Slashdot Mirror


AMD Says Upcoming Zen CPU Will Outperform Intel Broadwell-E (hothardware.com)

Reader MojoKid writes: AMD has been talking about the claimed 40% IPC (Instructions Per Clock) improvement of its forthcoming Zen processor versus the company's existing Excavator core for ages. Zen's initial availability is slated for late this year, with lager-scale roll-out planned for early 2017. However, last night, at a private press event in San Francisco, AMD unveiled a lot more details on their Zen processor architecture. AMD claims to have achieved that 40 percent IPC uplift with a newly-designed, higher-performance branch prediction and a micro-op cache for more efficient issuing of operations. The instruction schedule windows have been increased by 75% and issue-width and execution resources have been increased by 50%. The end result of these changes is higher single-threaded performance, through better instruction level parallelism. Zen's pre-fetcher is also vastly improved. There is 8MB of shared L3 cache on board now, a unified L2 cache for both instruction and data, and separate, low-latency L1 instruction and data caches. The new archicture offers up to 5x the cache bandwidth to the cores versus previous-gen offerings. However, after all the specsmanship was out of the way, AMD actually showcased a benchmark run of an 8-core Zen Summit Ridge procesor versus Intel's Broadwell-E 8-core chip, both running at 3GHz and processing a Blender rending workload. In the demo, the 8-core Zen CPU actually outpaced Intel's chip by a hair. Blender may have been chosen for a reason but this early benchmark demo looks impressive for AMD and its forthcoming Zen architecture.

188 comments

  1. Interesting... by fyngyrz · · Score: 3, Funny

    Zen's initial availability is slated for late this year, with lager-scale roll-out planned for early 2017.

    You know, although a tank lager looks big from the outside, there are usually no more than a hundred or so tanks in one. So this doesn't seem like a very large rollout.

    On the other hand, if one of the tanks rolled over the editor(s), that would be a service to humanity.

    --
    I've fallen off your lawn, and I can't get up.
    1. Re:Interesting... by PopeRatzo · · Score: 3, Funny

      lager-scale roll-out planned for early 2017

      I prefer the porter-scale roll-out. Maybe in IPA-scale roll-out on a hot day. Because that's the way I roll out.

      --
      You are welcome on my lawn.
    2. Re:Interesting... by Anonymous Coward · · Score: 1, Funny

      Because that's the way I roll out.

      I don't always roll out, but when I do I transform.

    3. Re:Interesting... by Anonymous Coward · · Score: 0

      They went lager-scale for a fighting chance.

  2. Kind of rigged test by Anonymous Coward · · Score: 0, Informative

    We win, yes so we underclocked the intel processor by 7%. But we managed to beat them just barely.

    1. Re:Kind of rigged test by Anonymous Coward · · Score: 2, Insightful

      Yeah. Some rumors suggest that the Zen line will be disappointingly slow at launch (under 3 GHz). On the plus side, if we can get Broadwell-E performance out of a chip that costs $200, AMD will have a hit on its hands.

    2. Re:Kind of rigged test by Anonymous Coward · · Score: 0

      And of course the real comparison - at ship time in 2017 will likely be a skylake-e or kabylake-e.

    3. Re:Kind of rigged test by GuB-42 · · Score: 1

      Beaten a CPU that is already out with one that isn't yet, using a benchmark of their choice. The only area where they can hope to compete is price.

    4. Re: Kind of rigged test by Type44Q · · Score: 4, Insightful

      The real comparison will depend entirely on price-points.

    5. Re:Kind of rigged test by Rockoon · · Score: 1, Troll

      Beaten a CPU that is already out with one that isn't yet, using a benchmark of their choice.

      I see you here purportedly caring about veracity...

      The only area where they can hope to compete is price.

      ...and then you reveal your true intentions.

      --
      "His name was James Damore."
    6. Re:Kind of rigged test by WorBlux · · Score: 1

      I doubt the price point of $189 The high-bin four-core eight-thread SKU will probably come out at that price to compete with haswell and skylake i5's. Double and then some say $489 for the eight-core to compete with broadwell-E. There will probably be a 6-core SKU made more the 8-core where one or two cores didn't pass quality control for around $299 to compete with some of the desktop i7's. (poorly in single threaded performance anyways, but should do well in parallel test especially DX12 or Vulkan games. AMD really needs cash flow now and won't want to get into a price war. Anyways I think the heavy hitter will be Zen APU's with Polaris and HBM2 likely the drop Q3 next year.

    7. Re:Kind of rigged test by DuckDodgers · · Score: 3, Interesting

      The Broadwell-E part they benchmarked against is probably a (slightly underclocked) Core i7-6900k, and it's $1100.

      I'm taking wild guesses with the numbers here, but "15% slower than a Broadwell-E at a 45% lower cost and a similar TDP" is a valid market strategy. I haven't spent more than $240 for a CPU in over ten years, if in spring 2017 there are Zen parts at the $250 price point that are 15% behind the 2017 spring equivalent of the Intel i7-6700k or i7-6800k I will buy it.

    8. Re:Kind of rigged test by Anonymous Coward · · Score: 1

      Are you implying they should have asked Intel R&D to provide a chip that isn't out yet? You fanboys are scary...

    9. Re:Kind of rigged test by Guybrush_T · · Score: 1

      No. Only that the outcome should be : our next-gen CPU is 20% faster/cooler/... than current gen CPUs. Being equivalent to current products is kind of disappointing.

    10. Re:Kind of rigged test by shaitand · · Score: 2

      Only if they are trying to be the highest performing chip on the block. Being the highest performing chip at a reasonable price point (the $200 mark is generally the sweet spot only the money is no object crowd wastes more on a cpu) is far more important.

      AMD needs a more efficient core. If they have done it, bravo to them.

    11. Re:Kind of rigged test by sexconker · · Score: 1

      Kabylake E at Zen's (volume) launch? No way.

    12. Re:Kind of rigged test by Anonymous Coward · · Score: 0

      zen at volume at launch? no way.

    13. Re:Kind of rigged test by Blaskowicz · · Score: 1

      HBM2 will be on the high end server (maybe workstation) socket. Specs will be 16 cores (likely 12 cores possible) with a mid-sized GPU on a multi-chip-module, with eight channels of DDR 4 (unbuffered, ECC, ECC registered)
      Most home users will likely not be interested, they'll prefer to go with a faster GPU for less expense on a traditional home desktop. This thing allows you 512GB RAM and a lower latency between CPU and GPU.

      I speculate this may go in a Mac Pro, because there is a lot of hardware under a single heatsink.

    14. Re:Kind of rigged test by gweihir · · Score: 1

      CPU speeds have hit a wall some time back. You may notice that Intel cannot fulfill your expectations speed-wise either.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    15. Re:Kind of rigged test by Luthair · · Score: 1

      Talk about not getting it, the entire point of the presentation and demo was to show that their new architecture has equivalent IPC to Intel's best.

    16. Re:Kind of rigged test by WorBlux · · Score: 1

      The thing is right now, as far as I understand it, APU's are largely memory bandwidth limited for game performance rather than silicon limited, hence why Intel put 128MB of L4 (eDRAM) on thier Iris Pro lineup. There's also a potential console market as well. The Heterogeneous Compute Environment IS AMD's wet dream, but if they come up with the fast APU-APU and APU-GPU interconnect fabric to make it happen, I'd imagine that it would end up in an enthusiast system sooner rather than later.

    17. Re: Kind of rigged test by Anonymous Coward · · Score: 0

      And compatibility. AMD is notorious for making CPUs that are almost x86 compatible but not quite and exhibit massive problems in certain areas.

    18. Re:Kind of rigged test by lsatenstein · · Score: 1

      Only if they are trying to be the highest performing chip on the block. Being the highest performing chip at a reasonable price point (the $200 mark is generally the sweet spot only the money is no object crowd wastes more on a cpu) is far more important.

      AMD needs a more efficient core. If they have done it, bravo to them.

      I'm running a 2009 processor that runs at 49C / 90watts. Its fast enough for desktop work, but not as a server with database. I am looking forward, as the other gentleman stated, to not spending more than $250. I want a mother board that supports the current technology and supports 6 sata 2 or 3 drives. Is it a dream, or future version possibility, that the MB will come with 16Gig ddr4 type ram soldered to the MB? If there is that possibility, I would plop down my credit card for such a combo now

      --
      Leslie Satenstein Montreal Quebec Canada
  3. Sure 'tis by Anonymous Coward · · Score: 1

    Pics (benchmarks) or it ain't so.

  4. Good to hear. by jellomizer · · Score: 3, Insightful

    AMD has been behind Intel for about a decade now ever since Intel released their "Core" processors. Because back in the early to mid 00's AMD CPU finally were considered serious chips in the desktop environment, outpacing intel. Then it just fell further behind.

    --
    If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    1. Re:Good to hear. by Anonymous Coward · · Score: 1

      Mostly due to money to invest in R&D with. it was also during that time that Intel was illegally giving kickbacks to companies to use their slower CPU's over AMD's. I'm not sure just how much that ended up impacting a market, but it definitely locked AMD out of some machines and possible income avenues that it could have if the market was not rigged against them due to Intel.

    2. Re:Good to hear. by mwvdlee · · Score: 2

      How about performance-per-dollar?

      --
      Slashdot social media options: AIM, ICQ, Yahoo, Jabber and Mobile Text. Why no MySpace?
    3. Re:Good to hear. by avandesande · · Score: 1

      They have had plenty of time to catch up... for Intel at best there is a 50% improvement over their own offerings from 7 or 8 years ago.

      --
      love is just extroverted narcissism
    4. Re:Good to hear. by Anonymous Coward · · Score: 0

      Nigplz. Phenoms were well competitive in terms of microarchitectural features. What AMD did mess up was QC on the very first Phenoms: the MMU/cache fail was enough to remove what advantage they got from having an integrated memory controller -- while Intel's offering was still relying on a north bridge.

    5. Re:Good to hear. by pr0fessor · · Score: 1

      I had a k6-2+ with maxed out ram on a super socket 7 it would never run win7 but the OSs it ran amazing with 95, 98, win2k, and later slack I haven't seen anything like it since.

    6. Re:Good to hear. by PIBM · · Score: 1

      Have you been living under a rock ?

      In the system to my left, an 8 years old Q9550 ( http://ark.intel.com/products/... ) which while running Windows 10 will only put out 4000 points on cpubenchmark.net (and it shows!). To my right, a 1 year old i7-6700K ( http://ark.intel.com/products/... ) using less power than the Q9550, yet dishing out 11000 points in cpu benchmark. Then, there's this year i7-6850k, handing out a Q9550 more performance over the already performing i7-6700k (got 15500 points!). That's an improvement of 288% over something that was considered top of the line 8 years ago.

    7. Re: Good to hear. by Anonymous Coward · · Score: 0

      Let's not forget AMD memory management and overheating issues. They had cools stuff, but they were a mess. Being faster wasn't actually a good thing when they couldn't deliver stability.

      Oh, there was also the chipset issues. AMD made an awesome CPU but the choice of chipsets was horrible. USB support was absolute shit... The nVidia chipsets were ok, but their USB host controller driver crashed like mad.

      Then there was development support. They simply didn't bother. They made the hardware and didn't invest heavily enough in developer tools. Intel almost always kicks AMDs asses because they focus on performance in the compilers.

    8. Re:Good to hear. by Anonymous Coward · · Score: 0

      Your maths is shockingly shit!

    9. Re:Good to hear. by Cealestis · · Score: 1

      Improvement in this case is somewhat ambiguous. Real world performance vs benchmarks isn't as telling as we would like in CPU comparison. Certainly the processors are more efficient, which allows higher clocks (in this case ~150% higher). New architectures usually bring improvements to performance. You have software utilizing the additional threads and and cores far better then the Q9550 would allow. Improvement yes, but the score encompasses a lot more then what the average user might see as improvement in their usage.

    10. Re:Good to hear. by Anonymous Coward · · Score: 0

      Does that include performance per-dollar-spent-on-each-watt? Because much as I love AMD, Intel owns the server space principally because of its power-performance ratio. You could cut the throughput of Intel CPUs in half, and as long as you maintained the performance-per-watt, Intel would probably still own the server space.

    11. Re:Good to hear. by Anonymous Coward · · Score: 2, Funny

      He's probably doing it on an old Pentium with the floating point bug.

    12. Re:Good to hear. by GuB-42 · · Score: 4, Interesting

      In fact it was more like they took advantage of the P4 fiasco.
      The NetBurst architecture was a failure, it could barely compete with Intel's own previous generation. They made a few bad design decisions. Perhaps they were blinded by the MHz race, perhaps they really thought long pipelines were the future, I don't know. However, they learned from their mistakes and their next generation (Core) was a success.
      At the same time, AMD took a more sensible approach and the K7/Athlon was a worthy "next-gen" CPU. But maybe the lack or craziness also caused them to stand still when Intel advanced. Intel's commercial practices probably didn't help either...

    13. Re:Good to hear. by avandesande · · Score: 1

      Should have been more specific/accurate... i7 family and 7 years old... but the point still stands.

      --
      love is just extroverted narcissism
    14. Re:Good to hear. by avandesande · · Score: 4, Insightful

      Incorporating the memory controller in the CPU and adding the 64 bit instruction set were AMD innovation and had nothing to do with Intel making mistakes.

      --
      love is just extroverted narcissism
    15. Re:Good to hear. by PIBM · · Score: 1

      15500 / 4000 = 3.875 =387.5%
      I'm sorry, I rounded to 388% (it was around 15520-15540 anyway for the few runs I did)

      Thus, the improvement is 288%. Would a 50% improvement be slower?

    16. Re:Good to hear. by PIBM · · Score: 1

      Ok, lets talk real world improvement then. With the same video card and hard drives and in game settings, Battlefield 4 went from an average of 28 FPS to VSync locked at 60fps when I moved from a Q9550 to an i5-4670k (already a 3 years old computer, 7,606 points in cpubenchmark). While the cpubenchmark yielded lower than 100% improvement, I got more than that in game. I also went from 8GB to 16GB of memory but BF4 wasn't using it all. The extra memory speed could have helped though, along with the faster PCI lines, so the whole system ended up being much faster than the old.

    17. Re: Good to hear. by Ravaldy · · Score: 1

      Everything you stated is the reason my IT department did not even consider AMD processors for our business. We were willing to pay more for Intel just to have piece of mind. 10 years later some of those processors were still active and needed to be decommissioned. I can't say the same for the AMD unit we had purchased for special performance requirements.

    18. Re:Good to hear. by PIBM · · Score: 1

      I still have an i7-960 between my five `live` computers, which had been released late Q4 2009, so not yet 7 years old. At that time, the CPU pricing was the same as the i7-6850K. It was about 30-40% faster in games than my 9550 (it had 12GB of fast triple channel memory too!). It gets a score of 6K on cpu benchmark. It was running a pair of geforce GTX260 core 216 in the time :) Currently a headless linux server, so a tad harder to compare to my W10 boxes to get a real world comparison.

      If you want to compare to that one, I`d say the improvement is closer to 150%

    19. Re:Good to hear. by corychristison · · Score: 2

      This is the unfortunate truth.

      I use AMD in my personal machines (Laptop, Desktop, HTPC) because they are cheaper and I get what I need out of them. I dont play video games, and don't really do anything CPU or GPU intensive. A browser, some terminal emulators, file manager, email client, etc. all run great.

      The servers I own are all Intel Xeon E5's. You really can't beat them, and more per watt is definitely better in this space. I run dozens of virtualized servers on each machine and they just keep chugging along without issue.

    20. Re: Good to hear. by johnsmithperson123 · · Score: 1

      It looks hopeful, but with AMD TDP is a matter of debate. There is a 95W TDP listed as opposed to Intel's 140W TDP. If actual power consumption is lower than Intel, say hello to the new era of Opteron. Back before Dozer, AMD was still often favored in server environments.

    21. Re:Good to hear. by DuckDodgers · · Score: 1

      Same video card?

    22. Re:Good to hear. by avandesande · · Score: 1
      --
      love is just extroverted narcissism
    23. Re:Good to hear. by PIBM · · Score: 1

      Yes, that specific video card (GTX 580) was overpowered for the old Q9550 thus it wasn't the limiting factor.

    24. Re:Good to hear. by Tough+Love · · Score: 1

      AMD clobbers Intel in graphics, so if they get their cpus even close to competitive performance wise with good value, I'm in. Still running a Phenom X4 desktop, it's been a reliable workhorse and great value.

      --
      When all you have is a hammer, every problem starts to look like a thumb.
    25. Re:Good to hear. by jcdr · · Score: 2

      For the 64 bit instruction set, AMD have take advantage of the Intel Itanium mistake and it was there smarter move ever as this enabled AMD to negociate the instruction set symetric cross licencing agreement with Intel.

      AMD did make a lot of technical innovations on CPU before, like copper metal interconnect, silicon on insular, integrated memory controller, hypertransport, multiple cores, exclusive caches, etc..., and lately integrated graphic. At some point Intel haved nothing to compete but there extremly agressive marketing worked long enough to develop a completely new generation of Intel core processors to gently catch up and finally surpassed AMD product. Intel HKMG and smaller gate process finally destroyed AMD that at the same time almost stopped his progression.

      The announce look like there used good design decisions. Time will tell.

    26. Re:Good to hear. by PIBM · · Score: 1

      There's clearly something wrong with this benchmark since my i7-6700k with GTX 970 sli is much faster than my i5-4670k with the same video card and amount of ram, yet that site say that it's a much lower performance. I do not know what they use as metrics, but everything I look at seems off. It`s like it has been filled up by hand.

    27. Re:Good to hear. by Major+Blud · · Score: 1

      Maybe, but Intel does bear some responsibility for letting themselves fall behind during that time frame. I can remember back in the K7/Athlon days, there were some Athlon chips that outperformed the P4 chips that ran at the same clockspeeds (Some of the PIII chips even outperformed it). Intel eventually tried to compensate for this by upping the clock speeds and adding cache to the already flawed NetBurst architecture (like the P4 Extreme Editions). I can also remember Intel stating early on that they expected NetBurst to scale up to 10 Ghz, when it in fact never got about 4. The decision to only support RamBus in the initial P4 Williamette rollout also hampered things.

      --
      If you post as Anonymous Coward, don't expect a reply.
    28. Re: Good to hear. by Anonymous Coward · · Score: 1

      "peace of mind"

    29. Re: Good to hear. by Immerman · · Score: 1

      Are you sure? Maybe they liked the accountants giving them "a piece of their mind" for overspending...

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    30. Re:Good to hear. by shaitand · · Score: 2

      Intel also played games with TDP while AMD gave max power consumption figures. Intel were/are still the more power efficient chips but not by nearly as wide a margin as the stated power figures lead one to believe.

    31. Re: Good to hear. by Anonymous Coward · · Score: 0

      1 is not a good sample

    32. Re:Good to hear. by aliquis · · Score: 1

      How about performance-per-dollar?

      Depends on the task.

      In non-FPU easily parallellized tasks maybe they even have an advantage.
      In FPU tasks or single core performance definitely not.

      In general they aren't attractive even on price I'd say.

      The problem is that they can't keep up on performance and they aren't really interesting enough to pick over Intel even at the cheaper tiers (except the APUs if you have no graphics card.)

      If this benchmark was well-chosen (or if say they will clock lower) and actual performance is more in line with an 8 core Zen performing as a 6 core Intel processor then Intel can just say "be 6 core consumer desktop processors" and they will make it so and .. well.. That was that. It brought 6 core processors to the mainstream consumers and one can still support AMD but not necessarily a compelling reason to pick them because of performance or price / performance.

    33. Re:Good to hear. by Anonymous Coward · · Score: 0, Troll

      AMD outpaced Intel starting in 1995 with the 133MHz "586" when Intel peaked at 100MHz. They have always been the price/performance leader. People buy Intel because imbeciles like you don't know any better and live in make-believe land to try and justify paying 100% extra for less than 2% performance difference.

    34. Re: Good to hear. by Anonymous Coward · · Score: 0

      Hey now, I miss my Q9550 chip, it was a great chip that lasted a long time.

    35. Re:Good to hear. by Anonymous Coward · · Score: 0

      does your new cpu protect you from the trillions of aimboters currently playing battlefield 4 pc?

    36. Re:Good to hear. by hairyfeet · · Score: 1

      Not interesting on price you say? Well I'd really like to you show me an Intel at $135 that gets these kinds of numbers because the last time I checked all you can get from Intel for $135 is a crappy Pentium dual or an even crappier Atom. When you look at the bang for the buck it really isn't even close, I mean you can get an octocore from AMD for around $135 that has a 4Ghz turbo clock OOTB and which can easily hit 4.4Ghz-4.6Ghz on air, you aren't gonna find anything from Intel that competes until you at least double the price. Then when you look at motherboards and see how you can get a much nicer motherboard for less when you go AMD? Its really a no brainer, you can get a hell of a lot nicer system for a lot less money by going AMD.

      This isn't even bringing up the elephant in the room which is that software hasn't kept up with hardware in quite a few years so even for gaming you can pair that chip with a $200 GPU and enjoy your games at 1080P with maxxed out graphics and high FPS, and for those that don't game one of the AMD APUs will give them all the power they need while having excellent picture quality and hardware accelerated video and both will spend a lot less.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    37. Re:Good to hear. by Anonymous Coward · · Score: 0

      yes intel was naughty, back when they were playing catch-up. not only was athlon faster, but amd had nearly overtook intel's marketshare, too. intel was absolutely desperate and scared, of course they played dirty and bribed the oems (and later got off with a comparatively tiny drop-in-the-proverbial-bucket judgement to amd afterwords).

      and then, the 'core' happened. and instantly amd was trash once again; and it's been that way ever since.... AMD HAS BEEN SHIT SINCE 2006.

      only sometimes since then, and if price is the absolute ONLY concern, would amd even be on your radar for consideration.

    38. Re:Good to hear. by CajunArson · · Score: 1

      You can buy a used Sandy Bridge Xeon system for cheap that will not only annihilate that thing but be a better bargain than Zen will be.

      http://www.techspot.com/review...

      Be careful comparing ancient parts (that in AMD's case are being sold as "new" because they haven't had anything really new for 5 years) in the bargain bin to new stuff. It won't look good on a price/performance basis for AMD next year when Zen finally launches if you do.

      --
      AntiFA: An abbreviation for Anti First Amendment.
    39. Re:Good to hear. by Luthair · · Score: 1

      During the early a64 era Intel P4 CPUs were considerably less efficient.

    40. Re: Good to hear. by cyberchondriac · · Score: 1

      Yes, a lobotomy.
      But seriously, I've flirted with AMD, back in their K6 and K7 days.. but prefer Intel. And definitely nvidia over (AMD) Radeon.

      --

      Look back up at my post, now look back down, you're on the Internet. Now look back up. I'm a signature.
    41. Re:Good to hear. by PIBM · · Score: 1

      I play on a relatively rare mode, hardcore only, 64 players. We know each other abilities, and when a cheater joins he's spotted and banned very quickly. Most of the time they don't read the rules and just start by breaking them, which gets them kicked. So I've been enjoying this little corner of BF4 :)

    42. Re:Good to hear. by shaitand · · Score: 1

      Yes but intel started playing games with how they reported power consumption with the core 2's.

    43. Re: Good to hear. by Anonymous Coward · · Score: 0

      I guess my currently running amd slot A is a figment of my imagination.

    44. Re:Good to hear. by Anonymous Coward · · Score: 0

      No, not really. AMD's Athlon chips were beating the pants off Pentiums even before AMD singlehandedly won the gigahertz race. Intel got to where it is through abusive OEM marketing strategies, which choked out AMD's R&D budget.

    45. Re:Good to hear. by WorBlux · · Score: 1

      For FPU it depends on what kind of operation. New high-end intel SKU's are supporting AVX512, whereas Zen will only support AVX 256 natively, However very few consumer application would notice the difference anyways. Zen is an entirely new design and shouldn't suffer the crippling FPU limitation of Bulldozer and it's refinements.

    46. Re:Good to hear. by unixisc · · Score: 1

      AMD has been behind Intel for about a decade now ever since Intel released their "Core" processors. Because back in the early to mid 00's AMD CPU finally were considered serious chips in the desktop environment, outpacing intel. Then it just fell further behind.

      Given the fact that Intel has the world's greatest state of the art fabs, while AMD is now fabless, I just don't see how that's gonna change

    47. Re:Good to hear. by Trogre · · Score: 1

      Well to be fair their Opteron line until about four years ago was competing well with Xeons on performance and absolutely blew them away in terms of price and sheer number of cores.

      Unfortunately that all changed with the Bulldozer architecture which, instead of cores, had "dual core" modules with one FPU shared between two ALUs.

      --
      "Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
    48. Re:Good to hear. by aliquis · · Score: 1

      Here in Sweden:
      FX-8370: 1890 SEK
      FX-8370E: 1620 SEK
      FX-8350: 1690 SEK
      FX-8350 with Wraith cooler: 1811 SEK.
      i5 6400 Box: 1868 SEK.
      i5 6500 Box: 2006 SEK.

      I'd definitely take the i5 6400 over the FX-8350.

      I've seen FX-8300-series processors closer to the i3 pricing though.
      FX-8320: 1390 SEK
      FX-8320E: 1328 SEK
      FX-8300: 1299 SEK
      FX-6350 with Wraith: 1308 SEK
      FX-6350: 1249 SEK
      FX-6300: 990 SEK

      So with those you're in i3 territory:
      i3 6320: 1594 SEK
      i3 6300: 1390 SEK
      i3 6100: 1149 SEK
      G4400: 565 SEK

      I think the i5 6500 and the i3 6100 is the processors to consider (and the older G3258 maybe?), those are the Intel processors which have the best value. Similarly I think the FX-8320 may have the best value of that line?

      Anyway, they are close enough and given a choice I'd pick the i5 6400 over the FX-8350 easily. And I'd pick an i3 6100 over the FX-6300 - FX-6350 too.

      The Pentium Dual is like 1/3 of the price (565 SEK vs 1690 SEK), I have no idea where you've "seen" your numbers but they are completely off.

      It is close and the Intel processors are most likely better for games which is where they matter for me. And i5 6500 will likely keep up with the over-locked FX-8350 in gaming.
      https://youtu.be/xKR04WMP9sw?t...

      Both chips are competitively priced against each other just as the graphics cards are. There's no performance advantage in picking the AMD chip over the Intel one in a given price category unless for integrated graphics because they are faster there. As said.

      What is a "much nicer" motherboard? If anything the cheap AM3+ motherboards have huge issues running the FX processors even at stock speeds. Plenty of people who have motherboards which is supposed to support the FX-6300 but which can't run it anyway. AMD doesn't have any processors with PCI-express 3.0 support. You can get cheap H110 motherboards too and at-least your build will actually WORK with those.

      The FX-8000 series is worse for games than i5,
      http://www.gamersnexus.net/gam...
      Fallout 4, 1440p, Ultra, 980Ti
      i3 4130: 53 FPS average, 33 FPS 0.1% low.
      FX-8370E: 51 FPS average, 33 FPS 0.1% low.
      FX-8320E: 50 FPS average, 31 FPS 0.1% low.
      The i5?
      i5 4690K: 71 FPS average, 45 FPS 0.1% low.

      An i3 keep up in the 6300-8300 territory and an i5 beats it. An overclocked i5 the AMD chips can't keep up with.

      An AMD APU doesn't cost "a lot less" than an i3 with integrated graphics. They are pretty expensive I'd say. Compare the Athlon II X4 860K vs the A10 7850K, you pay a lot for that pretty shitty but functional integrated graphics. The G4400 and the i3 6100 cost similarly and you will of course have "excellent picture quality and hardware accelerated video" with the i3 6100 too.

    49. Re:Good to hear. by aliquis · · Score: 1

      For FPU it depends on what kind of operation. New high-end intel SKU's are supporting AVX512, whereas Zen will only support AVX 256 natively, However very few consumer application would notice the difference anyways. Zen is an entirely new design and shouldn't suffer the crippling FPU limitation of Bulldozer and it's refinements

      My post wasn't about Zen at all. It was about the current processors.

      A lot of idiots think AMD processors are better and cheaper and offer better performance simply because they are branded AMD not because that actually is the case.

    50. Re: Good to hear. by untoreh+ · · Score: 1

      They flopped with phenom1, came back with phenom2, then flopped again with bulldozer, and came back with vishera. Both phenom2 and vishera were great for their time but the fail on the first iterations of such archs hurt their sales quite badly on the subsequent ones.

    51. Re:Good to hear. by Anonymous Coward · · Score: 0

      Opposite now. Intel's TDP is max power and AMD's TDP is "typical". Of course this leads to Intel CPUs with 120watt TDPs consuming 70watts during Prime95 and 120watt AMD CPUs consuming 140watts during Prime95.

    52. Re:Good to hear. by DuckDodgers · · Score: 1

      Thanks for answering. I just wanted to make sure that wasn't the deciding factor.

    53. Re:Good to hear. by Anonymous Coward · · Score: 0

      Intel made a few major blunders during that era

      - NetBurst as you mentioned

      - Itantium (required too much smarts in the compiler)

      - Itanium failed to run x86 as fast as older chips

      Opteron and Athlon64 chips, OTOH, were less expensive and were faster at running older 32bit x86 software at the same time that they provided a 64bit path forward. So if you were looking ahead a few years at 8/16/32 GB of RAM and needing to run 64bit software in a year or three, you would buy a 64bit Opteron for the server instead of an Intel 32bit Xeon.

      The AMD 64bit chips were a way to future-proof your server investment. It was obvious that 64bit was coming, but the timeline was fuzzy as to when it would be needed.

      Plus the AMD chips offered multi-cores for under $200 a year or three prior to Intel multi-core chips breaking that price point. Intel was still charging $300+ for their least expensive multi-core CPU at the same time we were rolling out Athlon64 x2 CPUs to the user's desktops for under $150.

    54. Re:Good to hear. by Anonymous Coward · · Score: 0

      Intel also made another mistake during this time period. All of the Athlon64 and Opteron64 chips came with HVM (hardware virtualization) enabled. On the Intel chips, you had to do research to figure out whether a particular Intel chip supported HVM or not.

      It was yet another reason (along with good 32/64 bit performance on the Opteron) to go AMD during the 2007-2010 time frame.

    55. Re:Good to hear. by Anonymous Coward · · Score: 0

      Not really so much any more in the integrated graphics space.

  5. "lager-scale" rollout? by Bearhouse · · Score: 1

    So, like, beer and hookers at the launch?

    1. Re:"lager-scale" rollout? by TomR+teh+Pirate · · Score: 1

      Will it be free, as in beer?

    2. Re:"lager-scale" rollout? by ilsaloving · · Score: 2

      Will it be free, as in beer?

      Better than free hookers. I'm not sure I'd trust those.

    3. Re:"lager-scale" rollout? by mobby_6kl · · Score: 1

      Well how else do you expect AMD to get some positive press for this thing?

  6. I'm ready to upgrade by Anonymous Coward · · Score: 1

    But the question is price. To upgrade from an old sandy bridge, it's going to take 500€~ to change the CPU/Mobo and memory if going the Intel route, for what I think is a good performance improvement, but it's not worth it yet.

    If they can deliver an 8 core CPU that outperforms a high end 4 Core Skylake and costs around the same, I'll nab it, thought I'm guessing getting 8 Cores for the price of 4 is out of touch with reality at release.

    Even if the 4 core chips can't compete with Intel, I'm hoping they force Intel to drop the eye-gouging prices.

    1. Re:I'm ready to upgrade by pjrc · · Score: 1

      I'm still running a Sandy Bridge E with 6 cores. I'm guessing quite a lot of non-gamer folks who do serious work on their machines are too.

      The main 2 reasons I haven't upgraded are only modest increases in performance since then, and the time & effort & disruption of upgrading. Cost is a much lesser concern.

  7. Any official word on the Win 10 requirement? by Anonymous Coward · · Score: 0

    Early 2017 will be the perfect time to upgrade my rig but I'm hesitant due to Microsoft making douchebag exclusive deals with chip makers which has already ruled out Skylake for me. Still can't find any word on if Zen will also require Win 10 or not, but that's a make or break issue.

    1. Re:Any official word on the Win 10 requirement? by sexconker · · Score: 1

      Skylake processors do not require Windows 10. They run fine on Windows 7.

      Intel's newer USB controllers intentionally fucked over interactive Windows 7 (and older) installation by forcing xHCI mode. If you tried to use a keyboard or mouse over an Intel USB port it simply wouldn't work when the installer got to a certain point.

      Because of the ensuing bitch fit and the Taiwanese motherboard manufacturers implementing their own fixes, Intel and MS were forced to release an official fix - tools for integrating the XHCI drivers into the Windows 7 installer.

      Other fixes included using a PS/2 keyboard (all glory to PS/2 and its interrupts) and an optical drive (praise be to physical installation discs) or doing an unattended installation with an answer file.

      AMD won't try to fuck people over with Zen. Windows 7 will be supported just as it is for their existing platforms.

    2. Re:Any official word on the Win 10 requirement? by gweihir · · Score: 1

      No x86 CPU requires Win10. The only process where MS support for a specific CPU has some impact is on installation, which may run a bit slower without that support. After that, just install the manufacturer driver to have all the support that makes a difference.

      The thing about "CPU support" MS is claiming is just more FUD for the clueless to force them onto the wholly unappealing and badly broken Win10.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    3. Re:Any official word on the Win 10 requirement? by yuriklastalov · · Score: 1

      This sounds eerily familiar to WinXP and SATA. Have Disk, will BSOD.

  8. Article image? by ilsaloving · · Score: 2

    Why does the article's logo show Intel?

    Has AMD falling so far that Slashdot can't even be bothered to show their logo anymore?

    1. Re:Article image? by acoustix · · Score: 1

      I thought the same thing. The logo should reflect the focus of the article, not their rival.

      --
      "A plan fiendishly clever in its intricacies"- Homer Simpson
    2. Re:Article image? by Gadget_Guy · · Score: 1

      Why does the article's logo show Intel?

      It actually shows both the Intel and AMD logos (with links to filter based on the companies) at the top of this story page, but it only shows one of the logos on the front page. They stupidly put Intel's logo first.

  9. Re:In related news by MiniMike · · Score: 5, Funny

    Bulldozers, Piledrivers, Steamrollers, and Excavators do use diesel engines. Maybe they hired the VW software engineers too late?

  10. Plus the extra money you spend on power supply... by Anonymous Coward · · Score: 0

    ...and electricity costs, you more than make up for in heating savings in the winter with AMD processors. With a smart fan arrangement you can even keep your coffee warm as the liquid coolers attempt to keep the chip from melting.

    By plugging it into the same circuit as your overhead lights it can also act as a light dimmer.

  11. Outperform in generating surprise heat? by BenJeremy · · Score: 0

    It's been a while since I had AMD on my main PC.... I got tired of their processors suddenly surging in heat.

    Simply put: I don't trust AMD.

    1. Re:Outperform in generating surprise heat? by SScorpio · · Score: 2

      Not when their main competitors Intel and Nvidia beat them in performance while generating less heat.

    2. Re:Outperform in generating surprise heat? by spire3661 · · Score: 1

      AMD components run hotter for the same performance. My Intel/Nvidia machine is already a 400 watt space heater at full load. With AMD at equivalent performance it would be closer to 500 watts or more.

      --
      Good-bye
    3. Re:Outperform in generating surprise heat? by Anonymous Coward · · Score: 0

      It really is just the age of AMD CPU's holding them back in this area; they're focused on the budget and APU market and it shows in the infrequent releases of true desktop processors.

      I will say that somehow nVidia seems to keep performance per watt up, but in that case its because they're getting more gaming performance per FLOP. In more generic workloads the AMD offerings are still more competitive on performance per dollar and per watt.

    4. Re:Outperform in generating surprise heat? by D00MSlayer · · Score: 1

      I've been running AMD's 8350 for the last 3 years with a liquid cooler(antec) and I couldn't be happier. I understand that the new wraith heatsink they developed will be more efficient, but I never trusted stock heatsinks in the past with AMD. No sudden heat surges here!

    5. Re:Outperform in generating surprise heat? by cheesybagel · · Score: 1

      AMD has had a manufacturing process disadvantage for years. Add to that the Bulldozer architecture was designed for high-end servers were multi-threaded integer performance was the main concern and power consumption back then wasn't really a priority. Zen is a whole new architecture so it should be optimized differently.

  12. SoC stuffs. by Z80a · · Score: 1

    Well, if they actually got a CPU that can actually at least don't get shamefully annihilated by the intel offerings, they will do quite well on SoC solutions with their superior GPUs, which means nice laptop deals.

    1. Re:SoC stuffs. by Anonymous Coward · · Score: 0

      But only if their SoC solutions have comparable energy budget.

    2. Re:SoC stuffs. by Z80a · · Score: 1

      Indeed.
      A 300 watt SoC laptop would fare better as a flame thrower than a laptop.

    3. Re:SoC stuffs. by spire3661 · · Score: 1

      Every time i look at a gaming laptop all i think of is 'How is this going to dissipate all that heat?'

      --
      Good-bye
    4. Re:SoC stuffs. by Z80a · · Score: 1

      Very proudly (and loudly).

  13. Alpha by Anonymous Coward · · Score: 1

    Sounds a lot like the DEC Alpha EV6 (21264)?
    https://en.m.wikipedia.org/wiki/Alpha_21264

    1. Re:Alpha by cheesybagel · · Score: 1

      Nah even AMD K8 was better than that. The 21464 (which never came out) was the one which was supposed to have SMT and large vector instructions. But then again Intel has had two-way SMT for years already and IBM even has four-way SMT in POWER 5 and later.

  14. Of cores not by fyngyrz · · Score: 1

    Even if the 4 core chips can't compete with Intel, I'm hoping they force Intel to drop the eye-gouging prices.

    Oh, come now. You don't think that Intel has actually achieved a reasonable yield in their manufacturing processes, do you? That's unpossible.

    --
    I've fallen off your lawn, and I can't get up.
    1. Re:Of cores not by Rockoon · · Score: 2

      Given the massive layoffs and 2-year delay on 10nm...

      Intel knows its in trouble. Their main two competitors (by semiconductor manufacturing market share) are TSMC and Samsung and both will have 10nm chips rolling out the door a year before Intel even begins testing its 10nm fab. Even Toshiba might beat them to 10nm.

      Zen will be at process size parity with Intel and that will probably equate to similar performance, but this fact is of only minor consideration to Intel. AMD isn't even close to Intels primary competitor. AMD catching up is a symptom of a much bigger problem. This time next year Intel wont be #1 in the market any more...it will be TSMC.

      --
      "His name was James Damore."
    2. Re:Of cores not by Kjella · · Score: 1

      Given the massive layoffs and 2-year delay on 10nm... Intel knows its in trouble. Their main two competitors (by semiconductor manufacturing market share) are TSMC and Samsung and both will have 10nm chips rolling out the door a year before Intel even begins testing its 10nm fab. Even Toshiba might beat them to 10nm.

      Intel has been testing 10nm for a long time, but they never announce milestones and test chips only finished volume products. They've lost their lead, but I doubt they'll be far behind if any.

      --
      Live today, because you never know what tomorrow brings
    3. Re:Of cores not by gweihir · · Score: 1

      Actually, I disagree. The main problem Intel has is that CPUs have not had significant speed improvements for years and that the high prices Intel asks (and apparently needs to ask) have less and less of a basis in reality. AMD however seems to have started to change to be able to survive in a commodity-market some years ago, while Intel has not even begun to do that. When it finally dawns on the last moron that Intel is asking way too much for their products, they will be in serious trouble. also because they, unlike AMD, have the actual chip-making still in-house and that has got to be pretty expensive.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    4. Re:Of cores not by Rockoon · · Score: 1

      When it finally dawns on the last moron that Intel is asking way too much for their products, they will be in serious trouble.

      You are making the mistake that I specifically tried to help you not make, by pointing out who Intels competitors really are.

      Intel makes 80x86 CPU's.. yes... BUT THAT IS NOT THEIR CORE BUSINESS

      Intel is a semiconductor manufacturer and is in more competition with names that you have never heard of than it is with AMD.

      Intel isnt downsizing because their desktop and mobile x86/x64 chips arent selling well enough.... Intel is downsizing because TSMC has taken most of their business.

      ...and as you may be aware TSMC doesn't make any fucking x86/x64 chips even though AMD has certainly want them to for more than a few years now.

      The reason high end desktop chips are so expensive isnt the demand for the chips, its the demand for the fab time. CPU's are only a small fraction of the silicon run off these fabs.

      --
      "His name was James Damore."
    5. Re:Of cores not by Bengie · · Score: 1

      The main problem Intel has is that CPUs have not had significant speed improvements for years and that the high prices Intel asks

      And yet Intel is still the fastest and best value. Here's hoping for some competition.

    6. Re:Of cores not by Anonymous Coward · · Score: 0

      You couldn't possibly be more full of shit if you tried. If you think TSMC and Samsung will have real 10nm chips rolling out first you are an absolute gullible idiot.

    7. Re:Of cores not by Anonymous Coward · · Score: 0

      You are a literal fount of misinformation. Intel's business is selling x86 CPUs. That is literally their core business and how they make the vast majority of their money. Couple this with your risible concept that they are suddenly behind TSMC/Samsung in 10nm and you should really just stop posting - you sound like an idiot. Protip: You can call any shitty old process you want "10nm", and do a paper release of chips based on it.

    8. Re:Of cores not by gweihir · · Score: 1

      That does not even deserve a reply. Arrogant and incompetent is the friendlies thing I can say.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  15. Downclocked by Anonymous Coward · · Score: 0

    Since the Broadwell-E in question has 8 cores as mentioned, one can assume it's the 6900K, which has a base frequency of 3.2 GHz and turbos to 4 GHz if not overclocked. The 3 GHz referenced in the article means that the 6900K is downclocked, either because the Zen CPU cannot reach 3.2 GHz, much less 4, or because the Blender benchmark used stops favoring Zen as soon as you run Broadwell-E at its native frequency. Either way the "result" is not to be taken seriously.

    1. Re: Downclocked by Type44Q · · Score: 1

      Sure, 'cause a "clock-for-clock comparison" is too complicated for you to grasp...

    2. Re: Downclocked by Anonymous Coward · · Score: 0

      It's a meaningless benchmark; I care about real products I can buy, not how it compares to an existing product that's been hobbled in some way to make AMD look better. How a processor delivers better performance is ultimately meaningless to me as an end user; all that matters is that it does.

    3. Re: Downclocked by PIBM · · Score: 1

      Let's say that the AMD chip only reach 2GHz and is still only marginally faster at that speed, would a clock-for-clock comparison still be interesting ? What about 1GHz ? When does it stop being useful to you ?

      The slower your chip runs the more work you can do per clock, btw.

    4. Re:Downclocked by WorBlux · · Score: 1

      Consider this benchmark. http://www.anandtech.com/bench... wherin the 6900k K beaks the same clocked and cored Piledriver by about 300%. So what is proven in this benchmark -Zen has a 50% IPC improvement for CineBench, and that it's SMT (symmetric multi-threading) solution scales quite well in the benchmark.

    5. Re:Downclocked by sexconker · · Score: 1

      Or they wanted to do a comparison against parts at the same frequency to demonstrate the IPC and architectural improvements they're talking about.
      We don't know what Zen will be clocked at. We know this engineering sample runs at 3 GHz. Typically, engineering samples are clocked significantly lower than retail parts. Keep in mind that AM4 is new as well, and the motherboards in use are also test units, not mature retail units.

    6. Re: Downclocked by Anonymous Coward · · Score: 0

      "The slower your chip runs the more work you can do per clock, btw." WHAT??!! So I'll just set it to 1GHZ and it should be blazing fast right? You fucking moron.

    7. Re: Downclocked by PIBM · · Score: 1

      How obtuse can you be. If I design a chip currently with the top of the line tools so that it can run at 1ghz, I can do more than if I was to design it with the same tools to run at 3ghz. You should try learning on the subject before commenting.

    8. Re:Downclocked by gweihir · · Score: 1

      You have no clue how engineering works. This is exactly how you do a professional benchmark.

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

      Hardly surprising since the Bulldozer and derived processors have a shared FPU between each two cores. Zen just replicates the FPU unit. I'm more interested in how it runs multi-threaded integer benchmarks. The FPU heavy applications I use nearly all run on the GPU anyway.

    10. Re:Downclocked by gustygolf · · Score: 1

      Indeed, integer performance is what I look for in a CPU too. There are large swathes of applications that have little to no need for floating point.

      That is not to say that the Piledriver's (or whatever's) FPU is in any way deficient. I'm not sure why it's getting the bad rap it gets. Is it because compilers emit 80-bit FP maths by default? [1] (The FPU can't do the 64/64bit split then, so it effectively runs at 'half speed'. Or so I guess.)

      [1] I recall there being some medium-size brouhaha happening a number of years ago when GCC decided to move to 80-bit over 64-bit for most FP operations. Even when the value will be truncated down to 64 bits anyway.

      --
      "Slow Down Cowboy! It's been 58 minutes since you last successfully posted a comment" -- slashdot, driving users away.
  16. As someone who's followed desktop CPUs by Anonymous Coward · · Score: 1

    for, let's see, a couple decades now, I welcome even the whiff of competition in this stagnant space. With Intel delivering practically non-existent top-end growth in desktop CPU performance for 5+ years now (compare an overclocked Sandy Bridge to Intel's latest to see what I mean), it's about time lowly AMD came along to give them a bump.

  17. Re:Plus the extra money you spend on power supply. by blackomegax · · Score: 1

    You're thinking pentium 4. ;p

  18. In the Ballpark by segedunum · · Score: 1

    If Zen is in the ballpark of Intel then people are going to buy it. Intel chips and especially chipsets have become too expensive and frankly, we need some competition. If the Athlon64 hadn't come along goodness knows where we'd be.

  19. Yeah, but Broadwell-E 8-core procs run at 3.2GHz.. by WilliamGeorge · · Score: 1

    So they artificially downclocked an Intel processor, and are able to *barely* beat it clock-for-clock. But that Intel processor should be running at a higher clock speed, and if they have it fixed at 3GHz then they also turned off Turbo Boost - which would have pushed the Broadwell-E chip up to 3.5GHz when all 8 cores are active. At those speeds, presumably, the Intel chip beats the AMD; if not, they wouldn't have bothered to downclock the Intel processor.

    To sum up then: AMD's next-gen, unreleased processor still cannot outperform Intel's existing model. This doesn't surprise me, though: ever since Intel caught back up with (and then passed) AMD's performance - starting back with the first Core series processors - AMD has been trying to catch up... and failing. They do have a role in the low-price / budget system market, and maybe this new Zen stuff will cost a lot less than Intel's offerings. It doesn't look like it will truly rival Intel for high performance applications, though, where the fastest speed is worth a few hundred dollars more. After all, in the grand scheme of a several thousand dollar workstation a few hundred dollars isn't a huge deal - and 5, 10, or 20% performance difference could easily pay for itself over time.

    --
    William George
  20. Sacred cows by lucm · · Score: 1

    There was a time were mentioning problems with AMD processors on Slashdot would have been as dangerous as wearing blue in a red part of the hood in Los Angeles.

    --
    lucm, indeed.
    1. Re:Sacred cows by dddux · · Score: 1

      What problems with AMD processors, may I ask? I've never had any problems with AMD processors, nor with Intel's for that matter, either. Processors are just processors. PSU/VGA/mobo on the other hand...

      --
      "It is no measure of health to be well adjusted to a profoundly sick society." - Jiddu Krishnamurti
  21. Re:Yeah, but Broadwell-E 8-core procs run at 3.2GH by Tough+Love · · Score: 1

    AMD's next-gen, unreleased processor still cannot outperform Intel's existing model.

    It only needs to get close and offer good value, then I'm in because of the superior GPU.

    --
    When all you have is a hammer, every problem starts to look like a thumb.
  22. right.... by Anonymous Coward · · Score: 0

    This article was worth a laugh. AMD will never match Intel clock-for-clock until they change their CPU's design radically. This article did make me ROFL.

    1. Re:right.... by gweihir · · Score: 2

      You seem to be stupid, because they just _did_ match Intel clock-for-clock.

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

      Zen is a ready CPU redesign in case you did not hear about. AMD just showed it actually has more IPC than Intel in a heavy FPU benchmark which was were they were weakest to begin with.

    3. Re:right.... by Trogre · · Score: 1

      They did, and they did.

      --
      "Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
  23. AMD says lots of things... by Anonymous Coward · · Score: 0

    AMD says lots of things, they never actually happen. Well not in the last 10 years, last time they were competitive on price or performance was the original Athlon (Which was a superb chip design for the time).

  24. Re:In related news by jcdr · · Score: 1

    The new CPU is "Zen", so the new staff might be buddhism monks :-D

  25. So let me get this straight by MerlynEmrys67 · · Score: 1

    AMD can beat an Intel processor from Last year with their processor that is due out next year. By that time Intel will have jumped 2 generations (almost) and where will AMD be?

    --
    I have mod points and I am not afraid to use them
    1. Re:So let me get this straight by Anonymous Coward · · Score: 0

      Broadwell-E is still the fastest offering from Intel, with Skylake-E not expected until next year. If the paltry performance increases from skylake continue on their -E platform then AMD won't have anything to worry about next year.

      Props to AMD for actually comparing to the fastest current offering instead of finding some overpriced underpowered newest offering.

    2. Re:So let me get this straight by sl3xd · · Score: 0

      If the paltry performance increases from skylake continue on their -E platform then AMD won't have anything to worry about next year.

      Except for the fact that AMD had to downclock and handcuff the Broadwell chip to even compare it on a clock-per-clock basis.

      Having approximately equal performance per tick is meaningless when your competitor is able to run 20% more ticks in the same time period.

      --
      -- Sometimes you have to turn the lights off in order to see.
    3. Re:So let me get this straight by sl3xd · · Score: 1

      This is what, the 10th season of this show?

      Every year AMD boasts next year's chips are going to beat out Intel.

      Every year AMD fanbois get frothy at the mouth because Intel will finally get what they have coming.

      And every year, two things happen:

      - AMD overestimated the speed of next year's chip by a wide margin
      - AMD underestimated Intel's performance by a wide margin

      I'm rooting for AMD, but they talk a much bigger game than they play.

      --
      -- Sometimes you have to turn the lights off in order to see.
    4. Re:So let me get this straight by Anonymous Coward · · Score: 0

      Intel is still trying desperately to recover from their bug-ridden Skylake fiasco and pushing hard to get back into mobile to save their balance sheets.

    5. Re:So let me get this straight by Anonymous Coward · · Score: 0

      This is a fair critique if we're talking about something that's currently in production. AMD has certainly had problems keeping up in clocks and general fabrication quality with Intel, and hopefully they can become more competitive in this area in the future. That said, this isn't a production chip, so I think it's more relevant to compare the architecture now, and the fabrication and clocks once it gets closer to release.

      Having a stable unit that runs at only 15% lower than a target for production hopefully indicates the fabrication won't be too difficult.

    6. Re:So let me get this straight by Luthair · · Score: 1

      They only people I've heard claim Skylake as buggy are Microsoft folks. Oddly no other brand of laptops appears to have any issues with Skylake. So what does Occam's razor say?

    7. Re:So let me get this straight by gweihir · · Score: 1

      Well, if Intel produces "speed ups" as they did for the last few years, they will be at most 10% faster with those 2 "generations".

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    8. Re:So let me get this straight by Anonymous Coward · · Score: 0

      AMD will still, as always, be at best value for money, for the people who realize that you don't need the fastest CPU available to be able to turn the computer on. People like you for some reason make it sound like AMD CPUs are delivering tiny fractions of the performance that Intel CPUs have, or that "2 generations" means ten times faster, while it's usually around 10-20%.

      You can either buy the AMD, or spend 100% more money on the Intel for 10-20% more CPU power.

    9. Re:So let me get this straight by Anonymous Coward · · Score: 0

      really?

      http://www.theinquirer.net/inquirer/news/2441458/intel-admits-to-skylake-bug-that-freezes-windows-and-linux-systems

    10. Re:So let me get this straight by cheesybagel · · Score: 1

      No if you actually read AMD's own press releases they have been claiming at least since Piledriver that they are competitive in price/performance (which is true) and that were going to only do minor IPC improvements until their new architecture (i.e. Zen) came out.

      Of course AMD is limited by available manufacturing technology regardless of how good their chip design is. They don't own fabs anymore...

    11. Re:So let me get this straight by Bengie · · Score: 1

      $180 PSU, $150 mobo, $150 memory, $400 few SSDs, $60 case, $200 monitor, $300 GPU, $70 Intel network card. Not to mention the $30 each for mag-lev bearing fans. Yep, I really want to save $50 on a CPU with heat and power issues.

      I came from a poor family, I had to earn my own money to buy computer parts when I was a child. I've learned to appreciate quality. If AMD can get within 10% of Intel in performance per core and efficiency, I will support the underdog. I really want a bunch of cores and ECC memory on my desktop, but Xeons are too expensive.

  26. Re:Yeah, but Broadwell-E 8-core procs run at 3.2GH by serviscope_minor · · Score: 1

    So they artificially downclocked an Intel processor, and are able to *barely* beat it clock-for-clock.

    In other news, AMD who has been languishing behind severely in the clock-for-clock performance actually beats an Intel CPU clock for clock for the first time since the Core 2 processor.

    Matching it clock for clock is a HUGE step forwards. They're beating it, just, but then 3.2 is not much more than 3, so it'd be neck and neck with that taken in to account. And I expect that AMD will do some bin sorting when the volume ramps to get higher clocking parts.

    --
    SJW n. One who posts facts.
  27. Hair by Anonymous Coward · · Score: 0

    "by a hair". Knowing companies specifically choose benchmarks to make their own products look good, winning "by a hair" makes this product look awful. And AMD is betting everything on this? AMD will be sold next year.

  28. Re:Plus the extra money you spend on power supply. by armanox · · Score: 1

    My Athlon certainly heated a room up. I have a FX-8120 that makes a good bit of heat too (compared to my Intel systems (Sandy Bridge and Ivy Bridge)).

    --
    I'm starting to think GNU is the problem with "GNU/Linux" these days.
  29. Will it have a trustzone (AMD PSP) processor? by Anonymous Coward · · Score: 2, Interesting

    I wonder if it will have AMD's equivalent to Intel AMT, the Platform Security Processor. If so, it may be a no-go for some people.

  30. Re: In related news by Anonymous Coward · · Score: 0

    Only if the engineers implant software which when turned on show the chip at amazing speeds and tiny power consumption but when handed to retailers it's turned off. Then I would believe they are VW engineers.

  31. Re:Yeah, but Broadwell-E 8-core procs run at 3.2GH by sl3xd · · Score: 1

    Matching it clock for clock is a HUGE step forwards.

    And here I thought we had finally dispelled the notion that clock speed was all-important.

    What matters is throughput per unit of time. It doesn't matter if they get throughput by using higher clock speeds or by more work per tick.

    AMD is still being beaten badly at throughput.

    --
    -- Sometimes you have to turn the lights off in order to see.
  32. Re:Yeah, but Broadwell-E 8-core procs run at 3.2GH by Anonymous Coward · · Score: 0

    And better open source support with AMD

  33. Re:Yeah, but Broadwell-E 8-core procs run at 3.2GH by Anonymous Coward · · Score: 0

    lolwut

  34. Re:Yeah, but Broadwell-E 8-core procs run at 3.2GH by serviscope_minor · · Score: 1

    AMD is still being beaten badly at throughput.

    How on earth did you work that out? It's marginally faster clock for clock and 7% slower on clock speed. Sounds like a dead heat on throughput to me.

    --
    SJW n. One who posts facts.
  35. Intel by backslashdot · · Score: 1

    Intel doesn't know how to design GPUs .. thats where they need to compete. I need 8K VR gaming at 120 fps.

  36. Re:In related news by gweihir · · Score: 3, Insightful

    You AMD-haters are really the most stupid morons around. Don't you realize that the only thing that AMD folding will do is that Intel improvements will vanish and Intel prices will skyrocket? Or maybe you people are into self-abuse?

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  37. Re:Yeah, but Broadwell-E 8-core procs run at 3.2GH by WhoBeDaPlaya · · Score: 1

    You DO get the point of an IPC comparison, no?

  38. oh please AMD by Anonymous Coward · · Score: 0

    they are the ryan lochte of chip companies they say all kinds of things then it turns out to be baloney

  39. Re:Yeah, but Broadwell-E 8-core procs run at 3.2GH by wbr1 · · Score: 1

    The high end WS market is a small part of the market. If AMD pulls off 75-85% of the performance in a reasonable power envelope for 45-50% of the price, it is a win-win for a huge segment of the market.

    --
    Silence is a state of mime.
  40. All their new chips do :( by Anonymous Coward · · Score: 0

    I was going to write a long post explaining exactly that. Put simply: AMD now has the same disadvantage as Intel, meaning any chip newer than the AM3+ generation or the 990FX chipset should be considered suspect. Additionally, you cannot get IOMMUv2 support WITHOUT a newer chipset, meaning your only options for 64bit BAR support needed for the Heterogenous systems architecture (Be it AMDs implementation, or the Xeon Phi implementation) are LGA2011/current gen 115x socket, or FM2+/FM3+ AMD socket hardware. Both of which have the signed management engine firmware, and going off the last generation firmware's exploits, probably have all sorts of holes to allow system level memory probing without knowledge of the x86 processors. Completely defeating any trust you would have in those computers for secure computing (not to be confused with Secure ComputingTM, which is only slightly secure and only for content providers DRM.)

  41. There was that, but they did well before by Sycraft-fu · · Score: 1

    The Athlon was very competitive with the P3, which was an exceedingly solid processor. So it wasn't just that Intel screwed up, but AMD had a well performing product to start with.

    But then ya, they really slowed down and stopped improving. They kept rehashing the same architecture over and over. They introduced new features, like 64-bit, but the computational architecture was fundamentally the same. Meanwhile Intel was hard at work making the Core series and just continually improving.

    Also AMD had a real problem in that while the Athlon was a good performer, the motherboard chipsets for it were fucking garbage. So the experience of owning an Athlon was a real mixed one and turned some people off. I got burned really badly by the original Athlon and compatibility issues with their motherboards and was turned off to AMD for some time because of it.

    1. Re:There was that, but they did well before by cheesybagel · · Score: 1

      The AMD chipsets that they themselves designed and manufactured were pretty solid if kind of poorly featured. It was the VIA chipsets which were crap. Especially the southbridges.

    2. Re:There was that, but they did well before by dbIII · · Score: 1

      But then ya, they really slowed down and stopped improving

      They sacked a lot of people to cut down on "cost centers" such as the development people designing stuff that would make them profitable in the future.

  42. Re:In related news by ArhcAngel · · Score: 0

    WOW! What stupid brain-dead comment. I don't hate AMD at all. I have used AMD products repeatedly over the years. This was a joke. It appears the SJW's have taken over the moderation system and the cry babies are coming out of the woodwork. I have even had numerous comments deleted in recent months despite this site claiming it never does. It's a sad, sad day for /.

    --
    "A person is smart. People are dumb, panicky dangerous animals and you know it." - K
  43. Re:Yeah, but Broadwell-E 8-core procs run at 3.2GH by gweihir · · Score: 1

    That is complete nonsense. You are _really_ clueless. First, you do comparisons at the same clock, anything else is unprofessional. The actual clock-rate does not matter. And second, for this type of architecture-benchmark, you always go for a "round" clock figure. Fortunately, the target for these benchmarks is people with an actual clue.

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

    Well, maybe entertain the notion that the problem is on your side? Because to everybody else that is rather obvious at this time.

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  45. Not so optimistic by Anonymous Coward · · Score: 0

    Intel chips score twice as much as an equivalent AMD chip on the single-threaded on the Cinebench R15 benchmarks (see http://www.anandtech.com/bench/CPU/1028), so AMD would need much more than a 40% improvement over their present chips to be competitive.

  46. True apples to apples comparison even possible? by Anonymous Coward · · Score: 0

    Anything compiled with the Intel Compiler will make AMD performance drop by default. Is there a true benchmarking test that does not provide boosts to any chip maker?

  47. Re:Plus the extra money you spend on power supply. by cheesybagel · · Score: 1

    Zen is a new architecture designed by Jim Keller, the same guy who designed the AMD Sledgehammer and Apple A4/A5 mobile processors. It shouldn't run as hot as Bulldozer/Piledriver/etc.

  48. Below bottom of barrel by dbIII · · Score: 1

    You can buy a used

    You are seriously taking that line?
    That's a pretty massive shift of the goalposts to get price-performance numbers to work the way you want instead of the way they are likely to be.

  49. Re:In related news by unixisc · · Score: 1

    Intel's biggest competitor has not been AMD for at least 10 years. It has been their own existing base - the new Intel CPUs have been unable to displace the existing Intel CPUs, once they went multi-core. Intel's newer CPUs like Skylake only have a chance if they are priced attractively.

  50. Re:In related news by yuriklastalov · · Score: 0

    Get back on narrative, knave.

    Repeat after me "There is no God but Social Justice. We have nothing to lose but our chains"

  51. Finally, an AMD CPU that's Good Enough by Shirley+Marquez · · Score: 1

    Zen probably won't be able to beat the very best chips that Intel has to offer. And it's going to lag a bit more in single threaded benchmarks; part of Zen's strength is in multithread performance. But it's going to be far more competitive then the company has been for years. Combine that with appropriate pricing and they will have a winner. They won't be able to charge $1000 or more for any of the Zen products but they can still enjoy healthier margins than they've been able to get for the FX line or for the current A-series APUs.

    As for choosing Blender, they probably deliberately chose something that makes full use of all the cores and that uses the FPU heavily; those are AMD strengths. And also an application that isn't built with an Intel compiler. Intel's compilers are known for producing code that works badly on AMD processors.

  52. Re:Yeah, but Broadwell-E 8-core procs run at 3.2GH by dddux · · Score: 1

    Let me put it this way for you: let's say you have a choice of two cars, a Porsche for $60000 and a Ferrari for $100000. A Porsche has a top speed of 260km/h and Ferrari has a top speed of 300km/h. Also, Porsche consumes a bit less fuel. Which one would you choose? Do you go for more expensive car just because you like Ferrari even though you don't need a 40km/h faster car? And it's certainly better to have a car that consumes less fuel, too. That's the problem I have with Intel fanboys. I really, really don't care if my car is a *bit* slower. It's not an earth-shattering difference. But I do care if I save some money and use it for something more important than having a 10% faster CPU, you know? It's a ridiculous pissing competition. What's the really most important thing these days is price and performance/watt ratio, nothing else. I couldn't care less about the 10% more speed. That's the last point people should consider nowadays.

    --
    "It is no measure of health to be well adjusted to a profoundly sick society." - Jiddu Krishnamurti