Question Pegged low core speed with i5 4690S

[mrbrunes]

Just acquired this CPU but the core speed as reported by CPU-Z remains stuck at 1200MHz (clock should go up to 3200MHz). The multiplier varies though (not that I've much idea about that).

Mobo is an ASUS H81M-PLUS with all cpu settings at auto that I can see, with the latest BIOS. "System Performance" is set to ASUS Optimal which according to their graphic seems to be the Performance-oriented option.
4690S is on the compatible CPU list.

I ran Passmark and the cpu gets a low score compared to others. Then I ran Prime95 and the cpu pegs at 20% util. Never seen that before!

Any ideas?

 

[Dark Lord of Tech]

Used?

 

[mrbrunes]

Yup.

 

[NightHawkRMX]

You could try clearing your CMOS just to double check a bios setting isn't causing any issues. I doubt it tho since it is an H81 mobo tho.
Intel says:
"Steps to clear CMOS using the jumper method

1. Power off the system.
2. Open up the system so you can see the motherboard.
3. Locate the CMOS jumper by referring to the motherboard manual and how to clear the CMOS. In general, the CMOS jumper is three pins located near the battery.
4. In general, CMOS jumper has positions 1–2 and 2–3. Move the jumper from the default position 1–2 to position 2–3 to clear CMOS. Wait 1–5 minutes then move it back to the default position.
5. Power on the system. In some systems, you might need to enter BIOS to reset to the factory defaults.

Steps to clear CMOS using the battery method

1. Turn off all peripheral devices connected to the computer.
2. Disconnect the power cord from the AC power source.
3. Remove the computer cover.
4. Find the battery on the board. The battery may be in a horizontal or vertical battery holder, or connected to an onboard header with a wire.
5. Remove the battery:
   1. If the battery is in a holder, note the orientation of the + and – on the battery. Gently pry the battery free from its connector.
   2. If the battery is connected to an onboard header with a wire, disconnect the wire from the onboard header.
6. Wait 1–5 minutes, then reconnect the battery.
7. Put the computer cover back on.
8. Plug the computer and all devices back in."

https://www.intel.com/content/www/us/en/support/articles/000025368/processors.html

 

[mrbrunes]

Tx. I cleared CMOS and the cpu got up to 2GHz core speed and the cpu peaked at about 80% util but it wasn't sustained. But after another reboot it remains stuck at 1200MHz,and I can't get cpu util to go above 30%, even with Prime95.

Is this normal with the low power Haswells? I've not seen any cpu behave like this before i.e. it won't go to 100% util and the core speed remains stuck, irrespective of load.

 

[NightHawkRMX]

No its not normal. What are your cpu temps at? Make sure its not throttling.

 

[mrbrunes]

About 70 Deg. Ran Intel CPU diags which passed tests ok.

 

[NightHawkRMX]

70c is fine.
Might be worth checking to make sure no socket pins are bent. Reseating the cpu in this process may help aswell.

 

[mrbrunes]

I ran some more tests and discovered that SpeedFan was reporting 100% cpu utilisation when Task Manager was only showing 30%.

I then tweaked some more BIOS options including disabling any power saving features and the base speed improved a little, but still showed poor benchmark speeds (below a 1/3 of expected).
Then I tried reseating the cpu, however on power on the fan only spun briefly. After removing the chip I found part of a pin stuck to the chip. :-(
I'm pretty careful with the handling so I don't know if I did something to break it, it's the first time anything like this has happened to me. :-(

With not much in the way of expectation, I put the old chip in (i3-4370T) and this booted with no issues! So at least my son is back up and operational again - phew!
Core speeds are fine and vary up and down according to CPU-Z - however Task Mgr still has an anomaly in that it initially peaked at 100% wth Prime95 running but then tailed off, leaving SpeedFan showing 100% on all cores.

So it looks like I'll pick up another mobo and try the 4690S again. However there's obvs some issue with Task Mgr and these lower power chips. Does it know more than SpeedFan somehow and is reflecting the power saving freq reduction under load ?