Note: I wrote this in June 2000, shortly after the release of the 1.1.0.0 patch, the core of which was a rewrite of the Half-Life netcode, introducing lag compensation. It was written as a response to an article by TeaTime, a commentator at what was then Fort Bundy but has since been renamed to The Fort, a PlanetFortess-hosted fan site. Short extracts from TeaTime's article are included here for context. The text of this article has been slightly revised to make it more suitable as a standalone essay, but is broadly unchanged from the June 2000 original. I have added a few historical comments about how things have changed since this was written.
The Patch
The New Netcode
[The new netcode] was impatiently awaited by almost everybody. And I think most people are quite satisfied with it. [...] The class most profiting from the netcode is the sniper, followed closely by the HWG.
Overall, I found the new netcode a huge improvement on a modem connection. Latency was reduced, and everything felt much smoother.
Lag Compensation
The main "gameplay" change from the new netcode, appears to be the lag compensation, which provides a huge improvement in the effectiveness of hitscan weapons. This obviously has a big impact on those classes whose primary weapon is hitscan - the sniper and the HWG. It also has a far larger impact for players with high latency. This improved effectiveness is for one simple reason - you no longer have to lead your target.
The rocket launcher, nailgun and other projectile weapons have always required you to lead the target, due to the time-in-flight of the projectile. You get visual cues (the visible rocket/grenade/whatever) to help you do it, and the effect is mostly dependent on the range to the target. It's pretty predictable, and soon becomes second nature.
For hitscan weapons, however, there were no reliable visual cues, and the primary factor was your own latency, which could vary depending upon how much action was happening in your field-of-view. You had to shoot at where you estimated the server would think your target was when your fire signal arrived at the server (or something equally complicated).
It was unpredictable and surprisingly tricky. At 200+ Latency (which is a typical figure for casual players), it was quite difficult to hit _any_ moving target, let alone anything as fast as a scout or medic.
The revised netcode means that, for hitscan weapons, if the target is in your sights when you fire, you will probably score a hit. This is MUCH more intuitive and HPB friendly. Now, an HWG or sniper just has to point and shoot, without worrying about leading the target. This also affects other hitscan weapons, such as the engineer's super shotgun.
I suspect that the new improved Sniper and HWG are simply closer to the way they were always intended to be and that, on a LAN, they were always closer to this ideal than on a modem connection.
Conc Jumping
The more I play the more I ask myself why Valve bothered to remove the handheld jump in the first place. I'd bet my soul (if I'd had any to begin with :-)) that in one month most people will c-jump as easy as before.
I don't think so. I'm not really a c-jumper (for a start I rarely play scout or medic), but I have done it a few times. The handheld jump was ridiculously easy to do. Even I could do it fairly reliably online, with virtually no practice. Especially the water-launched ones.
The drop-the-grenade-and-run-over-it jump, on the other hand, is somewhat harder. It takes far more practice, so fewer people will bother. Another difference is that anyone watching the jumper had little warning that a handheld jump was about to happen - with the thrown grenade, the grenade is a hint, which gives nearby enemies a chance to interfere, e.g. by knocking the jumper around with rockets, or by physically obstructing the jump.
In the time since the patch, I have seen a lot fewer jumps being attempted, and a far higher failure rate, than before. It takes more skill, and even a good jumper messes it up quite often.
The change also rules out the mad rocket-grenade double-jump that soldiers used to be able to do (e.g. ramp room to attic on the well). The dropped-grenade jump will also be much harder to pull off under fire than the hand-held, which will render certain conc-jumps much less viable than they used to be (e.g. across the yard in Rock2).
I think this is a good thing - as Yahn [Bernier, of Valve Software] said in an interview a month back, conc jumps break maps. Epicentre suffers badly from this.
How Conc Jumping Breaks Maps
Consider the following example: in The Well, a good offensive soldier can be stopped, a lot of the time, by a single, equally skilled, defensive soldier. Put TWO skilled defenders in his way, and he has his work cut out, even if the defenders don't co-operate very well.
Pre-patch, a good offensive medic using handheld conc-jumps could only be stopped by 2 or 3 good defenders working as a team. And given the speed at which the medic could cross the level, it was only a matter of time before he could strike while one or more of the defenders were resupplying. If the defenders weren't good, or if they didn't work as a team, that one medic could slice through a 4 or 5 man defence with ease. Two CJ'ing attackers were a defender's nightmare.
In short, on several maps, the team with the most/best conc-jumpers won - a very small number of skilled players determined the result of the match. Equally skilled players who chose to play other classes had little impact on the result. This was not good for game balance.
After 1.1.0.0, conc-jumping is a lot harder to do. It takes more practice, and even good jumpers mess it up more often. This makes things a lot fairer.
[2002: conc jumps are still a problem, but it's still not as bad as it was right before the 1.1.0.0 patch]
The Heavy Weapons Guy
Faster and more deadly now. [...] But the question is if the increased power of the HWG lies more in the new netcode (faster spin-on, easier to lead) or the fact that he moves faster.
As per my comments above, I think the netcode is the main effect here. I have to say that I have not noticed any increase in speed for the HWG. I agree that they needed reining in a bit, so I'm not too traumatised to see them tweaked in 1.1.0.4 (I speak as a regular HWG player). I hope they're not completely neutered, though. [2002: they weren't]
Grenade Respawn Limits
In my unworthy opinion this is the change with highest impact on gameplay (given that those darn snipers let you get close enough to use your grens). Some people claim Valve did that to reduce spam (also a very popular opinion when they limited the mirvs in 1.0.1.6).
I can't speak for Valve, but the obvious effect, to me, is to allow map designers more control over how many grenades there are in a map. It would be difficult for a map designer to reduce spam if players were constantly spawning with 5-8 grenades each (and as for demomen before 1.0.1.6: Spam on legs). Reduce that to 2-4 grenades per respawn, and there's much less of an impact. 2fort certainly seems to be less of a spam-fest after the patch.
Let me give you my theory: Less type1 grens force you to use them more sparely and therefore smarter.
Indeed. It also encourages the major spam culprits to try pick up the grenade pack - which they may have to wait for - thereby delaying a spammer between attacks. The grenade timer as standard also encourages players to apply a little more thought when using grenades.
Offence vs. Defence
It has been stated several times that the sum of the changes favour [Defence] to [Offence]. Though I'm not completely comfortable with this statement I think it's safe to say that Offense will have to be better organized and working closer together than before to get through enemy [Defence].
It's not so much that the balance has been shifted from Offence to Defence, as that a particular style of offence (the solo conc-jumping medic) has had its teeth pulled - it's now much harder for a lone medic to scythe through a skilled 5-man def. A couple of good soldiers with a medic as flagrunner will still be formidable.
We have also seen a return of the kind of attack I haven't seen since the early days - the massed HWG charge. Yeah, they're slow, but the FIREPOWER!
Teamwork
I think it's safe to say that Offense will have to be better organized [...] I'm sure I will be immediately corrected by clan people stating that in clan matches teamwork is already excellent with [little] room for improvement.
I agree, you probably will be "corrected", but I think they're wrong. In my clan-match experience, clan team-work is improved to some degree by more efficient planning, but the only clans with _really_ good teamwork have other advantages - like all sitting in the same room so they can talk directly to each other, or other Voice Comms. None of these routes are available to casual players, and many clan players have to go without, too.
[2002: Voice Comms were introduced in a subsequent patch, and proved that voice comms help most if you're organised to start with. Public servers are still pretty chaotic.]
In a post on the PF forum I mentioned compromises clans would have to endure in favour of pub games. This IMO is one such compromise.
Actually, a whole set of compromises. But, yes, I agree. Public server players outnumber clan players. Probably by a huge number. Why should the many casual players, on modems, have to make sacrifices for a minority of fanatics? Having played both, I think I actually prefer a good Public server match to most of the clan matches I've played - there's more variety, and hence more fun.
Remember, people, we're supposed to be doing this for fun. An awful lot of players seem to forget this at times. The main reason I gave up playing at clan level was that I ceased to enjoy it. Most players simply do not have the time to commit themselves in the way many clans demand. I have a full time job, groceries to buy, laundry to do, dishes to wash, and something vaguely resembling a social life. I'm doing well if I can manage 4 hours of TFC a week, and 2 hours would be more typical.
In my experience, clans usually require at LEAST 3 hours a week, often a lot more. Don't forget to allow for time loitering in ICQ waiting for the server, pre-match mode, and tactical discussions via various channels (ICQ, E-mail, web fora, whatever). I suspect many clans require a good deal more time than that. This is the reason that many clan players behave like 15-year-olds; many of them are 15-year-olds - they're the only ones who can afford the time to play online as much as many clans ask.
Some of the changes in the patch have cramped the l33t 5|<illz of the fanatics, but many of them have raised the baseline - newbies can be more effective than they were pre-patch.
When I first played TFC (the week it came out), I was relatively inexperienced as an online gamer (though I'd been playing single player First-Person Shooter games since Wolfenstein 3D) so I followed the advice in the manual, played HWG, and parked myself in the ramp room on 2fort.
I hosed lots of attackers, made an effective contribution to the team, and had great fun in the process. Any newbie who tried that immediately before the 1.1.0.0 patch would have had rings run round them by LPB soldiers/medics. After the patch, that newbie stands a chance again.
Right now I feel people [are] overreacting to the changes without really evaluating how the changes will work out in the end.
There's a lot of that, especially within the community. I also think there are a _lot_ of (silent) casual players out there who are very happy with the changes.
It could also be sour grapes. Experienced LPBs, who have become used to charging into a public server, running rings around HWG Player(1) and laughing as the tracer fire hits the wall behind them, are slightly taken aback now that there is a real chance of Player(1) mowing them down with Assault Cannon fire, due to lag compensation.
I say : get used to it. The playing field has been levelled a bit. You want a game where your fast connection gives you more of an advantage than using your brain, go play Quake III Arena.
Personally, I welcome the changes. Things were stagnating badly - I was starting to get sick of the sight of 2fort, and the toe-to-toe slugging matches that it usually played host to. The class-balancing tweaks provide a kick up the backside which will hopefully give new playing styles a bit of encouragement, and the new maps provide some much-needed novelty.