2700x Performance in PBO (Precision Boost Overdrive) only - Page 8 - Overclock.net - An Overclocking Community

Forum Jump: 

2700x Performance in PBO (Precision Boost Overdrive) only

Reply
 
Thread Tools
post #71 of 77 (permalink) Old 02-06-2019, 11:43 AM
New to Overclock.net
 
Join Date: Jan 2019
Posts: 49
Rep: 0
Quote: Originally Posted by Saiger0 View Post
I think I confused micron with hynix naming my bad. Nontheless increasing tRTP might help you out too.

1.4 Vcore is definelty too high for 24/7 stay at 1.38 if you want to avoid degrading anytime soon. And make sure you really are stable with prime95
I was able to reduce it to 1.38 and score a 1946 on Cinebench. Fails on the 2nd attempt, however. I'm OK with that, but as you allude to, the memory testing will be the fun part. Thanks again for pointing that out! I will update here when I have something more to add... hopefully more good than bad.

Memory: HX433C16PB3K4/64 @ 3000 14-16-14-17-28 1.3v / CPU: 2700x Ryzen @ 4.275 1.375v & 1.025 vSoC / Motherboard: msi x470 Gaming Pro / Power Supply: 650w Snow Silent / HDs: 250gb Samsung 970, 500gb Samsung 970, 4x 500gb 850 Samsung EVO (1tb mirror RAID).
Mike-EEE is offline  
Sponsored Links
Advertisement
 
post #72 of 77 (permalink) Old 02-06-2019, 12:44 PM
New to Overclock.net
 
CelticGamer's Avatar
 
Join Date: Jun 2015
Posts: 143
Rep: 2 (Unique: 2)
I really do think Ryzen overclocking is sensitive to temps. I spent a lot of time trying to get my 2700X stable at 4.2ghz on my MSI X470 Pro Carbon. I noticed that at 1.425V, it would pass Cinebench runs all day long, but only if I let the CPU cool in between runs. I believe it was around 68 to 70 degrees is where it would crash.

Finally, out of frustration, I just gave up trying to get it stable, as it required more voltage than I was comfortable with. I simply undervolted it, enabled PBO and just let it do it's thing.

I'm on a Corsair H60 with push pull fans at max speed, which does just fine for gaming loads, but not ideal for overclocking. Maybe one day if I get a better cooler I will try again to see if I can get a stable 4.2ghz.

Main Gaming Rig
(9 items)
Alienware Alpha R1
(5 items)
CPU
Ryzen 2700X
Motherboard
MSI X470 Gaming Pro Carbon
GPU
ROG STRIX RTX 2080
RAM
16GB ADATA XPG Z1 DDR4 3000mhz @ 3466mhz
Hard Drive
WD BLUE 1TB
Power Supply
Thermaltake Grand Toughpower RGB 750W GOLD
Cooling
Corsair H60
Case
Corsair 570X RGB
Operating System
Windows 10
CPU
Intel i7 4785T
GPU
Nvidia 860M 2GB+BlindRageVbios+335Core+650Memory
RAM
16GB G.Skill Ripjaws
Hard Drive
1TB 7200RPM
Operating System
Windows 10
▲ hide details ▲
CelticGamer is offline  
post #73 of 77 (permalink) Old 02-06-2019, 01:30 PM
New to Overclock.net
 
Join Date: Jan 2019
Posts: 49
Rep: 0
Quote: Originally Posted by CelticGamer View Post
I really do think Ryzen overclocking is sensitive to temps. I spent a lot of time trying to get my 2700X stable at 4.2ghz on my MSI X470 Pro Carbon. I noticed that at 1.425V, it would pass Cinebench runs all day long, but only if I let the CPU cool in between runs. I believe it was around 68 to 70 degrees is where it would crash.

Finally, out of frustration, I just gave up trying to get it stable, as it required more voltage than I was comfortable with. I simply undervolted it, enabled PBO and just let it do it's thing.

I'm on a Corsair H60 with push pull fans at max speed, which does just fine for gaming loads, but not ideal for overclocking. Maybe one day if I get a better cooler I will try again to see if I can get a stable 4.2ghz.
Awesome...well, not for you but for ME and my sanity. What you describe is what I have encountered. Having it freeze is the most maddening thing, too. You have no idea what caused that freeze: was it memory? CPU voltage? SoC? Something else? MAYBE A BUG?! LOL... wouldn't THAT suck.

I am indeed having all the trouble in the world with getting anything other than XMP Profile 1 (2933) to work. I tried your tRTP @Saiger0 , and while it might have fixed one problem, there is another and more that are keeping me from error-free test runs.

So I am sticking with 4.3 @ 2933 with 1.38 (I'll see if I can get that down, too). That gets me a 1948 which is plenty, considering the grief I have been through and will have to continue to "improve." FWIW anything over 1935 is "good enough" in my book.

EDIT: Ah I forgot this link: https://en.wikichip.org/wiki/amd/ryzen_7/2700x It would seem that Double Rank 4 of 4 is only 1866??? I am lucky to get 2933? Another reason to sit tight, especially with the score. Kinda sucks that I spent so much money on 3300 that I won't use, though. Doh.

Memory: HX433C16PB3K4/64 @ 3000 14-16-14-17-28 1.3v / CPU: 2700x Ryzen @ 4.275 1.375v & 1.025 vSoC / Motherboard: msi x470 Gaming Pro / Power Supply: 650w Snow Silent / HDs: 250gb Samsung 970, 500gb Samsung 970, 4x 500gb 850 Samsung EVO (1tb mirror RAID).

Last edited by Mike-EEE; 02-06-2019 at 01:38 PM.
Mike-EEE is offline  
Sponsored Links
Advertisement
 
post #74 of 77 (permalink) Old 02-06-2019, 11:20 PM
New to Overclock.net
 
Join Date: Nov 2018
Posts: 76
Rep: 0
Quote: Originally Posted by Mike-EEE View Post
Awesome...well, not for you but for ME and my sanity. What you describe is what I have encountered. Having it freeze is the most maddening thing, too. You have no idea what caused that freeze: was it memory? CPU voltage? SoC? Something else? MAYBE A BUG?! LOL... wouldn't THAT suck.

I am indeed having all the trouble in the world with getting anything other than XMP Profile 1 (2933) to work. I tried your tRTP @Saiger0 , and while it might have fixed one problem, there is another and more that are keeping me from error-free test runs.

So I am sticking with 4.3 @ 2933 with 1.38 (I'll see if I can get that down, too). That gets me a 1948 which is plenty, considering the grief I have been through and will have to continue to "improve." FWIW anything over 1935 is "good enough" in my book.

EDIT: Ah I forgot this link: https://en.wikichip.org/wiki/amd/ryzen_7/2700x It would seem that Double Rank 4 of 4 is only 1866??? I am lucky to get 2933? Another reason to sit tight, especially with the score. Kinda sucks that I spent so much money on 3300 that I won't use, though. Doh.
your 4.3 ghz with 1,38V is NOT stable if its still crashing after just 2 cinebench runs. Try running prime95 (large fft´s) for 1 hour instead.
Saiger0 is offline  
post #75 of 77 (permalink) Old 02-07-2019, 12:58 AM
New to Overclock.net
 
Join Date: Jan 2019
Posts: 49
Rep: 0
Quote: Originally Posted by Saiger0 View Post
your 4.3 ghz with 1,38V is NOT stable if its still crashing after just 2 cinebench runs. Try running prime95 (large fft´s) for 1 hour instead.
Agreed! I finally found the source of all my ails: putting the clock speed on anything past 4.275. Doing this speed + 2933mhz DRAM overclock allowed me to finally run Cinebench however many times I needed to do so. This allowed me to finally try different settings and pull samplings from Cinebench (running 5 times).

So using this, I was able to settle on vCore 1.37. I do have this on LLC1 which yields 1.38 under load.

Cinebench scores are 1948-1952 which is what I was angling around.

Also good news is that I was able to drop down to 1.35 vCore no problem @ 4.275 clock speed, but this yielded 1930 scores.

So now I will ruminate/research a little more to find out if the extra voltage is worth it. It's more of a psychological thing at this point.

Oh, I should also mention that I was able to bump up my DRAM to 3000, and with the Ryzen Calculator get the timings down to: 14-16-14-17-28. SoC is at .9875.

Running MemTest now to make sure the memory is OK with this. Then I will look into Prime95.

EDIT: Attaching my Ryzen Calculator Memory profile for more context.
Attached Thumbnails
Click image for larger version

Name:	1.png
Views:	21
Size:	70.3 KB
ID:	251968  

Click image for larger version

Name:	2.png
Views:	13
Size:	43.8 KB
ID:	251970  


Memory: HX433C16PB3K4/64 @ 3000 14-16-14-17-28 1.3v / CPU: 2700x Ryzen @ 4.275 1.375v & 1.025 vSoC / Motherboard: msi x470 Gaming Pro / Power Supply: 650w Snow Silent / HDs: 250gb Samsung 970, 500gb Samsung 970, 4x 500gb 850 Samsung EVO (1tb mirror RAID).

Last edited by Mike-EEE; 02-07-2019 at 01:15 AM.
Mike-EEE is offline  
post #76 of 77 (permalink) Old 02-19-2019, 02:27 AM
New to Overclock.net
 
PrivotSponge's Avatar
 
Join Date: Jan 2019
Posts: 9
Rep: 0
Hi there,

In what applications/games do i have an advantage with very high ram clocks?

I testet 3333mhz 16-16-16/ 3600 mhz 16-16-16 and 3733 mhz 16-17-16 (not yet stable).

I always get arround 1900 Points in cb.
Timespy Bench remains also at ca. 8300 points.

Only thing which is better, is the aida memory test:

Raised from 52000 kbs read to 58000 kbs
Also latency is a bit better: 61,4 ns to 60 ns.

So where is my benefit?

Greetz Sponge

Gesendet von meinem Moto Z (2) mit Tapatalk
PrivotSponge is offline  
post #77 of 77 (permalink) Old 02-19-2019, 05:33 AM
ORL
H20 only!
 
ORL's Avatar
 
Join Date: Sep 2011
Location: STL
Posts: 348
Rep: 18 (Unique: 17)
Quote: Originally Posted by PrivotSponge View Post
Hi there,

In what applications/games do i have an advantage with very high ram clocks?

I testet 3333mhz 16-16-16/ 3600 mhz 16-16-16 and 3733 mhz 16-17-16 (not yet stable).

I always get arround 1900 Points in cb.
Timespy Bench remains also at ca. 8300 points.

Only thing which is better, is the aida memory test:

Raised from 52000 kbs read to 58000 kbs
Also latency is a bit better: 61,4 ns to 60 ns.

So where is my benefit?

Greetz Sponge

Gesendet von meinem Moto Z (2) mit Tapatalk
One perfect example where you will see larger increases is games like ESO in cities with high player counts. These entities are tied to the CPU/Mem relationship more so than the GPU. ESO is one of the more extreme gaming examples where you can see gains of 30% FPS in this scenario. After tweaking the client a bit and moving from 2133 to 3200Mhz on my memory, the lowly 28-32 fps suddenly turned into 50-60FPS. Of which I can confirm 10-15 or so happened immediately after changing the memory settings. The rest came from optimizing the ini file to fully utilize the CPU/GPU. As I cannot control the operations of each player, there is also some margin of error in these estimates. But this is the biggest witnessed gain for me so far in any game that is repeatable to a degree.

There are other situations where you will not gain any FPS as well. When your CPU/Mem are pushing a GPU to 100% saturation and the game doesn't rely heavily on the CPU itself, you will see very little gain to even none.
ORL is offline  
Reply

Tags
2700x , Overclock , ryzen

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off