The Lounge is rated Safe For Work. If you're about to post something inappropriate for a shared office environment, then don't post it. No ads, no abuse, and no programming questions. Trolling, (political, climate, religious or whatever) will result in your account being removed.
Recently I have suffered with a wifi adapter issue related to win10 drivers and win10 precludes me from rolling back to a driver that worked in the past. win10 controls everything. quite annoying.
Windows will use the date and driver version to make this decision... if you attempt to 'roll-back' and then Windows locates a more recent driver then you may end up having the most recent driver reinstalled.
It takes less than 60 seconds to remove it from the driver store. Open a Command Prompt and do:
dism /online /get-drivers /format:table
Locate your Wifi driver in that list. If you are unable to locate the driver in this list then you can also get a more detailed list by doing:
Once you locate the Wifi driver then do:
pnputil /delete-driver [INF FILE NAME]
Reboot and reinstall your old driver. Also... if your Intel Wireless driver has been published to 'Windows Update' then make sure you are not connected to the internet when you do this. Otherwise the latest driver will be downloaded from Windows Update.
I never mentioned this... but Intel will probably blame it on your "INSYDE Corp" BIOS. Make sure you have the latest BIOS. Even if you do have the latest BIOS... reboot and go modify your power management settings. Go in there and look for an 'S4'[^] and try disabling that. try running with that for a week and see if your problem goes away.
Also... go into Windows Device manager and right click on your wireless card in the list. There should be a setting on one of the tabs to prevent the wireless card from entering the low power state during sleep/hibernation.
Think outside the box... you don't really need to mess around with BIOS and drivers... you can just instruct the operating system to not allow your wireless driver to enter the low power state.
This might be completely unrelated, but it's food for thought anyway.
There is one common username/password I prefer to use to log onto my different machines. However, I have a laptop (a netbook), on which I cannot use this password with some Linux distribution (I forget which--maybe some version of Lubuntu, because it's supposedly low-overhead) because it simply refuses to let me type a specific key (lowercase 'm') when attempting to log in.
No "." character gets added in the password field, unlike every other key for the password. The button isn't broken; when I initially entered it during the account creation, the key was functional, and at no other time does the key ever misbehave. For example, in any plain text field, if I press 'm', there's no problem. I even reinstalled the OS, after I realized what was going on, and paid close attention to the password field as I entered the password when (re-)creating the account. It gets accepted at that time. Yet once I'm at the actual login screen, it's rejected.
Being a laptop, some of the keys serve double-duty and are used as the numeric keypad as well - 'm' gets assigned to '0', and the NumLock key is used to switch between numbers and letters. However, no matter how many times I press NumLock at the login screen, pressing '0'/'m' registers no additional keystroke.
I've also seen it where a key is misbehaving. Especially on laptops, I've seen stuck alt or shift keys, etc.
One thing that happens on our iMacs is that if you hit space to wake the computer up and turn on the monitor to show the login page, it actually puts a space before your username and won't log in properly. It's really hard to see because the text is barely moved over.
This is probably not related to your problem at all, but I will bring it up anyway as somebody else may have some idea about what caused it.
When I travel on holiday I don't take my laptop as I don't need it. I just take an ancient HP netbook running XP, mainly to back up my photos each night and check my email via webmail.
I went to the Philippines for about a month over Christmas, and imagine my chagrin when it told me the password was wrong, yet I had been using it for years - I only use this HP for holidays and the only things on it are my photos which are removed as soon as I get home.
On a close examination of the input box I noticed that the *s appeared to jump around a bit, and established that the keyboard was producing the correct letter, but occasionally was putting the nth character at the start of the password. I later established that this was pretty random - it might happen, it might not, it may chose a different character each time, it might not.
The only solution I found was to use the 'impaired' keyboard, and use the mouse to click on each character in turn, which always worked. Once logged in, the keyboard behave normally.
That COM error definitely doesn't have anything to do with it, those can often be ignored unless you're having an app-specific problem associated with that Class ID.
If you'd like, feel free to shoot me an email jonmbutler at outlook dot com and I'll see what I can do to assist. Worst case, if we can't figure it out, I'll hook you up with a free support case (I work for MSFT.)
Thank you very much, Jon. I really appreciate it.
You are, of course, correct that the COM errors have nothing to do with it.
I actually cleared those up (following the set of steps in the youtube video I posted in the original post).
I was just very curious about this password problem because I've seen it a number of times lately and it seems to be when my computer goes offline.
I "fixed" it by cycling my computer off/on and then it let me log in.
Additional Weird Stuff - and the Weirdness Is All Me
Here's the deal why I was very interested in why this might be happening. This is going to take a bit of explanation.
I use a 64 character password (which is a SHA-256 hash) for my windows login.
It looks something like the following: 1f0bf4b762443fd0bbed59bc1d583a3ac80d45c7bcf308376be2593784244173
I told you this explanation was going to get weird.
That's because I use a program I wrote called C'YaPass which allows me to draw a pattern and choose a key to generate a random password (based upon SHA-256).
Add a site/key, draw your pattern and you'll see that a SHA-256 hash is generated for your password.
Weirdness Part 2
Okay, so since a 64 char password of random numbers and letters is difficult to type, I created :
1. android app which allows me to draw a pattern, pick a site/key and generate the same SHA-256 based password (as long as I draw the same pattern and pick the secret sit key).
2. will send the password over bluetooth
3. a device (arduino) which emulates a keyboard and is plugged into my USB port which receives the password over bluetooth and "types" the password in and "presses" <ENTER>
Finally, The Point
The point is that I was sending my password over bluetooth and every time my computer was rejecting the password, even though the app was sending the password properly. My additional point is that I couldn't have been typing it incorrectly because the app sends the password and the computer receives it.
Could there be a bug? That is entirely possible, however I've been using this app for over 1 year with no problems, except periodically when my machine goes offline, it thinks my password is incorrect.
Way Too Long
I know that is way too long but just thought it might be interesting. Or not.
I'm at a loss to explain why this is happening based only on what you describe -- clever approach though.
All of that aside … have you considered using a Windows Hello PIN to log in instead of a big fat password every time? That's even a bit more secure, since PINs are local to your machine and wouldn't work anywhere else even if they were compromised (unless, of course, it's your machine itself that gets compromised.)
Thanks for taking your valuable time to read my too-long post.
Also, I was just curious about the PIN thing and didn't understand how it could be secure.
Makes sense that it only works on the local machine and that is how it is secure.
Thanks for mentioning that.