aminomancer
commented
5 years ago I think I've discovered the root cause of the problem. Windows 1903 makes some unknown change that affects compatibility between different samsung NVMe SSD models. That can apparently manifest in pretty random ways. For me it was just sleep mode problems. I'm not sure why hackBGRT has any effect on it at all, but in any case samsung released an updated driver that circumvents the issue entirely and installing this fixes the issue even with hackBGRT installed
This might sound strange but my experience strongly indicates that hackBGRT was the culprit. I've been for one hell of a ride since updating to 1903. I first updated to it while grub2 was installed with both a linux boot drive and a clover+mac boot drive installed in the PC. This was pretty stupid I guess but I got away with it prior to this so I thought my initial fears were probably unfounded. Anyway it broke my windows bootloader entirely and all forms of recovery failed so I just reinstalled windows altogether. I usually have a better experience with major windows updates when I just start clean with them instead of updating from a previous build anyway. Set up all my stuff again and somewhere in there I installed hackbgrt as I normally do. I should say I've never had any problems with it until this, in about 4 or 5 years of using it, and love this program. I think there might be something in the may 2019 update that adversely interacts with it though. Now, it's happened a couple times in the past where something broke my sleep/hibernation, so pretty quickly after my clean install I hit sleep to make sure it was working, and it woke up just fine. It wasn't until I finished installing all my programs and stuff that sleep broke. In the past, in some cases I'd get a BSOD when waking up, in some cases I'd get a reboot to UEFI, this was actually the first case where the fans and hard drives started up, the keyboard and mouse were initialized (judging by the LEDs) but the screen just remained black. It would stay like this indefinitely and I would have to restart. Upon restart, I'd post fine, then go into a brief BGRT screen where I see my custom boot screen, then it would do this quick fade-to-black that I've never seen and just remain black again. Normally the transition from the custom boot screen to windows login screen is instant, or a cut to black if anything. So upon restarting from there, it would boot normally and I could log into windows. I don't know if that's useful to you but I'll try to be detailed. I knew it wasn't just a graphics output issue because my mouse has a special LED profile that it only loads if windows boots properly. It doesn't need me to log in or anything, it's set up as an administrative task, so even without any display output, I'd know if Windows successfully loaded. And in this case my mouse LED was set to the stock colors which tells me the bootloader was hanging for whatever reason.
Since I installed a lot of software and made some modifications with winaero tweaker and similar programs, and have had sleep related issues in the past, I didn't assume it was hackBGRT at first, so I spent hours trying different things. There is a lot of stuff to possibly blame here, my frankenstein system is a threadripper 1950X on an ASRock x399 professional gaming MB, 4x16GB of corsair LPX 3733mhz ram, a GTX 1080Ti, three samsung 970 evo+ drives, and a rocketraid 3740A raid card with 16 hgst helium 6's. First I did all of the stuff people normally suggest for sleep problems. Tried every possible combination of disabling/enabling hibernation, hybrid sleep, fast startup, wake timers, wake on LAN, etc., tried fixing graphics drivers with DDU, tried unplugging all my various hard drives, etc. So I knew something on my boot drive was messing stuff up. I uninstalled a ton of apps but I didn't want to reinstall windows yet since I already JUST did a clean install and spent so much time recovering all my programs and modifications and so on. So somewhere in there I had the bright idea to just run hackBGRT and type d. I made some other network changes too since it seemed too small of a change to bother with a whole trial & error session. This was a huge bitch since every time sleep failed I'd have to reboot TWICE to get back into windows for some reason. But yeah this time I went to sleep, clicked the mouse, and it finally worked. I went to sleep again and tried wake on lan from my phone and it worked. So then I re-enabled hackBGRT and it was right back to the black screen and so on. I went ahead and undid every single change I made in the process of troubleshooting this, but left hackBGRT off, and my system worked fine.
So I feel pretty confident that hackBGRT was causing this. Of course I doubt anyone else is having this problem - there was probably something I did during this install process that caused some kind of adverse interaction with hackBGRT. Possibly it's exclusive to my unusual setup. But I never had any problem with hackBGRT on this system until the may 2019 update, so it's not like my hardware is incompatible somehow. I also didn't do anything unusual this time. By all accounts it should have worked, except for the version of windows I installed. So on the off-chance that version 1903 changed the bootloader in some way that contraindicates hackBGRT (maybe for certain systems) I'm just posting my experience. Could just be a strange coincidence but now it's on your radar at least.