Fiery

Thank you for the dump. We've verified the registers content, and also the relevant part of AIDA64, and everything looks fine. The only thing that may cause such issue would be the proprietary fan muxing feature that ASRock uses on your motherboard (and a number of other motherboards as well). It effectively muxes the RPM measurement signals of the CPU fan1 and the CPU fan2 headers into the same fan register of the Nuvoton sensor chip. When another software (for example ASRock's own monitoring software called Extreme Tuner) alters the GPIO setting for the fan mux, then the sensor chip will reflect the fan RPM for the CPU fan2 header. I'm afraid we don't have a solution for such issues ASRock should simply stop using that weird GPIO fan muxing on their boards.

1) On mobile systems you may need to alter the default setting of power management features (like SpeedStep) to make sure your computer will be running at HFM (High Frequency Mode) under heavy load. With certain settings that some manufacturers call "Battery saving mode" the computer will always work at LFM (Low Frequency Mode), no matter if the computer is running from AC or DC power. You have to make sure to check the current power management settings, although in most cases to get a mobile computer to work at HFM is easy: just connect it to AC power 2) The FPU stress test usually heats up all system components in 15-20 minutes. 3) AIDA64 System Stability Test is a lot more sophisticated and is better optimized to the latest processors than what EVEREST used to offer -- that's why with AIDA64 it's normal to reach higher temperatures due to the heavier workload it puts on the system. EVEREST is an obsolete and discontinued product that e.g. doesn't support any processors released in 2011 (e.g. Sandy Bridge, Bulldozer, Llano). 4) The FPU stress test uses a heavy workload, based on floating-point calculations, to fill up all the floating-point processing units of the processor with calculation threads. It stresses all available CPU cores and CPU threads, and it utilizes every relevant modern x86 extensions such as SSE, FMA4 and AVX. Modern processors can be heated up to the highest level by running exactly the type of worload that AIDA64 uses in the FPU stress test. 5) Most mobile computers cannot measure their own fan speed, but only able to control the fan speed based on the measured CPU temperature. Unfortunately most mobile computers are not equipped with a sophisticated sensor chip (due to size constraints on the motherboard PCB). 6) I42IL1 is most likely produced by Elitegroup, which manufactures a lot of different motherboard and notebook brands, e.g. ECS, PCChips, Uniwill, etc., and their products are rebranded by an even bigger number of computer assembly companies. Elitegroup a huge company, but since we're not sure which brand the notebook product called "I42IL1" belongs to, we show what DMI reflects. And DMI holds the manufacturer name as "OEM" on your system. Regards, Fiery

Mobile computers implement serious constraints on case space and fan sizes to make sure they are portable. Hence you cannot expect the same cooling performance from a mobile computer than from a desktop computer. Generally speaking, hard disk drives should be the same in mobile and desktop computers, but the mobile motherboards and mobile processors can stand a bit more than their desktop counterparts. So as long as your CPU isn't overheating, you should be fine. You can check CPU overheating in the AIDA64 System Stability Test by enabling only the FPU test, and pressing the Start button. Watch the bottom graph where Throttling should never show any non-zero percentage activity. When your CPU is throttling, it means it is overheating. A properly designed computer should be able to run such a heavy workload like AIDA64 FPU stress test without the processor overheating. Regards, Fiery

Temperatures in AIDA64 may be missing and then re-appear because the temperature readout is unstable due to sensor device issues. GMCH sensors of Intel chipsets are notorious of providing unstable temperature readings When AIDA64 reads an invalid temperature (e.g. 200 Celsius), it removes it from the Sensor page, to avoid displaying bogus/invalid/meaningless values. While HWiNFO doesn't seem to have such tricks, and so it displays bogus values too.

Yes, it utilizes the general data crunching capabilities of modern GPUs rather than their ability to render high-resolution 3D images. When you "just" process a huge amount of data with the GPU, the whole video card is stressed in a different way than when running e.g. a 3D game. We picked the GPGPU way to let AIDA64 users stress their GPUs without the need to switch to full-screen mode.

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. Make sure to use the latest AIDA64 Extreme Edition beta release: http://www.aida64.com/downloads/aida64extremebuild1714gcn7wyrhdzzip After upgrading to this new version, make sure to restart Windows to finalize the upgrade. Thanks, Fiery

Thank you. Due to the fundamental difference between a full-screen 3D app and a GPGPU application (AIDA64 SST), the GPU won't be loaded or stressed the same way with those apps. On our test systems it took 15-30 minutes to make the video adapter heat up with GPGPU to the same levels that a full-screen 3D app could reach in a few minutes. It's definitely a slight issue, but since we generally recommend our users to run AIDA64 System Stability Test for at least a couple of hours, I don't think this is a show-stopper issue.

1) You can activate the DIMM temperature measurement in AIDA64 / main menu / File / Preferences / Stability. After altering that option you need to restart AIDA64. 2) As for the HP ENVY EC access, I'm not entirely sure if it's relevant on your system. Based on the measured values in HWiNFO, the values seem to show a huge fluctuation. And that usually means the readings are invalid (bogus). Regards, Fiery

Thank you. Have you tried to check it with GPU + FPU test both enabled? I wonder how would that affect the overall system temperature and the GPU temperatures too.

DDR3-1600 is an out of spec (overclocked) memory grade on most chipsets, and our reference results database do not feature any overclocked results due to our internal policy. We may include a DDR3-1600 or DDR3-1866 system featuring an AMD FX-8150 processor -- when that CPU becomes widely available. As for Z68, it's not much different to the P67 which we already have in our reference database. We may add a Z68 system to the database, featuring a Celeron G440 CPU.

Thank you for the feedback

They aren't outdated at all. What are your concerns about the reference results?

Thank you for the feedback Regards, Fiery

Thank you for the feedback Regards, Fiery

Maybe the fact that you have multiple monitors makes a difference... As for the 5-digit bar limits issue and the bar frame colour adjustment bug, we've just rolled out a new AIDA64 beta that should solve both of them: http://www.aida64.co...14gcn7wyrhdzzip After upgrading to this new version, make sure to restart Windows to finalize the upgrade.

Thank you for the feedback

Since it is a bogus reading, we've removed the Aux temperature on your motherboard in the latest beta version of AIDA64 Extreme Edition available at: http://www.aida64.co...14gcn7wyrhdzzip After upgrading to this new version, make sure to restart Windows to finalize the upgrade. Regards, Fiery

As a first step, we've added Battery Voltage to the Computer / Sensor page of AIDA64, also to the hardware monitoring module (OSD Panel, Sidebar Gadget, SensorPanel, etc). Please upgrade to the latest beta version of AIDA64 Extreme Edition available at: http://www.aida64.co...14gcn7wyrhdzzip After upgrading to this new version, make sure to restart Windows to finalize the upgrade. Let me know how it works. We'll also work on the UPS power readout, which apparently is specific to APC UPSes, hence it cannot be done via the standard Windows Smart Battery API...

Thank you for the dump. We've fixed the DIMM TS sensor support in the latest beta version of AIDA64 Extreme Edition available at: http://www.aida64.com/downloads/aida64extremebuild1714gcn7wyrhdzzip After upgrading to this new version, make sure to restart Windows to finalize the upgrade. Let me know how it works. Thanks, Fiery

We've fixed voltage readings on your motherboard in the latest beta version of AIDA64 Extreme Edition available at: http://www.aida64.com/downloads/aida64extremebuild1714gcn7wyrhdzzip After upgrading to this new version, make sure to restart Windows to finalize the upgrade. Let me know how it works. Thanks, Fiery

We've fixed pipeline count detection for Radeon HD 5000 and 6000 family of GPUs in the latest beta version of AIDA64 Extreme Edition available at: http://www.aida64.com/downloads/aida64extremebuild1714gcn7wyrhdzzip After upgrading to this new version, make sure to restart Windows to finalize the upgrade. Let me know how it works. Thanks, Fiery

We've further improved the hot-plug devices support in the latest beta version of AIDA64 Extreme Edition available at: http://www.aida64.com/downloads/aida64extremebuild1714gcn7wyrhdzzip After upgrading to this new version, make sure to restart Windows to finalize the upgrade. Please let me know how it works. Thanks, Fiery

We've added the supported and active DMA modes information to the Storage / ATA page in the latest beta version of AIDA64 Extreme Edition available at: http://www.aida64.com/downloads/aida64extremebuild1714gcn7wyrhdzzip After upgrading to this new version, make sure to restart Windows to finalize the upgrade. Regards, Fiery

We've just made the new GPU stress testing feature public in the following AIDA64 beta release: http://www.aida64.co...14gcn7wyrhdzzip It can be used to put stress on one or more GPUs in the system, by utilizing the GPGPU computation capabilities of OpenCL API. A couple of things to note about this new feature: 1) First of all, it is in a beta stage right now. We'd like to hear your feedbacks about it, and collect any improvement ideas. We will only roll out this new feature in an AIDA64 stable release in the first half of 2012. This feature will be disabled in any upcoming new stable releases of AIDA64 that will be rolled out in 2011. 2) Even this first public beta is capable of putting computation threads on multiple GPUs; supports CrossFireX, SLI and AMD DualGraphics configurations; supports mixed GPU configurations (one ore more nVIDIA GPUs + one or more AMD GPUs) under Windows 7; and of course it supports AMD APUs (Llano, Brazos, Trinity) as well. Putting stress simultaneously on an APU and one ore more discrete GPUs is also supported. OpenCL-capable IGP chips and mobile GPUs are supported. Multi-GPU video cards (e.g. GeForce GTX 590 and Radeon HD 6990) are supported. 3) Any OpenCL devices can be stressed, but right now it's not possible to select which devices to stress. Hence all GPU-type OpenCL devices that can be found in the system will be stressed simultaneously. 4) CPU-type OpenCL devices are not supported. We don't plan to support them, since for CPU stressing we have our own solution, the usual 64-bit System Stability Test module. 5) You can check the list of available OpenCL-capable devices on the Display / GPGPU page in AIDA64. All GPU-type devices listed there will be stressed by the AIDA64 System Stability Test. 6) Due to the way GPGPU implementations work, stressing the GPU also puts stress on other system components, e.g. system memory, CPU caches, CPU cores, and PCI Express links. Hence it's normal to have considerable CPU utilization while running the GPU stress test, even when CPU and FPU tasks are not enabled on the System Stability Test interface. 7) Due to OpenCL not being capable of providing a device utilization measurement, AIDA64 System Stability Test is not able to show the GPU load% while the stress test is running. It is however possible to check the GPU load in other places in AIDA64, e.g. on the Display / GPU page, or on the OSD Panel or SensorPanel. 8) We currently have no plans about supporting other GPGPU layers (e.g. CUDA, Stream, Direct3D CS) for GPU stressing purposes. 9) The current GPU stress module doesn't implement error-checking or results validation, hence any computation issues or glitches will not be detected or reported. 10) On old nVIDIA MCP chips (e.g. MCP73) the current GPU stress test may not work properly. We suspect it is due to a ForceWare driver bug or improper OpenCL support in ForceWare drivers. We'll investigate this issue later this year. Please let us know what do you think about this new feature, and what improvements would you like to see about it. Thanks, Fiery

AIDA64 already implements Areca RAID support, but it may or may not work with the latest Areca controllers. We have plans to improve the Areca and LSI RAID module of AIDA64 next year. Regards, Fiery