R7 1700 OC temps are strange
R7 1700 OC temps are strange
I'm trying to grasp this situation.
My stock clock is showing around 34°C after a short while, scans are complete, etc., and it jumps into the 40s.
After an OC at 3.8 OC with a voltage bump to 1.325V, temperatures drop to 32-33°C and it stays close without much activity.
Curiously, on the stock curves, the dynamic range varies from about 1.5 up to 3.10. When I change the OC, the range expands to around 1.4 all the way to 4.2.
I don't understand how it cools down and runs at a lower frequency after the OC. Nothing else has been adjusted—no changes to fan profiles. My CPU fan doesn't seem to cool more when I increase it; instead, temps rise if I lower it, so I left it unchanged. The case fans are always running at full speed.
Can you execute an HWMon instance and display screenshots of both idle and OC outputs? Please use a format like imgur to share them so we can examine the exact details. Also, kindly provide your PC specifications in full detail.
Also, please note the following system specs:
I'm referring to what the core temp is displaying on idle stock and OC. What are you trying to confirm that I didn't misunderstand? The listed high temperatures come from CPU-Z benchmark or stress tests. It's interesting that the tool used for UserBenchmarks doesn't match up closely, yet still provides varied results. My current high readings aren't a worry anymore. I can push the 3.8 setting all day and it stays around the 70s, but at 3.9 it quickly rises to higher values. I turned it off when it reached 84°C.
I received three parameters from many that HWMon/Info provides. As mentioned earlier, we can now examine the situation in greater detail. Prime 95 small FFTs are recommended for stress testing; CPUZ offers a stress option but doesn’t push your CPU beyond stable limits or high temperatures. It should maintain similar temperatures to a heavy gaming workload. Try running your 3.8 OC with Prime95 and observe the temperatures—expect values above 80 if you’ve only tested with CPUZ. Regarding the temperature reporting issue, have you updated your BIOS and motherboard drivers to the latest versions? Finally, please list your PC specifications so we can analyze your hardware setup.
^ Giggle, rarely used most of that, so I’ll check some later.
I’m searching for a balance between performance and cooling, and it’s odd the system runs cooler than stock clock. I often run into problems when gaming and overloading my AC, so I’ve been tweaking Afterburner curves and adjusting clock speeds to discover a comfortable setting. Mostly, it’s Ashes of the Singularity—great for testing a system.
Regarding the build:
R7 1700
2x8 DDR4 2400
ASRock fatality AB350 gaming ITX AC
GTX 1080
600W PSU
Running 1080p/60
I know the graphics card is a bit too powerful, but the price was worth it for what I needed.
Additionally, the most recent two version bios appear more appealing and stable compared to earlier versions, according to ASUS.
Nice system. Yes, Ashes can really test some systems, especially when it comes to OC attempts. Are you adjusting settings in the BIOS or using Ryzen Master? It looks like the performance might not be stable after applying the changes. Unless you're using a very aggressive fan curve, raising the vcore and clockspeed shouldn't lower temperatures. This could point to a BIOS glitch (possibly involving CMOS reset) or an issue with the monitoring software.
I received updates to the "3rd" bios (4.5) as noted in the comments, aiming to implement changes first and avoiding stability problems unless upgrading to 2xxx.
I’m utilizing the BIOS for overclocking. It seems Ryzen Master displayed odd readings with certain voltages, configurations, and temperatures—more pronounced than what I observe now.
I reviewed my voltage data and found that at stock clock with auto settings 1.324, the highest voltage appeared, so I set it as a baseline for a more aggressive clock speed. After the update to 4.5 (bios), performance improved significantly.
I experimented with different settings and voltages, but values around 1.325–1.350 provided the best stability and cooling at 3.8, though the latter tends to get hotter than I prefer at 3.9.
It was quite surprising when Ashes made me realize how much the system could change my approach. When I first started, I could handle crazy settings and run smoothly—though more enemies and bigger maps eventually caused crashes with certain configurations.
GPU issues only appear if I enable Extreme and disable multi-sampling with DX12.
Yeah, on the stock auto (I have a 1600x) I noticed a big jump, especially when it reached 4GHz, and the voltage was in the 1.4 and higher range. Of course, this was just a brief moment, but it still caught me off guard.
Yes, Ryzen Master tends to push performance, particularly when the Vcore isn't changed. Auto increases the Vcore.
When I first started overclocking this Ryzen, I began with my VID from Coretemp (1.375). I estimated that starting with a Vcore of 1.3 would be best for testing. After some trial runs at 95 MHz with small stress, I reached 3.9GHz consistently. When I hit a failure at Prime, I reduced the frequency by 100MHz and tested overnight. That’s my stable setting. Temperatures stayed around 77°C max at 3.9GHz. To reach 4GHz in my Asus motherboard, I had to raise the Vcore from 1.3 to 1.35! That was a huge jump for the extra 100MHz. So right now, 3.9 is my consistent OC. 24/7.
Are you doing the same? At 1080p/60Hz, you really don’t need to overclock if you’re not. It will consume all your power at that setting and then more. Make sure your CPU runs smoothly with low voltage, stable temps, optimized RAM, and with the GTX1080 you’ll be fine for years at that frequency.
By the way, 3.8 is a solid OC for the 1700. If you get it stable, you’ll have a powerful CPU running.
Also, didn’t mention your cooler? If it’s stock, consider upgrading to a third-party solution. With the right cooling, you could push 4GHz with the correct unit.