Ryzen 2600x OC

[hollyh88]

28 minutes ago, Tristerin said:

And you really are only getting a 35 point improvement for 4.2ghz OC, vs a 3.6ghz run repeated and you were also beat by the same 3.6ghz run.

 

Something is not right with your data.

 

 

if you read what i wrote in other comments. i only use cinebench to test heat and stability. not to get the most points. 

[hollyh88]

12 minutes ago, Mathieu9836 said:

You can do whatever your are pleased, but i would not throw 1.45volts to a 2600X just to get 3 more fps average.... that's playing with fire just because it's fun. Get your clockspeed back to 4.0ghz with the lowest stable voltage possible, test it with either aida64 or Intel Burn test for about 15 minutes. Then work those memory timings down, even the cheapest Team Group 3200mhz kit can run at CL16, you probably can run at CL14 or maybe CL15 which will probably give you more FPS than having you CPU running hotter with a shit ton of voltage.

 

Download Typhoon Burner and see what dies your RAM kit has, then try Ryzen DRAM Calculator.

 

Wasting money for a 3700X when your 2600X get the job done is just, wasting money.

And what motherboard are you using? If it's a weak VRM that might explain why you need to increase the voltage to an unsafe level.

yeah i took it off the 1.45v its at 1.42v 

3 fps in some games. more in others.  
more performance is more performance. 

typhoon burner gives a massive red flag on my virus scanner.

also, the difference between the 2600x and the 3700x is not little though. i mean about 20+ more fps. thats not little  
my mobo is in my signature. 
Its the msi b450 gaming pro carbon ac. which as far as i know has excellent vrm

 

[Tristerin]

So your all core OC dips to 1.3ghz and peaks at 5.2ghz on the same software you are taking your temperature reports from.  Odd choice.

 

Please someone else, help me understand how a 3.6ghz run, outperforms a 4.2ghz run on CinebenchR15 with the same CPU and setup - unless its thermal throttling, or VRM throttling because its not keeping the OC constant because it cant due to temps or budget power delivery.  This is what I am seeing, its pretty straightforward.  Id like to be educated on this as my understanding of how things work must be wrong.

[Tristerin]

Also a Hyper 212 can put you in the top 5 world wide for that CPU on Cinebench R15 - that's insane, or inaccurate.  Im still in disbelief.  @nick name @Zando Bob can someone help me understand how this is accurate.  The way I understand dissipation, OCing, etc is either wrong and I need re-educated or I need people smarter than me to take a glance at these numbers and correct my behavior or tell me if what I am saying is accurate.

 

@OP - My Ryzen 7 1700 is in the top 10 (on R20), but it took a custom loop.  I still do not understand how this performance is achieved because mathematically I just don't see it as possible, as Ive done my OCing on Air, AIO, and only a custom loop could tame those voltages and take away that heat.  

 

Can you do a R20 run on that @hollyh88 as its more brutal than R15.  Maybe that is where my disconnect is perhaps, as I don't use R15 because its not brutal enough. 

 

 

[Zando_]

6 minutes ago, Tristerin said:

Also a Hyper 212 can put you in the top 5 world wide for that CPU on Cinebench R15 - that's insane, or inaccurate.  Im still in disbelief.  @nick name @Zando Bob can someone help me understand how this is accurate.  The way I understand dissipation, OCing, etc is either wrong and I need re-educated or I need people smarter than me to take a glance at these numbers and correct my behavior or tell me if what I am saying is accurate.

 

22 minutes ago, Tristerin said:

Please someone else, help me understand how a 3.6ghz run, outperforms a 4.2ghz run on CinebenchR15 with the same CPU and setup - unless its thermal throttling, or VRM throttling because its not keeping the OC constant because it cant due to temps or budget power delivery.  This is what I am seeing, its pretty straightforward.  Id like to be educated on this as my understanding of how things work must be wrong.

Uh what now? A 4.2Ghz OC on X58 can beat some 4.5Ghz OCs if the uncore and RAM are tuned properly, but Ryzen doesn't have that. If you ran really shit RAM on the 4.2Ghz run it might be outperformed but stiiiiil that's a large MHz lead. I'm as bamboozled as you are. 

OP, do you have the chipset drivers installed? Also try using Ryzen Master, if HWMonitor is reporting wack speeds (I've never had an issue with it though), Ryzen Master will usually report accurate numbers and always the most accurate temps. 

Also 1.45v is really not safe for the 2000 series, like the other guy said at the beginning, 1.4v is the safe safe zone. Some claim 1.38v or so but most people doing 1.4v have been fine. 

[Mathieu9836]

26 minutes ago, Tristerin said:

Please someone else, help me understand how a 3.6ghz run, outperforms a 4.2ghz run on CinebenchR15 with the same CPU and setup - unless its thermal throttling, or VRM throttling because its not keeping the OC constant because it cant due to temps or budget power delivery.  This is what I am seeing, its pretty straightforward.  Id like to be educated on this as my understanding of how things work must be wrong.

Thermaltrottling can't happen with static all core OC, unless you hit the 95C wall. VRM can't trottle unless they hit and insane temperature and probably will never happen on a B450 Pro Carbon which is very good for a 2600X. I've had 3.8ghz better than 4.1ghz runs because of background apps running, but generally it's RAM related, i tweaked the RAM and lowered the CPU clockspeed before trying to OC the crap out of my RAM sticks.

 

I must admit that 5.2ghz on HWinfo is inaccurate, CPU would crash before hitting 4.5ghz.

 

Static all core can drop down to anything when you have windows power management ON, my 2700 hit 1.2ghz on idle and wake needed number of cores on load.

[Mathieu9836]

29 minutes ago, hollyh88 said:

typhoon burner gives a massive red flag on my virus scanner.

 

Run it as Administrator, or turn windows defender OFF.

[hollyh88]

32 minutes ago, Tristerin said:

Also a Hyper 212 can put you in the top 5 world wide for that CPU on Cinebench R15 - that's insane, or inaccurate.  Im still in disbelief.  @nick name @Zando Bob can someone help me understand how this is accurate.  The way I understand dissipation, OCing, etc is either wrong and I need re-educated or I need people smarter than me to take a glance at these numbers and correct my behavior or tell me if what I am saying is accurate.

 

@OP - My Ryzen 7 1700 is in the top 10 (on R20), but it took a custom loop.  I still do not understand how this performance is achieved because mathematically I just don't see it as possible, as Ive done my OCing on Air, AIO, and only a custom loop could tame those voltages and take away that heat.  

 

Can you do a R20 run on that @hollyh88 as its more brutal than R15.  Maybe that is where my disconnect is perhaps, as I don't use R15 because its not brutal enough. 

 

 

i would but even running stock settings out of the box cpu performance seems to crash r20. the only thing that so far worked is having pbo enabled and just that cpu wise. maybe a bug? maybe not. i dont know. pic below is with pbo

[Mathieu9836]

Turn that sh*t off it's giving your CPU 1.5volts.

[hollyh88]

53 minutes ago, Tristerin said:

So your all core OC dips to 1.3ghz and peaks at 5.2ghz on the same software you are taking your temperature reports from.  Odd choice.

 

Please someone else, help me understand how a 3.6ghz run, outperforms a 4.2ghz run on CinebenchR15 with the same CPU and setup - unless its thermal throttling, or VRM throttling because its not keeping the OC constant because it cant due to temps or budget power delivery.  This is what I am seeing, its pretty straightforward.  Id like to be educated on this as my understanding of how things work must be wrong.

im not a cinebench master as yourself. i just boot it up and use it if i want to test an oc.  or temps  

i do know that r20 is more unforgiving oc wise. and puts more stress on the cpu. so ultimately it would fail more or get hotter on there. dont seem to be an issue currently well.. only with pbo that is. 

[Tristerin]

Could there be a BIOS bug?  Pic shows 3.6ghz run on the App (cineR20) but 4.2 on HWmonitor.  

[hollyh88]

2 minutes ago, Mathieu9836 said:

Turn that sh*t off it's giving your CPU 1.5volts.

i know which is another weird thing. if i use pbo via mobo it can go up very high. if i use pbo via ryzen master it stays at save volts.

[hollyh88]

Just now, Tristerin said:

Could there be a BIOS bug?  Pic shows 3.6ghz run on the App (cineR20) but 4.2 on HWmonitor.  

i have the newest so maybe? 

[Mathieu9836]

How do you actually OC your CPU? Increase voltage and CPU clock speed? Turn everthing else off if you didn't.

[hollyh88]

32 minutes ago, Zando Bob said:

 

Uh what now? A 4.2Ghz OC on X58 can beat some 4.5Ghz OCs if the uncore and RAM are tuned properly, but Ryzen doesn't have that. If you ran really shit RAM on the 4.2Ghz run it might be outperformed but stiiiiil that's a large MHz lead. I'm as bamboozled as you are. 

OP, do you have the chipset drivers installed? Also try using Ryzen Master, if HWMonitor is reporting wack speeds (I've never had an issue with it though), Ryzen Master will usually report accurate numbers and always the most accurate temps. 

Also 1.45v is really not safe for the 2000 series, like the other guy said at the beginning, 1.4v is the safe safe zone. Some claim 1.38v or so but most people doing 1.4v have been fine. 

i have all the latest drivers at least im sure did a check in the weekend even. 
ryzen master does indeed report correct. but hardware monitor 99% of the time works fine. the other 1% it reports something wacky  

[hollyh88]

Just now, Mathieu9836 said:

How do you actually OC your CPU? Increase voltage and CPU clock speed? Turn everthing else off if you didn't.

cpu clock speed. and voltage yes. not offset just per amount it allows. 

[Tristerin]

3 minutes ago, hollyh88 said:

i have the newest so maybe? 

The reason I ask is because (the bug Im talking about is a ghost in BIOS where its keeping old values or reporting incorrect values, fixable by resetting the CMOS) all of the screenshots Im seeing reporting information is off across the boards.

 

You also hit the same temp of 71c between R15 and R20 - I do not understand this because that's not how it works.  And a Hyper 212 Evo cannot hold a CineR20 run with 1.5vCore.

 

Impossibru.

 

[hollyh88]

good news i managed to get it to work with simple out of box no pbo at all settings  here is the result. 

[hollyh88]

9 minutes ago, Tristerin said:

The reason I ask is because (the bug Im talking about is a ghost in BIOS where its keeping old values or reporting incorrect values, fixable by resetting the CMOS) all of the screenshots Im seeing reporting information is off across the boards.

 

You also hit the same temp of 71c between R15 and R20 - I do not understand this because that's not how it works.  And a Hyper 212 Evo cannot hold a CineR20 run with 1.5vCore.

 

Impossibru.

 

im just showing you what i got lol XD i am not manipulating anything. 

[Mathieu9836]

If you have latest BIOS update, i would update to an older stable version right before they released the 3000 series CPU bios update, thats what i did with my C7H which is now more stable. I would also turn everything off about the CPU overclock like CB15 optimizing stuff pbo and cool n quiet, anything that can mess with your overclock. I would also try offset voltage which will help you keep you CPU cooler at idle. But thats me, i like to have total control of my hardware and how it behave.

[nick name]

Don't use HWMonitor with Ryzen.  Use HWiNFO.  

[Tristerin]

10 minutes ago, hollyh88 said:

good news i managed to get it to work with simple out of box no pbo at all settings  here is the result. 

 

I still don't understand how Cinebench R20 is displaying 3.6ghz runs (while HWmonitor is showing 4.1ghz runs now).  Something is not reporting correctly.  In fact your R20 run is 3C less than an R15 run at the same "supposed" speeds.  

[Tristerin]

Example - this is how my runs look when OCing (this is a pic of mine from a few months ago) I.E. even Cine recognizes the OC.

 

 

 

[hollyh88]

20 minutes ago, Mathieu9836 said:

If you have latest BIOS update, i would update to an older stable version right before they released the 3000 series CPU bios update, thats what i did with my C7H which is now more stable. I would also turn everything off about the CPU overclock like CB15 optimizing stuff pbo and cool n quiet, anything that can mess with your overclock. I would also try offset voltage which will help you keep you CPU cooler at idle. But thats me, i like to have total control of my hardware and how it behave.

i could do that yes. go back to how it was before the 3000 series came out. but... the bios hasnt been giving me any issues so far. not even the new one. i actually recommended this mobo to a friend. he upgraded to this mobo with a 3600 and 2070 super. 

he hasnt been having issues either.

[Bacon soup]

@Tristerin I have a 2600 and a MSI B450 Mortar Titanium. If I use AMD's Ryzen tool, then cinebench does not see my overclock. Cinebench R20 is also able to crash my CPU where others can't. When Cinebench fails to see my cpu freq, I know it has overclocked because it does not crash on stock settings and I can see the heat rising. This has only been my first night of overclocking. I will try again in two weeks when my cooler, and other important parts, arrive.