Jump to content

Hagbard Celine

Members
  • Posts

    18
  • Joined

Reputation

10 Good
  1. The system in question is: Motherboard: MSI N3050I ECO CPU: Celeron N3050 GPU: Integrated Intel HD Brasswell GPU2: CLUB 3D Radeon R5 230 1GB PCIe*1x (CGAX-R5232LX1) (Does not belong in system, but sporadically mounted for testing stuff.) Ram: 2xG.Skill F3-1600C9-4GRSL HardDrive: CRUCIAL MX300 525GB SSD OpticalDrive: ASUS BW-16D1HT PSU: Corsair AX860i OS: Windows 7 Ultimate(With all Windows updates installed) I've also discovered that the disappearing sensors do not happen if I close iCUE SPACE before starting benchmark and open it again when benchmark finished. So it seems the issue is failing to "paint"/"render" the sensor output under high system load. Also another issue with iCUE SPACE with same version on both my systems: iCUE SPACE has a tendency to randomly put itself on top of screen-saver. Edit: Have entered this systems info prepended by "System2:" in forum "System Specs" form.
  2. Well, I just tested this claim, it does not hold up: Been running the following Nvidia DX11 demos (from NVIDIA_SDK11_Direct3D_11.00.0328.2105) in parallel with "Stress CPU" in CPU-Z and "CPUID Hardware monitor PRO"(Trial version) for more than 30 min: Island11, Compute Shader N-Body Simulation, OpacityMapping, FXAA, GregoryACC11, Hair, Multi-View Soft Shadows and DirectCompute Ocean. Result: Not one single glitch from "CPUID Hardware Monitor PRO"-sensor-readings. Log file attached. iCUE sensor readings broke down before 3 minutes passed on same type test in my second post(That was only DX11 samples, no CPU-Z even). Edit: Also, after this last test one thing seems clear to me: This is a bug in the Corsair front-end, not the logging back-end. Because I got the fan on the AX860i set to react to changes in CPU Core0 temperature, and that seems to happen even after the sensor output have disappeared in "iCUE SPACE". Also as mentioned in first post, the workaround is to restart the iCUE program, not the iCUE Service. HWmonPRO.zip
  3. This is a cleanly installed system. Only thing running in background is Intel HD Graphics Tray, MSI Live Update 6, Realtek HD Audio Manager, MSI Fast Boot, Logitech SetPoint and AMD Catalyst Control Center.
  4. OK, you got one example of another combination that experience a similar bug. That does not disprove the fact that it is a bug. USB is certainly not the issue, the sensors that disappear are motherboard temp and fan speeds, CPU temp and load, and the DRAM data(CAS, RAS and so on), and the temp of the PCIE Radeon card if I got that installed. The only sensor connected trough USB is the AX860i, which do not disappear ever. Else there is only a keyboard and two trackballs on USB-bus.
  5. Benchmarks was only an example, anything that loads the system considerably gets it to loose the sensors. What I did during the attached log was start the rendering test in GPU-Z plus pressing the Benchmark button in CPU-Z. Edit: On another note having problems running concurrently with other monitoring programs is a BUG in itself, especially since iCUE is a program that controls cooling. I've never experienced a sensor-monitoring-program on windows not behaving due to other sensor-monitoring-program running concurrently. And at times I've had up to at-least 6 running at the same time (open hardware monitor, hwinfo64, siv64, SpeedFan, sidebar diagnostics and GPU-Z, just of the top of my head), and this has never been an issue with any program. Edit2: Attached logs where the same thing happens with no other sensor-reading-programs running. Only started some dx11 samples to to load the system. LostSensorsWithOnlyDX11samples.zip
  6. Running iCUE v3.12.118 on Windows 7 Ultimate with all updates. When I start a benchmark or anything that loads the system considerably, all sensor output exept for the AX860i disappears from iCUE Space after a short time. Stopping and restarting the iCUE tray application gets the readings back. Logs and system-info dump attached(This is not my main system as described in my user info, but a MSI N3050i ECO). LostSensors.zip
  7. OK, this time I have actually tested my new fix to the point I'm positive it works on AX860i at least. Updated code here: https://github.com/Hagbard-Celin/cpsumon
  8. Edit2: Correction: I did NOT correct the problem, it just worked by fluke. Guess this is my cue to learn to use my DSLogic 16 logic analyzer(Don't hold your breath), I'll be back with updates IF I get somewhere.. Edit: Just ignore my request for copy of repo. I belive I got it working, updated code here: https://github.com/Hagbard-Celin/cpsumon Does anyone have a copy of this repo? I only got the cpsufancli.c file, which used to work when paired with https://github.com/ka87/cpsumon. But after a HD crash and reinstall I can not get that combo to work anymore.
  9. Hi, my Windows is suffering from hotkey-assignment-overload. I play some games that needs lots of "Alt-SomeKey" shortcuts and many programs wants the same shortcuts as default, so I'm running out of available keys to assign to unless I'm to remember lots of multi qualifier hotkeys. This is the main reason I bought a "K95 RGB". But the iCue software lacks one basic feature, when you map a key to "Launch Application" it will not accept arguments to the program being run. If I put "C:\WINDOWS\system32\notepad.exe" in "Launch application settings" all is OK. If I on the other hand put "C:\WINDOWS\system32\notepad.exe C:\test\TestOk.txt" it just fails silently. I need this for controlling DisplayFusion trough command-line interface, to free up hotkey-mappings for other programs and games. I know I can use bat files, but the extra delay for cmd.exe is painful when compared to doing the same thing on my Logitech G13 that actually supports command-line-arguments on executed applications.
  10. There is some code here: https://github.com/cloudkitsch/cpsumon that is reported to be half working. There is also https://github.com/audiohacked/OpenCorsairLink who has just started on AXi support. Disclamer: I've not had time to try any of them myself yet.
  11. Thank you very much, then I'll be able to actually use my gfx-cards in a sensible way:D:
  12. Yes, that is exactly what I need. For X distinction, they only need to identify with different names in /sys/devices/virtial/input/inputX/name or someting like that. According to this evdev can filter on any info under that directory.
  13. Been using the testing version from 06-18-2015 and except from regularly missing the first keystroke after booting or opening a terminal in X it seems to work with my K95 RGB in gentoo. (There is also a small issue with evdev saying "BUG: triggered 'if (axnum >= dev->valuator->numAxes)'" but my research points to that being the fault of evdev) I guess these are known issues, just wanted to report my experience when I was making this post anyway. He reason for the post; I was wondering if I could request a feature? I'm trying to get a usable multi-monitor setup with working. I'm able to get 2 separate X-servers up no the Nvidia and Intel cards. And I can control both with the same mouse and keyboard with x2x or synergy. The problem is with vt1-tv6 that is running on the radeon card. I can get the vt's on the radeon to stay active while in X by starting X with --sharevts but then the keyboard and mouse gets mirrored to both X and the active vt. To avoid this mirroring there is the option "GrabDevice", but with that on, one need a separate keyboard to control the console. One solution is conspy, which is impractical and painfully slow (it even slows down the output on the monitor that is connected to the radeon card to a crawl) as it insists on transferring the output of the vt to a X-window. The other solution I found was a perl script that uses the same method for keyboard-injection as conspy as far as I can tell. My first idea was to try using that with ckb, but I realize don't understand enough to make it work. So my humble request is; Could you please implement a option for the driver to create additional /dev/input devices that we can map keys to as "virtual keyboards".
  14. Hi, I have a problem with this driver in Gentoo. After starting the daemon, key-repeat outside of X (in vt1 to vt6) stops working.
×
×
  • Create New...