Memory speed is incorrect when running Memtest86

[PWSR]

Yeah, as I've tried at least 20 different make/model combinations, it's clear that each laptop acts a little differently.

On that note, can you provide a shortlist of laptop make/models/CPU-config that you use in-house to make sure the software is working correctly? Specifically looking for DDR5-capable laptops. If I had been using the same model you use, none of this would have come up. Ideally, they are high-spec, get-the-test-done-fast systems.

Alternate idea: A list of laptop make/models that are "certified", "approved", or "recommended". Something like, "Passmark has verified that these work as intended."

p.s. Apologies to original poster Anthony.Lee for taking over the thread. David, unless you responded separately, they are still waiting for an answer to their post.

[David (PassMark)]

Displaying the memory clock speed and timings in use when Memtest86 launches is not a bug, this is what we intended. So to say this isn't working correctly isn't wrong. It is working as designed.
(there are a few rare CPUs for which we still don't have support for this, but these aren't the machines we are discussing here, and in those cases we display the fixed SPD data values).

I understand you would prefer we display a value that isn't the one actually being used, if the max potential is higher. But maybe for this you can check the RAM's SPD data.
As noted above, the clock speed isn't updated during a test run at the moment. So maybe we can make the value dynamic in the future (or take the max value over time).

And none of this display information impacts the core function of MemTest86, which is to test the RAM works without error.

[PWSR]

Got it. It's the laptops that are causing the values that don't match what I'm expecting. I'm surprised because mine doesn't seem like a fringe use-case: bone-stock Samsung and SKHynix modules, no overclocking, super common enterprise Dells and HPs. But anyways, I'm okay with it as long as I can find a laptop/CPU configuration that works as intended.

To clarify, here's an updated "what I want" - Module says D5 4800 on the label; laptop/CPU can handle D5 4800; BIOS reports D5 4800; MemTest shows that it's testing close to D5 4800; MemTest performs tests up to D5 4800 at least some of the time. Is this a reasonable expectation?

My experience has been, 90% of the laptops I try do not do what I want. After participating in this thread, I know of no way to prevent it either: "laptop is gonna do what laptop is gonna do." Those are some bad odds if I go out and buy a laptop for testing...

So, can you help me better those odds? When you test successfully and everything goes right, what laptop/CPU config are you using? CHECKMEMSPEED must have been validated when it was programmed. What laptop/CPU config was used for that? Is there some kind of lab or stress-test area? Which laptop/CPU config is used in there? If you had confidence you could go produce an expected good result right now, which laptop/CPU config would you use?

[David (PassMark)]

When you test successfully and everything goes right

Software was designed to display the current clock speed. If the motherboard decided to run at a lower speed, then that is what gets displayed. This was the design. This is how it works. To say it isn't "right" is wrong. It is working as designed.

We don't have a list of laptops to that do or don't vary RAM speeds. A few years ago, it was a very rare thing. But I would expect nearly all new laptops to do it over the next few years.
We use mostly desktop machines for development work and so far none of them vary their speed (or don't during the boot process anyway). The laptops we have have mostly soldered RAM so there is never any possibility of testing different sticks of RAM. Soldered RAM and RAM built into the CPU package will also likely become more common in the future.

[PWSR]

Understood. I will adjust my expectations and move forward.

I can't say this is a satisfactory resolution but I sincerely thank you for all the time you took responding to my concerns.