http://images.akamai.steamusercontent.com/ugc/480023352499639640/D3B7B42ED68704F55EF7F3396018C1BD4356D027/

http://images.akamai.steamusercontent.com/ugc/480023352499639455/79FF3F03C912F33DD2803B662BD811F300DD115C/

http://images.akamai.steamusercontent.com/ugc/480023352499639189/346DC969E8610C2C4BAE78B3DBB311CC4559844A/

getting some pretty white textures when light is on it, any tips on reducing it? Really only affects reckoner and process. (DX8.1 and -15 LOD nvidia inspec)

ALL tests had nvidia spec LOD bias -15, res 1920x917:

-autoconfig

Comanglia's Config (shadows disabled)

felik's Config //before running I set the entire options/advanced to the lowest possible as it increased FPS

Using this setup: i5 760, GTX 470, here's my conclusion:

Comanglia's certainly give better performance, however I've noticed particularly on DX8.1 that there's discernible 'stutter' (the FPS is fine but it essentially feels less smooth); setting it to DX9.x always gets rid of this. The 'stutter' is most noticeable when launching local servers, or having a stream running in the background.

Using feliks or valve's default, there always seems to be smoothness regardless of DX level. Keep in mind that the lack of smoothness I've experienced on Comanglia's AND chris's have not bothered me and I've actually not thought about it at all until switching to DX9. Still, the lack of FPS using default settings have always kept me from using anything but the two mentioned configs.

I'll be switching to feliks using DX8.1, despite the 10~% FPS loss over Comanglia's because it feels like it's running smoother.

well, I started a lobby today on stadium with 4 other players. Took 90~ mins before it started despite going through my friendlist and posting on discord. i don't think anyone wants to be part of that experience. as of right now you can't really blame people for not going there, def needs to plan an event to get large groups of people over there. hopefully stadium will take off soon

oh, so they're literally stacked on each other; would be more detailed for comparing each if it wasn't like that

it looks like a standard line graph at first glance, i think the fact that there are lines in this causes confusion. Most cumulative charts I've seen are simply a solid color covering a specific portion of the graph.

https://www.youtube.com/watch?v=V8f2FKXCyV4

enemies accept lose lol

thought I'd compile the important bits for anyone who wants to watch without having to go through the broadcast: https://www.youtube.com/watch?v=hOYxbdJKnX4

Well it's a preference, if you're not used to having them on you'll have issues until you get used to it. For example, jukebox was telling me he had the scattergun off for a long time, presumably because he felt that he aimed better. But then he switched back as he said he felt that was better for him at that point in time. And I've personally been doing the same thing on every single class over the years, going back and forth as to what I've felt was best for me at that very moment in time. I don't believe that simply having all off is better, it's simply a preference as to how I feel will benefit me the best at any given point.

I'd like to add that I'm sure there are people out there that wouldn't be able to aim if they turned the viewmodels off :))

I don't quite agree with the notion that it gives you a huge advantage; it's still more obvious what weapon you're using with viewmodels on.

If you were to brush out some broad statistics as to how much better a player is with/without viewmodels it would be so insignificantly small that on a practical level there'd be no difference. You could sit here all day arguing about how some item or class in the game gives an advantage that is so enormously larger in contrast that this discussion being held becomes so insignificant to the point where it's ridiculous to even have it.

Is there going to be a player you'll miss hiding under you minmodels in competitive because of it? It's banal to even impose the idea that there's an advantage to be gained, I'd argue that a custom HUD is more beneficial. Focus on what actually matters in game balance: items, classes, maps etc. This is again, on a practical level, purely a preference and I believe changing your FOV above 90 is similar since that itself compresses the middle which makes it harder to see with the upside of a very small additive viewing spectrum on the sides.

I was never sure of the specifics but I assumed this worked on all sv_pure servers, but it was just 1 that they updated to allow? How would that prevent them from simply banning this on 2?

Valve's team should really take a look at how the Reflex developers work and the amazing customization they allow there.

I really appreciate you trying to fix TF2, yttrium. It's very unfortunate when valve and the people directly communicating with them is going out their way to officially fuck us over.

Here's the youtube link instead btw: https://youtu.be/I9nRPFRD1QY?t=1366

Yup, same for me, starting a listen server first then joining anything later will always work. Pretty sure this is related to sv_pure since I think badwater.pub uses "2" if I remember.

Try this:
badwater.pub/europe - 212.47.245.244

Maybe you're having the same issue as me. Are you getting a bunch of model errors in console when it doesn't work? Try connecting to servers with sv_pure 1 and then try sv_pure 2. It's important that every single time you connect to a server you restart the game.

It waits a certain amount of time if you switch to pipes right before the sticky (0.8sec) is suppose to be ready. However if you switch the precise frame after shooting a sticky you can actually avoid having any delay. It's a bit wonky as it seems that it's almost the opposite when switching to your melee, the faster you do so after firing the longer the delay seems to be.

Also keep in mind that if you put a sticky down on the ground and wait an undefined amount of time, the delay does not seem to kick in. Thus this only applies in crucial moments when you need fast execution.