So finally I had time to test what I suggested (too)many times Deployed in a native vhd, W7 64 In VMware in UEFI mode. I purposely used a not updated, no drivers integrated, no file replaced, W7 AMD64 enterprise ISO. To make the matter more difficult I even used a GPT VHD (something I never do even in W11) Well except disabling the secure boot I had nothing to do, no patches, no UEFI7, nothing. Now I'm going to test if kb5017361 allows me to boot W/O disabling the secure boot (another thing I never use normally)
Good Luck On Your Mission. Don't Forget To Extract These Files (Bootx64.efi, Bootmgfw.efi, memtest.efi) you'll find 'em in efi Partition (EFI\BOOT\) Waiting To Hear Your Results & Exchange My Results With you.
Those files are from Server Core 2012r2, the system on partition I used to start with. (I used it just because the core version is small and quick to install) I have no idea if they are to be replaced, but I hope not. Obviously the terminology here is a bit flawed given the whole machine is a VM. So for "physical partition" I mean a partition created directly on the .vmdk while the native vhd is a vhd created inside the physical one
Despite how did you started the installation process of windows 7, the more important thing is, if you managed to install the update that supports secure boot and updating efi boot files in efi partition, what is the role of clover in your testing after installing windows 7 secure-boot (kb5017361) update ?. can clover help to install windows 7 (updated or not) on uefi class 3 physical machine ?
Still you don't understand me. There is no Win7 EFI partition involved in the process. There is one EFI partition and that's the one created by the Server 2012 (but could have been anything else) No I didn't update, as I had to install another bunch of updates as prerequisites. But I moved there a fully updated vhd that I use in multiboot in my notebook. It's formatted in MBR mode, and it booted both with secure boot or w/o it. Well booted is a big word, as the boot process got stuck later (likely because incompatible drivers or alike), but for sure the boot didn't get stopped early because wrong signatures or alike. That proves my initial idea that most of the booting problems can be simply ignored using a native vhd, and the more time I spend with them, the more I'm surprised by the lack of interest/understanding that I see here on MDL about the matter. Native vhd[x] are just one of the best things MS did in 30 years. Maybe, but I'm not sure, I use it on my server to boot it from NVME (something that isn't supported natively), and It never failed in 4/5 years. But it's purpose is (as I'm using it) exactly the opposite: booting in UEFI mode from a MBR only (or early UEFI) system.
You're Absolutely Right, Your theory Can Be Applied On Vmdk Containers Too These Containers Are Powerful When Used In VHDX Format, old Generation (VHD) Has Too Many Problems Under Hyper-V Manager 2012. Now I Got Your Point, NVME Can't Operate Efficiently In Legacy Firmware Which Forces You To Move Your Setup To UEFI But With MBR Partition Scheme To Preserve Windows 7 Smooth Operation, Right ?
Just almost no NVME drive has the BIOS extension to make it bootable in BIOS machines. (the samsung 950 pro is a notable exception but it's almost impossible to find today) So you need UEFI and you need an UEFI implementation recent enough to have the NVME modules. Is not the case of my server. Clover (which was created originally to boot macos on BIOS machines) solves the problem loading it's own UEFI implementation, which is extensible with optional modules, one of which is a NVME driver. Duet is another similar tool Modding the UEFI firmware is another option, but it's risky, and also not feasible on Intel branded mobos.
Integrating EFI Loader Of Windows 10 Into The Installation Media Like Integrate 7 (uses EFI Files From Win 10 v1903) And Windows 7 Image Updater (uses EFI Files From Win 10 v1709) Only Passes The First Half Part Of Installation Then The Installation Got Stuck @ Starting Windows Animation Loop. If There Is A Small Chance To Find Such EFI Loader To Finish The Installation Process, I'd Be Happy To Test It On My Machine. Because I'm Totally Exhausted Trying Every Solution Starting From 1 June, But Without Any Remarkable Progress. I'd Be Grateful For Any Help In My Campaign.
Why you insist in looking how to solve a problem you don't have? Most of the guides are about booting UEFI from a *BIOS* machine
Update : Tried To Get It Work For 14 Days but The Result Still The Same (Black Screen With 4 Tiny Green Dots), Still The Developer Does not Want To Respond To My Mails, The Reseller Didn't Want To Refund Me Either (Sadly). Advice 1: Everyone, Don't Waste Your Hard-Earned Money On Faulty Software. Advice 2 : Stay Away From PrimeExpert (Developer) Or PayPro Global (Reseller).
I was reffering to your follwing comment that you also metioned DUET: the mentioned guides do offer a solution to boot directly out of an NVME drive. I am sure all UEFI Class 3 systems can boot directly from an NVME drive just fine