Best of luck.
Just about all RAM should work together and usually do but sometimes different chips just don't play nice together.
Unfortunately they don't advertise what chips they use on there RAM and worse often change them within the same model name.
Sorry seems I made the mistake of taking my info from Amazon which is incorrect unless kingston has released a few revisions of that model RAM.
So being 1.5v will make things easier.
Your old RAM is rated 1.6v while the new kit is only 1.5v
It may be possible that they will work ok together at 1.5-1.6v 1600c10 but by using different models of chips on the RAM with different v and timing rating you are increasing the chances that it will be unstable even with some manual...
It depends on the program.
I have found 3000c12 to perform just behind 4000c17 (tweaked subs for both) and much better than 3200c14 XMP with a 6700K.
For most the games I play I find that the final latency is the most important but if you can keep the latency low and squeeze in a little more...
I think any decent kit of b die will do 4000c17 ok weather it be 3200c14 or 4000c17 but the lower bin may need higher vdimm ~1.45v vs ~1.4v so less headroom to push higher.
At least that is the case with 8GB sticks not so sure about 16GB.
If the CPU\MB are your limit I don't think it will...
These two go hand in hand with higher frequency equaling more bandwidth but latency is also very important which is why for some programs quad channel systems with massive bandwidth are actually slower than dual channel systems.
This chart is good for seeing where frequency meets timings for...
The Asus MB has 4x8GB kits on the QVL up to 4500 while the MSI UNIFY has no QVL and the more expensive ACE only goes up to 4266 with 4x8GB.
Also G.Skill has no Z490 MSI or GB MB on the QVL for 4x8GB 4000c15.
I have always found MSI BIOS to be a pain when overclocking Intel systems although they...
If I was buying 10900K with 32GB RAM for overclocking it would be with.
ASUS ROG Maximus XII Hero
With two kits of
Patriot Viper Steel Series DDR4 16GB (2x8GB) 4400MHz PC4-35200 Dual Memory Kit - PVS416G440C9K
But I suspect I will end up with Ryzen 4000\5000 next month reusing two 2x8GB kits of...
Viper Steel DDR4 4400 16GB (2x8GB) C19
I have seen people running these 3733 14-15-15 and the A2 PCB with b die...
I would say it is as important but you have to spend more on RAM and push to higher speeds than you do with AMD to see the same kind of performance gain.
If you replaced your 3600c18 RAM with say 4000c15 RAM or just 4400c19 and overclocked it then you can expect to see a bigger performance gain...
Most 8GB sticks will be single rank while most 16GB sticks will be dual rank.
Crucial Ballistix sticks using Micron e die seem to be some of the easiest to hit high frequency with on AMD Ryzen.
Also your MB is fairly decent so that should work in your favor with dual rank 16GB sticks tested...
The more sticks you add the harder it is on the memory controller especially on cheaper MB which is why you wont see top speeds with 4 sticks only with 1-2.
If the RAM is dual rank then that makes it harder again to run.
For example this is with older Ryzen CPU but you get the idea.
If you want to take the time to OC\tweak the RAM you want Samsung b die as I suspect being quad channel you already have ample bandwidth for games but quad channel usually comes with a latency penalty and once tweaked b die will get you the lowest latency.
Any idea if your platform has a...
I found that with a 16GB GPU the game can use over 12GB VRAM and only ~10-12GB RAM at 3440x1440 ultra which with what the system uses can put it close to needing more than 16GB unless you keep your OS tidy.
However when it used 12GB RAM it only used ~10G VRAM and when it used 12G VRAM it only...
You shouldn't be able to tell the difference between single and dual rank outside of synthetic benchmarks.
Run a memory benchmark before and after with something like AIDA64 or userbenchmark.
Results will likely be within margin of error unless you close down a heap of background processes.
Need 32GB or would a faster 16GB kit be preferable?
Is the MB 2 slot or a 4 slot with T-Topology?
Overclocking or just running XMP?
This would have to be one of the fastest options
ROG MAXIMUS XII APEX
Does aida64 benchmark show a change in latency with 3200 vs 4000 and if you could show what timings changed would be helpful.
At least you know 4000 is possible.
If you take those settings and input everything manually I wonder will the other sticks then boot and allow you to tweak down timings...
Works on mine but wouldn't work on somebody's ASRock MB when I asked them for a screenshot of it one day.
If the two kits work alone but not together it isn't covered by warranty but otherwise there is no real difference other than it may be more likely to set automatic timings that just work...
If you want to run 4000+ a kit with A2 PCB like the the 4400 kit I suggested above would be best but I wouldnt be in a rush to buy more RAM as you should have no trouble reaching at least 3600 with the kits you have so I suspect your MB is trying to set some of the sub timings way to tight...
You want 1.45 to 1.5vdimm
Make sure it is set to T2
start with 3600 and try work your way up.
It would be interesting to compare the 3200c15 kit too see if it needs more v to reach the same frequency as it will probably be exactly the same RAM just a slightly lower bin.
Who knows maybe the old...
Is it good value? well it depends on how you look at it.
3900XT and 10900K cost a lot more than Ryzen 3600 but as far as most current games go they don't provide any more performance than a high speed kit of RAM tuned up vs a 3200c16 kit at stock.
The faster CPU or the faster RAM wont be all...
It is not just the frequency but also the timings.
You will notice that the tests run
3200c14 is fairly quick and would be faster than a 3600c17 kit for example and would probably have a lower real world latency than 4000c19 although the higher bandwidth that...
Some tests I ran with 6700K 3200c14 RAM
Time taken to export 150 24mp .nef to jpeg
Time taken to build 150 24mp 1:1 previews
Here is some tests I ran with this photoshop benchmark...
Then one of us are not using it right as all the G.skill 3200c14 kits listed in my first link are in the b die finder.
8GB sticks are single rank and 16GB sticks are dual rank which basically means double the number of b die chips on the PCB.
If you are looking to OC the RAM b die will allow you to tighten down timings and subs more than any other RAM for maybe another 5% CPU performance vs a tweaked kit of Micron chips. Maybe more with Zen3.
Newegg has 3200c14 2x16GB kits starting at $188 these should all be using Samsung b die...
Tweaking the subs alone can bring you around 10% take a look at what Ryzen DRAM calc suggests for your kit for a starting point.
If you chose the right RAM you can get some good speeds like 3200c14 kits using Samsung b die can reach around 3600-3733c14 or 4000c16 with very tight subs.
With that T-topology MB I would try running four of these sticks myself.
Patriot Viper Steel Series DDR4 16GB (2 x 8GB) 4400MHz Performance Memory Kit...
3466c14 is faster than 3600c16.
See if it passed HCI memtest which is in the RAM calc.
If so feel free to leave it at that for now and maybe do some more tinkering in the future as that kit should do 3733 no problem once you find the right settings.
There will be a timing that is just to tight and while I have seen others make recommendations to get trouble MB working I cant remember what they are.
One thing that may be worth a go is using the 3800 safe timings at 3200\3600 with 1.45vdimm 1.1SOC