F5F Stay Refreshed Hardware Desktop CPU-Z displays clock speeds varying from BIOS configurations

CPU-Z displays clock speeds varying from BIOS configurations

CPU-Z displays clock speeds varying from BIOS configurations

R
62
07-31-2016, 05:51 AM
#1
I'm experimenting with a few outdated Crucial modules, a 2666 16G and an 2133 4G, both running somewhat shaky default settings like cl20 and cl16. No XMP profiles are set. They function adequately at 3200MHz, but I’m unsure if the BIOS numbers (15-15-15-35) match what CPU-Z shows (22-22-22-53). Which source should I rely on? I think CPU-Z gives a more accurate picture, but I still need verification.
R
Rainbow_Spikes
07-31-2016, 05:51 AM #1

I'm experimenting with a few outdated Crucial modules, a 2666 16G and an 2133 4G, both running somewhat shaky default settings like cl20 and cl16. No XMP profiles are set. They function adequately at 3200MHz, but I’m unsure if the BIOS numbers (15-15-15-35) match what CPU-Z shows (22-22-22-53). Which source should I rely on? I think CPU-Z gives a more accurate picture, but I still need verification.

_
_eptastic_
Member
54
07-31-2016, 06:57 AM
#2
Set cl to 12 in bios and you’ll immediately see if the system is working properly or if cpuz isn’t posting. You can also use thaiphoon burner to review their iics so you don’t overclock blindly. If those boards are bare pcb, just inspect the iics directly—especially with the micron fbga decoder—to confirm what they are. Most microns (8gbit rev e, 16gbit rev b) should easily clear the 4400 threshold. Spoiler: those boards should definitely outpace my DDR3 gdie at 3400.
_
_eptastic_
07-31-2016, 06:57 AM #2

Set cl to 12 in bios and you’ll immediately see if the system is working properly or if cpuz isn’t posting. You can also use thaiphoon burner to review their iics so you don’t overclock blindly. If those boards are bare pcb, just inspect the iics directly—especially with the micron fbga decoder—to confirm what they are. Most microns (8gbit rev e, 16gbit rev b) should easily clear the 4400 threshold. Spoiler: those boards should definitely outpace my DDR3 gdie at 3400.

B
Brandonboy97
Junior Member
3
07-31-2016, 08:26 AM
#3
Sure, happy to help. Thanks for the advice!
B
Brandonboy97
07-31-2016, 08:26 AM #3

Sure, happy to help. Thanks for the advice!

S
saver206
Junior Member
15
07-31-2016, 10:15 AM
#4
What kind of motherboard is this? If it's a Gigabyte model, the BIOS seems to be showing incorrect timing data—only displaying JEDEC settings when everything is auto-configured. The reading of 22-22-22-53 at 3200MT/s makes more sense if you haven't adjusted the timings yourself.
S
saver206
07-31-2016, 10:15 AM #4

What kind of motherboard is this? If it's a Gigabyte model, the BIOS seems to be showing incorrect timing data—only displaying JEDEC settings when everything is auto-configured. The reading of 22-22-22-53 at 3200MT/s makes more sense if you haven't adjusted the timings yourself.

C
C4rol
Junior Member
10
07-31-2016, 03:51 PM
#5
Not too confident about that, huh? The system crashed during Cinebench multi at 3200 MHz. The 4GB 2133 DIMM uses Micron B-dies (D9VHP), which should perform well compared to someone who used 5400 cl18. Regarding the 16GB 2666 DIMM... Micron H-dies (C9BHC) have limited details available; one report mentioned they can handle stable 2933. I plan to test the 4GB DIMM separately in single-channel mode to see if it delivers any value. By the way, the CPU is a 5700X3D.
C
C4rol
07-31-2016, 03:51 PM #5

Not too confident about that, huh? The system crashed during Cinebench multi at 3200 MHz. The 4GB 2133 DIMM uses Micron B-dies (D9VHP), which should perform well compared to someone who used 5400 cl18. Regarding the 16GB 2666 DIMM... Micron H-dies (C9BHC) have limited details available; one report mentioned they can handle stable 2933. I plan to test the 4GB DIMM separately in single-channel mode to see if it delivers any value. By the way, the CPU is a 5700X3D.

M
mariokisi
Junior Member
45
08-02-2016, 10:35 AM
#6
Perfect, it's a solid AX370 Gaming 5 Well; I should have been more careful when choosing the board.
M
mariokisi
08-02-2016, 10:35 AM #6

Perfect, it's a solid AX370 Gaming 5 Well; I should have been more careful when choosing the board.

A
Aerithix
Member
182
08-04-2016, 11:19 AM
#7
It seems only the 16Gb Rev.B model is top-notch; the 4GB version falls far short. For X370 boards, there weren’t many significant improvements, which is just part of the quirks you adapt to with Gigabyte boards.
A
Aerithix
08-04-2016, 11:19 AM #7

It seems only the 16Gb Rev.B model is top-notch; the 4GB version falls far short. For X370 boards, there weren’t many significant improvements, which is just part of the quirks you adapt to with Gigabyte boards.

J
jdclay
Member
154
08-04-2016, 11:39 AM
#8
The 4Gbit rev B version seems questionable, I’m not sure about the rev H setting. I’ll start with the 8Gbit rev B as a baseline. Timings can be adjusted later, especially when testing at higher frequencies. If you plan to push voltage, a fan might be necessary—generally over 1.5 to 1.6 volts is needed for stress tests. Depending on the voltage, a fan could be required to prevent overheating. If you run it at 2.2 to 2.4 volts, a fan may be useful. For the secondaries, they usually don’t need manual tuning unless there’s an issue, but sometimes they do. Overclocking can happen unexpectedly with certain chips; I once reached 3400 after thinking it was bad. It turned out the secondaries needed adjustment. I pulled some secondary settings from a forum thread and they worked well. If your RAM lacks OC data or documentation, you can look at similar models—like the Samsung 1Gbit or 2Gbit variants. I managed to get it running at 2200 on a P6T Deluxe V2 with 2.46V, but only at 2600 under 2.6V. I’m planning more tests once I have the right tools and hardware to push beyond 3V safely.
J
jdclay
08-04-2016, 11:39 AM #8

The 4Gbit rev B version seems questionable, I’m not sure about the rev H setting. I’ll start with the 8Gbit rev B as a baseline. Timings can be adjusted later, especially when testing at higher frequencies. If you plan to push voltage, a fan might be necessary—generally over 1.5 to 1.6 volts is needed for stress tests. Depending on the voltage, a fan could be required to prevent overheating. If you run it at 2.2 to 2.4 volts, a fan may be useful. For the secondaries, they usually don’t need manual tuning unless there’s an issue, but sometimes they do. Overclocking can happen unexpectedly with certain chips; I once reached 3400 after thinking it was bad. It turned out the secondaries needed adjustment. I pulled some secondary settings from a forum thread and they worked well. If your RAM lacks OC data or documentation, you can look at similar models—like the Samsung 1Gbit or 2Gbit variants. I managed to get it running at 2200 on a P6T Deluxe V2 with 2.46V, but only at 2600 under 2.6V. I’m planning more tests once I have the right tools and hardware to push beyond 3V safely.