Precision 7530 & Precision 7730 owner's thread

I purchased this card from the seller (i am UK based)

Ordered the 7740 heatsink from a seller in America

All fitted in no problems, now just trying to sort drivers which seem a little glitchy (but i know i have seen a thread about modifying the inf file)

 

I found and reinstalled the drivers that were in use on the dates when the battery was good (460.89), but they don't seem to be making any difference. Might well be a Windows update that caused this behaviour.

Yes, I've been following your posts on this forum for a while now and tracking your progress on the matter. I was hoping that someone else had the same experience as I had, with the battery life suddenly improving massively so I could compare notes. Most of the time I am plugged in during the day, so it doesn't really matter, but it was nice to be able to unplug and not worry about the 2 hour limit for the few weeks that it lasted.

One thing I would add from the experience, is that Dell support, once you get past the initial "check this and check that" have been excellent. I've had 2 battery replacements along with a few other issues, and they have been very good at sending an engineer out and replacing parts very quickly. That pretty much makes the extra expense of the Precision line worth it to me, although I do need the Quadro GPU so it's my only choice really.

 

Hello everyone, recently I've started struggling for longer battery life on my 7530 and here are current results: in Win 20H2, with latest BIOS and drivers, I also have the issue of Quadro draining battery when not in use. But in fact the only thing needed (for my system) to fix GPU drain - is having any program with OpenGL context tied to an invisible window and assigned to Nvidia GPU. I've compiled a small .exe which creates a window, does wglMakeCurrent for an OpenGL context and hides window (so that the program is visible only in Task Manager); and assigned this program to Quadro in control panel. According to HWInfo, this reduces power consumption from 20W which I used to have before to 10W, the same that I see with Lightroom launched & minimized. Actually, I came to this solution today, so still have to see if this really works long-term.

 

This has mostly been my opinion: excellent, fast and quality service. It does vary with location, though, from what I gather: many Redditors in the US don't fare very well, even with ProSupport. Pity it's so expensive, though: I wanted to renew my warranty for another two years, and it cost a whopping S$700 per year.

So that's why minimising the Lightroom window worked, rather than leaving it simply open: likewise with Paint 3D. Any chance of sharing the source code?

 

So, I've been having a play around after updating to Windows 20H2 this afternoon, and by process of elimination, I've found that opening the Geforce Experience app and minimising it is giving me the same 10W power draw, according to HWinfo, along with the corresponding increase in battery life. I have to assume that it's to do with the OpenGL solution described above by Leeming and effectively the same as the lightroom solution from Ionising Radiation. Seems like we have a partial solution!

 

If we can isolate the fact that a hidden OpenGL instance that doesn't perform any load, and is locked to the NVIDIA GPU, and causes the notebook GPU to power down, then we can attempt to do some stack traces to see where the ACPI methods are defined wrongly. For the record, this appears to happen on Linux, too.

 

With what @Leeming has figured out we can probably come up with a pretty "convenient" solution even if Dell never gets around to fixing this. (Thinking about a hidden background app that runs at startup, some sort of redundancy to recreate the window if needed [i.e. after NVIDIA driver update, sleep, ...?]; maybe a system tray icon present to kill it if you need to, and maybe it could even be set up to run as a Windows service so that it can be active as soon as the machine boots ..... not sure if something running outside of the desktop context would be able to trigger the "power off" behavior.) I would be willing to help out with this, I think that something could be put together pretty quickly...

 

Below is minimalistic working code sample which creates invisible window with tied OpenGL context.
Interestingly, after some time torturing the machine (multiple sleeping/waking, disabling and enabling Quadro in device manager) I somehow managed to make Windows stop recognizing the card and show it as generic video adapter. Then I deleted the device, rebooted the system (it recognized Quadro again) and now I see GPU power drain only if I disable the GPU in device manager, otherwise power consumption is about 10W regardless of having any app using GPU. The issue most probably will return, but for now cannot do more experiments with the code unless someone finds a reliable way to bring the bug back.

Code:

#include <Windows.h>
#include <gl/GL.h>

#pragma comment (lib, "opengl32.lib")

LRESULT CALLBACK WndProc(HWND hWnd, UINT message, WPARAM wParam, LPARAM lParam);

int WinMain(__in HINSTANCE hInstance, __in_opt HINSTANCE hPrevInstance, __in_opt LPSTR lpCmdLine, __in int nShowCmd)
{
    MSG msg = {0};
    WNDCLASS wc = {0};
    wc.lpfnWndProc = WndProc;
    wc.hInstance = hInstance;
    wc.hbrBackground = (HBRUSH)(COLOR_BACKGROUND);
    wc.lpszClassName = L"activateGPU";
    wc.style = CS_OWNDC;
    if(!RegisterClass(&wc))
        return 1;
    CreateWindowW(wc.lpszClassName, L"activateGPU", WS_OVERLAPPEDWINDOW | WS_VISIBLE, 0, 0, 640, 480, 0, 0, hInstance, 0);

    while(GetMessage(&msg, NULL, 0, 0) > 0)
        DispatchMessage(&msg);

    return 0;
}

LRESULT CALLBACK WndProc(HWND hWnd, UINT message, WPARAM wParam, LPARAM lParam)
{
    PIXELFORMATDESCRIPTOR pfd =
    {
        sizeof(PIXELFORMATDESCRIPTOR),
        1,
        PFD_DRAW_TO_WINDOW | PFD_SUPPORT_OPENGL | PFD_DOUBLEBUFFER,
        PFD_TYPE_RGBA,
        32,
        0, 0, 0, 0, 0, 0,
        0,
        0,
        0,
        0, 0, 0, 0,
        24,
        8,
        0,
        PFD_MAIN_PLANE,
        0,
        0, 0, 0
    };
    HGLRC hglrc = 0;
    switch(message)
    {
    case WM_CREATE:
    {
        HDC hdc = GetDC(hWnd);

        int  pf = ChoosePixelFormat(hdc, &pfd);
        SetPixelFormat(hdc, pf, &pfd);

        hglrc = wglCreateContext(hdc);
        wglMakeCurrent(hdc, hglrc);
        PostMessage(hWnd, WM_USER + 1, 0, 0);
        break;
    }
    case WM_USER + 1:
        ShowWindow(hWnd, SW_HIDE);
        break;
    case WM_CLOSE:
        PostQuitMessage(0);
        break;
    case WM_DESTROY:
        wglDeleteContext(hglrc);
        break;
    default:
        break;
    }
    return DefWindowProc(hWnd, message, wParam, lParam);
}

 

Thanks for posting...
Are you just using settings in NVIDIA control panel to get this program to attach to the NVIDIA GPU instead of the Intel one?

 

Compiled the source code above, for anyone who wants to give it a try.
(Might need Visual C++ 2019 redistributable. https://aka.ms/vs/16/release/vc_redist.x64.exe)