Re: What PC hardware diagnostic stress-testing freeware can you recommend?
Paul <[email protected]>
| Newsgroups | alt.comp.os.windows-10,alt.comp.microsoft.windows,alt.comp.hardware |
|---|---|
| Organization | A noiseless patient Spider |
| Message-ID | <[email protected]> |
micky wrote: > In alt.comp.os.windows-10, on Fri, 04 Sep 2020 17:51:26 -0400, Paul > <[email protected]> wrote: > >> John McGaw wrote: >>> On 9/4/2020 1:35 PM, Arlen Holder wrote: >>>> What PC hardware diagnostic stress-testing freeware can you recommend? >>>> ........ >>>> >>>> If it's not the RAM, it's something else hardware related (the HDD has >>>> been >>>> re-imaged multiple times in its entirety and all extraneous boards and >>>> peripherals have been disconnected). >>>> >>>> What other hardware diagnostic stress-testing freeware can you further >>>> recommend? >>>> >>> Why settle on a hardware problem without proof? In my experience a bad >>> driver or glitched OS is equally likely if not more so. Try booting from >>> a Linux CD/DVD/STICK and run it for at least a few days and see what >>> happens. >>> >> You can run Prime95 from Linux. That's my favorite. >> >> Normally you'd do that to avoid corrupting the >> registry on an (unstable) Windows install. You should >> not run Windows or do Windowsy things, until >> "proving it's a computer" first. >> >> The Linux ones are static. Which is what you want. >> Nice and portable. >> >> https://www.mersenne.org/download/ >> >> Make sure your CPU cooler works first, before >> running that. Or, the power will go off on overheat >> (THERMTRIP). > > "We are not responsible for lost prize money, fame, credit, etc. should > someone accidentally or maliciously test the number you > are working on and find it to be prime. " > > Paul, if they are saying they're not responsible, who is? Will you > chip in if I lose the prize I should have won? >> Paul > It's a contest, if you snooze, you lose. First person to evaluate a number and determine primality claims the prize. Someone also duplicates the effort, as it looks really bad if a prize is claimed for no reason. I ported a piece of code using GMP (bignum package), and it runs at a lethargic pace. Not with the speed of the FFT method used by the "Prime Guy". However, if you use Prime95, then the money you get is reduced, as it's a pool of sorts. If I used my home grown, slow as molasses program (does math with 40 million digit numbers), and I happen to finish first and notice a particular number is a Mersenne Prime, then I don't share the money with the "Prime Guy". Maybe I get the money and you don't. However, i can only evaluate a single number, once in a blue moon, so I'm not much of a threat. My computer can only do around twenty 40 million digit calculations per second. Paul