Question Mixing two sets of DDR4 Ram, same manufacturer, but different capacities and different dies

[BuilderGuy]

Has anyone had success mixing two sets of DDR4 Ram, same manufacturer, but different capacities and different dies? I have 16GB (2 X 8) G.Skill Trident Z Royal Gold DDR4 CL(16-16-16-36) with a Samsung b die and 32GB (2 X 16) G.Skill Trident Z Royal Gold CL(16-19-19-39) with a hynix cjr die. Both have the same voltage. I'm planning to mount them on a ASUS Rog Strix x570-E motherboard, each set in two separate channels. I'm running a AMD Rysen 9 5900x CPU.

The WEB is all over the map with this. Some claiming that you can't even mix separate sets of identical RAM. Others claiming BS, with videos, demonstrating that this is generally not a problem.

Do manufacturers of RAM really have such poor QA/QC, that even identical sets come off incompatible? Are motherboards really so unforgiving? Is this a manufacturer's scam to make more profit?

 

[USAFRet]

The only real answer is a 100% maybe.
It might work, it might fail.

QA tolerances?
Lets assume there is a +/- 5% tolerance. 100 being the mean.
RAM stick A is at 96, stick B is at 103. They may have come from different fab plants...1 in Malaysia, 1 in Thailand.
Both completely within tolerance. Tested, absolutely working.

But they may not work together, at all.

 

[BuilderGuy]

5% would not be tolerable with most manufacturing, much less electronics, where trial runs can test millions of times for MTTF and tolerances are on the nano scale.

 

[USAFRet]

5% would not be tolerable with most manufacturing, much less electronics, where trial runs can test millions of times for MTTF and tolerances are on the nano scale.

I just used that as an example.
0.005% would be exactly the same.

If you want XXGB RAM, buy XXGB as a set. Don't count on 2 different sticks or sets to work. Especially with different timings, etc.

It might, it might not.

 

[geofelt]

Since you already have the two sets, go ahead and try it.
No damage will happen if it fails.

Ram is sold in kits for a reason.
A motherboard must manage all the ram using the same specs of voltage, cas and speed.
The internal workings are designed for the capacity of the kit.
Ram from the same vendor and part number can be made up of differing manufacturing components over time.
Some motherboards, can be very sensitive to this.
This is more difficult when more sticks are involved.
You can sometimes compensate for errors by increasing the ram voltage in the motherboard bios.

Since ryzen is tied closely to ram for proper performance, I have doubts that you will be successful.
Regardless test with memtest86
It boots from a usb stick and does not use windows.
You can download the free edition here:

Memtest86+ | The Open-Source Memory Testing Tool

Memtest86+ is an advanced, free, open-source, stand-alone memory tester for 32- and 64-bit computers (UEFI & BIOS supported)

www.memtest.org

If you can run a full pass with NO errors, your ram should be ok.

Running several more passes will sometimes uncover an issue, but it takes more time.
Probably not worth it unless you really suspect a ram issue.

 

[BuilderGuy]

I'll give it a try. Thanks. If the two aren't compatible I'll go with the 32 GB set. I do hope for the best, since both sets run at the same speed, voltage, and CAS latency, although secondary timings, i.e. TRCP, TRP, and TRAS differ. You'd think that motherboards could be designed to robustly handle these.

 

[DSzymborski]

5% would not be tolerable with most manufacturing, much less electronics, where trial runs can test millions of times for MTTF and tolerances are on the nano scale.

Not 5% of course -- that was an illustration -- but semiconductor fabrication does have a hard time making precisely identical parts because of the scale. There's a reason that budget and mid-range parts are largely components that failed "qualifying" to be more expensive SKUs and were binned as cheaper ones. Yields have been one of AMD's issues getting out more 5950x and 5900x chips.

 

[MeanMachine41]

Has anyone had success mixing two sets of DDR4 Ram, same manufacturer, but different capacities and different dies? I have 16GB (2 X 8) G.Skill Trident Z Royal Gold DDR4 CL(16-16-16-36) with a Samsung b die and 32GB (2 X 16) G.Skill Trident Z Royal Gold CL(16-19-19-39) with a hynix cjr die. Both have the same voltage. I'm planning to mount them on a ASUS Rog Strix x570-E motherboard, each set in two separate channels. I'm running a AMD Rysen 9 5900x CPU.

The WEB is all over the map with this. Some claiming that you can't even mix separate sets of identical RAM. Others claiming BS, with videos, demonstrating that this is generally not a problem.

Do manufacturers of RAM really have such poor QA/QC, that even identical sets come off incompatible? Are motherboards really so unforgiving? Is this a manufacturer's scam to make more profit?

Serious Overclockers and enthusiasts all say "DON'T MIX RAM" and manufacturers warn you of doing this.
Most RAM purchases are based on budget so some take the risk which is OK if you can return them when frustration sets in.
Most RAM related issues here at Tom's are due to mismatched DIMM's

RAM kits are binned at the factory to match density/Latency of each DIMM into one kit.
It's all about the Silicone quality just as with CPUs. Some DIMMs bin to a higher frequency and cost more others bin lower and cost less. Some have better ICs and SPDs with lower CL.

What you are trying to do is pair 2 kits from different manufacturers with differing Timings and differing Die sets. You will have issues.
Try to learn the basics of what Timings are all about cause the technology behind it is like rocket science.
https://www.gamersnexus.net/guides/3333-memory-timings-defined-cas-latency-trcd-trp-tras
And that's just the basics.

 

[BuilderGuy]

Are all you guys so arrogant? Your not getting my issue at all. Besides the fact is that these two kits are from the same manufacturer, have the same CAS latency, the same voltage, and the same speed. The secondary times differ. I am really raising a question about why RAM manufacturing has such poor QA/QC and why motherboard manufacturers fail to make more robust systems. Its about time that the so called serious overclockers demand better quality products. And it will happen, once manufacturers are taken to task. Your hobby is is not about serious engineering, but is about making due with what is bestowed upon you by the engineers designing for a very few overseas factories. Silicon quality is not a holy mystery, but is a function of engineering design for manufacturing, including reducing manufacturing cost and maximizing profits. If they can bamboozle everyone into accepting poor quality then its more profit in their pockets.

 

[plausible_deniability]

We are all arrogant, yourself included. (after reading the above post) Two kits with the exact same specs that finish production one after the other are not (guaranteed) to work together because of the aforementioned tolerances in components. Forcing manufactures to make anything work with anything until tech advances farther would price most of us right out of the hobby.

 

[DSzymborski]

Are all you guys so arrogant? Your not getting my issue at all. Besides the fact is that these two kits are from the same manufacturer, have the same CAS latency, the same voltage, and the same speed. The secondary times differ. I am really raising a question about why RAM manufacturing has such poor QA/QC and why motherboard manufacturers fail to make more robust systems.

Since you've descended from a regular question to angry conspiracy theory that betrays a fundamental ignorance on how semiconductor fabrication works, this thread has run its course.