• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Quick and dirty early r3 1200 OC results

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Ruiner

Ruiner

Member
​
Joined
Jan 23, 2001
I picked up a pair of these from MC for the kids, 99 each plus around 30 for the Asrock b350 boards with bundle discount. Using the OE Stealth coolers, both are priming away at 3900, stock Vcore. I tried for 4000, but got a black screen after a few seconds of small fft. Pleasantly, the Ripjaw 3200 kits I bought are working at 3200, albeit at xmp 1.35V (2x4gb kits). Somewhat disappointingly, the clock speed is locked and doesn't drop at idle. I guess it's a bios issue with C&Q.
The internal cpuZ benchmark is near identical to my Ivy at 4.3.
 
The most surprising thing in your experience is that the Ripjaws RAM is operating at the full 3200 mhz. First I have heard of that. Usually, it takes the more expensive Flare X or Trident Z to run at 3200. Wonder if this is due to recent bios tweaks or improvements via CPU stepping?
 
I wasn't optimistic when I picked them up on sale, but this was a budget build and with ram prices being what they are...
IIRC 4gb sticks are more likely to run at full speed. I did update the bios out of the box before overclocking the ram or cpu.
Vcore at these speeds is defaulted at 1.1875 in bios, and shows 1.22 under load in HW monitor.

Digging further into the advanced Zen settings shows custom Pstate settings that change from 'auto' when the clock speed is changed in the main 'OC tweak' menu. This is probably where the clock speeds are changed to fixed. Vcore does drop at idle however to ~.9v.
 
trents said:
The most surprising thing in your experience is that the Ripjaws RAM is operating at the full 3200 mhz. First I have heard of that. Usually, it takes the more expensive Flare X or Trident Z to run at 3200. Wonder if this is due to recent bios tweaks or improvements via CPU stepping?
Click to expand...
I was thinking the same Mr. T!

Ruiner maybe have a looksy here Johan45 wrote a pretty good article on modifying the P-states
 
Should I be optimistic that I can hit 3900 at such low Vcore? Not that I would go further on stock cooling.

One unrelated item that had me pulling my hair out for a while: Win10 auto-installed onboard NIC drivers and in all it's settings looked fine, but my router refused to assign it a DHCP address until I installed the proprietary drivers.
 
Ruiner said:
Should I be optimistic that I can hit 3900 at such low Vcore? Not that I would go further on stock cooling.

One unrelated item that had me pulling my hair out for a while: Win10 auto-installed onboard NIC drivers and in all it's settings looked fine, but my router refused to assign it a DHCP address until I installed the proprietary drivers.
Click to expand...
When you say stock V core, is it set manually at stock or is it still on Auto? If it's on auto it's likely the board is raising it to compensate for the increased demand.
 
Mandrake4565 said:
When you say stock V core, is it set manually at stock or is it still on Auto? If it's on auto it's likely the board is raising it to compensate for the increased demand.
Click to expand...

By stock I mean a default numerical value displayed in bios/efi on a fresh cmos reset , which in this case is 1.1875. There is no 'auto' setting as such although I haven't tried typing in 'auto'. The HWmon readings I mentioned are in windows at idle and load.
 
Ruiner said:
By stock I mean a default numerical value displayed in bios/efi on a fresh cmos reset , which in this case is 1.1875. There is no 'auto' setting as such although I haven't tried typing in 'auto'. The HWmon readings I mentioned are in windows at idle and load.
Click to expand...
OK well I suppose there is only one way of finding out if it will do 3.9 at that voltage. :D
 
Ruiner said:
Should I be optimistic that I can hit 3900 at such low Vcore? Not that I would go further on stock cooling.

One unrelated item that had me pulling my hair out for a while: Win10 auto-installed onboard NIC drivers and in all it's settings looked fine, but my router refused to assign it a DHCP address until I installed the proprietary drivers.
Click to expand...

Network problems are usually like that in my experience. Networks play by their own rules and some of them have not been shared with humankind.
 
Hi Ruiner
Try running CPUZ or HWMonitor and watching the Vcore while stress testing the CPU using prime95 or OCCT or Aida64 and observe what the Vcore settles in at under heavy load. I'd say you did well with your cpus at 3.9ghz on stock voltage. I had to raise the vcore a fair bit to post at 3.8 or 3.9. So far i've gotten my own Ryzen 3 1200 at 4.0ghz at 1.4v stable in stress tests and general usage
 
I tried updating to Asrock's 4.50 bios (posted for the new APUs) and it gave me an odd bug: rubberbanding lag in multiplayer games (PUBG starting island for instance) but only after a wake from sleep. It doesn't happen on a fresh restart. Changing cpu clocks back to default (but ram still at 3200) kills the bug, as does reverting to bios v3.30 at 38x cpu clocks.
No sign of Bclock changes in cpuZ on wake from sleep.
 
AMD provided last AGESA too late and motherboard vendors had issues to release BIOS on time ( for the APU premiere which was yesterday ). As a result, it's not guaranteed that everything will work like it should, even though brands like ASRock care to deliver fully stable products.
 
You must log in or register to reply here.
Back