Total logs seen: 2761285
Sixes overall data:
red:
Wins: 687879/1505856 = 0.45680264248374347
Loss: 702667/1505856 = 0.4666229705894853
Ties: 115310/1505856 = 0.07657438692677122
blu:
Wins: 702667/1505856 = 0.4666229705894853
Loss: 687879/1505856 = 0.45680264248374347
Ties: 115310/1505856 = 0.07657438692677122
By map:
process
red:
Wins: 211045/472281 = 0.44686320220377274
Loss: 221659/472281 = 0.46933711074550954
Ties: 39577/472281 = 0.08379968705071769
blu:
Wins: 221659/472281 = 0.46933711074550954
Loss: 211045/472281 = 0.44686320220377274
Ties: 39577/472281 = 0.08379968705071769
((product)|(viaduct))
red:
Wins: 144088/293807 = 0.49041717862406276
Loss: 144784/293807 = 0.4927860806583914
Ties: 4935/293807 = 0.01679674071754587
blu:
Wins: 144784/293807 = 0.4927860806583914
Loss: 144088/293807 = 0.49041717862406276
Ties: 4935/293807 = 0.01679674071754587
((snake)|(snack))
red:
Wins: 121982/269120 = 0.453262485136742
Loss: 124347/269120 = 0.4620503864447087
Ties: 22791/269120 = 0.08468712841854935
blu:
Wins: 124347/269120 = 0.4620503864447087
Loss: 121982/269120 = 0.453262485136742
Ties: 22791/269120 = 0.08468712841854935
ultiduo_baloo
red:
Wins: 77280/162308 = 0.47613179880227713
Loss: 81457/162308 = 0.5018668211055524
Ties: 3571/162308 = 0.022001380092170442
blu:
Wins: 81457/162308 = 0.5018668211055524
Loss: 77280/162308 = 0.47613179880227713
Ties: 3571/162308 = 0.022001380092170442
villa
red:
Wins: 2089/4807 = 0.43457457873933847
Loss: 2246/4807 = 0.4672352818805908
Ties: 472/4807 = 0.09819013938007073
blu:
Wins: 2246/4807 = 0.4672352818805908
Loss: 2089/4807 = 0.43457457873933847
Ties: 472/4807 = 0.09819013938007073

https://clips.twitch.tv/CrypticWimpyRuffOMGScoots

Patched as of August 21st; server operators can change sv_maxupdaterate if they want old behavior.

https://www.teamfortress.tv/10097/when-is-it-time-to-quit

Remove CPU cooler + GPU (and any other PCIE devices), probably also hard disks/disk drives/anything else shock sensitive

Pack all of that into your carryon along with peripherals. Computer case (with motherboard and other remaining 'secure' hardware) goes into a large suitcase packed with clothing.

Can also ship the case in its original box/padding ahead of time, but that's more expensive than checking it in my experience.

Source: I've done this >15 times in the last 4 years.

Immediately discarding reero's video, calling everything a placebo, and immediately doubling down when contested were all very much not "trying to figure things out." I don't doubt that mastercoms has a very good understanding of how the game works, but in this case she was wrong and it's massively harmful for a trusted community figure to double down on something that's simply incorrect.

In this case the result of playing with zero lerp was well known by the jumping community for a long time, regardless of how well understood the mechanism was. I'm only trying to make very clear that there is a real benefit to sub-15.2ms lerp.

The patch is not necessarily a bad thing; it would be strictly better to have servers run a higher sv_maxupdaterate than to rely on the exploit- this seems to fix some of the jitter from the exploit. However, it's important that we respond to this patch by doing that instead of just leaving the default.

It's worth figuring out what the actual downsides to very high sv_maxupdaterate settings are; sub-1ms lerp requires around 15x the default so it's worth checking that it doesn't harmfully increase server load or allow for any funny hitreg sillyness.

Your "specific test cases" consist of shooting pipes at a wall in itemtest. You have made absolutely zero attempt to argue in good faith, and you're acting as though a well understood and commonly used trick is "just a placebo" because you personally don't understand it (despite making zero effort to do so!).

Here's your sanitized testcase: https://www.youtube.com/watch?v=5Lih1aNxDaE

It's a very real difference and you're making yourself look like an idiot to a very large group of people.

The higher your ping, the more projectiles appear behind where they actually are. For soldier, this is mostly something that you can get used to providing that your ping is stable. For demo, it's a massive difference since 5 ping -> 30 ping is enough to make it so you don't see stickies when airpogoing at all, which makes it ridiculously harder.

For the purpose of seeing projectiles accurately, lerp is essentially additional latency, so 0 lerp helps a lot. Thirdperson is actually a cheat, but it's tolerated since it levels the playing field for people with 60 ping who otherwise don't really have a chance.

yes i cant wait to play on all the valve, community, and competitive servers which will surely respond to this update by setting sv_maxupdaterate 99999 rather than leaving it at the default

https://youtu.be/TlOQplgue1Y?t=2829

The reason that phase walls can be shot with rockets but don't collide with players is because the operative part of the wall is a func_physbox_multiplayer.

The reason that explosions radiate through phase walls is because the texture .vmt (usually Metal/metalgrate011a) has

"%compilepassbullets" 1

Each "phase wall" is usually a physbox with physics disabled, textured with metalgrate011a, with render mode set to none. On top of that physbox is a func_illusory with the desired texture.

The physbox is both the source of many hammer problems and completely irrelevant to any uses in 6s maps since there isn't much need for walls that block rockets but not players.

It might be fun to play around with walls that don't block explosions, but they would need to be very clearly indicated since players wouldn't expect it.

Also clipping stairs with blockbullets (or just making actual ramps, adding rendermode:none ramps, etc) for stairs is much simpler, easier, and is already done on pretty much all maps for a few years. If you notice rockets getting annoyingly eaten in any actively developed maps, you should report it to the dev and they'll probably fix it.

11:41 PM - warm rat: http://logs.tf/2609732 div 1 match

https://www.gamestop.com/video-games/retro-gaming/gamecube/accessories/products/nintendo-gamecube-controller-assortment/10131394.html

I've bought >10 from here and all but one have been perfectly usable.

If you want to try the game, any controller you can use with Windows will work fine. If you want to learn the game at all seriously, get a GCC.

(if you want massive unfair advantages coupled with a few awkward downsides and a weird learning curve, pay hax $200 for his all-digital controller)

If you can't find such a button, find the coin battery on it and take that out for ~1 minute with the computer unplugged; it will do the same thing.

mat_trilinear 1