NovaProspekt
New Member
- Joined
- Nov 9, 2017
- Messages
- 20 (0.01/day)
Here is my system:
ASRock Z77 Extreme4 motherboard running firmware 2.90p (adds NVMe boot capability)
Core i5 3570k with Hyper212 Evo cooler
AMD Vega 56 (Gigabyte reference model)
16 gigs Mushkin Blackline DDR3 1600 (4x4) running at stock XMP settings
ADATA XPG SX8200 Pro 1TB NVMe drive
850 watt Seasonic Focus Plus Gold PSU
Windows 10 64 bit
I have been trying to get the 3570k running stable at 4.4 ghz for quite a while now. The two games I am primarily playing right now are Destiny 2 and The Division 2, both of which experience random sporadic crashes to desktop. Sometimes there is no crash for hours to days, sometimes they crash in 5 minutes. I had flashed my Vega 56 to Vega 64 BIOS, then due to instability flashed back to the stock Vega 56 BIOS and just undervolted, now I have the card just running at the stock "Turbo" setting to eliminate that as a source of the instability.
I have Speedstep enabled in UEFI so that the CPU can drop down to 1.6 Ghz and <1 volt at idle, and I am using the additional turbo voltage feature to supply extra voltage under load. My methodology is every time I get a crash, I increase the turbo boost voltage by one step. I have done this 12-15 times, and my VCORE under a full gaming load is now approaching 1.35 volts which seems to be the point most people recommend not exceeding for Ivy Bridge. My temps while gaming are topping out at 72 degrees Celsius, but I am still experiencing crashes. The crashes are always exactly the same. In Destiny 2, the game suddenly locks up and stays on screen for a few seconds before closing, in the Division 2 the game suddenly shrinks to windowed mode and then closes.
The last time I tried a Prime95 stability test at 4.4 ghz temps shot up to 95 degrees Celsius pretty quickly (and that was before I started increasing voltage), so I can't run an extended Prime95 session. Maybe this is a sign that my thermal paste has gone bad (MX4 which I applied in 2012 when I initially built the PC, I have not had the heatsink off since then). But I figure 72 degrees C while gaming is acceptable, so I don't know about the thermal paste. I would really like to not replace it because I remember the heatsink being a pain to put on.
I guess my question is, at this point should I just accept that I'm not going to hit 4.4 ghz stability and start decreasing frequency? I see lots of posts with people claiming 4.5 ghz overclocks with 1.25v on this chip. Maybe it's just starting to die on me.
ASRock Z77 Extreme4 motherboard running firmware 2.90p (adds NVMe boot capability)
Core i5 3570k with Hyper212 Evo cooler
AMD Vega 56 (Gigabyte reference model)
16 gigs Mushkin Blackline DDR3 1600 (4x4) running at stock XMP settings
ADATA XPG SX8200 Pro 1TB NVMe drive
850 watt Seasonic Focus Plus Gold PSU
Windows 10 64 bit
I have been trying to get the 3570k running stable at 4.4 ghz for quite a while now. The two games I am primarily playing right now are Destiny 2 and The Division 2, both of which experience random sporadic crashes to desktop. Sometimes there is no crash for hours to days, sometimes they crash in 5 minutes. I had flashed my Vega 56 to Vega 64 BIOS, then due to instability flashed back to the stock Vega 56 BIOS and just undervolted, now I have the card just running at the stock "Turbo" setting to eliminate that as a source of the instability.
I have Speedstep enabled in UEFI so that the CPU can drop down to 1.6 Ghz and <1 volt at idle, and I am using the additional turbo voltage feature to supply extra voltage under load. My methodology is every time I get a crash, I increase the turbo boost voltage by one step. I have done this 12-15 times, and my VCORE under a full gaming load is now approaching 1.35 volts which seems to be the point most people recommend not exceeding for Ivy Bridge. My temps while gaming are topping out at 72 degrees Celsius, but I am still experiencing crashes. The crashes are always exactly the same. In Destiny 2, the game suddenly locks up and stays on screen for a few seconds before closing, in the Division 2 the game suddenly shrinks to windowed mode and then closes.
The last time I tried a Prime95 stability test at 4.4 ghz temps shot up to 95 degrees Celsius pretty quickly (and that was before I started increasing voltage), so I can't run an extended Prime95 session. Maybe this is a sign that my thermal paste has gone bad (MX4 which I applied in 2012 when I initially built the PC, I have not had the heatsink off since then). But I figure 72 degrees C while gaming is acceptable, so I don't know about the thermal paste. I would really like to not replace it because I remember the heatsink being a pain to put on.
I guess my question is, at this point should I just accept that I'm not going to hit 4.4 ghz stability and start decreasing frequency? I see lots of posts with people claiming 4.5 ghz overclocks with 1.25v on this chip. Maybe it's just starting to die on me.