These bug reports are all for "Max FPS for High Quality PCs".

cl_loadondemand_default 

no longer exists on the latest version of TF2 and causes a warning to be printed.

For some reason (I was previous on rhapsody's config) alt-tabbing out of TF2 while in the main menu causes music stuttering. EDIT: It was caused by the snd_mixahead buffer being too small. Better leave it at 100ms so it doesn't stutter while tabbed out.

This one is a minor one, but

mat_disable_d3d9ex

causes a warning on first launch (with an empty config.cfg) that a restart is needed to apply the change.

The "Enable facial features section" has convars set for disabling that aren't present in the enable section such as blink_duration. Here's the default values for this section:

// Enable facial features -- turning them on lowers framerate by 5ish%
r_eyes 1
r_flex 1
r_lod -1
r_rootlod 0
r_teeth 1
r_eyemove 0
r_eyeshift_x 0
r_eyeshift_y 0
r_eyeshift_z 0
r_eyesize 0
blink_duration 0.2 // Not in the convar list?

This is also happening in other sections like shadows (nb_shadow_dist) and gibs (cl_phys_timescale). I take it you expect the user to delete or comment out the entire "disabled" section and not just the ones that are set in the "enabled" section, right?

Maybe you should consider GitHub so people can file bug reports and see what's changed?

Sure is one hell of a coincidence you and I have posted similar threads announcing high-level pubs within hours of each other :)

I'd like to talk to you privately about joining forces. Please friend me on Steam! I'll be up after 2PM EST.

When it came to cp_coldfront variants, I went with what I thought was the "latest" version. I can switch to cp_warmfront if you think that's better. Not too big on the joke maps... I could always make this a quickplay server if I wanted it populated (is that a given?)

I would like this server to be the place the best pubstars and comp players go to, and to keep things balanced at high levels I felt like the whitelist was the right way to go. They're used by leagues for a reason, no?

I remember Reddit East had the pregame mayhem thing going on their server... I could turn it down to a minute. That's more reasonable. I thought people would love the pregame dm. I was even thinking of expanding it to be a dm server if the population is 6 players or fewer.

Still going to keep TFTrue, but I can keep the FOV at vanilla.

Also about money... don't worry about it. This is my personal server for other stuff besides games, too. It costs me a whopping $10/mo to keep going. It's hosted on DigitalOcean.

I haven't touched the big blob of custom maps yet since May, I should probably update those too.

:(

But yeah, again, would doing quickplay keep it populated, even though I'm limited to vanilla maps?

Star_'s server is living proof that the formula works. In fact, the reason I even decided to try this is because of how disappointing that server can be. It has connection issues (despite my often single-digit ping), it's missing some essential features (like votekick).

I guess its success is thanks to Star_ being so popular among the community?

EDIT GROUP LINK CLICKY: http://steamcommunity.com/groups/athairus

I've been tinkering with a server since around May. Here are some of the biggest features:

• 24 slot, No crits/spread
  
• Pro map variants: cp_warmerfront_rc2, cp_prolane, koth_product, pl_badwater_pro_v8, pl_barnblitz_pro6
  
• Tons and tons of custom maps to choose from + map rating
  
• Current mapcycle (subject to change): http://pastebin.com/4BrQ8Rcx
  
• UGC s17 whitelist
  
• Full array of vote options (mute, kick, ban, scramble, map voting) all done via native votes so as to not block slots 1 and 2 during gameplay
  
• 1 minute "Pregame mayhem": SOAP DM during the extended "Waiting for Players" phase
  
• TFTrue (restore score on reconnect, whitelist)
  
• SourceTV with demos downloadable from site, replay support
  
• Issue tracker on GitHub

If this sounds appealing to you, then I need your help. Do any of you have ideas as to how to avoid the kiss of death new servers experience? Here are some of my ideas:

• Run biweekly events via the server's community page. Do users get notified an event is starting? Any tips for keeping people from chickening out when they see the server not full yet?
  
• Get a community going.

The server is currently live (and passworded), but half-finished. If there's interest, I'll continue.

Looking forward to 16:10 support so I can actually use this!

So the conversation has really shifted to this: it's less the game (with class composition and weapon balance) but the players themselves that make a pub balanced. The reason? Good communication and teamplay makes for a balanced game.

TBH this just seems like an idealisitc way of looking at it. Yeah, it'd be wonderful if we could have a set of good people occupying each and every slot on the server, with no major social issues, communicating and working as a team to make the game fun. But here's the problem: is such an expectation realistic?

Can such a server exist today (and by exist I mean be consistently full)? If so, how?

Wareya, I don't like class limits. I've encountered situations where class limits have really gotten in the way of helping my team. What if the class slots for class ___ are taken by horrible players, and the fact good players didn't get the chance to play those classes sometimes seals that team's fate, like if the team NEEDED a good ___ to counter whatever the enemy team's got. I've also had the situation where I wanted to quickly switch to engi (or pyro) to kill a sapper at spawn, but I've been blocked. That's bad. I think class limits are a ham-fisted solution to the problem.

EDIT: whatever that problem is (??)

> but every classlimit has to be the same because every map and gamemode is different. restricting spies and snipers lower than pyros and heavies are can totally stack A/D maps for example.
uh...

> if you want to go the route of blocking items, only pick ones that are legitimately overpowered
I agree. For me, minis are first on the chopping block. A gunslinger with a mini who's really, really, good with the shotty is OP IMO, particularly vs scout. What about you guys? Can you think of others, or would you just advocate a full comp whitelist like someone suggested earlier?

> don't bother with excessive scrambles
The old reddit east did this, and I loved it. It kept things fair. In addition to aggressive scrambling, it also scrambled automatically at the beginning of the map, as some people pick a particular team by habit.

> don't bother with intrusive autobalance
That IS annoying. FirePowered servers have the option to let anyone on the opposing team volunteer to move to the other team for some FP funbucks. That's an improvement, though it still falls back to just normal autobalance.

> run maps with low long term inertia where solo players can shine
Low long term interia? Anyway, I'd like to hear everyone's ideas on how to encourage team play further.

The idea of a reserved slot perk for medics is great! That is definitely a legitimate reason to have them. Play medic for one full round (and win?), gain a reserved slot for the entire day (and the next?). How does that sound, guys?

Honestly, this is a valid concern! People are so downvote-happy around here...
There are pro versions of a lot of Valve maps, which add additional flanking options. Pubs should really be using them more. Recently a pro version of pl_badwater came onto the scene, and it has a lot of very welcome changes.

I've seen that happen plenty of times, too. Very annoying. One way to ensure people can't stack teams is to prevent them from stacking teams :)

Have an aggressive scramble plugin (trigger happy, scrambles at beginning of map) that takes into account "skill". I've seen skialbalance in action myself. Any idea how it works? Aren't all sourcemod plugins supposed to be open-source?

How would you guys determine "skill"? I'd use factors such as total TF2 hours played (somehow having to compensate for idle time?), average score/min which will be weighted based on which class the player is playing when it's time to update it (classes that generate less points/min on average will be boosted to compensate). Bonus for that last one: friendlies and idlers will have their skill rating lowered :)

Now, how about encouraging medic play? Forcing a lucky winner to play medic seems... a bit heavy-handed. I was thinking of perhaps some incentive system being better. But honestly, without seeming like just another junk server with "perks", what can I do besides end-of-round immunity for the map for those who play medic a whole round? And how about discouraging battle medic play? And how about discouraging it without discouraging ubersaw chains (or should I?)

In addition, how about a reminder upon team loss that your team did NOT have a medic? I bet that would gently prod someone on the fence to go medic next round!

I understand. I still want to pose the question because I'd love to see comp players fall in love with pubs again!

Great idea!

I'm working on a pretty big writeup on this subject, but I'd like to put this question out there for you guys to talk about now.

As I play tf2 more and more, I tend to play on servers that tend to have very skilled players. These servers tend to be imbalanced, though. Teams without medics. Teams with more snipers and spies than anything else. Games that, despite multiple scrambles, are consistently skewed towards one team. What kinds of imbalance do you guys encounter in pubs?

I have all kinds of ideas as to how to balance them, but I'd like to hear yours!

What things out there that few (if any) servers use that would help alleviate this problem?
What things out there that seem to be made to solve these problems do you feel not work well?
What things out there that don't even exist yet do you think would help?

Yeah. That's been removed. It's as simple as (on Windows) dragging the .sp file into compile.exe inside addons/sourcemod/scripting, then digging the .smx file out of the "compiled" folder and putting it where the plugins usually go.

I'd say it helped a good amount. People have been saying "if you use this, it won't help because you'll be focusing on the cursor". To them I say... no?

You'll aim/flick at the cursor, but you're pretty focused on the enemy's position because that's where the cursor will be near. Honestly, it's all on the person doing the training to use this properly.

Not quite yet. I don't want people to use it until I feel it's ready, and I'll update OP and the cfg file itself to say that.

I really wish I could edit the thread's title.

EDIT: I added a special tl;dr version for those who don't care about the explanations and just want the important stuff.
wareya, I added the cl_interp equivalents to all aliases.

Sorry about bringing up your guide that was not really meant to be a guide but an argument-stopper. But since that's not meant to be definitive, I definitely want to develop my cfg file further so it'll be useful for those ***whose jitter is less than 10ms 99% of the time***.

> Of course, that's actual effort.

The subtle little jabs you've added to your comments in this thread don't help to keep things constructive. I'm not claiming that I'm innocent of them, but let's both just take it down a notch, huh?

> Considering that the explanations give the impression of being necessarily correct despite giving misadvice like ...

Something I missed when I was trimming earlier based on what I learned so far. It's gone.

> not to mention that they're ridiculously long and full of jargon
> Or you could put out a networking essay

Why not both. Handy to have explanations in the same file as you edit.

> This literally retorts nothing that I've said about how telling people to "base their settings on their jitter" is useless. If anything it only compounds how ridiculous it is to tell people to bother doing so.

Here's the way I see it: play on your favorite servers, notice how high the jitter goes
Server 1: 4ms
Server 2: 1.5ms
Server 3: 5ms
etc.

Take the max, add a little extra (or double it, why not) and that sounds like an entirely reasonable value. Do you have a better alternative?