Trident Royal Z 5333Mhz and Asus Maximus Glacial

[daz_ni]

Hello,

I've recently put together a new PC build, based on an Asus Maximus XIII Glacial, an intel i9-11900k processor and the GSKILL Trident Royal 533MHZ ram however I'm having problems. No matter what I do, I can not get the mother board to post bios on anything higher than 4000Mhz. It will certainly run at the default 2133 once I clear the CMOS.

I have updated the bios to the latest, tried the default XMP settings along with other tweaks such as system agent and memory controller voltage but this hasn't worked.

This memory is listed as being on a QVL for this motherboard so I'm not sure what I am doing wrong. Is there anything I can do to get this to work any faster than 4000MHZ?

This is an example of one pair of my RAM

Code Number	04320X8820D
Date of Manufacture	2021/05/27
Series	Royal
Product Model Number	F4-5333C22D-16GTRG
*Memory Type	DDR4-5333
*Tested Latency	22-32-32-52
*Tested Voltage	1.60v
Capacity	8Gx2
Features	Intel XMP 2.0 Ready

[GSKILL TECH]

Have you tried each module individually to see if one has different maximum results?

Do you have memory populated in slots A2 B2?

[daz_ni]

Hello,

Yes, I have all 4 slots populated as I bought two sets of this RAM

I did not try the memory individually, I will maybe try that next. I guess I just expected it to work as I bought this more expensive memory as I thought faster was best. I'm starting to wonder if this memory is compatible with the Maximus XIII Glacial as it's not listed on the Asus website.

[emissary42]

Even if your CPU can handle that extremely high frequency, the motherboard will only do so in the two primary memory slots (A2+B2). Because of the daisy chain layout, the secondary slots have a (considerably) lower frequency limit. So when using four modules you have to reduce the frequency, with the motherboard being the limiting factor.

From a quick search they probably don't have a set of F4-5333C22D-16GTRG available for qualification testing, since the part number does not turn up in any of their QVLs.

Also ASUS only lists up to DDR4-4000 with four modules for 11th gen, so your result is pretty much in line with that.

[daz_ni]

hi,

Thanks for your reply, all very useful and given me some things to look into.

The memory is shown as on QVL by G.Skill => https://www.gskill.com/qvl/165/299/1...22D-16GTRG-Qvl

Where are you seeing ASUS only listing DDR4-400 support? Its listed as a speed option on the manual.

I wasn't aware of the A2+B+ channel limitation so I will try running a pair first. However, if that is the case then I'll be quite upset as I wasn't aware I could only run a single pair at 5333Mhz. I went for the fastest RAM thinking I would have a higher performing machine.

Thanks

[emissary42]

Where are you seeing ASUS only listing DDR4-400 support? Its listed as a speed option on the manual.

The highest frequencies are only support with two modules. Just sort the ASUS QVL for 11th gen by 4x8GB and then by max frequency = 4000.

I wasn't aware of the A2+B+ channel limitation so I will try running a pair first. However, if that is the case then I'll be quite upset as I wasn't aware I could only run a single pair at 5333Mhz.

It is normal for all dual channel motherboards that utilize a daisy chain DIMM layout and actually the norm nowadays, both for AM4 and LGA1200. That is also one of the (maybe not so obvious) reasons, why the QVL is valid for single kit use. With that being said, you are definitely not the first one, that got baited by the big numbers.

Your options would be to either run both with a reduced frequency or run a single kit with XMP at the max memory ratio your CPU can handle (which can be lower than DDR4-5333). Also keep an eye on SA and IO voltages, if your CPU requires really crazy values (some do), it might be preferable/safer to settle with a lower frequency and tighter timings instead.

[daz_ni]

thanks for your suggestions. It certainly triggered some more reading and I now better understand the whole QVL piece. I tested the machine with single stick and then two sticks and the motherboard immediately picked them up using the 533Mhz XMP profile. I however noticed the system was a bit unstable on games so decided to run memtest. Needless to say that on an otherwise set of standard settings and 5333Mhz ram (which it is supposed to run) it gives a lot of errors. As soon as I reduce the memory speeds, the errors go away. Overall, a bit disappointing that a high-end motherboard with expensive memory results in an unstable system.

[emissary42]

Since the memory is factory tested to clear its specs on compatible motherboards, the culprit usually is the CPU (or in some cases the motherboard, if you are unlucky and got a bad sample).

When suspecting a memory issue, test both modules individually. If one works at XMP and only the other one has issues with it, a replacement might help. In that case I'd recommend contacting tech support.

[completebricks]

Hey Daz, I came here to post the exact same issue but with the 4800Mhz version of this memory.

Here are my personal findings:

When set in XMP to above 4000Mhz, I can get the computer to post. I can even get it to run, but only for a few minutes before it crashes again. The way I do this is by overvolting the thing. I've gone as high as 1.85v (stock being 1.6v) and it'll work up until the point it won't. Why? Well, because the voltage exceeds the set amount. If I set it to 1.6v it'll exceed this and run at 1.611v. And so on. I can get it to 1.85v and eventually it'll get up to 1.869v before crashing. My only solution was to drop the speed down to 4000Mhz and it ran without any issues. I did a memtest, no problems, I did a burn-in on the CPU, no issues. I have also swapped out the CPU for a known good CPU (went from a 10700K to a 11700K) and the exact same issue is present. I also swapped out the RAM (same model, different sticks). I even went so far as to change out the MoBo for one of the exact same model.

Because I have literally done all this I can only assume two things:

1. It's somehow the PSU as that's the only thing I didn't change.
2. It's somehow the model of motherboard that's allowing the RAM to go above the voltage limit.

I can think of no other reasons that the voltage on the RAM is eventually hitting its maximum and then exceeding it.

If you guys can think of something, please let me know. And if you don't believe me, I'd be happy to post all of my RMA's to Newegg as proof.