I just ran a few tests of a brief demo of graphics intensive pub play and got surprising results

DX9
___
dx9frames:73.60 fps 8.985 fps variability
maxframes: 76.16 fps 8.953 fps variability
highframes: 62.94 fps 7.651 fps variability
default: 115.47 fps 8.673 fps variability

DX8
____
dx9frames: 112.34 fps 8.330 fps variability
maxframes:115.47 fps 8.673 fps variability
highframes:87.82 fps 6.042 fps variability
default: 83.38 fps 6.941 fps variability

"default" is what the game sets things to with -dxlevel 95/81 still in the launch options

so according to this, for my system, DX9 default performance and DX8 maxframes are identical. what even. also dx9frames gets better performance in dx8, even on my system which seems greatly biased toward dx9? Yeah I have no idea.

I have it scripted so Q and E handle everything (1 calls medic). Took some getting used to to have things changing depending on what I have out, but it has its advantages. I like the way it feels.

same bug here and I actually like this better than either the regular beam or complete lack of a beam. nice job breaking it, heroes

great game. most novel thing I've played since portal. (no crashing for me, fortunately.)

sigma saves the day. thank you, I now understand 100%

thanks for clearing up my misconception, too. I thought it was like n(1.5) per heal target rather than n+50% per heal target.

It's pretty hard to measure and calculate this information on my own so I was wondering if any other statistic nuts/long-time medics could take the time to tell me exactly how uber splitting affects the rate of uber draining. It's kind of hard to explain these scenarios, but I'll try my best. Thanks for your patience.

here are some pretty basic stats I'll establish just to make this thread easier to communicate in

let's say a "fat" uber drains at 100% rate. An uber split across 2 players would be 150% rate, and 3 players would be 225%.

Does the game just count how many different players you've touched during an uber and drain at the rate dictated by that number, or does it constantly check the current number of ubered players and modify the drain rate throughout the uber? And if you flash someone, then return to them later after their flash wears off, do they count as a third uber recipient and bring the rate to 225% or does the game remember that they've been ubered before?

that seems kind of hard to read, so to provide an example let's take a hypothetical scenario: If you pop on your pocket, flash the roamer, and return to your pocket, when the flash wears off of the roamer, does your uber continue draining at 150%? If so, putting the beam back on the roamer later on would have no downside, right? Also what if instead you give the beam to a third player? Does your uber now drain at 225% even though you never had more than 2 people simultaneously ubered?

tbh I was pretty much tunnelling on cookye's play. But even with the push he didn't get much support at all. And the enemy combo just completely turned their back on cookye like he wasn't going to do anything.

his DM is pretty crazy in clutch situations. Like that spire backcap, wow. But he gets so greedy at times.

also 19:50? He goes in without team support, dodges for a while doing decent damage, and then everyone he was shooting simultaneously forgets about him, kritzes into mid, and lets their med get killed??? I am genuinely baffled.

entire thread filled with compliments
all compliments -fragged
tf.tv, everyone.

I've never been lucid during this but I've been told that I used to fairly often get up in the middle of the night and wander around speaking gibberish that sounds like farsi.

seems like it'd struggle to find any players who aren't already busy with ESEA and UGC.

maybe they could do an ultiduo tourney/ladder? as far as I know, that's pretty much an untapped market in NA and I for one would love to get into that.

link fixed, thanks.

I'm pretty sure that the transparent viewmodel plugin .dll is only bannable because it can be used to increase your fov and get limited wallhacks.

I did. it's a legitimate opinion. I'm fine with it. I just thought this was cool and would appeal to some.

I don't really understand why the response for this is so negative. I expected people to think this was a good idea for the same reason people want transparent viewmodels.
Is the OP misleading on how much screen space it takes? here's how much of the screen
it actually takes up:
http://puu.sh/1d5Qz

or maybe I should change my marketting strategy and change the title to "QUAKE MOD FOR TF2"

some people sometimes forget what weapon they have out. and this would hypothetically work for any weapon you want, not just the grenade launcher.