DDR5-7600 memory is compatible with the Core i9-13900K.

you know the question was about maximum ddr5 speed, not max memory controller speed ?
oh boy, it really looks like your first day to pc forums, probably spent previous couple of years in youtube comment section circlejerk.
you don't know jack about memory tuning comapared to nizzen.

 

The smartest would be to go for max performance and not highest number clock.
I am pretty sure tuning at 6200 or 6400 is waste of time compared to just set 6000 and get that speed optimized.
AM4 memory at 3800 unlinked to fabric, runs slower then 3200 memory at 1:1 at least on my setup, so Nizzen would probably need to hit 7000-7500 unlinked to reach the same performance as 6000 at 1:1, because of the fabric performance hit.
I am looking forward to see realistic results instead of insulting people.
Had to add realistic, because there has previously been posted a good deal of suicide runs, chilled runs and water cooling out the window and other non daily setups, with the text "look it is easy everyone can do it."

 

except he is not measuring performance here lol only answering the question about max ram clocks.
and how do you know he's not running 1:1 on that 6200 ?
here are your realistic results, where the test is not limited by running a card that's too fast for the resolution in normal, cpu light scenarios (see how even an i3 gets 150fps in the useless HUB test) but rather limited by the cpu heavy game location that actually takes a heavy hit on the cpu perfomance
https://www.purepc.pl/test-procesor...szy-porownanie-w-grach-i-programach?page=0,58
4c/8t 12100 sits in the 40s here, that's representative of a cpu test done properly.

same here,where older architectures fall behind 12th gen/r7000 massively
https://www.purepc.pl/test-procesor...szy-porownanie-w-grach-i-programach?page=0,55

or here, where cache and latency are the limiting factor for most cpus,even pretty decent ones like said 12100 struggling to keep 60fps
https://www.purepc.pl/test-procesor...szy-porownanie-w-grach-i-programach?page=0,51

7700x is slower than 12600k w. ddr4 in those cpu heavy scenarios.that's realistic. It's something I'm likely to see on any gpu because it's limited to 40-60fps in certain game locations on older/lower tier cpus , it's not a contest what cpu gets closer to 200fps with a 3090Ti at 1080p in cpu light games, where even a locked i3 was still enough to get 150fps out of the fastest gpu available to buy (and 1% above 100fps). That's what techtubers would have you believe a cpu review is cause they can't do a proper one.
all it tells me is that there are many people on this forum who can't interpret the data they are presented with in the slightest.

 

lol that's actually 1:1 mode at 6200mhz, this guy was crying so much about how you're a pc-illiterate imposter, nizzen
same as 1:1 mode for 3000/5000 meant 1900mhz with 3800mhz ddr4

 

Are those results stable from run to run because Hilberts Aida 64 results are all over the place.
Memory 6000 30 38 38 96
7600 Read 58 Write 79 Copy 59
7700 R 58 W 81 C 59
7900 R 87 W 76 C 69
7950 R 74 W74 C 69
All of them roughly same 70ns

 

latency is what matters most.it's very high compared to r5000,even running basic 3600 c18
https://www.reddit.com/r/Amd/comments/jjnnya/aida64_cache_and_memory_benchmark_r9_5950x/

 

They are stable yes.
Difference in bandwidth has to do with different amount of CCX and different amount of cache.
7950x result from Hilbert is the same as mine on 6000 "xmp"

 

Someone needs to let Nizzen know that running memory above 6000mhz on Zen 4 breaks the 1:1 infinity fabric ratio. So much for being a die hard enthusiast and all knowing.

you do know that latency from GPU memory especially on those that are using 384-bit bus is astronamically higher than desktop memory latency. you are literally going high places with this system memory stuff to make up for the insane memory latency from the GPU. and then on top of that you are using monitors with 8ms-16ms which is at the other side of the universe when comparing to nano seconds of zen 4 memory latency. it is meaningless to worry about 50-70ns on zen 4 when you are dealing with ms on the output anyway lmao.

but better days are coming with improved cache on both gpu and cpu. 4090 with 96mb of L2 cache paired with 13900k's 32mb of L2 cache will improve everything no matter if you get 20ns or 80ns on your desktop memory, stop wasting your time in places that deal with nano seconds,

with 13900k and 4090 the ddr5 4800mhz and ddr5 8000mhz will have little to no impact in fps due to new and improved L2 cache. the 30 series had like just 2-4mb of L2 cache. the next architecture is very different than what is out now and literally nobody is fully aware of it. the secret weapon of raptor lake wont be higher clocks or being able to run higher clock ddr5 memory but it will be the doubling of the L2 cache. zen 4 will further improve over alder lake when used with 4090 due to zen 4 having 2x more of L2 cache per core than alder lake but raptor lake will have 2x more of L2 cache per core over Zen4.

 

Someone is trying to learn me something when I'm trying to benchmark 13900k with a-die ddr5. Oh well

 

I should shorten this to "blablabla Intels rep guy" but...
Yeah.
Teach us.
Nizzen knows nothing xD

 

question for you, does bandwidth scale past 6000/6200mt/s with the uclk:memclk ratio in 1:2 mode? just curious since the fabric of zen4 should handle up to 115-120gb/s or so.

 

Frame Chasers YT compare DDR4 4100C15 (quad ranks) vs DDR5 7600C36, 11 games.



Looks like DDR4 still has a long life ahead of it, DDR4 2x16GB 4000C16 are selling for slightly cheaper than DDR5 2x16GB 6400C32 and cheap DDR4 motherboards can take 4400MT/s no problem

 

Which DDR4 RAM kit is that?! Just asking...

 

Frame Chaser gave the link to this 3600C14 kit

 

@Valken
so far i havent seen any 3600C14/above 4000, that doesnt run b-die, especially if its a quad kit.
just depends if you get die that likes clocks or timings (at same V),
the few i had, were good with lower timings, but only 2 stick kits.

Patriot
https://smile.amazon.com/gp/product/B09NLBY8XB/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1
Corsair
https://smile.amazon.com/gp/product/B07VGJGKX1/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1