It was a strange feeling being involed in the early matchmaking beta. At first, Valve seemed to be working based on the feedback that was given by the beta-players (mostly top-level players) by listening through players' streams, but as time went on there were less and less updates to the system, even though it had some pretty big structural flaws. Suddenly, the matchmaking system was released to the public, without any big improvements to the system. I suspect this release was done because players gradually stopped beta-testing. But on the other hand the beta-testing had already been done and people were waiting for the feedback to be implemented. Of course, having no direct communication between the players and the developers makes this a lot harder to communicate.

One of the reasons why matchmaking failed is because it restricted a lot of settings and forced players on higher quality settings. It was theorised back then that having players stream on higher quality settings would make competitive TF2 more watchable, and bring in more viewers. But on the other hand, playing TF2 with high settings causes a lot of FPS issues on older hardware, and generally having the game look nicer can negatively affect gameplay. So there are pros and cons to this, but it seems that a lot of players weren't willing to migrate to matchmaking and this was one of the reasons.

Another reason was the ruleset. Valve did not like any of the competitive rulesets, so they had their own version, simply a best of three on 5cp maps. I have no idea why this ruleset was implemented, it doesn't seem like removing the timer would solve stalemates (assuming they would even happen that much in a matchmaking environment), but it also made games either really short or really long. A game could be 3 minutes or 40 minutes long with that ruleset. Also, since most of the playerbase was already playing their own ruleset, playing matchmaking with a completely different ruleset and no whitelist was simply not worth it because of how different it was.

Soon after the release of matchmaking, it was apparently riddled with cheaters. TF2 didn't have the overwatch tool that CS:GO had to weed out the most blatant hackers. So these people could cheat mostly without consequences, which obviously completely destroys people's experiences. I think this was the biggest factor why it flawed even with the public players who didn't previously play competitively.

It's not a pretty story. It shows how difficult it really is to create a matchmaking system, especially in the environment TF2 had. I think matchmaking could have been successful in TF2, but I think it would have required more open-mindedness and communication from Valve in the early phases in beta testing. I don't want to be overly critical of Valve because the situation was difficult to begin with, so I guess my attitude is, "at least they tried".

Apologies if I got some things wrong, this was three years ago so I don't remember things exactly how they were.

I already knew the answer and have for at least 6 years. The only reason I madd the video is to keep you from spreading misinformation. Perfect example of the bullshit asymmetry principle.

Don't worry, I made a video so you can more easily verify the results!

https://youtu.be/qU7D8JCP9CI

Magnitude of velocity in 0 gravity after getting hit by 50 damage, hitscan, uncrouched: 300 u/s
Magnitude of velocity in 0 gravity after getting hit by 50 damage, hitscan, crouched: ~400 u/s

Proof of experiment: https://www.dropbox.com/s/8rsessqs2yj9b4d/knockbacktest.dem?dl=0

Conclusion: Crouching or staying uncrouched does have an effect on the knockback you receive.

A rigorous way to test knockback would be to set sv_gravity to exactly 0 and get shot while you are in mid-air. Write down the damage you took, the magnitude of velocity on all axes after you got hit (cl_showpos 1) and whether you were crouched or uncrouched. Velocity doesn't change after you get hit in zero gravity unless you airstrafe or collide with something, so it's fairly practical to measure.

You have simply misunderstood the code. There is an extremely noticable difference from crouched knockback and uncrouched knockback. Everyone knows that crouched rocket jumping has far greater knockback than uncrouched rocket jumping in any given situation.

Staying uncrouched while rocket jumping makes your hitbox 33% bigger, so it does make you easier to hit while in the air. If you bomb correctly you are going to have so much speed that the additional knockback you get from being crouched is going to be negligible. You can also use the knockback to your advantage a lot of the times.

Unfortunately you haven't listed the main roster of SVIFT. Please fix this sviftly.

unfortunately that bastard izon went to a party instead of attending important team meetings, as such he will be replaced with dennia effective immediately

http://etf2l.org/teams/31115/

saints comeback confirmed, rest of team moved to backup

Jesus christ, you must have built up quite a lot of resentment after all of those seasons in Highlander you ended up finishing 2nd place.

It's kind of ironic that someone who pereptually finishes 2nd would be defending banning the best team.

Computer engineering

This reeks of entitlement issues. He has the right to pull out for any reason, without an explanation. It's really that simple. It's not a punishment, it's a revocation of privilege.

cl_updaterate 66
cl_cmdrate 65 

cl_interp_ratio 1
cl_interp .0152 (a value of 0 would be fine as well)

cl_interp_ratio 2 (possibly higher)
cl_interp .0228 (possibly higher but this should work fine)

cl_interp_ratio 3
cl_interp .0152

net_compresspackets 1
net_compresspackets_minsize 128
rate 196608 // 1.57 Mbps default rate from CS:GO
net_splitpacket_maxrate 196608
net_splitrate 1 // How many packets to split per frame
net_maxcleartime 0.020346

net_maxroutable 1200
net_maxfragments 1200

Correct me if I'm wrong but isn't interpolation related to packet loss, and not ping?

https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcSCL13I1JLUPlesKJSNsxIvnnwRfYTt4mQcFxGIyYZxjRBSKcgNUA

You can twist the 3.5mm plug that goes into this thing to untangle it.