Squall Leonhart

This topic involved a crash in the application that occured when it was left on that page of aida64, this has been fixed for a long long time now. PS: Don't turn off the page file, windows can't do some things with one turned off, such as allocate large virtual blocks (whether it intends to fill them or not). Set a Min/Max pagefile instead.

Drive type and model would be agood idea too ;), as well as the ahci driver in use.

Fiery is a busy man.

I believe the nvidia kernel mode driver is suffering a double free or use after free which is brought on via applications using nvapi, thats why the faulting process is one which accesses the card using this api. windows will often say ntoskernl or ntkrnlmp because of an imprecise timing issue in the exception handler, in my case its pointing at X64_0x1E_nt!KiDispatchException+1c7 which is definitely not the actual fault lol.

I don't, not without recompiling the claymore binary.

not nicehash itself, the claymore binary and a few others have nvapi interaction to monitor card thermals and disable p2 cuda.

i can't say i ever had issues with 391.35, it might be just your unique mining configuration in that case Xixou,

But now Driver is corrupting memory, I've had bsods with aida64 and gpu-shark so far, others have with precision and msi ab Fiery, does AIDA64 refresh sensors when running in the tray but no output panels or lcd's are set up? asking because if nvidia tries to repro the bsod issue on aida64 they'll have to do it on the GPU info or Sensors if the refreshing is paused when the window isn't active.

Blast from the past, but i just found an nvidia txt file covering this issue XD http://download.nvidia.com/open-gpu-doc/gk104-disable-underflow-reporting/1/gk104-disable-underflow-reporting.txt

Fixed with 397.64

I've just done some file drop trickery to get aida64/gpuz to use an older nvapi to stop the leaks, same with the nvidia containers

r397 fixes one thing and breaks everything else on windows 7, any sensoring app you open leaks GDI objects Aida64 included (as of 397.55)

Follow up post since others are reading this "Commit Size" is titled "Page File" in windows XP/9x and certain utilities made by a Unwinder, but it does not reflect the amount of of actual on-disk page file in use.

ooo.... so the lack of an actual register is why my 10-10-10-27 ram shows up as 8-10-10-27 in CPU-Z/PC Wizard and 11-10-10-27 in AIDA64/HWInfo after resume from S3 It's pulling a last known value from somewhere in different ways.

I was reading a review just now from a certain website, and the reviewer has blamed Aida64 for not being compatible with Threadrippers UMA mode on its less than perfect memory bandwidth. This same reviewer went on to bench a bunch of video games in UMA, just because it was the default setting completely ignoring the fact that the NUMA setting is intended for gaming appication. He might have well have been advertising for Intel with the gaming scores. IMO, its time these has beens leave the hardware journalism business if they don't even understand whats going on anymore.

They do, however only the VRAM of the 1080 is returned, this probably comes back to the fact only the primary display driver is listed though, so nothing can be done about that

Windows 7 x64 with latest patches. was running a heavily modded skyrim last night, and video memory up till 4096MB was reported correctly, anything over would start again at 0MB. This might actually be a limitation of windows, i see in Process Explorer the dedicated memory value caps at 4096MB. GPU-Z / Nvidia Inspector must be getting memory usage info from nvapi then :<

Also, AIDA64 doesn't at all show any DirectX Video information for secondary display drivers (devices) I have a 1080 and 1060 in this machine, only the information for the 1080 is shown.

It seems when vram usage passes 4GB, the aida64 values for the used GPU begin showing very low numbers. This i what i see in Aida64, where GPU-Z and Nvidia Inspector show >4.8GB used. Utilization GPU 98% Memory Controller 46% Video Engine 0% Bus Interface 5% Dedicated Memory 609 MB Dynamic Memory 699 MB DirectX Video info shows Direct3D Device Properties Total / Free Video Memory 8192 MB / 3385 MB

Ok, i've got some interesting information on this. The chipset page lists the following of a system currently running the marvel sata, jmicron sata, and nec usb 3, with a geforce 550ti and renesas usb 3 card installed in the bottom 16x slot PCI Express Controller PCI-E 2.0 x2 port #1 In Use @ x1 (Marvell 88SE9123 SATA 6Gb/s Controller) PCI-E 2.0 x2 port #2 In Use @ x1 (NEC uPD720200F1 USB 3.0 Host Controller) PCI-E 2.0 x16 port #3 In Use @ x16 (nVIDIA GeForce GTX 550 Ti Video Adapter, nVIDIA GF116 - High Definition Audio Controller) (top most pci-e 16x 2.0 port) PCI-E 2.0 x16 port #7 Empty PCI Express Controller PCI-E 1.0 x1 port #1 Empty PCI-E 1.0 x1 port #3 In Use @ x1 (Renesas uPD720201 USB 3.0 Host Controller) (bottom most pci-e 16x 2.0 port) PCI-E 1.0 x1 port #5 In Use @ x1 (JMicron JMB363 SATA-II RAID Controller) The location of the added usb 3 card is confirmed by device manager, which shows it running off the ich10 pci-e root port, even though it is in the bottom most 16x 2.0 slot. in my present Rampage II gene setup, i have a XFI soundcard in the second 16x slot, and it is shown as operating off the x58. I am therefore led to believe, that unless a card is placed in the 3rd pci-e 16x slot, the 4th one is not actually able to connect to the x58 root hub.

both are connected to the JMB chip presently, one connected to the rear esata port, and another connected on the case esata port. both of these will be going internal when my new build is ready.

JMB363, 1 esata, 1 internal sata, and a pata port for 2 devices.

All im doing is selecting the MX300 after the MX200 and vice versa and the smart fields come back wrong, some of them duplicated or missing. This could just be a Jmicron glitch, or one that only occurs when used with MSAHCI drivers, jmicron drivers are slower and buggy, and don't work with SSD tools well at all.

it hasn't, im guessing it might be the controller doing it

The MX200 is the first in the list