Slashdot Mirror


AMD One-Ups Intel With Cheap Desktop Chips

CWmike writes "Advanced Micro Devices on Tuesday announced inexpensive desktop microprocessors with up to six cores to put pricing pressure on rival Intel. AMD's new chips include the fastest AMD Phenom II X6 1075T six-core processor, which is priced 'under $250' for 1,000 units, AMD said. AMD also introduced a range of dual-core and quad-core Athlon II and Phenom II desktop microprocessors priced between $76 and $185. By comparison, Intel's cheapest six-core processor is the Core i7-970 processor, which is priced at $885 per 1,000 units, according to a price list on Intel's website."

362 comments

  1. I like AMD by Haedrian · · Score: 0, Offtopic

    Never had any problems with their chips and they're far cheaper than intel. I like them.

    1. Re:I like AMD by smash · · Score: 1

      I have had problems with AMDs in the past, but it wasn't the CPUs. The CPUs have always been fine, but often to support them you need to go to some busted-arse chipset from VIA, SLI or Nvidia.

      Admittedly it has been some time now since I've had a non-intel box (because of previous bad experiences with non-intel machines), but the most overlooked aspect of building a box is, imho the chipset.

      Now AMD appear to be building a lot more of the chipset either into the CPU or GPU (now they've purchased ATI) i might give htem another shot.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    2. Re:I like AMD by smash · · Score: 2, Funny

      SLI = something else. I forget the company, probably gone bust now. Chipset vendor from the late 90s onwards...

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    3. Re:I like AMD by Anonymous Coward · · Score: 0

      MSI?

    4. Re:I like AMD by Elbereth · · Score: 1

      ALI (Acer Labs, which eventually got spun off from Acer, I believe). There was also SIS (Silicon Integrated Systems) and OPTi.

      But, yeah, the chipsets were usually the problem. I never had a problem with Nvidia's chipsets, personally, but I hear other people have. I'm not sure what kind of wonky add-in cards they were using, but I loaded my systems with lots of stuff, with no issue. There was that one Nvidia chipset that caused hard drive corruption when you used it with a Sound Blaster, but I skipped that one.

    5. Re:I like AMD by smash · · Score: 1

      Yeah, MSI i think. Either way, chipsets are the problem more often than CPUs...

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    6. Re:I like AMD by Vectormatic · · Score: 1

      ALI/ULI

      They were bought out by nvidia i think, after they did a last hura for the amd 64, with a chipset which had both PCI-e and native AGP, asrock built some very good boards with their chips.

      anyway, as long as you stick to the pick of the day, AMD has some good chips, in the athlon days, VIA KT266/333, Athlon XP > Nforce 2, amd 64 > Nforce 3/4, after that ATI picked up the gauntlet, and these days you just want an AMD/ATI chipset, excellent integrated graphics and performance.

      Just keep in mind that picking the right chipset isnt enough if the mobo was designed crappy, i've had a MSI nforce 2 mobo which was SHIT, despite previous good experiences with the NF2 chipset

      --
      People, what a bunch of bastards
    7. Re:I like AMD by smash · · Score: 1

      Yeah thats the problem nvidia chipset i was talking about.

      Its a damn shame, because AMD have had some fine CPUs in the past - but typically the motherboard selection was pretty bad. I lost count of the number of broken VIA chipset problems I dealt with and just got sick of it.

      Went to an intel BX chipset P2-350 (back in the day) and never looked back.

      Its not just AMD affected though i might add - other motherboards for intel i have had wierd buggy shit on. Ended up just going for intel desktop board + intel cpu and have had very minimal problems since.

      If there's a decent AMD CPU + motherboard offering available, I'll give it a shot.

      Problem is, plenty of hardware review sites focus on benchmarks, and differences in SiSoft benchmarks, or FPS ratings, when i just don't care about 1-5% performance increases. I just want the box to behave properly and work.

      I don't want to be dealing with wierd and wonderful bios/chipset issues. I guess its why I like Macs so much these days... if it doesn't work, there's only one vendor to blame :)

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    8. Re:I like AMD by Calinous · · Score: 1

      Frankly, there was a time when motherboard reviews from Anandtech presented the number of times the board crashed during testing. It then went lower and lower, then they only crashed when using interleaved memory banks, then they didn't crash at all during normal use.
            Or maybe the crashes weren't reported any longer.

    9. Re:I like AMD by mr_mischief · · Score: 1

      AMD and NVidia are really the only players in AMD board chipsets these days, and AMD chipsets are on all the really well-performing boards anyway. The tight memory timings you can get with the recent AMD chipsets are great. Now, if only they'd leapfrog Intel's triple-channel and go quad-channel for memory modules, you'd see some really nice performance.

      I know some Intel chips outrun AMD chips, but after the whole Randal Schwartz fiasco, Intel refusing to do 64-bit extensions to ia32 because they were counting on Itanium (then bringing out EMT64 as a knockoff of AMD) and messing around with the FSB for so long until they took AMD's lead again for the on-die memory controller, I'll keep rewarding AMD for the way they do business and push the market forward.

      I do buy Intel chips sometimes, usually used. Most of my systems are AMD, though, and through the years although I've been mnostly happy with both brands I've been more happy with AMD.

    10. Re:I like AMD by mr_mischief · · Score: 1

      Try an AMD 790 or newer chipset board. DFI, ECS, Asus, Gigabyte, and Foxconn all make them. You should be able to find a decent board from one of those companies. MSI, AsRock, MSI, Biostar, Jetway, and even more companies also make them.

      AMD's chipset takes take of the northbridge and soutbridge, the USB, SATA, PCI and PCI-E, etc. Any other chips on the board will be sound or Ethernet unless it's a board with something somewhat unusual like more than one onboard Ethernet or more than half a dozen or so SATA ports.

      So if you get any board with an AMD chipset, there are really only a few other things to consider:

      1. build quality, of course
      2. which southbridge chip is paired with the northbridge, because sometimes they cheap out and put an older SB on a board than the main chipset normally pairs with
      3. claimed memory speeds (Some manufacturers using the same chipset rate their boards as less capable than others and even less than the specs for the chipset. I always doubt the quality of the board if they can't say it'll do what the chipset can normally do.)
      4. customer support in case you do get a bad board (Newegg's reviews usually work well for guessing this, even on other products from the company.)
      5. reported problems with the board (Newegg reviews are usually handy for this for a particular product if it's been on the market for a while.)
      6. price (of course)
      7. return policy at your merchant of choice for that board (again, just in case)

      You should of course check CPU and memory compatibility carefully if it's been a long time since you've done an AMD build. That's no fault of the chipsets or boards, but just a fact when there are so many chipsets and CPUs on the market at the same time.

      My main desktop rig right now is an ECS A790GXM-AD3 Black Edition with a Phenom II x4 955 Black Edition processor and 2x2 GiB Crucial Ballistix DDR3 1333 rated at CL6 (6-6-6-20). I've had it for quite a while now, and I'm pretty happy with the reliability and with the speed for the price. It'll overclock pretty darn well, too. I don't like the extra fan noise and it's plenty fast enough for what I do with it without overclocking.

      I don't play a lot of graphics-heavy games right now, so I'm running the integrated graphics. I average near 100 FPS in AssaultCube 1.1.0.1 with that, and it spikes to over 130 and never falls below 80. I should mention that's on Mandriva 2010.1 for AMD64 with the Catalyst blob. The board does have two x16 PCI-E 2.0 slots that will run full speed together. I just haven't put anything in either of them for right now because my fastest discrete PCI Express video card is a 1600 Pro.

    11. Re:I like AMD by wildstoo · · Score: 1

      I bet you're actually thinking of SiS, since MSI never designed any chipsets themselves and are purely manufacturers.

    12. Re:I like AMD by Anonymous Coward · · Score: 0

      You meant SIS probably.

    13. Re:I like AMD by anUnhandledException · · Score: 1

      Nvidia (nforce) SATA controllers also don't pass TRIM commands which is essential to maximizing SSD performance.

      The best solution (for me) was simply getting new motherboard. This time it has AMD chipset.

      The problem w/ NVIDIA chipsets has always been proprietary garbage. How/why Nvidia SATA controller doesn't play nice w/ SSD I don't know but the issue is well documented and there is no solution (other than never again trusting Nvidia w/ chipset duties).

      It is good AMD got into chipset business. AMD platforms were always entrusted w/ 3rd party for the most critical aspect (chipset). With Intel you had the reliability of Intel chipset. Even when Nvidia and SIS made chipsets for Intel boards you at least had the OPTION of going with Intel solution.

    14. Re:I like AMD by anUnhandledException · · Score: 1

      Just go w/ AMD chipset + AMD cpu or Intel chipset + Intel CPU.

      There is no reason for using 3rd party chipset anymore. They tend to have lower performance, more issues, and more crashes.

      Anyways the recommendation is academic anyways. Nvidia crappy solutions has edged them out of consideration. Taking a look at newegg for AMD newest sockets (AM3) the only chipset solution is AMD (700 series or 800 series). For Intel newest sockets 1156 & 1364 the only chipsets are made by Intel

    15. Re:I like AMD by Anonymous Coward · · Score: 0

      the company name you seek is SIS

    16. Re:I like AMD by raw-sewage · · Score: 1

      I have had problems with AMDs in the past, but it wasn't the CPUs. The CPUs have always been fine, but often to support them you need to go to some busted-arse chipset from VIA, SLI or Nvidia.

      ...

      Now AMD appear to be building a lot more of the chipset either into the CPU or GPU (now they've purchased ATI) i might give htem another shot.

      As another poster mentioned, there are now really only two chipset vendors for AMD CPUs, nVidia and AMD itself (at least in the desktop space, there may be others in the server space). I haven't done a detailed study recently, but last I checked, both were still sub-par compared to Intel's chipset offerings. nVidia had fairly widespread problems with the actual manufacturing of their chips; it tended to affect laptops more than desktops, but both were at risk. In general, the nVidia chipsets at least used to be quite power hungry.

      The AMD chipsets just don't compete with the Intel equivalent. There was a website that benchmarked AMD's SATA 3 against Intel's SATA 2; the Intel won! Also, I don't know if they still exist, but AMD's SATA performance used to change depending on whether you used the native AMD interface or switched to AHCI mode. I forget the details, but you can google for it. AMD also doesn't produce their own ethernet chips. That's not a problem, but it means that AMD boards usually come equipped with an el-cheapo Atheros or Realtek ethernet implementation. I think that Intel makes the best ethernet hardware, hands-down. Yet, outside of the server space, it's impossible to find an AMD board with an Intel NIC on it.

      Finally, while AMD's on-board graphics are some of the best you can get (definitely better than anything Intel currently makes), they don't support dual digital output. I have two monitors, and prefer to run them both digitally (i.e. HDMI or DVI). All of AMD's on-board GPUs allow only one digital device to be connected at a time. You can do two monitors with one digital and one analog, but I can see a quality difference; analog sucks.

      Overall, though, I think AMD still offers a lot of value. The issues I pointed out will go unnoticed by the casual user. And the power user should understand his needs well enough to know if they make a difference. E.g., I know the Realtek NIC is inferior to the Intel, but is it going to make a difference in day-to-day usage? For the average desktop? Probably not. Another nice feature AMD provides is ECC memory support for all non-Sempron CPUs. With Intel, you have to shell out the big bugs for a Xeon-branded CPU to get ECC support. The unfortunate thing, though, is that most consumer motherboards don't actually provide the traces that allow you to actually use the ECC support! The Biostar A760-G M2+ is a noteable example that actually does let you use the ECC features of the CPU (though it's unofficially supported).

    17. Re:I like AMD by Anonymous Coward · · Score: 0

      SLI was originally marketed by 3dfx with the voodoo3 (maybe voodoo2) line of graphics cards if i remember correctly.... They have since gone bust and the ip was bought by NVidia if i recall

  2. cache difference by phantomfive · · Score: 0, Offtopic

    The difference in cache size is real, the Intel i7-970 has 13.5MB of cache (L2 + 'smart'), whereas the AMD 1075T has 9MB (L2+L3) cache. I don't understand 'smart' cache well enough to say what that means for performance, but an extra 4.5MB of cache on the chip will surely push the price up.

    --
    Qxe4
    1. Re:cache difference by Ecuador · · Score: 4, Interesting

      The cache difference would not explain the price (or the transistor difference, 1.1bil vs 0.9bil), since we are talking about 3x+ the price. It is just that Intel enjoys a speed advantage so AMD has to pit its hexa-cores against Intel's quad-cores. And because, as it has always been, Intel is the more "recognized" brand, AMD makes sure that it gives you more performance for the price.
      It has been the same deal since my first ever PC: I could get, for about the same price, either an intel 486@66 or an AMD 486@100. My next was an AMD-K6 @ 233 which cost as much as the Pentium MMX 200 (yeah, the K6 lagged behind a PII, but it was no match for the Pentium MMX). Then I went with some Athlons, you remember how those did vs P3 at first, and then, even easier against P4. I am not a fanboy, but on a budget so I did get a Core 2 E8400 at some point because that was the only time I was buying a PC and AMD did not have a performance advantage at my desired price point. Now I am mainly on a Phenom II X4.
      But I digress, the point is that the Intel CPU's have traditionally been priced based on how much they can go for, not how much they cost. So right now they can get away with things like $1000 CPU's. If it wasn't for AMD, it would be like the 90's where they had mainstream cpu's at $1000, not just high end ones.

      --
      Violence is the last refuge of the incompetent. Polar Scope Align for iOS
    2. Re:cache difference by Eskarel · · Score: 2, Interesting

      AMD might try to give you more performance for the price now, and when they started they certain did, but remember that AMD are in the boat they are now largely because they used the advantage they gained from Intel's Itanium blunder to sell $400 mid range chips. Intel won their market back because AMD got greedy and Intel under cut them by about 50% with faster chips.

      AMD have no high end, with no high end they cannot survive because today's high end is tomorrow's mid range. You need to be tooling up that process 6 - 12 months in advance to compete. As much as I love AMD(I bought AMD for years, until my most recent PC), they're done.

    3. Re:cache difference by Corporate+Troll · · Score: 2, Insightful

      AMD have no high end, with no high end they cannot survive because today's high end is tomorrow's mid range.

      I would agree, 5 to 10 years ago. Alas, I don't anymore. We are at a performance plateau, where the user (normal, we're not talking special-case) can be perfectly happy with 5 year old machines (I'm a dumpster diver, good P-IV or AMD XP machines can be found there). Any machine in the 2.0GHz range (give or take) will cover the needs of users.

      CPU makers are at the point where people who need more CPU power will have to be willing to pay for it. All the rest can go with whatever is cheapest. Intel knows this, hence the Atom. I built an Atom desktop based on the D410PT motherboard for my mother in law running Ubuntu 10.04. At no point performance has been a problem.

      Tomorrows "desktop" CPUs won't be the "top-of-the-line" of today. They will be the scaled-down, power-efficient CPUs that won't deliver as much power, but enough for the end-user. All other will have to pay premium to get more power.

      Unless we suddenly get a big craving for extra CPU power, that's how it's going to go.

    4. Re:cache difference by beelsebob · · Score: 1

      No, the difference in performance is all it takes to explain it. The 1075T is roughly as fast as a Core i5 760, which is about $40 cheeper than it.

    5. Re:cache difference by cyberjessy · · Score: 1

      AMD is not done, in fact quite the opposite.

      2011 will have some of AMDs most anticipated releases ever, Bulldozer and Bobcat. Also decent graphics are no longer a luxury, it is a necessity even in phones and tablets. Intel makes terrible graphics cards, while AMD makes the best performing ones available today. 2011 will see more widespread adoption of integrated CPU+GPU solutions (from both Intel and AMD), and guess who will hold the advantage there.

      --
      Life is just a conviction.
    6. Re:cache difference by tyrione · · Score: 4, Interesting

      AMD might try to give you more performance for the price now, and when they started they certain did, but remember that AMD are in the boat they are now largely because they used the advantage they gained from Intel's Itanium blunder to sell $400 mid range chips. Intel won their market back because AMD got greedy and Intel under cut them by about 50% with faster chips.

      AMD have no high end, with no high end they cannot survive because today's high end is tomorrow's mid range. You need to be tooling up that process 6 - 12 months in advance to compete. As much as I love AMD(I bought AMD for years, until my most recent PC), they're done.

      Not even close. Bulldozer architecture, merged with their rock solid GPGPU structure in OpenCL is a reality and a fundamental architecture design shift that Intel will work at copying.

    7. Re:cache difference by Anonymous Coward · · Score: 0

      The cache on the Intel chip is only there because the can.

      Intel's magor advantage over AMD in the past, but still to an extent today, has been cache size. The processes they use are so bullet proof that they can pack cache in and still not get defects.

      AMD on the other hand have to keep killing cache units due to defects, space them out more and even worse, build their chips on bigger processes for longer.

      So Intel pack extra cache on not because it gives a massive performance boost, but just because they can and AMD can't.

    8. Re:cache difference by Anonymous Coward · · Score: 1

      If it wasn't for AMD, it would be like the 90's where they had mainstream cpu's at $1000, not just high end ones.

      Would that be a bad thing? If regular people couldn't afford powerful CPU's, software makers would be force to optimize for less juice. Everything, from Flash and browser and Javascript, to games and office apps, would have to run well with less CPU power. Would that be all bad?

    9. Re:cache difference by barbazoo · · Score: 1

      Cache is king

    10. Re:cache difference by bfree · · Score: 2, Informative

      AMD have no high end

      Ever heard of the Opteron, particularly the 6100 series released in March? 12 cores from 1.9GHz to 2.2GHz with 115W TDP or you can go up to 137W TDP and 2.3GHz or drop to 1.7GHz and closer to 65W TDP.

      --

      Never underestimate the dark side of the Source

    11. Re:cache difference by rbarreira · · Score: 1

      No, the difference in performance is all it takes to explain it. The 1075T is roughly as fast as a Core i5 760, which is about $40 cheeper than it.

      And the i5 760 is just a little faster than the Phenom II X4 965 which is $40 cheaper than the i5... See what I did there?

      The problem is that most benchmarks do not take advantage of six cores, right now you'd only get a six core CPU if your applications are heavily multithreaded.

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    12. Re:cache difference by Pranadevil2k · · Score: 2, Interesting

      AMD's chips perform threaded tasks faster than Intel's - if you want to talk about tomorrow's chips you should also look at tomorrow's software, where heavy threading is going to be the norm. From that perspective, the value of the AMD chips easily doubles.

    13. Re:cache difference by CajunArson · · Score: 3, Interesting

      Only on Slashdot can Powerpoint slides of what AMD's chips will be doing in the second half of 2011 be called "rock solid" and get modded interesting in 2010.....

      1. Bulldozer will NOT have GPGPU until 2012 at the earliest unless you think AMD is lying.

      2. When it comes to GPGPU ATI is nowhere near NVIDIA. Oh don't get me wrong, when it comes to making a good graphics card that plays games (which is what 98+% of the market actually wants) ATI is definitely ahead of NVIDIA. When it comes to GPGPU that the HPC sector wants, NVIDIA is still way ahead not only on hardware but also on software. I know all about the hype around OpenCL, I also know people who do this stuff for a living and CUDA is simply better and while Fermi sucks for playing games, it shreds anything ATI has for GPGPU.

      3. Note the "98%" figure I gave above about what the market cares about. For all the hype on Slashdot, the number of applications that can actually take advantage of GPGPU is vanishingly small and inside of that small subset the biggest niche that exists is for video transcoding. Guess what? Using 3 square millimeters of silicon Intel's Sandy Bridge (that will be out at least half a year before Bulldozer) already does this. Also when it comes to normal Floating point performance, an equivalently clocked Sandy Bridge with 4 cores will have TWICE the AVX computing power of an EIGHT core Bulldozer... yes you heard me right 4 cores Intel vs. 8 cores AMD, due to AMD only including 1 full AVX unit in each "module" that contains 2 "cores" in Bulldozer. This is true unless AMD is intentionally lying about Bulldozer to make it sound worse than what the actual architecture will be. Oh.. and before you say that nobody will ever use AVX just remember that you called openCL "rock solid" a few minutes ago. GCC is already able to emit AVX instructions that existing code can be tweaked for RIGHT NOW while OpenCL is stil a pipe dream in many ways.

      --
      AntiFA: An abbreviation for Anti First Amendment.
    14. Re:cache difference by anUnhandledException · · Score: 1

      Double Edged sword.

      It would cut out a lot of the bloatware in everything from office apps to websites (javascript/flash) to antivirus and even OS.

      However somethings like games can only be optimized so far. I think performance/realism/immersion of games would be significantly scaled back.

      Also you seem to forget than when Intel ruled the market it wasn't just the high end that was expensive. The "low end" was $500+ CPU. So you get less performance AND higher prices. No way to say that is a winner for consumer.

    15. Re:cache difference by mlts · · Score: 1

      What I see happening is that we will be seeing more specialized functions put on the CPU silicon as opposed to extra CPU power primarily. Of course, CPUs will get more powerful, but it won't be Moore's law doubling. Instead, what we will see are more specialized tasks (AES encryption for example, perhaps RSA) given specialized hardware. We will see more caching. We definitely will see more cores. Perhaps we will see specialized cores, so a CPU would have six cores that are mainly designed to do integer stuff (but can do FP in a pinch, not optimally though), and two cores optimized for FP work, but can do integer performance. Then operating systems will get a process scheduler that can take advantage of this and put floating point intensive tasks on the FP-heavy cores, unless the FP heavy cores are completely used, then they go on the integer cores.

      I also see eventually a hypervisor being put in at the CPU level. This way, a VM sees that it has 12 cores available to it, when in reality, it is handed CPU allocation by a min/weight/max configuration, so the 12 cores might be 25% of a single physical core at minimum, and 100% of 4 cores at maximum.

    16. Re:cache difference by Nemyst · · Score: 1

      As someone who switched from a 8800 GTX to a HD 5850, I'm always struck by how little OpenCL is being used currently. I have seen and heard lots about CUDA, but not about OpenCL. Also, AMD/ATI's graphics drivers really suck in comparison to Nvidia's ones.

      AMD will have to do a lot of work to manage to gain the upper hand again. People don't care that they have an Intel CPU and Nvidia GPU (integrated or not) or AMD/ATI, as long as it works.

    17. Re:cache difference by mlts · · Score: 1

      Knowing most software makers, they would just have their programs refuse to run if they are not installed on the latest and greatest CPU if push came to shove and they couldn't continue to code on the cheap. Or they would just have their stuff run poorly and users would get used to clicking on a button and waiting 30 seconds for a result, just like the early days of X-Windows on some machines. Marketing people will call it accustoming the user to the new software experience where one types, then 20-30 seconds later the text will show up on the screen.

      It makes me wonder what life would be like if a core enterprise UNIX (BSD, Linux, Solaris or AIX), became the mainstream OS of choice for home PCs. People would never tell a computer novice to reboot unless it was a hardware or kernel upgrade, and a reinstall just wouldn't be done unless the box was hacked. Probably one of the most common problems would be replacing the battery powered chips because people set the boot PROM password then forgot it. Viruses wouldn't be an issue, although there would still be break-ins and compromises due to unpatched machines, or Joe Sixpack running a Trojan in effort to view the dancing bunnies.

    18. Re:cache difference by Eskarel · · Score: 1

      Opteron is not and has never been a desktop chip, and it's not high end in comparison to the offerings Intel has in the same space.

      AMD's foray into being a major player was almost entirely due to Intel screwing up by making the Itanium a chip no on actually wanted. They failed to capitalize on that market share and they lost it.

      There's no evidence whatsoever that Intel is going to do something stupid like the Itanium again any time soon, or that AMD has a product on its road map to do what they've never managed before, which is take serious market share from Intel in a race Intel is actually playing in.

    19. Re:cache difference by Eskarel · · Score: 1

      And nobody cares.

      Even if AMD came out with an amazing GPGPU architecture tomorrow and could sell it at a profit for $1 it really wouldn't matter.

      Programming for a GPGPU is an entirely different paradigm than programming for a standard CPU and GPU combination, it involves totally different ways of thinking, totally different ways of designing, and totally different tools. You're just not going to see programmers making that sort of leap until there's serious motivation for them to do so, which there really just isn't at the moment.

      We're only just starting to see applications take advantage of 64 bit and that's been around for the better part of a decade. The kind of programming models required to take advantage of a GPGPU are far stranger and more alien than converting from 32 to 64 bit.

      By the time there's any serious usage of GPGPU chips, AMD will have long run out of money.

    20. Re:cache difference by CajunArson · · Score: 2, Interesting

      AMD's chips perform threaded tasks faster than Intel's

      Please provide one shred of evidence that supports this claim.. and I don't mean a comparison of a 6 core AMD CPU to 2 core Intel chip from 2007.

      --
      AntiFA: An abbreviation for Anti First Amendment.
    21. Re:cache difference by Belial6 · · Score: 1

      AMD will likely never be done until X86 is. AMD keeps Intel out of monopoly position while never really giving it a run for it's money. If AMD starts to fall too far behind, Intel will just raise their prices until AMD catches up a bit, or they will stockpile improvements until they need them. Then if AMD starts to get too close to being real competition, Intel will use those resources to pull back ahead.

      Intel doesn't set the pace of x86 improvements, AMD does. In the end, that relationship works out for everyone. Intel keeps making insane amounts of money without too much in the way of anti-trust issues. AMD gets to exist. Consumers get improvements from both AMD and Intel at whatever pace, AMD can roll their improvements out.

    22. Re:cache difference by Eskarel · · Score: 2, Interesting

      I'd also like to see the evidence that heavy threading is going to become the norm any time soon in most applications. It's had the better part of a decade to get there and still substantially less than half of all software supports multi-threading and most of that is lucky to support 2 cores.

      People don't seem to realize how much more difficult writing properly multi-threaded software is, or that not all software lends itself to multi-threading.

    23. Re:cache difference by cheesybagel · · Score: 1

      Bulldozer has half the floating point units of Sandy Bridge, but the Bulldozer floating point units have fused multiply add. This means the peak floating point performance of Bulldozer and Sandy Bridge is the same.

    24. Re:cache difference by cheesybagel · · Score: 1

      Intel is not going to do something like Itanium? What do you think Larabee was?

    25. Re:cache difference by cheesybagel · · Score: 1

      PS: The difference between Intel and AMD is that Intel has pockets so deep, they can throw multiple teams at the problem. Even if some of the products fail (i860, Itanium, Timna, Tejas, Larabee) they still manage to continue doing business just fine (Pentium, Xeon, Pentium III mobile, Intel Core, GMA graphics). AMD only needs to make one mistake and they are screwed.

    26. Re:cache difference by cheesybagel · · Score: 1

      The software which does not support multi-threading is usually not performance sensitive anyway. If the software is performance sensitive, the developers will spend time optimizing for it.

    27. Re:cache difference by Eskarel · · Score: 1

      I don't think you really understand exactly what it takes to code for a GPGPU, it's not just about optimizing it, it's about totally different ways of writing code.

      Even pure multi-threaded isn't all that easy, it's one thing to split a task into two parts, it's totally another to split it into 4 or 6 or 8 or whatever arbitrary number of parallel execution tasks. It's a bit like the old story about using 9 women to get a baby in 1 month, some processes just don't work that way.

  3. AMD One-Ups Intel? Another misleading Slash story. by Futurepower(R) · · Score: 4, Informative

    The PassMark Intel vs AMD CPU Benchmarks - High End show the AMD Phenom II X6 1075T as being nothing unusual in speed or price.

  4. $885 per 1,000 units? by Anonymous Coward · · Score: 0

    Shouldn't that read "$885/unit when buying 1,000 units"?

    1. Re:$885 per 1,000 units? by Anonymous Coward · · Score: 0

      No, it's really that cheap.
      --
      Tavis Ormandy, asshole extraordinare.

    2. Re:$885 per 1,000 units? by Anonymous Coward · · Score: 0

      When posting AC, don't include your sig ;)

  5. Pricing error in this article by jcrawfordor · · Score: 4, Informative

    There's an important data error in the pricing information in this article. The bulk price quoted by Intel ARK and the AMD catalog is the price per unit for 1000 units, not the total price for 1000 units. Otherwise, Intel's high-end six core processors would have retail prices of $10!

  6. Their ULV processors are pretty impressive, too by mykos · · Score: 4, Interesting
    1. Re:Their ULV processors are pretty impressive, too by ray-solomon · · Score: 0

      What games? FarmVille?

    2. Re:Their ULV processors are pretty impressive, too by Anonymous Coward · · Score: 1, Insightful

      You know, if you clicked the link....

    3. Re:Their ULV processors are pretty impressive, too by FranTaylor · · Score: 0

      de gustbus non est disputandem

      we are talking about how well it runs, not whether you like it or not

    4. Re:Their ULV processors are pretty impressive, too by Jedi+Alec · · Score: 1

      I was just about to yell at you about your irrational hatred of Farmville when you managed to turn it into praise for your (most likely irrationalas well) love for EVE.

      So, about that POS...any shiny labs in there? ;-)

      --

      People replying to my sig annoy me. That's why I change it all the time.
    5. Re:Their ULV processors are pretty impressive, too by Gruturo · · Score: 1

      de gustbus non est disputandem

      Please..... if you choose to quote latin to sound important, you might want to consider taking the time and attention to avoid 2 errors in 5 words.

      --

      Vacuum cleaners suck. Kings rule.
    6. Re:Their ULV processors are pretty impressive, too by Arimus · · Score: 1

      Nah, just the usual pile of shooty things and a hanger - purely a ratting/staging tower.

      --
      --- Users are like bacteria -> Each one causing a thousand tiny crises until the host finally gives up and dies.
  7. nothing new by ILuvRamen · · Score: 5, Interesting

    This is really nothing new. Everyone can say AMD is worse than intel all day until you actually look at the prices. I've put together computer quotes for people and I can't even put in a wolfdate core2 for remoately close to a 3.0GHz AM3 Regor which is around $62! And for an i3 board and processor together, it's over double an AMD board and processor even with a Phenom in it instead. I mean if you want something so fast that AMD doesn't even make it, only Intel does, go for it otherwise there's a darn good reason why AMD has been "losing" and isn't out of business yet. Their chips are better speed for the price in most cases!

    --
    Google's Super Secret Search Algorithm: SELECT @search_results FROM internet WHERE @search_results = 'good'
    1. Re:nothing new by 0123456 · · Score: 3, Interesting

      I mean if you want something so fast that AMD doesn't even make it, only Intel does, go for it otherwise there's a darn good reason why AMD has been "losing" and isn't out of business yet. Their chips are better speed for the price in most cases!

      Haven't AMD's recent profits come from a) ATI and b) Intel?

      Their chips are lower priced for the same speed because that's the only way they can sell them. If AMD could make faster CPUs than Intel's, they'd be charging $1000+ as well.

      Obviously that's good for us, because you can get a decent AMD system for less than Intel at the same performance level, but their low prices certainly aren't keeping them in business... I'm pretty sure that everyone at AMD wishes they could be selling their chips for twice as much.

    2. Re:nothing new by zippthorne · · Score: 2, Interesting

      AMD's lower prices are because of Intel's brand, not because of actual performance. You said it yourself: lower prices at comparable performance levels. The same holds true in the high end.

      Further, because of the market share of Intel, other the software giants don't do very much in the way of optimizing code to run on AMD, so they're always going to be compared on the subset of chip features that Intel also supports.

      Multiple equal giants would be a better situation for the rest of us, because then they'd be able to compete on architecture instead of just implementation. And yes, I'm aware of the Itanium fiasco, which i'd bet was driven by intel's x86 market inertia as much as anything: even Intel can't compete with Intel...

      --
      Can you be Even More Awesome?!
    3. Re:nothing new by mike260 · · Score: 1

      Google's Super Secret Search Algorithm: SELECT @search_results FROM internet WHERE @search_results = 'good'

      search_results
      +----------------+
      good

      1 row in set (0.00 sec)

    4. Re:nothing new by brucmack · · Score: 1

      Have you factored in power consumption? AMD's chips are still built on a larger process, so they use significantly more power than Intel processors at a given performance level. Depending on the purchase price, cost of electricity in your area, and how much the computer is used, it may be cheaper to invest in a bit more expensive processor to save on the electricity bill.

    5. Re:nothing new by Anonymous Coward · · Score: 1, Insightful

      The same holds true in the high end.

      Except AMD doesn't compete with Intel's high end. They simply don't offer product in the same category as Intel's best performing (and most expensive) processors.

    6. Re:nothing new by Methuseus · · Score: 0, Troll

      how does a 65W TDP part use more power than a 115W TDP part? I know those are absolute top end power values, but this still makes no sense.

      --
      Two things are infinite: the universe and human stupidity, though I'm not yet sure about the universe. - A Einstein
    7. Re:nothing new by Nemyst · · Score: 1

      I'm sorry, but AMD has no high end. They compete very well in low and mid ranges, but high end is just Intel.

    8. Re:nothing new by Chris+Burke · · Score: 1

      Haven't AMD's recent profits come from a) ATI and b) Intel?

      No, ATI is a small portion of AMD's revenue, and yes, without the Intel settlement AMD was in the red.

      Of course the Intel settlement is but a tiny fraction of the money AMD lost over the years due to Intel pressuring the OEMs to not sell AMD no matter how attractive they were. While this was a serious dampener for AMD since the K6 days, it's particularly ludicrous that they were unable to get more marketshare in the K8 days when they were far ahead in perf/watt, perf/dollar, and perf/watt. There was a brief period where they were fab limited, but in fact a big factor in their current financial woes and the reason they had to spin off their fabs is because they spent all this money on a building a new fab, but the marketshare didn't come so the fab was mostly empty.

      Obviously that's good for us, because you can get a decent AMD system for less than Intel at the same performance level, but their low prices certainly aren't keeping them in business... I'm pretty sure that everyone at AMD wishes they could be selling their chips for twice as much.

      I'm sure they would. To enable that, they need to get back in the game on a perf/core level. Or perf/core needs to stop mattering, but frankly even with the increase in multithreading thanks to multi-core mania "sure the cores are slower but you make it up with thread count!" is unlikely to be a generally true statement any time soon, imo. This is on the desktop, obviously. Anyway, part of why their prices are so low is because their top of the line is priced to look good compared to Intel's 2nd or 3rd stringers.

      But really, their current margins are keeping them in business. A company AMD's size can last a long time losing "only" tens of millions of dollars a quarter. When it was hundreds of millions, that was rather troubling. :)

      --

      The enemies of Democracy are
    9. Re:nothing new by Bert64 · · Score: 0, Troll

      And neither Intel nor AMD compete with the high end chips offered by IBM.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    10. Re:nothing new by jon3k · · Score: 1

      Typically for every AMD CPU there is a comparable Intel chip thats either slightly faster and slightly more expensive or slightly slower and slightly less expensive. The one place AMD really is doing well seems to be at the absolute bottom end of the market. If you need a CPU that is literally as inexpensive as you can get, them AMD is worth considering.

    11. Re:nothing new by rbarreira · · Score: 1

      Intel only recently started using 32nm, most of their CPUs are still 45nm.

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    12. Re:nothing new by afidel · · Score: 1

      AMD does make $1,000+ chips, just not for the desktop market because that's a few thousand units a month. The Opteron 6174 is a very capable chip depending on the workload. Unfortunately for AMD it came out more than a year after the Nehalem based Xeon's so we standardized on those for our current generation of servers. Previously all our CPU demanding processing was done on AMD (DL585 series boxes).

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    13. Re:nothing new by Anonymous Coward · · Score: 1, Insightful

      Nonsense. x86 has eaten the other boutique high end CPUs and it will eat Power too. IBM gets killed in price/performance by x86 chips, especially the high end Nehalems. And raw performance isn't _that_ far off, either.

    14. Re:nothing new by h4rr4r · · Score: 1

      What do you call those 12 core opterons?

    15. Re:nothing new by Decollete · · Score: 1

      I call them not-for-me processors. When looking at price lists or benchmarks, I disregard Opteron and Xeon completely.

    16. Re:nothing new by brucmack · · Score: 1

      I wouldn't say they only recently started using 32nm - they released the first 32nm processors at the start of the year, and 8 months is still a long time in the processor world. The entire Core i3 series is 32nm, along with all of the dual-core i5s and the hex-core i7s. That basically only leaves the quad-core i5s and i7s, and a few value Pentium processors that aren't really relevant to this discussion anyway.

      Even ignoring the process advantage, Intel also has a large IPC advantage, so their processors are going to do more work per watt. The only way AMD has remained competitive is by offering more cores than Intel at the same price point, which can only result in higher power consumption. This is also why you can't just compare TDP values to compare power consumption - for the same amount of work, the Intel processor is going to be idle longer than the AMD processor, and use less power as a result.

    17. Re:nothing new by brucmack · · Score: 1

      I highly doubt anyone is comparing parts with such a large difference in TDP. When shopping for a processor, the first thing you determine is what performance level you need, and then you look at what you can get to meet that performance level. Do this with AMD and Intel processors will not give such a large TDP gap.

      But even with that said, it is perfectly possible for a 115W TDP part to use less power than a 65W TDP part, for a given task. As an extreme example, take a 65W part from five years ago and a 115W part from today, and set them to perform the same video encoding job. The likely result is that the modern 115W part will be done in a fraction of the time, thereby spending most of its time in an idle state and using much less power than the 65W part. The same situation does occur when comparing modern processors - Intel has a large IPC advantage, so their processors will do more work per watt, all else being equal.

    18. Re:nothing new by h4rr4r · · Score: 1

      Foolish, they are often cheaper than the top of the line desktop chips.

    19. Re:nothing new by Decollete · · Score: 1

      Not talking about their price specifically, but does anyone actually recommend Opteron/Xeons for their grandmother?

    20. Re:nothing new by zippthorne · · Score: 1

      Well, yeah, for a gaming machine. But both have some very important features that you won't want to be without in a server setting.

      I believe an automotive analogy is customary: Both a semi and a high-end sports car can have 500 hp engines. Even their prices will seem comparable. But for the purposes for which you'd get a semi, the sports car would be highly inappropriate.

      --
      Can you be Even More Awesome?!
    21. Re:nothing new by cheesybagel · · Score: 1

      Power7 is good at what it was designed to do. Top performance for server tasks. If we are talking about consumer electronics, IBM manufactures lower power (hah) processors which are used in games consoles (PS3, XBox 360, Wii) and the like. The market for PCs keeps shrinking as people rely increasingly on smaller web connected mobile devices.

    22. Re:nothing new by Anonymous Coward · · Score: 0

      Phenom IIs turn off unused cores when idle or even if the load is not very high.

  8. Budget by cheekyjohnson · · Score: 3, Interesting

    I like AMD because their processors are usually fast enough for me and are usually much cheaper than the processors that Intel sells. I really can't afford to pay nearly as much for the processor as I do every other part for the computer combined, so I go with AMD.

    --
    Filthy, filthy copyrapists!
  9. Re:That would be all nice and dandy if only... by tacarat · · Score: 4, Interesting

    When selling to a non-tech person, though, such things make little difference. Most aren't savvy enough to know the difference and mostly look at the number of cores and speeds as final arbiters on performance. It's like explaining that while a motorcycle engine may have higher RPMs, a truck has more torque and can move big loads better. Hell, that's the simplest analogy I know for modern chip comparisons and it still goes over some people's heads.

    Then, of course, is the SUV-that-never-goes-offroad-computing crowd that throw down big bucks so they can have 3D accelerated, multicore/non-multithreaded MS Spider Solitaire. God bless them.

    --
    "Common sense will be the death of us all"
  10. The thing I think you miss by Sycraft-fu · · Score: 1, Interesting

    Is that AMD's 6 core chip only competes against Intel's 4 core ship speed wise, and then only for apps that can use all 6 cores.

    Currently Intel just has an untouchable performance advantage. Now, we'll have to see what the future holds. Both companies have new architectures coming soon. Intel's Sandy Bridge is in final production and slated to launch beginning of 2011, though only in the mainstream market at that time (the high end will come later in the year). AMD has a new architecture called Bulldozer that is supposed to come out in 2011 though they haven't been more specific as to when.

    However as of right now, the Core i series kills anything AMD has. Their 6 core CPUs can only keep up with Intel's quads, and then not even the high end. They have nothing that touches Intel's 6 core line.

    As such their prices are lower for any given part. They are a more budget solution, not a performance one.

    1. Re:The thing I think you miss by chichilalescu · · Score: 1

      you seem to know these things. I have two types of problems I'm interested in: pseudo-spectral simulations and point particle trajectory integration. for the first, I need to perform many FFTs; for the second, I have many small single-process jobs (can only assign individual jobs to individual cores).
      would I be better off with the 6 core AMD, or the 4 core Intel?

      --
      new sig
    2. Re:The thing I think you miss by Anonymous Coward · · Score: 0

      You'll probably want to go with i7 1366 board. I'm also doing Physics simulations using MPI, and the higher memory throughput helps a lot. i7 also has hyperthreading, which further helps if you can fully utilise. Intel chips do more per clock cycle, so you're not really losing on the non threaded applications. If you're using a 1 socket board, go with the W series, which can easily overclock to 4.0ghz and stay cool, they cost on par with i7 930. If you're going for 2 socket, then still Intel... If you go for a 4 socket with an extension board for another 4 sockets, then go with AMD 6+ cores. In the 4 socket setup its worth going AMD for the price advantage.

    3. Re:The thing I think you miss by seifried · · Score: 2, Interesting

      They have nothing that touches Intel's 6 core line.

      I'm pretty sure the AMD 12 core CPUs will "touch" Intel's 6 core CPUs quite nicely (locally they are both about $1200 retail from the store).

    4. Re:The thing I think you miss by obarthelemy · · Score: 2, Informative

      "performance" by itself doesn't mean anything. You have to refer to "performance per dollar" or "performance per watt" or ... that's why you don't see everyone buying the top of the line CPU... There's a bunch of stuff that offers the highest performance, and that doesn't really sell much, if at all: fighter jets, formula 1 cars, thoroughbreds... We might feel all sexy at the idea of owning one of those, but the bare fact is, we can't afford it, and, when push comes to shove, we would be stupid to, anyway.

      talking about perf/price, AMD is not that bad, especially for run-of-the-mill levels of performance, and once you take the MB cost into account (why are Intel's MB so much more expensive ?). On a fixed budget, I'd rather scrimp on the CPU/MB, and put more into the GPU and disks, because that's what limiting my PC, right now.

      talking about perf/watt... I'm not running a server farm, I don't care. Intal is ahead, though.

      --
      The Cloud - because you don't care if your apps and data are up in the air.
    5. Re:The thing I think you miss by dbIII · · Score: 0

      and then only for apps that can use all 6 cores

      Or multiple applications like you will see in your process manager all of the time. Also you can also bet that developers are all being dragged into the 1990s and they'll eventually start using those extra processors. This is beginning to look like a post I made from a dual Celeron 300 when Slashdot was only a year or two old, but most of the time there is something another processor can do which lets the main task finish more quickly.

    6. Re:The thing I think you miss by Jah-Wren+Ryel · · Score: 1

      Sounds like you might have a job for a GPU. GPUs absolutely rock at FFTs and if your single-process jobs are simple enough you may be able to run hundreds of them simultaneously on each GPU. Current generation GPUs support double-precision floating point if you need it and also ECC memory to guard against random bit errors corrupting your calculations. ECC is definitely the way to go if you need reliable results, whether it be GPU or CPU.

      --
      When information is power, privacy is freedom.
    7. Re:The thing I think you miss by FranTaylor · · Score: 3, Interesting

      Price/performance is not the criterion here. There are applications where this is important, but an average desktop user is not one of them.

      These systems are all quite bloody fast enough for "normal" desktops. The question is :how much does it cost? and AMD will get you a much better price for this class of machine.

    8. Re:The thing I think you miss by Boycott+BMG · · Score: 1

      If you want to do those things I suggest an Nvidia Fermi GPGPU. Of course, in order to use them you will need to know how to take advantage of them. Fortunately, Nvidia has a written CUDA plugins and other ways to use them in, for instance Matlab, so you don't have to write everything yourself.

    9. Re:The thing I think you miss by Anonymous Coward · · Score: 0

      Sounds like you might have a job for a GPU. GPUs absolutely rock at FFTs and if your single-process jobs are simple enough you may be able to run hundreds of them simultaneously on each GPU.

      Now if only my regular compiler would allow me to take advantage of the GPU.

    10. Re:The thing I think you miss by Pentium100 · · Score: 1

      Yes, the best AMD CPU is slower than the best Intel CPU, and probably slower than the second best Intel CPU too. However, AMD usually has better price/performance ratio, so if I do not want the fastest CPU there is, I should probably buy an AMD product.

      Same thing was when I was looking for a video card ~3 years ago. The new ATI HD2900XT was slower than the best nVidia card, but the nVidia card was much more expensive. HD2900XT was faster than the nVidia card that had a similar price.

    11. Re:The thing I think you miss by Anonymous Coward · · Score: 0

      For single-threaded tasks, Intel's Core i series is still the winner, but for multi-threaded apps (and multiple applications) I'll stick with my my dual 12-core Opteron system (the one that will hopefully turn up at my door in a couple of weeks).

      I'm looking forward to putting that mother through its paces.

    12. Re:The thing I think you miss by Anonymous Coward · · Score: 0

      However, AMD usually has better price/performance ratio, so if I do not want the fastest CPU there is, I should probably buy an AMD product.

      Yes, they do in a certain price segment. With the chips discussed in the article they don't. A Core i7 860 (2.8 GHz, 4 cores) is in the same price segment as the Phenom II X6 1075T (3.0 GHz, 6 cores). The Intel chip is at worst about the same performance as AMDs offering and at best has pretty big performance advantage. That's for both single- and multi-threaded tasks.

      http://www.anandtech.com/show/3937/amds-fall-refresh-new-phenom-ii-and-athlon-ii-cpus-balance-price-and-performance

    13. Re:The thing I think you miss by tepples · · Score: 1

      Or multiple applications like you will see in your process manager all of the time.

      Most of which are blocking on something. Even a dual core can run your foreground app on one core and all the little background apps using 0-1% CPU on the other.

      Also you can also bet that developers are all being dragged into the 1990s and they'll eventually start using those extra processors.

      The 1990s? When exactly did the home edition of Windows start supporting more than one core?

    14. Re:The thing I think you miss by anUnhandledException · · Score: 1

      For $100 more.

      A CPU needs a matching motherboard. An i7 motherboard is generally $60 to $100 more than a comparable AM3 motherboard. So i7 CPU is $20 more and comparable motherboard is $80 (average) more. So for $100 (30% price premium) you get equal or maybe slightly more performance.

      The macro-architecture contributes to i7 performance (triple channel ram, high speed link between CPU and chipset, large number of PCIe lanes) however those improvements also come at a cost in dollar terms. You can't look at the performance without looking at the associated cost.

    15. Re:The thing I think you miss by dbIII · · Score: 1

      There are were platforms including other Microsoft ones. Also even a cheap bastard like me had a dual CPU setup at home before 2000, let alone stuff like the Sparcstation 10 from 1992 (still got one here to fire up and run some legacy software every year or two).

    16. Re:The thing I think you miss by sjames · · Score: 1

      Especially considering that AMD no longer jacks the price to the sky for CPUs that can go in a 4 socket system.

    17. Re:The thing I think you miss by tepples · · Score: 1

      There are were platforms including other Microsoft ones.

      How much commercial off-the-shelf software intended for home or home office users was made to run on these?

  11. Technology by Jethro · · Score: 1

    Looks like we're still adding cores and cache to the CPUs, but we're not really coming up with anything really revolutionary. Last really interesting idea was Transmeta's ill-fated effort. Come on, people, innovate!

    That said, I think it might be time for me to upgrade my desktop. It's still got an AM2 CPU!

    --


    In the land of the blind, the one-eyed man is kinky.
    1. Re:Technology by Vectormatic · · Score: 2, Informative

      next year AMD will be launching its "bulldozer" architecture, which from what i've seen takes a rather novel approach to cores/hyperthreading (two 'cores' which share some execution units. I'm not saying bulldozer will suddenly revolutionize anything, but it is an interesting take on multi-core

      as for your am2 cpu, yeah man, if your board takes a am3 cpu, go for it, you can pick up a quad core (which will trounce whatever you have in single threaded performance too), for under $100, if you still have a single core am2 chip, put in a $60-70 X3 and be amazed!

      --
      People, what a bunch of bastards
    2. Re:Technology by Jethro · · Score: 1

      Oh I've got a quad-core - when I built that desktop AMD had just released the AM3 CPUs which made the Phenom II X4 940 a LOT cheaper. And frankly I have zero performance issues with this machine, so I really have no plans on upgrading (;

      As for the Bulldozer core, I mean, seriously the last REAL leap on x86 was, what, when Pentium went from pure CISC to a RISC/CISC hybrid kinda thing? We're still running the same basic architecture. Sure, we're 64-bit rather than 32, but you know.

      Not saying the new CPUs aren't cool, but I just want to see what the next leap will be already! I didn't spend most my life reading science fiction to NOT have light-based computers yet!

      --


      In the land of the blind, the one-eyed man is kinky.
    3. Re:Technology by Nursie · · Score: 1

      Well, IBM/Sony and others tried the Cell architecture thing, couple a few general purpose cores with vector processing units. It seemed to make a bit of an impact on the supercomputer scene for a while, though apparently games programmers don't much care for it.

      I suspect that's because they have to target multiple architectures though, and Cell demands it's own attention.

    4. Re:Technology by Vectormatic · · Score: 2, Interesting

      ah right, when you said AM2 chip i assumed non-am2+, rulling out any phenom I/II based thing. enjoy that 940, it is a killer chip (my gf has one, i get by with my x2 7750)

      and yeah, bulldozer wont be a leap like pentium>pentium pro, but it might take multi-core in an interesting direction

      --
      People, what a bunch of bastards
    5. Re:Technology by Jethro · · Score: 1

      Yeah, whatever the Next Big Thing is, it'll have to emulate x86/x86_64 pretty darn well...

      --


      In the land of the blind, the one-eyed man is kinky.
    6. Re:Technology by Jesus_666 · · Score: 1

      Wasn't the last real leap on x86 when AMD provided an upgrade path to AMD64 aka EM64T aka x86_64?

      --
      USE HOT GRITS WITH STATUE OF NATALIE PORTMAN (NAKED AND PETRIFIED)
    7. Re:Technology by mr_mischief · · Score: 1

      I think the L3 cache on the chip, multicore chips, and the on-die memory controller are good candidates for significant past upgrades to the architecture. Also, don't forget SSE and SSE2. One way to get a huge speedup from many apps is switching from 387/487 coprocessor logic to SSE2 math.

    8. Re:Technology by corychristison · · Score: 1

      I'm running an AMD X2 5200+ w/ 4GB DDR2 RAM (OCZ Platinum I think)

      Perhaps its time to start looking at an upgrade...

    9. Re:Technology by cynyr · · Score: 1

      hehe i have a AMD-athlon X2 2300BE (before BE was black edition and was the marker for low power) in an AM2 uatx motherboard. Its time for an upgarde, but i'd like to move to mini-itx, and there is only 1 am3(95w max) mini-itx board with a pci-e slot in it and it used so-dimms. AMD needs to make one themselves, AM3(125w), regualr ddr3 slots, PCI-E 2.1.

      IT's a shame because there are a few decent h55 based ones for intel.

      p.s. i should look at OCing my 2300+ X2

      --
      All of the above was encrypted with a Quad ROT-13 method. Unauthorized decryption is in violation of the DMCA.
    10. Re:Technology by zwarte+piet · · Score: 1

      Budget dictates I'll stick to the original socket A for now. But 1200Mhz is still a lot faster than my first pc.

    11. Re:Technology by zwarte+piet · · Score: 1

      Nah, unless you're gamer or a video artist.

    12. Re:Technology by Vectormatic · · Score: 1

      yeah, i am seriously drooling over the thought of a mini-itx amd system as well, i have in my head:

      mini-itx, quad core (can be a low power chip for all i care, just want 4 threads), 4gb of ram (preferably non-sodimm), a SSD for bootdisk, a 500gb laptop hdd for local storage (rest goes on the fileserver anyway), all of that stuffed in a case half the size of a shuttle barebone

      too bad there arent any decent AM2+ boards for sale (i could reuse my 4gb of ddr2), well zotac has a dirt-cheap am2 board which would work, but is only sold in the states...

      --
      People, what a bunch of bastards
    13. Re:Technology by Jethro · · Score: 1

      Nah, that's just more bits. It's still x86. Again, not complaining, just saying there wasn't a huge leap anytime recently.

      --


      In the land of the blind, the one-eyed man is kinky.
  12. Re:That would be all nice and dandy if only... by joss · · Score: 5, Funny

    Motorcycles, trucks... hmm, your analogy is nearly there, but there's something missing, I can't quite put my finger on it..

    --
    http://rareformnewmedia.com/
  13. Re:AMD One-Ups Intel? Another misleading Slash sto by Lueseiseki · · Score: 1, Interesting

    Yeah, because you should take aggregate data from three whole samples seriously.

  14. Legacy Implementation by Anonymous Coward · · Score: 2, Insightful

    Kinda sad really, only reason I buy AMD is because the AMD motherboards still support more legacy features than Intel boards but still support competitive modern processors (4x PCI slots for legacy video capture equipment but fast processor for encoding).

    1. Re:Legacy Implementation by petermgreen · · Score: 2, Informative

      4x PCI slots for legacy video capture equipment but fast processor for encoding
      hmm, when I go on newegg the most PCI slots they sell on an AMD board or a current gen intel board is 3 while they have LGA775 boards up to 5 PCI slots.

      --
      note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
    2. Re:Legacy Implementation by NJRoadfan · · Score: 3, Informative

      They also tend to still have serial and parallel ports as well. No X58 board has them and very few P55 boards still have them. Despite what people here might say, some of us folks still use these ports. USB adapters don't work very well with most bi-directional parallel port devices and USB serial adapters have issues with timing sensitive devices.

    3. Re:Legacy Implementation by PingSpike · · Score: 1

      Considering the cost of intel motherboards, one would think they'd at least be able to give you a header on the board to install these things.

    4. Re:Legacy Implementation by NJRoadfan · · Score: 1

      Gigabyte is one of the few companies that still equips their P55 boards with parallel and serial headers. Ironically Intel still sells a desktop board with a parallel and serial ports right on the backplate: http://www.intel.com/products/desktop/motherboards/db-DQ57TML/DQ57TML-overview.htm

  15. Re:That would be all nice and dandy if only... by X0563511 · · Score: 4, Insightful

    Oh, so you've already got a hand on and benchmarked these new chips?

    No, you haven't. We'll have to see.

    --
    For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
  16. Re:And 3 hours after reading this... by haruchai · · Score: 5, Interesting

    I love AMD ( and buy them ) as they are good enough for what I do and have really been the ones driving x86 innovation for the last 10 years. They've made Intel a better Intel by forcing them to keep up and cutting cost. Things would be even better for the consumer if AMD were closer to Intel in fabrication prowess - Andy Grove's company isn't called
    Chipzilla for nothing

    --
    Pain is merely failure leaving the body
  17. Re:And 3 hours after reading this... by Anonymous Coward · · Score: 3, Insightful

    Intel could compete on price... but..

    Someone has to pay for intels retarded quasi-futuristic commercials playing on tv all day long.

    And it won't be me.

  18. Re:That would be all nice and dandy if only... by cbope · · Score: 4, Insightful

    Not completely true, it depends on the application. In highly threaded tasks, AMD's 6-core will handily beat that i7 running at 2.4GHz (and even the higher clocked models without HT). Just check the latest benchmarks at Anandtech or Tom's hardware. In apps that are not heavily threaded, yes, Intel may win. But more and more apps are becoming multi-threaded and this will only increase in the future. AMD's current 6-cores are more future-proof than Intel's current platform. Not to mention that Intel loves to switch sockets every fucking generation, while AMD is able to keep sockets the same across many generations while staying competitive.

    I use both AMD and Intel, so I am not terribly biased one way or the other, but AMD deserves a lot of credit for keeping the processor market competitive. Without AMD or another strong competitor, we would all be paying $1000 for our CPU's form Intel and we would still be stuck with Netburst.

  19. It's not just about tech-savvyness... But time by Anonymous Coward · · Score: 1, Insightful

    When selling to a non-tech person, though, such things make little difference. Most aren't savvy enough to know the difference and mostly look at the number of cores and speeds as final arbiters on performance.

    I'm a software engineer who has taken several courses on computer/processor architecture, etc... So I could look into the subject, read manufacturers' datasheets, google forum discussions and be able to distinguish what is brand evangelism and who actually seem to know what they're talking about and so on...

    But am I really going to go through all that trouble? No.

    When a friend asks me "Which one of these processors should I choose?", I'll look at the clockrates and the number of cores and make a suggestion based on how much multithreading I think he needs. If he is the type of a person who I should recommend to overclock (which does have its downsides, too) I might also do a quick google search about "[the model] + overclocking" to see if I see anything special.

    It takes a lot of effort to keep up with the latest series from each brand in each area of hardware (display adapaters, processors, etc.)... So unless you actually need the $800 processors, work with them or they're your geek-specialization... You are most likely not going to care enough. ("Okay, the $300 processor wasn't the most optimal one? Damn. Well, it's probably good enough for the next few years, anyways.")

  20. Re:That would be all nice and dandy if only... by LingNoi · · Score: 1

    How can you say they're slower when they've just announced them?! There is no possible way you have demoed these 6 core chips.

  21. Re:That would be all nice and dandy if only... by smallfries · · Score: 1

    Is it like he is standing in the garage, about to drive to work, when he realises that his keys are still in the hall?

    --
    Slashdot: where don knuth is an idiot because he cant grasp the awesome power of php
  22. Re:And 3 hours after reading this... by hairyfeet · · Score: 5, Insightful

    Well after it came out that Intel was paying off OEMs not to use AMD chips I switched all my builds for customers and myself to AMD after being a lifelong Intel+Nvidia man, and my customers and I couldn't be happier. The bang for the buck is just insane as is seen in TFA, their 95w quads give damned good performance without turning my apt into a space heater, and when paired with an ATI chipset you have a great platform at a great price.

    I currently use my 925 quad for video editing and audio creation, and even with multiple realtime Cubase amp sims it just purrs like a big kitten, the Radeon onboard was powerful enough I played SWAT 3 and Bioshock on it with decent framerates until my HD4650 arrived , and I've been selling AMD Neo based netbooks to those customers that were thinking of Atom. After getting their Neo and seeing how nicely it runs compared to an Atom all they do is rave, with the Radeon onboard making it a smooth multimedia portable.

    So please, if you care about having real competition in the market as I do, give AMD a try. We really don't want to go back to the bad old days, when Intel would charge insane money for even their shitty chips, and the new AMDs will do any job you throw at them quite well and quite affordable. And where else can you buy a dual kit for $250 a quad for $300 or a fully loaded monster 6 core for $580?

    --
    ACs don't waste your time replying, your posts are never seen by me.
  23. Initial review... by Freddybear · · Score: 5, Informative

    Overclockers.com has a review of the Phenom II x6 1075T processor. Looks like it's got pretty good overclock potential and performs well against similarly priced Intel chips.

    http://www.overclockers.com/amd-phenom-ii-x6-1075t-review/

    1. Re:Initial review... by 0111+1110 · · Score: 1

      In that benchmark an overclocked I7 beat an overclocked 1075T. I7s can be had pretty cheap these days so I don't see a compelling reason to get the 1075T. I think AMD is mainly selling to the mainstream folks who don't overclock or run very demanding software. I do like AMD and they are kicking ass in the GPU market, but they have yet to catch up to Intel, which shouldn't even be that hard since Intel has been moving forward at a snails pace for a while now. AMD has been getting lots of catch-up time and they have been blowing it. I hope they can turn that around. Instead of beating intel at their own game of massive SMP they should be trying to beat them at single threaded performance. The majority of the apps I use on a daily basis are still single threaded.

      --
      Quite an experience to live in fear, isn't it? That's what it is to be a slave.
    2. Re:Initial review... by thoromyr · · Score: 1

      "I7s can be had pretty cheap these days" is not a comparison to the AMD price, but instead is implicitly indicating that I7 processors don't cost as much as they used to. Few (if any) people are saying AMD out performs intel, they are saying that the *ratio* of price/performance is better for AMD than intel. If true then for *best* performance buy intel. If you want or need to save money, buy AMD.

      I don't buy systems very often, but I do watch for a while before I do. What I have seen (which may or may not be the case at the moment) is that at the low end Intel is the buy, mid range AMD gives the best price/performance and for the high end it is Intel. This has led me to buy AMD because I need more than low end performance but, try as I might to justify it, I can't really afford the high end Intel.

      As others have pointed out it isn't just the CPU cost, its the motherboard. And the fact that the sockets don't change constantly. Its also the RAM. I had a motherboard go bad on me and because it was an AMD system I was actually able to buy a new motherboard that supported the old CPU I had *and* was compatible with the just released latest and greatest from AMD. If I had had Intel I would've had to spring for new CPU, new motherboard and new RAM. I couldn't have afforded it. And depending on how long it goes before I look at upgrading/replacing the system I may be able to get by with simply buying a new CPU and adding more RAM.

      This last spring I noticed a *12* core AMD processor that, on some server class motherboards, you could go dual processor with hideous amounts of RAM. I don't recall the pricing now, but at an approximation I could buy both processors and motherboard for less than the best Intel with plenty left over to get a goodly amount of RAM. I don't recall with certainty how many cores the Intel had, but in some cases hyperthreading decreases performance (even Intel only claims ~10% best case performance boost from hyperthreading so don't bother trying to claim that a 4 core Intel is equivalent to 8 cores).

  24. Re:And 3 hours after reading this... by kestasjk · · Score: 2, Interesting

    I wonder what the $/performance ratio looks like, rather than $/core...

    --
    // MD_Update(&m,buf,j);
  25. Re:That would be all nice and dandy if only... by Anonymous Coward · · Score: 0

    We already know how slow the Phenom II quad core is. This "new" six core design is almost identical, only with two more cores slapped on. So yes, we can easily gauge its performance without ever touching it.

    Also, this.

  26. Re:That would be all nice and dandy if only... by kestasjk · · Score: 1

    True, but realistically AMD aren't going to have developed & released a radically new arch at a budget price without telling anyone.

    --
    // MD_Update(&m,buf,j);
  27. Re:That would be all nice and dandy if only... by iamhassi · · Score: 1

    **Citation needed

    --
    my karma will be here long after I'm gone
  28. ? Do you really think Intels are 4x faster by SmallFurryCreature · · Score: 4, Interesting

    Intel provides currently the highest buyable performance. But AMD provides the best performance for value. If you buy a 200euro amd you get the best bang for your buck. If you buy a 800 euro Intel you get more bang but pay more bucks per bang.

    Intel offers no chip that provides the same bang for buck ratio as AMD. Hasn't done so in a long time.

    That is why AMD is the choice for price concious buyers who want high performance on a budget and Intel for the rich people who simply want the most powerful CPU.

    There are plenty of reviews comparing AMD vs Intel, Intel comes out ahead often but only by a small margin and for a HUGE price difference. Your choice wether you pay top money for minor gains.

    Just as a super car costing 10x as much as a regular one isn't going to go ten times as fast. By that logic the Shuttle would have to break the speed of light.

    --

    MMO Quests are like orgasms:

    You may solo them, I prefer them in a group.

    1. Re:? Do you really think Intels are 4x faster by Barny · · Score: 1

      Yeah, I have been building a TON of PCs of late with the Athelon II 635 in them, so much horsepower for so little cash, and you can throw 8G of ram at it and it will be a missile of a rig :)

      --
      ...
      /me sighs
    2. Re:? Do you really think Intels are 4x faster by somersault · · Score: 5, Funny

      If you buy a 200euro amd you get the best bang for your buck. If you buy a 800 euro Intel you get more bang but pay more bucks per bang.

      That's all very well, but how many bangs could a bangbuck buck if a bangbuck could bang bucks?

      --
      which is totally what she said
    3. Re:? Do you really think Intels are 4x faster by Kjella · · Score: 2, Informative

      But AMD provides the best performance for value. If you buy a 200euro amd you get the best bang for your buck. If you buy a 800 euro Intel you get more bang but pay more bucks per bang.

      The Core i7-860 spanks everything AMD has at $280 @ newegg, there's only a few odd benchmarks AMDs $300 top six-core CPU wins. Then entire market from $250+ and up is Intel, Intel, Intel. The $100 market AMD wins, but their value gets worse the closer you come to the high end. You make it sounds like Intel only owns the Ferrari market, when in reality they own the whole $50,000+ car market.

      --
      Live today, because you never know what tomorrow brings
    4. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      If you buy a 200euro amd you get the best bang for your buck. If you buy a 800 euro Intel you get more bang but pay more bucks per bang.

      That's all very well, but how many bangs could a bangbuck buck if a bangbuck could bang bucks?

      500! We've discussed this before.

    5. Re:? Do you really think Intels are 4x faster by IBBoard · · Score: 1

      Maybe I just hit a good patch, but the Core i5s didn't seem as bad as they used to be. I recently gutted my computer, upgraded the processor (i5), mobo (can't remember) and memory (4GB), and it was ~£300. Yeah, it's not in everyone's price range for normal desktop use, but that seemed good for a development machine and with better performance than I'd have got from an AMD. I guess these new chips might shake that up a little, though.

      That said, I probably would upgrade the mother-in-law's computer to an AMD, if it weren't for the fact that she insists she's poor and wants me to do it for free or not at all (i.e. getting my wife's machine when we replace it).

    6. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 3, Informative

      You should also consider the price of the motherboard. Core i7 motherboards are very expensive.

    7. Re:? Do you really think Intels are 4x faster by rbarreira · · Score: 4, Insightful

      The Core i7-860 spanks everything AMD has at $280 @ newegg, there's only a few odd benchmarks AMDs $300 top six-core CPU wins.

      Except that:

      1- The top AMD six core is actually $275, not $300.
      2- The AMD motherboards are cheaper, you can easily save at least $100 on that.
      3- The AMD motherboards are more likely to work with future CPUs (Intel has already changed sockets between Nehalem and Sandy Bridge... again).
      3- A 6 core CPU is probably more future proof than a 4 core one (even if those Intel cores are more powerful individually than the AMD ones, not arguing that).

      I agree with you that the AMD advantage is smaller at this price point than at the $100-$200 one, but the advantage is still real.

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    8. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 2, Informative

      Except that Apple is known to put "crippled" processors in some of their products. A perfect example is the current generation Macbook Pro. It has an i7 in it, but its not a quad core. It's a dual core with hyperthreading. Good luck finding that detail on their website, though. Also, there are very few Apple products these days that let you swap out the CPU if you want. From what I've seen, it's not the scientific community that likes Intel Macs, its the scientific community's managers. Most of the heavy lifting seems to be IBM hardware (from what I've seen).

    9. Re:? Do you really think Intels are 4x faster by OeLeWaPpErKe · · Score: 1

      Why do people ever make the argument that apple devices (anyone of their devices, really) are "cheap" ?

      That's like saying Ferrari's the cheapest brand once you require 400 kph+.

      Who cares ?

    10. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      la la la la (undoing mod!!!)

    11. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      buck off

    12. Re:? Do you really think Intels are 4x faster by dogmatixpsych · · Score: 1

      The i7 cores are hyperthreaded so you effectively have 8 cores. I'd rather have 6 real cores than 4 real + 4 hyperthreaded though. However, in practice (I use a computer with an i7) I haven't found a case when I've maxed out all "8" i7 cores before I run out of RAM (even with 8GB).

    13. Re:? Do you really think Intels are 4x faster by Tanktalus · · Score: 1

      You must not run Gentoo, then. :-P

      I regularly max out all "8" i7-930 cores here, though almost only when compiling. "make -j25" (with distcc spreading it out to my quad-core phenom box) generally will max it out when compiling openoffice, firefox (xulrunner) and kdelibs, and many others. 12GB RAM, though I doubt that to make a difference.

    14. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      You blew it. Should have been How many bangs could a bangbuck buck if a bangbuck could buck bangs?

    15. Re:? Do you really think Intels are 4x faster by OeLeWaPpErKe · · Score: 1

      And ferrari's are really cheap if one factors in that sometimes you have to run from a raging tyrannosaurus ...

      Same argument.

    16. Re:? Do you really think Intels are 4x faster by somersault · · Score: 2, Informative

      I'd actually gone a less conventional route to start, then tried to fit it more in with the scheme. However, I find banging things to be a more entertaining image than bucking them, so I'm happy with my choice.

      --
      which is totally what she said
    17. Re:? Do you really think Intels are 4x faster by Colonel+Korn · · Score: 1

      The Core i7-860 spanks everything AMD has at $280 @ newegg, there's only a few odd benchmarks AMDs $300 top six-core CPU wins.

      Except that:

      1- The top AMD six core is actually $275, not $300.
      2- The AMD motherboards are cheaper, you can easily save at least $100 on that.
      3- The AMD motherboards are more likely to work with future CPUs (Intel has already changed sockets between Nehalem and Sandy Bridge... again).
      3- A 6 core CPU is probably more future proof than a 4 core one (even if those Intel cores are more powerful individually than the AMD ones, not arguing that).

      I agree with you that the AMD advantage is smaller at this price point than at the $100-$200 one, but the advantage is still real.

      If you're willing to overclock, the i7 can easily get a 50% boost for the $50 cost of a larger heat sink that has the side effects of being silent and keeping the temperature of your heavily overclocked chip well below the stock chip with the stock cooler.

      --
      "I zero-index my hamsters" - Willtor (147206)
    18. Re:? Do you really think Intels are 4x faster by Nemyst · · Score: 2, Informative

      You're mistaking the i7-9xx for the i7-8xx. Only 9xx are on the more expensive X58 chipset. I'm not saying P55 motherboards are cheap, but they're certainly cheaper.

    19. Re:? Do you really think Intels are 4x faster by gottebag · · Score: 1

      17

    20. Re:? Do you really think Intels are 4x faster by Orestesx · · Score: 1

      Yeah, but now you are talking $150 more than an AMD solution. If you covet the i7 for the giant epeen that it will give you, fine, but you are throwing your money away.

    21. Re:? Do you really think Intels are 4x faster by Bert64 · · Score: 1

      You also have more resale value on a mac. In many cases its cheaper to buy a top end mac, use it for a year, ebay it and buy the latest model than it is to do the same thing with a generic machine. You may have paid more up front, but you will pay less each time you upgrade once you've deducted the sale price of your old machine.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    22. Re:? Do you really think Intels are 4x faster by Cornelius+the+Great · · Score: 1

      True, logical cores can't replace physical cores; a maxed out quad-core with hyperthreading is nowhere near the performance of a maxed-out 8-core system. But the public doesn't realize that- they bring up task manager, and see OMG-8 CPUs!!!

      Hyperthreading is just a fancy marketing term for Simultaneous MultiThreading (SMT). SMT allows the processor to issue two threads simultaneously in one superscalar pipeline so that threads can alternate whenever delays are encountered in the pipeline- these two threads aren't actually executed simultanously. Try explaining that to B&M computer shoppers or even the salespeople, and you'll get blank stares. I've made the mistake, once, trying to correct a know-it-all jackass at bestbuy who claimed that the i7 in a laptop I was looking at was actually an octocore cpu- I won't make that mistake again.

      There are benefits and drawbacks, but the important thing to note is that multiple threads within a single core will share and starve each other of resources- for that reason you would with a true multicore design. For current implementations, you'll end up with a ~35% performance gain at best. At worst, you'll lose 15% performance on threads that share resources and/or cause each other to cache miss more often.

      Currently typing this on an i7 920 with 9 gigs of ram (triple-channel). My home PC is a Phenom II X4 w/ 4 gigs of dual-channel ram. 99% of the time, I can't tell the difference between the two, and when I do, it's because of thrashing.

      --
      Sigs are for losers
    23. Re:? Do you really think Intels are 4x faster by iamhassi · · Score: 1

      "Intel provides currently the highest buyable performance. But AMD provides the best performance for value. If you buy a 200euro amd you get the best bang for your buck. If you buy a 800 euro Intel you get more bang but pay more bucks per bang."

      I tell friends this: AMD is 90% of the speed for 1/4th the price.

      You want 100% speed? Pay 4x more.

      No one buys Intel after hearing that.

      --
      my karma will be here long after I'm gone
    24. Re:? Do you really think Intels are 4x faster by iamhassi · · Score: 1

      "If you're willing to overclock, the i7 can easily get a 50% boost for the $50 cost of a larger heat sink"

      AMD can do the same thing: "Whew, what a ride. This processor (1075T) was a blast to benchmark and overclock. 4.5 GHz on air was just icing on the cake."

      So Intel and AMD are even when it comes to overclocking, and six cores at 4.5ghz sounds pretty nice.

      --
      my karma will be here long after I'm gone
    25. Re:? Do you really think Intels are 4x faster by Kjella · · Score: 1

      2- The AMD motherboards are cheaper, you can easily save at least $100 on that.

      The chepest AM3 socket board I could find here in Norway including 25% VAT was 403,- NOK ($68). The cheapest LGA1156 socket board (what you need for a Core i7-860, confusing I know) was 599,- NOK ($101) and looks quite comparable. Of course it wouldn't make any sense to pair a high end processor with a low-end mobo, but the difference is not $100. Maybe $50 if you spend $100 on the AMD and $150 on the Intel board.

      --
      Live today, because you never know what tomorrow brings
    26. Re:? Do you really think Intels are 4x faster by iamhassi · · Score: 1

      "The Core i7-860 spanks everything AMD has at $280 @ newegg, there's only a few odd benchmarks AMDs $300 top six-core CPU wins. "

      Why was this informative instead of troll? There's no links, no data, no proof, just a blanket INTEL RULZ! statement but of course he throws in that there are some tests where the AMD wins so anyone that attempts to refute his INTEL BESTEST OMG!!! statement with proof could be told "well, that's the few odd benchmark I was talking about".

      Guess PassMark must be one of those "few odd benchmarks", where the 1090T is 10% faster than the i7-860 and it's only $8 more, or save $72 (more ram anyone?) with the 1055T which is only 7% slower than the i7-860.

      --
      my karma will be here long after I'm gone
    27. Re:? Do you really think Intels are 4x faster by DurendalMac · · Score: 1

      Intel offers no chip that provides the same bang for buck ratio as AMD. Hasn't done so in a long time.

      Except that the i5 750 and Phenom II x4 965 are neck and neck in performance and were both priced right around $200 for a long time. It was only fairly recently that AMD cut the prices to compete.

    28. Re:? Do you really think Intels are 4x faster by rbarreira · · Score: 1

      I was actually thinking about the i7 9XX motherboards (I forget what socket that is), my bad.

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    29. Re:? Do you really think Intels are 4x faster by DurendalMac · · Score: 1

      He's comparing the socket 1156 i7 860, not a 1366 930. The boards are only a bit more expensive than an equivalent AM3 board, which renders #2 moot. A bulldozer chip will NOT work in AM3, only AM3+. The only advantage is being able to stick a current AM3 CPU in an AM3+ board...and I don't know why you'd want to do that. AMD's current boards are about as future-proof as Intel's, which makes #3 moot. And how the hell can you say the six core AMD chips are more future proof than the i7? Has multithreading not come around yet? Tests that blast the hell out of all cores (both physical and virtual) on both chips often show the i7 coming out ahead. It's just stupid to say that the AMD is more future-proof when you top out both and Intel is the winner. #4 is just idiotic.

    30. Re:? Do you really think Intels are 4x faster by DurendalMac · · Score: 1

      Except that the boards are not $100 more expensive for the 860, which sits in the 1156 socket. So really, you're talking maybe $75 more, or $25 more if you slap the same heatsink on the AMD chip and overclock it.

    31. Re:? Do you really think Intels are 4x faster by rbarreira · · Score: 1

      True, I was thinking about the 1366 socket motherboards (but even the older motherboards are still more expensive than the Intel ones).

      > A bulldozer chip will NOT work in AM3, only AM3+.

      Has this been confirmed? Not as far as I've seen. The point still stands that Intel changes sockets more often.

      > Tests that blast the hell out of all cores (both physical and virtual) on both chips often show the i7 coming out ahead.

      It seems to me that they don't. Link. This at a time where most applications are probably optimized for dual/quad cores.

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    32. Re:? Do you really think Intels are 4x faster by dch24 · · Score: 1
      Hmm, just for fun I priced out the highest end systems I could find on newegg:
      This is just what it would take to boot the board. They have identical on-board video chips which are great for KVM-over-IP but you may want a different graphics card, and then there's RAID, an optical drive, etc.
      AMD

      Skip the DVD and hard drive(s) since that should not affect the comparison.
      Total price: $3,439.96


      Intel

      Total price: $4,098.26


      Conclusion: AMD's Magny-cours option is awesome at the high end. Best of all is its price. Anandtech sums it up: "Here the choice is less clear. At this point, we believe both server CPUs consume about the same power, so that does not help either to make up our minds. It will depend on how the OEMs price their servers."

      Newegg prices put things way over on the AMD side of things. What am I missing?

    33. Re:? Do you really think Intels are 4x faster by mkiwi · · Score: 1

      I'd point out that the processors are "crippled" just like some of their graphics chips are "crippled" so that the computer does not overheat. It has nothing to do with Apple wanting to sell you bad hardware, and everything to do with giving you a computer that doesn't break down every 6 months.

    34. Re:? Do you really think Intels are 4x faster by dch24 · · Score: 1

      After researching some more, a socket G34 motherboard will support Bulldozer, so that's another huge plus (in my mind).

    35. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      how many bangs could a bangbuck buck if a bangbuck could bang bucks?

      Damn prosecutors. We all knew that shutting down the Craigslist adult services pages would merely push the trading of bangs for bucks off onto other, non-adult websites, but they just wouldn't listen.

    36. Re:? Do you really think Intels are 4x faster by waferhead · · Score: 1

      If you buy a 200euro amd you get the best bang for your buck. If you buy a 800 euro Intel you get more bang but pay more bucks per bang.

      That's all very well, but how many bangs could a bangbuck buck if a bangbuck could bang bucks?

      Pathetic losers, everyone know the answer is 42 ;-)

    37. Re:? Do you really think Intels are 4x faster by BlackSnake112 · · Score: 1

      It is a dual core or an i5? We got a macbook pro it says an intel dual core T9600. It is not the latest macbook pro line. Or else when you install win 7 on it the CPU changes.

      Anyway I have seen many i5 based laptops they are nice. Not as nice as the i7 based ones. For most regular people an i5 with 4GB (or more) of RAM in a laptop is just fine.

         

    38. Re:? Do you really think Intels are 4x faster by sexconker · · Score: 1

      That's all very well, but how many bangs could a bangbuck buck if a bangbuck could bang bucks?

      Between 0 and infinity.
      If a bangbuck could buck bangs, then it would be between 1 and infinity.

    39. Re:? Do you really think Intels are 4x faster by sexconker · · Score: 1

      You should also consider the price of the motherboard. Core i7 motherboards are very expensive.

      You should also consider the price of the motherboard upgrade when Intel changes the socket for no fucking reason, with no forward/backward compatibility, every other Tuesday.

    40. Re:? Do you really think Intels are 4x faster by sexconker · · Score: 1

      3- The AMD motherboards are more likely to work with future CPUs (Intel has already changed sockets between Nehalem and Sandy Bridge... again).

      It's been "tock-tock-tock" from Intel, instead of "tick-tock-tick". Stop changing the fucking socket you tools!

    41. Re:? Do you really think Intels are 4x faster by queequeg1 · · Score: 1

      >> A bulldozer chip will NOT work in AM3, only AM3+.

      >Has this been confirmed? Not as far as I've seen. The point still stands that Intel changes sockets more often.

      Anandtech just ran a similar article on these chips and made the same assertion regarding the inability of Bulldozer to work on AM3. Anand cited AMD itself as the source of the information. He made the point of noting that nothing is 100% until the actual products are finalized but that's the source of the info. Kind of a bummer since I'm in the market for a new machine (after 60 months with my 939 system) and was hoping to get something that was slightly future proofed. Oh well.

    42. Re:? Do you really think Intels are 4x faster by rbarreira · · Score: 1

      I searched a bit and found this:

      http://www.xbitlabs.com/news/cpu/display/20100826225852_Desktop_Bulldozer_Processors_Will_Require_New_Platforms_AMD.html

      Advanced Micro Devices said that its next-generation desktop processors code-named Zambezi will use socket AM3+ platforms, which will be backwards compatible with the firm's existing AM3 products.

      It's from a month ago, do you have the link to the newer info?

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    43. Re:? Do you really think Intels are 4x faster by espiesp · · Score: 1

      Well the Opterons don't come with heatsink/fan's, but that certainly doesn't tip the scales in Intel's favor...

    44. Re:? Do you really think Intels are 4x faster by queequeg1 · · Score: 1

      This is the article I was referring to (read through the comments for discussion about AM3+ compatability):

      http://www.anandtech.com/show/3937/amds-fall-refresh-new-phenom-ii-and-athlon-ii-cpus-balance-price-and-performance

      Here's another blurb about the same issue:

      http://www.tomshardware.com/forum/289211-28-official-socket-support-bulldozer

    45. Re:? Do you really think Intels are 4x faster by mr+exploiter · · Score: 1

      The Core i7-860 spanks everything AMD has at $280 @ newegg, there's only a few odd benchmarks AMDs $300 top six-core CPU wins.

      Except that:

      1- The top AMD six core is actually $275, not $300.
      2- The AMD motherboards are cheaper, you can easily save at least $100 on that.
      3- The AMD motherboards are more likely to work with future CPUs (Intel has already changed sockets between Nehalem and Sandy Bridge... again).
      3- A 6 core CPU is probably more future proof than a 4 core one (even if those Intel cores are more powerful individually than the AMD ones, not arguing that).

      I agree with you that the AMD advantage is smaller at this price point than at the $100-$200 one, but the advantage is still real.

      1-So the difference is 5 usd that doesn't change anything.
      2-100 usd, really? i7-860 use 1166 socket, and there are some 1166 mothers that cost less than this so I call this bullshit.
      3-Who cares, unless you are a hardware enthusiast you're probably going to use a i7-860 for so many years that this doesn't make any difference. And if you are you want the best not the cheapest.
      3(sic)- What? A chip that need 6 cores to archive the same performance than one that has 4 is obviously technically inferior.

      I really think AMD should stop astroturfing it's dishonest and doesn't speak well of the company.

    46. Re:? Do you really think Intels are 4x faster by EXrider · · Score: 1

      I'm not an Intel fanboy, but what if you figure power efficiency into the equation? I chose a Core2 over a Phenom on my MythTV backend a few years ago based on efficiency.

      --
      grep -iw skynet /etc/services
    47. Re:? Do you really think Intels are 4x faster by _Sprocket_ · · Score: 1

      I think the answer is geographically sensitive. For example, the answer might be different if the buck banging was in Bangkok vs. Timbuktu.

    48. Re:? Do you really think Intels are 4x faster by bhcompy · · Score: 1

      At least 939 had legs. I just got out from under a 754 system that was essentially obsolete as a socket in 6 months. Now with AM3 I've got a ton of scalability that should last for another few years a lot better than the 754 a64's did. Running 3 cores now an can easily drop in a phenom 2 when I'm ready for more. Mobo has usb3/sata3(6gb) as well as multiple 16x pcie2.0 slots, for 80$. Scalability is the best thing about AMDs offerings currently

    49. Re:? Do you really think Intels are 4x faster by Sporkinum · · Score: 1

      Passmark has a great bang per buck graph.
      http://www.cpubenchmark.net/cpu_value_available.html
      The PC I just bought my wife has a AMD Athlon II X4 620 in it. It rates pretty high in bang per buck, and I will say it's worked well for her. Bone stock, her PC boots Kubuntu in 30 seconds. Win7 takes 1.5 minutes.

      --
      "He's lost in a 'floyd hole"
    50. Re:? Do you really think Intels are 4x faster by Rockoon · · Score: 1

      These 6 core AMD's can be clocked over 4ghz on the stock cooler. ...and I know its stable. Been running a 1055T at 4.1ghz for several months now.. stock heat sink and fan.

      Yes you can overclock Intel chips, but you can overclock AMD chips too. I'm sorry for you that Intel ships with such a crappy heat sink and fan that you have to buy another set to break 4ghz.

      The fact remains that Intel's comparable-performing chips are more expensive. Put together a shopping list for the best basic system's (motherboard, CPU, power supply, and case) for each of the $300, $400, $500, $600, and $700 price points. You just wont be able to match AMD's performance for the same cost. I know, I recently tried.

      ..and just to be clear, all these Thuban's are insanely overclockable... so dont try to pull that "I can overclock Intel" card... its weak grasping.

      --
      "His name was James Damore."
    51. Re:? Do you really think Intels are 4x faster by hairyfeet · · Score: 1

      Mind a little advice from an old greybeard PC builder? Go AM3 and you should be good for at least 5 years, and here is why: You can buy a nice business class AM3 motherboard quite cheaply, and with AMD having nice low end quads like the Athlon II 635 you can build a nice rig quite affordable, and then drop in a 6 core when the price gets dirt cheap.

      Since we are just now seeing programs that are dual core aware, and running an AMD Phenom II 925 I can tell you VERY few programs take advantage of quad cores, barring some miraculous breakthrough in parallel programming even the low end quad will be a nice system for awhile and it will be most likely another 4 to 5 years before 6 cores are anywhere near the norm.

      So if you go AM3 and it turns out you can't run Bulldozer, who cares? Unless you are doing extreme number crunching (in which case you might want to go for broke and get an i 970) the quad will last you easily another couple of years, and add a few more for 6 cores. Finally add in the fact that you can OC the Athlon 635 quite high with a decent board and HSF and you should be good to go without seriously damaging your wallet. For most day to day jobs (video and audio editing/transcoding, gaming, graphic design, etc) the AMD quads are real champs. I have been pounding my 925 pretty hard with Cubase and Virtualdub and with the performance I'm seeing I can't picture myself replacing this for another 5 years. if you are still on 939 the speed increase will be well worth it, trust in da feet.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    52. Re:? Do you really think Intels are 4x faster by rbarreira · · Score: 1

      > A chip that need 6 cores to archive the same performance than one that has 4 is obviously technically inferior.

      It achieves a higher performance on benchmarks that are actually using all of the cores. Obviously not many applications right now, but no one is exactly saying that everyone should go out and get a six core CPU. It's obviously a niche product right now.

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    53. Re:? Do you really think Intels are 4x faster by somersault · · Score: 1

      The location in this instance is ever moving, as it is in fact a train journey from Buckingham to Bangalore. Possibly with a further leg to Phucket.

      --
      which is totally what she said
    54. Re:? Do you really think Intels are 4x faster by dch24 · · Score: 1

      Ok, thanks! (I didn't notice because the last 2 servers I've built I OC'd so I had to get a different cooling solution anyway.)

      For the Socket G34:
      Dynatron A6 77mm heatsink/fan $34.99

      For the Socket LGA 1366:
      Intel BXSTS100C heatsink/fan $29.99. Be warned: most of the reviewers don't like this one because it's loud, but it's the only one I could find rated at 130W TDP.

    55. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      Neither. It's a dual core i7. The T9600 is from late 2008/2009 models, not the current model.

    56. Re:? Do you really think Intels are 4x faster by Omestes · · Score: 1

      In January I built a new computer with a decent budget. I had to decide between an AMD 965 and some flavor of i7 (I forgot what was in the same price range at the time). With the AMD I could get a decent mobo for around $100, with Intel it was around $150. Not a game changer. But I would have to replace all my RAM with DDR3 for the Intel, adding around $250 to the mix (8GB to match what I would be carrying over from my old box). Getting a new Phenom II x4 965, plus a mobo, and keeping my DDR2 (which works fine, upgrading isn't worth the extra $400) cost around $500. Getting the Intel at the same price point cost around $750, including RAM and mobo. Moving up one or two percentage points in benchmarks wasn't worth it for that much.

      But then again I always shoot for mid-high, and not bleeding edge, since the cost/performance curve gets heavily skewed towards cost at the very top.

      As for the 965, I have no complaints (well one, the stock cooler SUCKED, which is often true, and Fry's Electronics is evil, selling both a bricked processor and mobo). I'm sure a mid-range i7 would be a bit faster, but I really haven't actually noticed any speed issues. Game performance is about as good as expected (even with an aging video card, the next project), no problems with encoding, or processing RAW format images. Etc...

      In other words its good enough. The only benefit I would have gotten from the Intel is a longer upgrade cycle (with the need to completely rebuild my system again at the end, thanks to their socket schizophrenia). But then again I probably will end up upgrading CPUs in 2-3 years anyway.

      --
      A patriot must always be ready to defend his country against his government. -edward abbey
    57. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      That's a lame excuse. Tons of other OEMs aren't having much of an issue with the quad core i7's. Anyway, its the fact that Apple doesn't mention how many cores or what i7 model it is. I assume most people seeking an i7 know they are typically quad core and would be caught off guard.

    58. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      3- A 6 core CPU is probably more future proof than a 4 core one (even if those Intel cores are more powerful individually than the AMD ones, not arguing that).

      Step back from Intel vs. AMD for a moment and consider a thought experiment. Take two CPUs. One has two cores, the other has four. The 2-core CPU has cores which are individually 2x as fast as the 4-core CPU, so total available computing throughput is equal. Which one should you buy?

      The answer is always the 2-core CPU. Why? Because even with 4 or more threads, it runs just as fast as the 4-core. And if you have fewer than 4 threads, it runs faster, as much as 2x when you go down to a single thread. (And don't make the mistake of thinking that software which depends on single-thread performance is going away any time soon.)

      This is why, when 4 Intel cores match or beat 6 AMD cores on very parallel throughput benchmarks (which is usually the case), it's obvious which one you should prefer.

      Also, even without the benefit of that thought experiment to clarify matters, most Intel 4-core CPUs gain performance as you scale beyond 4 threads due to hyperthreading.

    59. Re:? Do you really think Intels are 4x faster by rbarreira · · Score: 1

      The problem with that thought experiment is that it's not true that Intel's 4 core CPUs beat equally priced AMD's 6 cores on heavily multithreaded tests.

      Hyperthreading is a 10%-performance-boost thing in most cases...

      --

      The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
    60. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      The problem with that thought experiment is that it's not true that Intel's 4 core CPUs beat equally priced AMD's 6 cores on heavily multithreaded tests.

      Name one. I've seen plenty of heavily multithreaded tests where Intel CPUs equal or beat AMD despite having many fewer cores.

      Here's one where AMD wins by a tiny bit. HP 4-socket servers, AMD's best 12-core CPU (6176 SE) versus Intel's best 8-core (X7560) (48 AMD cores vs. 32 Intel cores), both tests sponsored and run by HP:

      http://www.spec.org/cpu2006/results/res2010q2/cpu2006-20100608-11617.html
      http://www.spec.org/cpu2006/results/res2010q3/cpu2006-20100719-12577.html

      Now, SPECint_rate is as perfect a case for multithreaded scaling as you're going to get, because the methodology is to take a single-threaded benchmark (SPECint) and run as many independent, parallel copies of it as you have hardware threads. There is literally no better case than this.

      If more cores are more future proof against heavily threaded workloads, why is it that Intel is keeping pace with 2/3 the cores here?

      (Note that it probably is roughly equal despite AMD having a slightly higher score. If you look at the individual benches which make up SPECint, it's quite likely that AMD is picking up some points by having figured out a really aggressive compiler optimization for 462.libquantum, since their peak score is almost 3x base on that one. (The difference between base and peak is that 'base' runs of SPEC benchmarks are only allowed limited compiler tricks while 'peak' is closer to no-holds-barred.) This kind of optimization game is always going on in SPEC, so until Intel does that optimization in their compiler, that one is going to skew things in favor of AMD in a way which doesn't reflect real world performance.)

      Hyperthreading is a 10%-performance-boost thing in most cases...

      This isn't Pentium 4 hyperthreading anymore. They've gotten a bit better than that.

    61. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      Except that Apple is known to put "crippled" processors in some of their products. A perfect example is the current generation Macbook Pro. It has an i7 in it, but its not a quad core. It's a dual core with hyperthreading. Good luck finding that detail on their website, though.

      Wow, hand in your geek card. Intel makes different kinds of mobile i7. One of the primary ways to categorize them is by power consumption. Most of the products are dual-core at 18W, 25W, or 35W (depending on target market: 18W for ULV with really low clock speeds, 25W for clock speeds of about 2 GHz, 35W for ~2.5 GHz). There are quad core mobile i7s, but only in the very highest mobile power bins Intel offers, 45W and 55W.

      The i7 Apple uses in MacBook Pros is the i7-620M, a 35W 2-core/4-thread 2.66 GHz processor. It is a standard Intel part and is not crippled in any way. It's the best performance you can get in the 35W power bin.

      Since Apple's philosophy in notebook design is to shoot for very long battery life and thin enclosures, they simply do not offer any products based on the hotter quad core CPUs. This is not crippling. It's just an engineering tradeoff. With limited internal room for battery capacity and cooling, 45W and higher processors are impractical.

      (Also, the quad core i7 mobile processors have a harsh tradeoff: their base clock rate is much lower than their peak turbo clock rate, and if you actually load all four cores they're very likely to throttle down close to the base clock rate to keep power under that 45W or 55W limit. It's still the best thing you can get if you want to crunch numbers on a notebook computer, but it won't be nearly as fast as a desktop i7. I don't call that crippling either, it's just a tip o' the cap to the laws of physics.)

    62. Re:? Do you really think Intels are 4x faster by Anonymous Coward · · Score: 0

      The earth would be pleased if you moderate your retardness.

      How many fucking watts are you wasting compiling shit you can get already compiled and usable on other OSs? Or are you trying to justify buying the i7 (or any other high ePenis CPU) so you can do unnecessary shit faster?

      Do my kids a favor a die.

  29. a tip on buying AMD processors by keeboo · · Score: 3, Interesting

    If you use your computer for heavier stuff like Qemu emulating weird architectures, heavy compilation, HD video and things like that:
    Go for the X4 models with 6MB L3 cache, it will do wonders with your aging AM2 motherboard (check for compatibility first, of course).

    Really, forget the 1 or 2MB L2-only models. Those are quite a disappointment for such tasks (to me? they're rubbish).
    I was considering a full upgrade to a Intel i5 (processor, mobo & memory) because my annoying sluggish old AMD Athlon 64. Frankly, my previous bad experiences with AMD processors (K6-1, old Athlons) did not help to form a positive opinion about the brand. But, hey, that Phenon processor was so cheap that I thought "heck, why not" and I was quite surprised.

    1. Re:a tip on buying AMD processors by Anonymous Coward · · Score: 0

      I wouldn't make such a blanket statement. I'm running on an Athlon II X3 425, and I've been quite happy with it. For anyone considering buying from AMD, check benchmarks to see whether or not the additional L3 cache makes a difference. There actually are tasks out there that do not benefit so much from additional cache, believe it or not.

      http://www.anandtech.com/show/2775/6

      http://www.xbitlabs.com/articles/cpu/display/phenom-athlon-ii-x2_9.html

      Depending on what tasks you care about, it may matter more than you think. Or it may matter less than you think.

    2. Re:a tip on buying AMD processors by FranTaylor · · Score: 1

      If you're feeling more spendy you can get a multi-socket Opteron 2000 or 6000 series server motherboard from SuperMicro or TYAN. With lots of RAM and 12-16 cores, you can emulate a whole room full of computers.

    3. Re:a tip on buying AMD processors by zippthorne · · Score: 3, Funny

      What tools are you using to determine that on-die cache is the bottleneck for your usage?

      --
      Can you be Even More Awesome?!
    4. Re:a tip on buying AMD processors by keeboo · · Score: 1

      Just regular software I normally use, among the most dramatic cases I remember: GCC, Qemu (like ARM emulation, for example), bzip2, my hoggish proprietary Samsung printer driver...etc. That compared against the performance I had with an Athlon 64 X2 (512+512 L2 only).

      There's also the fact the machine does not "hiccup" under load like before, but that's subjective and hard to demonstrate.

    5. Re:a tip on buying AMD processors by zippthorne · · Score: 1

      I don't know why this is funny. cache is one of the many knobs you have for designing your system.

      Among the easily accessible advertised metrics are:

      Processor
              Speed
              Number of cores
              Last level cache
              clock multiplier

      RAM
              size
              speed
              bus width

      Disk/SSD
              Size
              speed
              seek time
              rpms (if disk)

      I wouldn't expect it would be too hard to write a tool that just catalogues cycles spent waiting on each slower tier, to determine where the bottleneck really is for what you're using it for. There's no point in wasting money on extra L3 cache, or even RAM if your particular problem is CPU bound.

      But... I'm sure it's quite a bit harder than I think, because I haven't yet encountered such a tool, despite the fact that the vast majority of people don't have the money to buy a matrix of computer components and test out the combinations to find the price-optimal one. Really, the ideal case is to guess, test, and find it's close enough, or only needs a small amount of tweaking.

      --
      Can you be Even More Awesome?!
  30. Re:And 3 hours after reading this... by beelsebob · · Score: 2, Interesting

    Intel already does have chips that compete with at the same price. The i5 760 will beat these in pretty much any task, and costs $208. So it doesn't have 6 cores, but it is sufficiently faster on a per-core basis that it doesn't matter.

  31. Re:That would be all nice and dandy if only... by Anonymous Coward · · Score: 0

    I assume you haven't bothered to look at those latest Anandtech benchmarks yourself. If you did you would have noticed that AMD doesn't generally win with heavily threaded apps in those benchmarks. As you can see on the latest Anandtech benchmarks a Core i7 860 (2.8 GHz, 4 cores) will get similar results as a Phenom II X6 1075T (3 GHz, 6 Cores), often it will be ahead of the AMD chip. Since the Core i7 are usually benchmarked with Turbo mode disabled the numbers should even be a little better in real life than in those tests. Then there is Hyperthreading which does help for heavily threaded apps, as tests with x264 have shown.
    The AMD CPUs look better on paper, but that doesn't hold up in the real world. It's just like with Intels Pentium 4 CPUs that boasted huge clock rates but weren't much faster than lower clocked Athlon XPs. Unlike Intel CPUs back then the current AMD offering is a little cheaper, so it's still a good buy.

    http://www.anandtech.com/show/3937/amds-fall-refresh-new-phenom-ii-and-athlon-ii-cpus-balance-price-and-performance/1

  32. Re:And 3 hours after reading this... by Anonymous Coward · · Score: 1, Informative
  33. Re:AMD One-Ups Intel? Another misleading Slash sto by Anonymous Coward · · Score: 0

    Could you explain that?

  34. Sharing FP units between cores is nothing new: T1 by IYagami · · Score: 1

    The UltraSparc T1 shared his single floating point unit between its 8 cores and 32 threads

    See http://en.wikipedia.org/wiki/UltraSPARC_T1#Physical_characteristics

  35. Re:And 3 hours after reading this... by Anonymous Coward · · Score: 1, Informative

    Intel still charges insane money for their chips.. they only lower it to a fair price after AMD forces them ...

  36. Re:And 3 hours after reading this... by Vectormatic · · Score: 3, Interesting

    for the price segments where both AMD and intel are active in (so, that is the below $250 segment), $/performance is roughly equal, with AMD stealing some leads (and in some cases, very significant leads, in some segments intel only offers some insanely slow old celeron, where amd offers a x3 or so). AMD mostly wins because they are offering more cores/$. In single threaded performance, a c2d chip might just beat that athlon II x3, but as soon as threading comes into play, the 3rd core wins the battle for AMD

    Taking all things into account (cpu/mobo), a performance equivalent AMD system will be somewhat cheaper then a comparable intel build

    Personally i prefer AMD for that reason (not to mention i got into PC building in the amd 64 days, which might have contributed to my AMD preference)

    --
    People, what a bunch of bastards
  37. Vital missing info... by advocate_one · · Score: 1

    Article neglects to mention that the processor prices are xxx dollars "each" if ordered in quantities of 1000 units or more not xxx dollars for 1000 units... and summary merely repeats the mistake...

    processor chips are not really this cheap...

    --
    Donald 'Duck' Dunn: We had a band powerful enough to turn goat piss into gasoline.
    1. Re:Vital missing info... by AK+Marc · · Score: 1

      That error is done thousands of times a day all over the world (so much so that if you believe in descriptive language, not prescriptive language, it's no longer an error). Those that price things in lots get used to "each" pricing when "each" isn't specified. The only reason it's an issue here is that people aren't used to "each" pricing for lots of 1000+. And based on the numbers given, it's quite clear what is meant.

  38. Re:That would be all nice and dandy if only... by Vectormatic · · Score: 1

    The 1075T is a new chip, but the 1055T and 1090T have been out for months, same six cores, same cache, just a slightly lower/higher clock speed

    --
    People, what a bunch of bastards
  39. Re:And 3 hours after reading this... by kestasjk · · Score: 3, Insightful

    Personally i prefer AMD for that reason (not to mention i got into PC building in the amd 64 days, which might have contributed to my AMD preference)

    Ah yes, the very same reason I like Lotus for office productivity software and FoxPro for databases; once a computing great, always a computing great, and you've got to stay loyal of course.

    --
    // MD_Update(&m,buf,j);
  40. Re:And 3 hours after reading this... by beelsebob · · Score: 1

    Yup
    and
    Yup
    and
    Yup

    So basically, the i5 760 will beat it handily at pretty much everything except for one embarassingly parallel task.

    Good game though.

  41. Re:And 3 hours after reading this... by Vectormatic · · Score: 1

    considering you say you like lotus, i guess my sarcasm detector must be on the fritz...

    anyway, i prefer AMD, but that is not to say i will ONLY buy AMD, that wouldnt make sense anyway (in fact, my file server is a core 2 Duo, my small experimentel webserver is a dothan based celeron)

    --
    People, what a bunch of bastards
  42. CPU manufacturers and I have a history by RogerWilco · · Score: 4, Interesting

    I think AMD really only One-upped Intel twice: When they were the first past the 1GHz mark and when they developed AMD64 while Intel was mucking around with Itanium.

    I've owned many non-Intel machines, the full list goes like this: Intel 8086-4.77, NEC V20-8, Cyrix 286-20, AMD 386-40, Cyrix 486DX2-66, AMD DX4-120, Cyrix P166+, AMD K6-300, AMD Duron-700, AMD K7-1,400, Intel PIV-3,06 Intel PentiumM-1,7, AMD Athlon64 X2-2,0, AMD Phenom X6-3,2

    I've never had any trouble with any of them, even though some had motherboard chipsets from SIS or VIA. The DX4-120, K6-300, K7-1,4 and all the newer ones are still running. (The DX4 is a stand-alone DOSbox for my dad to run some ancient software (on 360k floppies!), The K6 serves as a firewall somewhere, the K7 is used when my mom needs Windows (she's got 2 macbooks), the P4 is now in a laptop and now a media server, the PentiumM is in my current laptop, the Athlon64 is in my dads current computer and I run on the X6).
    Now I look at it, even though I left my parents over 15 years ago, they are still a kind of dumping ground for my old computers. :-D

    --
    RogerWilco the Adventurous Janitor
    1. Re:CPU manufacturers and I have a history by mr_mischief · · Score: 1

      How about when AMD was the first of the two to do on-die memory controllers? Or maybe when they were the first of the two to have a quality high-end graphics division? Even though that was through purchase, I think it was a smart purchase. AMD also did 386s up to 40 MHz, a DX4 up to 120 MHz, and forced Intel into the model number marketing when AMD processors were running faster clock-for-clock than Intel chips back in the Athlon XP+ days.

    2. Re:CPU manufacturers and I have a history by CajunArson · · Score: 1

      How about when AMD was the first of the two to do on-die memory controllers?
          You mean AMD invented the 486 before Intel? Despite what you read on your AMD fanboy websites, on-die memory controllers weren't invented by AMD.

      Or maybe when they were the first of the two to have a quality high-end graphics division?
            Oh yeah.. that whole $5.4 Billion buyout of another company that already did graphics. Also for all the buzzwords AMD's "fusion" parts will still make it to market later Intel's own stuff and Intel will be the first to admit that they suck at graphics to boot! Talk about massive "innovation" there, but in case Intel ever does buy Nvidia I'm sure you'll be the first to loudly proclaim that buying out another company to get technology is perfectly fine....

      --
      AntiFA: An abbreviation for Anti First Amendment.
    3. Re:CPU manufacturers and I have a history by VGPowerlord · · Score: 1

      That depends on what you mean by one-up.

      The AMD Athlon was significantly faster than the Pentium IIIs of the time, simply by changing the FPU design.

      --
      GLaDOS for President 2016! "Well here we are again. It's always such a pleasure." -- GLaDOS, 2011
    4. Re:CPU manufacturers and I have a history by Anonymous Coward · · Score: 0

      AMD also "one upped" on:

      Glueless MP.
      True multi core (not dual die, single package)
      Better IPC (back during the P4 marketing based clock rates)
      Silicon on Insulator (for better power usage/lower heat-- again back at the P4 time)
      Integrated Memory controller

    5. Re:CPU manufacturers and I have a history by Wolfrider · · Score: 1

      [[
      The DX4 is a stand-alone DOSbox for my dad to run some ancient software (on 360k floppies!)
      ]]

      --Brother, you would be well advised to see if you can Virtualize that PC by now -- any part of it could literally DIE at any moment. And good luck finding replacement parts for anything resembling low-cost...

      --
      .
      == WolfriderV6 == I'm willing to admit that *I just might* be wrong... Are you??
    6. Re:CPU manufacturers and I have a history by Anonymous Coward · · Score: 0

      I'd say long lived CPU socket types are worth at least half a one-up, too. Intel has always been kind of schizophrenic on sockets, and that combined with their aggressively tiered prices means you never really get a chance to run a CPU for a sane amount of time and then swap it for a better yet competitively priced CPU later - because it's damn near guaranteed it'll require you to also replace your motherboard, which aside from being a pain in the ass is also expensive enough that the new CPU is no longer a good deal anyway.

      But my year old AMD system's mobo, which wasn't a new socket when I got it, will be able to run the stuff coming out for the next couple years. If I find myself with a few hundred extra bucks in the next six months, I can swap out my current triple core chip for this new six core chip, for example. (And I only limit that speculation to six months because they've got new stuff coming out after that... which, unless I'm terribly mistaken, is supposed to continue using the same socket type). So it's a double-win. The price:performance ratio is there even if you consider only the CPUs, but not having to also replace the motherboard effectively drops the price at least a hundred bucks more for me, plus it's a great saving of aggravation.

    7. Re:CPU manufacturers and I have a history by afidel · · Score: 1

      Also the fact that they had a decent processor interconnect for almost a decade before Intel caved and copied them with QPI.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    8. Re:CPU manufacturers and I have a history by phoenix_rizzen · · Score: 1

      First to 1 GHz. First to lose the frontside bus. First to 64-bit x86. First to true multicore (on same die). There's probably a few more. AMD has been first to a lot of neat technology.

    9. Re:CPU manufacturers and I have a history by mr_mischief · · Score: 2, Informative

      I said it before Intel did it, jackass. I'm not a fanboy, BTW. I'll buy Intel when it makes sense. I just prefer AMD after about two decades of experience buying products of both companies.

      The 486 had a DRAM controller on its die? I'm going to have to ask for a citation. I think you're thinking of either the on-die L1 cache or the MMU (memory management unit), neither of which is a system main memory controller. Here's a citation to the counter: List of Intel Chipsets at Wikipedia. See how the chipset determines the memory specs up until the Core i Series, including the 80486? Here's another: List of Intel Chipsets at World IQ. Here's another: Intel CPU and Chipset History at Overclock 3D courtesy of a forum post there by "PV5150".

      An MMU has nothing to do with controlling the actual SIPPs, SIMMs, or DIMMs. It's a multiprocessing ("multiprocessing" doesn't mean "multi-core") feature that allows the processor to enforce memory address range protection so that program A doesn't stomp on program B's memory range. That's a separate concern from getting data into and out of the processor from main memory.

      Why don't you go get a copy of something like Upgrading and Repairing PCs and inform yourself? Here's the ISBN.nu link for the 18th edition in case you're a bargain shopper: 18th edition. I have the fifth edition myself. I might get an updated version for the handy reference tables in the back featuring things like POST codes and error codes for SCSI controllers.

      BTW, have you ever actually built a PC older than, say, a Pentium 4? Or owned one?

    10. Re:CPU manufacturers and I have a history by phoenix_rizzen · · Score: 1

      AMD did a 486DX4 133 MHz CPU. That was my first PC. :)

    11. Re:CPU manufacturers and I have a history by dwinks616 · · Score: 1

      Take a look at http://en.wikipedia.org/wiki/Memory_controller Notice that AMD has had on-die memory controllers starting with the AMD64 line of CPUs? And Intel with only their i7? AMD has had on-die memory controllers for something like 7 years, Intel is more like 7 months.

    12. Re:CPU manufacturers and I have a history by Anonymous Coward · · Score: 0

      As far as engineering goes, Intel was definitely first. Remember when Intel was pushing RDRAM? They designed and validated a Pentium 3 family CPU model with an on-die RDRAM controller. There was real silicon and it was developed to the point where it could have been flipped over to mass production. The reason it wasn't? It was supposed to be a low cost CPU, but RDRAM prices never dropped enough for a budget platform (*), so Intel killed the project.

      That episode may have played a role in Intel being late to go back to on-die memory controllers. Once they got burned, they probably didn't want to risk being wedded to any particular new memory interface technology before the market had shaken out enough to pick a winner.

      * - It turns out that the DRAM mfrs colluded to keep RDRAM prices high to keep it from taking over, basically because they didn't want to have to pay patent royalties to Rambus. Rambus sued several large DRAM suppliers over it and has won some large settlements as a result.

  43. Re:And 3 hours after reading this... by mr_mischief · · Score: 1

    You know "parallel" when you're not running a just a benchmark and you're not running a big server or HPC load means "desktop", right? How often do you only have one application running? Multiple processes go to multiple cores just like multiple threads do.

  44. Re:And 3 hours after reading this... by Anonymous Coward · · Score: 0

    My current build is using an AMD Phenom II X6 1090T CPU. My previous build was an AMD X4 Phenom. My previous 3 builds were Intel and before that were a few AMDs. I started it all with an Intel 286 many many years back.

    I buy whatever is the best bang for the buck for a full system (including motherboards which has firewire, and now SATA3/USB3). Intel full builds with similar capability / speed tend to be about 200-300 bucks extra usually. I am not a fanboy but AMD has always helped in more ways then one. At best it gives us better bang for the buck, at worst it keeps Intel on it's toes and dropping prices. I do a new build for myself on an almost yearly basis. I wonder how the PC landscape will look after Bulldozer is out.

    I also have a bunch of friends / clients who get what I recommend. And guess what I have been recommending the pass couple of years? AMD / ATI of course!

  45. Re:And 3 hours after reading this... by Anonymous Coward · · Score: 0

    C64? It was great back in 1982.

    http://en.wikipedia.org/wiki/Commodore_64

  46. Re:And 3 hours after reading this... by h4rm0ny · · Score: 5, Insightful

    Intel will have an offering which provides equal performance for approximately the same price.

    You're joking aren't you? Intel currently owns the highest performance segment of the Desktop chip market. AMD doesn't produce any Desktop chips that can match Intel's best in any impartial benchmarking. But AMD has been confidently out competing Intel on "bang for buck" for some time now. I doubt Intel will suddenly lower prices to AMD's levels. If Intel are going to lower prices to compete, they've had just as much reason to do so for some time already. And don't forget motherboard support. AMD has traditionally been friendlier to separate motherboard and CPU upgrading than Intel which is a hidden cost.

    --

    Aide-toi, le Ciel t'aidera - Jeanne D'Arc.
  47. Re:And 3 hours after reading this... by beelsebob · · Score: 1

    Yes, but 1) this is not all parallel tasks, it's one specific parallel task, and 2) how often do you really load up *all* cores at once running multiple desktop applications. Instead, it's *much* more common to load up all cores at once by running one well written process.

  48. Overclocked, Intel and AMD have similar price by viking80 · · Score: 2, Funny

    Have built systems for quite a few years, and it seems like you can overclock the hell out of intel chips using just good air coolers while AMD pretty much are running at peak speed. Both regarding heat and not crashing.

    Built a dual core core 2 Duo 1.83GHz. It is running stable at 3.5GHz.
    Intels 32nm i5-650 3.2GHz easily overclocks to 4.7GHz (not sure if stable yet)

    If you compare Intel with AMD after you take this headroom into account, intel is on par with if not more cost effective than AMD.

    --
    don't cut it off www.mgmbill.org
  49. Where are they buying their processors at? by gravis777 · · Score: 0

    AMD's new chips include the fastest AMD Phenom II X6 1075T six-core processor, which is priced 'under $250' for 1,000 units, AMD said. AMD also introduced a range of dual-core and quad-core Athlon II and Phenom II desktop microprocessors priced between $76 and $185.

    I am not sure about the 1075T, but I picked up a 2.8Ghz 6-core and motherboard combo six months ago for $200. I would say that is WAY less than than "under $250 for 1,000 units", Also, at the time, the comparable Intel was around $480, and did not include the motherboard.

    As other people have said, AMD gives you more bang for your buck, but Intels are cheaper. I will take this one step futher. Not sure if this is still true, but my experience have been that Intels are better at raw-number-crunching, whereas AMDs seem to handle multitasking better. On my six core, I can be processing h264 video over all six cores, still run photoshop and have it exporting a movie to my TV all at the same time. I would think the i7 could do that as well, but man, what a price jump!

    1. Re:Where are they buying their processors at? by gravis777 · · Score: 2, Funny

      DANG IT! That's what I get for getting on Slashdot at 6:30 in the morning. Meant Intels are FASTER, not cheaper!

  50. Re:And 3 hours after reading this... by OeLeWaPpErKe · · Score: 1

    Actually you should see what microsoft has created with microsoft visual foxpro. It's a beautiful product, certainly much more so than access.

    Just my 2c.

  51. The mind boggles by ThatsNotPudding · · Score: 1

    Just think how fast that AMD cpu will be once you get the unlock card from Best Buy!

    Seriously; AMD should run an ad campaign pointing out this latest Intel craven shenanigan.

  52. Re:And 3 hours after reading this... by The_mad_linguist · · Score: 2, Insightful

    Have there been any real innovations in word processing software in the last ten years?

    I mean sure they've gotten shinier and bloatier, but I haven't seen any real groundbreaking features.

  53. Apart from Hackintosh, why go Intel? by Anonymous Coward · · Score: 0

    I went Intel on my last build as I wanted the box to double as a Hackintosh multi-boot.
    The idea was that the kids could bring their Elementary school-work home.
    In practice, they prefer to use Windows and the box has spent next to no time running OSX.

    (Usage probably breaks down 50% Windows 7 for Steam, 25% Ubuntu for Ubuntu upgrades, 25% XP for PIC and ARM programming.)

    Back to AMD next build.

  54. Re:And 3 hours after reading this... by mr_mischief · · Score: 1

    Well, I'll admit I'm not a typical desktop user for starters. I'm also not running the typical desktop OS on most of my machines.

    Most of the time that I'm filling the four cores on this box to capacity, it's running something that forks into multiple processes rather than threading inside one. Usually something like running make against a C or C++ program. My browser also uses separate processes, one per tab. Sometimes I'll be testing server software or debugging it and the client, server, and all the middleware together will peg all four cores. Most of the software I personally write uses forks rather than threads when I need multiple streams of execution. Linux balances those out among cores very well, and a fork on Linux is cheaper than the equivalent process launch on Windows.

    Other people I know do things like run a CPU and GPU intensive game while running antivirus, a host firewall, some communications program (IM or voice chat) and streaming music. Until you get to a really monster number of cores, more cores will give you a speed boost even if most of those tasks only use one or two threads or processes.

    Given that more and more people are using SSDs and more and more people are using multi-core chips, there's no reason a full anti-virus scan has to check single files serially. Especially with multiple SSDs in RAID 0, 5, or 6 there's no reason you couldn't fill a couple of cores with just a full filesystem scan.

    I run IDS software, backups, logging, update checker, crypto services, mail server, name server, ssh server, time synchronization, database server, intranet web server, X server, window manager, and miscellany on this box all the time, and it spreads those things out over all four cores evenly. When I, say, build a new Linux kernel with make running 32 or 64 jobs at once, having four cores really speeds things up over having two. Having six would speed it up more, I'm sure. Having SSDs or building in a RAM disk wouldn't hurt, but I've yet to pay for enough RAM to do that and I'm trying to pick which SSD I want as I type this.

  55. The money I saved I bought a SSD by Anonymous Coward · · Score: 0

    I bought a black six core AMD for the home and the money I saved I was able to install a SSD. Wow! Now my $1,500 home rig now beats my $3,800 four core Intel at work hands down. It nows pains me to have to sit at work waiting on this Intel computer to complete its tasks. I feel like seaking my AMD machine to work and showing it off. AMD home machine loads Windows 7 in about 11 seconds while my Intel work machine takes almost a minute - I typically can go get a cup of coffee and come back only to find my Intel work rig is still booting. Thank you AMD!

    1. Re:The money I saved I bought a SSD by Issarlk · · Score: 2, Funny

      If you had bought an *Intel* SSD it would boot in 10 seconds!

  56. Will it require $50 activation like Intel? by Borealis · · Score: 4, Informative

    Frankly if they don't do the cheesy-as-hell activation fee like Intel is proposing I'm sold.

    --
    Unbreakable toys can be used to break other toys.
    1. Re:Will it require $50 activation like Intel? by scharkalvin · · Score: 1

      Nothing new. Big Blue used to do the same thing with their 360/370 mainframes. When a customer ordered a memory upgrade the tech would show up with a wirewrap gun and install a few jumpers on the backplane to activate the memory that was ALREADY INSTALLED IN THE COMPUTER AT THE FACTORY!

    2. Re:Will it require $50 activation like Intel? by alen · · Score: 1

      "Enterprise" products have been doing this as long as i can remember. last year i was looking for a new tape library and most of them license the slots at ridiculous prices. ended up buying an HP MSL 8096 with no nonsense all licensed slots. same thing with fiber SAN switches. they will sell you a "low end" $6000 switch that is the same hardware as the $50,000 switch. only difference is that most of the ports and features are disabled. you buy the keys to unlock the features.

      all this nonsense started around the time that RSA lost some patents due them expiring

    3. Re:Will it require $50 activation like Intel? by Anonymous Coward · · Score: 2, Informative

      They still do this, and it's a nice feature :)

      They ship completely loaded out machines, then we use Capacity Upgrade on Demand to enable the cores via software (though the HMC console). It sounds like idiocy to do this, but when this means that during our peak season we can enable/disable cores and save literally hundreds of thousands of dollars versus buying a fully loaded and enabled machine, it makes sense.

      Keep in mind that this is not only for the hardware. Oracle licensing is HIDEOUS. We pay millions to license their database across our enterprise. We also pay $500K and more for Tivoli software (an IBM company). We pay for scheduler software. All these are based either on the processor capacity, number of processors, memory size, disk size, number of IO adapters, etc.. So adding a few cores here and there has an immediate impact on our licensing costs.

  57. Re:And 3 hours after reading this... by PopeRatzo · · Score: 2, Interesting

    Intel could compete on price...

    That's for sure.

    From TFA:

    Core i7-970 processor, which is priced at $885 per 1,000 units

    Somebody is marking those things way up by the time they get to my local store.

    --
    You are welcome on my lawn.
  58. Hmm... by MacGyver2210 · · Score: 0, Troll

    You could buy one of these cheaper procs... ...but then you're still stuck using an AMD chip.

    They're garbage from my experience. Open, cheap, and absolutely shitty fabrication.

    --
    If the only way you can accept an assertion is by faith, then you are conceding that it can't be taken on its own merits
    1. Re:Hmm... by Anonymous Coward · · Score: 1, Interesting

      absolutely shitty fabrication.

      What, do the bits fall out when you shake them? Or do you have an STM in your house?

      (captcha: oxides)

  59. Re:And 3 hours after reading this... by Pojut · · Score: 1

    What kind of innovations can you think of? You type, and the words appear on screen. Break out the bubbly :-)

  60. Re:And 3 hours after reading this... by c6gunner · · Score: 0, Redundant

    You're joking, right? Have you actually looked at any benchmarks, anywhere? The ONLY way you can say that AMD beats out intel on price is if you consider the mobo as part of the cost. Otherwise, Intel spanks AMD on pretty much every front.

  61. Re:That would be all nice and dandy if only... by hairyfeet · · Score: 1

    But benchmarks are often to reality what bullshit is to a debate. I am typing this on a Phenom II 925 quad which is running a dozen tabs in FF, playing my fav playlist in WMP 12 AND doing a 16:9 to 4:3 conversion in Virtualdub and everything is smooooth as butter and only running at 123f on stock HSF. Oh and the ENTIRE rig, with 8Gb of DDR2 800, a 1Gb HD4650, a nice business class motherboard, and 2 500Gb HDDs with Windows 7 HP x64 cost a grand total of $650 before MIR and around $585 after.

    So one can point out various benchmarks giving test X to AMD, and test Y to Intel, but how often are normal folks actually running benchmark conditions? I pound the living hell out of my CPU, with serious number crunching tasks like video transcoding and audio recording and amp sims with Cubase, and through it all this AMD has handled it like a champ without needing a dozen fans or heating up my apt. Likewise I have many customers that pound their PCs, such as the ones built for graphic design for the local copy shop or the one being abused in a construction trailer, and they couldn't be happier with the performance they are seeing in real world tasks they do every day. And in the end, doesn't that matter more than who gets the score in X?

    --
    ACs don't waste your time replying, your posts are never seen by me.
  62. Where is the open CPU? by Anonymous Coward · · Score: 0

    "...but we're not really coming up with anything really revolutionary.... Come on, people, innovate!"

    ''...Not saying the new CPUs aren't cool, but I just want to see what the next leap will be already! ..."

    "...but AMD deserves a lot of credit for keeping the processor market competitive..."

    When do we stop letting profit oriented corporations define the age's technological standards? It's no secret that a company will hold back technology in order to maximise profit.
    I wonder where micorproessor thechnology would be today if there were more competition, real competition, where hardware companies are forced to keep up with technology instead of holding things back.(Think Microsoft and IE)
    It's not like hardware manufacturers are doing anything special, they're just applying knowledge that is freely available.
    Competition should be based on who can make the best designs.
    So with all the knowledge out there, why can't I pick an open design and have non-profit-manufacturer-X make it for me at cost?
    What's going on with open hardware?

    1. Re:Where is the open CPU? by zwarte+piet · · Score: 1

      Hardware guys shouldn't earn money?

    2. Re:Where is the open CPU? by VGPowerlord · · Score: 1

      So with all the knowledge out there, why can't I pick an open design and have non-profit-manufacturer-X make it for me at cost?

      What non-profit-manufacturer-X?

      Why would any manufacturer eat their sunk costs just to produce something for you for the price it costs them to produce it?

      I've never heard of a non-profit company in the manufacturing segment, simply because manufacturing has a ton of sunk costs.

      --
      GLaDOS for President 2016! "Well here we are again. It's always such a pleasure." -- GLaDOS, 2011
    3. Re:Where is the open CPU? by westlake · · Score: 1

      When do we stop letting profit oriented corporations define the age's technological standards?

      The day when the Genii of the Lamp grants you Three Wishes.

      The first wish for a labor force, materials, production facilities and distribution networks that can be built and maintained without spending a dime.
         

  63. Stop with the "bang for the buck" already! by Anonymous Coward · · Score: 0

    It's hunting season around here, you insensitive clods!

  64. Re:And 3 hours after reading this... by Rogerborg · · Score: 4, Insightful

    If you don't factor in purchasing a motherboard, then both chips offer zero bang for infinite bucks.

    --
    If you were blocking sigs, you wouldn't have to read this.
  65. THAT'S BECAUSE INTEL CHEATS! by Anonymous Coward · · Score: 4, Informative

    Some of the benchmark programs are compiled with Intel's C++ compiler, which generates CPUID checks for the manufacturer string 'GenuineIntel' and redirects all other manufacturer's CPUs to the slowest code path. So if you can't compile the benchmark yourself with a trusted compiler, its not worth the paper its printed on.

    Intel also releases several libraries that other software vendors use in their products; these libraries contain the same manufacturer check which cripples their performance on chips by AMD, Via, etc. Commercial software products such as Matlab have unintentionally or intentionally shipped with these checks, with the result that they run slower than necessary on AMD CPUs. When the manufacturer test is patched out of the program, it is un-crippled and runs as fast or faster than a comparable Intel chip.

    Intel settled out of court with AMD over this, and are in the process of also settling with the FTC, but have not actually stopped the practice.

  66. About time... by hesaigo999ca · · Score: 1

    Thats is probably what I like best about AMD, able to turn one up on Intel everytime, so that intel (with the better product anyways) needs to lower their prices...and therefor will allow me to get their next gen CPU for much cheaper, way to go AMD, keep it up....

    On a serious note, has anyone tried the phenom chip, can anyone say they actually like it, I have seen some benchmarks, nothing to get exited over, except the price...

    1. Re:About time... by Ant+P. · · Score: 1

      I bought an X3 (now an X4) just over a year ago. It's not the fastest thing in the world, but for encoding videos/compiling code it's more than good enough for me. Also it runs cool enough that you can safely run it with the stock fan powered off for average desktop use.

    2. Re:About time... by hesaigo999ca · · Score: 1

      So i guess getting a cpu with 8 cores on it will be superb!...I just hope they make it to be able to be understood by today's motherboards, or are we talking another 6 months before we see any avail. motherboards supporting it

    3. Re:About time... by Ant+P. · · Score: 1

      FWIW my mobo (an AM2+) supports the AM3 X6 chips, which didn't exist when I bought it. An X8 would probably work in it too.

  67. Re:And 3 hours after reading this... by Joce640k · · Score: 1

    AMD is *always* more bang per buck then Intel - they wouldn't be in business otherwise.

    --
    No sig today...
  68. Paying of OEMs is not their only trick.. by Anonymous Coward · · Score: 5, Informative

    For many years, the Intel C++ compiler has discriminated against non-Intel chips by detecting their manufacturer using CPUID and redirecting all chips not manufactured by 'GenuineIntel' to a slower code path. (And that manufacturer ID is their trademark so other manufacturers may not use it). C++ libraries available from Intel (such as their math libraries) also contain the same discriminatory code checks. This artificially decreases their performance on AMD's chips.

    It makes Intel chips look better, by slowing down the program on all of their competitor's chips. So the safest thing is NOT to use Intel's compiler for anything (most especially benchmarking). This is a problem because it has a reputation of producing larger, but faster, code. (Faster on Intel processors at least!).

    The code it produces is actually quite decent on AMD chips too, as long as you patch out the generated version checks to un-cripple the performance on AMD chips. You can do it as a post-build step after compiling. It's a hassle that most software vendors don't bother with -- in most cases they aren't even aware that Intel's compiler generates the manufacturer-checks and redirects their program through slower code paths on AMD chips.

    1. Re:Paying of OEMs is not their only trick.. by afidel · · Score: 2, Interesting

      No, it compiled them for a *different* codepath, when you were comparing Athlon to Netburst that was a *good* thing because the Netburst stuff was optimized for the crazy long pipeline whereas the regular i586 codepath was optimized for a more sane pipeline length and so had superior performance on the Athlon.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    2. Re:Paying of OEMs is not their only trick.. by Anonymous Coward · · Score: 0

      If this is true, why isn't this a massive anti-competition lawsuit?

    3. Re:Paying of OEMs is not their only trick.. by hairyfeet · · Score: 1

      Because Intel paid AMD to the tune of 1.25 billion to shut up and make the problem go away? And by taking the money AMD just about cleared any ATI debt left on the books which helped their bottom line. The bigger question is whether the US DOJ or the EU antitrust commission are gonna do anything or not.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    4. Re:Paying of OEMs is not their only trick.. by Rockoon · · Score: 3, Informative

      Bullshit.

      The fact that AMD's could run unoptimized code faster than the netburst optimized code is only testimonial to how crappy the netburst design was.

      FACT: Intel's compiler is more than happy to selectively test for each of SSE, SSE2, and SSE3 at runtime and will use the best path it can, if and only if the vendor ID is "GenuineIntel." If it is not "GenuineIntel", then it runs a 80386 codepath intead. Thats right.. you heard me.. it runs a codepath for 25 year old CPU's.

      Stop making excuses for the bullshit that is Intel's compiler shenanigans.

      --
      "His name was James Damore."
    5. Re:Paying of OEMs is not their only trick.. by afidel · · Score: 1

      -xO, enable SSE1,2,3 for non-Intel CPU's, use it if you want.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    6. Re:Paying of OEMs is not their only trick.. by haruchai · · Score: 3, Informative

      That's not the whole story - there was a deliberate attempt to not use optimized instructions.

      See a long discussion at http://www.agner.org/optimize/blog/read.php?i=49

      --
      Pain is merely failure leaving the body
    7. Re:Paying of OEMs is not their only trick.. by sjames · · Score: 1

      And that is in itself a good reason to cheer for AMD. certainly don't want the entire industry depending on a single company that is so willing to pull dirty tricks like that.

    8. Re:Paying of OEMs is not their only trick.. by Rockoon · · Score: 1

      That doesnt effect the standard library, or the extended libraries.

      From Agner Fog's Blog

      Version 10: Up to SSE4.2 paths for 32-bit Intel, only 386 path for non-Intel.
      Version 11: Up to AVX paths for 32-bit Intel, only 386 path for non-Intel.
      Version 12: Up to AVX paths for 32-bit Intel, only 386 path for non-Intel.

      The story is similar with the Vector Math Library.. If (!GenuineIntel) { run shitty 25 year old code path }

      --
      "His name was James Damore."
  69. Re:And 3 hours after reading this... by Joce640k · · Score: 1

    Somebody is marking those things way up by the time they get to my local store.

    They're called distributors, AKA middlemen.

    If you've got a spare $885,000 you could buy 1000, keep a couple for yourself and sell the rest on eBay. :-)

    --
    No sig today...
  70. Re:And 3 hours after reading this... by jedidiah · · Score: 1

    It seems that the one main thing that benefits from running on a single fast core versus smaller ones with more combined performance is Flash.

    Yes. Our friend Flash is there to ruin your day.

    Despite all the nonsense about GPU acceleration (on Win7 no less), it still seems to only do well when running on a single core fast enough to decode a BluRay in software.

    Running across a couple of slower cores won't cut it.

    For anything else, my own experience favors more cores even for "simple desktop use".

    --
    A Pirate and a Puritan look the same on a balance sheet.
  71. Re:That would be all nice and dandy if only... by rbarreira · · Score: 1

    The core i7 860 is more expensive than the 1075t and it requires a more expensive motherboard to boot.

    Not really a fair benchmark if what you care about is performance per dollar.

    --

    The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
  72. Re:And 3 hours after reading this... by kenh · · Score: 1

    Intel paying off OEMs for exclusivity deals is nothing new, they've done it for years and have been very forthright about it - what you seem to consider as a bribe has been spun as "loyalty discounts" for years.

    MS subsidises advertising budgets for PC Mfgs. if they include Windows 95/98/200/XP/Vista/Win7 logos in the ads and ship every model advertised with the OS mentioned. They have been doing this for over 15 years (since Win95 AFAIK).

    Intel makes nice chips, and some great deals can be found if you go with last gen technology. Intel has a dual core Celeron E3300 that supports virtualization that I can buy retail, with heatsink/fan, for $40 at http://www.microcenter.com/single_product_results.phtml?product_id=0327583 - I don't really consider that a price premium over AMD, and MBs are available at comparable prices to AMD models. The cheapest dual core AMD chip I can find retail is $61 http://www.microcenter.com/single_product_results.phtml?product_id=0346724 with a single core Sempron available for as low as $33 http://www.microcenter.com/single_product_results.phtml?product_id=0317380 .

    AMD & Intel both make some very nice chips, and with a little bit of effort you can buy a quad core CPU from either mfg. for about $100-120 (AMD 620 and Intel Q8300)

    --
    Ken
  73. Re:And 3 hours after reading this... by Joe+The+Dragon · · Score: 1

    only if you pay Intel $50 more for the unlock code to get the full power of that cpu!

  74. Re:And 3 hours after reading this... by Halifax+Samuels · · Score: 1

    If that's true then how is Intel in business at all?

  75. Re:And 3 hours after reading this... by fast+turtle · · Score: 5, Insightful

    how often do you really load up *all* cores at once running multiple desktop applications.

    Lets see:

    I've got three Java background apps running and unlike the Folding Client, they do not back off when I want to do something else. This means Firefox, Word, Outlook, One Note, XMPlay all have to fight them for any ticks on the CPU though I rarely see more then 50-75 avg. cpu loading. That's on an E6300 (1.8GHz) Core 2 running Win7-64 on 8GB and this is a typical situation for my system.

    My system is 3 years old and I've just started looking at upgrading but I have a problem. There are no CPU's now available from Intel that are compatible with my board and no a Bios update wont solve the problem. They changed the damn socket 6 months after I built it. Intel has a habit of changing things ever 6 months so you can't upgrade you CPU to gain the performance boost needed when the time comes. In my case, the only option if I could find one is a Q6600, which has already been discontinued (18 months ago) so I'm now forced to look at building a new system.

    Due to Intel's policy, I'm looking at AMD for my next system because they don't obsolete Sockets and Chips 6 months after you build the system, forcing you to buy the most chip you can afford and then replacing the entire system in two or three years when it can't keep up with demands. That's right. It's Intel that drives the business upgrade cycle because they can get more money from companies selling all new chips such as north/southbridge, nics and everything in between unlike AMD who prefers to see you buy more CPU's and gives us a gradual upgrade path by simply ensuring their new chips can run in at least the "+" series of sockets even though you may not have access to all features.

    --
    Mod me up/Mod me down: I wont frown as I've no crown
  76. Go AMD by Anonymous Coward · · Score: 1, Interesting

    My last system was an AMD build and I just built another right before he 6 core Phenoms came out. I play games and got Phenom II X4, Crosshair III motherboard, and a AMD5850 video card, 800W Corsair PS and a samsung LED. IBasically it amounted to the best an AMD system could offer at the time. The best Intel could offer was more than 1k more. Maybe some Intel systems benchmark higher than mine, but you CANNOT see the difference. That what pisses me off so much about benchmarks is that they are just numbers and people start saying what is what. Try a system, try the operations YOU will be performing on that system and then tell me whats worth it. Benchmarks are a guideline, nothing more. You think because Intel can encode faster that it is better? When was the last time John Doe encoded anything? When was the last time John Doe did anything other that look at porn and type in word. Most people could live with current Netbook performance for their daily tasks. Hell, the iPad seems to be enough for most people nowadays. You should always value your dollar, with any purchase, for anything. I will keep buying AMD until their performance does not match the price, but that won't be for a while cause I'll probably be able to put any next gen chip of theirs in my AM3 socket.

  77. Re:And 3 hours after reading this... by anUnhandledException · · Score: 1

    Not sure why some people find it hard to believe.

    The i5-750 & i5-760 provide great value for performance.
    Generally speaking they match or exceed AMD offering at that price point.

    IMHO they provide good value (which is unusually for an Intel offering). Only reason I passed on that is because I had 8GB of DDR2 ram. Using AM3 CPU I could resuse the ram, going intel would require new ram (and $$$). Still the i5-7xx is a nice series of chips.

    The large point remains. In the $50 - $274.

  78. Re:And 3 hours after reading this... by pandaman9000 · · Score: 3, Informative

    EXCEPT:

    The intel motherboard for those less expensive CPUs has zero forward-looking value. They are already obsoltete when you buy them now. Additionally, the memory on the board will be obsolete DDR2. Furthermore, there will be no USB3/SATA3 option.

    The AMD board will be closer to current, can have DDR3 should you choose it, and will support AMDs newest quad core/six core processors currently out.

    This is like me pointing out that a prius makes a hella good race car, because it can get over 15 laps out of a gallon of gas. A PC is a bigger picture than the CPU.

  79. Not really an apples to apples comparison. by anUnhandledException · · Score: 3, Informative

    The i7-860 costs a lot more than a the X6 1075T

    Sure the chip is only $20 more $250 vs $270 however you need a motherboard. i7 MB are notoriously expensive.

    One can find a decent crossfire/SLI capable AM3 motherboard for $80 - $100. Not so with i7. Prices start at $160 and tier 1 brands are more like $180 - $200.

    When you consider the additional cost of the motherboard your i7 solution is running 30% higher than the X6 platform. 30% higher for maybe 10% more performance.

    Performance per $ is what matters. Sure you can get that i7 but it will cost you $100 more. Someone buying an X6 could spend that $100 on a better GPU making an overall better system.

    Pure power Intel has always been a leader but many of their solutions fall apart in a price per value metric.

    1. Re:Not really an apples to apples comparison. by FreelanceWizard · · Score: 1

      Mm, no. The i7-860 is an LGA 1156 processor, and boards for that are in the same range as the AM3 boards. You can grab an ASRock H55 for $75 on NewEgg, and $109 will get you an ASUS board with 2x PCIe and CrossfireX support. nVidia SLI support, of course, comes at a higher price point because the manufacturer must license the tech from our favorite space heater vendor.

      LGA 1366, which is where the i7-9xx processors are, is indeed much more expensive, and probably for questionable benefit; HT isn't that great, though the 920 and 930 have an almost legendary reputation for easy and extensive overclocking (though one must be wary of unbalancing certain voltages that result in the on-die memory controller burning out). However, LGA 1366 is probably more future-proof, as it's currently the high-end non-server socket.

      --
      The Freelance Wizard
  80. Re:And 3 hours after reading this... by rbarreira · · Score: 1

    And a cheaper 4-core CPU from AMD will be at par with the i5 760 while being cheaper...

    --

    The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
  81. Re:And 3 hours after reading this... by psbrogna · · Score: 1

    Because even after those in the know recognize a better value proposition there is significant latency before the rest of the world to catches up- decades in some cases.

  82. Re:And 3 hours after reading this... by spinkham · · Score: 2, Informative

    The features are small, but often quite worthwhile.
    For example, in Word 2010, you can choose based on the program you're copying from how you want styles to be handled, whether to use the original style, local style, or paste as plain text. That saves me tons of time I would have spent messing with manually choosing paste type before.
    2010 also has built in "dataleakage" detection, warning you about metadata you might be sending on accident.

    The last version of Word I used was 2003, then switched to Open Office (which I still often use).
    For collaboration reasons, I have to use Word for work and was quite pleased at many of the changed made in 2010 vs old versions. Yes, there's no groundbreaking features, but UI changes can a big deal and big time saver in programs also.

    --
    Blessed are the pessimists, for they have made backups.
  83. As fun as compiling may be by e065c8515d206cb0e190 · · Score: 1

    You may have to consider the cost (CPU+power) of your i7 vs the improvements you actually get. I typically throttle my i7 920 from 2.6GHz to 1.6Ghz in the summer to avoid heat (and also cut down on my power consumption).

    I absolutely love launching make -j32, but it is rare. If you're a developer in the OpenOffice.Org team, it could make sense though.

  84. Virt? by bill_mcgonigle · · Score: 1

    Anybody know what the current state is of virtualization hardware support across the AMD line? I could certainly go look it up for these individual processors, but I'm curious to know, "all X's of the Y line have it".

    I've been really happy with a couple recent builds I did with their 12-core parts, even at $750 each. Can these 6-core units handle dual-processor configs (clearly I'm just coming over to AMD from a long stint with Intel)?

    --
    My God, it's Full of Source!
    OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
    1. Re:Virt? by Rockoon · · Score: 1

      AMD's system with regards to AMD-V (hardware virtualization) is real simple. All the chips they continue to make support hardware virtualization except the Sempron's, and a few of those also support AMD-V.

      All Athlon 64's, Athlon II's, Phenom's, and Phenom II's...

      Some of the older Turions didnt support it, but it will probably be hard to find one for sale. Essentially, just get any Phenom II.. its got AMD-V.

      --
      "His name was James Damore."
    2. Re:Virt? by bill_mcgonigle · · Score: 1

      All the chips they continue to make support hardware virtualization except the Sempron's, and a few of those also support AMD-V.

      That's exactly the kind of wisdom I was hoping to hear. Thanks!

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
  85. Re:And 3 hours after reading this... by pandaman9000 · · Score: 1

    ^^^^^^^^^^^^ this ^^^^^^^^^^^^^^

  86. still expensive by zwarte+piet · · Score: 0

    $76 per 1000 units is still more than 7,5 cent a piece. *sigh* and then you just got the cpu. Will computers ever become affordable?

  87. Re:And 3 hours after reading this... by raw-sewage · · Score: 1

    I run IDS software, backups, logging, update checker, crypto services, mail server, name server, ssh server, time synchronization, database server, intranet web server, X server, window manager, and miscellany on this box all the time, and it spreads those things out over all four cores evenly...

    I don't disagree with the overall theme of your post, but just a bit of a nit: I often see people saying I run all these things on my computer, so I need as many cores as possible. The number of processes (or threads) really doesn't mean anything; it's how much work each of those processes are doing. My firewall/router computer is an OpenBSD system running on an AMD Geode processor (500 MHz I think). This is a single core CPU that's slower than even an Atom (I've seen people compare it to a 486!). On this machine, I run a name server, ssh server, time server, DHCP server, and a firewall. The Geode is plenty powerful for all these things for a small home network. Of course, if I was, say an ISP with 100s of customers, the Geode wouldn't be sufficient.

    So, in general, any one of the processes you listed can be virtually no CPU load, or require a whole cluster by itself. Consider a mail server. If you're not doing virus scanning, I would guess that you can support 100s of typical users on a pretty wimpy computer, as it's really mostly I/O load. But to run, e.g. gmail.com, that requires a bit more power. The same goes for a web server: if you're serving mostly static pages or relatively simple dynamic pages with a limited number of users (for example, a hobby web development server), it's mostly I/O load, and doesn't take a powerful CPU at all. But, on the other hand, pick any high-profile site, and it's obvious you need a lot more horsepower. The point is, simply saying "I run a webserver" doesn't really actually say anything about the load you're putting on your system.

  88. Re:And 3 hours after reading this... by pandaman9000 · · Score: 1

    It is very sad that some folks refuse to consider the MANDATORY motherboard change cost. The exceptions favor AMD, with the socket AM2 lasting for about 4 years, and 2+ lasting close to 3, when considering the MOST current CPU of the moment. Intel boards cost morer and last less time. Any top tier (parts-wise) AMD PC built in the last 3 years can move to the latest and greatest processor, by a imple CPU swap. Intel cannot even get you in the i7 GENERATION without a board.

  89. Re:And 3 hours after reading this... by DurendalMac · · Score: 1

    They already do. It's called the i7, and it still manages to beat the six-core Phenom IIs.

  90. Does not compute! by DarthVain · · Score: 1

    OK as someone that follows this pretty closely I am not all that impressed just yet. I have done the research, and no AMD does not have the price/point value or advantage.

    Firstly Intel's i5 will beat anything AMD has in 90% of benchmarks at the same price range. Secondly most people are hard pressed to use 2 cores let alone 4 or 6. Thirdly most developed software has a hard enough time using 2 cores, let alone 4, or 6. AMD's technology and die sizes is less advanced, and larger (which is not good). In certain specific situations AMD can have an advantage, but this is rare, and that must be all you plan on doing. So for instance if your using it for work, doing rendering, and that is all you are doing, and AMD has a benchmark advantage in the software you use for that, then sure go out and buy one. However for general usage, video games, etc... advantage Intel.

    Someone did make the point that the AMD mother boards are cheaper, and this is true. However at the low or mid end the differance is only maybe 20% or 20-30$ so not all that significant really (at the high end yes their are some stupidly expensive intel MB out there, but who buys those anyway, not someone that is really worried about price or value I assume). Also the past has shown us that AMD makes just as many socket changes as Intel, they have simply delayed it due to not being on par with intel technology. It will happen, and then your 20% less costly MB will be 100% less compatible when they do change the standards, as you just bought at the end of a lifecycle.

    Anyway like anything, it really comes down to your requirements, what you need, and what other components you plan on buying to make a complete system. Taking by their definition "components" and passing judgement may not be all that useful as you need to look at the whole system and what you plan to do with it. That said, if I were to go out and buy a new CPU tomorow it would be an Intel i5. AMD did have the advantage years ago, but ever since the Intel Core 2 Duo, AMD has been struggling to keep up. I sincerly hope that they will come out with something that trumps Intel technology, and takes back the technological advantage, as that would be the best thing for the market and the consumer. However simply slapping more cores on a solution it does not make.

    1. Re:Does not compute! by anUnhandledException · · Score: 3, Interesting

      i5 starts at $180. So IF (big IF) you want to spend $180+ on CPU then yeah you really can't go wrong with Intel i5-xxx series.

      However for many users they aren't CPU limited. CPU power makes very little difference in game performance once you get past dual core 2.5Ghz. It makes a difference but not as much as a GPU.

      In the $100 to $200 segment is where AMD really shines.

      From tomshardware.
      http://www.tomshardware.com/reviews/best-gaming-cpu-core-i5-760-core-i7-970,2698.html

      Best Gaming CPU for ~$70 - Athlon II X3 425
      Best Gaming CPU for ~$85 - Athlon II x3 445
      Best Gaming CPU for ~$115 - Core i3-530
      Best Gaming CPU for ~$140 - Phenom II x4 945
      Best Gaming CPU for ~$160 - Phenom II x4 955
      Best Gaming CPU for ~$180 - Phenom II x4 965
      Best Gaming CPU for ~$200 - Core i5-760
      Best Gaming CPU for ~$290 - Core i5-930

      When someone has a fixed budget spending $50 to $100 LESS on CPU allows them to buy $50 to $100 more GPU and that system will offer better overall game performance.

      I do agree that $200+ Intel dominates and AMD doesn't really have anything that answers but that is a small segment of the market. One also has to consider that i5 boards tend to run $10 to $20 more than comparable AMD board and i7 boards run $60 to $100 more.

    2. Re:Does not compute! by DarthVain · · Score: 1

      1) Tomshardware is not a good review site. The reviews there I take with a large helping of salt, the only thing I like there is the charts to amalgamate the larger scope of results. Even those I take in the most general of sense. [H]ardOCP is much better, as is Arand, and a host of others.

      2) "Best Gaming CPU" for say 140$ is like saying the Ford Fiesta is the best hauling truck for under 14k. It doesn't exist. If you are buying the severely low end, what does it really matter what you pick, they will all really suck, and/or do what job they can reasonably do about the same. So it isn't really a choice. No one is buying a CPU for gaming for 70$, the fact they even mention it shows how little understanding they have. Just like when a friend wanted advice on sub-500$ laptops. There is VERY little difference, they will all sort of suck, and they will all do the job you want it to do more or less. An even more extreme version is netbooks, as they almost always have the exact same components making comparing them a bit silly in most cases. This is because they are so low end.

      Considering what you are spending on building a system (which is likely what you are doing unless upgrading), and how important your CPU is (I would argue it is the MOST important), I personally don't understand why someone would spend X amound dollars on a system and go cheap on your CPU for the sake of 40-60$.

      A Hockey analogy (because I am from Canada eh? and have been looking at gear recently), if I am going to spend say 800-1500$ on gear, why would I only spend 100$ on skates which is the most important part.

      Same with building a computer system, if your going to spend 1000-1500the money to build one, why only spend 100$ on a POS CPU?

      If you argue, well if they are building a 500-600$ system it makes a difference, then I would argue that they don't really care about performance then, and it doesn't really matter then does it. So long as you can use facebook and send an email or two who cares.

      But then your not looking a a Price/Performance value. Your just looking at price, because you really don't care about performance.

      So that was pretty long winded to boil that down... :)

    3. Re:Does not compute! by DarthVain · · Score: 1

      GPU is definitely more important these days if you are gaming. However if your not gaming it is moot and in many cases even if you are, you can be easily CPU limited. Also if your getting a dedicated video card, your typically not getting a really cheap system to begin with.

      I would argue that the 170$-250$ (which is the mid range) is the Largest part of the market. I am talking custom system building, not the millions of POS low end garbage that Dell and others spit out each year.

      Yeah the intel board are a bit more expensive, but then again they also use a newer standard, so you are paying for it.

      Yeah i7 is too rich for my blood and the high end MB are just silly. i5 + 120-130$ MB is way more value for your money.

      Upgrading to a low end CPU makes no cost sense either (unless you are replacing a dead, not trying to improve or increase the life of your system). Typically your limited to your MB anyway, and once that is toast, you will need to spend a lot anyway, so why would you not maximize that as much as you can.

    4. Re:Does not compute! by anUnhandledException · · Score: 1

      ") "Best Gaming CPU" for say 140$ is like saying the Ford Fiesta is the best hauling truck for under 14k. It doesn't exist. If you are buying the severely low end, what does it really matter what you pick, they will all really suck, and/or do what job they can reasonably do about the same."

      Hardly once again virtual all games are GPU limited not CPU limited and especially so at higher resolutions.

      Given a finite budget spending less on CPU allows more to be spent on GPU.

      If someone had $400 on CPU & GPU combined (more than capable of playing most games) they would be better served spending $150 on CPU and $250 on GPU. Spending $250 on higher end i5 CPU makes little sense as the system would be EVEN MORE GPU limited. Even if GPU/CPU budget is $500 one would be better suited spending $200 on CPU and $300 on GPU.

      GPU is far more the limiting factor. For games spending >$200 on CPU buys you very little (and even less at higher resolutions).

    5. Re:Does not compute! by DarthVain · · Score: 1

      Speaking from personal experience I am wishing now that I spent a bit more on CPU.

      Years ago when I built my last system I combined a C2D 4200 and a 1950Pro thinking when the time came I would just OC the 4200 or upgrade it as needed.

      However I bought the one 4200 that couldn't OC worth a damn (or its my MB or something), and shortly after I bought it Intel changed the standards for my socket limiting my upgrade options.

      So it is actually kind of easier to upgrade your video card, which to me is a bit backwards but whatever. As it is unlikely that the PCI express interface is a lot less likely to change in the next 5 years compared to the various MB socket changes by both AMD and Intel.

      So while yes you are right, game PLAY is GPU limited, actually loading the game is still CPU dependent, and unless you enjoy sitting around forever waiting to play it might be better to get a decent CPU first, and when you need to, replace your video card.

      Anyway that's my personal experience. I could go out tomorrow and buy a 500$ video card, however to upgrade my CPU I would need not only buy a new CPU, but now a new MB, Memory, etc...

  91. Re:And 3 hours after reading this... by bberens · · Score: 1

    I feel more bad for suppliers/retailers who have stuff in stock. When companies semi-announce a future price drop from $800 to sub-$250 that has to kill sales. The only way those vendors/suppliers are going to sell the chips they bought for $800 is to sell them as if they had paid $250. I realize this is an age-old problem with technology, but it just struck me as a startling drop in this case.

    --
    Check out my lame java blog at www.javachopshop.com
  92. Re:That would be all nice and dandy if only... by pandaman9000 · · Score: 2, Insightful

    It doesn't help that Anandtech has become intel and Nvidia-biased of late.

  93. Re:And 3 hours after reading this... by beelsebob · · Score: 1

    Just for reference –AMD's next major upgrade is coming soon (no I don't mean fusion, I mean bulldozer) and will do away with Socket AM2/2+/3/3+ totally, so you're in the same boat there.

  94. Re:That would be all nice and dandy if only... by Machtyn · · Score: 1

    You know... I was feeling that way with my rig's really nice sound card and front panel dials and inputs. Thankfully, though, I've had need of them... I've finally taken my soundcard "off-road" in the past year. Of course, if utilizing the EAX or THX features of my card is off-road, then I've been taking it off-road for awhile now. My graphics card goes off-road everyday.

  95. Re:And 3 hours after reading this... by Archangel+Michael · · Score: 1

    So, what you're saying is that WP and SS are mature products.It happens

    --
    Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
  96. Re:That would be all nice and dandy if only... by Anonymous Coward · · Score: 0

    Lol, you said big loads

  97. Re:And 3 hours after reading this... by Anonymous Coward · · Score: 0

    Additionally, the memory on the board will be obsolete DDR2

    DDR3 LGA775 boards are available and has been since 2007, but you have to specifically look for them.

  98. Re:And 3 hours after reading this... by webheaded · · Score: 1

    how often do you really load up *all* cores at once running multiple desktop applications.

    Lets see:

    I've got three Java background apps running and unlike the Folding Client, they do not back off when I want to do something else. This means Firefox, Word, Outlook, One Note, XMPlay all have to fight them for any ticks on the CPU though I rarely see more then 50-75 avg. cpu loading. That's on an E6300 (1.8GHz) Core 2 running Win7-64 on 8GB and this is a typical situation for my system.

    My system is 3 years old and I've just started looking at upgrading but I have a problem. There are no CPU's now available from Intel that are compatible with my board and no a Bios update wont solve the problem. They changed the damn socket 6 months after I built it. Intel has a habit of changing things ever 6 months so you can't upgrade you CPU to gain the performance boost needed when the time comes. In my case, the only option if I could find one is a Q6600, which has already been discontinued (18 months ago) so I'm now forced to look at building a new system.

    Due to Intel's policy, I'm looking at AMD for my next system because they don't obsolete Sockets and Chips 6 months after you build the system, forcing you to buy the most chip you can afford and then replacing the entire system in two or three years when it can't keep up with demands. That's right. It's Intel that drives the business upgrade cycle because they can get more money from companies selling all new chips such as north/southbridge, nics and everything in between unlike AMD who prefers to see you buy more CPU's and gives us a gradual upgrade path by simply ensuring their new chips can run in at least the "+" series of sockets even though you may not have access to all features.

    This. A thousand times this. I got lucky and my friend's dad swapped me my E6600 and $100 for his Q6600. Otherwise, I'd be screwed, which severely pisses me off. At the same time, I've got an AMD machine for my media center and I'm free to upgrade with chips that have just come out. I built the media center a year or so after my Intel machine. I'd say I'm starting to sour on Intel, but for me, the performance is worth it on my gaming PC. I'm pretty much never going to use them for anything else though. PCs for friends, family, other uses around the house...they're going to be AMD. I'm tired of Intel's bullshit. They don't need a new socket type every few months. It's just getting ridiculous now.

    --
    "Those who would sacrifice essential liberties for a little temporary safety deserve neither liberty nor safety." - BenF
  99. Small Differences In Price Can Scale by ifiwereasculptor · · Score: 1

    Especially in third world countries, you'll find a US$25 difference can be huge. I know for a fact that in Brazil a US$70 chip is sold for US$150 because of absurd taxation. Now when you factor in the monetary convertion, you'll see that chip being sold for R$300. In a place where the average wage is R$1000, you'll find a US$25 difference in a motherboard to be huge. So AMD will have an enourmous advantage on these kinds of market. Funny thing is they tend to be all overrun with Intel processors, from what I've seen.

  100. Re:And 3 hours after reading this... by beelsebob · · Score: 1

    No it won't, AMD's fastest 4 core consumer chip (the X4 965) gets beaten handily by the i3 540 in pretty much every test –the only place it claws anything back is in heavy multithreading tests where it can *just* beat the i3.

  101. Re:And 3 hours after reading this... by rbarreira · · Score: 1

    AMD's fastest 4 core consumer chip (the X4 965) gets beaten handily by the i3 540 in pretty much every test

    You must be confused, or we're looking at very different benchmarks:

    http://www.anandtech.com/bench/Product/102?vs=143

    --

    The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
  102. Re:And 3 hours after reading this... by sexconker · · Score: 3, Insightful

    Intel could compete on price...

    That's for sure.

    From TFA:

    Core i7-970 processor, which is priced at $885 per 1,000 units

    Somebody is marking those things way up by the time they get to my local store.

    $899.99 through newegg.
    With a free piece of shit game no one cares about.

    $15 / $885 = 1.7%.

    So it's pretty obvious - INTEL is the one marking up these prices. No major shop pays anywhere close to that $885 figure. And no smaller shop has to either if they "just sign here" and agree to flog only Intel chips.

    My last few purchases have been Intel chips, because of the whole "Core 2 Duo > Anything AMD has" thing. But the prices have been jacked up sky fucking high, the sockets have changed way too often with no backwards compatibility, and the performance difference isn't all that great.

    And don't forget Intel's latest rapejobs:
    "Yes this chip has virtualization instructions. No you can't use them."
    "Download an upgrade to your CPU today! Only $49.99!!!"

    I'm going back to AMD, and I'm taking everyone I build / recommend for with me.

    If I really want my encodes to go faster, I'll buy a dual-socket mobo and drop 2 AMD cpus in there, and still save money.

  103. Re:And 3 hours after reading this... by sexconker · · Score: 1

    I feel more bad for suppliers/retailers who have stuff in stock. When companies semi-announce a future price drop from $800 to sub-$250 that has to kill sales. The only way those vendors/suppliers are going to sell the chips they bought for $800 is to sell them as if they had paid $250. I realize this is an age-old problem with technology, but it just struck me as a startling drop in this case.

    Do you really think the stores aren't given plenty of advance notice or buyback offers?

    And smaller shops don't order in large enough quantity to be hurt too much, and they don't have customers who know the MSRP of the specific CPU in the PC / parts bundle they're getting.

  104. Re:And 3 hours after reading this... by rbarreira · · Score: 2, Informative

    That is just complete bullshit, Bulldozer will use AM3+ and probably be compatible with AM3 (just as the AM2+ CPUs were compatible with AM2 motherboards).

    --

    The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
  105. 4X faster ? by Anonymous Coward · · Score: 0

    I currently use my 925 quad for video editing and audio creation, and even with multiple realtime Cubase amp sims it just purrs like a big kitten, the Radeon onboard was powerful enough I played SWAT 3 and Bioshock on it with decent framerates until my HD4650 arrived , and I've been selling AMD Neo based netbooks to those customers that were thinking of Atom. After getting their Neo and seeing how nicely it runs compared to an Atom all they do is rave, with the Radeon onboard making it a smooth multimedia portable.

    Thanks

    guitare occasion

  106. Re:And 3 hours after reading this... by MobileTatsu-NJG · · Score: 1

    Intel still charges insane money for their chips.. they only lower it to a fair price after AMD forces them ...

    This is why you don't want *either* company to 'win'.

    --

    "I like to lick butts!" by MobileTatsu-NJG (#32700246) (Score:5, Informative)

  107. Re:And 3 hours after reading this... by michrech · · Score: 1

    Why *wouldn't* you consider the mainboard as part of the cost?! The chip can't operate with all your components just by being in contact with them!

    --
    bork bork bork!
  108. Re:And 3 hours after reading this... by beelsebob · · Score: 1

    No, that benchmark bears out exactly what I said –the phenom wins in the multithreaded benchmarks, loses in all the less threaded ones, and costs $40 more. Now lets look at the intel chip you claim that phenom is on a par with...
    http://www.anandtech.com/bench/Product/102?vs=191

    Oh look, the phenom beaten by between 1 and 40% depending on the benchmark.

  109. Re:And 3 hours after reading this... by jtownatpunk.net · · Score: 1

    The upgrade path is key. I hemmed and hawed when I built a new gaming rig about 3 years ago. There was a some fairly convincing hype about how this new AM2 socket was the wave of the future and the key to a long upgrade path. I was skeptical but ended up going that way. I bought a top of the line motherboard with true dual 16 lane PCIe slots. I'm still using that board two processors and three video cards later. And I can still upgrade to the latest class of CPU because Asus keeps cranking out BIOS updates. I could even put a six-core CPU in there, tho I'd only be able to use 4. But that means I could upgrade the CPU now, save up some more money, then upgrade the motherboard and memory later. And, if I'm building a new system with an AM3 socket, I can put any CPU in there from a single-core Sempron to the best six-core they've got.

    AMD's management of the AM2/AM2+/AM3 socket progression has sold me on their commitment to system builders.

    Meanwhile, I look at Intel and they have THREE current sockets in the consumer market. What a mess.

  110. Re:And 3 hours after reading this... by dwinks616 · · Score: 1

    How often do you use a CPU without a mobo? Not only are comparable AMD motherboards a fair bit cheaper, but AMD doesn't require a new socket every time they release a new processor.

  111. Water is rising by snadrus · · Score: 1
    ARM, the "low water mark", has the A15 (not yet in production):
    • 4MB L2
    • Quad 2.5ghz
    • hardware virtualization support

    Power/heat wins causing mass-production for servers could reduce price. Plus licensees price compete.
    A Windows port would only need slow x86 emulation for non-.NET apps and could gain perf by using multiple CPUs during emulation, JIT, and mapping calls to native libs for performance.

    --
    Science & open-source build trust from peer review. Learn systems you can trust.
  112. It always amazes me... by supernova87a · · Score: 1

    Just as a side thought, amidst all this talk of how $900 is an obscene price for a 6-core processor --

    It always amazes me how much we have grown to expect the price of amazing things to approach mundane everyday objects. Just think about how little you get for $900 in some of the other things you buy. For $900, you could probably buy a leather couch, a piece of hardware that you yourself could probably build if given a few months, no experience, a hammer and some wood.

    Yet we still gripe about we can't believe how a 6-core processor is selling for the extortionate price of $900, a piece of hardware that took trillions of dollars in investment, many hundreds of thousands of people to develop, the great minds of our generations.

    By some measures, then, $900 is cheap. But of course, it's all relative to what you come to expect...

  113. Re:AMD One-Ups Intel? Another misleading Slash sto by Anonymous Coward · · Score: 0

    Ahh, but for the $180 Fry's was (is? May be over today.) selling them for it was a hell of a deal at the speed and price.

  114. Re:And 3 hours after reading this... by hairyfeet · · Score: 1

    Allow your old pal Hairyfeet to help you out with that buddy...Pow! And since according to the graph the ultimate space would be bottom right, you are looking at Phenom II on AMD side (which is what I got, yay me!) or the Core i5 on the Intel side.

    --
    ACs don't waste your time replying, your posts are never seen by me.
  115. Re:And 3 hours after reading this... by rbarreira · · Score: 1

    > -the phenom wins in the multithreaded benchmarks, loses in all the less threaded ones

    What do you mean by "all the less threaded ones", given that the phenom loses two or three benchmarks out of that whole list (at most)?

    > Now lets look at the intel chip you claim that phenom is on a par with...

    Seems much more even than the comparison with the i3 that you made...

    --

    The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
  116. Re:That would be all nice and dandy if only... by Anonymous Coward · · Score: 0

    Without the gigahertz wars, Netburst would not exist in the first place.

  117. Re:And 3 hours after reading this... by WebKing279 · · Score: 1

    I am glad to see ... there is someone who can demolish the monopoly of Intel. I have always admired the awareness of AMD processors while utilizing the resources. Great to see it coming back with force.

    --
    Regards, Sam I Love My Reliable Service Provider Do you?
  118. Re:And 3 hours after reading this... by beelsebob · · Score: 1

    You suck at reading stats. The i3 beats the phenom in 7 tests, and if you look at only the not-massively-parallel ones stays within 5% of the phenom.

    Meanwhile the i5 gets beaten only 3 times, and averages 20% faster than the phenom, in some tests even managing 60% faster!

  119. Re:And 3 hours after reading this... by rbarreira · · Score: 1

    beelsebob 4 hours ago: "the only place it claws anything back is in heavy multithreading tests where it can *just* beat the i3."

    beelsebob a few minutes ago: "if you look at only the not-massively-parallel ones stays within 5% of the phenom."

    Seems you're singing a quite different song now. Of course you had to, since the multi-threaded tests show a huge advantage (pretty obvious since we're comparing a quad-core vs a dual-core... hey it's not my fault, you were the one who brought the i3 into the discussion).

    --

    The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
  120. Re:That would be all nice and dandy if only... by Iverach · · Score: 1

    He forgot the intertubes!

  121. Re:And 3 hours after reading this... by beelsebob · · Score: 1

    Yet you're the one who claimed that you could get a 4 core phenom "on a par" with the i5 760 – yet the benchmark clearly shows it to be anywhere up to 60% faster.

  122. Re:And 3 hours after reading this... by wwfarch · · Score: 1

    I think PopeRatzo was under the impression that it was $885 for 1,000 units instead of $885 per chip when bought in batches of 1000

  123. Re:And 3 hours after reading this... by PopeRatzo · · Score: 1

    I think PopeRatzo was under the impression that it was $885 for 1,000 units instead of $885 per chip when bought in batches of 1000

    What else does "$885 per 1,000" mean?

    I was joking, of course.

    --
    You are welcome on my lawn.
  124. Re:And 3 hours after reading this... by haruchai · · Score: 1

    I doubt AMD can accomplish this alone - they just don't have the fabrication prowess and it's big money to build chip factories. I think the Global Foundries spinoff was a good move but they still too far behind Intel.
    However, Chipzilla is getting hit on several fronts and I never thought that the ARM would be the architecture to knock them back a step. And now that Marvell has announced a triple-core ARM(-like?) chip that'll run very-low power when not stressed, this might be another nail in Intel's coffin.
    But, on the PC desktop, they are the current champs, except for the price-performance category.
    If they swallowed Nvidia like AMD did with ATI, I think they would outright own the desktop again.

    --
    Pain is merely failure leaving the body
  125. Re:And 3 hours after reading this... by Nysul · · Score: 1

    Yep, I got an e6400 and with a bios update I can go to an e8400, maybe even an e8600. Except these chip prices have remained stable and overpriced for nearly 2 years and for $40 more I can get an i5 760, except now I need new memory and a new board.

  126. Re:That would be all nice and dandy if only... by tacarat · · Score: 1

    Why is it that porn sites aren't rated in hertz?

    --
    "Common sense will be the death of us all"
  127. Re:And 3 hours after reading this... by mr_mischief · · Score: 1

    I used to support 14,000 customers on two dual-processor single-core MX boxes with spam filtering, two tow-proc single-core SMTP outbound box with spam filtering (to catch our customers and keep them from spamming) and four single-proc single-core POP/IMAP boxes that accepted mail only from the MX or SMTP outbound boxes and did antivirus. Webmail was on a separate web server which was also the company's web site server and hosted a couple of specific clients.

    I used to support over 100k users at a different employer with four MX boxes with rudimentary spam filtering, four SMTP outbound boxes, four dedicated AV/anti-spam boxes, and 12 POP/IMAP/Perdition/webmail boxes. All of these boxes were dual processor single-core 3.06 GHz Xeons.

    Yeah, I get the concepts of percentage used, load average, RAM constraints, and per-user load.

    Guess what? I still like to have more cores when I'm doing something massively parellel like running 64 make jobs at once. That example is something I already mentioned, and can be done with GNU make's -j switch and a number of maximum jobs. It can also be done with a bare -j switch by itself, but that tells make it has no maximum, and on a big build it will make your machine pretty much unusable until the build is done, even on a quad-core with the preemptible kernel Linux configuration.

  128. Re:And 3 hours after reading this... by WebKing279 · · Score: 1

    Yes, that's true. Who has the power to spend can only lead the industry. i-core left AMD very much behind the race, but for time being AMD got something to hold on to. Now it just need the right time to get settled down with the latest technology and trend.

    --
    Regards, Sam I Love My Reliable Service Provider Do you?
  129. AMD is great by mrdtr · · Score: 1

    I've been using AMD for about 4 years now, and recently purchased a new desktop, and decided that AMD is just was the way to go. Intel products (are great) but expensive, and as the frugal type of guy I am, I just can't justify the extra cost for Intel products.

  130. and the winning guess... by smash · · Score: 1

    Was ALi...

    --
    I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
  131. Re:AMD One-Ups Intel? Another misleading Slash sto by indi0144 · · Score: 1

    Yeah, Everyone knows Word and Facebook benefits a LOT from HT and multicore so average computer user RUSH to the store so they can buy the latest and greatest and they finally get good FPSs in farmvile. Oh and p0rn is more snappier (pun pun pun) with 12 cores /sarcasm

    I simply take any benchmark as a grain of salt, it is well known the Intel optimizes their CPUs so they tweak themselves if they detect a benchmark application. Excuse me if I pass on the eV1agr4 I have better things in which spend the money for desktop components.

  132. Re:And 3 hours after reading this... by Anonymous Coward · · Score: 0

    Sure they can, at least I can smear my i7 on my face and get the pimps gone, I smear the whole mobo in my back to treat my backne. And I completely replaced viagra just by looking at very high deff pics of the i7 internals. AMD is the fa1lxzorz olololololololol