Slashdot Mirror


AMD FX-8350 Review: Does Piledriver Fix Bulldozer's Flaws?

An anonymous reader writes "AMD just officially took the wraps off Vishera, its next generation of FX processors. Vishera is Piledriver-based like the recently-released Trinity APUs, and the successor to last year's Bulldozer CPU architecture. The octo-core flagship FX-8350 runs at 4.0 GHz and is listed for just $195. The 8350 is followed by the 3.5 GHz FX-8320 at $169. Hexa-core and quad-core parts are also launching, at $132 and $122, respectively. So how does Vishera stack up to Intel's lineup? The answer to that isn't so simple. The FX-8350 can't even beat Intel's previous-generation Core i5-2550K in single-threaded applications, yet it comes very close to matching the much more expensive ($330), current-gen Core i7-3770K in multi-threaded workloads. Vishera's weak point, however, is in power efficiency. On average, the FX-8350 uses about 50 W more than the i7-3770K. Intel aside, the Piledriver-based FX-8350 is a whole lot better than last year's Bulldozer-based FX-8150 which debuted at $235. While some of this has to do with performance improvements, that fact that AMD is asking $40 less this time around certainly doesn't hurt either. At under $200, AMD finally gives the enthusiast builder something to think about, albeit on the low-end." Reviews are available at plenty of other hardware sites, too. Pick your favorite: PC Perspective, Tech Report, Extreme Tech, Hot Hardware, AnandTech, and [H]ard|OCP.

259 comments

  1. How about idle?? by Anonymous Coward · · Score: 5, Interesting

    90+% of my CPU is idle time.

    How much power does the new chip use at idle and how does that compare to Intel?

    50W at the top end means about $25/yr if I was running it 24/7. But since typical desktop is idle, what is the power difference there??

    And yes, I don't care about single thread performance as I care about multithread performance. Single thread performance has been good enough for desktop for almost a decade, and the only CPU intensive task I do is running those pesky `make -j X` commands. No, not emerging world or silly things like that ;)

    1. Re:How about idle?? by Animal+Farm+Pig · · Score: 5, Insightful

      I agree about multithreaded performance being important thing moving forward.

      Regarding power consumption, anandtech review puts total system power consumption for Vishera tested at 12-13W more than Ivy Bridge. Scroll to bottom of page for chart. Bar and line graphs at top of page are misleading-- they put x axis at 50W, not 0W.

      If you are concerned about power consumption, find 100W lightbulb in your house. Replace with CFL. You will have greater energy saving.

    2. Re:How about idle?? by ifiwereasculptor · · Score: 4, Informative

      Idle power seems pretty competitive with Intel's Core offerings. Anand found little difference and attributed it to their selection of a power-hungry motherboard.

    3. Re:How about idle?? by ifiwereasculptor · · Score: 2, Informative

      So true. AMD isn't competitive energetically in any way anymore, but the desktop is probably the only place while it doesn't matter. When you're thinking mobile, saving energy is a priority. On huge server farms, little relative gains can mean a tangible different in absolute numbers. But on desktops, their difference is about a light bulb, at load, and hardly anything when idle. And with PSUs under 350W being incresingly harder to find and an FX-8350-based system only gobbling about 200W at its most intensive, it's probably "low enough" power consumption for most people.

    4. Re:How about idle?? by viperidaenz · · Score: 2

      Its not just 50W at full load, there is another graph that shows the same workload on Intel used 220.8watt-hours, the AMD took 352.5watt-hours. Not only did the Intel system use less power under load, it finished the work quicker.

    5. Re:How about idle?? by amorsen · · Score: 1

      If you are concerned about power consumption, find 100W lightbulb in your house. Replace with CFL. You will have greater energy saving.

      Surely you already did that 3 years ago, so now you're preparing for the switch to LED?

      --
      Finally! A year of moderation! Ready for 2019?
    6. Re:How about idle?? by Anonymous Coward · · Score: 0

      You know they just labelled all the 350w power supplies 500w, right?

      And that they don't draw that much at idle?

    7. Re:How about idle?? by war4peace · · Score: 4, Insightful

      I play games maybe 1h 30m a day on average. My 5 year old dual-core E6750 overclocked at 3.2 GHz handles most of them gracefully, but there are some new releases which require more processing power. However, in choosing a new platform, I'm mostly looking at TDP, not from a consumption perspective, but heat dissipation. I hate having to use a noisy cooler.
      My current CPU has a TDP of 65W and a Scythe Ninja 1 as cooler, and the fan usually stays at 0% when the CPU is idling. While gaming, I can't figure out whtehr it makes noise, because my GPU cooling system makes enough noise to cover it. And I'd like to keep it that way when I pick my new CPU.

      You're saying that graphs are misleading. No, they're not, if one has half a brain. I'm not looking at the hard numbers and the power consumption difference is of about 100W. The i5 3570K draws about 98W and Zambezi and Vishera (who the fuck names these things?) draw around 200W. if you put TWO i5 on top of the other, they barely reach ONE AMD cpu power consumption. Thanks, but things DO look bad for AMD. I'll just have to pass.

      --
      ...gis sdrawkcab (usually not responding to ACs; don't bother posting as AC)
    8. Re:How about idle?? by Lumpy · · Score: 3, Insightful

      "Surely you already did that 3 years ago, so now you're preparing for the switch to LED?"

      Why? so I can get less light output for the same watts used but instead of spending $8.95 per bulb I get to spend $39.99?

      LED is a joke for home lighting, only fools are buying it right now. CFL is still way more efficient.

      --
      Do not look at laser with remaining good eye.
    9. Re:How about idle?? by Kjella · · Score: 3, Insightful

      I think the biggest factor for your home desktop is noise - it takes a lot more airflow to remove 125W of heat than 77W of heat. In Anandtech's tests he actually measures 195W versus 120W total system power consumption. Sure it might not matter much if you plan to put a noisy 200W+ graphics card or two in it, but for non-gamer use I'd say that's pretty significant.

      --
      Live today, because you never know what tomorrow brings
    10. Re:How about idle?? by Anonymous Coward · · Score: 0

      Fools and people with powerline networking. Some CFLs cause interference with powerline networks whereas LEDs are fine. I've also had several CFLs release nasty substances while they run turning my light fixture yellow and covering the bulb along with burn marks around the base. Before you ask, they were GE bulbs.

      I might be a fool to use LED, but I feel safer doing so. If you leave lights on for long periods, some CFLs aren't that safe.

    11. Re:How about idle?? by amorsen · · Score: 1

      Why? so I can get less light output for the same watts used but instead of spending $8.95 per bulb I get to spend $39.99?

      LED can come much closer to proper full-spectrum light than CFL ever will. CFL is just a stopgap technology we have to deal with until LED gets there.

      Also, it is possible to make LED spotlights to handle that strange modern trend of building lots of spotlights into ceilings. CFL cannot do that.

      --
      Finally! A year of moderation! Ready for 2019?
    12. Re:How about idle?? by Anonymous Coward · · Score: 0

      Fools and people with powerline networking. Some CFLs cause interference with powerline networks whereas LEDs are fine. I've also had several CFLs release nasty substances while they run turning my light fixture yellow and covering the bulb along with burn marks around the base. Before you ask, they were GE bulbs.

      I might be a fool to use LED, but I feel safer doing so. If you leave lights on for long periods, some CFLs aren't that safe.

      Or, in the absence of a meddling government, you could just use a 50 cent incandescent with none of those problems.

    13. Re:How about idle?? by Anonymous Coward · · Score: 0

      total system power consumption for Vishera tested at 12-13W more than Ivy Bridge. Scroll to bottom of page for chart. Bar and line graphs at top of page are misleading-- they put x axis at 50W, not 0W.

      Umm.. 12W difference is for an IDLE system. At load the difference is clearly closer to 80-100W. Maybe for a desktop this doesn't matter. But given the decline of desktops and consumers moving to laptops (and even more mobile devices), these results are downright TERRIBLE for AMD.

    14. Re:How about idle?? by h4rr4r · · Score: 1

      And spend more money than the cost of the bulb to keep it lit, and I get to replace them all once a year.

      I loved CFLs, but LEDs are all I buy now. Why bother changing lightbulbs?

    15. Re:How about idle?? by h4rr4r · · Score: 1

      You do realize they last like 20 years, right?

      Please explain why you think someone would be foolish to buy a longer lived and nearly as efficient device?

    16. Re:How about idle?? by Anonymous Coward · · Score: 0

      I have to agree, AMD needs to take what they have in a 135watt part and make it 80 watts or less for them to be competative, otherwise all these parts are only going to end up in corporate desktops where they all go to sleep after work, and places where electricity is cheap (eg mid-west)

      These aren't going to end up in data center machines due to to being way too hot/power-hungry. Even the APU A series are too hot to be used effectively in a desktop or laptop, and only become a 'win' in a few very specific use cases (eg netbook/ultrabook, media centers) where the intel's video is a complete joke.

    17. Re:How about idle?? by Anonymous Coward · · Score: 0

      I use nothing but LED lights. I replaced them all when I moved in, and they run for 16 hours a day for 2 years, and they're they same brightness as when I bought them. They beat CFL's handsdown, and completely kill Incandescent lights.

    18. Re:How about idle?? by ifiwereasculptor · · Score: 1

      True. Especially for hotter cities. Here, My undervolted, underclocked 95W Athlon II already manages to annoy me with its 3500+RPM fan (it goes to about 5000RPM at full load and clock, it's madness). Maybe AMD should start putting coolers with 120mm fans on their boxed processors. If they plan on selling such behemoths, that would provide more airflow and less noise at the same time. Fitting that on a motherboard would likely be a problem, though. Water cooling would be good, but it's much too expensive to make sense.

    19. Re:How about idle?? by squiggleslash · · Score: 1

      I still have CFLs that I bought in 1998 working properly - that's when I moved to the US. The few CFLs I've had to replace have generally done so for external reasons (such as a bizarre one I bought that started to fall out of its base. I assume it wasn't glued together properly or something. Bizarre both in the sense that it did it, and that it still actually seemed to work OK.)

      I'd be very surprised, at current prices, if LEDs actually represent value for money against CFLs if the quality and capabilities of both types of "bulb" are close enough to not be a consideration for the task at hand (like lighting a living room.)

      --
      You are not alone. This is not normal. None of this is normal.
    20. Re:How about idle?? by negRo_slim · · Score: 4, Informative

      I think the biggest factor for your home desktop is noise - it takes a lot more airflow to remove 125W of heat than 77W of heat.

      Larger fans with slower rotational speed.

      --
      On the Oregon Cost born and raised, On the beach is where I spent most of my days
    21. Re:How about idle?? by negRo_slim · · Score: 2

      Get an aftermarket cooler if CPU fan noise is really a concern.

      --
      On the Oregon Cost born and raised, On the beach is where I spent most of my days
    22. Re:How about idle?? by Bryansix · · Score: 1

      I'm fairly certain the light output is basically the same per watt for LEDs compared to CFL. In fact here, I looked it up http://cleantechnica.com/2011/09/01/led-vs-cfl-which-light-bulb-is-more-efficient/ and the LED's actually produce MORE lumens per watt.

      Now, lets expand and point out something this article got wrong. They say you would only have to replace a CFL three times over the course of an LED lifespan. In my experience, this is at least an order of a magnitude on the low side. I have dimmers in all my sockets and the dimmable CFL lights will fail sometimes within a year to a year and a half. In that case, I'd be replacing 15-20 CFL lights per LED light I put in. Nevermind that CFL has a serious issue with diminishing light quality over the lifespan. Yes, there are CFLs without these problems, and they are 50-100% more expensive then the normal ones.

    23. Re:How about idle?? by Bryansix · · Score: 1

      And then I would be charged in the highest tier for my Electricity ( I live in California where electricity is expensive) and then that would be compounded by the fact that incandescent produces massive amounts of heat which would raise my electricity bill even further to cool down. Again, I live in California where I need to use A/C 9 months out of the year. I'm glad I don't use incandescent.

    24. Re:How about idle?? by Animal+Farm+Pig · · Score: 1

      Umm.. 12W difference is for an IDLE system. At load the difference is clearly closer to 80-100W.

      No shit, Sherlock. Look at comment subject line.

      Maybe for a desktop this doesn't matter. But given the decline of desktops and consumers moving to laptops (and even more mobile devices), these results are downright TERRIBLE for AMD.

      If you think desktop chips (AMD or Intel) are used in laptops, you are an idiot.

    25. Re:How about idle?? by cyberjock1980 · · Score: 1

      Same here. Used to buy CFLs, but switched to LEDs because they never burn out. Now I have a single circuit in my whole house with a single UPS(300W) for lighting. If I lose power all the lights stay on for about 10 minutes. Very useful at times. Not to mention they never burn out. I have yet to have an LED fail me and my whole house has been LED since 2009.

      Do I think I'll make my money back someday? Maybe, maybe not. But I'm glad to have given money to a technology that needs to become more common and get cheaper, so why not invest in it? If everyone is buying LEDs for their homes in 5 years then I'm glad to have done my very meager but useful part to help make LEDs a daily reality.

    26. Re:How about idle?? by PopeRatzo · · Score: 1

      My undervolted, underclocked 95W Athlon II already manages to annoy me with its 3500+RPM fan

      Get a better fan.

      --
      You are welcome on my lawn.
    27. Re:How about idle?? by hoboroadie · · Score: 1

      We only use incandescent heaters for illumination in the studio. Stratocaster pickups hum plenty already.

      --
      They feared that it could be used to suppress protest or support unpopular rule.
    28. Re:How about idle?? by LordLimecat · · Score: 2, Insightful

      You live in an apartment and dont plan to be there for 20 years?

      I imagine for a lot of people, dumping $40 into each light socket is a losing proposition for you, and a winner for your landlord (who I am sure would greatly appreciate the gift).

    29. Re:How about idle?? by Anonymous Coward · · Score: 0

      Moron.

    30. Re:How about idle?? by jgrahn · · Score: 0

      And yes, I don't care about single thread performance as I care about multithread performance. Single thread performance has been good enough for desktop for almost a decade, and the only CPU intensive task I do is running those pesky `make -j X` commands.

      You don't mean multithread, exactly. Make isn't multithreaded; it uses real processes.

    31. Re:How about idle?? by afidel · · Score: 2, Insightful

      LED's running on AC will fail for the same reason most CFL's fail, the ballast.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    32. Re:How about idle?? by Anonymous Coward · · Score: 1

      A Bing link, a Hotmail email address, and you misspelled Zalman?

      Are you sure you're on the right site?

    33. Re:How about idle?? by zenith1111 · · Score: 1

      That may be true, but all the good quality fluorescent bulbs I've bought lately have a color quality at least as good as the good quality LED ones and with very similar power efficiency.

      I've found a place were LEDs are superior, tough, we have a corridor with a motion sensor with 5 bulbs and, with dozens of daily starts, some CFLs don't even last 6 months there, others quickly get very dark near the heating filament. I replaced them with cheap LED bulbs and they are going strong after one year.

    34. Re:How about idle?? by Anonymous Coward · · Score: 0

      The whole point it to reduce pollution in the long run. How does being wasteful with power help? f'n tard.

    35. Re:How about idle?? by tftp · · Score: 5, Insightful

      Light bulbs are easily removable, and there can't be that many of them in a rented apartment - or even in a rented home. Take them with you, since you will need them at the new place anyway, and leave the old ones (that you saved) in their place.

    36. Re:How about idle?? by afidel · · Score: 1

      What?!? There are CFL's with a CRI of 95, where is the LED bulb that competes? Oh, that's right they're the same, because they use the same phosphor based system to create full spectrum light.

      As far as the use in cans, both LED's and CFL's have the same problem with overheating ballasts, unless you have DC power run to your LED can light you'll run into the same power and a similar lumen ceiling that you do with CFL's in that application.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    37. Re:How about idle?? by Anonymous Coward · · Score: 1

      Except LEDs use a much simpler solid-state resistor based ballast. Might as well compare a F1 engine's durability to a diesel engine.

    38. Re:How about idle?? by epyT-R · · Score: 0, Offtopic

      yeah you're overcharged because of a meddling government.

    39. Re:How about idle?? by Anonymous Coward · · Score: 0

      And then I would be charged in the highest tier for my Electricity

      Huh? How does a light bulb change what you pay per KWh?

      I live in California

      Oh.

    40. Re:How about idle?? by guises · · Score: 4, Insightful

      Well if that's no good, how about this: http://www.silentpcreview.com/Ninja2

      This is what I use on my Athlon 2, works perfectly, is very quiet, and it's rather old now so you could probably pick a used one up pretty cheap.

    41. Re:How about idle?? by Lumpy · · Score: 3, Insightful

      Because the incandescent ban is for the old out of date crap that sucked and worked better as a heater than a lightbulb. Halogen bulbs are the replacement. Who did you get your education on the ban from? Because they did not know what they are talking about, you should stop listening to the uneducated news network.

      --
      Do not look at laser with remaining good eye.
    42. Re:How about idle?? by Lumpy · · Score: 1

      Really, what ones have lasted 20 years, Because I have found none that do. Note: I have worked for a LED bulb distributor. the return rate is nasty high. They last about 1-2 years on average. Warranty replacement rates are high as hell on them.

      --
      Do not look at laser with remaining good eye.
    43. Re:How about idle?? by Anonymous Coward · · Score: 1

      And spend more money than the cost of the bulb to keep it lit, and I get to replace them all once a year.

      I loved CFLs, but LEDs are all I buy now. Why bother changing lightbulbs?

      If you have to change your incandescents once per year you're doing it wrong.

      Conversely, check the TCO on CFL's once you consider they die within a year or two when you just flip them on for a few seconds at a time in a closet (or other similar scenarios). Are people coming out ahead on those high cycle rate -> early CFL death deals? I doubt there's much wasted energy on brief use of an incandescent, say, in a closet because the light's only on for a few seconds or minutes. Never going to recapture the upfront cost of a CFL or LED there. To add insult to injury, the CFLs burn out much more frequently than incandescents in these scenarios.

      Besides, all that additional landfill waste (with mercury, natch) from early death CFLs can't be good for the environment.

      Just say no to meddling government. I should be allowed to decide what bulb type I want to use & where, based on my own assessment of my needs.

    44. Re:How about idle?? by hairyfeet · · Score: 4, Informative

      The problem, and as someone that has been selling AMD exclusively since I found out about Intel bribery and compiler rigging really hurts, is they are going ass backwards when it comes to design. they are betting everything on having as many cores as they can, even when the cores are weak, when most people frankly aren't even slamming a dual or triple, much less a hexacore.

      You see the problem is this: When we had the MHz wars it was easy for programs to take advantage of better single core performance but now that we've (IMHO) gone off the rails with number of cores what we are seeing is frankly a LOT of the software people use every day? Really doesn't spawn that many threads. Heck even games have trouble using more than a couple without running into sync issues.

      Frankly a better choice would have been less cores but give those cores crazy single threaded IPC, in fact I would argue the "perfect CPU" but be one with one or two insanely powerful cores and then 4 to 6 lesser cores to handle background tasks and an OS who understood the layout so it could schedule appropriately.

      Instead what AMD did with Bulldozer was to give us a huge number of really weak underperforming cores, in fact it reminded me of Intel's Larrabee GPU in that sticking a whole bunch of weak CPUs was supposed to be "good enough". So while I'm glad AMD seems to be pulling back from the weak cores somewhat I find it telling that on the 2 links I picked at random it looks like nobody is testing it against Phenom II anymore because i have a feeling full cores will still do good compared to the half core designed used in BD/PD.

      So while I do hope AMD rights the ship until i can say across the board "This chip is in every way better than Phenom II" I'll be sticking with the AM3+ Phenom IIs in my builds. The prices are better, Frankly the Athlon quads have been going for $80 and the Phenom Hexacores for $110 so the bang for the buck is there, and they give the average user more cores than they can use at a great price. While its good that PD isn't using more power than BD frankly BD was a hot chip to begin with so that's not saying much, and frankly even heavy multitaskers like myself have trouble keeping a 6 core fed, even a lot of video encoding software don't support 6 cores yet. Until software catches up 8 cores, heck probably 6 cores like mine, will be total overkill and when the software does catch up the changes in CPUs will make the 6 and 8 cores released today just not good. I just don't see a selling point for the BD/PD arch AMD, sorry but I don't.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    45. Re:How about idle?? by Anonymous Coward · · Score: 0, Troll

      And, as I've stated elsewhere, it should be up to me to assess my needs. If the government has to meddle in order to choose a winner (or, in this case, a loser) that's not really a strong endorsement of the competing technology.

      As to your point about halogens: please point me to an A19, 60 to 100 watt halogen bulb that costs a dollar or less. Oh wait, you can't. I'm well aware of the actual US ban (*cough* "efficiency standards"), so don't make presumptions.

      Stop drinking the statist kool aid: incandescents are the best choice in certain everyday scenarios—and this is dealing solely with TCO, without even addressing subjective complaints about alternative technologies.

    46. Re:How about idle?? by incer · · Score: 1

      Yeah, but desktops are being replaced by notebooks and tablets, so power consumption is very important nowadays!

    47. Re:How about idle?? by Lonewolf666 · · Score: 1

      CFLs and LEDs are still better in terms of efficiency than halogen bulbs.

      Now I had mixed results with CFLs myself, depending on brand (I'm living and buying the things in Germany). Osram are quite OK so far, I had only one die on me and that after several years. Megaman have truly disappointed, 3 out of 4 broken within a few months.

      Gonna try LED next, even if they are a bit more expensive. I've seen a 60W equivalent with a pleasant semi-warm white light, and at 18 Euros it is not excessively expensive :-)

      --
      C - the footgun of programming languages
    48. Re:How about idle?? by DragonWriter · · Score: 1

      Also, it is possible to make LED spotlights to handle that strange modern trend of building lots of spotlights into ceilings. CFL cannot do that.

      If you have a lightsource and an appropriately curved reflector, you have a spotlight. This can definitely be done with CFLs. I'm pretty sure of that, since my house has those ceiling spotlight mounts (originally with halogens, but replaced with CFLs when we bought the house.)

    49. Re:How about idle?? by Fjandr · · Score: 1

      There are many areas of the USA where your cost per KWh go up depending on how much you use per month. There are even places you can get seasonal or time-of-use metering. The latter are great for those who produce their own solar power, since rates are highest during midday in the summer, which is the time you're actively producing more power than you use. A prime location for that is California. :)

    50. Re:How about idle?? by Lonewolf666 · · Score: 2

      Compared to the Phenom II in 40 nm, I guess that Piledriver is finally faster. There were a few benchmarks where a Phenom II X6 could beat a Bulldozer, but IIRC only by a few percent. Which is not enough to beat the Piledriver in the same tests.

      Now I still wonder how a Phenom II in 32 nm would have performed. That hypothetical chip might still embarass the Piledriver ;-)

      --
      C - the footgun of programming languages
    51. Re:How about idle?? by hairyfeet · · Score: 1

      That is actually quite easy and cheap to fix friend. I would recommend the same thing my family and I use, the Hyper N520 which is small enough to fit in most mATX mini-towers, and paired with a little arctic silver will keep even the hottest chips under control without the whole "jet engine" problem.

      My 1035T which is naturally a hotter chip than yours, having 6 cores and Turbocore, idles around 85F-95F, depending on how hot i have it in my apt at the time, and after hours of video transcoding it maxes out around 128F, which is really good for just a $30 cooler and a $5 tube of arctic silver, especially since i removed the exhaust fan as I saw no point with the temps i was getting. I tried using the tube of paste that comes with it on my nephew's quad and it only adds 5f-7F so its up to you if you want to buy the arctic for that little difference, but I figured if i was gonna have the cooler changed out having some arctic around would probably be nice to have and it was on sale.

      But once installed frankly you can't hear it unless you just lay your ear next to the case, in fact when i installed the N520 into my youngest boy's 1045t gamer PC he called me back in there before I could pack up my tools saying 'I think the fans are busted!" and I had to pop the side off to show him yes, the fans are working just fine, because it was so quiet he was sure there had to be something stopping the fans from turning. like you he had an AMD stock cooler and that thing was a noisy little beast,that is why I urge folks on my new builds just to spend the extra money and get the N520, the drop in noise and temps is well worth it.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    52. Re:How about idle?? by fnj · · Score: 1

      That's more than a bit pedantic. A process is one or more threads, with private memory not normally accessible from other processes. Multiple processes are ALWAYS involve multiple threads. Gnu make is most assuredly multithread capable. The point that the threads are also processes does not change that fact.

    53. Re:How about idle?? by Anonymous Coward · · Score: 0

      So while I'm glad AMD seems to be pulling back from the weak cores somewhat I find it telling that on the 2 links I picked at random it looks like nobody is testing it against Phenom II anymore because i have a feeling full cores will still do good compared to the half core designed used in BD/PD.

      One of the reviews does test the new chip against the Phenom II lineup
      But I won't tell you which one, because laziness shouldn't be rewarded.

    54. Re:How about idle?? by ifiwereasculptor · · Score: 1

      I know it can be fixed and thank you for your suggestions, but there are two complications: a) ambient temperature here goes up to 36 degrees Celsius, which means I need lots of airflow; b) import taxes render aftermarket coolers prohibitive. A $30 cooler is sold here by about $65, which is almost what I paid for the processor itself. I could drop $30 in a Hyper 101, but given it's still an Athlon, an upgrade is just around the corner. So I'll just hang in there for only this summer, then I'm going for a 65W Trinity or 55W Ivy.

    55. Re:How about idle?? by nazsco · · Score: 1

      > noise
      that's because you put your $3000 parts into a $15 case that you've been reusing for 27 years...

      i did that too. until 3 years ago i got an antec case for almost 200, a massive cooler and several high end larger fans, all obscenely priced --- and moved to raid0 instead of 6 :(

      it's now totally silent. and i live in a place where it makes 80F and i don't like A/C

    56. Re:How about idle?? by Anonymous Coward · · Score: 0

      Has any ever stopped to think about the use of "land lord". What are we in the medieval ages? Yes me lord.....

    57. Re:How about idle?? by Osgeld · · Score: 1

      Thats cause most LED bulbs are at best wishful thinking in how they are powered

      "toss a cap in there and let those fuckers rectify the mains, good enough" mentality you see in LED household bulbs is not nearly enough, considering you can get electronics from the 70's and their LED's still work fine

    58. Re:How about idle?? by hairyfeet · · Score: 1

      Personally I prefer the Hyper N520 for both AMD and Intel builds, why? Its quiet, fits into the mATX towers, unlike many of the heatpipe coolers, and with a little arctic silver (although to be fair it comes with coolermaster heat paste in a tube and its only around 5f difference between the two) it keeps even the hexacores nice and cool without being loud.

      The problem with those like the ninja, and having a couple of gamer customers that love having me build them crazy gamer rigs I've gotten to play with most of the high scoring coolers, is they are fricking HUGE and will have trouble fitting into anything but the really large gamer cases. Since the vast majority of PCs come with the mATX mini-towers its nice to have a cooler that will do the job without having to toss the case and the N520 fits that description perfectly. In fact with most of my customers (as well as myself) that go for the N520 I don't even bother with an exhaust fan, the N520 will push enough air that frankly it isn't needed. We are talking 85-105F light load on the Hexacores, with 128F seeming to be the highest they'll hit after hours of heavy loads, and you can knock 5-12F for the quads and triples, just depending on the model.

      Truly a great little aftermarket cooler at a great price, i highly recommend and like it so much not only is my own Hexacore running it but so is my dad's quad, and my boys quad and hexacore. All of us are quite happy with how low it keeps the temps and how quiet it is, just great.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    59. Re:How about idle?? by Anonymous Coward · · Score: 0

      The original question is more about average power consumption for a given workload.

      Say, you save $50 or $100 by buying AMD vs. Intel. Now, you run it at a given power profile for 5 years or 10 years or whatever. Assuming a low 5% discount rate (for capital investments), cost of the AC and heat per W of heat, does it make sense to buy AMD or Intel?

      This all may seem trivial, but multiply $25 overall price difference by 1000 units, or 10,000 units.

      This is not about how much money you save on power. Or how much you save on a chip upfront. It is what is the total cost of owership of a chip for a given workload profile and power costs. Sometimes paying $50 more for a CPU will save you money. Sometimes, it will not. It all depends on how you use it ;)

    60. Re:How about idle?? by hairyfeet · · Score: 1

      The problem is look at the prices, last I checked the only way that BD/PD beat the Phenom II Hexacore was with the Octocore, the BD/PD Hexacore being beaten by a few percent. so you have to buy the Octocore to beat the Phenom II hexacore, and you have to buy at least the BD/PD hexacore to beat the quad, again because of lower per core performance.

      What's the problem with that? because with the higher cost to make the BD/PD chips frankly they have to sell them at too high a price. You can get the Athlon II triple for $60, quad for $80, Phenom II quad BE for $90 and the Hexacores for any but the BE are selling for $110. Meanwhile the FX8350, which as you noted is ONLY winning by a couple of percent, is $195, and the Hexa and quads are $132, $122. so you have to spend $195 to beat a $110 Phenom II, $132 to beat a $90 BE quad, and frankly they don't really have anything that competes with the Athlon price points anymore.

      One of the reasons I've been sticking with AMD, besides the fact that the Intel bribery and compiler rigging left a bad taste in my mouth, is I have a LOT more normal customers than I do gamers. We are talking office workers, home users, small business, even some using their PCs in home theater setups, and in all those areas frankly even the Athlon triples give more performance than they are ever gonna really need, with the quads and hexas being super overkill.

      The real bitch, one that is seriously hurting AMD, is they are reportedly having serious issues with the yields on the top end chips, yet all the benches show you really need those top end chips to beat the previous Phenom II. You see the bitch i bet you don't know is because they had so much time to refine the Phenom series that they were looking at near 100% yields and thanks to the way they had the SKUs set up they were having pretty much ZERO wasted chips. Phenom X6 have a bad core or two? Make it an X4. Three bad cores? An X3, bad cache? An Athlon. But because they were getting such good yields from the chips frankly most of their lower SKUs weren't bad chips, they just had them turned off to meet a quota, so guys like me were seeing 70%+ when it came to turning on disabled cores! It made the bang for the buck just insane.

      So while I hope that AMD is able to fix the chips and right the ship ATM there is really no reason for me to switch from Phenom II and Athlon II to BD/PD chips, it just doesn't give any advantages to my customers. To beat the Phenom IIs you have to have more cores than my users will ever need, and frankly most of them won't even be able to stress that $60 Athlon triple, much less that $110 X6. Heck my youngest is gaming on my hand me down 4 year old 925 Phenom II and it spends more time idling than anything, and that is with him playing those large MMOs.

      We just have more power than we know what to do with now, throwing more cores at it just doesn't change that.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    61. Re:How about idle?? by hairyfeet · · Score: 1

      Ouch, they are stickin it to ya huh? But I don't think you are looking at this the right way friend, see you can just save the instruction booklet and the old HSF and then move the new cooler to whatever unit you get later, as the new coolers will fit pretty much ALL Intel and AMD chips. So you see you can get this now and when you go Trinity or Ivy down the road you can just take that cooler with you, no need to replace it when you replace the chip. I have a customer with one of the Zalman coolers, he's changed chips 3 times, still on the same cooler.

      And if you need something a little lower cost this unit works pretty well, about 5-7F hotter than the N520 but its also half the price so with your import taxes it might save you a pretty penny. As you see here its also a heatpipe but with the added advantage of this baby will fit ANYTHING, from Socket 754 on the AMD side to LGA 775 on the Intel, so it really doesn't matter what system you get down the road, just keep the adapters and instruction booklet and take the cooler with you.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    62. Re:How about idle?? by Anonymous Coward · · Score: 0

      Get a better heat sink.

      I use this heatsink (ZALMAN CNPS 8900 Quiet, low profile, $40) with a couple different AMD Phenom II boxes (125W TDP). The CPU runs on average 10-15 degrees cooler (celcius) than with the stock heatsink. It's incredibly quiet at full load and imperceptible otherwise. It's quieter than the heat sink that comes with the Intel i7-3770 (77W TDP).

      If I didn't need a low profile heatsink I could have spent half as much and gotten something that would have worked just as well or better.

    63. Re:How about idle?? by hairyfeet · · Score: 1

      Uhhh...They don't have a Fred's where you are at? I get incandescents at Fred's, 4 for $1.10 I believe was what i paid for the last batch and I use them for just what you describe, places like the closet and bathroom where they are only gonna be on a few minutes while using the CFLs in places like the living room where the light is on longer.

      I can't really tell any difference in the number of times i change them because as you noted CFLs don't do well with "flip on and off" situations either, and the bulbs are cheap enough i really don't care.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    64. Re:How about idle?? by Anonymous Coward · · Score: 0

      .

      Also, it is possible to make LED spotlights to handle that strange modern trend of building lots of spotlights into ceilings. CFL cannot do that.

      They make CFL PAR 40 bulbs for recessed lights/can lights/ or "spotlights". I don't get why you say a CFL can't do that because the only problem with either a CFL or LED is an overheated ballast. I have 30 CFL recessed lights and they last about 5-7 years (I've lived here for 10) as I've only replaced each bulb once. The possible difference may be that my can lights don't have covers on them so there is some air movement. Really, don't seal your lights and you can stick whatever you want in there.

      Where are you getting your information from?

    65. Re:How about idle?? by Anonymous Coward · · Score: 0

      Most people that return them probably stuck them in the wrong kind of fixture, as in no airflow or exposed to the elements.

      Was there any diagnosis on the failures? I know that as a distributor you most likely remove yourself from this and just issue replacements or refunds and that's the end of it, but I'm willing to bet that 80% of failures were "overheated ballast". You should ask the manufacturer.

    66. Re:How about idle?? by hairyfeet · · Score: 1

      Well I found it and on the few tests where the new chips win they are looking at just a few percent, maybe 7% on the tests practically design for PD.

      Hmmm...less than 7% on a few cases, and even or below the margin for error on the rest...for a chip that costs $70+ more than the Phenom II X6? I think I'll pass, the Thuban is the better bang for the buck.

      They have already announced we AM3+ owners will be getting one more chip, excavator, so I think I'll wait. Personally the fact they are gonna put out another AM3+ chip looks like a bad sign to me, as dual channel memory has to be a bottleneck but its like someone at AMD is saying "Look these chips just won't sell with such MOR performance if we make them buy a new board as well and the AM3+ boards are dirt cheap, best stick with it until we can get something better".

      --
      ACs don't waste your time replying, your posts are never seen by me.
    67. Re:How about idle?? by ifiwereasculptor · · Score: 1

      A very valid point. It honestly hadn't occurred to me to look at it from this angle, since I'm already counting on my next CPU's stock cooler being quieter and eliminating the problem. But a better cooler, with heatpipes, is always a plus, and it does linger. I'll check its orientation (my airflow is front to back, not top to bottom) If it works well and quietly, I might just get a 100W Trinity. Thanks.

      (Also, your specific suggestion was very helpful. Turns out a TX3 costs about the same as a 101 around here and is quite a bit heftier.)

    68. Re:How about idle?? by Anonymous Coward · · Score: 0

      30 bucks for a cooler? I think I'll pay the extra 30 bucks and get an Intel CPU instead.

      Most gamers will have GPUs with fans that are noisier...

    69. Re:How about idle?? by tirefire · · Score: 1

      war4peace:
      I was in your exact position a few months ago. My Core 2 Duo E4300 wasn't cutting it for new games like Crysis 2 or LA Noire even though I had it overclocked +66% to 3000Mhz, fairly close to the same performance as your E6750. So I decided it was time for an upgrade, and I set out looking for the cheapest CPU that could handle any modern game at around 40 fps or more.

      I was all set to buy a brand new cpu+motherboard+ram kit from newegg, but at the last minute I checked on eBay and I found it was cheaper to just get a quad-core cpu for my old socket 775 system. After all, I already had plenty of fast DDR2-800 memory and I was 100% satisfied with my Gigabyte 965P-DS3 motherboard, so why chuck them out if I can still make use of them? Instead of paying $230+ for a Sandy Bridge Core i3 2100 kit, I just bought an old Core 2 Quad Q6700 for about $125, and sold my old E4300 a week later for $25 to recoup some of the money I spent. The Q6700 is great... it's basically two E6750s glued together on one package, and at stock clocks it is about 90%-100% as fast in modern multi-threaded games as a core i3 2100, which would have cost just as much as my Q6700 for the cpu alone, never mind the new board and memory I would need. And really I probably paid a little too much for the Q6700 at $125... if you are patient I bet you can get a used Q6700 for $80-$100 off eBay or maybe Craigslist. One tip if you go the used route: make sure the cpu was just pulled out of some some boring office workstation computer from 2007. You do NOT want a cpu that once lived in some 14-year-old's gaming pc that was overclocked to the max with some $20 off-brand power supply.

      The TDP for the Q6700 is 95W because it uses the dual-die 65nm "Kentsfield" design, but even so I think your Scythe heatsink (good choice on that, by the way...) should have no trouble handling 95W quietly. But if 95W still sounds like it's just too much heat for you, you might want to look into the 45nm "Yorkfield" chips, if I recall correctly they vary between 65W and 95W. Be aware, though, that not all socket 775 motherboards will support 45nm CPUs, even if you flash to the latest BIOS. You'll want to check with your motherboard vendor to make sure that your model supports 45nm chips (my Gigabyte 965P-DS3 does not support them, else I would have purchased a 45nm Q9650 with a better clock rate and more L2 cache than a Q6700).

      Oh, just remembered something about the Q6700 - if you leave it at or below stock frequency (2666 Mhz), it will almost certainly tolerate a good deal of under-volting, (your motherboard almost certainly supports voltage control if you can overclock with it). In my case, stock voltage is 1.27500V, but I got it down to 1.12500V, totally stable through 72 hours of torture testing. This reduced the cpu power consumption from 95 W to 73 W, based on this wattage calculator.

      Lately I've been gaming with headphones, so noise is not much of issue anymore, and I've found that I can get my Q6700 up to 3333 Mhz at 1.40000V (this is on a mediocre Thermaltake i1 heatsink) completely stable through 72 hours torture testing, temperatures between 44 C idle and 77 C full load, (last I checked, the "Kentsfield" Core 2 Quads are good to go at any overclock so long as they stay both under 1.50000V and under 80 C). At this overclocked speed, my 5-year-old Q6700 beats out a Core i3 2100 pretty handily. Overclocked, my Q6700 consumes a ton of power at 143 W (+51% more than stock power consumption), but it's getting chilly at my latitude this time of year and I could use a space heater anyway :P. Still, the +25% boost to clock speed is very noticeable when paired with my Radeon 7850 2GB, especially in cpu-heavy games like LA Noire and Deus Ex: Human Revolution. Btw, the Radeon 7850 is an unbeatable mid/mid-high range graphics card if heat/power consumption is a concern... it's noticeably faster than a GTX 560 Ti and it trades

    70. Re:How about idle?? by Anonymous Coward · · Score: 0

      That's precisely what I'm stating: incandescents are the best choice for many common, everyday scenarios. However, they're being banned (sorry, "efficiency standards'ed out of existence"). The ban in the US began with the 100 watt bulb and will continue, in steps, down to 40 watt bulbs, so your current approach will be stymied soon.

      In effect, the government is tying your hands in choosing the bulb you believe is most suitable. All of the replacement technologies will underperform in terms of TCO for this type of closet/bathroom use (et al). Sure, you could drop in a halogen if you don't want to pollute the earth (and drain your wallet) with dead CFL's, but each A19 halogen bulb is going to cost you $10 rather than the 25 cents for your incandescent.

      I'm just sick of hearing the green zealots' tired, demonstrably-false trope that incandescents are bad and serve no good purpose. I'm in favor of consumer choice, and this nanny state ban is the opposite of that.

    71. Re:How about idle?? by war4peace · · Score: 1

      Yeah, I have a Gigabyte P35 DS4 which is pretty good.

      --
      ...gis sdrawkcab (usually not responding to ACs; don't bother posting as AC)
    72. Re:How about idle?? by Anonymous Coward · · Score: 0

      If you are concerned about power consumption, find 100W lightbulb in your house. Replace with CFL. You will have greater energy saving.

      Why do you hate Freedom?

    73. Re:How about idle?? by TheRaven64 · · Score: 1

      Not to mention they never burn out. I have yet to have an LED fail me and my whole house has been LED since 2009.

      The lifespan of an LED is supposed to be about 40 years (not sure how much to trust manufacturer numbers though), but your second sentence doesn't really support the first. I've not had a CFL burn out in under 3 years, but I don't think I've had any last more than a decade either.

      If I were doing a new build now, I'd be tempted to put in a DC main for lighting and for powering small appliances and then LEDs would be a big win. Retrofitting them in existing houses seems like the wrong place to go for initial adoption.

      --
      I am TheRaven on Soylent News
    74. Re:How about idle?? by h4rr4r · · Score: 1

      Are you soldering them in?

      Just take them with you. I did that with CFLs last time I moved.

    75. Re:How about idle?? by h4rr4r · · Score: 1

      So then they get new bulbs without paying for them? Seems like that works for me.

      The reality is good quality bulbs in non-enclosed fixtures last as described.

    76. Re:How about idle?? by Anonymous Coward · · Score: 0

      DEY TUK ERR BUUULBS!

    77. Re:How about idle?? by hairyfeet · · Score: 1

      Well the nice thing about these heatpipe coolers is you really don't have to worry about orientation, at least not with the coolermasters. You can't see it in the picture but They design the base so YOU decide which way it is gonna blow, up/down or front/back. Most of the systems I build use the mTAX cases so its also front/back which is one of the reasons I stick with coolermaster, as some of the others you do get stuck on one orientation or another, but with these you bolt the base for whichever type of chip you've chosen onto the base of the HS and that way can point it anyway you like.

      But you really should look at these long term, since the fans are standard, you can change bases, so frankly the HSF can last as long as you want and through as many machines as you want. Simply keep the old HSF along with the parts to switch bases and the instruction booklet and you can then just slap the old HSF on and move the coolermaster to the new unit. Simply wipe the old paste off, apply new, bolt it on, and you're back in business.

      If you really want your system to shine? You really do need one of these. Frankly I have yet to see a decent stock cooler, even the best ones from Intel and AMD suck horribly compared to even the low end heatpipes. My place is kinda hot right now as I've had multiple PCs getting re-installs and my N520 has my hexacore at just 101f and its not even spun up. If you live in a hot climate the TX will definitely help drop those temps and will last you for many years, just hang onto the booklet and parts and you're golden. Enjoy.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    78. Re:How about idle?? by bingoUV · · Score: 1

      Interesting. From the picture it might appear it is difficult to get rid of dust on this - the aluminium part. How does it work out to be?

      And Arctic Silver - how does it perform during its "setting period" of upto 200 hours ? Does one need to be careful?

      --
      Bingo Dictionary - Pragmatist, n. A myopic idealist.
    79. Re:How about idle?? by bingoUV · · Score: 1

      Is there an after-market for the base of these coolers? I mean if I buy one, and later Haswell turns out to be tempting enough a year from now, Haswell CPUs most likely will be a slightly different size and shape. Would I be able to find a base to fit on the CPU, using the same heat sink?

      thanks

      --
      Bingo Dictionary - Pragmatist, n. A myopic idealist.
    80. Re:How about idle?? by Anonymous Coward · · Score: 0

      My 11W LED bayonett 'bulb' cost $12NZD a year ago

    81. Re:How about idle?? by bingoUV · · Score: 1

      But considering the weight, expense, hassle**, and inefficiency of the reflector, LEDs would come out ahead being naturally "spotlighty" ?

      ** : To retain efficiency, now not only the light bulb but the reflector also has to be cleaned.

      --
      Bingo Dictionary - Pragmatist, n. A myopic idealist.
    82. Re:How about idle?? by guruevi · · Score: 1

      The rated lifespan for a CFL is ~6,000-15,000 hours (1 year if you left it continuously on). Much better than incandescents but not nearly as good as LED that have a 20-30 year lifespan (if you left it continuously on).

      Also CFL's have all kinds of nasty side effects like mercury and other gasses inside the bulb, leave a really dirty signature on the net and have a really bad power factor. They're fluorescent tubes after all except where you could've gotten a decent ballast with a good coil on the big things, the designers of CFL tubes have all but stripped everything but a cheap electrolyte capacitor (which blows it's guts the moment the net browns out) and a badly coiled piece of wire (which is usually smashed during assembly to fit it in the base).

      Besides that, the flickering on that thing (50 or 60Hz depending on your location) gives many a headache and screws with fast shutter speeds on camera's.

      --
      Custom electronics and digital signage for your business: www.evcircuits.com
    83. Re:How about idle?? by DragonWriter · · Score: 1

      But considering the weight, expense, hassle**, and inefficiency of the reflector, LEDs would come out ahead being naturally "spotlighty" ? ** : To retain efficiency, now not only the light bulb but the reflector also has to be cleaned.

      I don't think you understand. CFL spotlight bulbs have the reflective surface as part of the inside of the bulb around the flourescent tube. LED spotlights might be lighter (never compared that) than CFLs, but they are definitely more expensive, and there is no difference in hassle.

    84. Re:How about idle?? by bingoUV · · Score: 1

      Ok, never seen such integrated product, but that might make sense.

      --
      Bingo Dictionary - Pragmatist, n. A myopic idealist.
    85. Re:How about idle?? by hairyfeet · · Score: 1

      IIRC you can contact Coolermaster and pay a couple of bucks and they'll send you a collection of bases, but most of these things are pretty standard so unless they REALLY change the layout you can usually fit say an Intel base onto a new Intel socket or a new AMD to an AMD, for example i had a friend that couldn't wait to use his new 1045T AM3+ but they screwed up and forgot to send his heatpipe cooler so he used a socket 939 on it no problem, and according to Wikipedia the LGA 1155 socket plate WILL fit the new chip, citation here.

      But when you think about how long these sockets last I'd say the heatpipe is a good investment, as its doubtful you are gonna need to do more than 1 or 2 max full socket changes over say a decade and by the time you can no longer find something your heatpipe will fit on frankly the cooling tech will have progressed far enough you'll probably want a new more efficient cooler anyway. And I can tell you that after switching a few years back I will NEVER go back to stock coolers, I've tried the best both Intel and AMD have to offer and frankly they don't hold a candle to even the lower cost heatpipes like the TX. The heatpipes are quieter, move more air (because you are pulling air through the heatsink instead of simply blowing down upon the base) and frankly the heatpipes simply pull more heat away from the chip than a standard stock plate and that translates to longer life and/or higher OCing if you are into that.

      So even with you paying the higher import cost if you were my customer I'd say go for it, you can use it this very day with what you have and then later on use it on your next chip, and in both cases it'll give you a much cooler and quieter system. Remember its not just the CPU and your ears that will benefit, a cooler chip equals cooler caps and phases which will help extend the life of the board. So just think of it like a GPU or CPU, its a long term investment that will last quite awhile and give you a better running system overall.

      --
      ACs don't waste your time replying, your posts are never seen by me.
  2. So they are not dead by Krneki · · Score: 1

    Finally something positive from AMD, while I'm not interested in this CPU since I'm a gamer, the lack of competition kept the Intel CPU price stagnant. This new AMD CPU seems to have some strength in multi-threaded applications. But then again, a 2 year old o/c Intel i5 eats any game you give him, while the same can't be said for a same priced video card. So Intel is not all that evil (watching AMD marketing troll campaign for Bulldozer on the other hand made me hate AMD).

    --
    Love many, trust a few, do harm to none.
    1. Re:So they are not dead by ifiwereasculptor · · Score: 1

      I'm a gamer too, and I'm actually interested, mainly because the FX-4300 seems to be now a fierce competitor to Intel's i3 while costing quite a bit less. The FX-8xxx still sucks, but this is a major improvement for AMD on the mainstrem segment. They were losing to cheaper Pentiums with the FX-4100, it was embarrassing.

    2. Re:So they are not dead by Anonymous Coward · · Score: 0

      Not dead, but if they're having to sell a chip with twice as many transistors as an i7 for two thirds of the price, they must be spraying blood all over the room.

    3. Re:So they are not dead by Synerg1y · · Score: 2

      You do realize that most of your processing occurs via your gpu in a game right...?

    4. Re:So they are not dead by Synerg1y · · Score: 1

      I miss the FX series... we may never have gotten the i7 if not for that war between intel and amd, but AMD never answered...

    5. Re:So they are not dead by Z00L00K · · Score: 1

      It certainly will make things more interesting - and increased number of cores is the way to go today. Most applications and operating systems will benefit from multiple cores - even though some applications may benefit from additional tuning.

      However - most bread&butter applications today runs well on even the cheap processors on the market, it's only when you start to look at things like gaming and advanced processing that you really will benefit from a faster processor. Most computers will be a lot faster when you replace the ordinary hard disk with a SSD. And if you really want some computing power you should look at using GPUs for computing. Sooner or later it's time to get rid of the x86 architecture and look at new ways to raise the performance.

      --
      If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
    6. Re:So they are not dead by Kjella · · Score: 1

      Not dead, but if they're having to sell a chip with twice as many transistors as an i7 for two thirds of the price, they must be spraying blood all over the room.

      Twice as many transistors? No, it's 1.2B versus Ivy Bridge's 1.4B - but the die size is almost double, 315mm^2 to 160mm^2. That is both because Intel is on a 22nm process and because they have higher transistor density, Intel's 32nm Sandy Bridge has 1.16B transistors but is only 216mm^2. I guess die size is one of the many things AMD didn't have the time or resources to optimize for and yes they're going to hurt selling these chips for $200 and down.

      --
      Live today, because you never know what tomorrow brings
    7. Re:So they are not dead by Anonymous Coward · · Score: 0

      You do realize that most of your processing occurs via your gpu in a game right...?

      And?

      The majority of modern games are quite bottlenecked by CPU. Thank shit console ports and/or general shit programming, combined with the industry's complete fucking lack of how to properly use multiple cores.

    8. Re:So they are not dead by Anonymous Coward · · Score: 0

      Depends on the game.

    9. Re:So they are not dead by baka_toroi · · Score: 1

      No, that depends on the game. AI processing isn't done on a GPU. Of course, it's almost always better to improve your GPU instead of your CPU.

    10. Re:So they are not dead by Synerg1y · · Score: 1

      Most AI / Game processing can be handed on a single core 2 ghz processor, why do we need multi-core for gaming? ... ... ...
      Windows background processes, yep it's as lame as that.

    11. Re:So they are not dead by Synerg1y · · Score: 1

      Something like Physx rendering gets kicked over to your cpu if your gpu can't do it... which at this point in time puts it in the bottom bracket of gpus, but besides the point even w crappy coding cpus are leagues and leagues ahead of any bad practices programmers may deploy. Read below for why mutli core.

    12. Re:So they are not dead by Anonymous Coward · · Score: 0

      There are games coming to market now that are *very* CPU dependant. A current example would be the Planetside 2 beta (which is also very dependant on the memory bus).

      There's likely to be more coming along with the next 'next-gen' consoles to boot - there's normally a period of a year or two after a new family of consoles is launched where system requirements for the PC ports surge.

    13. Re:So they are not dead by Anonymous Coward · · Score: 0

      My rig for the last year has been built on an AMD X6 1090T Thuban processor and is running great. I do a fair amount of gaming, recently I've been playing Skyrim and Boarderlands 2. I don't quite understand what people have against AMD for gaming. All of these games run great, I don't have any issues with lag and the system definitely does not feel slow. When playing multilayer with my friends (who run a mix of Intel i5/i7 and AMD Thuban/Bulldozer systems) I am always one of the first people to load into a map.

      My question is can anyone give any reproducible demonstrations as to what I am missing out on using an AMD processor vs an Intel when it comes to gaming. Sure, benchmarks can show that an i7 can have X number more points then the AMD, but what does that translate out to in reality. Can anyone give an example where if I put my rig up beside someone with an equivalent Intel system I would get clearly "smoked" outside of a synthetic benchmark?

    14. Re:So they are not dead by Anonymous Coward · · Score: 0

      Too bad my 2.16ghz i7 can't feed data to my GPU to push my 6950 past 10% load in many games at 1080p ultra settings. Single threaded performance still matters until deferred-rendering becomes standard.

    15. Re:So they are not dead by Anonymous Coward · · Score: 0

      Only if you have a shit CPU, or you're playing a really bad port. Most games are designed around hardware specs that are six years old, so as to be compatible with consoles. Any relatively modern CPU, even my 1090T, can slam anything currently on the shelf at 16xAA with even a mediocre video card.

      If you're buying the high-end i7s "because gaming hurr", then you're wasting your money -- your CPU isn't even breaking a sweat. Game engines have simply not caught up with even the last generation of CPUs/GPUs.

    16. Re:So they are not dead by Anonymous Coward · · Score: 0

      Not only games. Google Chrome, Firefox, and others don't make use of threads in efficient ways, and these are the programs people actively use.

      Where is our multi-threaded Javascript? Oh firefox doesn't do it: https://bugzilla.mozilla.org/show_bug.cgi?id=392073 , neither does chrome: http://www.chromium.org/developers/design-documents/multi-process-architecture

      Why are we still using 1970's "multi-process pre-fork" models? Waste ram, waste cpu, waste power.

    17. Re:So they are not dead by 0123456 · · Score: 1

      Guild Wars 2 seems to be CPU-limited as well; no graphics setting change other than supersampling makes more than 1fps difference on my system.

    18. Re:So they are not dead by Synerg1y · · Score: 1

      That's probably your gpu not being challenged, gpu's have exceeded game requirements in recent years by quite a wide margin (also you failed to mention your cpu load)... there does seem to be a bunch of confusion on the issue, so here's a great link on the subject: http://www.thinkdigit.com/forum/hardware-q/154674-what-role-cpu-gpu-relation-gaming.html

      Now, going back to my previous statement: unless you run a p4, you're probably fine on the cpu and the gpu is what does the heavy lifting.

    19. Re:So they are not dead by Anonymous Coward · · Score: 0

      No, they cannot, because there isn't a game on the market right now that can really stress even the 1090T, never mind the current-gen i5 or i7 series. Even if there were, you'd have a bottleneck at the PCIe bus, so what does it matter?

      I also have the 1090T, and even Skyrim with high-texture mods and the LBA patch, at maximum settings plus driver-level AA enhancements, my CPU never inched above 50% across all cores. My GPU was sweating bullets, sure, but even with active monitors running on my second screen, I never had a bit of lag. The whole rivalry BS is just that... games quite simply haven't caught up to the hardware, but neither Intel nor AMD want you to figure that out -- they just want you to buy "the next big thing!"

    20. Re:So they are not dead by Anonymous Coward · · Score: 0

      My CPU tends to run with 1-2 logical cores pegged and the rest idle. Unless you're using DX11, you can only have one thread supplying instructions to the GPU. There is a fixed overhead associated with the system calls to transfer instructions to the GPU. Entirely limited by context switching and system calls, a single thread has a cap on your graphics performance.

      My GPU runs about 10% and my CPU about 12-20%. Game engines are crap right now.

    21. Re:So they are not dead by DocSavage64109 · · Score: 1

      Try reading some reputable websites where they test modern games with modern cpus (like hardocp.com) and you will see that at high resolutions and high graphic settings, there are huge differences in frames per second between processors.

    22. Re:So they are not dead by Anonymous Coward · · Score: 0

      it's 1.2B versus Ivy Bridge's 1.4B - but the die size is almost double, 315mm^2 to 160mm^2. That is both because Intel is on a 22nm process and because they have higher transistor density, Intel's 32nm Sandy Bridge has 1.16B transistors but is only 216mm^2.

      Remember also that the Intel Ivy/Sandy Bridge die size and transistor counts include an integrated GPU. AMD's FX series CPUs don't have an integrated GPU (only their "APU" chips do).

    23. Re:So they are not dead by Tough+Love · · Score: 1

      I'm interested in this cpu because I'm a dev. It looks like a pretty respectable game workhorse too. I'm glad I didn't waste my money on an i7 last time around, though I did go Intel with a 4 core i5. AMD just got me back. I just don't like feeding money to the cynical old monopolist.

      --
      When all you have is a hammer, every problem starts to look like a thumb.
    24. Re:So they are not dead by Tough+Love · · Score: 1

      Not dead, but if they're having to sell a chip with twice as many transistors as an i7 for two thirds of the price, they must be spraying blood all over the room.

      Not if they're booking a profit on the chip. Comparing AMD's income to Intel's monpoly-boosted margins used to make sense when you knew Intel would be turning a lot of that windfall profit into dirty tricks aimed at choking off AMD's air supply. A couple things changed: Intel has antitrust watchdogs in its hallways now; and Intel has to fight a war on another front, ARM.

      Consider this: AMD is able to produce competitive chips with an uncompetitive process. What if Intel actually had to compete on a level playing field, as might well come to pass with multiple competitors banding together in the push to bring up 20nm and 14nm nodes.

      --
      When all you have is a hammer, every problem starts to look like a thumb.
    25. Re:So they are not dead by Tough+Love · · Score: 1

      Remember also that the Intel Ivy/Sandy Bridge die size and transistor counts include an integrated GPU. AMD's FX series CPUs don't have an integrated GPU (only their "APU" chips do).

      The HD 2500 on the Ivy Bridge desktop chips is a joke, it might as well not be there at all. Any $50 fanless Radeon will blow it right out of the water.

      --
      When all you have is a hammer, every problem starts to look like a thumb.
    26. Re:So they are not dead by Ironhandx · · Score: 1

      Try actually reading some of those sites yourself... even the most modern games the difference between stock AMD processors and stock Intel processors is usually something along the lines of 150 fps vs 200+ fps. Neither matters, your monitor can't display all of the frames EITHER ONE is rendering.

      Therefore, he is absolutely correct. CPU doesn't matter at all.

    27. Re:So they are not dead by Ironhandx · · Score: 2

      Actually check out the Anandtech review. They do it on windows 8 with the new scheduler... the difference in performance of the FX series from Windows 7 to Windows 8 is fucking mind-boggling. The FX-8350 actually trounces all but the higher end I5s and gives the I7 a run for its money in several games. For Sub-$200 if you're getting Windows 8(which I apparently will be now) you can't get anything thats even close performance wise.

    28. Re:So they are not dead by Anonymous Coward · · Score: 0

      The X6 is an amazing processor that got less credit than it deserved. Hell, I've even seen a few benchmarks where it beats a i7-3770k by a few points!

      Unless you absolutely must have the fastest possible system out there, I don't think there's anything worth upgrading to right now if you already have a 1090T.

    29. Re:So they are not dead by hairyfeet · · Score: 1

      If you want gaming on the cheap you need to look at the Phenom II X6 chips, you can pick them up for like $110 and in most benches they are just a few percentage points away from the FX 8s and usually beat the FX 4s pretty handily.

      Since i have two boys that also game and i wanted us to be pretty evenly matched i had to be seriously picky when it came to price, me and the oldest who play more hardcore games got X6s while the youngest who is more into MMOs took the 925 quad and paired with some HD4850s I got cheap we blow through pretty much any game we want at the monitors 1600x900 resolution with plenty of bling and nice framerates. Just Cause II, Saints Row 3, the Crysis series, Payday the Heist, everything plays nice and the systems were pretty cheap to build.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    30. Re:So they are not dead by hairyfeet · · Score: 1

      Which just proves most mainstream games aren't gonna struggle on pretty much ANY CPU made in the last 5 years short of Sempron or Celeron.

      In fact just for shits and giggles I decided to see what it would take to drag down my Phenom II X6 and the only time I could really get the system to drop below 30FPS (although to be fair, with all the fireballs it may have been the HD4850 choking) was in Just Cause II when I did my own version of "cool guys don't watch explosions" by rigging an entire oil refinery with remote charges and set the whole thing off while doing the cool walk.

      Frankly with the price of AAA graphics fests hitting over 100 million i have to wonder if the days of CPU and GPU slamming games are gonna be more of a rarity than the rule, because of the crazy costs associated with making them. Wasn't it one of the guys at Epic that said they made more on their iPhone game than their last AAA titles? With chips from even 5 years ago being so insanely overpowered the amount of crap you gotta put on the screen to really slam one of these chips is just nuts, and I'd hate to see the bill on the coding required to slam the latest Core i7s.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    31. Re:So they are not dead by Anonymous Coward · · Score: 0

      Get yourself a monitor with 2560x1600 resolution, apply a little antialiasing, and you will see the difference can be very noticable -- as in between choppy and smooth. If you have a 1280x1024 screen, then I concede.

    32. Re:So they are not dead by Anonymous Coward · · Score: 0

      At 1920x1080, a very reasonable resolution, and 16xAA, the difference is negligible and completely undetectable to the human eye. Period. End of fscking story.

      You can quote specs and benchmarks to the contrary, but I'm running that system right now, with at 1090T, and it's perfectly acceptable. If you want the highest possible benchmark numbers, regardless of human perception, that's your prerogative... but it's utter BS, and provides you the consumer jack shit for your investment. True fact...

    33. Re:So they are not dead by Anonymous Coward · · Score: 0

      What about the Civ games? I always assumed a better CPU would speed up between turns, as it's not really rendering anything. I hope I'm right, as it's something I look forward to when i upgrade.

      I've heard rumors of dwarf fortress too.

    34. Re:So they are not dead by TheLink · · Score: 1

      Doesn't matter for _desktops_. For most desktop users the HD 2500 is more than enough. The gamers who need GPU performance will be buying GPUs for their desktops anyway.

      --
    35. Re:So they are not dead by TheLink · · Score: 3

      How the heck is using a higher resolution relevant to a _CPU_ test? High resolutions stress out the GPU way more than the CPU.

      You use a low resolution so that the bottleneck becomes the CPU and not the GPU.

      --
    36. Re:So they are not dead by TheLink · · Score: 1

      chrome does multi-process and so can use multiple CPU processing threads.

      Multiprocess = safer.

      And if you have a crappy plugin that leaks memory, you close the offending window and the memory is freed. You don't need to close the entire Chrome browser and lose your sessions etc. Whereas with Firefox since they have a single process model, you have to close the entire browser to free up the leaked memory, even if it's just one problematic tab out of many.

      --
    37. Re:So they are not dead by Tough+Love · · Score: 1

      For most desktop users the HD 2500 is more than enough.

      Not really. Even screensavers suck on this little thing.

      --
      When all you have is a hammer, every problem starts to look like a thumb.
    38. Re:So they are not dead by Anonymous Coward · · Score: 0

      What screensaver are you using? Crysis?

    39. Re:So they are not dead by DocSavage64109 · · Score: 1

      You are correct in saying that the video card makes a huge difference. But so does the processor. For that matter, so does having sufficient memory. If you don't believe me, read http://www.hardocp.com/article/2011/10/11/amd_bulldozer_fx8150_gameplay_performance_review/4 and notice how even the i5 2500k has nearly 50% more average fps (from 40.8 to 57.2) than the amd fx8150 on a fairly high-end AMD Radeon HD 6970 even at only 1920x1200 resolution and noAA/noAF. Then try extrapolating that to a 2560x1600 screen (1.78x the pixels) and that would be 23 average fps vs. 32 average fps -- the difference between smooth and choppy. Note that the same review has other games with no noticeable difference between the processors, but the only game of the few reviewed that I play happens to be Civ V.

    40. Re:So they are not dead by TheLink · · Score: 1

      I didn't even say what you claimed I said.

      --
    41. Re:So they are not dead by Anonymous Coward · · Score: 0

      "...unless you run a p4, you're probably fine on the cpu and the gpu is what does the heavy lifting."

          Sadly I do. It's to bad ati/amd open source drivers under freebsd suck so bad.
      Actually it's the drivers and gpu that do the heavy lifting because if either one of them suck the end result is the same.

      glxgears
      radeon x600 - x1600, 4670 on several p4s 1.6-3.2Ghz 1GB ram 8x agp, pcie-- 2000 frames/sec regardless of system or card mixture.
      nvidia 5900xt on athlon XP 1.6Ghz 1GB ram binary drivers 4X agp only -- 5500 frames/sec solid.

      Ut2004 on radeons is playable with everything turned off. Nvidia is playable with settings up 3/4 way and higher.

    42. Re:So they are not dead by Anonymous Coward · · Score: 0

      Actually check out the Anandtech review. They do it on windows 8 with the new scheduler... the difference in performance of the FX series from Windows 7 to Windows 8 is fucking mind-boggling.

      Er, what? I think you need to reread the review a little more carefully. It looks like some of the benchmarks were run under Windows 7, some under Windows 8, but at no point did they run the same benchmark on both Windows 7 and 8 to compare the schedulers.

      What you're seeing in AnandTech's numbers is that the performance fixes and clock speed bumps in Piledriver have improved it quite a bit compared to Bulldozer. The original Bulldozer FX-8150 still looks like crap in AnandTech's review, even in tests run on Windows 8. It's only Piledriver CPUs like the FX-8350 which look better.

      The idea that the Windows 8 scheduler was going to be the savior of Bulldozer has always been a fanboy pipe dream, and it's gotten a lot more credibility than it deserved. AMD needed to make silicon changes, and that's what they've done with Piledriver.

      The FX-8350 actually trounces all but the higher end I5s and gives the I7 a run for its money in several games.

      That's a pretty biased take on the data AnandTech presented. In highly parallel tests, the FX-8350 is great, yes, but in less parallel tests i5s often trounce the FX-8350. Which is pretty much what you'd expect given the strengths and weaknesses of each CPU design.

      There's a reason why AMD is pricing the FX-8350 to compete directly against i5s instead of i7s. Lots of people would actually be better off with an i5. Gamers, for example:

      http://www.anandtech.com/bench/Product/701?vs=697

      An i5-3570K "trounces" the FX-8350 in every game AnandTech tested, and it only costs $10 more on newegg right now.

      If you want to encode videos all day, on the other hand, there is no question that the FX-8350 would be a better choice.

    43. Re:So they are not dead by Anonymous Coward · · Score: 0

      Way to miss my point entirely. Which was that if you're going to compare die size and transistor count between Ivy Bridge and Piledriver and you're interested in comparing apples to apples, you should probably try to estimate the area and transistor count of the Ivy GPU and subtract those from the Ivy Bridge totals. Regardless of how much a joke you might think it is, the Ivy GPU is a pretty large block and the die would be substantially smaller if it was left out.

      http://www.anandtech.com/show/5876/the-rest-of-the-ivy-bridge-die-sizes

      Take a look at the difference between 4-core Ivy GT1 and GT2 variants. Things are complicated because it isn't just the graphics that changes, but also the L3 size. However, if you look up annotated die shots of 4-core Ivy elsewhere on the web, it's clear that cutting the GPU is responsible for changes in the length and cutting L3 for changes in the width. With that, we can try to estimate how large an Ivy "GT0" with 8MB L3 cache would be.

      Assume GT2 graphics are exactly twice as long as GT1 (this is probably untrue but still reasonably good for back-of-the-envelope purposes) -- that would mean you'd take 4.024mm off the length of the GT2 die to remove graphics completely. Multiply by the width of the 8MB die and you get 125 mm^2.

      So, Intel needs 125mm^2 of 22nm silicon to make a CPU which is more than competitive with AMD's 315mm^2 32nm design.

      Transistor count is a bit more involved to estimate.

    44. Re:So they are not dead by Anonymous Coward · · Score: 0

      Not if they're booking a profit on the chip.

      Er, I don't know if you'd noticed, but AMD's had a tough time making profits most years.

      Consider this: AMD is able to produce competitive chips with an uncompetitive process.

      Needing more than 2x the transistor count and die area, 1.6x the power, and a higher clock speed for roughly the same throughput running embarrassingly parallel code (while being way behind on single thread performance) is not exactly what I'd call "competitive". It means that for any given performance level, Intel's cost to build that chip is far, far less than AMD's -- and that's when AMD can even match Intel's performance. At the high end, they simply can't, so they can't sell high end CPUs at all these days.

      Which is why AMD has a tough time making profits most years.

      What if Intel actually had to compete on a level playing field, as might well come to pass with multiple competitors banding together in the push to bring up 20nm and 14nm nodes.

      You don't really understand the trends in the semiconductor industry. Alliances are not a new thing. Intel's lead today exists in spite of several long-standing alliances, and it's weird to even say "in spite of" because the alliances aren't actually intended or expected to topple Intel's technical lead. They exist because over time it's become more and more necessary to share development costs.

      The thing you have to understand is that Intel is in a unique position. They're the only one with a major customer (themselves) interested in building enough high performance desktop CPUs to fill multiple fabs. Everybody else has a customer mix which is mostly interested in low power, so low power processes are what they're developing. Intel's playing a different game. They're the only one which can justify being so aggressive on new technologies (such as HKMG and FinFET) that they do volume production 3 to 4 years before anyone else. (Yes, this means Intel 32nm is better than anyone else's 32nm -- the feature size is not the only relevant figure of merit for a process.)

      AMD used to optimize their processes for high performance CPUs too, but when they spun off their fabs to create GlobalFoundries, GF began restructuring to chase the same low power markets as all the other merchant foundries. Because that's where the volume and money is. (Also, it's probably easier to take advantage of industry partnerships when your technology goals are more aligned with theirs.)

      At this point, the only plausible way AMD can get back to a level playing field in process tech is if the bottom drops out of the market for super high performance desktop and laptop CPUs, destroying Intel's margins and ability to keep outspending everyone else on process R&D. But the endgame there is not exactly what you seem to desire. That would mean the market had shifted to lower performance ARM SoCs once their performance became "good enough", killing sales of high performance CPUs. Instead of AMD once again jockeying with Intel to make the most badass CPUs on the planet, Intel would be going through a painful transformation similar to what's happened to AMD over the past several years and nobody would be making badass CPUs any more.

      (Which is a thing Intel does fear quite a bit, because their high margin business model depends on sustained demand for badass CPUs.)

    45. Re:So they are not dead by Anonymous Coward · · Score: 0

      The X6 is an amazing processor that got less credit than it deserved. Hell, I've even seen a few benchmarks where it beats a i7-3770k by a few points!

      No you haven't. Don't be absurd.

      http://www.anandtech.com/bench/Product/203?vs=551

      (Take note, some benches are larger-bar-is-better, some are smaller-bar-is-better.)

    46. Re:So they are not dead by Ironhandx · · Score: 1

      I said higher end. 3570 is where it starts to win a bit again. The games that were tested in Win8 that particular i5 lost most of the benches.

      The scheduler is making big improvements however. The 8150 is in those benchmarks seeing huge gains over similar tests run 6 months ago with much the same configuration for GPU but on Win8. The overall data points to a 20-30% performance gain, which puts it where AMDs marketing team has basically said they are shooting for right now. They just screwed up releasing bulldozer too early. They should have continued development and released a Phenom III then released bulldozer now, as piledriver.

    47. Re:So they are not dead by Anonymous Coward · · Score: 0

      I said higher end. 3570 is where it starts to win a bit again.

      The FX-8350 is priced very close to the 3570K, so the 3570 is an appropriate comparison don't you think? Also, even if you select the i5-3470 in AnandTech's Bench database, you'll find that Intel still wins all but one game benchmark over FX-8350 by medium-to-large margins. (The one exception is a virtual tie, as we'll see below.)

      Once again, this is not a surprise. Few games scale well above 2 to 3 CPU cores. It's hard to parallelize lots of the code in games, especially the part which generates commands for the GPU. Whenever the software you want to run can't scale to 8 cores, the FX-8350's weak individual cores are likely to show up as a limitation.

      The games that were tested in Win8 that particular i5 lost most of the benches.

      Dude. There is no way you can justify claiming any i5 AnandTech tested "lost most" Win8 gaming benchmarks to FX-8350. AT only tested two games in Win 8, Skyrim and Diablo 3. All three i5s tested handily beat the FX-8350 in Skyrim (by 7% or more), and all 3 basically tie the FX-8350 in Diablo. (I mean, technically you can say the FX-8350 "beat" the i5-3470 and i5-2500K in Diablo... 215.8 fps to 215.5 or 214.2 fps. Yeah. Great. Ask AT to rerun that bench 10 times and I'm dead sure you'll see the order change a few times because these differences are down in the noise.)

      I hate to say it but it really seems to me like you're seeing what you want to see, not what's there.

      http://www.anandtech.com/show/6396/the-vishera-review-amd-fx8350-fx8320-fx6300-and-fx4300-tested/5

      The scheduler is making big improvements however. The 8150 is in those benchmarks seeing huge gains over similar tests run 6 months ago with much the same configuration for GPU but on Win8.

      Show me, and be sure to prove there aren't any other factors (such as 3D driver updates). I'm all ears. So far you seem content to just say that these things are happening without proof, or just flat out invent shit that isn't there (like your claim that the AT FX-8350 review supports this idea).

      A CPU-only test would be a much better method of proving your point than a game benchmark, by the way.

      The overall data points to a 20-30% performance gain,

      Oh good lord, you really are one of those delusional fanboys, aren't you? No, no data points to a FX-8150 gaining 30%. I guarantee it.

      Do you guys not understand that even AMD at its most optimistic never said tweaking the Windows scheduler was going to be OMG HUEG? Here's an AMD guy saying the Windows 7 patches (which covered the most important things long before Windows 8) were not likely to deliver more than a 10% boost to anything, and a mere 1-2% on average. (The low average figure is because the original Win 7 scheduler actually wasn't bad for Bulldozer 100% of the time, just under some workloads.)

      http://blogs.amd.com/play/2012/01/11/early-results-achieved-with-amd-fx-processor-using-windows%C2%AE-7-scheduler-update/

      which puts it where AMDs marketing team has basically said they are shooting for right now. They just screwed up releasing bulldozer too early. They should have continued development and released a Phenom III then released bulldozer now, as piledriver.

      No, the screwup was doing Bulldozer at all. It was a risky decision to pursue an unusual architecture, and it didn't pay off the way they hoped. But they kept pursuing it even after some problems (it was delayed a very long time as it was), and that pretty much committed them to sink or swim with it.

      You have to keep in mind that (out of economic necessity) AMD uses the same die for both server and "enthusiast" desktop products. FX-8150 Bulldozer is the same as "Interlagos" Opterons -- the only difference is that in Interlagos, more HyperTransport links are activated and two dies are assembled in one MCM packag

  3. Lowers barrier to entry by Gothmolly · · Score: 5, Interesting

    I put together an 8way,32GB machine (no local storage) for $400 to play with ESXi. Courtesy of the freebie VMWare download and a reasonably priced 8way machine, I can get into some pretty serious VM work without spending a ton of dough. I don't need massive performance for a test lab.

    --
    I want to delete my account but Slashdot doesn't allow it.
    1. Re:Lowers barrier to entry by h4rr4r · · Score: 2, Interesting

      Get an SSD.
      Local storage is a must for performance. iscsi cannot hold a candle to local SSDs. In a lab you won't need to share the storage with multiple machines anyway.

    2. Re:Lowers barrier to entry by rrohbeck · · Score: 4, Informative

      Same here. I built a Bulldozer machine for compiling projects in VMs last year and it works very nicely. If Intel had had a CPU with ECC memory and hardware virtualization support at a reasonable price I would probably have bought it, but I would have needed at least a $500 Xeon for that, with a more expensive motherboard, and I wouldn't be able to overclock it. For the same performance I have now I would probably have needed a $1k CPU.

    3. Re:Lowers barrier to entry by Anonymous Coward · · Score: 2, Informative

      Local storage is a must for performance

      This is hyperbole. If what you're doing is mostly CPU or Memory intensive and requires very little disk activity having fast local storage isn't going to help much, if at all.

      Besides, apparently it isn't a must for the grandparent as he stated he doesn't "need massive performance for a test lab."

      Don't get me wrong, using an SSD to provide storage for a handful of VMs is a great idea (massive read/write IOPs), but it isn't necessary.

    4. Re:Lowers barrier to entry by __aardcx5948 · · Score: 1

      8-way would mean 8 sockets on the motherboard, like http://www.supermicro.com/products/system/5U/5086/SYS-5086B-TRF.cfm .

    5. Re:Lowers barrier to entry by Anonymous Coward · · Score: 0

      8-way would mean 8 sockets on the motherboard

      This is incorrect. N-way can apply to the number of sockets, cores, or threads in a system.

      An 8-core AMD FX processor is considered to be an 8-way processor.
      A 4-core Intel i7 processor could be considered either a 4-way or 8-way system, depending on if you're looking at physical cores or total threads.
      A SPARC T4 could be either an 8-way or 64-way system, once again if you're looking at cores or threads.

    6. Re:Lowers barrier to entry by __aardcx5948 · · Score: 1

      I've always looked at it as number of CPUs (physical chips), number of cores (total, on all CPUs), number of threads (total cores, with HT or not etc).

      -way was a socket thing for me, but perhaps it's not.

    7. Re:Lowers barrier to entry by LordLimecat · · Score: 1

      Without iSCSI you cant really use shared storage, which means 90% of the features of ESXi cant be used. Kind of dampers the whole "for a lab" thing.

      SSDs ARE quite sweet for VMs, Id recommend setting up a VM that serves out a local SSD as iSCSI over an internal ESXi storage network-- thats actually how things were done during my VCP training. I believe they were using FreeNAS (MIGHT have been openfiler) to serve up iSCSI and NFS targets. Its a little buggy but sufficient for a lab.

    8. Re:Lowers barrier to entry by Anonymous Coward · · Score: 0

      This is the best definition I can find:

      N-ways

      Multi-core compute nodes can be described by the number of execution units, or cores. A computer with 8 cores would be described as an 8-way node. This machine can have 8 independent processes running simultaneously. A 32-core system would be called a 32-way node.

      Some hold that it only pertains to the number of physical cores and sockets, but I've seen it used when describing a system based on the number of threads as well.

    9. Re:Lowers barrier to entry by Anonymous Coward · · Score: 0

      Without iSCSI you cant really use shared storage, which means 90% of the features of ESXi cant be used

      s/iSCSI/& or NFS/

      With only one node, which the original commenter described, you can't use those features anyways.

      If you are using network attached storage that is already in place you can decrease the cost of the VM host itself and have less hardware to maintain. That's a big benefit.

    10. Re:Lowers barrier to entry by Simulant · · Score: 3, Interesting

      I converted about 20 physical servers to hyper-v VMs running on two 2 socket/32 core Bulldozer, hyper-v hosts at the beginning of the year and have been thrilled with the results. Those two servers provide more horsepower than my company needs and, in a pinch, either one of them can run all of the critical VMs itself. Not a single problem since deployment. I paid around 11k/box. Intel cost quite a bit more in hardware as well as for the per socket licensing of Hyper-v for that many cores using intel chips. I was converting >4-5 year old hardware and as far as the users are concerned, everything is faster now.

      I will keep buying AMD as long as they are cheaper and "good enough", if only to keep some competition alive.

      Still running a quad-core AMD gaming machine at home as well and it is still playing every thing I throw at it.

    11. Re:Lowers barrier to entry by Anonymous Coward · · Score: 0

      At the risk of inviting the inevitable "yo, dawg, I heard you liked..." comments...

      With ESXi 4 and later, you can virtualize ESXi *hosts*...which means, with one suitably powerful server, you can have three virtualized ESXi hosts, and run something like FreeNAS to have your iSCSI target. This lets you play with clustering, vMotion, etc even if you have only one physical server to use.

      Just a thought...

    12. Re:Lowers barrier to entry by Anonymous Coward · · Score: 0

      I'd like to know more about your set up, if you're willing to share:

      Make (Dell, Lenovo, IBM, etc)
      Model
      CPU
      RAM
      VM storage (local or network attached)
      Applications/Services you are providing
      The size of your user base
      etc

    13. Re:Lowers barrier to entry by LordLimecat · · Score: 1

      That is precisely how our VCP lab was set up. It was ESXi 5, however, which plays much nicer with nested hypervisors. Each student lab was a vApp with a storage VM, a VCVA vm, two ESXi hosts, and a Windows VM (we RDP'd into that host for access to our infrastructure). Adding a new student was just a matter of cloning the whole shebang and doing a little tinkering with the Windows VM. The whole setup used under 16GB RAM and 30GB storage, IIRC-- our test vms were all win2k or 2k3, and only needed a few GB of RAM each/

    14. Re:Lowers barrier to entry by strikethree · · Score: 1

      I have an 8way (16?) machine with 24 gigabytes of RAM. With Windows Server 2008 R2 and Hyper V, I can not run more than 2 VMs without noticeable lag. I hope you are using some other technology.

      --
      "Someone needs to talk to the tree of liberty about its ghoulish drinking problem." by ohnocitizen
    15. Re:Lowers barrier to entry by Anonymous Coward · · Score: 0

      Here are a couple of whitebox builds I've done over the past year:

      AMD Phenom II X6 1090T , 16 GB RAM, single 1 TB hard drive (two of these boxes total)
      Intel i7-3770, 24 GB RAM, 1x 1 TB hard drive, 2x 1.5 TB hard drive

      Both running VMware ESXi.

      I've ran numerous VM guests with very little noticeable performance issues. The biggest bottlenecks have been network performance (running over a single gigabit ethernet interface, so I added more) and disk I/O. Booting up several VM guests at once, or performing multiple OS installs over several VM guests at once, was always met with slowness, but it's because I was running multiple guests off a single drive. If I were to spread the VM guests over multiple disks, or used an SSD instead, I'm sure it would have been fine.

    16. Re:Lowers barrier to entry by Simulant · · Score: 1

      Sure...
      Server specs are:

      Dell R715
      2 x AMD Opteron 6282 SE CPUs
      96 GB RAM
      4 x 600 GB 10k SAS drives (RAID 5 data)
      2 x 146 GB 15k SAS drives (RAID 1 system)
      4 x onboard Broadcom 1Gbps NICS
      4 x 1Gbps Intel NICS (quad port server adapter)

      Most production VMs are stored on an Equalogic PS4100 (2 x 1Gbps multipath iscsi connection) but I run some non-critical stuff locally. The local storage was spec'ed mainly for testing and backup.

      I support a little over 400 users, 250 of which use thin clients.

      4 of the VMs are Server 2003 x64 terminal servers in a cluster config, used by the thin client users.
      Other servers are mostly 2008R2 running various things including, DC/DNS (we have a physical DC as well), Blackberry BESX, Nagios, Spiceworks, Document management, ISS/MSSQL (internal and external web sites & apps), a dedicated print server.

  4. Here's the problem... by CajunArson · · Score: 4, Informative

    These chips "excel" at big, heavily threaded workloads. Which is to say that they can beat similarly priced Intel chips that are simply up-clocked laptop parts. Move up to a hyperthreaded 3770K (still a laptop part) and Vishera usually loses. Overclock that 3770K to be on-par with the Vishera clocks while still using massively less power than Vishera and the 3770K wins practically every benchmark.

      Unfortunately, if you *really* care about those workloads (as in money is on the line) then Intel has the LGA-2011 parts that are in a completely different universe than Vishera, including using less total power and being much much better at performance/watt to boot. I'm not even talking about the $1000 chips either, I'm talking about the sub $300 3820 that wins nearly every multi-threaded benchmark, not to mention that $500 3930K that wins every one by a wide margin.

        So if you want to play games (which is what 90% of people on Slashdot really care about): Intel is price competitive with AMD and you'll have a lower-power system to boot. If you *really* care about heavily-multithreaded workloads: Intel is price competitive because the initial purchase price turns into a rounding error compared to the potential performance upside and long-term power savings you get with Intel.

          Vishera is definitely better than Bulldozer, but AMD still has a long long way to go in this space.

    --
    AntiFA: An abbreviation for Anti First Amendment.
    1. Re:Here's the problem... by Synerg1y · · Score: 1

      GPU's benchmark game performance...

    2. Re:Here's the problem... by CajunArson · · Score: 2

      GPUs are very important for games, which is why an Ivy Bridge with an on-die PCIe 3.0 controller is going to do better at keeping next-generation GPUs running full-tilt than the PCIe 2.0 controller on the Northbridge of an AM3+ motherboard.

      --
      AntiFA: An abbreviation for Anti First Amendment.
    3. Re:Here's the problem... by Anonymous Coward · · Score: 2, Insightful

      AMD has never been about pure performance. It's all bang for buck. You can : buy an AMD system for much less than an Intel one, get a motherboard that has a lot more connectivity than the equivalent Intel board for less, AND get a true 2 x PCI-Ex 16x (while Intel force you to get an LGA2011 board, much costlier). The tradeoff? You'll get a machine with a CPU that perform maybe 10-20% less in benchmark than the Intel equivalent. But seriously, who cares in 2012? Most game are GPU starved, so you're much better spending that 50-70$ that an Intel would have costed you on a better GPU. And most day to day operation can be sped up by getting a fast SSD and more RAM. Get used to it, the CPU is simply not the main bottleneck anymore.

      HOWEVER, if you're an overclocker that looks to shatter the world benchmark record or a mad scientist that need workstation-like horse power AND you have the cash to spend on it, I'd say go for Intel. It'll worth every penny.

    4. Re:Here's the problem... by Anonymous Coward · · Score: 4, Insightful

      You are missing an importan point when it comes to "money is on the line". No one in their right mind would use a desktop processor from Intel for anything critical. Why? All non-xeon processors have been crippled to not support ecc memory. If money really is on the line there is just no way that is acceptable.

      Amd on the other hand does not cripple their cpu's as all. The whole Vishera lineup support ecc memory, as did Bulldozer.
      The xeon equivalent of 3820 is in a completely different price league.

      So please, when you compare price and use cases make sure you fully understand which processors are the actual alternatives.

    5. Re:Here's the problem... by CajunArson · · Score: 1

      Sure Vishera theoretically supports ECC memory, but you need a motherboard that takes ECC memory to tango... and those are a rare beast in the consumer space, meaning you really are looking at Opteron socketed motherboards and Opteron chips (which are nowhere near as cheap as Vishera). So there is no free lunch.

      --
      AntiFA: An abbreviation for Anti First Amendment.
    6. Re:Here's the problem... by Enderandrew · · Score: 1

      When you talk about Intel being price competitive, it depends.

      AMD clearly wins on budget gaming systems where the processors and motherboards are much cheaper, but Intel has the fastest high end systems out there right now.

      Just a couple months ago I priced two builds with similar benchmark numbers on NewEgg, and the AMD budget gaming rig was around $800, and the Intel equivalent was around $1100.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    7. Re:Here's the problem... by Anonymous Coward · · Score: 0

      Here in Norway, in the most popular consumer-oriented web-stores there are more ECC-enabled motherbords for AM3+ than not. If this is not true for your location I send my sincere condolences.

    8. Re:Here's the problem... by rrohbeck · · Score: 1

      Look at the Phoronix benchmarks. Vishera beats the 3770k in many benchmarks as long as you're running multithreaded code.
      http://www.phoronix.com/scan.php?page=article&item=amd_fx8350_visherabdver2&num=1
      And do the math about power savings. Unless you're a folder you'll need several years to recap the additional cost of an Intel CPU.

    9. Re:Here's the problem... by Kjella · · Score: 1

      You are missing an importan point when it comes to "money is on the line". No one in their right mind would use a desktop processor from Intel for anything critical. Why? All non-xeon processors have been crippled to not support ecc memory. If money really is on the line there is just no way that is acceptable.

      Well if it's critical then I wouldn't want to use a desktop processor or system in any case, then you should get a proper Opteron/Xeon server with all the validation and redundancy and managed environment. As for the general employee, I've yet to see anyone working on a "workstation" class machine. Unless they need the horsepower for CAD or something like that my impression is that 99.9% from the receptionist to the CEO use regular non-ECC desktops/laptops to do their work and I'm pretty sure that means money is on the line. Of course it could be that we live in an insane world, but I think the risk of someone making a disastrous typo is equal or greater than the risks of a bit flip.

      --
      Live today, because you never know what tomorrow brings
    10. Re:Here's the problem... by Anonymous Coward · · Score: 0

      My mid-grade Asus motherboard supports ECC memory.

      http://www.asus.com/Motherboards/AMD_AM3Plus/M5A97/#specifications

    11. Re:Here's the problem... by CajunArson · · Score: 1

      You have a very interesting definition of "many". I would say that the non-overclocked 8350 beats a 3770K in "a few" multi-threaded benchmarks, usually by a small margin, while still losing by much much larger margins in many other multi-threaded benchmarks (in fact the Phoronix article barely has any lightly-threaded benchmarks in the mix at all).

          The Vishera OC'd to 4.6 GHz wins a few more benchmarks, but, as I said above, all you have to do is apply an easy OC to Ivy Bridge to get up to 4Ghz and Vishera's lead in the few benchmarks it excels at quickly disappears.

      --
      AntiFA: An abbreviation for Anti First Amendment.
    12. Re:Here's the problem... by lopgok · · Score: 5, Informative

      By rare, you mean all Asus motherboards. You have heard of Asus? I presume they are in the consumer space? Look at any AM3+ motherboard, or even AM3 or AM2+. All of them support ECC.

    13. Re:Here's the problem... by Rockoon · · Score: 1

      I would say that the non-overclocked 8350 beats a 3770K in "a few" multi-threaded benchmarks, usually by a small margin, while still losing by much much larger margins in many other multi-threaded benchmarks (in fact the Phoronix article barely has any lightly-threaded benchmarks in the mix at all).

      Yes, thats what you would say, and the last thing you would do is mention that the 3770K costs a whopping 50% more than the FX-3850.

      Obviously your fanboyism has clouded your ability to grasp the situation. The AMD chip which beats this Intel chip in "a few multi-threaded benchmarks" is doing it for significantly less cost.

      If you really want to do a point-for-point comparison, and count up how many benchmark tests a CPU wins or loses by, then you need to note the cost as well, or whats the point of "point-for-point". At $200, the best Intel has to offer is the i5-3470, which the FX-8350 beats the snot out of.

      --
      "His name was James Damore."
    14. Re:Here's the problem... by 0123456 · · Score: 1

      Obviously your fanboyism has clouded your ability to grasp the situation. The AMD chip which beats this Intel chip in "a few multi-threaded benchmarks" is doing it for significantly less cost.

      So Intel are presumably beating the snot out of AMD's margins, as the AMD chip is twice the size of the i7 yet has to sell for a fraction of the price because it's unable to compete.

      And unless you're overclocking, I believe the i7 3770 will have the same performance as the 3770K at a lower price (about $30-40 less last I looked).

    15. Re:Here's the problem... by Rockoon · · Score: 1

      So Intel are presumably beating the snot out of AMD's margins, as the AMD chip is twice the size...

      Typical fanboy knows a fact or two but doesnt know what he is talking about (transistor counts are about equal), and why would I care about AMD's margins anyways?

      You seem to want to poke at AMD without considering what it means to you. I guess if you want to ignore the fact that you have to bend over and let Intel stick it right into you in order to "enjoy" an equally performing chip.. thats fine by me. Have fun paying more.. it makes you look really smart.. really.. it does.. paying more for the same performance is obviously a wise thing to do.

      --
      "His name was James Damore."
    16. Re:Here's the problem... by CajunArson · · Score: 1

      I'm an engineer not a fanboy. I takes a whole lot more engineering talent to design a laptop chip like the 3770K that is faster in the large majority of benchmarks, including the large majority of multithreaded benchmarks, than a chip which is twice the size, has a much much higher transistor budget, uses much larger caches, has a 50 - 100% larger practical power envelope, and runs at 15% higher clockspeeds.

      I could hand a price gun to a homeless guy out on the street and have him slash the price of practically any CPU to be less than an Ivy Bridge. I guarantee you that the same homeless guy isn't going to come up with a brilliant design tweak to turn the FX-8350 into a power sipping breakthrough. Those are just the facts, I even pointed out in my original post that the 8350 is making positive forward progress for AMD, but unlike Trinity that has a decent GPU, there is no real advantage for Vishera in a technological sense, only in a marketing sense. Even there, if you want to play games there are cheaper Intel CPUs that do a better job than Vishera does while also not being power hogs, so now you are down to absolutely needing 8 "cores," not caring one whit about power consumption costs and being unwilling to pay anything extra for even a consumer grade Intel CPU.

      --
      AntiFA: An abbreviation for Anti First Amendment.
    17. Re:Here's the problem... by Rockoon · · Score: 1

      ..blah blah blah..

      What does any of that have to do with the choices facing me? The choice as framed is between the FX-8350 and the i7-3770K. Thats the choice facing me based in this thread of replies that you decided to get involve in.

      You are exhibiting exactly the behavior of a fanboy. You are flailing around the topic on wild tangents that do not pertain to the decision in front of me. The pertinent information is the cost and the performance. How great the Intel engineers are dont matter. How much higher or lower the clock speeds are don't matter. The physical size of the chips don't matter. What matters is what I get and for how much.

      Maybe you didn't realize that you were being a fanboy.. but you were. You did everything *but* talk about what matters.

      --
      "His name was James Damore."
    18. Re:Here's the problem... by CajunArson · · Score: 1

      I'm sure you made these *EXACT SAME* arguments back when the FX-62 cost over $1000 and the Core 2's came out at lower prices while being substantially faster. I'm getting tired of the "AMD is always right drink the Koolaid and die like a good little believer" line.

      --
      AntiFA: An abbreviation for Anti First Amendment.
    19. Re:Here's the problem... by Anonymous Coward · · Score: 0

      I want something to play games with and to transcode Blu-ray discs to mkv. Any recent CPU paired with a good GPU will play all the games fine at maximum detail in the res I use (1920x1200), but transcoding requires more power. 6-core Sandybridge-E chips are powerful, but outrageously expensive, so those are out for me, leaving basically the i7-3770k and the FX-8350. The FX-8350 actually beats the 3770k in the x264 conversion test, plus it's considerably cheaper. Electricity is cheap where I live, so I don't care that it uses a few extra Watts.

      I don't care the means with which it was done. The AMD chip fits my needs and at a better price than Intel, and that's what matters.

    20. Re:Here's the problem... by Anonymous Coward · · Score: 0

      And if you don't get the K you can't overclock, so then the FX8350 can be OC'd and beat it.

    21. Re:Here's the problem... by Rockoon · · Score: 1

      blah blah blah...

      I dont make the kind of argument that you think I make. For starters, I have never considered a $1000 processor to be a good value unless it was a server chip and had something desktop variants didnt (for instance, significantly better I/O) You, however, seem to think its normal to compare a significantly more expensive processor to a significantly less expensive processor, while conveniently completely ignoring the price.

      You are making an obvious statement in a deceptive way. Instead of saying "if you spend more, you get more" you go on about Intel engineers and other bullshit that doesnt fucking mean anything to the truth.. that if you spend more, you generally get more. Well no duh! Thank you captain obvious for conveniently not saying the obvious in order to make a specific brand look better than it should. Now shut the fuck up, fanboy.

      --
      "His name was James Damore."
  5. Shared FPU? by Anonymous Coward · · Score: 1

    Does it still share a floating point unit between two normal cores? For HPC users, that was the worst possible thing to hear about AMD's Bulldozer product. Almost everything in HPC needs good floating point performance and AMD didn't deliver, thus if you buy AMD you have to buy twice the number of cores just to get the FPU count up.

    1. Re:Shared FPU? by OneAhead · · Score: 5, Informative

      I can't believe people keep on complaining about the bulldozer's FP performance. Does nobody realize the shared floating point unit is a 256-bit one, TWICE AS WIDE as the usual 128-bit? Also, assuming you're an HPC user, did you run actual benchmarks? Because we did, and for our (only modestly parallelizable) HPC workload, compiled with a bleeding-edge compiler (not Intel) that supports AVX and running on a bleeding-edge Linux kernel, Bulldozer was remarkably competitive with Intel's offerings at the time, with Interlagos and Westmere getting about the same amount of useful work done per clock cycle. There are some HPC benchmarks on AMDs website that seem very unlikely in light of the mainstream press. However, in light of our own benchmark results, they seem quite reasonable (although we never quite could make it look that good for AMD; probably because AMD didn't go to the same lengths to squeeze the maximum performance out of the Intel systems). Either way, AMD simply blew Intel away on a per-node-price basis, even when compared to Romley. All the way, I was the one arguing that "we should try Intels" based on reviews I saw online, but once we got all the benchmark results in, I simply couldn't argue anymore.

      Also, if AMD's FP performance is truly that abysmal, please explain this? AMD bribed Dell more than Intel so that they now market Bulldozer-based Opterons as "excellent for oil and gas exploration, scientific and medical research, video rendering and other challenging HPC projects"???

      Of course, this is all for a very specific workload and may not hold for all HPC workloads, but I have a strong feeling that even generally spoken, the Bulldozer's FP performance for HPC applications is just fine. It's just that most FP-intensive applications used in most of the benchmarks we're seeing in "end-user" space are not compiled to take full advantage of it and/or not running on an Operating System that takes full advantage of it and/or not very relevant test cases for the Bulldozer's parallel HPC potential. For example, one of the things we found out is that Intel's frequency scaling is more aggressive than AMD's, so Intel suffers badly if you put all the cores on a die to work at once. Also, Intel's improved HyperThreading still ain't worth shit if you saturate the FP units, while AMD's "clustered multithreading" succeeds to squeeze out a significant advantage owing to the fact that not all of our FP code is easily vectorizable so that sharing the 256-bit FP unit between 2 execution threads works better than trying to keep it busy with 1 thread's vectorized instructions.

      /rambling rant

    2. Re:Shared FPU? by Anonymous Coward · · Score: 0

      I suspect the mainstream reviews are getting different benchmarks based on the testing suites - particularly if using the Intel Compiler for any code testing as it won't make decent use of many of the CPU extensions.

      Which means it's a superior chip only if you're not using Windows - great for Linux desktop, servers (particularly non-MS hypervisors), HPC, not so great for Windows desktops (or laptops but then it's not a mobile processor anyway!).

    3. Re:Shared FPU? by Anonymous Coward · · Score: 0

      agree withthis mechanical simulations are great with amd, lots of memory slots and cheap .

    4. Re:Shared FPU? by Anonymous Coward · · Score: 0

      I can't believe people keep on complaining about the bulldozer's FP performance. Does nobody realize the shared floating point unit is a 256-bit one, TWICE AS WIDE as the usual 128-bit? Also, assuming you're an HPC user, did you run actual benchmarks? Because we did, and for our (only modestly parallelizable) HPC workload, compiled with a bleeding-edge compiler (not Intel) that supports AVX and running on a bleeding-edge Linux kernel, Bulldozer was remarkably competitive with Intel's offerings at the time, with Interlagos and Westmere getting about the same amount of useful work done per clock cycle.

      That's pretty sad for AMD considering Westmere doesn't even implement AVX. Did you try Sandy Bridge, which does support AVX?

      Also, if AMD's FP performance is truly that abysmal, please explain this? AMD bribed Dell more than Intel so that they now market Bulldozer-based Opterons as "excellent for oil and gas exploration, scientific and medical research, video rendering and other challenging HPC projects"???

      Uh, you really think marketing copy is all that meaningful? Of course Dell is going to say that machines they sell are an "excellent choice" for some application. They want to sell them!

      Judging by sales figures available to the general public, the percentage of HPC customers for whom AMD is the best fit is fairly small.

      Also, Intel's improved HyperThreading still ain't worth shit if you saturate the FP units, while AMD's "clustered multithreading" succeeds to squeeze out a significant advantage owing to the fact that not all of our FP code is easily vectorizable so that sharing the 256-bit FP unit between 2 execution threads works better than trying to keep it busy with 1 thread's vectorized instructions.

      You have no idea what the fuck you're talking about here. AMD "clustered multithreading" is a different take on hyperthreading intended to provide a large speedup for integer code by providing dedicated integer execution resources to each thread. Because the FPU is shared, for FP code AMD CMT is effectively equivalent to Intel HT. Whatever benefit you see from saturating one FPU with two threads instead of one? It'll be there on Intel too.

      Judging by your sloppy ill-informed post I'd guess that you don't actually code (or at least are not one of the low level optimization wizards) and are severely garbling information you picked up from your coworkers.

    5. Re:Shared FPU? by OneAhead · · Score: 1

      That's pretty sad for AMD considering Westmere doesn't even implement AVX.

      It's not sad at all considering that our code only gets a modest boost from AVX (as I said, a lot of it is not easily vectorizable) and that that similarly-priced AMD parts are clocked way higher.

      Did you try Sandy Bridge, which does support AVX?

      FYI, Romley (which I talked about earlier) was Intel's code name for Sandy Bridge-EP, mister Anonymous Expert. To give a little bit more information, on a per-clock basis, it ran our code about 35% faster than than Westmere. But on a per-price basis, AMD still won comfortably.

      Uh, you really think marketing copy is all that meaningful?

      No, but it supports our benchmarks, which are (at least for us).

      Judging by sales figures available to the general public, the percentage of HPC customers for whom AMD is the best fit is fairly small.

      Gee, I wonder why. Can't be because all that bad press and people bashing the "shared FPU"... We almost bought it too; we were ready to skip Bulldozer in favour of Nehalem, if it wouldn't have been for some of our people who insisted on elaborate benchmarks.

      You have no idea what the fuck you're talking about here. AMD "clustered multithreading" is a different take on hyperthreading intended to provide a large speedup for integer code by providing dedicated integer execution resources to each thread. Because the FPU is shared, for FP code AMD CMT is effectively equivalent to Intel HT. Whatever benefit you see from saturating one FPU with two threads instead of one? It'll be there on Intel too.

      I apologize for going too fast, mister Anonymous Expert. Allow me to rectify my mistake by explaining it a bit more elaborately. So, as I said, our code is only modestly vectorizable. It also has a significant amount of branching. Therefore, we think that on AMD, a single thread will not saturate the 256-bit FPU (which can only be accomplished through vectorization). Starting a second thread on the same module leads to the FPU being used more efficiently. This manifests itself in a nice performance boost between using half the logical cores on the AMD chip and using all of them. By comparison, doing the same with the Intel part gave us no performance boost at all which we speculate is because the FPUs (or its data pathway) is already saturated with a single thread. For Westmere, you could say that this has nothing to do with CMT and everything with 2 threads sharing a 256-bit FPU on AMD vs a 128-bit FPU on Intel, but we're seeing the same thing on Sandy Bridge, so we speculate that SMT vs CMT may have something to do with it. I'm looking forward to hearing your opinion on this matter, mister Anonymous Expert. But then again, you're clearly not one of the low level optimization wizards either, are you?

    6. Re:Shared FPU? by Anonymous Coward · · Score: 0

      First, let me say... I was quite an asshole there, wasn't I? My only excuse was having my threshold for "probably an idiot fanboy" set to a hair trigger after slogging through a typical slashdot CPU flamewar (i.e. one full of fanboys), and that's not a good excuse. Apologies.

      FYI, Romley (which I talked about earlier) was Intel's code name for Sandy Bridge-EP, mister Anonymous Expert. To give a little bit more information, on a per-clock basis, it ran our code about 35% faster than than Westmere. But on a per-price basis, AMD still won comfortably.

      You wrote:

      Because we did, and for our (only modestly parallelizable) HPC workload, compiled with a bleeding-edge compiler (not Intel) that supports AVX and running on a bleeding-edge Linux kernel, Bulldozer was remarkably competitive with Intel's offerings at the time, with Interlagos and Westmere getting about the same amount of useful work done per clock cycle.

      You only mentioned Interlagos and Westmere when discussing what you'd tested. That's why I didn't think you'd tested Sandy Bridge.

      Gee, I wonder why. Can't be because all that bad press and people bashing the "shared FPU"... We almost bought it too; we were ready to skip Bulldozer in favour of Nehalem, if it wouldn't have been for some of our people who insisted on elaborate benchmarks.

      I think you'll find most serious HPC users are willing to do benchmarking instead of always taking someone else's word for it. If you're going to fill a few racks full of compute nodes, it's pretty easy to justify buying one small node of each flavor to do some competitive evaluation, or perhaps rent time on someone else's systems.

      So, as I said, our code is only modestly vectorizable. It also has a significant amount of branching. Therefore, we think that on AMD, a single thread will not saturate the 256-bit FPU (which can only be accomplished through vectorization). Starting a second thread on the same module leads to the FPU being used more efficiently. This manifests itself in a nice performance boost between using half the logical cores on the AMD chip and using all of them. By comparison, doing the same with the Intel part gave us no performance boost at all which we speculate is because the FPUs (or its data pathway) is already saturated with a single thread. For Westmere, you could say that this has nothing to do with CMT and everything with 2 threads sharing a 256-bit FPU on AMD vs a 128-bit FPU on Intel, but we're seeing the same thing on Sandy Bridge, so we speculate that SMT vs CMT may have something to do with it. I'm looking forward to hearing your opinion on this matter, mister Anonymous Expert.

      Warning, wall of text incoming.

      Sandy Bridge has a smaller branch mispredict penalty (because its execution pipeline isn't as deep as Bulldozer's -- BD targets higher frequencies and that's the tradeoff for it), and significantly better branch predictors too. Since you say your software is branchy, I suspect Bulldozer needs to fill branch mispredict bubbles with instructions from the other thread.

      Another possible source of more execution bubbles on Bulldozer is load-to-use latency. Past L1, Bulldozer's memory hierarchy is a lot slower than Westmere or Sandy Bridge, and its L1 caches are a bit undersized so the L1 hit rate isn't always great either. Usually FP HPC code is relatively insensitive to memory latency, but then it's usually not branchy either so maybe there's two ways your code is atypical.

      The thing which most annoyed me is that you spun this as a positive for AMD and negative for Intel. That just isn't a good way of thinking about it. Filling pipeline bubbles with instructions from the other thread is of course what hyperthreading (or sharing the FPU in AMD's hybrid design) is all about, so on the one hand it is a good thing that Bulldozer can do it. However, if Intel can avoid those bubbles (and the need t

    7. Re:Shared FPU? by OneAhead · · Score: 1

      First, let me say... I was quite an asshole there, wasn't I? My only excuse was having my threshold for "probably an idiot fanboy" set to a hair trigger after slogging through a typical slashdot CPU flamewar (i.e. one full of fanboys)

      No problem, I can relate. I tend to visit /. predominantly when I'm too tired to do anything useful (like some people plop in front of the TV) so I tend to be a bit short on patience and judgement myself. My original post is a good example of this; I got ticked off by the person starting the thread bluntly stating that (paraphrased) the shared FPU kills the Bulldozer for HPC applications. To me, this is simply false; the shared FPU was twice as wide as the (then-competing) Westmere, so all else being equal (which obviously isn't the case), the architecture of the bulldozer module with a 256-bit FPU supporting 2 "virtual cores" would be expected to give similar FP throughput than 2 Westmere cores (supporting 4 "virtual cores") with two 128-bit FPUs. In the case of poorly vectorizable code with non-optimal parallel scaling (which happens to be our case), the shared 256-bit FPU even has a distinct advantage because you need half as many parallel threads to benefit from, as you put it, branch mispredict bubbles being filled in with instructions from the other thread. In my opinion, the shared FPU by itself is fundamentally not a bad design and is totally not the reason for Bulldozer's competitive failure against Westmere; the real reasons here are many and slightly more subtle. Admittedly, that's a bit of a historical discussion since Sandy Bridge came out not so long afterwards touting one 256-bit FPU per core. But it still it infuriates me that people keep on blaming the shared FPU, which explains why my original post was quite rambling. I can easily understand how it set off your BS detector. Anyhow, I'm glad that we succeeded to turn a potential flamewar into a civilized discussion and I appreciate your insights.

      I think you'll find most serious HPC users are willing to do benchmarking instead of always taking someone else's word for it. If you're going to fill a few racks full of compute nodes, it's pretty easy to justify buying one small node of each flavor to do some competitive evaluation, or perhaps rent time on someone else's systems.

      There are large numbers of small to medium-sized HPC shops out there that don't even come close to the top 500 but probably contribute significantly to chipmakers' HPC sales. This is our situation; my boss was somewhat annoyed at the loss of time associated with our elaborate benchmarks. In the end, I think we did right, but it did delay our deployment by a couple of weeks. Time is money; unless there are clear red flags, bosses in this situation typically prefer speedy deployment over a small chance (remember our benchmark results were completely against expectations) of getting a financially slightly better deal. Therefore, they tend to default with what they know and trust, and that's often "Intel Inside". Again, we almost did just that. It's difficult to quantify the impact of this effect on overall sales, but I wouldn't dismiss it, especially given how overwhelmingly negative the Bulldozer's press coverage is.

      Sandy Bridge has a smaller branch mispredict penalty (because its execution pipeline isn't as deep as Bulldozer's -- BD targets higher frequencies and that's the tradeoff for it), and significantly better branch predictors too. Since you say your software is branchy, I suspect Bulldozer needs to fill branch mispredict bubbles with instructions from the other thread.

      The conditions for branching in our software are mostly based on simulation data that is inherently chaotic; it's hard for me to imagine a branch predictor doing much better than 50%, no matter how good it is. You're right that Bulldozer will have bubbles, but so will Sandy bridge. That (and the fact that I didn't think our code is that bra

    8. Re:Shared FPU? by Anonymous Coward · · Score: 0

      I got ticked off by the person starting the thread bluntly stating that (paraphrased) the shared FPU kills the Bulldozer for HPC applications. To me, this is simply false; the shared FPU was twice as wide as the (then-competing) Westmere, so all else being equal (which obviously isn't the case), the architecture of the bulldozer module with a 256-bit FPU supporting 2 "virtual cores" would be expected to give similar FP throughput than 2 Westmere cores (supporting 4 "virtual cores") with two 128-bit FPUs.

      Fully agree with that.

      Really, the only thing I've ever heard as a downside to the Bulldozer FPU has nothing to do with the fact that it's shared. Instead, it's that AVX instruction set support was a relatively late addition to Bulldozer and it isn't as efficient as it could be. Supposedly Bulldozer cracks each 256-bit wide AVX instruction into two 128-bit micro-ops, rather than being able to gang the two 128-bit execution units into a single 256-bit unit. While this is theoretically the same throughput, a thread using AVX instructions does generate 2x as many "instructions", putting extra pressure on out-of-order execution resources (rename registers, result forwarding networks, scheduler entries, etc).

      But it's not like I've ever heard of this being any kind of fatal flaw. In fact, from benchmarks I've seen it looks like Bulldozer usually gains performance if you switch to AVX instead of SSE2/3, so if the effect is real it's probably pretty small.

      The conditions for branching in our software are mostly based on simulation data that is inherently chaotic; it's hard for me to imagine a branch predictor doing much better than 50%, no matter how good it is.

      Not necessarily. If you have an inner loop containing an if/else which depends on sim data that's static across multiple iterations of the loop, that's good enough to work with. Branch predictors are mostly based on special purpose caches ("branch history tables"). BHTs are tagged by the branch instruction address rather than any data items the branch depends on, so all you need is multiple executions of the same branch using the same data input and you'll get useful tracking of branch history.

      But if a given branch instruction site never uses the same calculated data twice, you're absolutely right. It's a coin flip. History tables don't usually have enough bits of state to record something as sophisticated as the true average probability of taking the branch, either.

      The ideal computer does not exist. Faster cores take more silicon so you'll have less of them.

      For sure. The fact that we have multi-core CPUs at all is a consequence of that. I was basically just saying that even in the multi-core era, faster individual cores are marginally closer to the Utopian computer, so if there aren't any real world tradeoffs to get them it's a good thing. Ignoring end user price, Sandy Bridge easily passes that test -- SB cores are smaller than BD modules (both in a 32nm process), they're faster, and they use less power.

      The game AMD has to play (and is playing) is to make lemonade because they've got lemons. They can't compete head to head with Intel in all market segments, but by undercutting Intel's prices they can make themselves a very attractive value in niches. It just hurts a bit is all... an 8-core/16-thread SB-EP has a single 435mm^2 die, while an 8-module/16-core Interlagos has two 315mm^2 4-module dies or 630mm^2 total. AMD's package is likely to be more expensive too, since it's a MCM.

      Also thanks for pointing out the specifics of the SpecFP results. I'm not sure how to relate them to our results and my post is already more than long enough, but it's generally interesting.

      No problem, and do be wary of the comparison: the "rate" versions of SPEC benchmarks test multi-core computers by starting N copies of each nominally single-threaded SPEC sub-benchmark. There is no inter-thread communication or locking, and the "threads" don't even share an address space.

  6. tl;dr version by gman003 · · Score: 5, Informative

    New AMD processor, higher clocks than the last one but no massive improvements performance-wise. Still rocks at multi-threaded, integer-only workloads, still sucks at single-threaded or floating-point performance, still uses a huge amount of power. AMD giving up on the high end, their top-end parts are priced against the i5 series, not the i7. Since Intel's overpricing stuff, they're still roughly competitive. Might be good for server stuff, maybe office desktops if they can get the power down, but not looking good for gaming. Overall mood seems to be "AMD isn't dead yet, but they've given up on first place".

    There. Now you don't need to read TFAs.

    1. Re:tl;dr version by theswimmingbird · · Score: 0

      I built a new rig back in March, went with the Phenom II 975 Black Edition. Haven't been disappointed, as it's still a good bit faster than the Core2 P8400 in my laptop. I'm still holding out hope AMD comes out with a better gaming processor, though, so I don't have to change motherboards when I do decide to upgrade...

    2. Re:tl;dr version by Anonymous Coward · · Score: 0

      OMG... your brand new AMD system is faster than an intel system from 2008. Watch out Intel!

      http://ark.intel.com/products/35569/Intel-Core2-Duo-Processor-P8400-3M-Cache-2_26-GHz-1066-MHz-FSB

    3. Re:tl;dr version by Anonymous Coward · · Score: 0

      For varying definitions of "brand new," I suppose. You do know that the 975 launched in 2010, right? Doesn't quite read into your zinger as well that way.

    4. Re:tl;dr version by 0123456 · · Score: 1

      So AMD's fastest desktop CPU from 2010 beats an Intel laptop CPU from 2008? Call me impressed.

    5. Re:tl;dr version by Anonymous Coward · · Score: 0

      Not looking good for gaming?

      What are you smoking? The games run and run fine. If you HAVE to run at a bajillion FPS, that's enthusiast-level and for that you'll have already bought intel.

      Weak comment.

    6. Re:tl;dr version by gman003 · · Score: 1

      Check some of the benchmarks. In most, you're looking at a 10-15% drop in framerates - and that causes problems in at least a few cases. Starcraft 2, with graphics set low enough that it's purely a CPU bottleneck, the top Vishera gets only 48fps. Last year's much-loved i5-2500K gets 65, and every other Intel processor is higher.

      I would have loved to see benchmarks from more CPU-bound games, though. GTA IV is well-known to be a heavy CPU user, but none of the reviews I read used it. Same for Minecraft - although that's often memory-bandwidth-bound, not CPU-bound.

      But in any case, yes, it seems AMD processors aren't good for the more dedicated gamers, the ones who will actually be choosing a processor based on its gaming capabilities. Although they do seem to have an edge for the lower-end gaming laptops due to Trinity - their integrated GPU is a competent replacement for medium-low dedicated cards, quite respectable in the smaller gaming laptops, might also be good for pseudo-consoles.

    7. Re:tl;dr version by gman003 · · Score: 1

      The Phenom IIs were indeed pretty good processors. I was a particular fan of the X3s - they were quad-core dies that had one defective core, but were priced only a hair above their dual-cores. And they usually had all the cache of the quad-core they were based on. A very powerful low-cost processor.

      I'm still sort of surprised AMD didn't revert back to the Phenom design, which seemed like it had a good amount of life left in it. Make turbo a universal option, throw some deeper cache into it, see if you can widen the decoder or something, and you'd have something at least as competitive as Bulldozer/Piledriver, I think.

    8. Re:tl;dr version by yurtinus · · Score: 1

      Lost in this discussion is the cost benefit. I went with a Phenom II 945 when I rebuilt my computer (CPU, RAM, mobo) two or three years back and ended far cheaper than an i7 while not being too terribly far behind in performance. If you want the fastest, sweet - get an Intel. AMDs offerings do stack up quite well in performance per dollar. It's not their glory days of yore when they were cheaper *and* faster than Intel, but cheaper and not that much slower still isn't a bad place to be.

      --
      +1 Disagree
    9. Re:tl;dr version by Anonymous Coward · · Score: 0

      To be fair, it was an off-the-cuff comparison using on-hand equipment. The Core2 Duo of that era was literally half the processor that the Phenom II 975 was -- they weren't even in the same ballpark. You go right on trying to imply that the latter was barely better than the former, though, while the rest of us who understand the very obvious power consumption differences between laptop/desktop CPUs, especially of differing generations, laugh heartily at your ignorance.

      Stupid Wintel fanboi is stupid.

    10. Re:tl;dr version by Anonymous Coward · · Score: 0

      Bullshit. I'm a heavy SC2 player, and my 1090T/GTX560 has never had a problem giving me high frame-rates at maximum settings -- never a lag spike that cannot be attributed to network latency. Period. You're full of shit.

      And seriously... Minecraft? Really? That's your benchmark of choice? You're a grade-A idiot, bro...

  7. Nice CPU, high power usage by coder111 · · Score: 4, Informative

    I've read through some of the reviews. It looks like a nice CPU with a bit too high power usage for my taste.

    And please take benchmark results with a pinch of salt- most of them are compiled with Intel compiler, and will have lower results on AMD CPUs just because Intel compiler will disable a lot of optimizations on AMD CPUs.

    I don't know of any site which would have Java application server, MySQL/PostgreSQL, python/perl/ruby, apache/PHP, GCC/llvm benchmarks under Linux. Video transcoding or gaming on Windows is really skewed and nowhere near to what I do with my machine.

    --Coder

    1. Re:Nice CPU, high power usage by ifiwereasculptor · · Score: 2

      I think Phoronix took Vishera through some GCC tests. They have another article about GCC optimizations for the Bulldozer architecture in general (it seems to improve some workloads by quite a bit).

  8. athlon mp baybeeee by Anonymous Coward · · Score: 0

    I could love AMD again if they allowed Opteron overclocking.

  9. 8 cores + poor memory performance by Anonymous Coward · · Score: 0

    What were they thinking? The more cores, the more memory bandwidth you need.
    This is what AMD gets for outsourcing its engineering. (look up articles from 2007)

    If anyone has doubts or thinks this is trolling, go see the benchmarks for yourself.

    http://www.overclock3d.net/reviews/cpu_mainboard/amd_vishera_fx8350_piledriver_review/3

    1. Re:8 cores + poor memory performance by Billly+Gates · · Score: 1

      The bandwidth is the same. Intel is doing mathematical tricks by guessing the next instruction before it is even fetched in hardware with 90% accuracy. In essence it doubles performance for small burst of loads and since it is patented AMD has an inherent disadvantage.

      Thats what the ivybridge icores did that the sandy bridge ones didn't that nailed the coffen for AMD which was already struggling.

  10. For linux... by ak3ldama · · Score: 5, Insightful

    Here are a set of benchmarks that are more centered on the Linux world from phoronix and are thus a little less prone to intel compiler discrimination. The results seem more realistic: better and worse and similar to an i7 at different work, still hard on power usage, low purchase price.

    --
    "but money is the God of Algiers & Mahomet their prophet." - Rich. O'Bryen June 8th 1786
    1. Re:For linux... by turgid · · Score: 1

      Thanks.

    2. Re:For linux... by Anonymous Coward · · Score: 0

      Good. x264 encoding is better than the i7's. Might try one of these out for this.

  11. Perfect for the 99% by EmagGeek · · Score: 2

    The new Trinity, and now these FX Procs, are perfect for "the 99%," that is to say for what 99% of people do with their machines: surf the web, check email, maybe do some photo editing or piecing together home movies.

    They're cheap, reasonably fast, and support all the latest extensions and optimizations. Plus, even for enthusiast prosumers who want to screw around with things like virtualization, you can get into the IOMMU space cheaply with AMD, which is nice for a platform like ESXi that has a mature PCI passthrough implementation.

    I think the Trinity A-10 does really hit the sweet spot for most consumers. It has reasonably fast processing and reasonably fast GPU in a ~$130 package. If you're putting a box together on the cheap, it becomes quite compelling.

    1. Re:Perfect for the 99% by FreonTrip · · Score: 2

      Don't underestimate these things for video encoding or playing around with scientific computing. They're great for embarrassingly parallel computing problems, and the price is very good.

    2. Re:Perfect for the 99% by 0123456 · · Score: 1

      The new Trinity, and now these FX Procs, are perfect for "the 99%," that is to say for what 99% of people do with their machines: surf the web, check email, maybe do some photo editing or piecing together home movies.

      I did all those things on a Pentium-4.

      I agree about AMD's low-end CPUs, but why would you want an 8-core CPU to do any of those things? Does an email program really need eight cores these days?

    3. Re:Perfect for the 99% by JDG1980 · · Score: 2

      The new Trinity, and now these FX Procs, are perfect for "the 99%," that is to say for what 99% of people do with their machines: surf the web, check email, maybe do some photo editing or piecing together home movies.

      If you're building a "good enough" system for a non-technical user, why in the world would you even consider a Vishera FX CPU? It's expensive, power-hungry, and has a high TDP. And it doesn't even have integrated graphics, so you'd have to add the expense of a discrete graphics card.

      For an inexperienced user who doesn't need much processing power, a Bobcat board will get the job done for cheap, and if that isn't quite good enough, the Pentium G630T is very competitively priced and (with its 35W TDP) very efficient. For enthusiasts, especially gamers, the Sandy/Ivy Bridge "K" series CPUs offer better performance, better efficiency, and more overclocking headroom, for only a little more cash. (In fact, if you have a Micro Center in your area, you can score an i5-2500K for $160, or an i5-3570K for $190.)

      What's the intended audience for Vishera – people who do a lot of x264 encoding and don't want to step up to a server CPU or a Sandy Bridge-E? I'm having a hard time seeing a substantial audience for this thing.

    4. Re:Perfect for the 99% by Billly+Gates · · Score: 5, Interesting

      I am typing this on a Phenom II 6-core system. It is quiet, 45 watts, and at the time (2010) it was only 10-15% slower than an icore5. What did I get that the intel icore5 didn't?

      - My whole system including the graphics card was $599! Also an Asus motherboard by the way too and part of their extended warranty boards.
      - Non crippled bios where I can run virtualization extensions (most intel mobos turn this off except on icore7s)
      - 45 watts
      - My ati 5750 works well with the chipset
      - the AM3 socket can work with multiple cpus after bios updates.

      What the icore5 has
      - It is made by intel
      - It is 15% faster
      - The cost of the cpu alone is 2x the price and I can pretty much include a motherboard as well if you are talking up to icore7s.

      An icore7 system costs $1200 at the store. An icore5 gaming system similiarly specced cost $850 and does not include virtualization support to run VMWare or Virtualbox.

      The FX systems ... ble. I am not a fan. But for what I do AMD offered a quieter cheaper system that could run VMs of Linux and can upgrade easier. To me my graphics card and hard drive are the bottlenecks. I would rather save money on the cpu. I was so hoping AMD would use this to have a great graphics for tablets and notebooks :-(

    5. Re:Perfect for the 99% by Billly+Gates · · Score: 1

      These things are monsters. I do not like loud space heater machines that require large desktops but only match the lowest performing cpus in the market.

      Do not get me wrong I am typing this on a phenomII and I am an ATI/AMD fan. These are not the better phenomIIs. I would have preferred it if AMD admitted their new lineup is flawed and sold more up to date phenomIIs.

      I pray AMD survives but these things are even further behind icore3s. I like the extra cores do not get me wrong but the market is heading towards notebooks and tablets and these things just are not up to the task at 130 watts!

    6. Re:Perfect for the 99% by Rockoon · · Score: 1

      Agreed on the core count. I too have a 6-core Thuban and I would never go back to less.

      Sure, when all I am doing is surfing the web then it doesnt matter, but most of the time that I am surfing the web I've got other processes going that are actively doing work.. on top of the browsing I frequently have netflix/hulu running, and either a big compile or a video encode is also sometimes taking place (in fact, when something really time consuming begins, I am *always* browsing the web during it)

      The system remains highly responsive all the time, and I'm 100% certain that equally priced Intel chips are only 2 core at most and would never remain highly responsive. 6-core AMD chips can be found for $120 these days. Intel doesnt have a $120 part that competes with them. Intel fans make excuses about this, but thats the reality.

      --
      "His name was James Damore."
    7. Re:Perfect for the 99% by Rockoon · · Score: 1

      What's the intended audience for Vishera

      The audience is people that want the most performance they can get for their $140, and thats not the "I shopped around to find a great price on CPU's that have piled up in someones inventory" price.. thats the "I went to the place the most people trust" price.. The straight NewEgg introductory price for the FX-6300 Vishera is $139.99.

      --
      "His name was James Damore."
    8. Re:Perfect for the 99% by EmagGeek · · Score: 1

      I think you need to understand the meaning of Thermal Design Power.

      TDP is not the amount of power the CPU sits there consuming all the time. TDP is the maximum amount of energy that can be dissipated in the physical package with the factory-provided cooling device. A TDP of 100W does not even mean the processor will ever consume that much under the most heavy loads.

      In fact, there is a lot of margin built into the TDP rating, to ensure that normal degradation over time of the cooling device does not result in exceeding the maximum junction temperature of the processor.

      Believe it or not, the guys who design processors are pretty smart people, and a CPU sitting there doing nothing is not consuming its TDP.

      As far as not needing a Vishera, I mentioned both the Vishera and the Trinity, and the Trinity is what I said was cheap and covers all the bases. The lower-end trinities are $75 - half the price of a basic i3, and 99% of end users will see no difference whatsoever.

    9. Re:Perfect for the 99% by Kjella · · Score: 2

      I am typing this on a Phenom II 6-core system. It is quiet, 45 watts (...) - 45 watts

      I guess if the facts don't support your argument, make something up. All the Phenom II 6-core CPUs have either 95W or 125W TDP. But yes, the X6 was a quite competitive chip by offering you 50% more cores for about the same money as an Intel quad. Anandtech's conclusion did have a prelude to what was coming though:

      You start running into problems when you look at lightly threaded applications or mixed workloads that aren't always stressing all six cores. In these situations Intel's quad-core Lynnfield processors (Core i5 700 series and Core i7 800 series) are better buys. They give you better performance in these light or mixed workload scenarios, not to mention lower overall power consumption.

      Let's look at what has happened since 2010:
      Cinebench R10 single-threaded:
      Intel Core i5 750: 4238
      Intel Core i5 3570K: 6557
      AMD Phenom II X6 1090T: 3958
      AMD FX-8350: 4319
      Intel has improved 55%. AMD? 9%.

      Load power consumption:
      Intel Core i5 750 (95W TDP): 140W
      Intel Core i7 3570K (77W TDP): 101W
      AMD Phenom II X6 1090T (125W TDP): 201W
      AMD FX-8350 (125W TDP): 195W
      Intel has improved 28%. AMD? 3%. And Intel includes an IGP in those 77W, AMD doesn't.

      If this was a boxing match, I'd say Intel is throwing all the punches and AMD is taking them. Even if both are on their feet, one of them is in really big trouble.

      --
      Live today, because you never know what tomorrow brings
    10. Re:Perfect for the 99% by O('_')O_Bush · · Score: 1

      You are cherry picking numbers, we already know the single threaded performance isn't as strong as the multithreaded performance (where it matters, for many). Also, TDP is a measure of worst case power usage, not average case as the GP describes.

      --
      while(1) attack(People.Sandy);
  12. Perfect for parents PC by hibble · · Score: 1

    CPU's tends not to be the bottle neck these days its either the hard drive or slow internet connection. At work I priced up some AMD classroom PC's and in bulk we could have given each pc a 128GB SSD(all user work kept on a SAN and system image is about 60-80GB once MSI's have been deployed) with the savings of going AMD but as the boss had never used AMD and reviews keep going on about how intel is faster he went for a mix of I5 and I7 intel based pc's with little ram so as to keep the cost down and as a result startup times are no better than the old intel core duo pc's they replaced. To top it off the boss now has my demo AMD setup as 'its the fastest PC' admittedly that’s only due to a SSD not the CPU but I do get his I7. How do I find a excuse to put a large RAID of SSD's in it?

    1. Re:Perfect for parents PC by 0123456 · · Score: 2

      CPU's tends not to be the bottle neck these days its either the hard drive or slow internet connection.

      AMD fanboys have been saying that for years as AMD lagged behind Intel's CPU performance. But they then keep telling people they should buy AMD 8-core CPUs instead of Intel 4-core, when, if they really believed what they were saying, they would be telling people to buy the cheapest dual-core available, or an Atom.

      Seriously, why would you possibly think an 8-core CPU was 'perfect' for your parents' PC, no matter who makes it? If they're not CPU-limited, buy the cheapest CPU you can get.

    2. Re:Perfect for parents PC by Animal+Farm+Pig · · Score: 2, Funny

      For non-technical parents and other users, I actually like lots of cores and lots of memory-- more so than 'power users.'

      Have you ever seen some people boot up their machines? It will take 5 minutes because of the sheer amount of crap installed.

      It will be shit like two different anti-virus suites (the first one's subscription expired, and they installed a new one without uninstalling the old). There will be a update checker for every conceivable thing that's been installed. There's the preloader software that loads whatever crappy application it is into memory at boot time so that the application starts faster when you click on it's associated file type. Then, there are all the various helpful toolbars for Internet Explorer that "enhance your internet experience" and delivery dollars to your inbox. Of course, the user will probably go through two or more inkjet printer/scanner/copier all-in-one devices that require the manufacturer's 300 MB "printing experience suite" to enhance user experience by making pop-up windows when the ink is low and will helpfully tell you where you can order more special photo printing paper. There's the two different desktop search tools that got installed when the user downloaded the kitten screensavers pack and accompanying mouse cursors. Then, there is the sidebar app showing the weather outside and what time it is in Fiji.

      That's stuff I can think of off the top of my head. I have no idea how average users manage to cruft up their machines so much, but they do. Keep in mind also that they're probably not going to upgrade hardware for a few years, but they'll likely install iTunes 32, Internet Explorer 17, and Office 2016, which will have >1 GB memory footprints.

      By having lots of cores, and lots of memory, you can give them a decent user experience even on a machine that they've bogged to shit.

    3. Re:Perfect for parents PC by 0123456 · · Score: 2

      Have you ever seen some people boot up their machines? It will take 5 minutes because of the sheer amount of crap installed.

      So uninstall the crap or install an SSD. A faster CPU makes far less difference to boot time than an SSD, because boot time is mostly limited by the seek performance of the disk as it loads all that crap into RAM.

    4. Re:Perfect for parents PC by Billly+Gates · · Score: 1

      Oh I would also like to add the answer for multi-core are browsers! ... and applets as Windows 8 is about to come upon us. I thought a 6 core was excessive but I use VWare. When not I have 20 tabs open in Chrome or Firefox. IE 9 and Chrome makes each tab a seperate process and is very CPU scalable. Believe it or not I had 30+ tabs opened and have noticed 3 out of the 6 cpus doing a ton of work running some terrible actionscript or javascript several times.

      HTML 5.1 or 5 (depending on W3C or Whatg implementation) supports websockets and webworkers where elements in the page can be spawn into a seperate thread/process as workers and children. Metro does this with its apps and they are always running even if you close them. A multi-core cpu will make life easier if your system has been on for awhile when you have 8 metro apps running and 15 tabs opened with half running html 5/5.1 in the next few years ahead.

    5. Re:Perfect for parents PC by Smauler · · Score: 1

      My core 2 duo with 2gb RAM booted Vista in 15 seconds. What operating system do you need more cores or RAM to boot with? For that matter, how do more cores and RAM really help booting that much?

      It only had 2gb because of a bug in which it would not install with 4gb with my motherboard. It was soon upgraded to 4gb after a hotfix, and now has 8gb because one of the original sticks died, and there were 2 4gb sticks floating about. I'm not sure how long it takes to boot now, but it'll probably be a lot longer than 15 seconds. However, my current uptime is 1649 hours.

    6. Re:Perfect for parents PC by Animal+Farm+Pig · · Score: 1

      The boot time isn't the point-- it's an indicator of the problem. The problem is that Grandma doesn't know how to maintain her computer. It's full of crap (see above post). That crap typically loads at boot time. After it's loaded, it sits around stealing processor cycles and memory.

      My core2duo laptop with 2GB of RAM running Windows 7 works great. I'm not compiling a lot of code on it, but it works for basic computer tasks. If I give my laptop to Grandma, after one year I expect it to be painful to use for basic computer tasks. See above re: crap.

      So, if we want Grandma to have a computer to use that doesn't run dog slow, we have three options:

      1. Train Grandma not to install stupid shit, and not to click 'OK' and 'Accept' without reading what it's for, and how to use msconfig.
      2. Constantly maintain Grandma's computer to keep it free of crap.
      3. Give Grandma a computer with a powerful multicore processor and lots of memory.

      For most users, #1 is a lost cause. They just don't get it. If you have lots of free time, fuck it, do #2. Otherwise, it's better to do #3.

      TLDR For basic computer uses, "Grandma" needs a significantly more powerful computer than you.

    7. Re:Perfect for parents PC by Anonymous Coward · · Score: 0

      "Train Grandma not to install stupid shit, and not to click 'OK' and 'Accept' without reading what it's for, and how to use msconfig.
              Constantly maintain Grandma's computer to keep it free of crap.
              Give Grandma a computer with a powerful multicore processor and lots of memory."

      4. Put Grandma on a different OS(bsd/linux) so she can't download the crap in the first place. Even if she managed to download it, it won't do anything at all so she'll learn not to bother. Problem goes completely away.

  13. Finally by Spiflicator · · Score: 1

    a CPU for the low-end enthusiast. Quite the niche market.

    1. Re:Finally by kriebz · · Score: 1

      Me and the $60 A4 I'm typing this on disagree. Replaced the socket 754 1st gen Athlon 64 that wasn't quite cutting it 6 years later. Plenty of enthusiasts want a cheap 2nd PC for playing with OSs, home server, etc. The Pentium E5200 I bought a couple of years ago (Intel's version of this niche) is not aging very well with no virt and DDR2.

  14. Vishera Piledriver Trinity Bulldozer FX-8350 i5-25 by Anonymous Coward · · Score: 0

    As Hobbes predicted, language has become a complete impediment to understanding. Vishera Piledriver Trinity Bulldozer FX-8350 i5-2550K i7-3770K.

  15. Phoronix is great but I want more by coder111 · · Score: 4, Interesting

    I read phoronix a lot these days. I find more technical news in there than in Slashdot.

    However, their benchmarks are often flawed. For example they did a Linux scheduler benchmark recently which measured throughput (average this or that) and not latency/interactivity (response times), which was totally useless. Well, ok, you can consider it a test checking for throughput regressions in interactivity oritiented schedulers, but it did not measure interactivity at all.

    And regarding their Vishera benchmark, they measured most of their standard stuff, mostly scientific calculation, video/audio encoding, image processing, rendering. I very rarely do any of this.

    The developer related benchmarks they had were Linux kernel compilation times (Vishera won), and you might count OpenSSL as well. They didn't do PostgreSQL, they didn't benchmark different programming languages, nor application servers, nor office applications, nor anything that would really interest me. I wish someone would measure Netbeans/Eclipse and other IDE performance.

    And anyway, did you notice that AMD usually does much better in Phoronix reviews than in Anandtech/Toms Hardware/whatever sites? That's because Phoronix doesn't use Intel Compiler nor Windows, so results are much less skewed.

    --Coder

    1. Re:Phoronix is great but I want more by ifiwereasculptor · · Score: 1

      You're talking about the low-jitter Linux kernel testing where they didn't test jitter at all, right? I've seen that one, it was embarrassing. However, their often incomplete (and sometimes flawed) testing is still the best we can get for Linux.

      Linux already has scheduler fixes for Bulldozer, too, unlike Windows 7. Not a game changer, but it does grant another 1-3% performance, if I remember correctly. not the the Intel compiler doesn't cripple AMD quite a bit - just run Handbrake on Windows on both platforms and compare the delta of the results to the one you'd get on, say, Adobe's Premiére. I wonder if Mozilla uses the ICC for Widnows builds, because they are noticeably slower on my Athlon II.

    2. Re:Phoronix is great but I want more by Anonymous Coward · · Score: 0

      You used to be able to see the Visual Studio compiler at "about:buildconfig" now it points to some python script, but the windows version uses a different compiler than the linux version

  16. Re:Wattage costs by hoboroadie · · Score: 1

    More heat equals louder fans. and more dust on the vents.
    The limosine tax imposed by our ISP overlords is a couple orders of magnitude more painful.

    --
    They feared that it could be used to suppress protest or support unpopular rule.
  17. SATSQ by JDG1980 · · Score: 4, Informative

    AMD FX-8350 Review: Does Piledriver Fix Bulldozer's Flaws?

    No. It still guzzles power like crazy compared to Sandy/Ivy Bridge, and its single-threaded performance still sucks royally. (And that's still very important since many, many programs cannot and will not ever support full multithreading.)

  18. A space heater included by Billly+Gates · · Score: 3

    Man I really want AMD to win!

    I am typing this on a phenom II which is a better chip in my opinion and fast at the time in (unfortunately in 2010 standards). But these things run well over 130 watts, are loud with huge freaking fans, 4.4 ghz, and it seems AMD is trying to pump out as much speed as possible to beat intel's lowest end chips.

    Just call it pentium IV 2.0 while we are at it? I am not a fan of intel because I run vmware and hate that intel cripples its chips and the bios to exclude virtualization on all but the most expensive units. I hate the cost of a high end icore 7 which in 2010 was only 10 - 15% faster than a Phenom II but cost 400% more where I can buy a whole system for the cost of a single intel core 7 extreme.

    Well gentlemen. Expect dark days ahead and a return to $1000 desktops, $500 chips, and virtualization only available on xeon chips by next year. :-(

    With AMD junk status it is bound to happen now since these chips can't match intels offering.

    1. Re:A space heater included by evilviper · · Score: 1

      If you care, AT ALL, about power usage, you've just got to stay a bit back from the high-end. My quad-core 2.6GHz AMD CPU has a TDP of 45W. It costs a few dollars more for the low-power series of chips, but not much (unlike Intel's heavy ULV tax) and performs quite well. And this was over a year ago, better stuff with similarly low power is out there.

      I think there's something about the makeup of many AMD fans, where they want the high-end, and low-power, and aren't willing to pay a bit more for it.

      --
      Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
    2. Re:A space heater included by ak3ldama · · Score: 1

      AMD "fan" here... When I built my current tower build in January of 2010 I asked myself what I really needed out of it, and what things I did, and what I wanted to pay. I purchased an Athlon II x2 240 for $60. I had a few different times where I wished I could do some multi-threaded programming/testing on a quad core instead but ultimately it wasn't necessary. My at work i7 860 was sufficient to test with to see what would happen with OMG amazing multi-threading potential. Maybe some day I will upgrade to a Phenom II x4 965, or go down the FX-6300 route, or something like that but I do not "need" to yet.

      --
      "but money is the God of Algiers & Mahomet their prophet." - Rich. O'Bryen June 8th 1786
    3. Re:A space heater included by Billly+Gates · · Score: 1

      Holy cow.

      Get a Phenom II man before they stop making them? Seriously as the FX is an AM3+ and not socket compatible. Windows 7 has its slow moments during a re-image iwth Windows Updates taking forever to recompile all the .NET fixes. A new processor will also help with HTML 5/5.1 as browsers are using threading and separate processes more and more.

      My hex-core phenom II is great for 30+ tabs and I can run a few games too. As webworkers and websockets become standard in more and more webpages a newer phenomII with 6 cores will be beneficial.

      I like the phenom II when it was new. I just hate the newer chips.

  19. things need to change for both intel and amd by Vince6791 · · Score: 1

    Get rid of the l2 and l3 cache and beef up the l1 cache or just improve upon it. L2 cache was created to fill in for the speed gab between the cpu and ram. Now ram is running about 2.6ghz(multiple data transfers 1 clock) with hypertransport or quickpath buses.

    1. Re:things need to change for both intel and amd by gagol · · Score: 1

      On-die cache is much more about latency than bandwidth.

      --
      Tomorrow is another day...
  20. Ya I having trouble seeing where it is good by Sycraft-fu · · Score: 1

    Low end would be fine but it would need to be low power there too. Ivy Bridge CPUs are great at sipping power, particularly the dual core variety. So if what I'm doing is just real light usage like web surfing and so on, I'm better off with that.

    For heavier usage, well the Intel CPUs are better particularly at floating point calculations which is what most heavy performance is these days, at least on desktops. All the programs I can think of that I have which hit the CPU real heavy, are doing FP stuff, not so much integer.

    I just don't see what this is better for, unfortunately.

  21. What did Intel do? by gr8_phk · · Score: 1

    AMD is getting spanked badly in per-core performance. AMD was actually quite competitive a while back. From the benchmarks it looks like Intel had a very substantial leap in per-core performance with one generation of their core architecture. What did they do that made such a huge gain? And it's not that they're ahead on fab process. The single core per-clock performance jumped. What's up with that? How'd they do it?

    1. Re:What did Intel do? by Anonymous Coward · · Score: 0

      1. Shrunk their circuits with new fabrication that AMD couldnt' afford. Electricity moves faster and transistors switch quicker when electrical signals move in one direction rather than spread through thicker conduits.
      2. Can get data before it even arrives using mathematical tricks! AMD needs a ton of i/o to fetch the data while the intel ones can guess it with 85-90% accuracy without it even arriving and is already on the next data fetch without the first one even there yet.

      I know 2 sounds strange but it means you can get 8 mb l3 cash of performance and extra non existent bandwidth with just a 2mb l3 core chip virtually. This also makes intel smoking fast for servers too which are more i/o bound than fps intense. Intel has the math patented so AMD can't do squat about it and same is true with hyper-threading.

      This is why cutting R&D is bad which is what AMD has been doing while Intel is beefing it up. If AMD had this innovation instead and had it patented the bulldozers would be creaming the icore series.

    2. Re:What did Intel do? by Anonymous Coward · · Score: 0

      I didn't think you could patent math.

    3. Re:What did Intel do? by Billly+Gates · · Score: 1

      I didn't think you could patent math.

      LOL! You can patent corners and even look and feel of websites. Algorithms are math! Yes you can patent them in hardware too and not just in software. Should we? FUCK NO. Click on an address book is math that is the sole proprietary of Apple in which they are trying to nail Google for doing just as an example.

      Hardware yes I agree with Intel if they spent the R&D and want to protect as long as it is non commonly used styles or software where copyright can provide enough protection.

    4. Re:What did Intel do? by gr8_phk · · Score: 1

      1. Shrunk their circuits with new fabrication that AMD couldnt' afford. Electricity moves faster and transistors switch quicker when electrical signals move in one direction rather than spread through thicker conduits.

      Nope. That just allows higher clock speeds. AMD is getting nead 4GHz too these day. I also specifically said "single core per-clock performance". This means architectural changes - significant ones.

      2. Can get data before it even arrives using mathematical tricks! AMD needs a ton of i/o to fetch the data while the intel ones can guess it with 85-90% accuracy without it even arriving and is already on the next data fetch without the first one even there yet.

      Citation please. I don't buy that hand waving mumbo jumbo for a second without a reference - could be your wording but you even say it sound strange ;-) So I'm still wondering.

      Gotta agree that AMD cut all the wrong things under Ruiz and it's cost them dearly.

  22. Re:LED is a joke by DocSavage64109 · · Score: 1

    I have an outdoor porch light that seems to eat florescent bulbs. I've only had the LED bulb in for a few months, but so far it's doing fine.

  23. Another Meh... by Anonymous Coward · · Score: 0

    Too bad Intel had to go an stick Insider in all their chips. That's the only reason I am going with AMD this time around.

  24. Re:LED is a joke by djm · · Score: 1

    Same here. For several years, incandescents and CFLs in my front porch light would die almost every time there was a big storm, several times a year. Maybe thermal stress. I replaced the fixture, but the problem continued. Finally bought a 60W-equivalent LED several months ago and it's still fine after several storms. It's the one place I've found an LED is definitely worth it.

  25. intel pci-e io still sucks in most of there chips by Joe_Dragon · · Score: 0

    intel pci-e io still sucks in most of there chips

  26. Re: by Anonymous Coward · · Score: 0

    You should quit repeating their memes. Do you not understand that by saying "incandescent ban" you further their agenda? Those words are specifically framed to automatically generate an emotional response in people. Just by saying those words, you've lost the game.

    I guarantee I can find a box of incandescent in any hardware/grocery store (hint: bottom shelf ).

  27. I have to laugh by sa1lnr · · Score: 2

    When you call an 8 core 4GHz CPU low-end.

    1. Re:I have to laugh by evilviper · · Score: 1

      How about 2 pentium-d CPUs with hyper threading? 8 cores (half of them virtual), almost 4GHz, and yet very much low-end.

      --
      Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
  28. AMD and Intel do TDP differently by Anonymous Coward · · Score: 0

    Intel do "TDP when 100% load" and AMD do "TDP at maximum", in the latter case, finding out how much by design the TDP would be given a psychotic test case.

    AMD has always been better relatively on multicore than Intel, but at the expense of each cores' performance.

  29. Not the best choice. Just little different. by Anonymous Coward · · Score: 0

    Incandescents that are used infrequently are no better than LEDs that are used infrequently. In the latter case you may not quite make your investment back quick enough to make it financially worth it, but that doesn't make incandescents better to use. Just no different.

    1. Re:Not the best choice. Just little different. by Anonymous Coward · · Score: 0

      It's obvious by the context of my posts that I'm equating "best" to "optimal TCO", because that's an objective metric rather than a subjective one. In this "brief, infrequent" use case and others, incandescents are best.

      Furthermore, I allege that one will *never* reach breakeven when comparing a $40 LED bulb in the closet to a series of 50 cent incandescents—even when energy costs are included (ie. TCO). Even neglecting time value of money, it's just ludicrous.

      So, yes, incandescents are objectively best in certain circumstances. If we include subjectives, then it's always ultimately a matter of personal preference.

      I support the right to choose. The greens want to take away your rights, and they are winning.

    2. Re:Not the best choice. Just little different. by h4rr4r · · Score: 1

      No LEDs still beat them, because my time has a value. Not having to change the bulbs is worth it.

    3. Re:Not the best choice. Just little different. by Anonymous Coward · · Score: 0

      No doubt you're just trolling at this point; however, you're simply doing it wrong if you have your brief, infrequent use incandescents burn out more than once every 5 to 10 years. Might want to have your electrical system or structure's soundness checked out if this is a legitimate issue for you.

      Also, the expected time investment of changing an incandescent deployed in a closet for 10 years is what? 2 minutes? 5 minutes? Are you being paid $200/hour or more? Besides, there's no guarantee your LED bulb won't burn out (especially if you're as unlucky with lighting as seems to be the case).

      Now, if you're simply valuing your time higher for what you perceive to be an especially onerous task (changing bulbs), then that's into the realm of the subjective and I have no argument against that. Of course, if subjective arguments are valid and sufficient, then it's doubly inappropriate for the government to restrict personal choice in this matter.

      Seriously, if you want to spend $40/closet for LED bulbs that should be your prerogative. I know I would come out ahead, cost-wise, using incandescents, but the government is going to restrict my choice in the matter.

    4. Re:Not the best choice. Just little different. by hairyfeet · · Score: 1

      Uhhh...you charging $45 an hour labor friend? it takes less than 4 minutes for me to change the closet bulb, and I do so about once every THREE YEARS and that is with the cheapos, if the ban hadn't gotten rid of the higher quality bulbs I doubt I'd have to change more than once a decade.

      The simple fact is, and like it or not you can't make the math work, is that the LEDs simply won't ever break even in places like closets and bathrooms where the bulbs are rarely on more than 5-10 minutes max. When you figure in the cost of power, the price of the bulbs, and the lifetime of the bulb, I'm sorry but that 50c bulb ends up being the winner over your $40 LED every. single. time. This isn't some subjective thing like "Which CPU works best for me" but tables to figure TCO have existed for ages, its really not hard to do, and your math? it just don't work.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    5. Re:Not the best choice. Just little different. by bingoUV · · Score: 1

      if the ban hadn't gotten rid of the higher quality bulbs

      Are halogen incandescents not available in the US? My experience with halogens has been that they cost twice as much as regular incandescents, but failure rate has been - well I don't know. None have failed on me for 6 years when I started deploying them, I have total 3.

      --
      Bingo Dictionary - Pragmatist, n. A myopic idealist.
    6. Re:Not the best choice. Just little different. by bingoUV · · Score: 1

      Have you tried halogen incandescents? In my experience, they last way longer than regular incandescents.

      --
      Bingo Dictionary - Pragmatist, n. A myopic idealist.
  30. Honestly! by DarthVain · · Score: 1

    As someone who is pretty into this field, I find it somewhat surprising that no company has taken advantage of user trends.Perhaps it is just too expensive.

    I mean 64 bit and multi-threaded stuff has been out for some time. I mean a pretty long time. However typically, a user is still a single core user. Games, applications rarely seem programed to take advantage of multiple cores. Mostly the multiple cores are used by background processes, or running multiple applications at the same time. Most however don't really do all that much of even that. If you really wanted performance gains, you would think you would look at this usage and design something that fit that model.

    Like one core that is single and very fast. It would be used for 80% of everything a typical user would throw at it. Then you have several cores of low power slower speed that can pick up any spillover caused by mutiple applications, or the few applications designed for multicore use. I mean in many cases, you are buying a 4 core processor, which is using one core 80% of the time while the other 3 sit idle anyway (and they are all the same speed). Anyway logically it doesn't make any sense, however I conceed that likely it is a manufacturing issue more than anything else (i.e. they could build it, but it would be too expensive). Still for a nimble company looking for a competitive advantage, it seems a smarter route to explore than simply throwing more shitty cores that aren't being use anyway at a problem.