Dx12 On Windows 7

[Leoma Cianchetti]

I have already confirmed it only happens while i have eve clients running, playing any other games i dont get this issue. Nevertheless i did a full game/windows/all sorts of drivers reinstall and ran all the checks i could for faulty memory/gpu issues, i could not find any. After the full reinstall of basically everything the issue persists.

Yes, this is exactly what it seems to be like regardless of whether DX12 is ticked or not, every grid load seems to trigger reloading of all textures (based on disc activity and rough correlation with how populated the grid is).

i cant run direct x 12 stuff like diablo2r or some direct x 12 benchmarks etc.) all other stuff like older games up to direct x 11 runs perfect (i can see gpu acceleration on host system (the main windows VM -> taskmanager -> GPU)

i didnt make any special to get it work , just used the gpu-p script and via Hyper-V viewer i can interact with the vm + parsec etc, all worked out of the box instead dx12 + huge performance problems with the gpu on th e "host" machine (unraid->windows11 vm (gpu problems) -> hyperV->second windows vm

I have a new asus rog z790-e gaming i motherboard, 64gb 7200mhz ram, with i9-14900k processor. Everything runs great except dx12 games. In particular, I am running Fortnite. There are tons of people have this same issue with 13900k processors as well as 14900k processors. Since the 13900k issue started happening, many people have shared on Reddit and a few have found a work around by decreasing the pcore to 52x, which sucks to have to do, but it seems to work for some (not all). Some have RMA'd and think it's resolved, but it still comes back sometimes. How can I get help to solve this?

Just fwiw, I have the same issue. Built a new PC in January that has since just been bsod(ing) to high heaven when running any game. All the BSODs are to do with the watchclock and other memory failures. This even happens, albeit rarely, when using files that require processing like Excel.

We are glad to know you found a solution, thank you for taking the time to share this fix, and hopefully, it may help other community members experiencing similar behavior. Remember you can always check the Memory Specifications.

If I use Intel XTU to reduce my P-Cores from 57 (which they default to without overclocking on this setup) to 54... all DX12 Games run fine (my worst-case scenario was Immortals of Aveum, which wouldn't even launch no matter what settings I tried previously).

It's easy enough to switch this back and forth, but it's a hack, not a solution. I'd like to know from Intel what is going on; whether it is an issue in DX12 (which is still in beta, and has been for a very long time), or the Intel 14900KF, or the ASUS motherboard, or the RAM. The RAM is listed on the ASUS QVL list.

The symptom is identical to what others above have reported, if I _don't_ downlock the P-Cores, I get all sorts of crashes, BSODs in Windows 11 (I have tried this in Home, Pro and am now on Preview (Beta) Pro Build, which is ironically the most stable at the moment), and even got stuck in some Windows 11 "Automatic Repair Loops" which recommended Resetting Windows, i.e. reinstalling it from scratch and then reinstalling all my many apps, which is a multi-day effort, so not a quick solution.

Please, Intel, provide some hard info on this. I have warned others about it in various places on the internet, the ASUS ROG Forum in the 600-700 Series section, have seen brief info in the Steam Community Immortals of Aveum Community Forum, from a participant with the handle of Grimzky (12-23-2023), where he claims this is a common issue with 14900 and 13900 series CPUs, and I have been laughed out of court in the Microsoft Flight Simulator Forum/Community (Tech section), where respondents, including moderators, have professed they've NEVER heard of this problem and think I must be some kind of idiot for thinking it is happening.

The last concerns me most, purely out of non-selfish concern for the upcoming Microsoft Flight Simulator 2024. This is a program enjoyed by literally tens of millions of participants around the world. Microsoft is intending to _require_ that DX12 be used when it has been installed. I don't know whether or how far they will get with fixing DX12 beta issues at that point, but none have shown up for a very long time now, so that's unclear as to whether it will ever change.

In any event, I would like to take this moment for the last time to encourage Intel to work with Microsoft to resolve whatever it is that is causing this issue. If it is something specific to the ASUS Z790 motherboards, that company needs to be brought into the loop, too.

Meanwhile, I can hack with the downclocking tweak, though it seems a shame to do so. It's not like there's much performance lost, but if I've spent money on an unlocked 14900 CPU, I should be able to use it as it is spec'd to be used, at least. Not being able to use it even without overclocking it unless I downclock the P-Cores seems like a defect of some sort, whether in marketing, manufacturing or implementation. The error should be corrected. And this should be more widely known as an issue. I'll leave it to others to publicize it.

Just created an account to say I have the exact same issue. Completely new system, ASUS Z790-Plus MB, i9-14900kf, RTX 4090. My system seems completely unstable, crashes left and right in CS2, MSFS, Cinebench. I can run some things on an underclock, but I still get instability somewhere in the system, even on stock RAM (DDR5-6000 running at 4800). I've had it with tinkering for each title I want to run. FIX please.

wow the amount of time i've spent trying different things to fix my pc and never would I have thought its the brand new cpu I bought, I literally replaced every single part in my pc and now I find out its the cpu. Need intel to find a fix for this asap.

Intel does not verify all solutions, including but not limited to any file transfers that may appear in this community. Accordingly, Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.

Application,ProcessID,SwapChainAddress,Runtime,SyncInterval,AllowsTearing,PresentFlags,PresentMode,Dropped,TimeInSeconds,MsBetweenPresents,MsBetweenDisplayChange,MsInPresentAPI,MsUntilRenderComplete,MsUntilDisplayed

I created a version of the program which has v-sync disabled. It gives tearing but I would like to see what the latency will when measuring with the light sensor. I will pick this up tomorrow when I am back at the office.

Circling back to this, it does look like there's an off-by-one in the frame latency waitable object. A requested frame latency value of 1 means "give me the minimum frame latency possible from any present mode, not necessarily the current one." So a composed swapchain will get you 2 frames of latency, just like a fullscreen / independent flip swapchain will.

Regarding the present stats workaround, I've confirmed we've got a bug there which is causing the zeroes. The workaround is to avoid using the SetFullscreenState API and just adjust your windows to cover the screens manually. If you do this (and call ResizeBuffers afterwards), then present stats should work correctly and you should be able to use that to get down to 1 frame of latency when your swapchain qualifies for independent flip.

Note that there are scenarios where composition will still be used (e.g. the volume indicator pops up), and the minimum latency does become 2. If you go the route of using frame statistics, and wait for a frame to be on-screen before rendering another, this will cause your application's framerate to drop to 30hz or worse. I have, however, confirmed that this approach does allow a 16ms latency measured by PresentMon. This approach will work in D3D11 or D3D12 as long as you use one of the FLIP swap effects (mandatory in D3D12).

I was inspired by @STREGAsGate's interest in building a game on Windows with DirectX and Swift. DirectX is a large framework and provides multiple sub-frameworks including Direct 3D and Direct 2D for accelerated graphics as well as things such as Direct Input and Direct Audio for audio and input handling.

In order to demonstrate that using this functionality is both possible and how far Swift on Windows has actually come, I decided to write one of the time honoured traditional demos for 3D graphics with Swift (and HLSL) to show how to use Direct 3D on Windows with Swift - an orthographic projection of a cube.

Note that this proof of concept implementation does not always follow best practices (e.g. does not use v-sync) but it does show the viability of implementing 3D graphics on Windows with Swift. The code patterns are not entirely idiomatic Swift, but would be very familiar to someone who is experienced with DirectX.

Hmm, I would then guess that there could be a mismatch between the SDK version in use and where the module map is currently present. This can happen if you have more than one SDK installed or if you have a mismatch between the UCRT and WinSDK. Perhaps a recent update to Visual Studio put things out of sync?

Yes, the implementation of the demo leaves much to be desired. I tried to call that out in the README, but if its not clear, please send a PR. I think that there could be value in providing a better implementation to make it easier for people to get started with DirectX, but that was beyond the scope of that particular attempt - but I definitely see the value in a separate attempt to provide something which can serve as a starting point for others.

I think this example is fantastic as is. It just needs some adjusting specifically when someone wants to use a VM for development as the VM and host become unresponsive, however on a native machine it runs great. I've already used this example to help me make a basic window, which was really difficult for me to understand. As a mac/Linux programmer the way windows does it's references and com is a challenge for me and this example really helped me begin to grasp that stuff.

c80f0f1006