Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Tuesday October 19 2021, @04:06AM   Printer-friendly
from the 2021's-Pentium-Bug dept.

Windows 11 hardware requirements made a mockery of by an Intel Pentium 4 processor

As the screenshots below show, Microsoft considers the Intel Pentium 4 661 a supported processor. Intel released the Pentium 4 661 in early 2006, with a solitary core to its name. Apparently, Microsoft forgot to add any Intel Family 15 (Netburst) SKUs in its unsupported processors list for Windows 11.

Hence, the PC Health Check tool sees that the Pentium 4 661 has a 3.6 GHz boost clock, which satisfies one of Windows 11's requirements. Curiously, the tool states that the Pentium 4 661 has two or more cores, even though it lists it as having one.

@Carlos_SM1995 has even got Windows 11 (Build 22000.258) running on a Pentium 4 661. Supposedly, Windows Update still works too, highlighting the ridiculousness of Microsoft's overtures regarding Windows 11 compatibility.

Windows 11 final (Build 22000.258) running on Intel Pentium 4 (11m4s video)


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2) by iWantToKeepAnon on Tuesday October 19 2021, @02:27PM (8 children)

    by iWantToKeepAnon (686) Subscriber Badge on Tuesday October 19 2021, @02:27PM (#1188415) Homepage Journal

    Not all CPUs are created the same. Lot's of progress in the last 14 years. Why should M$ restrict their instruction set and not take advantage of those advancements? Especially if one of them would provide a speedup or otherwise enhanced experience? Checkout the specs below of a CPU on one of our work linux servers (not recommending this CPU in any way, just one that is handy at the moment). Do you think a CPU from 2006 has the same technical spec? Wouldn't you be upset if linux didn't take advantage of every bit of juice your CPU could provide?

    • vendor_id : GenuineIntel
    • model name : Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
    • microcode : 0x200005e
    • fpu : yes
    • fpu_exception : yes
    • cpuid level : 22
    • wp : yes
    • flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon nopl xtopology tsc_reliable nonstop_tsc eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch arat fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 invpcid rtm rdseed adx smap xsaveopt
    • ...
    --
    "Happy families are all alike; every unhappy family is unhappy in its own way." -- Anna Karenina by Leo Tolstoy
    Starting Score:    1  point
    Karma-Bonus Modifier   +1  

    Total Score:   2  
  • (Score: 2) by ChrisMaple on Tuesday October 19 2021, @04:19PM (1 child)

    by ChrisMaple (6964) on Tuesday October 19 2021, @04:19PM (#1188442)

    Software can be written to test for whether an instruction is implemented, and only then use routines that require the instruction.

    • (Score: 5, Informative) by DannyB on Tuesday October 19 2021, @04:44PM

      by DannyB (5839) Subscriber Badge on Tuesday October 19 2021, @04:44PM (#1188451) Journal

      Consider the configuration nightmare. How many different duplicate sets of routines do you want Windows to have? Consider just the Kernel for a second. How many different permutations of Kernel should Microsoft compile for? There are different dimensions of features that a processor may or may not have.

      Now that said, let me digress a moment and point out where Java wins hands down on processor compatibility. See this journal article: It's fashionable to hate Java [soylentnews.org], where I explain about the C1 / C2 native compilers.

      Summary:

      Your JVM bytecode starts up being interpreted. As soon as a function is identified as using disproportionate CPU, it gets immediately compiled quickly by the C1 native compiler into much faster but not optimal code. It is also added to a list of functions to be recompiled by the native C2 compiler soon. When the C2 compiler comes around, it will spend significant time and effort compiling your function to native code. C2 can take advantage of things known only at runtime, which cannot be known ahead of time, such as a C compiler. For example, maybe this one method call does not actually need to be a virtual call going through a vtable. There is only one instance of a subclass ever, at all in the entire program. And many other optimizations an ahead of time compiler cannot know. In particular, and of specific relevance here is that the C2 compiler can compile your function to the exact instruction set of the hardware it is currently running on. Does your processor have this or that other set of vector extensions? Or this or that other coprocessor, or some other feature? The code is compiled to the exact processor you are running on. Whether it is Intel, AMD, or even a different architecture, ARM (many flavors), Power PC, even IBM mainframes, etc.

      Thus, Java programs start up slow, then seem to "warm up" and run fast. This is great for a program that will run for a long time and wants excellent performance. It is horrible for quick command line programs such as 'ls'.

      --
      The anti vax hysteria didn't stop, it just died down.
  • (Score: 2) by Opportunist on Tuesday October 19 2021, @07:40PM (5 children)

    by Opportunist (5545) on Tuesday October 19 2021, @07:40PM (#1188538)

    Not at all, that would be logical. But obviously they didn't do that. As it has been demonstrated, the code does run on ancient hardware, and since Intel has been backwards compatible for as long as I can remember, to the point where newer CPUs had to include all the bugs the old CPUs came with and programmers programmed around, this would suggest that Windows should run on newer hardware, and it not running is only due to MS deliberately sabotaging it.

    • (Score: 2) by iWantToKeepAnon on Wednesday October 20 2021, @04:28PM (4 children)

      by iWantToKeepAnon (686) Subscriber Badge on Wednesday October 20 2021, @04:28PM (#1188807) Homepage Journal

      the code does run on ancient hardware

      I am not a M$ cheerleader, but ... showing screenshots or even a video doesn't prove that it runs for all software in all environments. It could easily throw unsupported instruction errors running on hardware at a level that is below the build process compile flags. CPUs may be backward compatible but they are not *forward* compatible. Apple has been breaking backwards compatibility for, well, since forever. I think M$ is well within their rights to require a modern CPU to run on, and it may even be overdue.

      How many times must this happen [https://www.joelonsoftware.com/2000/05/24/strategy-letter-ii-chicken-and-egg-problems/] :

      ... Jon Ross, who wrote the original version of SimCity for Windows 3.x, told me that he accidentally left a bug in SimCity where he read memory that he had just freed. Yep. It worked fine on Windows 3.x, because the memory never went anywhere. Here’s the amazing part: On beta versions of Windows 95, SimCity wasn’t working in testing. Microsoft tracked down the bug and added specific code to Windows 95 that looks for SimCity. If it finds SimCity running, it runs the memory allocator in a special mode that doesn’t free memory right away. That’s the kind of obsession with backward compatibility that made people willing to upgrade to Windows 95 ...

      Since I have to run Windows on work PCs, I'd be glad to get rid of all those old "special mode"s.

      --
      "Happy families are all alike; every unhappy family is unhappy in its own way." -- Anna Karenina by Leo Tolstoy
      • (Score: 2) by iWantToKeepAnon on Wednesday October 20 2021, @04:44PM

        by iWantToKeepAnon (686) Subscriber Badge on Wednesday October 20 2021, @04:44PM (#1188810) Homepage Journal

        Also of note to the backwards compatibility issue : https://www.joelonsoftware.com/2004/06/13/how-microsoft-lost-the-api-war/ [joelonsoftware.com]

        --
        "Happy families are all alike; every unhappy family is unhappy in its own way." -- Anna Karenina by Leo Tolstoy
      • (Score: 2) by Opportunist on Wednesday October 20 2021, @09:31PM (2 children)

        by Opportunist (5545) on Wednesday October 20 2021, @09:31PM (#1188960)

        Again: Nobody complains about them not "officially" supporting ancient hardware. You run it and it runs, fine, you run it and it barfs, too bad. No support.

        What they did, though, is deliberately sabotaging the ability of the software to run on certain CPUs. It doesn't just simply crash because of an illegal instruction or unsupported function, it checks for the type of CPU you have and if it detects one that it does not want to run on, it refuses to.

        Do you understand the difference?

        • (Score: 2) by iWantToKeepAnon on Sunday October 24 2021, @02:04AM (1 child)

          by iWantToKeepAnon (686) Subscriber Badge on Sunday October 24 2021, @02:04AM (#1189997) Homepage Journal

          > Do you understand the difference?

          Yes of course, let's not get condescending. But M$ doesn't want hordes of tweets and ig's about how horrible the new OS is and how it crashes all the time; public opinion is still important. Nip it in the bud and just don't allow it.

          But I agree, there should be a hack or a huge RED don't press this button to run on old hardware. There probably will be a hack, in time.

          --
          "Happy families are all alike; every unhappy family is unhappy in its own way." -- Anna Karenina by Leo Tolstoy
          • (Score: 2) by Opportunist on Monday October 25 2021, @04:15AM

            by Opportunist (5545) on Monday October 25 2021, @04:15AM (#1190264)

            If information control and PR was the name of the game, that's easy to do. Make sure everyone and their dog knows the minimum specs and use your army of fanboys to mercilessly mock anyone who complains about performance on unsupported hardware.

            Sorry, not buying it.