the partner core's performance. This is something AMD never claimed anyway.

      --
      Entia non sunt multiplicanda praeter necessitatem.
  10. Re:Damnit by NicBenjamin · · Score: 1

    Real life is not like software. The cost of developing a totally new produce is not a couple million for some really smart engineers and the computers they use.

    How much does it cost to set up a chip fab? Me memory from a late 90s factoid was it was the Billion$ range, and with the ever-shrinking nanometer processes they keep coming up with it can't have gotten cheaper.

    When a major player goes bankrupt in a market like that it makes it much harder for anew guy to enter, because all the banker isn't stupid. If a company with decades of experience, existing fabs, etc. goes bankrupt, then the odds against our innovative new start-up have to be pretty bad, and a loan in that range has to be extremely fucking risky.

  11. Re:The AMD chip by Anonymous Coward · · Score: 2, Insightful

    You're right! They're a joke! Hopefully they go bankrupt so you can buy an i5 for the price of a Xeon!

  12. May as well sue over GHz claims by Overzeetop · · Score: 1

    I mean, if you bought a 3.5GHz chip and it didn't perform as fast as an Intel 3.5GHz it must be misleading marketing, right? We all know that clock speed is all that matters when you compare a chip. And, I've been told, that they may be reducing clock speed dynamically when the processor isn't fully loaded - basically cheating you out of the speed you PAID for. I hear AMD also ran over your cat.

    --
    Is it just my observation, or are there way too many stupid people in the world?
  13. AMD vs Intel performance by Cowclops · · Score: 2

    Regardless of how many cores you have, 90% of the crap that 90% of people do doesn't even benefit from having more than two cpu cores, and it only benefits from that because you can use one core for "the app you're running in the foreground" and one core for literally all other background tasks so your system doesn't lock up if you really need to use 100% of a core, AMD can throw more cores at their CPUs but end users won't see real world performance benefits except in limited scenarios. Intel's super cheap Pentium G3220 at 3ghz (and with only two cores) beats everything AMD sells except for hand-picked benchmarks where AMD's core advantage overcomes its brutal per-clock disadvantage. Single threaded performance is whats king and you'd have to run an AMD cpu at like 5ghz to overcome its poor performance per clock to beat a 3ghz Intel (core-architecture) CPU. And its sad for me to say this because I bought nothing but AMD CPUs from like 2000 until 2014. I wish they could be competetive but the situation looks pretty bleak. I can't use their stuff when their 8 core 4.5ghz cpu is only "a little bit faster" than my 4 core 3ghz phenom from 2009, but my i7 4790k is like, a solid 3 times faster than that 2009 phenom in almost anything you throw at it.

    1. Re:AMD vs Intel performance by Lord+Apathy · · Score: 1

      Your observations on the AMD vs Intel performance gap seem to match my own observations.

      I have a 3.6 ghz fx-8150 and my friend bought a Intel i7 also at 3.6 ghz. On most desktop operations you can't tell the difference. When we start operations like video encoding the difference in performance is staggering. On the average he gets over a 40% performance increase over my AMD system.

      That is a hell of a performance gap.

      --

      Supporting World Peace Through Nuclear Pacification

    2. Re:AMD vs Intel performance by swalve · · Score: 1

      The PCs I support are mostly AMD now, and I've noticed the same thing. The numbers can say all they want, but they underperform compared to much older/slower/cheaper Intel CPUs.

  14. Question: Is this the CPU that's in XboxOne/PS4? by sbaker · · Score: 1

    Is this the same AMD "8-core" device that's in both PS-4 and Xbox One consoles?

    If so, the consequences may go much wider than this story makes it appear!

        -- Steve

    --
    www.sjbaker.org
  15. Re:Damnit by Lord+Apathy · · Score: 2

    No, the better option is that AMD gets their shit together, never giving Intel full monopoly on the x86-market even for a bit.

    Myself and a friend have been AMD fanboys for 20 years. I built my first system on AMD 133mhz '486. Since then he abandoned AMD on his last workstation build and went all Intel. I have a FX-8150 and a FX-8350 in my builds right now. He has given up on AMD and I'm not far behind. I'm looking at my next workstation build sometime in 2016. I'm not ready to completely abandon AMD. I'm gong to wait and see what the specs on the AM4 bring to the table but I expect to be disappointed.

    A few months back we looked to see who might be interested in buying AMD and keeping as a whole. We didn't come up with any one. What we got was a list of people that might be interested in parts of AMD. Sony or Microsoft might buy their chip facilities since they do use AMD chips as the CPU/GPU in their consoles. AMD has a few patents that both Nvidia and Intel might be interested in picking up.

    Realistically AMD is to far behind the technological curve in so many areas. Even looking at the specs to the new AM4 systems I can tell they are behind Intel in almost every aspect. AMD has always been behind Intel in the performance area for most of its life. But they where never so far behind that their lower prices didn't make up for. That is not true with the FX chips. The performance gap is not worth the lower cost.

    Lets hope AMD gets their shit together. As I've said before. The fat lady hasn't sung yet on AMD but she is warming up in the bullpit. Lets hope that as she waddles up on stage AMD pulls a rabbit out of the hat and she falls off the stage into a tuba.

    --

    Supporting World Peace Through Nuclear Pacification

  16. Re:i5, same thing? by Anonymous Coward · · Score: 1

    Wrong. Many i5 CPUs only have two physical cores that are identified as 4 cores due to threading.

  17. Same marketing BS as "hyperthreading cores" by Lumpy · · Score: 1

    Honestly they need to stop being allowed to advertise "cores" if they are not honest to goodness real separate processor cores.

    I hope AMD loses HARD on this and forces the industry to stop being misleading assholes.

    --
    Do not look at laser with remaining good eye.
    1. Re:Same marketing BS as "hyperthreading cores" by Luthair · · Score: 1

      As explained elsewhere what AMD is doing is not the same as HT.

    2. Re:Same marketing BS as "hyperthreading cores" by Lumpy · · Score: 1

      Correct, it's actually a little bit better.

      but it's still marketing bullshit.

      --
      Do not look at laser with remaining good eye.
    3. Re:Same marketing BS as "hyperthreading cores" by wagnerrp · · Score: 1

      The alternative is to try to call them "dual module" processors, and then go through a big long explanation to customers who really don't care what "module" actually is.

      Can it execute two separate threads simultaneously? Yes. At full performance? Mostly, although with a shared frontend, when both units are running at full load, instruction disp---- Is someone trying to make a practical judgement of a chip's performance based solely on its core count going to have a clue what any of that means? No. Then it's a four-core processor.

    4. Re:Same marketing BS as "hyperthreading cores" by superwiz · · Score: 1

      They do have 8 CPU cores. You can run 8 OS's (1 host and 7 guest VMs) on it. Yes, I tried it. It works on AMD. It doesn't on Intel (essentially freezes on Intel).

      --
      Any guest worker system is indistinguishable from indentured servitude.
    5. Re:Same marketing BS as "hyperthreading cores" by superwiz · · Score: 1

      BTW, the cores 2,4,6,8 ship parked. You need to go through a bit of initial effort to unpark them. So they are really NOT just additional features added to CPUs 1,3,5,7. They independent. Show me how to run 5 VM's independently at the same time on 4 hyperthreaded core processor. I tried it. It freezes even if you have plenty of memory left. Because it has to start context swapping between multiple VMs's. Which essentially means cache is invalidated. Not sure if hyperthreading ever happens either. On AMD, 5 VMs and 1 host don't even slow down. There is no difference in the performance of each VM if the others are turned off.

      --
      Any guest worker system is indistinguishable from indentured servitude.
    6. Re:Same marketing BS as "hyperthreading cores" by viperidaenz · · Score: 1

      AMD emphasize the core count in their advertising

      The industry's first and only native 8-core desktop processor for unmatched multitasking and pure core performance with "Bulldozer" architecture.

      They then offer almost no explanation of what is compromised in the Bulldozer architecture, even if you go searching for it. The only thing I could find on their website that explained what parts were shared was a PDF link to hotchips.com

    7. Re:Same marketing BS as "hyperthreading cores" by angel'o'sphere · · Score: 2

      You should be able to host as many VMs as you have memory for.
      Cores should not be an issue.

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    8. Re:Same marketing BS as "hyperthreading cores" by gweihir · · Score: 1

      I have them and I did run real-world benchmarks. For non-FP you get the performance of 8 cores. I do not see anything misleading here. I do see a lot of stupidity from people that have no clue what they are talking about.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    9. Re:Same marketing BS as "hyperthreading cores" by superwiz · · Score: 1

      You are either simulating instructions (binary translation in VMWare terms) or you are letting the processor run them directly. If you do binary translation, all performance is abysmally slow. So you essentially set it up to pass through instructions to the CPU. If you do that and the number of CPU cores you use up is higher than the number of cores you actually have, then performance slows to virtually nothing (even slower than binary translation). On Intel with 4 cores and hyperthreading you'll get to that point even with 4 VMs (you do need VT-x enabled in BIOS). On AMD (you do need VT-d enabled in BIOS) you can run 5 VM without binary translation and they perform. So the 8 cores are 8 actual CPU cores.

      --
      Any guest worker system is indistinguishable from indentured servitude.
    10. Re:Same marketing BS as "hyperthreading cores" by gweihir · · Score: 1

      I'm unaware...

      Yes, you are.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    11. Re:Same marketing BS as "hyperthreading cores" by gweihir · · Score: 1

      The point here is that AMD has 8 full CPU state sets and hence task switching up to 8 execution threads costs nothing. But my guess is this level of understanding is way, way beyond the typical Intel fanboi.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  18. Re:Damnit by whoever57 · · Score: 1

    How much does it cost to set up a chip fab?

    You know AMD doesn't own chip fabs, right? They were spun out into Global Foundries.

    But you are right about the cost of designing a new chip. I would expect the cost to design, validate, manufacture a competitive x86 chip is going to be hundreds of millions of dollars.

    --
    The real "Libtards" are the Libertarians!
  19. Re:Damnit by Luthair · · Score: 2

    No, the better option is that AMD gets their shit together, never giving Intel full monopoly on the x86-market even for a bit.

    Unfortunately its less about AMD getting their shit together, and Intel not being punished enough for their behaviour 10-20 years ago which blocked AMD. At this point unless AMD gets lucky (or skilled whatever) in a future architecture which places them ahead or on par with Intel, and Intel doesn't get a pass on monopolistic behaviour AMD is never going to be more than a niche pick.

  20. Re:Nothing new here by Luthair · · Score: 1

    Nope, as explained elsewhere what AMD is doing is not the same as HT.

  21. How important is FPU? by unixisc · · Score: 1

    Good point. It's one thing to base it on MHz, which usually was a good indicator of relative speed within a CPU architecture (even though a 150MHz Alpha was slower than a 150MHz Pentium). It's another thing to base it on #cores, since it's well known that parallelism in CPUs is hard to extract, in which case, beyond 4 CPUs, one is quickly hitting the point of diminishing returns.

    Also, to what extent have the loads on Floating point units changed? I recall that in the 90s, FPUs used to be heavily de-emphasized, due to the fact that then versions of Windows were very light on it. In fact, NexGen - one of the companies that AMD acquired - excluded them altogether, while if you needed high floating point performances, you went w/ RISC - Alpha, PA-RISC or Power. Today, people use GPUs for any floating point intensive firepower that may be needed, so that begs the question - how critical is it to have a high performance floating point unit for every core in the CPU?

  22. Re:Damnit by Gaygirlie · · Score: 4, Informative

    AMD has always been behind Intel in the performance area for most of its life.

    Well, there was the Athlon - era where they were sweeping the floors with Intel; the classic Athlons and Athlon XPs were phenomenal CPUs at the time and highly overclockable. It was glorious, but yeah, I think that was pretty much the only time they beat Intel.

    Lets hope AMD gets their shit together. As I've said before. The fat lady hasn't sung yet on AMD but she is warming up in the bullpit. Lets hope that as she waddles up on stage AMD pulls a rabbit out of the hat and she falls off the stage into a tuba.

    I really, *REALLY* hope they can manage to do it, but.. I just haven't heard any promising news in that regards anywhere. There's quite literally nothing to indicate that AMD has in any way or form stopped digging even further down the hole they are already in. I do dread the day when Intel becomes the sole x86 - vendor and can practically demand whatever they want, do whatever they want and laugh all the way to the bank.

  23. Re:Question: Is this the CPU that's in XboxOne/PS4 by Anonymous Coward · · Score: 2, Informative

    Cant be, as the PS4 and Xbox One use Puma derived cores, a completely different Architecture than the Buldozer derived Desktop CPU.
    The Puma Cores do NOT share the FPU. Puma CPU use a special low power Design that doesn`t scale up well above about 2 GHz

  24. Re:The AMD chip by Lisandro · · Score: 2

    AMD dying off would effectively make Intel the sole producer of high-performance x86/64 CPUs in the market. Sure, you have VIA as well, but good luck trying to crunch numbers on an C7 core.

  25. Re:Not everything requires an FPU. HT, anyone? by SScorpio · · Score: 1

    I facepalmed at the mention of how Intel allegedly doesn't do this... does anybody remember HyperThreading? :-P

    Intel's marketing for hyperthreading shows a quad, or hexa core processor that supports hyperthreading. So they advertise a four or six core processor that shows up as 8 or 12 cores in Windows.

    AMD advertises an eight core bulldozer processor, but you only have four FPUs on the chip.

    So Intel says you can run two threads on a single core, AMD says you have twice the number of FPUs so they aren't full cores.

  26. Re:blame windows not amd by beelsebob · · Score: 1

    No, blame AMD. They were doing the same thing as Intel is doing with Hyperthreading (scheduling two cores worth of instructions on one actual physical core), just they were claiming that that meant it had twice as many cores, rather than being honest that it's a trick to get more efficiency out of one core.

  27. Compare to NVIDIA by SoftwareArtist · · Score: 1

    They've got nothing on NVIDIA, who advertise the GTX 980 as having 2048 "cores", when by any standard definition it only has 16 (or if you're really generous, you could maybe argue it has 64, but that's pushing it). They count every lane of their vector unit as a separate core. By that standard, AMD (and Intel) should multiply all their core counts by 8, since each AVX unit can do 8 int or float operations at once.

    --
    "I'm too busy to research this and form an educated opinion, but I do have time to tell everyone my uninformed opinion."
    1. Re:Compare to NVIDIA by tempmpi · · Score: 1

      They've got nothing on NVIDIA, who advertise the GTX 980 as having 2048 "cores", when by any standard definition it only has 16 (or if you're really generous, you could maybe argue it has 64, but that's pushing it). They count every lane of their vector unit as a separate core. By that standard, AMD (and Intel) should multiply all their core counts by 8, since each AVX unit can do 8 int or float operations at once.

      I think 64 cores in 16 modules would be a fair assessment of a GTX980. Each SMM contains 4 register files and dedicated execution units, 4 instruction fetch and scheduling units. Not too many things are shared within each SMM: mostly caches and shared memory.

      --
      Jan
  28. Re:Damnit by Lord+Apathy · · Score: 1

    Well, there was the Athlon - era where they were sweeping the floors with Intel

    Those where glorious days, where they not? I remember the race to 1 ghz and AMD beat Intel to that mark. AMD didn't just inch ahead there ether, they rode Intel hard and put them away wet. Those where good days.

    Then Intel caught up and stayed ahead most of the time but AMD was still able to maintain a competitive edge in price. The performance for a Intel chip vs AMD was about 5% to 10% in most areas but AMD's prices where a hundred or more dollars cheaper than Intel. For that much of a difference a 5% gap in performance didn't matter.

    We still see that much of a price difference but the performance gap is like 40% or greater. That is not worth the discount of a AMD part. There there are other technological gaps like DDR4, USB3, and PCI 3.0 support.

    I looked at what AM4 was bringing to the table next year in 2016. Problem is, everything its bringing Intel had out last year. I think things are looking pretty bleak for AMD in 2016.

    --

    Supporting World Peace Through Nuclear Pacification

  29. Re:Damnit by TheRaven64 · · Score: 3, Informative

    Sony or Microsoft might buy their chip facilities since they do use AMD chips as the CPU/GPU in their consoles

    Given that AMD spun off its chip facilities a few years ago, I think we can probably ignore your market analysis.

    --
    I am TheRaven on Soylent News
  30. Re:The AMD chip by silverkniveshotmail. · · Score: 2

    You realize x86-64, the architecture nearly every PC uses was made by AMD, right?

  31. Re:The AMD chip by Martin+Blank · · Score: 3, Informative

    Intel doesn't want AMD to die off because that would subject its practices to monopoly scrutiny. It shares just enough patent information with AMD to allow it to trail a bit behind (in exchange for access to AMD patents, of course) and not completely collapse. The one time that AMD managed to move ahead of Intel (when the Athlon was the king of the hill), Intel pulled out all the stops to prevent it getting a solid foothold in the PC market until Intel's Core 2 Duo could come along and put Intel technologically back in the lead. AMD hasn't had the money to effectively compete since then in part because Intel ensured that its bank accounts couldn't build up too far.

    --
    You can never go home again... but I guess you can shop there.
  32. Re:Question: Is this the CPU that's in XboxOne/PS4 by Lord+Apathy · · Score: 2

    I don't see how this issue can affect PS-4 and Xbox One consoles. Both those are marketed with out disclosure to the public on how many cores their processors have. Most people that buy games consoles don't care about that as long as their a radical performance increase over their last console.

    I doubt that ether microsoft or sony will have anything to say ether. Both their engineering teams crawled all over the cpu designs they where planning to put in the console. They didn't just point at a cpu and said that one is good. They knew exactly what they where putting into those consoles.

    --

    Supporting World Peace Through Nuclear Pacification

  33. Re:The AMD chip by Lord+Apathy · · Score: 2

    I built a few workstations a few years back out of those VIA parts. God what a piece of shit. These where for low end office work too. We couldn't get off those things fast enough.

    --

    Supporting World Peace Through Nuclear Pacification

  34. Why ARM and Atom thrive by tepples · · Score: 2

    If slow processors get the work done fast enough, and they allow the device to be smaller, generate less heat, and run longer on a given mass of battery, then I'll take the slow processors.

    1. Re:Why ARM and Atom thrive by tepples · · Score: 1

      I'd rather have a faster processor that generates less heat and runs longer because it can do the work much more quickly and therefore needs less on time.

      True, the backlight often draws more power than the CPU nowadays. And if a program takes longer to run on the CPU, it keeps the backlight on longer. But if it's blocked on user input the vast majority of the time, then the backlight runs either way. If I'm typing a Slashdot comment into a text area in Firefox, even a five-year-old Atom N450 can more than keep up.

      Also SpeedStep eliminates any benefit of a slow CPU.

      Is SpeedStep on Pentium or Core i good enough to scale down to Atom-class power consumption nowadays?

    2. Re:Why ARM and Atom thrive by arglebargle_xiv · · Score: 1

      But if it's blocked on user input the vast majority of the time, then the backlight runs either way. If I'm typing a Slashdot comment into a text area in Firefox, even a five-year-old Atom N450 can more than keep up.

      My main desktop PC is nearly ten years old. It typically sits at 3-5% CPU, with only the case fan even running. For most users, who use their PCs to check email and read Facebook, anything even 5-10 years old will do just fine.

      Is SpeedStep on Pentium or Core i good enough to scale down to Atom-class power consumption nowadays?

      Actually you need to reverse that quote, for most of its lifetime standard (not mobile) Atom had pretty much nonexistent power management: C-states? Nope (well, C1, which doesn't really count). Idle states? Nope. Demand-based switching? Nope. Speedstep? Nope. It's only the most recent versions that have started to add this. Sure, the base CPU may only consume 10W, but you can't get it much lower than that. When I wanted to build a noiseless, low-power media PC I went through three different Atom-based ones before I finally switched to a mobile Celeron, which gave me the quiet, low-power operation that an Atom (at that time) couldn't. They've finally fixed this in the embedded E3800 series, but then I don't know whether those aren't just rebadged Celerons (64-bit, multi-core, ECC memory, AES instructions, that's not an Atom any more).

    3. Re:Why ARM and Atom thrive by mattventura · · Score: 1

      Well, AMD CPUs are notorious for their lack of efficiency (and therefore heat output), so I'm not sure what AMD has to do with CPUs that "allow the device to be smaller, generate less heat, and run longer on a given mass of battery".

  35. CPU Architecture by tempmpi · · Score: 1

    "The suit claims that Bulldozer's design means its cores cannot work independently, and as a result, cannot perform eight instructions simultaneously and independently."
    If the suit really makes this claim it is easy for AMD to defend, because an AMD Bulldozer with 4 modules / 8 cores can actually execute 8 independent floating point instructions per cycle.

    The two cores in each module share the floating point units, but each module contains 2 independent 128-bit FMAC units. Floating point throughput could be even higher, if both cores would have their own 2 128-bit FMACs, but would likely be lower if AMD decided to go with a single 128-bit FMAC per core. Sharing the floating point units allows higher execution speeds within the same area, as in most cases only one core is currently executing FPU instructions and that core can reach a higher speed with 2 execution units vs. one.

    --
    Jan
    1. Re:CPU Architecture by viperidaenz · · Score: 1

      Unless one core has a cache miss and needs to wait for the data to be fetched from main memory.
      Mean while the instruction pipeline that feeds the other core in the module gets stalled, meaning nothing executes in either core.

    2. Re:CPU Architecture by tempmpi · · Score: 1

      Unless one core has a cache miss and needs to wait for the data to be fetched from main memory.
      Mean while the instruction pipeline that feeds the other core in the module gets stalled, meaning nothing executes in either core.

      Something that can always happen. Stalls because of cache misses are happening in all high performance CPUs. This is completely unrelated to modules vs. cores. Stalls even help the shared FPUs approach: If one core is stalled, than the shared FPU is just as good as a non-shared FPU for the other core.

      --
      Jan
    3. Re:CPU Architecture by viperidaenz · · Score: 1

      The problem is a stall in one core can stop the other. This doesn't happen if the cores have their own instruction pipelines. It's a critical resource that's shared between the two cores of a module.

      The shared FPU is useless in this case, as the instruction pipeline that feeds both cores is stopped.

    4. Re:CPU Architecture by tempmpi · · Score: 1

      The problem is a stall in one core can stop the other. This doesn't happen if the cores have their own instruction pipelines. It's a critical resource that's shared between the two cores of a module.

      Nope, this will not happen. They are not completely stupid. I think you do not get how modern Out-of-Order CPUs such as Bulldozer are designed: There is not a single big instruction pipeline, that is either running or stalling, but many different pipelines with many waiting spots where stalled instructions can wait and non-stalled instructions can overtake stalled instructions.
      A instruction cache miss will also not stall the instruction fetch and decoder pipelines: Bulldozer will just try to fetch something from the other core in the meanwhile or even prefetches instruction blocks that are predicted to be required soon by branch prediction.

      --
      Jan
  36. Re:Damnit by Anonymous Coward · · Score: 1

    My rig just built is as fast as any Intel system even for gaming. I have an AMD FX 4350, MSI high end motherboard, 16GB of memory, all of which cost me less than 250.00 and it's all off the shelf (for AMD the most expensive gear you can buy but nothing custom ordered). I would have a hard time getting a Motherboard and memory for an Intel at the same price. Then I would have to dump out a few hundred more for a 4core 4.2GHz chip.

    I dual boot Linux and Windows, and there is nothing Intel could do faster or better than my rig for any workload I can do at home.

    Now for less than 250.00 what would you be able to do with Intel? What could you measure an increase over AMD with? Nothing.

    You are not an AMD fanbo, you are a shill trying to disguise yourself as one.

  37. Re:The AMD chip by Lisandro · · Score: 3, Informative

    I understand they're very popular on the embedded market due to their low cost, but i can't honestly tell why. Any half decent ARM platform will run circles around it.

  38. Re:Nothing new here by jones_supa · · Score: 1

    The point is that they still are not real cores.

  39. Polarized Plugs by Etherwalk · · Score: 1

    I was shopping for VCRs about 20 years back and asked the Future Shop guy how much better it was for having (quoting from the card beside the VCR) a "19 micron tape head". Turns out they ALL had 19 micron tape-heads (whatever the hell that *meant*) as it was the spec for a VCR tape head, at the time, at least. It was just another bit of science-y sounding technobabble to put on the card.
    Buying based on core count is like buying for the 19-micron thing; it's either a fast machine for your purposes or not. Absolutely the only way to tell that for sure is a test. The only thing that was ever useful with, say, "megahertz" was that it had for a decade or so there a correlation with the performance you'd get in real use. I've never found "cores" to have anything of the sort.

    When I went apartment-hunting five years ago, one of the buildings was advertising its unit features to include polarized plugs. And auto-defrost refrigerators.

    You know, in case you had been living in 1950.

  40. Re:i5, same thing? by Anonymous Coward · · Score: 1

    try msinfo32, try the performance tab in Task Manager

  41. Re:i5, same thing? by __aanfwt7763 · · Score: 1

    the shit you linked states exactly what the guy said. 2 physical and 2 logical cores. at least learn how to read your own links.

  42. Re:Damnit by Agripa · · Score: 1

    Those where glorious days, where they not? I remember the race to 1 ghz and AMD beat Intel to that mark. AMD didn't just inch ahead there ether, they rode Intel hard and put them away wet. Those where good days.

    Then Intel caught up and stayed ahead most of the time but AMD was still able to maintain a competitive edge in price.

    Leading to Intel releasing the flawed Coppermine Pentium 3 1.13 GHz which had to be recalled.

  43. Re:i5, same thing? by radarskiy · · Score: 3, Informative

    i3, i5, and i7 represent "good", "better", and "best" respectively. That's it. A *particular* SKU with an i5 mark may have 4 physical cores, but 4 physical cores is not a requirement to receive the mark.
    For example, this i5 has 4 physical cores: http://ark.intel.com/products/... while this i5 has 2 physical cores http://ark.intel.com/products/...

  44. Re:blame windows not amd by rubycodez · · Score: 1

    That's fine, as most serious computing is not done with FPU at all.

  45. Re:What about my drones that don't fly themselves? by rubycodez · · Score: 1

    Sure it can fly itself

    I can put my car on cruise control and it will drive itself, eventually into a wall or ditch or another car

  46. Re:i5, same thing? by viperidaenz · · Score: 1

    msinfo32 says "2 Core(s), 4 Logical Processor(s)" on this i5-3470T

    Performance tab in Task Manager says:
    Cores: 2
    Logical processors: 4

    Intel market it as a dual core CPU as well.

  47. Re:i think this lawsuit will get thrown out by swalve · · Score: 1

    That's all well and good, but you can't call it multicore. You are correct in that it will be pretty hard to prove damages.

  48. Re:i5, same thing? by viperidaenz · · Score: 1

    You can replace "physical cores" with "cores", Intel has never confused hyperthreading with CPU cores.

  49. Re:Damnit by Kjella · · Score: 1

    How much does it cost to set up a chip fab? Me memory from a late 90s factoid was it was the Billion$ range, and with the ever-shrinking nanometer processes they keep coming up with it can't have gotten cheaper. When a major player goes bankrupt in a market like that it makes it much harder for anew guy to enter, because all the banker isn't stupid. If a company with decades of experience, existing fabs, etc. goes bankrupt, then the odds against our innovative new start-up have to be pretty bad, and a loan in that range has to be extremely fucking risky.

    Except that AMD isn't in the chip manufacturing business anymore, that's TSMC, GloFo and Samsung and they split investment costs across the whole ARM market too. In fact, their 14/16nm process is more competitive with Intel than ever, it's AMD that needs to come up with some kick-ass designs. Of course laying out >1 billion transistors in an optimal way is roughly as hard as it sounds....

    --
    Live today, because you never know what tomorrow brings
  50. Re:blame windows not amd by viperidaenz · · Score: 1

    It's fine? Compared to a dual core Haswel with 16 integer units. They don't say it's a 16 core CPU

  51. Re:i5, same thing? by radarskiy · · Score: 1

    In this thread, it's not Intel's confusion that I am worried about ;-)

  52. Re: Damnit by Anonymous Coward · · Score: 1

    Your rig is slower for gaming than a top end i3 which would cost the same to buy and use less power. Its single thread or dual thread performance, which is most important for games, is 25% slower than that dual core i3 with hyperthreading and its 4 thread performance is only 1-2% better than a dual core i3. In 90% of real world workloads your amd is slower and by a big margin at 25%.

    In addition to that your amd has to run faster, hotter and use more power to get slower performance and doesnt even have a gpu on board. It doesnt support pcie 3 so top end graphics card performance is slower and its memmory latency is double that of the i3 and the memmory bandwidth is 30% less.

  53. Re:blame windows not amd by rubycodez · · Score: 1

    both AMD and Intel have their specs publicly known, and moreover there are benchmarks that actually do a good job of pointing out those architectural differences. I'll bet you three-quarters of the public don't buy on any of that, they buy a whole system for a certain price...heck even at one major chain they usually don't even put the amount of RAM on the description cards! You're saying the kind of numb-from-the-neck-up buying public should have some concern and worry over marketing hype, distortion, and buzzwords that are blasted at them? nah, fuck 'em. anyone who wants to do real research, even a ten year old kid, has the truth out there for the taking.

  54. Re:The AMD chip by silverkniveshotmail. · · Score: 2

    They made one and it sucked so much that they gave up on it and used AMD's.

  55. Re:blame windows not amd by cheater512 · · Score: 4, Informative

    Err except with HT there is one execution unit, and with AMD there are two.

    They are real cores, that share instruction decode and a FPU.
    Actual execution is parallel (unlike HT which is more interleaved).

    HT isn't similar to Bulldozer modules in any sense.

  56. Re:The AMD chip by KGIII · · Score: 1

    I actually prefer to buy AMD. I've got plenty of money but I might as well save it. I tend to buy a lot, an absurd amount actually, of hardware over a single year - often just to use it for a short time and play with it. I may keep that hardware or I may just donate it to the local elementary school's computer lab at the end of the school year (so it's in place for the next year). Why AMD? I get more than enough bang for the buck. I'm not a gamer so I don't even buy high end graphics cards. I'm content with an older GPU, even one a couple of generations (or more) back.

    I no longer do any compute intensive work. The little that I do really doesn't need to be quick, I'm not in a rush. I don't really notice much of a speed difference any more. The vast majority of time everything runs just fine. Hell, I don't even bother installing an OS oftentimes. I'll just use a USB drive and boot from that. It's not like I'm lacking in RAM or compute power. If I've got an installed OS then chances are pretty good that it's just a host and that I'm running a VM. With an SSD, a quad core, and just 16 GB of RAM then I'm pretty much good to go - regardless of who made the CPU. I just don't need anything faster and I don't really do anything where the difference shows.

    I also find that AMD seems to play nicer with the varied distros. I'm not talking about policy or code being open or whatnot. I'm talking about it just working without my needing to fuck with anything. It's not that I don't mind poking and learning but I've done enough of that with CPU architectures in the past. I'm learning new and different things and AMD helps me do that. I've experienced more issues with Intel CPUs than I have with AMD CPUs when using Linux as my OS. These days, that doesn't seem to matter nearly as much but I suppose it might be salient.

    *shrugs* I get more than enough horsepower from AMD. I tend to buy whiteboxes with AMD CPUs and even usually get laptops with AMDs in them because they're "good enough" for my needs. I do have a nice new laptop with a shiny new Intel in it, it wasn't an accident but wasn't really a primary concern. I do admit that it's pretty snappy but, again, I don't really note much of a difference between it and a five year old laptop or at least I don't recollect much of a difference. Hardware's fast enough. The only reason I buy new hardware is to play with new configurations, break things in new and interesting ways, and to just waste time and money doing something different for a little while.

    I have a couple of towers, back home, that are still sitting in their boxes. Those do have a dedicated purpose, I tell myself, but there are already PCs in their way. One is going to get an extra pair of NICs and connect to a hub and become a new firewall with pfsense and the other is going to make a new storage array - I'm going to remove the optical drive and fill it up with spinning platter drives and one SSD. The eventual goal is to keep it isolated from external access while allowing neighbors to access it for file sharing and off-site backup. I'll probably find some attached storage to add to it if it turns out that it gets any use.

    I've also been thinking of offering them some sort of cloud hosting - where they can spin up a VM and connect to do any one of a few things - I'll probably see if there's any interest with the aforementioned elementary school. I'm thinking that I can do a small server and they can spin up Edubuntu and Scratch and have a pre-set image and then just store their data locally at their site as well as retain backups here in my basement/server closet. Mostly, I just want to play with the tech because being retired means that I get to do that.

    --
    "So long and thanks for all the fish."
  57. sounds like bull shit by superwiz · · Score: 1

    I run at least 5 VMs on it. And they all run all the time (in addition to the host OS). It also took some initial effort to unpark every other core. Which means cores 2,4,6,8 can be parked independently of their cousins. AMD chips are better for running VMs, while Intel chips are better at running games. I couldn't run 5 VM's on Intel 5 (4 cores with hyperthreading). The actual CPU count is 8 on the AMD chip. You don't need an FPU to run basic operating system threads themselves. It also has more symmetric caching. Whereas Intel's caching is clearly geared at increasing the speed up of each running core.

    --
    Any guest worker system is indistinguishable from indentured servitude.
    1. Re:sounds like bull shit by PhunkySchtuff · · Score: 1

      I'm curious - what do you mean by the cores were parked, and you had to unpark every other core?

  58. Re:Not everything requires an FPU. HT, anyone? by superwiz · · Score: 1

    Yes, you get the Intel machine for high performance applications and AMD for running lots of VMs. FPU is not needed for basically anything but computational stuff. If you want to run a few systems and compile in a few different environments (linux/windows/32 bits/64 bits... that's already a combo of 4), you get AMD and lots of memory.

    --
    Any guest worker system is indistinguishable from indentured servitude.
  59. Re:Damnit by KGIII · · Score: 1

    My first was the K6-2 350 MHz that I OCed to just under 500 MHz. It would go a step higher but was not stable. It was awesome. I still, happily, buy AMD. I find that the CPU is fast enough for everything I do. They became fast enough about five years ago. I don't really need any more speed. Buying new PCs is just a luxury that I enjoy - I don't actually need any of them.

    --
    "So long and thanks for all the fish."
  60. Re:i think this lawsuit will get thrown out by viperidaenz · · Score: 1

    It's false advertising, the two cores in a module share an instruction pipeline.
    That leads to the possibility of one core being starved of instructions because the other one suffered a cache miss.

  61. Re: The AMD chip by Anonymous Coward · · Score: 1

    Itanium isn't x86. Try again.

  62. Re:Not everything requires an FPU. HT, anyone? by viperidaenz · · Score: 1

    Intel does so it.
    They just call it a single core with hyperthreading.
    Since Haswell, there are 8 execution ports in each core that can dish micro-ops to one of 20 execution units - which is made up of memory ops, ALU's and FPU's. They even added an extra branch unit, so hyperthreading can continue one thread while the other does complex FPU ops that take up multiple cycles.

  63. Because America uses jury trials for most things by rsilvergun · · Score: 2

    and Juries are really, really unpredictable. Even if the outcome is legally obvious it doesn't mean the jury won't get dazzled into blaming AMD.

    --
    Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
  64. Re:Damnit by Lord+Apathy · · Score: 2

    Ah yes, yepping of the disbeliever. We'll I guess I'm not. I guess the fact that my workstation is a fx-8150 and my linux box is a fx-8350. I have a A-5350 in my htpc and my old workstation is a Phenom II 965. The rig I build for my daughter is only a AMD 6300. I won't talk about the 9950, 9850, 1440XP, and the T'bird 950 before that. Yup your correct, I've never been a AMD fanboy.

    Look, your little toy box is nice but if you want to run with the big dogs you best know what you're talking about. I'm glad it works for you but some of us don't live in denial. When I upgrade I plan to go Intel. I will give AMD a chance with its new platform but I'm expecting disappointment.

    --

    Supporting World Peace Through Nuclear Pacification

  65. buying based on a single spec isn't wise by glsunder · · Score: 2

    If you're a consumer, you can't base your buying decision based on a single specification. You have to look at your needs vs what is important to you. For some people performance/$ is important. For others, performance $/watt is more important. You have to compare based on the applications that are important to you. If you, as a typical desktop/laptop user, mostly use application A and price is the main consideration, it doesn't matter if the CPU runs at 3 or 4 GHz, 4 threads or 8 threads, etc. What matters is performance/$. If you have $200 to spend on a cpu, it really doesn't matter who makes the better $700 cpu.

    There are plenty of resources available to help people make decisions. Only relying on marketing department information is just plain dumb.

  66. Re:Damnit by chmod+a+x+mojo · · Score: 1

    I do dread the day when Intel becomes the sole x86 - vendor and can practically demand whatever they want, do whatever they want and laugh all the way to the bank.

    The likelihood of that happening now is actually fairly small. ARM chips these days are getting pretty decent, with some ARM servers even being looked at for datacenters. If Intel jacks up the prices System vendors will just move to ARM.

    Add in that there is already a - rather crippled right now, but easily open-able - version of Windows that runs on ARM, as well as many Linux distros and O/S.X builds that run on ARM.... it would be pretty much a death sentence for Intel and X86. There might be some pain at first for hardware integrators, but with what is already out there right now, it wouldn't be impossible to phase out ( majorly expensive ) Intel X86_64 systems for ARM systems in a short amount of time.
      If Intel jacks up the prices System vendors will just move to ARM. Then Intel would be relegated to the high end server market, until ARM or another architecture ( IBM ramps up POWER CPU, someone releases cheap OpenSPARC chips ETC ) and either forces them down in price or just outcompetes them from the market.

    What would be truly interesting would be if you could get an ARM system that can use all of your existing external peripherals I.E. PCI / PCIE addon cards. You could run a few "high end" cores for stuff that is good for single threading, and have tons of lower / slower cores you can dynamically power up for stuff that benefits from massive parallelization, all in approximately the same thermal envelope as X86.

    --
    To err is human; effective mayhem requires the root password!
  67. Re:i5, same thing? by petermgreen · · Score: 1

    i3, i5, and i7 represent "good", "better", and "best" respectively. That's it.

    Worse still they only represent that within a given generation and within a particular "market segment".

    My theory is they designed this naming scheme to deliberately mislead customers into thinking ultraportables could be as powerful as desktops while the reality is an ultra-mobile i7 is less powerful than a desktop i3.

    --
    note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
  68. No, just no. by Redmancometh · · Score: 1

    Eight instruction can, in fact, be executed simultaneously. Run a parallel CompletableFuture task or something, and this becomes extremely obvious. Each parallel "run" use a ClassTransformer to see the ASM instructions being run with a locking (NOT a re-entrant lock) counter, and a \n every 8 locks/unlocks.

    You will see 8 ASM instructions at a time.

    That being said...physically they aren't cores, and shouldn't be referred to as such. They should have referred to them with their own term "module." If they didn't...they should be sued.

  69. Re:blame windows not amd by gweihir · · Score: 1

    AMD has 8 full integer cores in there. HT gives you one full and one 40% core (at best). This gives you 8 as long as you do integer loads. For float-loads things are different, but AMD never hid that at all. And it is not so long ago that floating-point units came as external add-on.

    Incidentally, the complaint is self-contradictory, as a non-expert cannot interpret a number like "8 cores" either as these may be slow or fast cores and have different architectures. The only thing a non-expert can go by is benchmarks, and AMD did not lie on any of them.

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  70. Re:i5, same thing? by EvilSS · · Score: 2

    You are not too smart. 4 HARDWARE THREADS in an I5. Get it?

    Not all i5's have 4 physical cores. Neither do all i7's (5500u for example).

    --
    I browse on +1 so AC's need not respond, I won't see it.
  71. Re:Nothing new here by gweihir · · Score: 1

    Run an integer-benchmark on them some time. These are 8 full cores performance-wise. HT cannot do that. It gives you something like 40% in addition. If you are lucky.

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  72. Re:AMD by gweihir · · Score: 1

    Addendum: You are stupid. Because if AMD goes away. Intel will become much, much worse than you perceive (wrongly) AMD to be.

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  73. Re:blame windows not amd by gweihir · · Score: 1

    There is not only AMD64 in the world. But what would an idiot like you know.

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  74. Re:Damnit by Billly+Gates · · Score: 2

    Sony or Microsoft might buy their chip facilities since they do use AMD chips as the CPU/GPU in their consoles

    Given that AMD spun off its chip facilities a few years ago, I think we can probably ignore your market analysis.

    Personally I think AMD should give up and sell their assets under bankruptacy to Samsung or Realtek or some ARM cpu manufacturer who is component and knows what they are doing and can write decent drivers. :-(

    AMD is damaged goods man and you know why they suck? No one including global foundaries wants to invest making chips that are not 2012 era technology! Global foundaries said there is no demand for AMD chips and their new cutting edge .18 NM plants only make chips for cell phones for Apple, LG, etc. Once they go out of date and are slow and big like .28 microns they use them for AMD chips. ... meanwhile Intel's skylake is .14 NM and moving forward. How can you compete with that? They are toast. I would still buy ATI graphics cards if people would not keep dropping them like Falcon Northwest (I think that was the one) whose customer service reps demanded they only support NVidia as 80% of callers had ATI problems with the drivers.

    If another company had some skin in the game they could use their chip designers to update their x86 line and give more up to date plants. Competing with Intel is like competing against Walmart these days. It can't be done.

  75. Re:Damnit by Gaygirlie · · Score: 1

    If Intel jacks up the prices System vendors will just move to ARM.

    That's hogwash. Almost *none* of all the available Windows-software is available for ARM, including games -- no chance in hell of people moving to ARM and letting go of all of their software, especially so for gamers and people with specialized software. Basically the only people capable of moving between x86 and ARM are those who only use F/OSS-software.

  76. Re:i5, same thing? by fyngyrz · · Score: 4, Interesting

    Mod parent up. It is correct. Some i5's do have 2 cores. The original AC post near the top is also correct, which I notice the mods have knocked to -1 in the usual frenzy of "I disagree, although I am uninformed about the matter."

    For a couple decades, "core" meant CPU. 4004, 1802, 8080, 6800, 6809, Z80, 68000... etc.

    Then it meant CPU+MMU.

    What's funny is that "core" now seems to mean CPU+MMU+FPU (which is great... I love me an FPU per core.) But some (often aimed at mobile or tablets or phones) also have GPUs.

    So how long before we won't accept "core" except as a label for CPU+MMU+FPU+GPU?

    Then it might be "Nah, that thing has no NPU (neural processing unit), that's not a "real core": CPU+MMU+FPU+GPU+NPU

    Then perhaps it'll be "Nah, that thing has no QPU (quantum processing unit)", that's not a "real core": CPU+MMU+ FPU+GPU+NPU+QPU

    Then... well, I have no idea. But I do think it'll be something. It's always something. :)

    --
    I've fallen off your lawn, and I can't get up.
  77. Re:Damnit by chmod+a+x+mojo · · Score: 1

    No one said ARM would instantly replace X86 hardware, there would be no reason for it.

    But software will follow hardware, if Intel _would_ suddenly, or even over a period of time, increase prices to an extreme amount both companies and consumers would opt for the much cheaper alternatives that would crop up. Especially companies, once the price points of hardware outweighed the cost of porting to a new architecture... it's porting time. And once companies start switching it will trickle down to consumers faster and faster, much like how windows is used the most common in business and so is also most used commonly in the home since it is what the PC purchasers were familiar with.
          Basically you have the time T1, the span of time where prices start to increase to the breaking point and the time T2, the span of time from the price breaking point, where life of the hardware in use should last. T1 + T2 would be years, possibly up to a decade or longer depending on hardware stockpiles.

    That is more than enough time for things to be ported. Even if it was a short lead up time, I am very sure Microsoft and Apple, much less Linux, would have very little trouble coming up with something like Rosetta, the program that let Power architecture Mac software run on Intel Macs on their own OS.

    --
    To err is human; effective mayhem requires the root password!
  78. Re:The AMD chip by spire3661 · · Score: 1

    You realize that AMD and Intel have a cross-licensing agreement, so its completely moot who came up with it. If your point was to show AMD somehow is a superior CPU manufacturer, well thats just not supported by the facts. There was a time when AMD came close to parity with Intel, but that was more Intel mis-stepping than AMD's prowess.

    --
    Good-bye
  79. This Discussion Proves It by friedmud · · Score: 2

    The fact that, years later, _WE_ are still arguing about this proves that the case has merit.

    If WE can't come to a consensus about this... then how is Joe Scmoe supposed to figure it out?

    The fact is: this was _misleading_ advertising. They could have easily come up with another name for it (like Intel did with Hyperthreads)... instead they consciously chose to call the extra ALUs _cores_... which does have a meaning to the typical consumer. They did this, on purpose, to muddy the waters... and they REALLY did.

    Does that mean that people shouldn't be more careful about what they buy? Sure. But that doesn't absolve AMD from putting out misleading advertising.

  80. Re: The AMD chip by silverkniveshotmail. · · Score: 1

    That wasn't my point, I never even implied that, I have zero brand loyalty and would have no problem buying a Via CPU if they somehow made one worth using.

  81. Re:i5, same thing? by toddestan · · Score: 1

    Well, your portable i7 probably has features like VT-d and turboboost whereas the desktop i3 probably doesn't. But in any case, Intel's product line is a total mess.

  82. Re:The AMD chip by toddestan · · Score: 1

    Probably because if you need an x86 chip, you're not going to use an ARM?

  83. Re:The AMD chip by Lisandro · · Score: 1

    The point being, most embedded applications don't need x86.

  84. Re:The AMD chip by Lord+Apathy · · Score: 1

    If you need x86 in a embedded application you should look at a atom processor.

    --

    Supporting World Peace Through Nuclear Pacification

  85. Re:Damnit by barc0001 · · Score: 1

    You're an idiot, to put it bluntly. AMD is Intel's only semi - credible competition and if they go tits up you can look forward to paying $500 for a middle to low grade processor again like the 80s and 90s. It wasn't until AMD's K5 came along and was "good enough" for most business tasks that Intel started pricing competitively and it wasn't until the K6 (briefly) took the performance crown that Intel actually got scared. If AMD goes under, there is NOBODY who can step into the void they will create at current. Starting up a chip company isn't like opening a 7-11, it's an unbelievably expensive and complex undertaking, and most investors would be extremely gunshy to back a new company that intends to compete with Intel.

    That means you can absolutely expect to pay double or even triple what you're currently paying for a processor and an nVidia card since if AMD is gone, so are the Radeon cards. The only place there would still be competition is in the tablet/phone space where a bunch of ARM manufacturers are established. And no, I doubt any of them would see the demise of AMD as an invitation to dance with Intel on the desktop.

  86. and yet... by slashmydots · · Score: 1

    In a completely synthetic benchmark, I recall the 8300 being double the speed of the 4300 and the 6300 coming in right in the middle. So I guess their theoretical limitation and pretend core complaint is just that, theoretical. Oh and also it's bullshit.

  87. Re:Not everything requires an FPU. HT, anyone? by serviscope_minor · · Score: 1

    AMD advertises an eight core bulldozer processor, but you only have four FPUs on the chip.

    That's not really true though. You have one full 128 bit FMAC unit per core, which can be temporarily be loaned to the other core in the same module. If you don't use AVX instructions then each core can independently perform 2xdouble operations.

    --
    SJW n. One who posts facts.
  88. Nomenclature X+Y by tohoward · · Score: 1

    I've always preferred to use the following to describe the Intel architectures: (# arbitrary parallel processes) + (possible # additional parallel processes).

    For most i7's that's a 4+4 core CPU. The i5 is either a 2+2 or 4+0, and so on. I'd think these new AMD offerings would be in the 4+4 range, although this description doesn't indicate how often a workload can satisfy the requirement. I'm sure we could complicate it by adding a percentage at the end or something, i.e. 4+4x.75 (75% of the common workloads have the equivalent of 8 cores). You'd think the industry could adopt this (or a similar) approach.

  89. Re: The AMD chip by KlaatuVarataNiktu · · Score: 2

    Come now AC, I suspect you are being intentionally obtuse. Of course Intel could have come up with some sort of 64 bit x86 architecture by themselves (if they cared about their installed base and had actually wanted to); but as you said, they already had the IA-64 monstrosity. AMD did us all the service of providing an acceptable alternative to Intel's plan of forced obsolescence. And, technically, AMD64 turned out pretty well, not sure IA-64 would have done the same as the only desktop\server ISA as it looked like things were headed.

    https://en.wikipedia.org/wiki/...
    "AMD64 was created as an alternative to the radically different IA-64 architecture, [by Intel and HP]. [...] AMD64 architecture was positioned by AMD from the beginning as an evolutionary way to add 64-bit computing capabilities to the existing x86 architecture, as opposed to Intel's approach of creating an entirely new 64-bit architecture with IA-64."

    https://en.wikipedia.org/wiki/...
    "but with x86-64, roles were reversed: Intel found itself in the position of adopting the ISA which AMD had created as an extension to Intel's own x86 processor line. [...Yamhill] After several years of denying its existence, Intel announced at the February 2004 IDF that the project was indeed underway"

    https://en.wikipedia.org/wiki/...
    "[...] Intel had originally hoped [Itanium] would find broader acceptance as a replacement for the original x86 architecture.

    AMD chose a different direction, designing the less radical x86-64, a 64-bit extension to the existing x86 architecture, which Microsoft then supported, forcing Intel to introduce the same extensions in its own x86-based processors.[15] These designs can run existing 32-bit applications at native hardware speed, while offering support for 64-bit memory addressing and other enhancements to new applications.[6] This architecture has now become the predominant 64-bit architecture in the desktop and portable market."

  90. But can it do blast processing? by wardrich86 · · Score: 1

    I'm still waiting for a PC that can blast process like my old Genesis.

  91. Re:The AMD chip by Anonymous Coward · · Score: 1

    The FX-8350 is not their "top of the line" processor.It's about 15% slower than the FX-9590. And the fastest i5 is the 5675C, to my knowledge, which is about 75% the speed of the FX-9590 and $100 more expensive. It takes a lot more than a "slight overclock" to add an extra 25% speed. In most cases you can't accomplish that with an overclock, even with a 40% clock boost, because performance doesn't scale with frequency that way in CPUs.

    Translation: Bad deal, financially.

    Yeah, the FX-9590 isn't great for power efficiency, it's not as fast as the newest generation i7s, but for a huge number of workloads it is as fast as an i7-4770k which is $200 more. And that $200 not only buys me a hell of a cooling unit but also pays for a 480GB SSD.

    And unless you get a braindead game like battlefield, the difference for an i3, and i7, and the goofy AMD A series CPUs is something like 2 FPS. I'm guessing that since you are talking about overclocking that you aren't using it for business, since it pretty regularly leads to stability problems, so you're probably a gamer and you're probably paying an extra $400 for crap that makes no difference at all every time you buy a new computer. Great job buddy!

  92. Re:The AMD chip by tempmpi · · Score: 1

    I doubt that. ARM and MIPS chips are optimized for high energy efficiency and low price point, not high performance. A top of the line ARM Cortex-A57 performs about the same as an Intel Silvermont Atom.

    --
    Jan
  93. Re:The AMD chip by Lord+Apathy · · Score: 1

    Yes. I did some quick research looking for that $7 VIA cpu. I didn't find it. What I did find was the specs for current VIA CPU's. From what I read they are worse than AMD. S

    So the choice between cheap crap and a reliable working system? You see cost alone doesn't just always seal the deal. Spend the extra money and get quality.

    --

    Supporting World Peace Through Nuclear Pacification

  94. Re:i5, same thing? by mattventura · · Score: 1

    But even if a CPU has an integrated GPU, it's not part of any specific core. It's just off on its own. Not to mention, for the foreseeable future, high end desktop/server CPUs won't have integrated GPUs.

  95. Re:The AMD chip by cthulhu11 · · Score: 1

    Which is the same reason Microsoft cedes a certain fraction of market share to Apple. They want them to be significant enough to claim that MSFT isn't a monopoly, but not a dollar bigger.

  96. Re:AMD by gweihir · · Score: 1

    You obviously have never heard of what a "monopoly" is and that it is universally bad for the market. Stupid.

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  97. Re:The AMD chip by LinuxIsGarbage · · Score: 1

    My 2007 laptop has a K8 based Turion64X2 CPU. It was competitive with the C2D on performance, and the nVidia Geforce Go 7200 graphics beat the Intel 945 series (for a laptop). Though like a lot of nVidia chips from the era, it has a tendency of delaminating from the motherboard, and HP released a BIOS update that runs the fan continuously, and the Wifi stopped working 5 years ago because of it.

    I later bought an AMD K8 based Neo netbook with ATI/AMD x1200 graphics, which beat the Atom N270/N280 and GMA945.

    Neither of these chips were great energy savers, but performance per dollar was hard to beat.

    A year ago I set out to replace the Turion laptop with a desktop for daily usage. I'm not a gamer, though I wanted it to be capable of playing HD videos, encoding videos, running VMs, etc.

    I wanted to justify AMD as the underdog, but in the end I couldn't deny the Haswell based i5-4690 beat the AMD offerings. For ordinary usage (eg even web browsing), single core performance is more important than many give it credit for (especially once you pass the 2-4 core boundary). Intel smokes AMD in single thread performance, and unless you regularly run heavy multithread loads, the i5 beats the i7 in cost. On video the Haswell isn't bad for desktop usage, and I can even run 3x 1080 screens (1x HDMI, 1x DVI, 1x VGA).

    As my username may suggest I'm no Linux pundit, but I was always under the impression that Intel video had better Linux support compared to either nVidia or AMD.

  98. Re:The AMD chip by KGIII · · Score: 1

    I don't game so I'm usually not bothering to stay near the bleeding edge with video cards. I can't really say. In my experience, however, anything that I do actually want to do is 'fast enough.' I watch documentaries. I don't even watch them so much as listen to them. Graphics aren't even all that essential. I can play HD and I do have a 4k monitor at home (several, technically) and those all work fine but the GPUs will all be about two years old by now, I imagine. I don't even own a BluRay device. Well, I don't have one hooked up. I did buy a write capable drive to store data but that turned out to be slow and, frankly, I already had better media storage options.

    I could be fairly unique in my use patterns. *shrugs* So, there's that. I'm not Linux exclusively. Sometimes I use BSD in a VM, after all. Err... I think I have a Minix VM that I've booted a few times. ;-)

    For stuff like what I do? Meh... It works, I've not really noticed much of an improvement in overall speeds in years. The Intel, as mentioned, in my laptop is pretty fast. It's not an amount faster that makes me go, "wow." I don't think I've had a speed increase make me 'wow' in a long time. It's kind of sad.

    --
    "So long and thanks for all the fish."
  99. Re:Damnit by Lord+Apathy · · Score: 1

    I'm sorry that you feel that way but I've been doing this for a very long time. I've had numerous AMD chips over the years and I can say only for that short span of period AMD chips have always trailed behind Intel in performance. That performance gap was only around 5% to 10% on average.

    The last figures I ran put my old phenom II 965 at 3.4 ghz against a 3.2 Ghz I5 with 4 cores. The result was exactly what I expected. The AMD 965 inched out the i5 by a very small margin, about 3%, due to higher clock rate. I estimated at the time if I slowed the 965 down to 3.2Ghz it would fit nicely with in the standard model 5% to 10% gap.

    We tested my FX-8150 at 3.6 ghz against a 3.6 ghz i7. I don't remember the model number of the i7. The performance gap stunned us. A 20% fo 40% gap on performance in the i7 favor.

    AMD has made great chips before, and thier chips are still good. But you just can't compare them to what intel offers now in terms of performance.

    --

    Supporting World Peace Through Nuclear Pacification