Yeah, but which game?

Different games have different file formats that may be named the same, so one game's .pak files can be completely different from another game's .pak files.

mat_viewportscale says what percentage of the screen the viewport (the view into the 3d world you see) should be sized to. This is always true, but it's mat_viewportupscale that determines how it's drawn to the screen.

mat_viewportupscale 0 says "the viewport only takes up half the screen, well let's just leave the rest blank and forget about it." This is what OP's problem was, he had it set to this.
mat_viewportupscale 1 says "the viewport only takes up half the screen, so let's resize it to fill the rest of the screen" This is what a lot of people complain about with comanglia's toaster cfg, that it makes everything look really blurred.

r1ce does a pretty good job of explaining it.

vgui_drawtree actually seems to suggest the opposite for me. Opening vgui_drawtree outside of the mainmenu shows extremely different components than inside the mainmenu. Though the "show hidden" option crashes tf2 for me and that might reveal something more, I can't say.

What's happening is that the engine is not prepared to redraw parts of the screen that aren't covered by anything, because normally it always should and it'd just make everything laggier if it had to check every time. Even though the mainmenu isn't actually enabled, it's still visible just because nothing actually replaces it and tells the engine to cover it up. If you have something you can drag over the screen (like the vgui_drawtree window) it can end up covering the whole screen, even though there's still only the one instance of it.

Basically, don't go removing your mainmenu hoping it'll help fps, because it probably won't (and if it does, valve fucked up something way weirder than always having the mainmenu enabled).

Incase you haven't used cheats:
Good plays will never result in a VAC ban. VAC isn't one of the anticheats that does that. The most common legitimate causes I've seen for VAC bans are

• Having cheatengine open, even if it's in the background and not doing anything to tf2
  
• Programs like hidmacros that you think wouldn't inject any dll's but do. You can check to see what dll's you have injected currently with https://www.nirsoft.net/utils/injected_dll.html (keep in mind that many dll's are allowed by VAC, and many are even used by the OS by default, but if you see anything suspicious there then that may be the issue).
  
• Running SAM while ingame. I don't think this applies to you because your SAM'd achievements were from years ago, but is worth mentioning.

If any of these are the case, there's not much you can do other than make a new account and not run whatever caused the ban.

If you've checked and none of these apply, just wait. Afaik they keep logs of all VAC bans and automatically reverse false positives, within a week or so. More about that here: http://www.teamfortress.tv/22346/i-got-vac-banned-and-it-was-a-mistake

http://i.imgur.com/aOSTFsu.png

I wanted my sensitivity to be low enough that moving my mouse an inch = moving an inch on my screen. It doesn't really work that way, since your perspective means a 5 degree angle will be much more pixels near the edge of your screen than near the center (don't know the details, but this is perspective distortion in photography), but ~0.98 ingame 400 DPI no accel worked for the most part. I thought it was extreme, but hitscan ended up playing like a dream for me, so I kept it.

I wouldn't recommend even 80+cm/360 without something to make you turn faster (like mouse accel or a sens bind) because at that point it really does limit you. But it's not as crazy as it sounds either, and you can definitely make it work.

• A lower sens means that an accurate shot is easier, since being a few mm off will be basically nothing on 100cm/360 but mean you missed by a lot on 5cm/360 for example. This isn't really wrist vs arm, since the wrist actually is more accurate than the arm at the same distances, but to have a lower sens / bigger distance you pretty much need to use your arm to aim and that's way more important than your wrist being slightly better at high sens / smaller distances.
  
• The arm is less prone to joint strain and RSI than using your wrist. A "low but still wrist aiming" sens is probably the worst thing you can do for this, just because you'll have to bend your wrist in more unnatural positions than a higher sens or a "arm aiming only" sens.
  
• Arm aim is easier to track / do slow movements I find, because you can be more conscious of your movements than your wrist. On the flipside flicking is better with the wrist, but most people used to a high sens flick too much anyway.

The switch shouldn't take too long, you just need to use it for the few days or so until it becomes more natural. I'm not a great aimer (and I need to use a sensitivity multiplier bind for rocketjumps) but I went from 9cm/360 for most of my time playing tf2, to 106cm/360 in around 10 hours of gameplay. I gotta say it feels a lot more natural and like I have a lot more control (and can aim better), but that might just be me.

G900 can be plugged in if you want, and though I'm not a fan of wireless myself logitech's easily got the best implementation of wireless mice currently.

Its latency is still pretty low (not as good as other logitech mice, but better than any zowie mouse for example; the actual internal hardware and firmware is the main cause of latency, not the connection itself), its battery life is pretty good if you're down for plugging it in when you're not using it, and its weight could be a lot lower but if you're considering a G900 anyway then low weights probably don't matter all that much to you. It's still tradeoffs for a feature a lot of people don't care about, but it's not a major issue or objectively bad design decision that kills the mouse either.

If you load any files with #base, you need to make sure they also have minmodeDisabled (or whatever you chose) instead of minmode. Not sure if that's your issue, but that's what it was when I ran into that problem.

If it turns out to not be that, you might need to post some snippets or a DL.

Valve's not really satisfying them either, though. They're also disappointed with the lack of updates, communication, and quality content. Both communities are slowly losing player numbers for the same reason, valve's just not putting out consistently good stuff in a timely manner like they were before.

What I always do is open every file up in notepad++ (you can just drag the entire hud folder into the notepad++ window) and ctrl+h replace-all of "_minmode" to "_minmodeDisabled" or whatever else so that it gets ignored. It may not look the nicest but it works, even if a basic program to remove those lines would be the best way of doing it.

A download link would just get outdated next major update, it doesn't take long to do it yourself if you know how.

Plugable fullsized if you like extremely clicky switches (outemu blues are very loud, and also the only outemu switch that's any good) and is well under budget.

Magicforce fullsized with gateron switches of your choice if you don't like extremely loud clickiness.

antialiasing itself can only be 0 or 1

You can add blur to it with "blur" "1" and that number can go higher, but it's pretty different from antialiasing depending on what you're trying to do

There are a lot of different things to base it off of, but fatal1ty is the greatest I'd say--no one compares in terms of consistent top-level performance for fps games. He's easily the most prominent player from the era when most people started taking notice of eSports, and is still one of the better known players despite retiring a decade ago.

Some other players of note:

• Thresh is probably the first prominent player. Unfortunately he stopped playing before eSports really started picking up, but he's got a Guinness World Record as the first professional gamer at least.
  
• Mew2King has by far the most tournament plays between any games at 534 currently, and is only a few wins away from having more #1 placements than the next most frequent player (Nerchio, a starcraft player) has tournaments entered.
  
• UNiVeRsE currently has the most prize money to his name at over 2.7 million, though it's always pretty close between top Dota2 players and with the International 2017 coming up soon, there's a lot of potential for another Dota2 player to take it up instead.
  
• It's personal, but the most impressive plays I've seen in any game(s) come from Jhlee0133 (namely Ascention to Heaven DT, Blue Zenith, A Fool Moon Night, and his unreadable BMS plays), even if it's niche and not what most people think of eSports as. Though some other vsrg players (jakad's Doppleganger DT or ABCD's 14K+2 abilities) are pretty damn good too.

https://www.youtube.com/watch?v=ucnYBA2xYdo

Pretty much just go in clientscheme.res and change every instance of surface-medium with the tf2 font of your choice (ctrl+h for the replace dialogue and selecting "replace all" helps). You might have to tweak the sizes too but it'll probably still look fine for most things, and without any tweaking it could be done in 30 seconds.

Keep those keys, learning how to do it yourself isn't hard

I think basically all disagreement is going to be whether or not you think only weapons actively bad to comp play should be banned, or if you should ban potentially annoying and/or ban useless weapons (in other words, only whitelist actually good weapons). And that's all personal preference.

My only issue with what you have (knowing what kind of whitelist you're going for, even if I'd like to see a different direction with it personally) is that it's a bit inconsistent. Some unbanned weapons (like the loose cannon imo) have a lot more potential to be broken than some banned weapons (like half of what's banned for soldier) and it doesn't feel like there's any rhyme or reason to which ones you decided to ban or not.