Slashdot Mirror


The Chip That Changed the World: AMD's 64-bit FX-51, Ten Years Later

Dputiger writes "It's been a decade since AMD's Athlon 64 FX-51 debuted — and launched the 64-bit x86 extensions that power the desktop and laptop world today. After a year of being bludgeoned by the P4, AMD roared back with a vengeance, kicking off a brief golden age for its own products, and seizing significant market share in desktops and servers." Although the Opteron was around before, it cost a pretty penny. I'm not sure it's fair to say that the P4 was really bludgeoning the Athlon XP though (higher clock speeds, but NetBurst is everyone's favorite Intel microarchitecture to hate). Check out the Athlon 64 FX review roundup from 2003.

259 comments

  1. The old days by Thanshin · · Score: 5, Insightful

    Those were the good old days. How I miss when it took me one day at most to learn about all options I had to build a gaming computer, with enough detail to make an informed decision about what bits and pieces to built it with.

    Nowadays just piercing the veil of lies, half truths, false reports and bought reviews, makes the entire process incredibly boring and frustrating.

    1. Re:The old days by Dunbal · · Score: 5, Insightful

      It's still pretty much common sense. You want a fast CPU, so not the top of the line $1000 chip, take a step back or two and go for the one selling in the $300-$500 range. Motherboard for that chip from someone you trust - ASUS, Gigabyte, etc. Again never the $500 "gamer" board, take a step back, there are some really nice ones for $200 or so. Latest generation graphics card, or top end from last generation (assuming the prices have come down), plenty of memory on the card. Power supply that can feed the card what it needs and then some. Plenty of system RAM. SSD hard drive. Water/Air cooling system for your CPU type. And you're set! Shouldn't take a whole "day" to check those out. An hour or two would suffice.

      --
      Seven puppies were harmed during the making of this post.
    2. Re:The old days by ArcadeMan · · Score: 4, Insightful

      The good old days was the 286 era, when all you needed to know what the clock speed of the CPU, EGA was four times better than CGA and SoundBlaster was AdLib compatible.

      Of course, you had to deal with XMS and EMS memory settings, loading your mouse driver into high memory and solving IRQ and DMA conflicts between your ISA add-on cards.

      Screw that, the good old days are today. Take out the iMac from the box, plug it in the wall socket and start using it right away.

    3. Re:The old days by Anonymous Coward · · Score: 1

      The only difference is that today you know you're being lied to. It's called growing up.

    4. Re:The old days by Nadaka · · Score: 1

      These days, aim for a price point of $1k with competitively priced components and you are almost certain get a decent gaming rig. PC hardware is far ahead of the curve thanks in part to extreme production costs of high quality graphics, and also in part to console hardware holding back the standard quality settings for multi-platform releases. That will give you medium of better settings at 1080p on all current and foreseeable games.

    5. Re:The old days by h4rr4r · · Score: 1

      $1000? You can beat the consoles for less than $500.
      If you keep your old case and dvd/bluray drive you can do even better. I tend to swap out MOBO, RAM, CPU and GPU in one shot every few years. I have not had to do this since I picked up my GTX465.

      The consoles are holding gaming so far back there is no point in spending even $1000.

    6. Re:The old days by Anonymous Coward · · Score: 0

      I can attest to this. The current computer I have(which is outdated, but still very fast), has DDR2 memory, the Q9550 Yorkfield Core 2 Quad Intel processor and an nVidia 260 GTX. None of the stuff I bought was high-end at the time, excluding the Yorkfield processor which was brand new, but even then it was 200-400(can't remember if it was 200-300 or 300-400, but somewhere in that range). My motherboard had the highest speed for DDR2 ram and cost ~$170. The graphics card was in the $200-300 range. At the time, the popular gamer benchmark was running Crysis, which this computer could do for hours on end without slowing down(because it took a long time to leak through 8GB of memory). This particular computer has never had a problem - ever - it destroys anything I throw at it. The only problem? Newer PC games are coming out that will utilize better graphics, and in particular my graphics card doesn't do directx 11 very well. In order to upgrade I really need new RAM(DDR3), which means new mobo, which means new processor...so I might as well buy a new computer outright.

      I'm currently planning on buying the parts to build a new computer, but only because I like building computers and would like better graphics. I've specced out a computer for ~$2,500(including 2 monitors, and Windows 7 64-Bit...not sure which edition just yet), but really my current computer has no problems at all. I'll probably sell my current computer for 800-1,500 to recoup some of the cost of my new computer. I probably spent 8 hours sifting through tom's hardware reports on various parts. Not necessarily so I could make an informed decision, but because I genuinely enjoy reading the articles comparing the latest hardware. I'm not buying the parts right now, probably Q1-Q2 next year, so I'll get to wait and see how Haswell pans out, and there will probably be slight modifications to what I've picked out so far.

    7. Re:The old days by Anonymous Coward · · Score: 0

      Unless you want more out of your computer than gaming. Or if you're a PC modding enthusiast. Or if you don't want to upgrade every year, and want a machine that will last for a decade.

    8. Re:The old days by OolimPhon · · Score: 3, Interesting

      You claim to be a geek and you're contemplating getting rid of an old computer?

      All my old computers ended up being used for something else. I only get rid of them when the architecture is so old that <OS of choice> won't run on it any more (or when the smoke comes out!). Device drivers are the things that limit usage to me.

    9. Re:The old days by asmkm22 · · Score: 3, Informative

      Not really sure what you're smoking. It's much easier to put together a computer (including a gaming computer) these days than it was 10 years ago. We don't really have to worry if we need PC-133, PC-2700, DDR1, DDR2, etc.. There's no need to choose between AGP, PCI, or that new-fangled PCI-Express, much less whatever multiplier is involved. Hard drives are straight up SATA now, and it doesn't matter if you choose a disk or SSD type. The graphics cards themselves aren't even as important since the console cycle has pretty much bottlenecked as a result of developers focusing on those consoles first and foremost. We don't need to do much more than make sure the motherboard is either an Intel or AMD socket.

      In fact, about the only real difficult decision you might need to make these days is finding a computer case that has enough room to use a modern video card.

    10. Re:The old days by bemymonkey · · Score: 1

      To be honest, I'm loving the ease of putting together a decent system these days. I actually owned an Athlon64 based system back in the day (with an expensive high-end SLI nForce-based board), and that sucker was never completely stable. Same thing with the AthlonXP generation, and K7 (Athlon/Duron) beforehand...

      These days, I just pick the Intel chip that fits my needs, by the cheapest name-brand board that fits my needs, slap it together and it's rock solid. Celerons, Pentiums, Core iX, whatever... hell, even overclocking 30-40% (I'm sitting next to a Sandy Bridge machine that's running at about 45% OC'd) only takes a few seconds and is rock solid.

      From what I've heard, AMD's current chips are similarly stable, but I'm not really willing to risk it.

    11. Re:The old days by h4rr4r · · Score: 1

      My machine also runs KVM, and various database software. I have not upgraded since that video card, which is more than a year old.

    12. Re:The old days by Nadaka · · Score: 1

      $1000 includes all parts of a computer, including monitor. Reusing previously bought parts does not greatly reduce the actual cost of the computer as you lose the opportunity to use that other hardware for other purposes.

      And that $1000 will likely play all current games at high settings, and have the ability to play foreseeable future games with at least medium settings.

      A new generation of consoles is coming out, and the base quality of graphics will rise to meet the abilities of that new hardware, and that may go beyond where your $500 machine can keep up.

    13. Re:The old days by interkin3tic · · Score: 1

      I'd need a day before I felt that I actually knew enough to start buying rather than just THINKING I knew enough.

    14. Re:The old days by Dagger2 · · Score: 4, Informative

      And then you end up either with an i7 4770 which has a locked multiplier, or a 4770K which doesn't do VT-d. Then you realize that there is no Intel CPU that'll do both. So then you start looking at AMD, in the hope that they don't pull shit like that with their CPU models. And then you're way over your hour or two budget.

    15. Re:The old days by h4rr4r · · Score: 1

      How many monitors do you need?
      If you had no other use for that hardware it sure does. SSDs, monitors, good cases those are expensive.

      We have seen the new consoles already and no they will not exceed a $500 PC at this point.

    16. Re:The old days by SB9876 · · Score: 2

      I'll second that. You haven't known pain until you try to get Ultima 7 to run on a system with a Proaudio Spectrum 16 sound card.

    17. Re:The old days by Anonymous Coward · · Score: 1

      800 to 1500 for your old computer?

      What do you think you have in there?

      And 2500 for a new computer?? Damn, you are way overpriced with your target there.

    18. Re:The old days by yurtinus · · Score: 5, Funny

      Y'know, I was enjoying reading all the little nuggets of wisdom (Video cards that could use as much as 512 mb of address space, $700 for 2GB of RAM). Then I was thinking "hey, the computer I had before this one was an Athlon 64, it wasn't *that* long ago!" Then I realized it was. Then I felt old. Now I'm crying.

      --
      +1 Disagree
    19. Re:The old days by Bigbutt · · Score: 2

      I did that for a bit but when I got to seven computers sitting idle in the closet, I took them down to the electronics recycling bin. Heck, I'm even looking at my old Sun box and considering punting that one as well. That will leave me with 4 computers that are regularly in use plus the tablet and phone.

      [John]

      --
      Shit better not happen!
    20. Re:The old days by yurtinus · · Score: 1

      Or if you don't want to upgrade every year, and want a machine that will last for a decade.

      ...like an Athlon 64!?!?!

      --
      +1 Disagree
    21. Re:The old days by Anonymous Coward · · Score: 0

      I'm not your buddy, buddy.

    22. Re:The old days by Anonymous Coward · · Score: 0

      Yeah, I had a Sparc 5, Sparc 10, and an Ultra 5 that were just jettisoned.. especially considering that I can run Solaris x86 in VMWare if I need to do Solaris work. A bit depressing, oddly enough. There was something sexy about RISC workstations back in the day. I've had to restrain myself from buying a dual-head SGI Indigo2 on Ebay for no good reason.

      But.. to your point.. they do eat up space, and if used, oft-times power as well. A friend bought a SPARC based Solbourne and after the next power bill the folly of running it as a toy machine showed him the error of his ways.

    23. Re:The old days by armanox · · Score: 1

      Unless you're running HyperV or Xen, VT-d doesn't matter.

      --
      I'm starting to think GNU is the problem with "GNU/Linux" these days.
    24. Re:The old days by nucrash · · Score: 1

      I had to retire my pair of 9800 GX2s. Apparently running folding on a computer that isn't properly ventilated pisses off EVGA after about the 3rd replaced video card. In other news, I managed get about 10 million points or so before retiring them. :/

      --
      Place something witty here
    25. Re:The old days by Anonymous Coward · · Score: 0

      Not to mention the fun of bragging rights being proportional with the amount of free low memory available. Secret incantations and tricks that freed 2k were treasured.

    26. Re:The old days by Anonymous Coward · · Score: 0

      lol he just said iMac. That's funny.

    27. Re:The old days by serviscope_minor · · Score: 2

      Unless you're running HyperV or Xen, VT-d doesn't matter.

      Or if you want to virtualise a random piece of hardware that your primary OS doesn't have drivers for. Like the heaps of hardware with XP only drivers, for example.

      --
      SJW n. One who posts facts.
    28. Re:The old days by adolf · · Score: 1

      You think that was pain? Try it with a Gravis UltraSound.

    29. Re:The old days by Burning1 · · Score: 2

      Would have been nice if you'd put a TL;DR at the top that this is an apple propaganda piece.

      The specs you listed above are for a gaming computer. Your Mac is a nice machine and it can certainly play some games, but it wouldn't be ideal for that purpose.

    30. Re:The old days by Anonymous Coward · · Score: 0

      I'll second that. You haven't known pain until you try to get Ultima 7 to run on a system with a Proaudio Spectrum 16 sound card.
      Been there, done that... and more. PAS-16 for digital and Gravis Ultrasound for MIDI. Fun times.

      Cutemouse was my savior. It was amazing how much extra stuff I could jam into upper memory once I started using it.

      Anyway, I always had more problems getting Falcon 3.0 running than Ultima 7. IIRC, it needed something like 615kb low memory to run with all the bells and whistles.

    31. Re:The old days by vux984 · · Score: 1

      I tend to make this years gaming rig next years home-office server.

      So the core2quad Q6600 I used to use in my gaming PC is now running XenServer 6.

      I realize I'm in the very distinct minority here, but still ... it would be nice if i could buy a product that does both.

    32. Re:The old days by Bigbutt · · Score: 2

      Two of the ones I punted this time were Ultra 60's. Back in 2004 when I moved, I got rid of an SLC I had for several years. This one is an Enterprise 250. I have to remove the power supplies in order to pick it up. It's fricking loud though. :)

      [John]

      --
      Shit better not happen!
    33. Re:The old days by 0123456 · · Score: 1

      Nah, forget it. As a computer software nerd, but not a PC building nerd, I'll just go with a 27" iMac for $1999. Granted only an i5 CPU and 8 GB, but comes with a great OS and a gorgeous 27" monitor. (BTO with i7, 16GB and 256GB SSD bumps the price to $2599.) It has a GTX 775M instead of GTX 660 - no idea which is faster. At least I know all the components will work together, and they're properly supported by the OS.

      I built a new games machine last year. That had the second-fastest i7 at the time, 32GB of RAM, the GTX660 GPU you mentioned, a 200GB-ish SSD, 3TB hard drive and a few other bells and whistles. Even including $100 for Windows, it only cost $1500.

      So that had better be a really, really gorgeous monitor.

    34. Re:The old days by Anonymous Coward · · Score: 2, Insightful

      So then you start looking at AMD

      Your problem is starting with Intel for a CPU, the company that thinks the only way to win a CPU benchmark is to cripple their own compiler (the best x86 compiler at the time and used in many benchmarks for this reason). It required a curt case for them to admit that the compiled binaries would cripple themselves* on any non intel CPU . Intels lack of trust into their own CPUs and their attempts to manipulate benchmarks speaks volumes

      * x86 and x64 CPUs contain optional instructions and feature flags to check for their existance. Compilers generate code based on different feature sets and put all of them into the program binary. At load time the program checks against the feature flags and loads the fastest available code path for the available features or falls back to a less optimal code. Programs produced by the intel compiler do an additional check against the vendor ID hardcoded into the CPU and load the worst code path for any non Intel CPU .

    35. Re:The old days by NJRoadfan · · Score: 3

      I miss the old days where I didn't have to consult Intel's website to figure out what the model numbers mean. It use to be easy, the CPU has a name and a speed rating which told you how fast the chip was and the number of cores at a glance. Now we get a jumble of numbers to decipher.

    36. Re:The old days by jandrese · · Score: 1

      $600 is way too much to spend for a CPU unless you're all about big dick benchmark numbers. $200-250 is a much better price-performance spot. $200-250 is also a good spot for a video card. You also don't need a $200 mobo most likely, the $100 or maybe $150 models are much more reasonable. $100 is plenty for the Power supply too. What the heck do you need 16GB of memory for too? 8GB will be plenty until it is time to retire this machine. $50-75 is more than enough for a nice case as well. Big cheap SSDs don't really exist, but you can find 256GB ones for ~$175 or so.

      Put all of that together and you're talking about ~$1000 for a nice gaming machine, minus the keyboard, mouse, monitor, or software. Most people who are building their own hardware already have all of those and upgrade them on a different timescale than their CPU. You can likely shave off some of the cost by dealhunting as well assuming you aren't heavily time constrained.

      --

      I read the internet for the articles.
    37. Re:The old days by jandrese · · Score: 2

      It's rediculous just how long a properly built machine will last, even if you are a gamer. I'm still using my 2.4Ghz C2D from 6 years ago and it's only now starting to fall below the minimum requirements for some games.

      --

      I read the internet for the articles.
    38. Re:The old days by elistan · · Score: 1

      The specs you listed above are for a gaming computer. Your Mac is a nice machine and it can certainly play some games, but it wouldn't be ideal for that purpose.

      Yes, you're right. I really wasn't even planning on mentioning the Mac at the end, I was trying out GP's suggestion that one could spec out a DIY computer in just a couple hours. So I followed their suggestions on how to pick parts as best I could, and I was rather surprised at the end price. Made me think of the stereotypically overpriced computer, the iMac. So that was a spontaneous addition at the end. My apologies. Anyway, yes, if I go back and redo the component list with cheaper parts, it will probably take me just another 25 minutes like before. Something with a sub-$200 CPU, sub-$100 motherboard, 4GB RAM, 256GB HDD, etc.? Probably under $800 for a headless system?

    39. Re:The old days by elistan · · Score: 0

      I built a new games machine last year. That had the second-fastest i7 at the time, 32GB of RAM, the GTX660 GPU you mentioned, a 200GB-ish SSD, 3TB hard drive and a few other bells and whistles. Even including $100 for Windows, it only cost $1500.

      That sounds like a much more reasonable price. Where'd I go wrong? GP suggested taking just a step or two back from the top-of-the-line, so instead of picking a $1000 CPU, I picked a $570 one. GPU was under $300 instead of $1000. Etc. How'd you manage to put together a machine that sounds like every bit as fast or faster as the one I spec'ed out, but for $443 less?

      Perhaps the GP's suggestion on how to spec a PC in an hour or two will result in a nice, but overpriced, machine? Perhaps GGP Thanshin's comment about being an informed builder of a PC taking more than a day is correct?

    40. Re:The old days by Anonymous Coward · · Score: 0

      Unless you're running HyperV or Xen, VT-d doesn't matter.

      Or if you want to virtualise a random piece of hardware that your primary OS doesn't have drivers for. Like the heaps of hardware with XP only drivers, for example.

      Or you can realize that we're talking about building a *new* computer, with *new* hardware.

    41. Re:The old days by Anonymous Coward · · Score: 0

      An SLC!?! OMG.. can you say 'tftp' and 'network boot' :)

    42. Re:The old days by Anonymous Coward · · Score: 0

      2 monitors, OS, mouse, keyboard, case, cooling, speakers, headphones, SSD, DVD/Bluray drive.

      Sure you can spend $500 on CPU, GPU, RAM, Mobo, PSU, and HDD. But to build a computer from scratch (involving purchases of all of the above) is gonna cost you about $1000.

      P.S. there's no way you're beating the next gen consoles in image quality without spending more than $500 on a PC. There's a lot more to consider than raw flops.

    43. Re:The old days by elistan · · Score: 0

      You can likely shave off some of the cost by dealhunting as well assuming you aren't heavily time constrained.

      Emphasis is mine. That was mostly the point of my spec-listing exercise. GP suggested a machine could be spec'ed out in an hour or two, but based on the feedback I'm getting, their methodology results in a pretty pricey setup. Seems like GGP was more accurate about the time it takes to put together an affordable but well performing computer. (However I'm not sure GGP was correct about it being easier to do back then than now. Seems like either way, a well thought out build will take a solid day or two of research?)

    44. Re:The old days by courteaudotbiz · · Score: 1

      And then, try running a 1982 GE laser engraving machine on a 286 through a proprietary ISA interface card, with help from DOS 6.2. These are the good ol'days... Now get of my lawn!

    45. Re:The old days by bored · · Score: 3, Insightful

      Or you can realize that we're talking about building a *new* computer, with *new* hardware.

      Which doesn't do him a lick of good if he wants the new computer to run his old $5000 data acquisition hardware that only has XP drivers. Or dozens of other pieces of hardware which may have newer versions supported by newer OS's but the price of replacement is significantly more than the computer.

       

    46. Re:The old days by bored · · Score: 2

      You haven't known pain until you try to get Ultima 7 to run on a system with a Proaudio Spectrum 16

      IIRC, with that version of ultima it wasn't your PAS that was the problem, it was the game. That darn game was a buggy POS even a couple years after release.

      My game machines from that era always had the latest Sound Blaster (even though I also owned a PAS and a Gravis (actually still have the gravis)) because then tended to "just work". That is until PCI came out, in which case nothing really worked for a couple years.

    47. Re:The old days by Anonymous Coward · · Score: 0

      What fucking garbage hardware are you using that has XP only drivers?

    48. Re:The old days by Anonymous Coward · · Score: 1

      Not to mention ECC support. Yes, inside games it may reduce performance some, but when not gaming I prefer not to lose the contents of main memory and it can readily be turned off when gaming. This then narrows it down to a few processors that lack other high-value features. Ugh!

      .

    49. Re:The old days by master_kaos · · Score: 1

      depends if you know absolutely NOTHING about hardware, yeah you will probably need to spend a week researching and probably even consulting friends or people on a forum. If you know even a little about hardware (say you swapped a video card in the past) the time could drop to a day. If you specced out a computer in the past, most you would be looking at is a couple of hours as you just need to catch up on sockets and video cards and bang for the buck range.

      GP price points are right on for a bang for the buck system. Can definitely go cheaper but you will be upgrading sooner, but going more expensive won't extend shelf life enough (maybe 6 months)
      $230 for CPU (i5)
      $230 for GPU
      $120 mobo
      $80 PSU
      Ram used to be cheap and could get 8 gigs for $50 but price shot up, think closer to $80 now
      $80 case
        SSD .. depends on what size you need. say a 120GB for $80ish or 256GB for $170
      $70 for 1TB HDD

      So yeah around $1100 give or take $200 depending on options gets you a great gaming rig that will play anything now and for next couple years (maybe not on ultra though). Like GP said, usually when you are upgrading your computer you don't need input devices or monitor

      People who are pricing out 2k+ systems for gaming are nuts. If you have 2k to spend you are much better off spending 1k now, then building another computer 2 or 3 years from now for another 1k. The 1k computer you build in 3 years time would be at least double if not triple the speed (for gaming) of the 2k one you build now, or better yet just upgrade the video card in 2 years instead and spend only $250.

    50. Re:The old days by operagost · · Score: 1

      Power supply that can feed the card what it needs and then some.

      That part's funny, because it used to be that you had to account for the hard disk and CPU consumption; now the GPU takes more power that either of those.

      --

      Gamingmuseum.com: Give your 3D accelerator a rest.
    51. Re:The old days by Belial6 · · Score: 1

      Yeah. I worked at a company where the VP wrote the compiler for the custom language we were writing in. It required 650k of base memory to run. That was fun. While he was on vacation, a couple of the system developers went in and adjusted the memory usage so that it could run with ~300k of base memory. When the VP got back, he was pissed and demanded that they restore the application to the old system that took 650k out of the base memory to function.

    52. Re:The old days by master_kaos · · Score: 1

      yeah at low graphics, I personally like to run everything on high, which means I generally upgrade ram and GPU every 2 years (although maybe I won't need to do ram every 2 years anymore), and a full system rebuild every 4.

    53. Re:The old days by operagost · · Score: 1

      Overclocking has always been, and will probably always be, expert-level shit. Except for OCing those Celeron 300As to 450... that was child's play.

      --

      Gamingmuseum.com: Give your 3D accelerator a rest.
    54. Re:The old days by Anonymous Coward · · Score: 0

      You are not suppose to give a crap for model numbers!

    55. Re:The old days by Anonymous Coward · · Score: 0

      Ah yes.. Orign aptly named their memory manager Voodoo.. Frickin black magic that was!

    56. Re:The old days by dow · · Score: 1

      The PC specced was mismatched. The processor he picked is a 6 core, and that socket it known for being the line for hardware enthusiasts trying to build something very special. It uses 4 channels of DDR3 for example. Better to go with the other current socket and a 4770k processor.

      Combined with this processor if wanting to play games the 660gtx isn't the best choice at the moment. Top of the line for previous generation would be a 680gtx which is exactly the same card as current 770gtx I believe, both great cards for gaming with a 1080p 60hz screen. Going a little above these cards would be a good idea if planning to play with a 120hz screen or use one of the higher resolution screens available now.

      Making the decision between higher refresh rates and higher resolution screens is a tricky one. I went with higher refresh rate which is great for online first person shooters, but with single player games the beauty which higher resolutions brings out in some titles makes me think I made the wrong decision sometimes.

      Good luck with your method, Hardware and constantly swapping components is to me still a hobby. Since I built this PC a few years ago I have used 4 hard disks, 3 graphics cards, 3 soundcards, 2 processors, 2 cpu coolers and 2 different sets of ram. No components were broken, and the ram was the same capacity just different speed ratings.

    57. Re:The old days by Anonymous Coward · · Score: 0

      Get off my lawn or my trusty Amiga 500 will vaporize your punk ass!

    58. Re:The old days by operagost · · Score: 1

      I miss my PAS. That, and Turtle Beach when it was in Pennsylvania and sold entry-level professional cards and software.

      --

      Gamingmuseum.com: Give your 3D accelerator a rest.
    59. Re:The old days by jandrese · · Score: 1

      It's not even that low. I did replace the graphics card a couple of years ago because my old one died, but anything that's a console port still runs at max settings just fine, and most AAA games these days are console ports. Indie games rarely push the graphics very hard either. I had some slowdown in Gone Home, but I'm not sure that wasn't just a bug in the game.

      --

      I read the internet for the articles.
    60. Re:The old days by David_Hart · · Score: 2

      It's still pretty much common sense. You want a fast CPU, so not the top of the line $1000 chip, take a step back or two and go for the one selling in the $300-$500 range. Motherboard for that chip from someone you trust - ASUS, Gigabyte, etc. Again never the $500 "gamer" board, take a step back, there are some really nice ones for $200 or so. Latest generation graphics card, or top end from last generation (assuming the prices have come down), plenty of memory on the card. Power supply that can feed the card what it needs and then some. Plenty of system RAM. SSD hard drive. Water/Air cooling system for your CPU type. And you're set! Shouldn't take a whole "day" to check those out. An hour or two would suffice.

      I would agree with you except for one thing, new technology. It can take a few days to get up to speed on the newest technology. I built a new system this past winter and it had been three years since I built my old one. It took time to research SSDs (brands, price, reliability, best practice, etc) as it was fairly new tech at the time, CPU and socket types, triple-channel memory, Video cards, etc. On top of that, anyone concerned about best bang for their buck will shop around a bit and look for deals.

      If you build systems for a living or stay on top of the latest tech, you can do it in a few hours. However, if you've been sitting on the sidelines for a while, it may take some time to shake off the rust and get up to speed on what's out there.

    61. Re:The old days by 0123456 · · Score: 1

      That sounds like a much more reasonable price. Where'd I go wrong? GP suggested taking just a step or two back from the top-of-the-line, so instead of picking a $1000 CPU, I picked a $570 one.

      I think my i7 was around $280, so that's $300 saved already. Looks like I was wrong aboui it being the second fastest, it's the second-fastest quad-core i7, I hadn't realised Intel had six-core i7s at that time, thought it was only Xeons.

    62. Re:The old days by Anonymous Coward · · Score: 0

      Unless you're refurbing them and donating them that's a huge waste. How many tasks do you really have that need a separate computer in separate locations? Do you know how much it's costing you in electricity to keep a half dozen or more obsolete PCs running all the time? What can they possibly be doing that can't be rolled into one or two relatively modern low cost boxes? Sure, keep the systems with some particularly unique aspect or architecture that may make for fun projects, but you really don't need a collection of bog standard P4s.

    63. Re:The old days by Burning1 · · Score: 1

      A courteous, polite response on Slashdot. Was not expecting that. :)

      $800 is definitely doable. With components, there's often a sweat spot on the price/performance curve where you can build a powerful system without spending more than $1200 or so. A high-end gaming computer tends to be expensive. With that said, the main benefit of building a machine from components is that some of those components will be re-usable down the road.

    64. Re:The old days by armanox · · Score: 1

      Are you actually using VT-d though? Nothing in the Core 2 line supported VT-d. Don't confuse VT-x with VT-d.

      --
      I'm starting to think GNU is the problem with "GNU/Linux" these days.
    65. Re:The old days by Bengie · · Score: 1

      Can't let the magic smoke out

      When your cell-phone has more memory and CPU power, then it's time to upgrade. The best reason to upgrade is to have a few newer computer that visualize several older computers. Old computers tend to use lots of power.

    66. Re:The old days by armanox · · Score: 2

      I've got 2 SGI Octanes and an O2 sitting in my room now. Without an actual purpose. I might be restoring one of the Octane boxes just to annoy a friend of mine on support (hey, you guys don't have a build for IRIX!).

      --
      I'm starting to think GNU is the problem with "GNU/Linux" these days.
    67. Re:The old days by vux984 · · Score: 1

      Are you actually using VT-d though?

      No. But I'd like to play with it, and I can't. And when I move my current i7 3770K from my main pc into I still won't be able to.

    68. Re:The old days by flimflammer · · Score: 2

      Why exactly would you exclude ever using any older hardware in a new computer...?

    69. Re:The old days by Dunbal · · Score: 1

      I'll just go with a 27" iMac for $1999

      As a gaming rig? Sure. You'll be one of those guys in the forum screaming about when the patch to fix the problems for Mac is coming out. Or why won't "x" work my Mac? I prefer to be absolutely compatible.

      --
      Seven puppies were harmed during the making of this post.
    70. Re:The old days by Anonymous Coward · · Score: 0

      The good old days was the Commodore 64 era, when all you needed was a WICO Command Control (or the Boss) joystick, and a 1541. Add a 1581 (or two) and a modem for BBS goodness.

      Ahh, back when the word "elite" still had meaning, and wasn't spelled with numbers.

    71. Re:The old days by elistan · · Score: 1

      Yeah, sorry for the iMac comment. When I totaled up the component costs and hit $2000 I was taken aback enough to neglect to recall that the comment of yours I was responding to was in turn a response to a comment specifically about building a gaming rig.

    72. Re:The old days by elistan · · Score: 1

      A courteous, polite response on Slashdot. Was not expecting that. :)

      Haha, thanks. I find that courteous, polite responses are good at getting more courteous, polite responses - which I often find much more informative and insightful than a harsh, condescending or sarcastic response. I read and comment on Slashdot to learn, is all. Unfortunately, there are often initially some harsh, condescending or sarcastic responses that one has to deal with...

      I guess the take-away from all this is that now, just like in "the old days" Thanshin talks about, an experienced builder can spec a reasonable gaming machine in a couple of hours - and that now, also just like in the old days, somebody new to DIY would still have to take a few days to learn what's what.

    73. Re:The old days by epine · · Score: 1

      their methodology results in a pretty pricey setup

      You're being pedantic in a way that's annoying and counterproductive, unless you're the kind of person who wallet thickens on discovering the following tidbit:

      Audioquest Everest are the most expensive speaker cables in the world at over $21000 for 3m.

      Oh come on. This does not properly belong in the category of "speaker cables". If it did, "taking a step back" would land you this:

      Pear Cable Corporation's ANJOU Speaker Cable, a 12 foot length of which retails for $7,250

      These aren't products, they are honeypots of rarefied nonsense. Navy Investigating Bills For $660 Ashtrays, $400 Wrenches:

      The Navy is investigating bills from Grumman Aerospace Corp. to determine why it was charged $660 each for two aircraft ashtrays and $400 each for two socket wrenches.

      The costs of the parts, manufactured by Grumman, were revealed during an inspection this month, said Cmdr. Tom Jurkowsky, spokesman for the Pacific Fleet Naval Air Force.

      Burch said the Navy officers who authorized the purchases would be disciplined and possibly dismissed.

      First cross off anything that will get you disciplined or possibly dismissed, then take a step back toward the sane center. Is that so hard?

      I don't build as many boxes as I once did, but it only takes a couple of hours a year to stay current if you know how to parse the tea leaves, and you pick a up quick booster shot from Ars Technica System Guide: July 2013.

      When I've purchased bundled systems, I usually discover surprising limitations and short-cuts down the road unless one pays the insurance premium of buying twice the box you really needed.

      Apple has a tiny product line, so at least with Apple any warts are soon well known, until the day comes that they change the software underneath you and deprive you of features you had come to depend upon, with little warning and no public rationale. Welcome to an ecosystem with benefits you can't refuse.

    74. Re:The old days by Molochi · · Score: 1

      Yes I takes more than a day. It always has. Unless you use one of the build a PC guides that, these days, only get updated about every six months. They pretty much tell you what to buy for the money but go obsolete pretty quick. Even so, guides usually have a comments section that often list better builds for the money or cheaper builds with better performance.

      Today there's a model number for each pricepoint from 100 to 1000 bucks for CPUs and Video cards. Manufacturers use the numbers to obscure the value of the component so a 760 might mean faster slower or the same as 660 of the previous generation. You have to research it.

      It's certainly more complicated than it was 13 years ago when Athlons and P3 were sold by their frequency and were almost identical in performance. Though even then it was easy to buy the wrong component.

      Forums are still probably the best place to quickly catch up on the art of building a capable sub$1000 box.

      --
      "The Adobe Updater must update itself before it can check for updates. Would you like to update the Adobe Updater now?"
    75. Re:The old days by Molochi · · Score: 1

      Right and if you upgraded the C2D to a ~3GHz C2Q you'd be right back in the game.

      --
      "The Adobe Updater must update itself before it can check for updates. Would you like to update the Adobe Updater now?"
    76. Re:The old days by laffer1 · · Score: 1

      I can't agree with this last part about sockets. There are two consumer sockets for Intel and AMD. That means you really do have to pay attention to the socket.

      AMD is pushing A-series chips for consumer which one 2 different sockets (depending on age) plus the AM3+ for FX series.

      Intel has socket 2011, 1150, 1155 in the consumer space right now.

    77. Re:The old days by Anonymous Coward · · Score: 0

      This, so much, this.

    78. Re:The old days by kangsterizer · · Score: 1

      that's just not true tho. yes you can throw $$$ and get whatever is called a "gaming pc". it'll work; but you won't know what you bought and/or why.
      you also won't have the best price/performance ratio.

      10, 15 years ago, you could do that easily if you wanted to. the information was there. it's not true anymore.

    79. Re:The old days by Molochi · · Score: 1

      I'm not sure about the weakness of next gen consoles. They may look like a low spec-ed PC but the code will be for the hardware. Remember the original Xbox was also just a very very weak PC. It had a 733MHz CPU and its GPU was in the GF2MX range. I think it even only had 64MB of memory. PC ports of Xbox games required quite a bit more than that.

      Still I'd expect 1080P gaming to be the target of this next generation of consoles and you can build a pretty damn cheap machine that games well at 1080P.

      --
      "The Adobe Updater must update itself before it can check for updates. Would you like to update the Adobe Updater now?"
    80. Re:The old days by Anonymous Coward · · Score: 0

      *whispers* The horror... the horror...

    81. Re:The old days by LordLucless · · Score: 1

      Geek doesn't necessarily mean hoarder. I've got three non-portable machines - a desktop for me and the wife, and one in the closet that runs all the other stuff. If I want to do something, the one in the closet is modern enough that it has the capacity to do pretty much anything I want, for personal/family needs. I'd rather manage one relatively modern machine than seven cranky old ones.

      --
      Just because you're paranoid doesn't mean there isn't an invisible demon about to eat your face
    82. Re:The old days by Anonymous Coward · · Score: 0

      Hipsters don't have old computers. They only have Macs, and nothing old there either because that wouldn't be fashionable.

    83. Re:The old days by Anonymous Coward · · Score: 0

      Your instructions are worth nothing. The CPUs and GPUs are numbered in so weird ways that there is no way of knowing if it has the chipset from couple generations before or current chipset, that's been crippled, unless you study.

    84. Re:The old days by Anonymous Coward · · Score: 0

      Yeah, i was excited about x-64, but never got around to build a machine with one. I bought some P4 machine from my school about 7 years ago. I think it has a geforce 2 MX440. There hasn't been that many interesting games in years. I might buy a new machine when there's a new interstate '76 type of game.

    85. Re:The old days by pecosdave · · Score: 1

      I used to be like that.

      Now I would rather get something new and low power. The money you save by running a Raspberry Pie, an Atom, or some other lower power processor to do an "old system" task in electricity alone vs. an old P4 or Athlon will pay for the hardware in a year or so, Raspberry pie probably less.

      --
      The preceding post was not a Slashvertisement.
    86. Re:The old days by Anonymous Coward · · Score: 0

      It's rediculous just how long a properly built machine will last, even if you are a gamer. I'm still using my 2.4Ghz C2D from 6 years ago and it's only now starting to fall below the minimum requirements for some games.

      What is worse is that beyond power savings, my (almost) five year old Athlon II quad core is pretty much on par for most of the current cpus that are around the same price today. To see any performance improvements I would need to go for a cpu that costs more then what my current cpu/motherboard cost when I purchased those.

      Mind you, this information was current as of a few months ago when I was looking around at upgrading, price drops and newer cpu tech may have outdated what I wrote already...

    87. Re:The old days by serviscope_minor · · Score: 1

      Why exactly would you exclude ever using any older hardware in a new computer...?

      For anything that's high end or obscure and hard to get a new version of.

      --
      SJW n. One who posts facts.
    88. Re:The old days by Anonymous Coward · · Score: 0

      The only thing using old computers accomplishes is pissing away money. They are extremely inefficient and the lost money from electricity could have been used to pay for a new efficient computer with 10x the power.

    89. Re:The old days by Anonymous Coward · · Score: 0

      Or you can realize that we're talking about building a *new* computer, with *new* hardware.

      Which doesn't do him a lick of good if he wants the new computer to run his old $5000 data acquisition hardware that only has XP drivers. Or dozens of other pieces of hardware which may have newer versions supported by newer OS's but the price of replacement is significantly more than the computer.

      People with $5000 data acquisition cards tend to not be interested in overclocking the computers said cards are plugged into, so they just buy a cheaper non-K (locked multiplier) CPU with VT-d enabled.

      That's if they build a new machine at all. The far more common option: said $5K DAQ hardware soldiers on in the ancient XP PC it was originally installed in, even as everything around it is upgraded to Windows 7/8/onward. Everybody ceases to use it as anything other than a front-end for the DAQ card. If the old box breaks, secondhand hardware is bought on eBay to replace it, so as to avoid the pain of trying to make it all work on a modern system and OS.

      This isn't really meant as a defense of the market segmentation games Intel plays. Just pointing out that most discussions of the issue tend to lack perspective on just how many users actually need VT-d and overclocking at the same time, and so forth. It's usually a very small set of people whom Intel leaves out in the cold -- which is precisely why they can get away with it, despite the loud objections from techy forums.

    90. Re:The old days by Anonymous Coward · · Score: 0

      If he's using $5000 data acquisition hardware then he probably doesn't want to add errors by overclocking the CPU.

    91. Re:The old days by Anonymous Coward · · Score: 0

      Your argument is illogical: "Intel cooks some benchmarks" does not imply "Intel CPUs are slow and therefore the worst choice" or even that Intel "thinks the only way to win a CPU benchmark" is to cheat. All CPU companies try to game benchmarks. They're all corporations who want your money, not good guys and bad guys in a morality play. They'll take an advantage wherever they can get one, even if they don't need it.

      Which arguably Intel doesn't, in this case, given that they've been wiping the walls with AMD even when using totally different compilers. For example, consider this pair of SPECint_rate scores run by Sun in 2009 on 8-core Sun Fire Xeon and Opteron systems running at about 2.9 GHz:

      AMD: http://www.spec.org/cpu2006/results/res2009q4/cpu2006-20090928-08758.html
      Intel: http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090413-07027.html

      Note that the AMD score was obtained using the closest thing AMD had to ICC (i.e. an in-house compiler), the PathScale Open64 compiler, and the Intel score used Sun's in-house Solaris compiler suite.

      Also, despite AMD fanboy lore, ICC has not infected every benchmark with evil tendrils of moustache-twirling villainry. There are even fair ICC-compiled benchmarks (i.e. they provably use the same code path on Intel and AMD).

    92. Re:The old days by Anonymous Coward · · Score: 0

      Socket 2011 isn't really "consumer", as in you don't need to even consider it unless you're aiming to build an expensive high end system. There are no low-cost 2011 CPUs, and the motherboards are all quite expensive.

      Socket 1150 and 1155 are true consumer sockets. But the only reason there are two is that Intel moved several voltage regulators into the CPU in their new Haswell chips, which required them to change the socket since power delivery has changed so much. Other than that, not much changed from the old Socket 1155. So if you want to build a Sandy Bridge (i3/i5/i7-2xxx) or Ivy Bridge (-3xxx) system, you pick 1155. If you want to build a Haswell (-4xxx) system you pick 1150.

      Unless you're an overclocking dragracer obsessed with that last 100-200 MHz, there's nothing which Sandy or Ivy does better than Haswell, and plenty of things which Haswell does better than Sandy/Ivy. 1150 is currently the right choice for almost everyone building an Intel PC, and if you need something else you probably already know what and why.

    93. Re:The old days by Man+Eating+Duck · · Score: 1

      I initially wrote this reply to another post, but only noticed that he was trolling right before submitting. I think my experiences through the years could be useful for a budget-conscious gamer building his own rig, so instead of deleting it I leave it here. Here are a few points to consider:

      • * Buy quality wear parts, mid-range performance parts, upgrade only what you need for a new iteration. I have done this for 15 years, and have always been able to play the latest games at decent settings. My upgrade costs have been pretty stable at around $800 on average every two to three years (running on three years with my current rig, upgrading these days), giving me solid performance for less than $300/year. The quality brand 600W PSU I got five years ago, for instance, will still be present in my new rig, and maybe even the next one after that as long as it's cable compatible with new components.
      • * Don't pay the huge premium on the latest GPU/CPU, you can get mid-range alternatives with 80% of the performance at 1/2 the price. Check out CPU and GPU charts (sort by rank), or actual game performance reviews to find good alternatives. No current game *needs* an i7 or a GTX 7XX video card to get great fps at high settings on a single 1920x1200 monitor. The ROI drops off in a ridiculous rate if you go for the newest hardware. It won't even pay off significantly in performance, see the next point.
      • * Realise that all games are made to run very well on computers far below high-end. At the end of my 2.5-ish year upgrade cycle I might lose out on some minor eye-candy in the latest games. I also get to laugh at the tools who post in forums with comparison screenshots discussing whether the shiny they were able to turn on and run at 90 FPS with their two new $1019 video cards is even *visible in comparison screenshots*. Seriously.
      • * Don't be stupid about what you really need. A 512GB SSD, for instance, is completely ridiculous in a budget gaming rig. A 120GB or even a 90GB one (as I have in my current rig and will keep for the next iteration) will hold a couple of OS'es and the 3-4 games you currently play. Just use mklink (or Steam Mover, which automates this for any game or application, not just Steam ones) to swap them in from your humongous bulk storage spinning drive. It will not degrade your gaming experience at all. If you pay a huge premium to avoid a three-minute coffee break every two weeks to shift in a new game that's your choice, but don't complain about it :)
      • * Spend a few dollars more to get a mainboard with an automatic overclocking analyzer along with a a $30 aftermarket silent CPU cooler. I've never bothered to overclock manually, but if the computer does the work for me, I'll take the 10-30% stable CPU performance increase you can get "for free". A $40 premium for this feature and a cooler is *a lot* cheaper than actually buying a chip specced from the factory for those increased speeds. Even my current el-cheapo ASRock board did this three years ago.

      • * Open up your case and clean your computer thoroughly once in a while. Remove all dust from air channels in cooling ribs and the video card. This one might be obvious, but very few people actually do it. The computer will run cooler, be more quiet and last longer.

      I completely agree with you about brand name components. After two bad experiences with a relatively cheap video card and a mainboard from unknown manufacturers I only buy brands myself (including EVGA for video cards and ASRock for mainboards).

      Yes, it does take a bit of research. No, you don't really have to spend more than a couple of hours on it if you don't want to.

      --
      Are you a grammar Nazi? I'm trying to improve my English; please correct my errors! :)
    94. Re:The old days by Anonymous Coward · · Score: 0

      In order to upgrade I really need new RAM(DDR3), which means new mobo, which means new processor...so I might as well buy a new computer outright.

      you could just get a new gpu; optimizations be damned

    95. Re:The old days by Anonymous Coward · · Score: 0

      GF3

    96. Re:The old days by Anonymous Coward · · Score: 0

      What is worse is that beyond power savings, my (almost) five year old Athlon II quad core is pretty much on par for most of the current cpus that are around the same price today. To see any performance improvements I would need to go for a cpu that costs more then what my current cpu/motherboard cost when I purchased those.

      This isn't even remotely true. From what I can tell a high-end AII x4 would have been around $120 when new, and you can easily find $120 CPUs from both AMD and Intel which beat it handily. If you go with AMD, you basically get the same capabilities with a significantly higher clock (and maybe an integrated GPU if you choose an APU). Yeah, it's a Bulldozer-derived core now, so some of the clock scaling is eaten by the lesser perf/clk ratio, but there has been enough clock scaling to more than make up for it.

      If you go with the Intel option, you get an i3, which is only dual-core. But it's so much faster per thread that it is still competitive with an old x4 in heavily threaded benchmarks, which is generally the way to go if you can get it. (Look up Amdahl's Law. Its implication is that if you're given a choice between two machines with roughly equivalent 100% parallel throughput, but one uses fewer CPU cores to get there, you should pick that one because it will perform better running most real-world software.)

      Oh, and the i3 uses about half the power too.

  2. Before AMD committed suicide by Anonymous Coward · · Score: 1

    They swooped in when Intel was being stupid, made the best chips in the world... then committed suicide and haven't built a competitive chip in 3 years. Sad times...

    1. Re:Before AMD committed suicide by bill_mcgonigle · · Score: 4, Insightful

      AMD is very competitive for many-cores workloads. To get an equivalent core count on Intel can be as much as a second AMD system. AMD has gone more wide, Intel has gone more deep. Both have their applications.

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
    2. Re:Before AMD committed suicide by Nadaka · · Score: 2

      They are still competitive on the performance per $ scale, and provide cpu's adequate for almost all standard needs.

    3. Re:Before AMD committed suicide by Billly+Gates · · Score: 3, Insightful

      Tell that to Tomshardware and others who use x87 benchmarks and games like skyrim showing an AMD 8 core being handed a smackdown by an i3?

      No one believes in AMD anymore

    4. Re:Before AMD committed suicide by h4rr4r · · Score: 4, Insightful

      For games sure, but there are lots of workloads that are not games.

    5. Re:Before AMD committed suicide by h4rr4r · · Score: 2

      and many games are not made by Bugthesda.

    6. Re:Before AMD committed suicide by 0123456 · · Score: 2

      Tell that to Tomshardware and others who use x87 benchmarks and games like skyrim showing an AMD 8 core being handed a smackdown by an i3?

      True, it's horrible that review sites benchmark CPUs using the kind of programs people actually run on them.

      I remember back when I bought my P4, the only thing a similarly priced Athlon XP really beat it on were x87-intensive games. Professional 3D apps using SSE were significantly faster on the P4, which is why I ended up buying it instead.

    7. Re:Before AMD committed suicide by Anaerin · · Score: 1

      And for those computations, at the desktop level, 1 Intel core is approximately as fast as 2 AMD clocks. Intel has MUCH better IPC (Instructions Per Clock) and better re-ordering and lookahead than AMD, and have since the Intel introduced their "Core" infrastructure. This is why a mid-range Intel part (Say, an Intel Core i5-4670K) can handily (and significantly) beat AMD's top-of-the-line desktop CPU (An FX-8350)

    8. Re:Before AMD committed suicide by h4rr4r · · Score: 1

      My desktop is used to run KVM, please tell me all about how a 1 core intel would be good enough. I shall wait.

    9. Re:Before AMD committed suicide by bill_mcgonigle · · Score: 2

      Tell that to Tomshardware and others who use x87 benchmarks and games like skyrim showing an AMD 8 core being handed a smackdown by an i3?

      Awesome, you found a workload where deeper is better. Now go try costing out a cluster with hardware virtualization and ECC RAM to support several thousand SMP virtual machines and see what you come up with.

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
    10. Re:Before AMD committed suicide by Anaerin · · Score: 1, Insightful

      It probably wouldn't. But a dual-core Intel processor would be as good as (or better than) a quad-core AMD. And a quad-core Intel would be as good as, or better than, an 8-core AMD. Especially with Intel's Hyperthreading enabling 2 cores-worth of processing to be handled on a single core.

    11. Re:Before AMD committed suicide by Anonymous Coward · · Score: 0

      I have a dual core pentium that regularly handles 8 windows xp virtual machines at once doing vairous tasks mainly used for testing browsers it handles them pretty good and thats 4 per core I dont have a newer amd but I had an old fx quad core that struggled with 2 virtual machines so in my book the battle for cores has become irrelevant a while ago oh and mhz to because the intel is 2.4ghz and the amd was 2.8ghz.

    12. Re:Before AMD committed suicide by FreonTrip · · Score: 1

      It depends an awful lot on the workload, though. For gaming it's one-sided in Intel's favor to the tune of around 2/3 more work done per clock on average (sometimes more), but for video encoding with x264 the sheer core count makes it better than competitive with Intel unless you're willing to pay noticeably more. It's a behemoth for virtual machines, it plays video games well enough, and for scientific computation I really haven't found myself wanting. Granted, I'm an edge case...

    13. Re:Before AMD committed suicide by bored · · Score: 5, Insightful

      The spec benchmarks tell a different story, and tend to be more representative because each vendor does their best rather than intel/nvidia providing "free" performance enhancement advice for game companies.

      So, from my own experience the Amd/Intel story is a lot closer than some of these benchmarks might lead you to believe. Especially for server applications.

      Its pretty easy with modern CPU's to make fairly small tweaks that give advantage to one CPU or another. We have a bunch of microbenchmarks for our application, and things like memcpy performance can be swung 2x-4x. Or even the depth of loop unrolling for some things. In one loop the intel it may like 2x and the AMD like 4x unroll. With each one tuned to run best on the platform the bottom line performance is often quite similar, but run the AMD optimized one on the intel, or the reverse and suddenly one or the other CPU appears to be trouncing the other.

    14. Re:Before AMD committed suicide by Anonymous Coward · · Score: 1

      This is why a mid-range Intel part (Say, an Intel Core i5-4670K) can handily (and significantly) beat AMD's top-of-the-line desktop CPU (An FX-8350)

      Really? CPU Benchmarks says

        i5-4670K - 7531
        AMD FX-8350 - 9091

      A comparable Intel chip would have to be closer to i7-3820, not your i5. Perhaps your benchmarks are a little crappy?

      http://www.cpubenchmark.net/cpu.php?cpu=AMD+FX-8350+Eight-Core&id=1780

      http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i5-4670K+%40+3.40GHz&id=1921

      Anyway, AMD is far more $$$ efficient for typical desktop. Yes, including any thermal envelope differences.

    15. Re:Before AMD committed suicide by Anaerin · · Score: 2

      Perhaps you should read the second chart here. That's testing encoding with Handbrake (Which is essentially a graphical frontend to x264). In that particular test, the i5-4670K wipes the floor with the comparably priced FX-8350, even without the former's huge overclocking potential.

    16. Re:Before AMD committed suicide by bored · · Score: 2

      True, it's horrible that review sites benchmark CPUs using the kind of programs people actually run on them.

      Yes, and no. If your a gamer, obviously having a CPU that the games are optimized for is a big win. But don't extrapolate general performance from a single benchmark. Especially when one of the CPU vendors is providing "free" performance help for the game/application.

      At this point, its pretty clear that choosing the Intel is the correct choice for big name games.

      We will see if this changes over the next few years with the consoles being AMD based. The game companies are going to optimizing for those platforms, we will see if any of that carries over into game benchmarks for desktop machines. In many cases its possible to get a 2x delta by optimizing for a particular CPU/platform at the expense of the general case.

      As for SSE, AMD has always been a little behind with SSE (well duh, they follow whatever intel is doing and it takes them a while to catchup), so if your application is built for the latest version of SSE, and its gaining something from it, then running a similar version without hurts. Recently seen with SSE 4.1/4a where there were a couple useful instructions for some code paths that didn't exist on the AMD and hurt it on benchmarks using SSE4.

    17. Re:Before AMD committed suicide by Anonymous Coward · · Score: 0

      Your mom is wide *and* deep.

    18. Re:Before AMD committed suicide by Anonymous Coward · · Score: 0

      Tomshardware and others are intel shills now.

    19. Re:Before AMD committed suicide by serviscope_minor · · Score: 4, Informative

      They swooped in when Intel was being stupid, made the best chips in the world... then committed suicide and

      If, by committed suicide, you mean that suffered when intel bribed people like Dell not to use the clearly superior products and so lots out on many billions of sales and hence the crucial R&D advantage, then yeah sure suicide.

      Assisted suicide.

      Like assisted like throwing a healthy happy person off a clifff.

      haven't built a competitive chip in 3 years. Sad times...

      Depends what for. For games, intel seem to be better IF you're prepared to buy a separate GPU. If you look on the Linux, not Window centric benchmarks, the top end AMD ones often lie somewhere between the top end i5 and the top end i7.

      Sometimes they lose out sometimes thy beat the i7s.

      For the kind of stuff I do, they're very competitive.

      But if you play skyrim, then no. But an i5 and an external graphics card.

      --
      SJW n. One who posts facts.
    20. Re:Before AMD committed suicide by Burning1 · · Score: 1

      Games typically don't benefit from horizontal scaling. Although many games have gone multi-threaded, there are only so many tasks you can hand off to additional cores.

    21. Re:Before AMD committed suicide by Anonymous Coward · · Score: 0

      games like skyrim showing an AMD 8 core being handed a smackdown by an i3?

      Does skyrim use the microsoft or the intel compiler? The intel compiler has been for a long time the best compiler and it produces binaries that deoptimize* on non intel CPUs (confirmed in curt after intel claimed that its competitors just produced inferior CPUs). If it uses the intel compiler then the benchmark is worthless for the general case.

      * as in x86 CPUs have optional features and program binaries can support multiple combinations of these features, binaries produced by the intel compiler will always disregard available features if the CPU vendor is not Intel. This guarantess that any benchmark using the intel compiler is biased towards Intel CPUs.

    22. Re:Before AMD committed suicide by FreonTrip · · Score: 1

      Ergh. Looks like Intel's managed to close the gap with Bulldozer in that stronghold through sheer clock ramping. You got me - I hadn't looked at Haswell benchmarks much.

    23. Re:Before AMD committed suicide by Anonymous Coward · · Score: 1

      It probably wouldn't. But a dual-core Intel processor would be as good as (or better than) a quad-core AMD. And a quad-core Intel would be as good as, or better than, an 8-core AMD. Especially with Intel's Hyperthreading enabling 2 cores-worth of processing to be handled on a single core.

      Hyperthreading does not enable 2 cores worth. A properly optimized OS will get 1.2 cores, MAYBE. An old decreped OS like RHEL 6 will get 1.8-ish, but you're losing CPU power by a poorly handled IO scheduler.

    24. Re:Before AMD committed suicide by Anonymous Coward · · Score: 0

      If, by committed suicide, you mean that suffered when intel bribed people like Dell not to use the clearly superior products and so lots out on many billions of sales and hence the crucial R&D advantage, then yeah sure suicide.

      Assisted suicide.

      Like assisted like throwing a healthy happy person off a clifff.

      I suspect the parent author was referring to AMD laying off a substantial portion of their processor development team. Either the CEO thought they could hire them back when times got better (alas, when you lay off engineers, the times will never get better) or that it was somehow possible to do the same quality job with fewer people. That portion is corporate suicide. I just hope ARM manages to move into desktops faster than Intel pushes into cellphones/embedded.

    25. Re:Before AMD committed suicide by electrosoccertux · · Score: 2

      I go by Cinebench myself. It seems completely neutral.

      http://www.bit-tech.net/hardware/2013/06/12/intel-core-i5-4670k-haswell-cpu-review/6

    26. Re:Before AMD committed suicide by Anonymous Coward · · Score: 0

      A few years back my cousin was telling me how his $60k Intel server with less than half of the cores was stomping on his $60k AMD server in nearly every workload from DB to Video-compressing, then the power usage was an issue. They can't even power up all of their servers. They had to jimmy-rig 250KW of power into the datacenter, which was all used within a week of powering up extra racks. He says his electrical engineer is freaking smart.

    27. Re:Before AMD committed suicide by Anonymous Coward · · Score: 0

      You've obviously never actually researched Hyperthreading scaling.

      From Intel: "Two logical processors do not provide the same level of performance as a dual processor-based system.", "Hyper-Threading Technology does not deliver multiprocessor scaling."

      HyperThreading allows one thread to 'spin'(enter a wait/idle state, waiting for memory, for instance) while another thread, memory 'in hand'(or cache, rather) executes, and they switch back and forth in this manner in an ideal situation(a typical core spends most of its life idling, followed by waiting, and finally executing). In less than ideal situations they both spin or both want to execute simultaneously, and the results aren't good in these cases as in the first case nothing gets done and in the second case contention arises, which can exacerbate each other into cycles of mutual deadlock and mutual contention thereby rendering HyperThreading nothing more than wattage wasting overhead.

      The great Hyperthreading lie is that the technology "enables 2 cores-worth of processing to be handled on a single core". Of course, if you don't understand what a serial processor is then you, well, repeat the lie on /. and somehow get modded for being insightful when really you're just lost somewhere between delusional and ignorant.

      Sincerely,
      Anonymous Coward

    28. Re:Before AMD committed suicide by Anonymous Coward · · Score: 0

      Nothing says awesome virtualization like consuming much more power for the same workload.

    29. Re:Before AMD committed suicide by Anonymous Coward · · Score: 0

      The great Hyperthreading lie is that the technology "enables 2 cores-worth of processing to be handled on a single core". Of course, if you don't understand what a serial processor is then you, well, repeat the lie on /. and somehow get modded for being insightful when really you're just lost somewhere between delusional and ignorant.

      The problem with this screed of yours is that you are the one who is delusional and/or ignorant. Intel x86 CPUs with Hyperthreading do in fact work on instructions from both threads at the same time, not just one thread at a time. What you've described is known in the field as "switch on event multithreading", or SoEMT, and is indeed distinct from "simultaneous multithreading" (or SMT). Intel has used the marketing name "Hyperthreading" to refer to both SMT and SoEMT, but the form found in their high performance x86 chips has always been SMT -- they've mostly used SoEMT in Itanium.

      Educate yourself with words written by an actual expert:

      http://semipublic.comp-arch.net/wiki/Hyperthreading

      (Please note: comp.arch is an old Usenet newsgroup where lots of people knowledgeable about CPU design have been known to post. Andy Glew, the author of this wiki, is one of them. As in, this is a gentleman who has held high level CPU architect positions at both Intel and AMD.)

    30. Re:Before AMD committed suicide by zixxt · · Score: 1

      This is why a mid-range Intel part (Say, an Intel Core i5-4670K) can handily (and significantly) beat AMD's top-of-the-line desktop CPU (An FX-8350)

      Really? CPU Benchmarks says

        i5-4670K - 7531

        AMD FX-8350 - 9091

      A comparable Intel chip would have to be closer to i7-3820, not your i5. Perhaps your benchmarks are a little crappy?

      http://www.cpubenchmark.net/cpu.php?cpu=AMD+FX-8350+Eight-Core&id=1780

      http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i5-4670K+%40+3.40GHz&id=1921

      Anyway, AMD is far more $$$ efficient for typical desktop. Yes, including any thermal envelope differences.

      Finally some facts.

      --
      ---- GENERATION 26: The first time you see this, copy it into your sig on any forum and add 1 to the generation.
  3. Made me miss the old Slashdot by ElementOfDestruction · · Score: 4, Insightful

    It seems we've lost a lot of quality in the comment fields in the last 10 years. Lots of expertise modded up carefully; now we seemingly have opinion-pieces moderated up by whichever group happens to be awake at the time, and the real expertise is hidden in the +2 or below.

    1. Re:Made me miss the old Slashdot by Billly+Gates · · Score: 5, Insightful

      Yeah like the hot grits down your pants, Natalie Portman naked and petrified, gay niggers association, penis bird registrations in ASCI, and of course who could have forgotten Goatse I mean LITERALLY forget! Ahoot one goatse troll had a +3 and got +90 responses with MY EYES?!! By a moderator trying to be funny.

      No I dont miss those days as we tend to remember only the good ones

    2. Re:Made me miss the old Slashdot by Anonymous Coward · · Score: 0

      It's Bush's fault.

    3. Re:Made me miss the old Slashdot by Beardo+the+Bearded · · Score: 1

      CHOCOLATE CHIP COOKIE RECIPE

      It was a nice break from the usual trolling. The recipe was legit, too.

      --

      ---
      ECHELON is a government program to find words like bomb, jihad, plutonium, assassinate, and anarchy.
    4. Re:Made me miss the old Slashdot by iroll · · Score: 3, Interesting

      You must have read different articles than I did, because 10 years ago it was "Micro$oft $hills," "Apple Fanboys," etc. You do know that this was the origin of "No wireless. Less space than a Nomad. Lame," right? And that was 2001.

      --
      Repetition does not transform a lie into the truth. - FDR
    5. Re:Made me miss the old Slashdot by Anonymous Coward · · Score: 0

      That is not exclusive.

      There was plenty of -justified- microsoft bashing. But the content was not the moronic drivel we get nowadays. Microsoft fanbois and appealing to the masses brought mediocrity as usual.

    6. Re:Made me miss the old Slashdot by Mr.+Flibble · · Score: 1

      I like how this is modded offtopic - as if the majority of users on now don't know or remember these things.

      --
      Try to hack my 31337 firewall!
    7. Re:Made me miss the old Slashdot by DG · · Score: 1

      Tell me about it...

      DG

      --
      Want to learn about race cars? Read my Book
    8. Re:Made me miss the old Slashdot by ElementOfDestruction · · Score: 1

      Oh couldn't you tell? I was trolling for Karma. Cuz that's what people do in New Slashdot.

  4. Just Replaced by MightyYar · · Score: 2

    I only just replaced my Athlon 64 motherboard and processor this spring. It was a good product, but not quite up to running Windows 8 IMHO.

    --
    W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
    1. Re:Just Replaced by Anonymous Coward · · Score: 1

      Your Athlon might have been up to running Windows 8 if it had been able to get over the acute nausea that resulted.

    2. Re:Just Replaced by toddestan · · Score: 1

      Actually, the first Athlon 64's would be the oldest chips that could run Windows 8, due to them being the first to implement the NX bit which is a requirement. On the Intel side, you need at least an LGA775 system, and even with those you have to be careful because not all the LGA775 chips implement it.

    3. Re:Just Replaced by MightyYar · · Score: 1

      The chip ran it (it booted s-l-o-w-l-y), but I needed more RAM and the video card was, um, ancient. I couldn't justify RAM and a video card when a whole new motherboard with integrated video, processor, and 8GB of RAM was under $200. I wouldn't have upgraded it at all, except that I really needed to reinstall Windows, and I couldn't justify the time spent setting up XP again. I would have used Windows 7, but I was ignorant and stupid and fell for the $40 price of Windows 8.

      --
      W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
  5. 10 years later and applications are still 32bit. by Anonymous Coward · · Score: 4, Insightful

    10 years later and we're still running games and applications that are 32bit that only use a single core.

  6. Great processor by the_humeister · · Score: 1

    Too bad AMD was just sitting on their laurels after that. Incidentally, in 2 more years, you can start making your own Pentium Pro compatible processor without violating any patents (assuming you're using the same patents that went into the Pentium Pro).

  7. Error in 32/64 bit libraries. Please reinstall by ObsessiveMathsFreak · · Score: 3, Interesting

    Wow. Ten years. And here I am still dealing with 64 bit incompatability issues every six months or so.

    Out of curiosity, how long did 16bit library problems linger after the 32 bit move?

    --
    May the Maths Be with you!
    1. Re:Error in 32/64 bit libraries. Please reinstall by osu-neko · · Score: 1

      Library problems? Were there ever 16-bit programs that were not statically linked to their libraries?

      --
      "Convictions are more dangerous enemies of truth than lies."
    2. Re:Error in 32/64 bit libraries. Please reinstall by Anonymous Coward · · Score: 0

      10 years and I yesterday discovered that Photoshop Elements 11/12 is 32-bit and limited to 3.2GB of RAM. Adobe's stirling work continues.

    3. Re:Error in 32/64 bit libraries. Please reinstall by Anonymous Coward · · Score: 0

      They barely existed.
      Usually most applications were largely self-contained, so there were never any linking issues.
      Writing shared libraries on 16 bit was difficult anyway and thus it was rarely done.

    4. Re:Error in 32/64 bit libraries. Please reinstall by Anonymous Coward · · Score: 0

      Until about 2 weeks ago. They seemed to clear up quite a bit then.

    5. Re:Error in 32/64 bit libraries. Please reinstall by sl4shd0rk · · Score: 1

      Wow. Ten years. And here I am still dealing with 64 bit incompatability issues

      10 years? Some of us are still waiting to reap the benefits of MMX extensions. Ha..

      --
      Join the Slashcott! Feb 10 thru Feb 17!
    6. Re:Error in 32/64 bit libraries. Please reinstall by 0123456 · · Score: 1

      You don't really think that all those 16-bit Windows apps statically linked in every Windows library, do you?

    7. Re:Error in 32/64 bit libraries. Please reinstall by Waffle+Iron · · Score: 1

      Library problems? Were there ever 16-bit programs that were not statically linked to their libraries?

      Yes, and it was a largely manual process. I suggest you find an old 16-bit Windows programming textbook and learn about the hoops people used to have to jump through to implement dynamic linking. When software versions changed, then as now, what could possibly go wrong?

    8. Re:Error in 32/64 bit libraries. Please reinstall by KliX · · Score: 1

      They're still here.

    9. Re:Error in 32/64 bit libraries. Please reinstall by FreonTrip · · Score: 1

      And in AMD64 MMX (along with 3DNow!) is officially deprecated in favor of the SSE instructions. Like a twisting of the knife.

    10. Re:Error in 32/64 bit libraries. Please reinstall by Anonymous Coward · · Score: 0

      There is zero truth in your statement.
      I just don't know whether or not it's intentional.

    11. Re:Error in 32/64 bit libraries. Please reinstall by jbeaupre · · Score: 1

      I've got rule of thumb on how long it will take to move completely to 64 bit. Basically every time we double the number of bits, the time to convert takes double the time. I'm sure someone could refine that but it makes a tiny bit of sense.

      Rectally extracted numbers:
      4 to 8 : 2 years
      8 to 16 : 5 years
      16 to 32 : 10 years
      32 to 64 : 20 years?

      --
      The world is made by those who show up for the job.
    12. Re:Error in 32/64 bit libraries. Please reinstall by Espectr0 · · Score: 1

      The "problem" is that 32 bit is still enough for mostly everyone, while 32 bit gained quick popularity after windows 95 (18 years old)

      Also, there is less incentive to upgrade your current working machine today than before.

    13. Re:Error in 32/64 bit libraries. Please reinstall by armanox · · Score: 1

      Except they don't. i386 was dropped a while ago. Red Hat/Fedora builds for i586, Slackware for i486, no idea for Ubuntu. But the kernel itself no longer supports i386.

      --
      I'm starting to think GNU is the problem with "GNU/Linux" these days.
    14. Re:Error in 32/64 bit libraries. Please reinstall by Bert64 · · Score: 1

      If you're running a 32bit distro then some do, most compile for something a bit newer (eg 686) and the kernel doesnt even support i386 anymore...
      If you run a 64bit distro (and what possible reason would you have for not doing so on compatible hardware) then it will usually be compiled for a baseline amd64 cpu, so sse2 etc...

      Ofcourse you could always run gentoo, then you can have the entire system compiled specifically for your cpu and using the latest compiler too.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    15. Re:Error in 32/64 bit libraries. Please reinstall by Anonymous Coward · · Score: 0

      Ubuntu 686 (after Debian).

      But more importantly, GCC (like most modern compilers) has the ability to generate separate code paths for different architectures, so even a distro compiled to be compatible with i486 will reap the benefits of MMX, SSE, SSE2,... So GP is completely wrong and deserved to be modded "flamebait".

  8. Bludgeoned? by Anonymous Coward · · Score: 0

    Someone is smoking the crack pipe again. That was the golden age of AMD. Intel had slower chips running hotter including the P4... Intel ate AMD dust until they did a complete redesign.

  9. Re:10 years later and applications are still 32bit by alen · · Score: 2

    for gaming, the GPU took over most of the work which is the way it should have happened

    for applications, most don't really need 2 cores. even running multiple apps at the same time you don't really need 2 cores. i was playing MP3's on a computer in the 1990's with minimal CPU usage. there is no way you need to dedicate a whole core to music while surfing the internet. or some of the other idiotic use cases people make up

  10. Re:10 years later and applications are still 32bit by ciderbrew · · Score: 1

    Fair point and the follow up is WHY?

  11. Still better IMHO by s.petry · · Score: 2, Insightful

    AMD still makes a better chip for many FP intensive applications, and the price is still superior to Intel to boot. Intel always made a big deal about clock speed, while AMD worked on actual performance. It is really a shame that people pay more attention to marketing than real performance.

    --

    -The wise argue that there are few absolutes, the fool argues that there are no probabilities.

    1. Re:Still better IMHO by Anaerin · · Score: 2

      Maybe you should look at the actual performance numbers. Intel is performing better than AMD, and at a cheaper price point. And unfortunately, I'm an AMD fan, running a Hexcore Bulldozer here.

    2. Re:Still better IMHO by bemymonkey · · Score: 1

      Aren't Intel's chips faster clock for clock right now? Not to mention much more efficient?

    3. Re:Still better IMHO by s.petry · · Score: 1

      Instead of relying on a web site which does testing based on their funding, perform actual bench marks yourself. I do, and see mixed results today. 5 years ago for math intensive apps, AMD was a hands down winner. Apps like Muses, Nastran, Abaqus, Dyna, etc...

      --

      -The wise argue that there are few absolutes, the fool argues that there are no probabilities.

    4. Re:Still better IMHO by s.petry · · Score: 0

      If speed was all that mattered in a chip you would have a point. Speed is not all that matters, it never has been and never will be. The memory buss is much better on an AMD chip. FP instruction is faster and better in AMD chips. If it takes .01ns to get an instruction to memory due to buss length and .001ns on an AMD because of shorter buss length, who's chip is better?

      Speed is a misnomer, and go back 10 years and AMD was telling you that the only rating on a chip should not be speed. You probably forgot to read it, or forgot that you did read it.

      It's like having a city with a highway around it, think Houston or DC. You can drive 65mph around the city on the freeway. To you, I'm going slow on a 35mph road. Who makes it from the South side of town to North first? Not you, assuming I know a better route. That analogy is why chip speed arguments are a measure of ignorance.

      --

      -The wise argue that there are few absolutes, the fool argues that there are no probabilities.

    5. Re:Still better IMHO by bemymonkey · · Score: 1

      Hmmm, I was actually using the word "faster" in a "get shit done more quickly" sense - not higher clock speeds. I.e. a 3GHz Haswell i5 being faster than a 3GHz Bulldozer (or whatever the latest generation is called) for a purely CPU-limited single-threaded workload. That's what I'm asking - is this not still the case?

      The fact that I can crank an unlocked i5/i7 to 4.5-5.0 GHz without any issues whatsoever is just icing on the cake.

    6. Re:Still better IMHO by s.petry · · Score: 0

      It honestly depends on the application using the CPU, see my initial post. Math intensive applications tend to perform much better on AMD chips. This is everything from Analysis and Simulation to some types of compression and encryption.

      --

      -The wise argue that there are few absolutes, the fool argues that there are no probabilities.

    7. Re:Still better IMHO by Anaerin · · Score: 4, Insightful

      Unfortunately, not everyone has the ready cash available to buy every CPU, set up a complete system for each, and benchmark them. We plebians have to rely on other people to do that kind of testing for us. And when a great deal of websites, all doing independent benchmarks and reviews, all show AMD getting their collective asses handed to them on a regular basis at the moment, I tend to lend those reviews some weight.

    8. Re:Still better IMHO by bemymonkey · · Score: 3, Insightful

      See, this is why I asked. Looking at benchmark lists (things like Cinebench) would lead me to believe that this is nowhere near the case, with the fastest AMD chip (with a 4.4GHz singlethreaded turbo vs. 3,9GHz on the fastest Intel chip in this benchmark) barely keeping up with good old 1st-gen Core i5/i7 chips.

      http://www.tomshardware.de/charts/cpu-charts-2012/-01-Cinebench-11.5,3142.html

      Multithreaded workloads are a different story, of course, what with AMD having consumer octacores on the market: http://www.tomshardware.de/charts/cpu-charts-2012/-02-Cinebench-11.5,3143.html

    9. Re:Still better IMHO by serviscope_minor · · Score: 0

      Yay duellng with URLs:

      http://www.phoronix.com/scan.php?page=article&item=amd_fx8350_visherabdver2&num=1

      That's versus IVB, not Haswell, but Haswell doesn't seem hugely faster than IVB. In some cases intel win. In some cases AMD beat the i5 handily. In others, AMD beat the i7 handily.

      --
      SJW n. One who posts facts.
    10. Re:Still better IMHO by Anaerin · · Score: 1

      Ooh, Duelling Benchmarks! Can I have a go?

    11. Re:Still better IMHO by s.petry · · Score: 0

      I trust Toms Hardware as far as I can spit. Once a cheat always a cheat.

      --

      -The wise argue that there are few absolutes, the fool argues that there are no probabilities.

    12. Re:Still better IMHO by s.petry · · Score: 1

      Most businesses will allow a budget for testing. You think I purchase systems at home to run enterprise analysis applications? I would have to pay for the applications as well, which are between 20K and 200K depending on options.

      --

      -The wise argue that there are few absolutes, the fool argues that there are no probabilities.

    13. Re:Still better IMHO by Anaerin · · Score: 1

      Well, I was personally talking about desktop parts. I do realise that in server and computing applications, AMD's HyperTransport system gives them massive capability for expanding the number of cores they can put on a CPU (and the number of CPUs in a system), which in some cases can make up the overwhelming IPC deficit they have. I would be waiting for them to produce their next series of CPUs, hoping they can (at least) make up the difference, but it seems that persistent, intermittent hardware failures on my current rig (along with "Incompatible" RAM, of all things) mean that I'm going to have to change CPU and motherboard now (As in, as soon as I can get the cash together). So I'm going to have to go with a Haswell CPU and Z87-based motherboard, as I simply can't justify paying the same amount for poorer performing AM3+ part.

    14. Re:Still better IMHO by s.petry · · Score: 0

      At home, I run an AMD. Much cheaper than Intel and I can honestly state that I have never had problems with AMD chips. Memory, depends on the manufacturer but that's not a CPU or motherboard issue. My motherboard at home cost 90 bucks, the CPU 150. Similar Intel gear would be double that (roughly).

      At home, since most parts of the computer work whether AMD or Intel, it would not be "that" expensive to benchmark Intel vs. AMD unless you planned on the AMD first. The additional 200 bucks for a motherboard and CPU is not "that" much, but more than most want to pay.

      When I look at CPU performance, why would I not test for work over home? A game may stress the CPU and GPU, but it's not what I consider important.

      --

      -The wise argue that there are few absolutes, the fool argues that there are no probabilities.

    15. Re:Still better IMHO by Anonymous Coward · · Score: 0

      Well, I was personally talking about desktop parts. I do realise that in server and computing applications, AMD's HyperTransport system gives them massive capability for expanding the number of cores they can put on a CPU (and the number of CPUs in a system), which in some cases can make up the overwhelming IPC deficit they have.

      HyperTransport doesn't do anything for fitting more cores on one chip -- it's a chip-to-chip interconnect. Also, for big systems Intel's QPI is actually superior to HT, thanks to being something like a 5+ year newer design. AMD used to have an edge when Intel had no answer for HT, but since 2008 the shoe's been on the other foot, and with AMD's struggles they really haven't had the resources to bring a new generation interconnect to market. They've more or less conceded the bigger systems to Intel these days, and are mostly trying to compete in the 1-to-2 socket market.

      There's two basic reasons AMD has more cores per chip right now. One is that ever since AMD's "Bulldozer" architecture, what they're calling one core is no longer an independent core. Instead, it's half of a 2-core "module" which shares many resources between the two cores, notably including (but not limited to) the FPU. When executing integer code, dual-thread performance of one module is about 1.8x single-thread, hence AMD's desire to market one module as two cores.

      Intel has a different approach, hyperthreading, in which one core supports two thread contexts and more resources are shared between the two threads than in an AMD module. This scheme realizes a more modest 2-thread throughput gain, maybe 1.3x to 1.4x on typical integer code. However, Intel's single-thread performance is far better than AMD's, so the reduced gain is more than made up for by a higher baseline.

      The second reason for AMD's lots-of-cores-for-cheap approach is simply that this is the only way they can stay competitive. They're tightening their belts and selling larger chips with more cores (equals higher manufacturing costs) for less money than Intel charges for smaller chips with fewer cores. (And even with this approach, the best they can do is be competitive in highly threaded applications. If you can't load all of the cores in an AMD chip, odds are good you'd see better performance with Intel -- sometimes much better.)

      AMD's bigger Opteron CPUs (12-core and 16-core) consist of two physical chips mounted in one package, interconnected with HyperTransport. That might have been what you were referring to. Intel could do that too, using QPI, but there's little reason to when they're matching and beating AMD's performance with single chip CPUs. HT and QPI are good but not perfect. Those big systems built on HT or QPI tend to need software that is performance tuned for NUMA -- non uniform memory access -- i.e. it interacts with the operating system to try to place data with the CPUs which are actively using it, to reduce the amount of HT or QPI traffic. Introducing NUMA into even a single socket system (as AMD's big Opterons do) is not actually a good thing, but it's what you do when that's the only way to stay in the race.

    16. Re:Still better IMHO by Anonymous Coward · · Score: 0

      It honestly depends on the application using the CPU, see my initial post. Math intensive applications tend to perform much better on AMD chips. This is everything from Analysis and Simulation to some types of compression and encryption.

      This is nonsense. Simply not grounded in reality.

      http://www.spec.org/cpu2006/results/res2013q1/cpu2006-20121204-25276.html
      http://www.spec.org/cpu2006/results/res2013q3/cpu2006-20130909-26269.html

      TLDR: 2-CPU 32-core 140W per chip AMD system loses badly on the FP throughput SPEC test to a 2-CPU 24-core 130W per chip Intel system. I chose Dell for both, and made sure the AMD score used the AMD compiler so nobody could whine about Intel compilers hamstringing AMD.

      The problem with AMD zealots is that you almost universally assume that no new developments have happened since dinosaurs ruled the earth -- i.e. when the competition was between first generation Opteron and Pentium 4. What you say was true back then, but it isn't today. In the high performance world, AMD has been reduced to selling lots of poorly performing cores for very cheap prices. This strategy can even win a few benchmarks here and there, but it isn't a strategy for market growth, just for holding on to what little share they have left.

      Their low end strategy is based on pairing superior GPUs with not-so-great CPUs, and is considerably healthier since at least one half of the chip is notably better than what Intel has to offer.

  12. Bludgeoning, vengeance, kicking, seizing? by Sir+or+Madman · · Score: 1

    Must we be so violent? This is CPU sales not some barbaric conflict.
     

  13. Comparison with current CPUs by IYagami · · Score: 1

    I was hoping to find a current review of the processor against current CPUs....

    However, in AnandTech bench you can compare an AMD Athlon X2 4450e (2.3GHz - 1MB L2) with current CPUs. If you compare this to an Intel Core i7 4770K (3.5GHz - 1MB L2 - 8MB L3, one of the best CPUs right now), you can find that the Intel CPU is between 3 times faster and 9 times faster. Most of the times is about 6-7 times faster.

    See http://www.anandtech.com/bench/product/37?vs=836

    However, if you could compare an AMD FX-51 with an Pentium 66 Mhz (best CPU in September 1993), I think that the difference would be way greater.

    CPU process is currently focused on efficiency and lower power. However, in the ARM field, you can still find progress in CPU performance.

    1. Re:Comparison with current CPUs by bjackson1 · · Score: 1

      The Athlon X2 4450e was released in April of 2008, so we are only looking at 5 years difference not 10 years. I think the more interesting comparison would be to what the Athlon FX-51 and the new Apple A7 chip look like, given they are the first 64 bit chips of their class.

    2. Re:Comparison with current CPUs by Anonymous Coward · · Score: 0

      Simply put, ARM has always put foward efficiency first, mostly because of it's RISC architecture. Now ARM is racing up the CPU performance with great leap as it's way easier to increase processing power than efficiency. I'm pretty sure that it will be able to catch up on AMD and should near Intel in the next 4-5 years.

    3. Re:Comparison with current CPUs by Anonymous Coward · · Score: 0

      Simply put, ARM has always put foward efficiency first, mostly because of it's RISC architecture.

      RISC architecture doesn't imply a focus on efficiency or low power -- just look at IBM's POWERn line, and the derivative PowerPC G5 formerly used in PowerMacs. Apple chose to switch to x86 rather than fund further development of PPC G5, because they felt Intel x86 had a much better roadmap for performance and power efficiency. (and, as it turns out, it did)

      Now ARM is racing up the CPU performance with great leap as it's way easier to increase processing power than efficiency.

      This is the exact opposite of the truth. Desktop x86 class performance is very difficult, far more difficult than low power. And ARM's present business model (a supplier of semi-generic CPU and other cores which are ported across many fabs) makes it extremely difficult for them to directly challenge AMD, much less Intel.

      I'm pretty sure that it will be able to catch up on AMD and should near Intel in the next 4-5 years.

      That's nice. Just one problem -- you're totally clueless, and unaware of it.

      The real problem facing AMD and Intel is not ARM catching up, it's that there's declining demand for the cost-no-object take-it-to-the-limit high performance cores they're used to designing. Advances in chip fabrication technology have made it relatively easy to put what was 200W desktop class performance 7-ish years ago into a very cheap ~2W mobile chip. That's enough performance for an awful lot of software, and the new popularity of operating systems targeting these small devices has noticeably reduced demand for PCs even though they can't match PC performance. So what Intel and (to a lesser extent AMD) are actually worried about is that demand for the big powerful cores goes away, and prices erode, making it impossible for them to invest so much in what they're good at (it takes a lot more money to design a cutting edge high performance CPU core).

      But they're already responding. It's not just theory that it's easier to go down the power curve from where Intel's at than up the performance curve from where ARM's at -- Intel has already done it. You can already buy Android smartphones with an Intel x86 chip inside. On the other side, are there any ARM-designed cores able to challenge Intel's high end performance? Nope. There isn't even a hint of a future roadmap to them.

  14. Re:10 years later and applications are still 32bit by Anonymous Coward · · Score: 0

    Then you must have not played Skyrim with 4k texture mods, crashing to desktop all the time because it reaches its magic max memory limit.

  15. P4 vs Athlon XP by Anonymous Coward · · Score: 0

    I wouldn't say that the P4 was bludgeoning the Athlon XP series except maybe right at the verrry end of its career. While P4s clearly had the advantage when it came to integer math, Athlon XP's Barton core processors had a clear edge when it came to floating point math, allowing it to just outshine in gaming benchmarks over the P4. AND the XP's cost a fraction of the price! For a gamer, the AthXP vs P4 debate was seen a bit differently.

    1. Re:P4 vs Athlon XP by Dputiger · · Score: 4, Interesting

      As the author of the article:

      In 2000 - 2001, the Athlon / Athlon XP were far ahead of the P4. But from Jan 2002 to March 2003, Intel increased the P4's clock speed by 60% and introduced Hyper-Threading. SSE2 became more popular during the same time. As a result, the P4 was far ahead of Athlon XP by the spring of the year in most content creation, business, and definitely 3D rendering workloads. Now it's true that an awful lot of benchmark shenanigans were going on at the same time, and the difference between the two cores was much smaller in high-end gaming. But if you wanted the best 'all around' CPU, the P4 Northwood + HT at 2.8 - 3.2GHz was the way to go. Northwoods were also good overclockers -- it was common to pick up a 2.4GHz P4 and clock it to 3 - 3.2GHz with HT.

      Athlon 64 kicked off the process of changing that, but what really did the trick was 1). Prescott's slide backwards as far as IPC and thermals and 2) The introduction of dual-core. It really was a one-two punch -- Intel couldn't put two Pentium 4 3.8GHz chips on a die together, so the 820 Prescott was just 3.2GHz. AMD, meanwhile, *could* put a pair of 2.4GHz Athlon 64's on a single chip. Combine that with Prescott's terrible efficiency, and suddenly the Athlon 64 was hammering into the P4 in every workload.

    2. Re:P4 vs Athlon XP by FreonTrip · · Score: 1

      The misleading thing about benchmarks is that they're generally prebaked - there's no chance for "surprise" physics interactions or various pipeline-stalling things that tended to trip up the Pentium 4. From personal experience I'll tell you that my old 2.8 GHz Pentium 4 generally didn't do as well as my Athlon XP 2400+ in Doom 3, Bioshock, or Unreal Tournament 3. The latter two should have been poster children for the Netburst chip by comparison. Also, the Pentium D 820 was a 2.8 GHz chip: it was the miserably hot 130W TDP 840 that ran at 3.2 GHz. But you're correct on the other counts - the higher IPC and integrated memory controller were both HUGE advantages over a latency-crippled, deeply pipelined architecture. The Pentium D was itself a flailing, mostly failed response to the surge in mindshare the Athlon 64 X2 created until the Core architecture could be prepared.

    3. Re:P4 vs Athlon XP by fuzzyfuzzyfungus · · Score: 2

      Not that it mattered to the P4-area contest(where desktop OSes and workloads would remain 32 bit for quite some time to come, and RAM fairly expensive); but the A64 was a real smack in the face for IA64...

      Intel has their grand, big-iron-class, future-of-enterprise-computing 64-bit architecture, then AMD pops up "Hi guys, who wants a 64-bit CPU, fully backwards compatible with your 32-bit x86 code and pretty damn fast at that, for only slightly more than the price of a nice desktop CPU?"

      Boom. Headshot. Game over man, game over.

      The overclockable 2.4-2.8 Northwoods kept up on the 32 bit side at the time, and Intel has since swallowed their pride and put out some genuinely brutal 'EMT64' parts; but IA64 was buried beyond hope.

    4. Re:P4 vs Athlon XP by Anonymous Coward · · Score: 1

      Absolutely right. And Intel really wanted the Itanium to succeed, because only Intel could build Itaniums and they really really wanted to be a sole-source for processors. That way they could more easily jack the price up.

      I'm very glad AMD was able to cut the legs out from underneath the Itanium.

  16. .dll16 in WINE seems to indicate so. by Anonymous Coward · · Score: 0

    And I'm pretty sure 'dll hell' started in the windows 3.x and below era. Although it's possible I am wrong.

  17. Re:10 years later and applications are still 32bit by Anonymous Coward · · Score: 1

    You mean modifying a game outside of its intended scope makes it work in unexpected ways?

    That's so weird.

  18. Toms = Intel PR by charnov · · Score: 0, Troll

    Toms got outed years ago as being paid by Intel. If you want good, unbiased reviews of games and gaming hardware, go to HardOCP.

    --
    [RIAA] says its concern is artists. That's true, in just the sense that a cattle rancher is concerned about its cattle.
    1. Re:Toms = Intel PR by sunami · · Score: 1

      I don't remember this, and can't find it noted. Can you link to where you saw that? I've been using tom's hardware for years.

    2. Re:Toms = Intel PR by Anonymous Coward · · Score: 0

      You are lying or grossly misinformed.

      Back in the 90's Intel attempted to influence Tom's Hardware (and other sites) through ad money which ultimately backfired and greatly embarrassed Intel.
      http://www.nytimes.com/1997/03/12/business/dispute-over-unauthorized-reviews-leaves-intel-embarrassed.html

      In 2000 Tom's (and HardOCP) reviewed the then-new 1.13 Ghz PIII so poorly that Intel pulled it off the market.
      http://www.theregister.co.uk/2000/08/28/intel_recalls_1_13ghz_pentium/

      There has been no "outing" about Tom's and Intel. In short, citation needed, troll.

  19. Re:10 years later and applications are still 32bit by Waffle+Iron · · Score: 1

    10 years later and we're still running games and applications that are 32bit that only use a single core.

    At least 64-bit OSes are widespread now.

    Almost ten years after the 80386 was introduced, most people were still running "OSes" which were little more than GUI shells running on 16-bit DOS.

  20. zombie computing by Anonymous Coward · · Score: 1

    And, of course, AMD is a much cheaper option to get ECC on the desktop. Which is less and less a luxury and more and more a necessity. Simply because the memory densities keep going up, yet the MTTF doesn't go down as much (if at all). Same problem with hard drives, actually, so RAID doesn't really cut it any longer. Need parity files on the disk itself as well as at least mirrored disks.

    Still and all, this sort of reminiscing makes me long for alpha and parisc. ARM isn't quite there yet, and mips isn't available in the top performing brackets any longer either. And POWER? Well, few people can take that much gouging.

    1. Re:zombie computing by bill_mcgonigle · · Score: 1

      Need parity files on the disk itself as well as at least mirrored disks.

      You need ZFS. :) No, really, it checksums all the writes, which reflects the modern reality. I've got a machine at home in the basement that is effectively just ECC RAM and a bunch of disks (RAID-Z on that one I think, RAID-Z2 at work), to store our home data. It's still cheaper to do it in one spot and then run non-ECC hardware elsewhere, accessing the reliable data over gigabit.

      On my laptop, where I have many fewer options, I've just got ZFS running on top of a single LUKS volume. But for the same reasons (and compression helps on the laptop).

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
  21. No not really by Sycraft-fu · · Score: 1

    On high end games, the CPU gets hit hard. AI, physics, etc, all need a lot of power. Battlefield 3 will hit pretty hard on a quad core CPU, while hitting hard on a high end GPU at the same time.

  22. AMD was king of the hill, but... by Anonymous Coward · · Score: 4, Interesting

    AMD, forgotten by most of you, purchased a CPU design company not long after it lost the right to clone Intel CPU designs. The people from this company gave AMD a world beating x86 architecture that became the Athlon XP and then Athlon 64 (and true first x86 dual core), thrashing Intel even though AMD was spending less than ONE-HUNDREDTH of Intel's R&D spend.

    What happened? AMD top management sabotaged ALL future progress on new AMD CPUs, in order to maximise salaries, bonuses and pensions. A tiny clique of cynical self-serving scumbags ruined every advantage AMD had gained over Intel for more than 5 years afterwards. Eventually AMD replaced its top management, but by that time it was too late for the CPU. Obviously, AMD had far more success on the GPU side after buying ATI. (PS note that ATI had an identical rise to success, when that company also bought a GPU design team that became responsible for ALL of ATI's world-beating GPU designs. Neither AMD nor ATI initially had in-house talent good enough to produce first rate designs.)

    Today, AMD is ALMOST back on track. It's Kaveri chip (2014) will be the most compelling part for all mains powered PCs below high-end/serious gaming. In the mobile space, Intel seems likely to have the power-consumption advantage (for x86) across the next 1.5 years at least. However, even this is complicated by the fact that Nvidia is ARM, and AMD is following Nvidia, and is soon to combine its world beating GPU with ARM CPU cores.

    At this exact moment, AMD can only compete on price in the CPU market. Loaded, its chips use TWICE the power of Intel parts. In heavy gaming, average Intel i5 chips (4-core) usually wallop AMD's best 8-cores. In other heavy apps, AMD at best draws equal, but just as commonly lags Intel.

    Where AMD currently exterminates Intel is with SoC designs. AMD won total control of the console market, providing the chips for Nintendo, Sony and Microsoft. Intel (and Nvidia) were literally NOT in the running for these contracts, having nothing usable to offer, even at higher prices or lower performance.

    AMD is currently improving the 'bulldozer' CPU architecture once again for the Kaveri 4-core (+ massive integrated GPU and 256-bit bus) parts of 2014. There is every reason to think this new CPU design will be at rough parity with Intel's Sandybridge, in which case Intel will be in serious trouble in the mains-powered desktop market.

    Intel is in a slow but fatal decline. Intel is currently selling its new 'atom' chips below cost (illegal, but Intel just swallows the court fines) in an attempt to take on ARM, but even though Intel's 'atom' chips are actually Sandybridge class, and have a process advantage, they are slaughtered by Apple's new A7 ARM chip found in the latest iPhones. A7 uses the latest ARM-64 bit design known as ARMv8, making the A7 and excellent point of comparison with the original Athlon 64 from years back.

    Again, AMD is now x86 *and* ARM. AMD has two completely distinct and good x86 architectures ('stars-class' and 'bulldozer-class'. Intel is only x86, and now with the latest 'Atom' has only ONE x86 architecture in its worthwhile future lineup. Intel has other x86 architectures, but they are complete no-hopers like the original Atom family, the hilariously awful Larabee family, and the putrid new micro-controller family. Only Intel's current sandybridge/ivybridge/haswell/new-atom architecture has any value.

    1. Re:AMD was king of the hill, but... by 0123456 · · Score: 1

      AMD is not ARM. ARM is ARM. Anyone can buy an ARM license and start releasing ARM chips. AMD are producing ARM chips because they can't compete with Intel in the x86 market.

      Nothing stops Intel releasing ARM chips, as they have in the past, except the margins would probably be awful compared to their x86 lineup.

    2. Re:AMD was king of the hill, but... by Anonymous Coward · · Score: 0

      sadly, top management killing companies happens often, and everywhere; and it might just happen again

    3. Re:AMD was king of the hill, but... by jonwil · · Score: 2

      The most telling thing about AMD is that their first generation Bulldozer-architecture CPUs were getting their pants creamed not just by their Intel competitors but by the last-generation AMD parts.

    4. Re:AMD was king of the hill, but... by Anonymous Coward · · Score: 0

      Sandybridge as you call it isn't the latest Intel processor design.

    5. Re:AMD was king of the hill, but... by Anonymous Coward · · Score: 0

      AMD is currently improving the 'bulldozer' CPU architecture once again for the Kaveri 4-core (+ massive integrated GPU and 256-bit bus) parts of 2014. There is every reason to think this new CPU design will be at rough parity with Intel's Sandybridge, in which case Intel will be in serious trouble in the mains-powered desktop market.

      That would be true if Intel was stuck on Sandy Bridge (a 2011 CPU) in the "mains-powered" desktop market. They're already +2 generations past Sandy, and are likely to be +3 (Broadwell) before Kaveri ships. You might want to consider rethinking your ludicrous over-optimism for AMD's future prospects and absurd pessimism for Intel's. How many times have we heard AMD fans getting excited about the next tweak of Bulldozer being poised to rule the world, only for each and every one to be late and a resounding "meh, will barely keep them afloat"?

      Intel is in a slow but fatal decline. Intel is currently selling its new 'atom' chips below cost (illegal, but Intel just swallows the court fines) in an attempt to take on ARM,

      Bullshit, Intel has not paid a single cent of fines for selling Atom chips below cost. Is it just that you can't imagine Intel selling any chip cheaply? Your lack of imagination does not imply that Intel is breaking the law.

      Also, you've set up a bit of a catch-22 for Intel in that it sounds like you'd only consider them to be aboveboard if they were charging more than their competitors. Which is absurd, the cell phone market is quite price sensitive. They're also facing resistance thanks to x86 -- it is actually a competitive disadvantage since x86 phones have to run many important Android apps through an emulator, and no matter how good Intel says the emulator is it isn't as good as native code. Since their products are competitive but not world beaters in terms of performance and power, they pretty much have to offer something on the price side to get design wins. Basically you're saying you'd only be happy if Intel was deliberately failing.

      And even offering lower prices is not necessarily a sign of evil. As the owner of its own notoriously efficient fabs, which are at a more advanced process node than anybody else's, Intel probably has a better cost structure than most ARM SoC suppliers. All else equal, they should be able to make a profit while charging less per chip. You will not find many government lawyers sympathetic to the idea that selling products for a profit is an anticompetitive action which should be prosecuted -- they'd get laughed out of court.

      but even though Intel's 'atom' chips are actually Sandybridge class,

      What? No they aren't, you're crazy. The Sandy Bridge core is from a different planet. It's larger, way more complicated, way more power hungry, and way faster. The two cores are not very related beyond both being x86 cores designed by Intel.

      and have a process advantage, they are slaughtered by Apple's new A7 ARM chip found in the latest iPhones. A7 uses the latest ARM-64 bit design known as ARMv8, making the A7 and excellent point of comparison with the original Athlon 64 from years back.

      64-bitness does not in and of itself make one CPU design like another.

      Again, AMD is now x86 *and* ARM. AMD has two completely distinct and good x86 architectures ('stars-class' and 'bulldozer-class'. Intel is only x86, and now with the latest 'Atom' has only ONE x86 architecture in its worthwhile future lineup. Intel has other x86 architectures, but they are complete no-hopers like the original Atom family, the hilariously awful Larabee family, and the putrid new micro-controller family. Only Intel's current sandybridge/ivybridge/haswell/new-atom architecture has any value.

      I am amused by the way that you reserve negatively slanted emotional descriptions for Intel products while effusively praising AMD's even though the two AMD archi

  23. Re:10 years later and applications are still 32bit by UnknowingFool · · Score: 1

    Damnit people, do not give Adobe's Flash developers any ideas!!

    --
    Well, there's spam egg sausage and spam, that's not got much spam in it.
  24. Re:10 years later and applications are still 32bit by Anonymous Coward · · Score: 0

    Trololol stop using Windows. Linux and Mac are generally 100% 64-bit for what most people are using.

  25. *THE* chip that changed the world? by Anonymous Coward · · Score: 0

    I'm sorry, that is the height of hyperbole. There are *MANY* chips that I would say had a bigger impact than AMD's introduction of x86-64.

    It was a very big impact, for sure, but not "the" chip...

  26. that first tyan dual cpu board by rs79 · · Score: 1

    so i was already feeling stoked about finally getting around to finding a matched pair of the fastest cpus that I can put into this board that's been sitting in a box for SIX YEARS, they boot and now I read this. /me - does the peacock strut happy dance thing.

    i was always a fan of AMD going back to the 8x300 bit slice stuff. they're clever boys.

    --
    Need Mercedes parts ?
    1. Re:that first tyan dual cpu board by Anonymous Coward · · Score: 0

      Ohai and congrats. No idea who you are. Building systems is fun! Nice homepage. Cheers!

  27. Those bastards by LodCrappo · · Score: 5, Funny

    Apple just released a 64bit processor, and now AMD is copying it TEN YEARS ago?!?

    Can the industry please do something original and quit just following wherever Apple leads it?

    --
    -Lod
    1. Re:Those bastards by tie_guy_matt · · Score: 0

      Is this intended to be a joke?

      Apple doesn't make processors. These days macs use intel chips. Back in the day they used to use powerpc chips made by motorola and IBM. If you go into the way-way back machine they once used m68k chips from motorola. On the iphone and such they just use ARM cpus. Then if you go all the way back to the days when Woz was making computers in his garage you will see that they used MOS 6502 chips. But they have never really been in the cpu industry.

    2. Re:Those bastards by FreonTrip · · Score: 1

      First: yes, it's clearly a joke. How can you copy something done ten years earlier? :P

      Second, Apple does make their own ARM CPUs these days. They build and design licensed ARM CPUs for their iOS devices these days, which includes AppleTV, iPhones, iPads, and iPods, but for their Mac / OS X business they are still 100% Intel. Their latest design is starting to turn some heads.

    3. Re:Those bastards by Anonymous Coward · · Score: 0

      Is this intended to be a joke? No? Oh dear.

    4. Re:Those bastards by slash.jit · · Score: 1

      Yes it is a joke and you ruined it by asking that question. Why can't you just enjoy the joke and not ask unnecessary questions!

    5. Re:Those bastards by ModernGeek · · Score: 1
      --
      Sig: I stole this sig.
    6. Re:Those bastards by unixisc · · Score: 2

      AMD acquired a company called PA Semi, which originally made Power CPUs, and changed its focus to ARMs - their 'A' line of CPUs. So when AMD was a part of the AIM alliance - Apple, IBM & Mot, they didn't make CPUs, but now, they make their own ARMs and fab them from TSMC & Samsung. Which is ironic in that had they done the same when they were part owners of PowerPC, they could have had all the power saving modes that they wanted IBM to design in.

    7. Re:Those bastards by Anonymous Coward · · Score: 0

      To be honest, Apple does not have a Fab, it is Samsung who actually makes the A7.

    8. Re:Those bastards by UnknowingFool · · Score: 1

      To be honest not everyone who has a chip makes it. Qualcomm uses fabs like Samsung and TSMC to make their chips.

      --
      Well, there's spam egg sausage and spam, that's not got much spam in it.
    9. Re:Those bastards by UnknowingFool · · Score: 1

      You mean Apple acquired PA Semi right? They also acquired Intrinsity. Between the two they got the chip design experience they needed.

      --
      Well, there's spam egg sausage and spam, that's not got much spam in it.
    10. Re:Those bastards by nbritton · · Score: 1

      Apple just released a 64bit processor, and now AMD is copying it TEN YEARS ago?!?

      The first 64 bit system I remember was the Atari Jaguar, circa 1993. Also the Super Nintendo used the same processor as the Apple IIgs, and the Macintosh had a 32-bit processor from day one. The PowerPC G5 was Apple's first 64-bit processor, circa 2003.

    11. Re:Those bastards by FreonTrip · · Score: 1

      By that metric AMD doesn't "make" its chips either. Nor do many companies, even if the IP is firmly theirs and the designs were entirely in-house. That's a strange, argumentative kind of quibbling.

    12. Re:Those bastards by Anonymous Coward · · Score: 0

      The first 64 bit system I remember was the Atari Jaguar, circa 1993.

      FYI, that was high-grade marketing bullshit. The Jaguar had one 68000 CPU, intended to handle joystick IO and other menial tasks, and two "main" CPUs which were both 32-bit RISC of some semicustom Atari design. You were supposed to use one of them for sound -- it had some DSP functions built in or something like that -- and the other for graphics. Atari got "64-bit" out of claiming that adding two 32-bit RISCs together makes 64. Yes, really. (By that logic you could call the PS4 a 512-bit system, since it has eight 64-bit cores...)

      Nintendo had the first true 64-bit game console by a wide margin. So wide that, as far as I know, it was the only 64-bit console until this year's PS4 and XBox One, including Nintendo's own successors to the N64. N64 used a 64-bit MIPS CPU mainly because the hardware was a collaboration with SGI. SGI loved itself some 64-bit MIPS, that being what they used in their workstation computers, and it really didn't cost much thanks to the existence of the MIPS R4000 (a low cost 64-bit MIPS design), so both parties figured "why not". Nintendo's subsequent hardware partner (IBM) had a low-cost 32-bit PowerPC "G3" core originally developed for Macs, but no low-cost 64-bit core. Rather than pay for developing one they quietly dropped the 64-bit marketing.

  28. Re:10 years later and applications are still 32bit by Anonymous Coward · · Score: 0

    From the steam hardware survey page around 10% of the computers still run a 32bit OS (Windows XP, Windows Vista 32bit,...) and the development of current games was started when that percentage was even higher.

      The multicore situation is another problem, getting multi-threading right and bug free is hard and getting more than a marginal performance increase might require an almost complete redesign of engines that have been in development for over a decade (just a few keyworks: shared mutable state, consistency, synchronisation overhead, deadlocks, visibility).

  29. Comment removed by account_deleted · · Score: 1

    Comment removed based on user account deletion

  30. Re:10 years later and applications are still 32bit by timeOday · · Score: 1

    The need for computers that can run multiple programs concurrently with a total of > 4GB RAM is more than the need for any single program to consume multiple cores or > 4 GB RAM.

  31. Re:10 years later and applications are still 32bit by Anonymous Coward · · Score: 1

    Not for much longer. DICE has already announced that some of their new Frostbite 2 games will be 64-bit only, due to memory requirements beyond 2 GB.

  32. Re:10 years later and applications are still 32bit by adolf · · Score: 1

    When I'm waiting for an application to do whatever that application is doing, and that application is only using one core, then yes, I really do need it to use more than one core.

    To suggest otherwise is also to suggest that computers are fast enough, and that general-purpose computing is a solved problem.

    I don't think we're anywhere near that point just yet.

  33. Re:10 years later and applications are still 32bit by yuhong · · Score: 1

    I mentioned that Caldera actually sued MS based on the fact that Win9x was still based on DOS in my blog article on the OS/2 2.0 fiasco, because OS/2 never depended on DOS.

  34. 16 to 32 was mostly incompatible by erice · · Score: 1

    Wow. Ten years. And here I am still dealing with 64 bit incompatability issues every six months or so.

    Out of curiosity, how long did 16bit library problems linger after the 32 bit move?

    16 to 32 was a much more radical change. Segments to flat. In the Unix line, this happened in the early 80's (late 70's?) when few systems were deployed.

    In the Wintel line, it was also cooperative to preemptive. Very painful. Very manual. It took 10 years just to let go of 16 bit device drivers and many were never ported.

    Classic Mac, Amiga, and Atari ST had an easier time since their "16-bit" systems were already 32-bit internally. Even then you had a few years of dealing with geniuses who stored data the upper 8 bits of pointers and some significant software (like AmigaBasic) were never ported.

  35. Re:64 bit - Really, what's the point? by 0123456 · · Score: 1

    Today, I do not see any apps in general use that need or require access to memory beyond a contiguous block of data beyond 4GB (frankly far far less).

    Then you're not looking very hard. There are plenty of games that fall over as soon as you install enough mods that you go over 2GB (the maximum amount of RAM the majority of 32-bit apps are allowed to access in Windows). There are also games which install crappy versions of textures in 32-bit mode because they would otherwise hit the 2GB limit.

    PAE is a disgusting kludge. There is simply no reason not to run 64-bit apps on a 64-bit x86 OS... unless you're stuck with a proprietary antique like Windows.

  36. Re:10 years later and applications are still 32bit by Anonymous Coward · · Score: 0

    It's too late, they already have a 64-bit version out.

  37. Bechmark what people use? Which people? by coder111 · · Score: 4, Insightful

    Ok, I'm a Java developer, working on Linux. 99% of benchmarks you see on-line are benchmarks of some stuff on Windows, which is immediately skewed towards intel because lots of Windows DLLs are built with Intel compiler which disables optimizations and advanced features on non-intel CPUs, even if those features are supported.

    Also, most benchmarks are some synthetic benchmarks (compiled with Intel compiler), or some 3D games, or some video transcoding. I do none of that, and what I do I do not do it on Windows. I am yet to see a site that benchmarks Java compilers, Java IDEs, databases, application or Java web servers, etc. on Intel vs AMD vs ARM platforms on Linux. The only site that comes close is Phoronix. And if you look at their Linux benchmarks, difference between AMD and Intel CPUs is much less than on Anandtech or Toms Hardware. Intel is still making faster CPUs, but not that much faster.

    --Coder

    1. Re:Bechmark what people use? Which people? by 0123456 · · Score: 1

      Meanwhile, in the real world, most people's primary reason for buying a faster CPU is... to play games on Windows.

      If AMD suck at gaming on Windows, they want to know that.

      Now, if a site is only using games as a benchmark, yes, that's a problem. But Toms Hardware, at least, usually covers a range of different types of applicaiton benchmarks, for the minority who aren't looking at the CPU as something to run games on.

  38. Re:64 bit - Really, what's the point? by MasterOfGoingFaster · · Score: 1

    Today, I do not see any apps in general use that need or require access to memory beyond a contiguous block of data beyond 4GB (frankly far far less).

    Lacking inside information, I could be wrong, but I believe you'll find Photoshop, Premier Pro, After Effects, Solidworks, PTC Creo, various FEA packages, and CAM applications all use blocks of RAM beyond 4GB. And that just a sample of MY machine.

    More knowledgeable folks (i.e. coder's for the above apps) can correct me if I've got it wrong. But there is clearly a need for >32bit memory address space.

    --
    Place nail here >+
  39. Comment removed by account_deleted · · Score: 1

    Comment removed based on user account deletion

  40. Re:64 bit - Really, what's the point? by 0123456 · · Score: 1

    Sorry, I do not have the time to educate you here on the word "contiguous" in context to OS memory management. Also, I preambled my comment re: memory access not instructions.

    Maybe you should spend more time reading than 'educating'. My post was specifically about memory usage, and 'contiguous' is irrelevant to software, because the MMU makes it look contiguous. All Windows gives a 32-bit app is 2GB of RAM, unless the app and every single DLL it loads has the 4GB-aware flag (good luck with that).

    Many, many 32-bit Windows apps are hammering hard against that 2GB barrier right now.

  41. What kills AMD is a per-core license. by emil · · Score: 3, Informative

    Oracle's Enterprise database costs $47,500 per processor core. There is no way in heck that I'd choose AMD over Intel when I have to run more cores to get the same performance.

    Microsoft SQL Server Enterprise costs $6,874 per processor core.

    AMD has a heavy investment in the server space. They should negotiate lower per-core license costs in these cases; license parity with Intel is throwing them out of the data center.

    As the developers of x86-64, they should have a patent portfolio to do serious damage to 64-bit x86 systems vendors. Use it.

    1. Re:What kills AMD is a per-core license. by TheNinjaroach · · Score: 2

      Microsoft SQL Server Enterprise costs $6,874 [microsoft.com] per processor core.

      Ha! Did you know that Microsoft reduces "per core" licensing costs for SQL Server on AMD processors, because otherwise nobody could justify buying them?

      --
      I went to eat some animal crackers and the box said, "Do not eat if seal is broken." I opened the box and sure enough..
  42. 10 years. Bloody hell! by GarethIwanFairclough · · Score: 1

    Subject pretty much says it all. Oh well, moving on.

  43. Re:64 bit - Really, what's the point? by markhahn · · Score: 1

    The point is the new register set. Registers being wider is a happy side-effect, as is greater virtual address space. But the main point of AMD64 is more registers. and it started a sequence of ISA extensions that have dramatically improved compute-bound throughput via SIMD.

  44. NOT! by Anonymous Coward · · Score: 0

    How did this chip change the world again? Answer: It didn't.

  45. Re:64 bit - Really, what's the point? by Anonymous Coward · · Score: 0

    I like the ability to mmap 50gig files (or create shared memory objects of that size). [on a box with 256gigs of ram that is].

  46. Re:10 years later and applications are still 32bit by walbourn · · Score: 2
    The Xbox 360 and PS3 pushed game developers to embrace multi-threaded gaming when they were primarily single-threaded before that. This has definitely spilled over to more multi-core PCs, although there's been a number of 'early' generation games that tired to require quad-core before the PC gamers were really running them in large numbers. These console ports were often tuned for 3-6 threads, which didn't always map well to just a dual-core PC. The Xbox 360 and PS3 are also 32-bit platforms, so they didn't push developers to move to 64-bit native. Xbox One and PS4 are 64-bit native platforms, so that's hopefully going to accelerate the move to 64-bit. Game developers are already using x64 native builds of their games internally, but publishers don't want to pay to test, release, and support both a 32-bit and a 64-bit version of their game--or they released it but stopped updating the 64-bit version. The biggest driver for this is that 32-bit applications can only manage 2GB of virtual address space, and AAA PC games were already hitting this limit.

    So the focus on consoles in the industry, plus the lingering zombie life of Windows XP, had basically kept everyone stuck at Direct3D 9/Shader Model 3 and 32-bit for years. With next-generation consoles moving to Direct3D 11/Shader Model 5 and 64-bit, we should see more PC titles that take advantage of the latest generation PC hardware. BTW, for some technical background on 64-bit and gaming, see http://www.gamasutra.com/view/feature/3602/sponsored_feature_ram_vram_and_.php

  47. With respect... author needs a history lession by Anonymous Coward · · Score: 0

    I think the original Opteron needs some love here. It was the 64bit instruction set gamble. It was the server processor gamble. It was a fantastic leap in performance that had Intel running scared. Opteron provived the foundation for knocking Intel off their hill.

    However far more important historically was the AMD 386. This is the CPU that brought AMD out to pace with Intel and offered exceptionally higher headroom on clock speed. It also gave Cyrix it's brief chance compete as well. Also, It lead to the lawsuit that Intel couldn't copyright the processor numbering ( which ultimately resulted in the Pentium naming convention )

  48. Re:10 years later and applications are still 32bit by ThreeKelvin · · Score: 2

    Most don't really need two cores, but that's not a reason not to want two cores.

    I fell in love with multiple core processors when I first got one, not because my computer in general became faster (I'll bet that all but one of my cores are idling most of the time) but because my computer wouldn't get unresponsive when I was doing computationally heavy tasks (or programs crashed).

  49. I think that's what benchmark fanboys forget by Sycraft-fu · · Score: 1

    While you may have the benchmark you like the best that you think is wonderful, probably because it shows your favoured product in a good light, big numbers aren't what most people care about. They care about wall time of execution, or FPS, or the like of what they run. Well that varies person to person, but it is actual applications they care about.

    That's the reason I have an Intel CPU. For the programs I use, it does better. I don't really care why. You can argue that it is lack of optimization, hating AMD, Intel nefariousness, whatever. Doesn't matter, what matters is it works faster for me. That is what I care about.

  50. Re:10 years later and applications are still 32bit by Bert64 · · Score: 1

    I don't know about xbox 360, but the ps3 is most definitely using a 64bit cpu, i was running a powerpc64 linux distribution on it.

    --
    http://spamdecoy.net - free throwaway anonymous email - avoid spam!
  51. The chip that sunk the Itanic by unixisc · · Score: 3, Interesting

    The instruction set itself was an yawner - I was looking forward to 64-bit being the point where all CPUs become RISC, and where Windows NT could go from being Wintel only to NT/RISC.

    However, one delicious piece of irony that I love about the Opteron/Athlon 64 is that this was the architecture that sunk the Itanic. If the Itanium sank far worthier chips before it - PA-RISC, DEC Alpha and MIPS V, this architecture brought out the Itanic in Itanium. Originally, the Itanium was supposed to be the 64 bit replacement for x86, but thanks to this gag from AMD, it never happened. Instead, AMD started stealing the market, and to add insult to injury, when Intel tried entering w/ 64-bit extensions of its own, Microsoft forced them to be AMD compatible. So that Intel was ultimately forced to let x64 be the successor to x86, and let Itanium wither on the vine.

    Once that happened, Itanium followed the same path as the better CPUs that it killed above. Microsoft dropped support for it after Server 2008 and XP or Vista were never supported, Monterrey collapsed and to add insult to injury, even Linux - the OS that boasts about being ported everywhere - didn't want to remain supported on the Itanic. Today, the Itanic has as many OSs as the DEC Alpha had at its peak - 3: HP/UX, Debian Linux and FreeBSD.

    So no, the x64 didn't change the world. But it sure sunk the Itanic!

  52. Re:10 years later and applications are still 32bit by Anonymous Coward · · Score: 0

    there is no way you need to dedicate a whole core to music while surfing the internet.

    Ever tried loading a popular slashdot discussion page on an old singe-core machine?

    ... idiotic use cases people make up

    Oh wait, never mind.

  53. Re:10 years later and applications are still 32bit by ciderbrew · · Score: 1

    I can't mod this up, so have a +1 thank you.

  54. The sad thing.... by Anonymous Coward · · Score: 0

    I tried to follow all the links in the previous slashdot article.... and I think the ONLY one that worked was the one from HardOCP. More than half of the other review sites were completely gone, and the rest didn't have the content available any more.

    As for the FX-51 itself, I very clearly remember reading those articles when they first came out. I was so in lust with them.... At the time I think I was running a Duron processor that was like almost 50% overclocked as my main rig. :) I've been an AMD FANatic since the 486 days, back when AMD was the only way to get a 486 processor > 100mhz, or at least the only way I could afford. Amd also let me keep using my ancient socket 5 and socket 7 boards... K6 and K6+ anyone? :)

    This almost make me nostalgic enough to go home and try to get one of those old systems running again... I'm pretty sure I have all the components, right down to PCI vid cards and SDR ram. Wonder just how slow my original slot processor Athlon would run....

  55. Re:10 years later and applications are still 32bit by GarethIwanFairclough · · Score: 1

    EnBoost my friend. Check it out on skyrim nexus. Works wonders if you have a 3d card with a lot of ram, but it helps a lot with stability even on a 1gb card!

  56. From the "burning" AMD Athlons by charnov · · Score: 1

    Tom's got nailed over the "burning" AMD Athlons. If you dig through the HardOCP forums from back in 2001-2002, Kyle Bennett and others managed to trace some not so savory relationships with Intel and the site.

    I've never trusted Tom's since and they have a well established Intel bias. HardOCP is much more transparent in it's testng.

    --
    [RIAA] says its concern is artists. That's true, in just the sense that a cattle rancher is concerned about its cattle.