1. You are viewing our forum as a guest. For full access please Register. WindowsBBS.com is completely free, paid for by advertisers and donations.

Wrong CPU Installing

Discussion in 'PC Hardware' started by neutropia, 2014/07/09.

  1. 2014/07/09
    neutropia

    neutropia Well-Known Member Thread Starter

    Joined:
    2002/08/26
    Messages:
    52
    Likes Received:
    0
    At some point, windows 7 32-bit upgraded my old AMD Athlon II X3 450 Rana Triple-Core 3.2Ghz processor to an AMD Phenom II X4 B50 Quad-Core processor and refuses to allow me to roll it back to the correct processor. Even BioS recognizes it as a quad core and notifies me all 4 cores are active despite it only being a three core.

    I've tried the basic things I can think of: Turning off auto-driver install and manually removing the CPU from the Device Manager. Windows STILL auto installs this processor and there is no options when trying to manually update/change the driver. AMD's website is no help as I can find virtually no driver's for my old processor, only video drivers. BioS does have options to disable cores, but this is beyond my scope of knowledge and I don't want to inadvertently disable an active core and leave a non-existent core active.

    I know this is an old processor I'm running, so even if I can't roll it back, I'd like windows to at least see the proper clock speed and number of cores as I'm not sure if it's causing performance issues in general.

    Does anyone have any ideas on how to either install the correct processor or at least tweak the currently "recognized" one to be utilized like the Athlon should be?

    Sorry if there's any lack of information that may be needed. I'm not well-versed on multi-core CPUs and legacy devices being recognized as something completely different.
     
  2. 2014/07/09
    Steve R Jones

    Steve R Jones SuperGeek Staff

    Joined:
    2001/12/30
    Messages:
    12,279
    Likes Received:
    246
    Maybe there is a new chipset drivers from the maker of the mobo?
     

  3. to hide this advert.

  4. 2014/07/09
    retiredlearner

    retiredlearner SuperGeek WindowsBBS Team Member

    Joined:
    2004/06/25
    Messages:
    7,150
    Likes Received:
    499
  5. 2014/07/09
    neutropia

    neutropia Well-Known Member Thread Starter

    Joined:
    2002/08/26
    Messages:
    52
    Likes Received:
    0
    Nice find retiredlearner. I went ahead and updated by BiOS to the latest version and it still see's my CPU as this new quad core. I had been manually disable the 4th core, but I went ahead and let my ASUS MoBo auto detect cores and it still feels there are 4. I ran a CPU speed tester and it clocks me at 3.2GHz & I'm experiencing no stability issues, so I guess W7 has inadvertently unlocked the 4th core on my CPU!
     
  6. 2014/07/09
    retiredlearner

    retiredlearner SuperGeek WindowsBBS Team Member

    Joined:
    2004/06/25
    Messages:
    7,150
    Likes Received:
    499
    It's the usual story - "If it ain't broke-----" I assumed that your comp was still working OK. You've got more than you knew you had. Neil.
     
  7. 2014/07/10
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    That tells me this is NOT a Windows thing. Remember, Windows is not in the picture until WELL AFTER all the BIOS information has been read in and the computer completes POST, then the boot process seeks out the boot drive with the OS. Windows is only taking the information provided to it, and trying to run with that.

    BTW, AMD is pulling a trick out of the old floppy disk makers hats. For you old flatulent-types ;), remember 5.25 floppies? Originally, they were all SSSD - single sided, single density. Then makers learned to jam smaller, thus more magnetic particles in the same space, and they were called, SSDD - single sided, double densities. Then they got real clever and learned to jam magnetic particles on both sides, for DSDD - double sided, double density.

    But the problem was, the manufacturing process was not perfect. In fact, it was not that good at all and it was hard to make a flawless DSDD disk - but EVERY disk they made, they produced as DSDD. Then it was tested. And if one side failed, the disk was not rejected, instead it was "downgraded" and labeled as SSDD and sold for less. If it failed the density test, then it was downgraded to SSSD and sold for less yet.

    Same with these quad AMDs. It is not easy to jam 100s of millions (well over a billion in some cases) of transistor gates in a single CPU and not have some fail. So AMD makes these quads, and if one core does not meet specs, they shut it down, and call a triple core, and reduce the price.

    This may sound like a gimmick, but it actually makes good business sense for both AMD and for consumers. AMD does not take a total loss on the device, and consumers still get an excellent and reliable CPU for a good price. It's a win-win.
     
    Bill,
    #6
  8. 2014/08/17
    hawk22

    hawk22 Geek Member

    Joined:
    2002/01/31
    Messages:
    1,991
    Likes Received:
    26
    Funny enough this is all legit, I did buy one of these 3 Core CPU's with a ASUS Motherboard that has a button on the Motherboard to enable that 4th core on that AMD CPU. That particular CPU is a deliberate act by AMD to sell it cheaper than the quad core, but it actually is the same CPU with 1 core disabled, Gigabyte and ASUS came to the rescue, to allow users to enable that core with just a push of a button.
    It is still running on 1 of my comps, I hardly ever turn that machine off, at this very moment I returned from holiday on July 10 and it has not been turned off since.
    hawk22
     
  9. 2014/08/17
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    "A deliberate act" makes it sound sinister. It is nothing of the sort. As I said above,
    It makes no sense to "deliberately" disable a perfectly functional core and sell the CPU for less profit.

    While it is true that some motherboards allow these cores to be unlocked, I personally would not expect the CPU to remain as stable if unlocked. And when doing important work on my PC, an unstable CPU is not acceptable.
     
    Bill,
    #8
  10. 2014/08/18
    hawk22

    hawk22 Geek Member

    Joined:
    2002/01/31
    Messages:
    1,991
    Likes Received:
    26
    I partly agree with you Bill, but not completeley, as I had posted, this CPU is in one of my machines that runs for weeks around the clock and has given me no trouble whats so ever, could'nt be any more stable, (maybe I was lucky).

    http://www.tomshardware.com/news/amd-phenom-cpu,7080.html

    I was made aware of the CPU by the Overclockers Forum, I have been a AMD fan since I changed from my CYRIX 333 to AMD, only ever had one Intel and that was the Pentium100 before the CYRIX.
    Give you an Idea how old I am.:eek::eek:
     
  11. 2014/08/18
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    You are still a kid! ;)
    Not sure what you "partly disagree" about. Are you suggesting they are being sinister about it? If so, then where's the logic?

    If AMD condemns the entire CPU because 1 core fails their rigorous testing, it is a total loss and they lose money big time.

    If they sell the quad as a quad knowing one core does not meet specs, and it fails during use, they inconvenience and pi$$-off their customer (possibly losing them to Intel for future CPU buys) and lose money big time to replace the failed CPU. Plus, if that upset user is vocal about his AMD CPU failing, AMD's image becomes tarnished.

    If they disable the core that does not meet specs and sell the unit as a triple-core at a reduced price, they still turn a little profit, or at least break-even and not take a total loss, and they keep a happy customer who will hopefully buy AMD again, and maybe recommend AMD to his friends.

    If you are suggesting because you enabled the 4th core and had no stability problem that all disabled cores can be enabled and have no stability issues, then, yeah, we disagree completely on that point. If that were true, AMD would not disable them, and take the full profit selling them as quads.

    I don't think you were lucky, just that your computing tasks did not "touch" the problem area that caused AMD to lock the core in the first place.

    Are AMD's testing procedures TOO rigorous? Maybe. But that is MUCH BETTER than not rigorous enough.
     
  12. 2014/08/20
    hawk22

    hawk22 Geek Member

    Joined:
    2002/01/31
    Messages:
    1,991
    Likes Received:
    26
    I know from what I had read at the time (a few years ago now) that some of the disabled cores when enabled can possibly show to be unstable under extreme conditions in certain computing areas, if memory serves me right you would have had to use AutoCad.
    In saying that, the average user would never have had a unstable CPU in my opinion that is.
    In my experience and opinion regular AutoCad users would not use AMD CPU's anyway, they are predominantly Intel users, mostly professionals.
    Bill you probably won't agree with me on that score, there won't be any hard feelings we just have to agree to disagree on that score.
    I would like though to tell you, that I highly respect your very professional assessments and opinions.
    hawk22
     
  13. 2014/08/20
    lj50 Lifetime Subscription

    lj50 SuperGeek WindowsBBS Team Member

    Joined:
    2003/07/04
    Messages:
    2,801
    Likes Received:
    137
    Orb\search box (type in)msconfig.exe\Boot\Advanced Options. Put a check mark in Number of processors the click the down arrow.
     
  14. 2014/08/20
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    It is absolutely true that re-enabled cores can be unstable. That is why they were disabled!!! And should be left disabled, if stability (and data integrity) is of importance to the user. And if stability and data integrity are not, then why not? :confused:

    But it is flawed logic to assume that every one of those disabled cores were disabled because they failed factory testing with the same fault, or that multiple dissimilar faults affected only the same small group of users.

    Considering the AMD Trinity line of CPUs, for example, has more than 1.3 billion (with a "b! ") transistors on a die the size of a postage stamp, it is statistically improbable - if not impossible - for the same group of "gates" to be defective, to have failed in the same manner such that all these failed cores affect ONLY AutoCAD users. That just makes no sense.

    If the defects, thus failures were that specific, they would have corrected them for subsequent production runs.

    Also, AMD offers several different triple-core processors with different transistor counts and architectures. It is illogical to assume, once again, that these different models would develop the same faults that affect only the same group of people.

    And while I do agree that Intels are the processors of choice for most professionals doing CAE/CAD work, it certainly is not exclusive. And again, it is flawed to assume only AutoCAD users are affected - especially since modern 3D animated gaming can tax systems to the same degree, if not more.

    CPU maker makers do cater to the "average" user, but they cannot afford to ignore the advanced users as you seem to be implying - because there is a lot of them too - and they tend to spend the big bucks, if budgets allow. And they also may be the family or neighborhood goto person for buying advice.

    More bad logic. "A few years" is ancient history. This is like assuming what you had to do with XP is still necessary with W7 or even W8. Or that SSDs still suffer from write limitations. Or that IE11 must not be secure because IE6 was not secure. Very bad logic all. It assumes software developers and hardware makers (including CPU code developers) are sitting on their thumbs instead of constantly looking at ways to improve their products.

    I don't let my opinions get in the way, unless personal insults are tossed directly at me. And there is no place for that in technical discussions IMO. So no worries about my feelings. But I have to say, IMO "agree to disagree" is a cop-out as it is often used when someone cannot, or is unwilling to justify rationally, or collaborate their opinions with "true" facts. :(

    I am not saying you are being stubborn and unwilling, I just would like you to consider that it is not a fact these locked cores only affect CAD users. And it is not a fact these locked cores (on several models of CPUs) were all locked due to the same gate or group of gates failing tests.

    It still seems as though you feel AMD is being deceitful. I totally disagree here but I can only use business "logic" (no evidential proof) to show how this is a win-win for both AMD and consumers. And the fact AMD is being upfront about these processors shows IMO, they are not hiding or being deceitful.

    I am not in favor of that. For one, msconfig should not be used for permanent changes. It is a troubleshooting tool. For another, that switch sets the number of processors that Windows will run at startup. With this switch, you can force a multiprocessor system to use only the quantity of processors (number) that you specify. This switch can help you troubleshoot performance problems and defective CPUs.

    With all unchecked, you are telling Windows it can use all. With 4 checked, you are forcing Windows to attempt to use all 4 and that can cause problems with disabled cores.

    At any rate, it only affects boot times, if that - not overall performance once Windows is fully booted.

    IMO, if someone wants a quad, they should buy a quad in the first place - even if it means delaying purchase until budget allows. And if they get a triple-core anyway and re-enable the disabled 4th core, they better not complain about stability, or start bashing AMD.
     
  15. 2014/08/20
    lj50 Lifetime Subscription

    lj50 SuperGeek WindowsBBS Team Member

    Joined:
    2003/07/04
    Messages:
    2,801
    Likes Received:
    137
    Bill: I am not instructing the OP to change anything. Just to check out how many processors are listed.
     
  16. 2014/08/20
    hawk22

    hawk22 Geek Member

    Joined:
    2002/01/31
    Messages:
    1,991
    Likes Received:
    26
    I have 4 cores running on that CPU right from the beginning, I had bought at the time (several years ago 2008-2009) the CPU for that purpose, it was selling for a good prize then. It is still in everyday use, but the system that I have just built using a AMD FX8350 Black Edition will replace it, partly because I have W XP on it.

    When I say a few years ago I mean a few years ago, that is when I looked into that CPU Phenom X3, and it is then that I had bought it from information I had gathered then, a few years ago (2008-2009), and that has nothing to with bad logic as you say.
    hawk22
     
    Last edited: 2014/08/20
  17. 2014/08/21
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    Well, XP needs to go away so that is good enough reason alone.

    2008 was before Windows 7, which came out in Oct 2009. So hardware bought at that time was built for Vista and legacy XP support. Ancient technologies compared to today's operating systems, and in particular today's modern hardware - which has ALWAYS been years ahead of software!

    So "bad" logic may not be the right word. "Flawed" may be more correct, as I used previously.

    Still, to clarify, my "issue" here is whether AMD is being sinister or not. I COULD BE READING YOUR COMMENTS WRONG - but it "appears" to me you [still] feel there is, or was some deceit or sneakiness going on here.

    If I am wrong, then my apologies. But I just don't think AMD is being deceitful, intentionally or otherwise. I also think these are common manufacturing techniques used in the making of many products, not just CPUs.

    As I mentioned before, I know floppy disk makers manufactured all their disks to be high density and double sided. Those that failed in-house testing for "high" or double density, or one side failed were then branded and marketed as single density or single sided. I even had a punch for DDSS (double density, single sided) 5.25" disks to turn them into DDDS (double density, double sided). And in most cases, I had no problems.

    I would not be surprised if memory device makers and drive platter makers do something similar as well as manufacturers in entirely different industries.

    What is the logic for AMD to purposely lock out a perfectly good core, then sell the device at a lower profit margin?

    And while AMD does not market and advertise these triple-cores as "crippled quad-cores ", they have made no attempt to deny what they are doing either, and have in fact, acknowledged it.

    Ah! Gotcha. Thanks.
     
  18. 2014/08/21
    hawk22

    hawk22 Geek Member

    Joined:
    2002/01/31
    Messages:
    1,991
    Likes Received:
    26
    Call it what you like, the fact remains, when AMD released the Phenom X3 they did not advertise it as a Quad with one core disabled.
    It was the consumers that lifted the lid, not AMD who afterwards admitted it.
    hawk22
     
  19. 2014/08/21
    Bill

    Bill SuperGeek WindowsBBS Team Member

    Joined:
    2002/01/11
    Messages:
    3,332
    Likes Received:
    389
    Why would they? I think you are seeing evil where there is none. I guess we WILL have to agree to disagree with that!
    No, it wasn't. IT journalists reported it during pre-release reviews. And as seen here, they surmised it was to compete and keep nipping at the heals of Intel - a VERY GOOD thing for us consumers!

    And BTW, Intel did something similar when the Core 2 line of CPUs came out. The Core 2 Duo was a dual-core, and the Solo was the same dual-core, but with one core disabled. See Intel Core Solo.
     
  20. 2014/08/21
    lj50 Lifetime Subscription

    lj50 SuperGeek WindowsBBS Team Member

    Joined:
    2003/07/04
    Messages:
    2,801
    Likes Received:
    137
    Bill: You're welcome.
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.