Ryzen 3600X ALL Core Overclock 4425Mhz DDR4 CL16 4200Mhz

[teppic]

what you ask is impossible.Fabric can not go higher than memory but i have run fabric up to 1900 testing fine and also 1.5v on b-die ram is fine,i a not sure where people come up with these things ,any back to game.

If you have IF at 1900MHz and RAM at 3800MHz then they're in sync. You're unlikely to make any gains pushing memory above 3800MHz because of the extra latency you get from having the memory clock higher than IF.

 

[Zeed]

Memory clock-2100 =4200Mhz
Fabric clock- 1800 =3800Mhz
All core clock - =4425Mhz

Did all core overclock 25Mhz more than whats printed on the box,I like writing that.

EDIT:As per request by MrPils,CPU 1.325v ,enjoy

At the 4.00Minute mark in video I answer some question in the video lol for fun.

So I been testing for a couple days.I only play games ,so the overclock will not be prime stable and I do not even care.

Anyway I will be using boost clock which boost up to 4525Mhz and that is perfect for me.

I just did overclock thru Ryzen master and you can see all DDR4 timings.
HWinfo64 for all voltages and fan speeds.

So I fired up Kingdom Come Deliverance a game known for the great hardware killer and made a quick video.

show me 1 hourt pass of Aida stress test and i give ya my respect and memory test numbers. What timings You are running ?

 

[gerardfraser]

show me 1 hourt pass of Aida stress test and i give ya my respect and memory test numbers. What timings You are running ?

Do not look at the video for timings.Your out of your mind if you think I want any respect from you.
OH sorry maybe you can not read ,I apologize ,here ya go.
2nd video-DDR4 CL16 4200Mhz
1st video- DDR4 CL14 3600Mhz

If you have IF at 1900MHz and RAM at 3800MHz then they're in sync. You're unlikely to make any gains pushing memory above 3800MHz because of the extra latency you get from having the memory clock higher than IF.

Sorry misread your comment,thought you said 3600ram and 1900fabric.
There is a bug in BIOS after memory goes higher than 1900+ and you will not gain much FPS untill fixed.

 

[Zeed]

Do not look at the video for timings.Your out of your mind if you think I want any respect from you.
OH sorry maybe you can not read ,I apologize ,here ya go.
2nd video-DDR4 CL16 4200Mhz
1st video- DDR4 CL14 3600Mhz



Sorry misread your comment,thought you said 3600ram and 1900fabric.
There is a bug in BIOS after memory goes higher than 1900+ and you will not gain much FPS untill fixed.

so basically braging about unstable system gr8 Im sitting here testing my full stable 1.5 hour aida 30 minute realbench 3733cl14 1:1:1 ratrio 3000% ramtest and 600% Hci memtest

So what stabnility test have ya ran ??

 

[Orange Nexus]

so basically braging about unstable system gr8 Im sitting here testing my full stable 1.5 hour aida 30 minute realbench 3733cl14 1:1:1 ratrio 3000% ramtest and 600% Hci memtest

So what stabnility test have ya ran ??

Several hours of fun gaming? Haha.

 

[Zeed]

Several hours of fun gaming? Haha.

**** I can play several hour of fun gaming on My firnd Pedro i think i could get 4.4 on all 12 cores stable to play it no problem just bump my volts by 50mv job done lol

 

[gerardfraser]

Dude i am not going to argue with you,call you names.I am bragging about an unstable system that only works when gaming.Come on you do not really believe that.
so when i said

So I been testing for a couple days.I only play games ,so the overclock will not be prime stable and I do not even care.

that makes you offended and all riled up and you think your system is better than mine after 1.5hours where I on the other hand be testing for days.

Should i actually post a video in this thread where my system is actually stable with prime/intel burn test etc ,which i already posted but not in this thread.

The video BTW is to show people it can be done and I purposely made a video with 25Mhz over Max rated boost clock written on the box.You should reread MrPils first post in here.
Also I bought 2x3600X and still testing the first one and it can do 4525Mhz boost.Hoping the second one can do higher.

Maybe I will lend you one when I am done testing but certainly it wont be in 1.5 hours.J/K I would not give you ****.

 

[Orange Nexus]

**** I can play several hour of fun gaming on My firnd Pedro i think i could get 4.4 on all 12 cores stable to play it no problem just bump my volts by 50mv job done lol

Try it. If the games do not crash to desktop for months on end, then that is the true test of stability.

Only if you game in your system. If you use it for work, then that's a diff story.

 

[Zeed]

Dude i am not going to argue with you,call you names.I am bragging about an unstable system that only works when gaming.Come on you do not really believe that.
so when i said

that make you offended and all riled up and you think you system is better than mine after 1.5hours where I on the other hand be testing for days.

Should i actually post a video in this thread where my system is actually stable with prime/intel burn test etc ,which i already posted have but not in this thread.

The video BTW is to show people it can be done and I purposely made a video with 25Mhz over Max rated boost clock written on the box.You should reread MrPils first post in here.
Also I bought 2x3600X and still testing the first one and it can do 4525Mhz boost.Hoping the second one can do higher.

Maybe I will lend you one when I am done testing but certainly it wont be in 1.5 hours.J/K I would not give you ****.

Mate I'm sp[ending -19012 hours a DAY on testing since wednesday. I play Benchmarks mem tests and stabiluity test. Last weekend was tweeking memory 20 hours on sunday lol

Everyone here know I'm playing benchmarks cause I got no life so waiting for Wow classic to play 12 hour as a day every day after work.

@Orange Nexus well I do get 30 hour renderuing session from my youtube after events. But with health problems i can go out onca a month for 4hours.... rest of time im stuck home doing well benching

And I dont need a spare 3600x I got full 3600 + 2070 super on x570 system next to mine. That I'w build and been Stres testing for 4 days at same time

 

[awaybreaktoday]

My PI is best gaming system in the world

 

[gerardfraser]

PI 4 baby

 

[gerardfraser]

Some context to go with this:
Really, you don’t have much play at all at the moment.

Thanks for the information ,I should have added it in my comments.

 

[MrPils]

Thanks for doing the test

So by leaving just the last core alone at a low multi you were able to run all the other cores at 4.45ghz 1.325v, a considerable voltage drop from your previous all core setting. That really is interesting, I get the impression that 5c/10t at higher clocks will more than make up for the single lower clocked 1c/2t. Just a shame the windows scheduler isn't clever enough to assign higher load tasks on faster cores and leave all the windows background rubbish to run on the slower ones. Removing cpu affinity on the lower clocked core for whatever game/app/bench is running may be the way round any frametime or run variability issues resulting from this config. I need to test this to see if its faster, if a single core is holding my other 5 back then this could be the missing puzzle piece for beating that weird vcore behaviour PB stuff in gaming tests.

I've been clocking from bios and haven't seen any need to install ryzen master yet, mainly as I hate bloat in my OS. I'm had Intel for almost the last 15 years, this is a learning curve so I guess i have to shed the comfort blanket of bios clocking at some point.

Also realised while typing this post that my last main AMD cpu was also a 3000 series cpu (San Diego Athlon 64 3700+ in 2005)

 

[Ste]

I'm being dense but where is the reference to dropping one core? What's been done? Sounds really interesting!

 

[MasterOC]

Stability is subjective anyway if it runs your usuall loads without issue it's stable if your running anything as intensive as p95 then that's fine as long as your workload matches that or are selling on unkowing about the clients needs. , I've had chips eat the voltage for years never missing a beat then I had a 2700 which died very early on pushing really little extra voltage. These are more fragile chips then any monlith design as there's just more points of failure . Normally having a chip dies means it wouldn't even post however my "dead" 2700 would. Rule of thumb for me personally is 1.3v air max 1.4v water max.( Although have ran Upto 1.45v ) Upto 1.5 for suicide runs with extremely exotic cooling on a platform I expect to die.

Wouldn't and couldn't afford to do this to a 3950x but can completely see why I'd do this to a 200quid ish chip

 

[MrPils]

I'm being dense but where is the reference to dropping one core? What's been done? Sounds really interesting!

If you watch the second video he posted using 1.325vcore at my request, you can see he sets multiplier on 5 out of 6 cores in ryzen master, but leaves/forgets the 6th. If you watch the FPS overlay in the game you can see the final 2 threads are boosting in the high 3ghz range with the other 10 threads at 4.45ghz. His previous stated stable setting was all 6 cores 4.45ghz @ 1.4v+. Given the 1.325v run looks to be stable, that's a big drop in voltage requirements.

My theory is as follows: given we know how threaded (or not) games currently are, would disabling your worst clocking core (or even just any one core) leave enough headroom in CPU to achieve some genuinely high stable "all core" speed, comparable even to single core boost speeds using PBO/AO? Any game/bench/app that leaves spare cores on the table in terms of cpu utilisation should see a boost from using this kind of setup as long as you remove its cpu affinity to the downclocked core in windows.

I'm not going to be able to test anything tonight, I'm working late. If anyone gives this a go please post up results

 

[gerardfraser]

No problem for the test at all.
So also for anyone interested,with AMD overclock Precision Boost Overdrive Up To 4525Mhz on my chip.This AMD overclocking voids warranty for AMD lol. and can be really aggressive and maybe too aggressive for some people.
So I made a video of the game i am currently playing Outlast 2,great game.

AMD overclock Precision Boost Overdrive jist of video
Boost clocks up to 4525 Mhz
CPU voltage average 1.53V-1.57v 95% of the playing for about 45 minutes
CPU temperatures hitting 78C
I am fine with all these numbers,it may not be the same for all.