[2001zhaozhao]

Yeah the only way to run 4 sticks of DDR5 decently is with Intel. It's a bit of a shame that you can't cram enough RAM to run big models.

The most I could get running on 10GB VRAM + 96GB RAM was a REAP'd + quantized version of MiniMax-M2.5

[mort96]

Got it running with 4800MT/s and literally 30 minute boot times in an AM5 machine. The 30 minute boot time could be worked around by enabling the (off-by-default) memory context restore option in BIOS, but it really made me think something was broken and it wasn't until I found other people talking about 30 minute boot times that I stopped debugging and just let it sit for an eternity.

It's so bad. I don't get why they sell AM5 motherboards with 4 RAM slots.

At least that system has been running well for like two years. But had I known that the situation is so much more dire than with DDR4, I would've just gotten the same amount of RAM in two sticks rather than four.

[WD-42]

I’m in the same situation! My machine will take 2-5 minute to post every few reboots, it seems random. The messed up part is the marketing material says this things can handle 256gb of ram or whatever absurd number, f me for thinking then 128gb should be no problem. Honestly this whole thing has soured me on AMD. Yea they have bigger numbers than intel but at what cost, stability?

[noir_lord]

Check you have MCR (Memory Context Restore) enabled, otherwise you train the RAM way more often than you need to (every boot).

[secondcoming]

Your machine takes 30 minutes to boot because of the RAM? Or it takes 30 minutes to load a model?

[WD-42]

It's the RAM. It needs to "trained" which takes some time but for for some reason these boards seem to randomly forget their training, requiring it to happen again.

[magicalhippo]

I've never had memory training be forgotten with my AM4 nor LPDDR5-based laptops and NUCs. Is this a new thing with AM5 or something? Or just a certain brand of BIOSes?

[sspiff]

It's a common issue on consumer boards with DDR5 and more than two DIMMs installed.

Doesn’t affect soldered memory or lower speed memory (like DDR4). Many memory controllers fail to achieve good speeds and timings at all on 4 DDR5 DIMMs, and fall back to running DDR5 at 3600MHz instead.

[magicalhippo]

Ok, so user selects too-high speed, controller tries for ages and fails, but doesn't save since it's overridden by user in BIOS?

I distinctly recall thinking my LPDDR5 NUCs were broken since they seemingly didn't boot the first time, until I recalled the training stuff. Took up to 15 minute on one of them. But neither has had any issues since, hence my question.

[justinclift]

Wonder if DDR5 ECC ram has the same problem? I'm meaning the real ECC stuff, not the "on chip only ECC" that all DDR5 has.

[jazzyjackson]

huh, its been a decade since i built a PC, whats changed?

[mort96]

DDR5 is much, much more fickle than DDR4 and earlier standards. I think it's primarily due to pushing clock speeds (6000 MT/s would be insanely fast for DDR4, but kinda slow for DDR5).

Memory training has always been a thing: during boot, your PC runs tests to work out what slight changes between signals and stuff it needs to adapt to the specific requirements of your particular hardware. With DDR4 and earlier, that was really fast because the timings were so relatively loose. With DDR5, it can be really slow because the timings are so tight.

That's my best understanding of it at least.

[WD-42]

My guess is bigger numbers, higher voltages, tighter timings.

[roboror]

It's an AMD thing

[noir_lord]

You need to enable MCR (which trains the memory once and caches the result for (iirc) 30 days) otherwise yeah, booting is horribly slow, even the 64GB I have can take several minutes but with MCR it boots basically instantly.

Some motherboards have it off by default.

[kenhwang]

Memory training seems to be getting faster with each bios update. In 2024 when I upgraded to AM5, 64GB memory training took like 15 minutes. Now the same setup takes about a minute when it needs to retrain, then near instant with MCR (Windows 11 takes significantly longer to load than the POST process).

[mort96]

From my comment:

> The 30 minute boot time could be worked around by enabling the (off-by-default) memory context restore option in BIOS

[WD-42]

I’m running 128gb on a 9550x now with 4x32gb sticks and it’s terrible. It’s unstsable, post time is about 2 minutes (not exaggerating)and I’m stuck at a lower speed. I’m considering just taking 2 of the sticks out and working with 64gb and increasing my swap partition. The nvme drive is fast at least.

This is my first time off intel and I have to say I don’t understand the hype.

[magicalhippo]

> It’s unstsable, post time is about 2 minutes (not exaggerating)

The long POST times must mean it's retraining the memory each time, which is not normal. Just in case you haven'ttried it yet, I'd start by reseating them, I've had weird issues with marginally seated RAM before.

Also you definitely have to go much slower with 4 sticks compared to two, so lower speed as much as you can. If that doesn't help, I'd verify them in pairs.

If they work in pairs but not in quad at the slowest speed, something is surely wrong.

Once you get them working in quad, you can start bumping up the speed, might need voltage boost as well.

[hxorr]

What ddr5 speed are you running? 6000 is technically an over clock, AMD only guarantees being able to run at something like 4800 or 5200.

You may need to bump up voltages slightly for your CPU's IMC (I needed to on my ryzen 8700F to run 6000 stable). Its CPU sample dependant.

Also as other commenter pointed out, typically 4 sticks will achieve lower stable clocks

[WD-42]

I just yanked two of the sticks out. Who knows, maybe I'll sell them. 64gb is sufficient most of the time anyway, and now I'm running at 4800 instead of 3600 and the boot is much faster. Thanks AMD!

[jodleif]

Threadripper is a good alternative. No point having a lot of dual channel ram for LLMs, too slow

[HauntingPin]

I had the same issue with Intel. It's not guaranteed there either.