Hello, the most recommended is to try it on another motherboard.

You can browse your processor here:
https://www.amd-k6.com/wp-content/uploads/201 … e_Datasheet.pdf

There is a jtag to verify correct operation. That is quite complex to manufacture if you don't have a lot of knowledge.

You can also search for "universal chip analyzer"

I am not involved in this, but the contribution can be useful.
Greetings.

You can't test whether a CPU works without testing it in a mainboard. But you can reduce the risk of damaging a mainboard by testing the CPU for shorts between Vcc, VIO and GND. If these pins are not shorted, the CPU is very unlikely to damage mainboards.

Are there any reports of CPU damaging the motherboard when it is visually and physically in a good condition?

In case there is a short the PSU won't power on.

mkarcher wrote on 2021-07-22, 21:54:

You can't test whether a CPU works without testing it in a mainboard. But you can reduce the risk of damaging a mainboard by testing the CPU for shorts between Vcc, VIO and GND. If these pins are not shorted, the CPU is very unlikely to damage mainboards.

I second that, you can still check CPU for Vcc to Vdd shunts or shorts, though some CPUs might have a naturally low resistance between Vcc and Vdd, thus you'd need to have a similar CPU for comparison. I had confirmed and ruled out several ESD damaged CPUs this way.

Then we have the least known and highly fallible method... divination. If you hold a pendulum over it and it begins to spin counterclockwise it's good, clockwise it's bad... the amazing thing about this method, is that you don't have to have the given CPU in front of you, if the eBay listing has a picture of the actual CPU, you can check it before you buy it, and get exactly the same degree of accuracy as if you already had it! 😉

But if you really, really, really gotta know 100% you gotta stick it in a known good motherboard at some point.

CPUs are traditionally the most robust components in a build and rarely fail; yes there are exceptions to this rule but unless the CPU looks toasted or burnt in any way, it's unlikely to be faulty - you'll need to test it in a board.

How common are motherboard-killing defective CPUs anyway ?

I mean, I've heard of one of them in my entourage about 25 years ago and I've read about them over the years, but I never got the impression that this was anything but a very rare occurrence .

That being said, having a low value sacrificial tester board seems like a very good idea if one is going to be sourcing lots of CPUs in unknown/untested condition . Pre-testing for obvious shorts, as suggested by previous posters, seems like a good idea prior to in-board testing .

BitWrangler wrote on 2021-07-22, 23:52:

Then we have the least known and highly fallible method... divination. If you hold a pendulum over it and it begins to spin counterclockwise it's good, clockwise it's bad... the amazing thing about this method, is that you don't have to have the given CPU in front of you, if the eBay listing has a picture of the actual CPU, you can check it before you buy it, and get exactly the same degree of accuracy as if you already had it! 😉

Just be aware that the directions are reversed in the southern hemisphere.

darry wrote on 2021-07-23, 00:48:

How common are motherboard-killing defective CPUs anyway ?

I mean, I've heard of one of them in my entourage about 25 years ago and I've read about them over the years, but I never got the impression that this was anything but a very rare occurrence .

That being said, having a low value sacrificial tester board seems like a very good idea if one is going to be sourcing lots of CPUs in unknown/untested condition . Pre-testing for obvious shorts, as suggested by previous posters, seems like a good idea prior to in-board testing .

Agree ! Odds of a bad cpu killing a board are so rare you might win a small lottery before you run into it 😀 More likely is a bad board will kill a good cpu than other way round 😀

Dude on OC back in the day said he'd got a killer Duron, killed 3 boards, 2 because he didn't believe it killed the first one, then put it away for a few months and forgot he had a wild killer chip (Quick, get Tom Selleck and the Runaway squad) and killed a 3rd one with it. Think there was a pic of him taking a hammer to it, not soon enough obviously 🤣

They may be rare, but they are potential serial killers .

First and foremost i'd like to thank everyone for their great tips and advices. I'm actually a bit overwhelmed. I didn't expect so many great people trying to help me.
That shows what a great community you guys are. I'm genuinely thankful for your help. I beg your pardon, as I won't be able to reply on most of your messages, but i've read them all.
Since quite a lot of you are pointing out, that the odds are very low for this CPU to be broken, I suppose there is no other way, except to put this CPU in a known good board.
In the moment I have 2 working boards. My PCChips M571 rev 3.2a and a MSI M5169 rev 2.1.
The M571 is likely going to be my candidate, since this board is well docummented. (I have flashed it to the latest bios)
Inspite not being a super socket 7 board, i've read some articles about people getting the k6-iii to run on this board.
Here is one of the articles: http://m571.com/m571/m571_amd.htm
On the website this CPU is listed amongst the processors that might be suitable for my system, but i'm unsure if this applys to rev 3.2a of this board.
Since this is a mobile chip, it has a vcore of 2.1v. However, the lowest voltage I can set on the M571 is 2.2v. Is this a big deal?
As long, as this board supports dual voltage and a multiplier of 4.5x, I should be good to go, right? Or am I mistaken?
Furthermore I was wondering, if this CPU runs well on lower FSB speeds, like 66mhz for instance.
What are your thoughts on using the M571 as my testbed? Am I out of luck with this board?

Sure. You can use any board that supports low enough Vcore (2.2V is fine). FSB doesn't matter for testing purposes. I'd use some standard FSB like 66Mhz and 3X multi. Than you can go from there and up the multi if you wanted to.

teclillass wrote on 2021-07-22, 21:08:

Hello, the most recommended is to try it on another motherboard. […]
Show full quote

Hello, the most recommended is to try it on another motherboard.

You can browse your processor here:
https://www.amd-k6.com/wp-content/uploads/201 … e_Datasheet.pdf

There is a jtag to verify correct operation. That is quite complex to manufacture if you don't have a lot of knowledge.

You can also search for "universal chip analyzer"

I am not involved in this, but the contribution can be useful.
Greetings.

The UCA currently only test up to 5x86. K6 will possibly never be supported.

Awesome, I will test the CPU in my M571 board tomorrow, if nothing gets in my way.
When i'm done testing it, i'll make sure to tell you guys, whether my project succeed.

I've changed my mind and did the test today.
I'm happy to anounce, that the CPU is indeed just working fantastic.
That is a big relief.
Once more i'd like to thank anyone who participated in helping me.
Just like that, all my questions were answered in no time and the problem is solved, that's just amazing.
You guys rock!

That's great... but I knew it will work 😉