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

gerardfraser · 16 Jul 2019 at 03:18

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.

gerardfraser · 16 Jul 2019 at 03:37

LOL youtube quality,I could not even read the 4K upload.I will re-upload for those interested
Memory clock 1.5v -2100 =4200Mhz
All core clock CPU 1.4v-4425Mhz

DragonQ · 16 Jul 2019 at 07:19

No offence but I'm sure most people could get a great overclock if they didn't care about stability. I mean it's kinda cool but I'd never trust it on a daily use system.

gavinh87 · 16 Jul 2019 at 07:23

1.4v is also too high for these new zen cpus.

Yaayuh! · 16 Jul 2019 at 07:32

FYI, when you uncouple the MEMCLK from FCLK you increase the memory latency as the data has to synchronise back.

Grim5 · 16 Jul 2019 at 08:36

gavinh87 said:
1.4v is also too high for these new zen cpus.

Fk using 1.4v on the daily.

Ste · 16 Jul 2019 at 10:33

Cheap processors, take more risk, 1.4v probably be ok for a couple of years before deg if you stay off the torture tests

MrPils · 16 Jul 2019 at 10:38

Regardless of peoples opinions on stability this is one of the highest all core OCs i've seen - this dude's chip is good silicon. Where it gets interesting is comparing it to my "bad" 3600(not x), I can get 4175mhz all core @ 1.425v custom watercooled. I imagine with another 0.025v you'd be able to reach whatever stability target people set provided it was cooled, so it looks like the OC spread is very roughly 250mhz best silicon to worst silicon. There will be outliers, I certainly wont have "the worst ever", but a 250mhz range doesn't leave much wiggle room. With so little room to work with its no wonder PBO can do 90% of what a manual overclock can achieve.

Just as a further point of comparison, and if you don't mind spending the time to do it, where can you get to all core with 1.325v fixed? This seems to be what people are recommending for super safe all core daily volts. Mine does 4050mhz @ 1.325v, 25mhz more and it bluescreens during realbench. Full load temps reach a mighty 47c running AVX, I'm definitely not temperature limited lol.

Lolik · 16 Jul 2019 at 10:52

1.4v ok (sort of) if your digi vrm set quite low, otherwise it will boost much higher and for longer periods ( bet that is how hardware unboxed burned his 3900x ).

Orange Nexus · 16 Jul 2019 at 11:42

You need to give reviewers a lesson or two in oc'ing the 3600X.

Drollic · 16 Jul 2019 at 13:23

Orange Nexus said:
You need to give reviewers a lesson or two in oc'ing the 3600X.

What, by illustrating that pushing volts up and possibly being lucky with a chip it leads to not 100% stable speeds? There's ocing, then there's 1.5v through ddr4, which long term I don't think anyone should advise.

Alan1969 · 16 Jul 2019 at 13:29

would you recommend OC via Ryzen master rather than manually in the bios. I have no experience with Ryzen master.

Steveocee · 16 Jul 2019 at 14:13

MrPils said:
Regardless of peoples opinions on stability this is one of the highest all core OCs i've seen

Yeah the reason for that is because he hasn't stability tested it. It'd likely fall over if tested.

teppic · 16 Jul 2019 at 14:32

You might get better performance if you drop RAM speed to 3800MHz and go for 1900MHz infinity fabric, keeping them in sync.

MrPils · 16 Jul 2019 at 15:39

Steveocee said:
Yeah the reason for that is because he hasn't stability tested it. It'd likely fall over if tested.

If its stable enough to be in windows and running a game without crashing then its at least close to stable (whether that would require dropping a few mhz or increasing volts). Don't get me wrong, that's not what i'd call a finished overclock, but that still leaves the fact that its the fastest all core oc I've seen shown. I cant even launch a game at 4.2 / 1.4v - instant BSOD. He definitely has a good piece of silicon imo, as the weeks go by we'll see more results to know just how good.

Orange Nexus · 16 Jul 2019 at 17:54

Drollic said:
What, by illustrating that pushing volts up and possibly being lucky with a chip it leads to not 100% stable speeds? There's ocing, then there's 1.5v through ddr4, which long term I don't think anyone should advise.

True. Max recommended by AMD is 1.325v. But it is only 250 quids. lol. As for the RAM, 1.5 is safe but lower is better. I set 1.43v for 3533 CL14. Anyways, presets in this guide includes one with 1.5v.

https://www.techpowerup.com/review/amd-ryzen-memory-tweaking-overclocking-guide/9.html

Sargatanas2511 · 16 Jul 2019 at 18:11

Orange Nexus said:
True. Max recommended by AMD is 1.325v. But it is only 250 quids. lol. As for the RAM, 1.5 is safe but lower is better. I set 1.43v for 3533 CL14. Anyways, presets in this guide includes one with 1.5v.

https://www.techpowerup.com/review/amd-ryzen-memory-tweaking-overclocking-guide/9.html

From videos and articles I have seen recently, it almost seems like undervolting is more beneficial than overvolting to be honest!

Orange Nexus · 16 Jul 2019 at 18:32

Sargatanas2511 said:
From videos and articles I have seen recently, it almost seems like undervolting is more beneficial than overvolting to be honest!

Yah I saw a 3900 benefit from lowering vcore. There is a point of diminishing return, Too much and the cpu boosts lower. Same with previous gens. BIOSES are still being worked on.

gerardfraser · 16 Jul 2019 at 19:43

gavinh87 said:
1.4v is also too high for these new zen cpus.

This is false,too low voltage is no good

Grim5 said:
Fk using 1.4v on the daily.

You need higher voltage on Ryzen 3K ,just go over to overclock.net and see what stilt has to say about low voltage

MrPils said:
Regardless of peoples opinions on stability this is one of the highest all core OCs i've seen - this dude's chip is good silicon. Where it gets interesting is comparing it to my "bad" 3600(not x), I can get 4175mhz all core @ 1.425v custom watercooled. I imagine with another 0.025v you'd be able to reach whatever stability target people set provided it was cooled, so it looks like the OC spread is very roughly 250mhz best silicon to worst silicon. There will be outliers, I certainly wont have "the worst ever", but a 250mhz range doesn't leave much wiggle room. With so little room to work with its no wonder PBO can do 90% of what a manual overclock can achieve.

Just as a further point of comparison, and if you don't mind spending the time to do it, where can you get to all core with 1.325v fixed? This seems to be what people are recommending for super safe all core daily volts. Mine does 4050mhz @ 1.325v, 25mhz more and it bluescreens during realbench. Full load temps reach a mighty 47c running AVX, I'm definitely not temperature limited lol.

Now you get it,so I done up video as you will see when i upload at 1.32v on cpu .I rather run higher voltage

teppic said:
You might get better performance if you drop RAM speed to 3800MHz and go for 1900MHz infinity fabric, keeping them in sync.

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.

DOHH I just noticed I forgot the last two cores ,it is ok though and game has 60FPS cap sorry .Well this could go on all day ,I am done for now back to testing this CPU.Thanks for the replies

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

Yaayuh! · 16 Jul 2019 at 21:44

gerardfraser said:
This is false,too low voltage is no good

You need higher voltage on Ryzen 3K ,just go over to overclock.net and see what stilt has to say about low voltage

Some context to go with this:

TheStilt said:
Its not wise to undervolt Matisse, unless you enable the OC-Mode (fixed frequency).
That's because the CPU is constantly monitoring its supply voltage. When it sees the supply voltage below the desired level, it starts increasing the VID request from the VRM controller. When the VID request increases above the maximum voltage allowed by FIT, the clock stretchers will kick in. Your performance will decrease, despite the change is not visible in the frequencies diplayed by the usual monitoring software.
For example on the 3900X sample I have, the default VID request during CB20 NT is 1.3125V. With the offset set to -25mV the VID request raises to 1.33125V or so. At -75mV the VID request is already 1.36875V and increasing the negative offset any further will trip the stretchers.
There is an undervolting margin available, but it is tiny. The exact behavior will depend on the silicon specimen and the workload.
Its also the very reason why this guy thinks he is running at > 4.2GHz with just 1.00V...

TheStilt said:
Matisse can easily reach intensity of > 1.5W/mmÂ² (120W+ @ 74mmÂ²). The second issue is, that beyond ~3.8GHz the V/F curve becomes extremely steep. According to FIT, the safe voltage levels for the silicon are around 1.325V in high-current loads and up to 1.47V in low-current loads (i.e ST)

Really, you don’t have much play at all at the moment.