Fiery

AIDA64 v4.30 follows the sensor registers layout specification provided by MSI for their LGA1150 boards. In case it doesn't match MSI's own software, then either their software works incorrectly, or they failed to provide accurate information with us BTW, does the CPU temperature seen in MSI's software happen to match any of the CPU-related temperature readings you can see on the Sensor page in AIDA64? Like CPU Package, or any of the CPU cores?

What version of AIDA64 are you using, and which MSI motherboard do you have? Thanks, Fiery

Ah, yes, my bad, it's an SMBus based sensor Then please right-click on the bottom status bar of AIDA64 main window --> Sensor Debug --> SMBus Dump (Full). Copy-paste the full results here. Thanks, Fiery

Thank you, but I'm afraid that dump didn't include any information for your onboard sensor chip. Are you sure everything was okay with AIDA64 when you made the dump? Or maybe you've stripped some parts of the dump output?

We'll remove the GPU TDP percentage decimal digits in the next AIDA64 beta release due next week. Regards, Fiery

Please right-click on the bottom status bar of AIDA64 main window --> Sensor Debug --> ISA Sensor Dump. Copy-paste the full results into this topic. Thanks, Fiery

Engineering Sample is not meant to be localized. Regards, Fiery

The latest AIDA64 update introduces a multi-threaded OpenCL GPGPU Hash benchmark to measure SHA-1 hashing performance of modern graphics processors and APUs. It also implements support for Microsoft Windows 8.1 Update 1 and Windows Server 2012 R2 Update 1, and supports the latest graphics and GPGPU computing technologies. New features & improvements Microsoft Windows 8.1 Update 1 and Windows Server 2012 R2 Update 1 support OpenCL GPGPU SHA-1 hash benchmark CUDA 6.0 support Socket AM1 motherboards support Improved support for Intel “Broadwell” CPU Preliminary support for AMD “Carrizo” and “Toronto” APUs Preliminary support for Intel “Skylake”, “Cherry Trail”, “Denverton” CPUs Crucial M550 and Intel 730 SSD support GPU details for AMD Radeon R7 265 GPU details for nVIDIA GeForce GTX 745, GeForce 800 Series What's new since AIDA64 v4.00 OpenCL GPGPU Benchmark Suite AMD Mantle graphics accelerator diagnostics Multi-threaded memory stress test with SSE, SSE2, AVX, AVX2, FMA, BMI and BMI2 acceleration Optimized 64-bit benchmarks for AMD “Kaveri”, “Bald Eagle”, “Mullins”, “Beema” APUs Optimized 64-bit benchmarks for Intel Atom C2000 “Avoton” and “Rangeley” SoC Optimized 64-bit benchmarks for Intel “Bay Trail” desktop, mobile and tablet SoC Full support for the upcoming Intel “Haswell Refresh” platform with Intel “Wildcat Point” PCH Razer SwitchBlade LCD support Preliminary support for Intel Quark X1000 “Clanton” SoC Improved support for OpenCL 2.0 Support for VirtualBox v4.3 and VMware Workstation v10 OCZ Vector 150, OCZ Vertex 460, Samsung XP941 SSD support GPU details for AMD Radeon R5, R7, R9 Series GPU details for nVIDIA GeForce 700 Series [ Press Release (English) ] [ Press Release (Deutsch) ] [ Press Release (italiano) ] [ Press Release (magyar) ] [ What's new in AIDA64 v4.30 ] [ Download ]

1) The lack of BCLK and SPD information is because AIDA64 is unable to use its kernel driver on your system. It's either because AIDA64 is not properly installed on your system, or because your Windows user has no administrator privileges. 2) The diode temperatures well under ambient (room) temperature is a common issue of modern AMD processors. AMD says the diode temperatures in such cases are only meaningful when the CPU is under load. At idle you should simply ignore the diode temperature readings. Regards, Fiery

AFAIK Mantle currently has no interface to measure FPS or draw an image layer via hooks (what RTSS does). AMD may implement those later this year, but I suppose such features aren't their top priority right now.

The difference is that AIDA64 v2.80 didn't support AMD Mullins/Beema APUs at all. It also didn't support the SMBus protocol of the integrated FCH of those processors, so SMBus readouts couldn't cause a lockup there

I'm afraid only MSI can tell you the truth and specifics about the Aux temperature reading. All I can tell is that it can be a bogus reading, and invalid reading, or a meaningful reading. If the latter, then it can be for the CPU, iGPU or some motherboard component. Try to ask MSI Support, maybe they will be able to provide you with the information you're looking for.

Currently AIDA64 can only monitor the video memory usage of the primary (first) nVIDIA video adapter. We'll make sure to improve on that soon Regards, Fiery

Please refer to question #18 at: http://www.aida64.com/support/knowledge-base In case the two CPU core temperatures remain normal while you're gaming, then you can simply ignore the Aux value, and focus on the Motherboard, CPU and CPU core temperatures only. Regards, Fiery

Thank you for your help. I've just sent you a private message about this.

Thank you for the info. Then maybe some other SMBus addresses cause the issue, but it's not easy to debug that remotely. Would it be okay with you if we made a special debug build of AIDA64 that would let us find out exactly at which SMBus address does the Discovery board crash?

I'm not sure how could replacing the PSU make the onboard sensor chip to report incorrect values. Quite frankly, I've never heard of such issue. The sensor chip used on your motherboard is Nuvoton NCT6776F. The base voltage (voltage resolution) of the sensor chip is 0.008V, but AFAIK that has nothing to do with the temperature measurement module of the sensor chip.

Thank you guys for the info. The BitDefender bit is quite interesting. What version of BitDefender are you using? Is it the BitDefender Antivirus, Internet Security or Total Security variant? BTW, since AIDA64 doesn't poll anti-virus software or virus database constantly, the only thing I can think of is that maybe the firewall module of BitDefender somehow collides with either the network traffic measurement layer of AIDA64, or with the external IP address detection part of AIDA64. Do you have any of those items enabled on your SensorPanel, Logitech LCD or OSD Panel?

Your motherboard sensor chip (Nuvoton NCT6776F) seems to have 2 temperature inputs utilized, labelled as Motherboard and CPU by AIDA64 on its Sensor page. Both temperature inputs indicate a very high temperature reading which I assume would be incorrect. That however is not the fault of AIDA64 or any software, but an issue with the sensor chip that is integrated on your motherboard. What you can do is try updating the BIOS, but I don't think it would help. You can also ask ASRock Support about this, maybe they know of a workaround or resolution. If none of those work out, then you can simply just ignore those 2 temperature readings and rely on the processor's own temperature diode readings which seem to be a lot more accurate in your system. Regards, Fiery

Exactly under what circumstances did you get those messages? Did you connect via Remote Control? If yes, then what did you do when you had those error messages come up? Do you use AIDA64 v4.20? (latest version)

Here's the new AIDA64 Extreme beta update: http://www.aida64.com/downloads/aida64extremebuild2833czvd4jfw8yzip Let me know if it works better than the previous one

It is not related to the fact that you're using a trial version of AIDA64. It only means your computer is not 100% stable. You need to tweak your overclock settings, or return back to factory settings to make your system stable again. Regards, Fiery

The next beta with the mentioned fix will be rolled out later this week. We're targeting today or tomorrow right now, but it's not 100% yet

Those options, except for #2, are not really relevant in your case. Your issue comes from the fact that the Discovery motherboard raises issues when an application scans for SMBus devices in the device range of 0x28..0x2F. In the next AIDA64 beta we'll disable the rest of the SMBus-based sensor device scans, and hopefully that will enable you to re-enable the SMBus stability option.

Thank you for the feedback