Harsh observation of things that will hold this game back

I've noticed some serious issues with this project that keep it from actually getting to where it needs. No, I'm not talking about the community or anything like that, I'm talking about the game itself. Below is a list of the major things I see holding this back and keeping it from being able to actually compete with other arena styled shooters that are about to hit the streets.


MAJOR issues need to be fixed and problematic aspects (such as unpolished or incomplete things) need to be swept under the rug come 1.0
----------------

Things under this category that I know REALLY need to be taken care of
---------
-CAMPAIGN MODE AND LACK OF PROPER TUTORIAL, NOT TOO MENTION USELESS BOTS. Currently as is, the campaign mode serves (poorly) as an offline training mode of sorts that pits inexperienced players against brain-dead bots that will frustrate those that don't know what they're doing (which is most newbies since this isn't a traditional FPS) and bore the hell out of those that have grasped the game mechanics. I'd say ditch the campaign mode entirely and go to work creating a somewhat lengthy training mode that teaches the player basic game mechanics as well as your more advanced techniques such as weapon combos. Said training mode could have multiple levels to it, ranging from a couple of simple tutorials (much like Nex 2.5 had) to having challenges that test a player's ability to conquer game mechanics (level speed runs, advanced weapon handling, a target practice map with bots running routes recorded by players and developers, etc.). Having such a training mode instead of the campaign that currently exists would remove one really bad impression that a lot of newbies get when they first pick this game up and remove the need to actually do anything about the bots (which will likely never happen, else something would've been done within the 7 or 8 years I've been playing).

-VEHICULAR ASPECT OF THE GAME. Potential, potential, and guess what? More potential. Right now the vehicular aspect of this game has everything that StarSiege: Tribes and Unreal Tournament 2004 should've had. HOWEVER the last I checked there was some serious need of content maps wise and the CTF aspect of it is just screaming for something Tribes-like (hint hint, it burns recharging fuel and allows you to go really high off of a jump). There's also the issue of actually balancing the damn things, but that doesn't seem like it will be as difficult as the other stuff related to this, especially given how much more satsifying these things are than what Tribes and UT2k4 has to offer combined with the fact that they are far from useless in this game.

-WEAPON SYSTEM AND AMMO DISTRIBUTION. To my knowledge this is getting worked on, but it still needs to stay at the attention of the developers. Currently as is, the machine could use a replacement weapon and the ammo distribution for all the guns is out of whack (two core weapons have their own ammo and the other two ammo types have three or so guns to them)

-WEAPONS AND VEHICLE COCKPITS GET VARYING LEVELS OF DISTORTION DEPENDING ON WHAT THE PLAYER FIELD OF VIEW IS SET TO. This is a serious problem, as it makes vehicles near impossible to use in first person (which is how they should be piloted in my opinion) when the FOV is jacked up to the levels needed make this game playable for most people. The weapon models and cockpits should look exactly the same at 90 fov as they do at 130.

-DEFAULT FIELD OF VIEW WILL GET YOU WRECKED. The default 90 fov in this game feels closer to the 65 that Call of friggin Duty given the pacing of the game. What's really bad about this is guess what? Most newcomers aren't going to look at it.

-OFFSET AND RELIEF MAPPING CAUSE ARTIFACTS IN TEXTURES AND MASSIVE PERFORMANCE DROPS. Relief mapping is a pretty neat effect that understandably eats up a bit of performance, HOWEVER it seems to have a nasty habit of blurring things and making a lot of stuff almost look like it's basically just a bunch of textures printed out onto something and stacked one on top another. Offset mapping by itself is very similar except it doesn't have nearly as kick ass of a 3D effect and still suffers from artifacts, thus making it pointless. With that said, these options, as neat as they are, should just be hidden from the user and kept in the advanced options until corrections are made to them. To my knowledge, these were hack fixed by master[mind] like 4 years ago (thread linked below), but the fixes weren't accepted as part of the game due to it actually being that assets that were faulty. Still, the code tweak should be in place at least until the material is fixed.

-ISSUE WITH WATER REFLECTIONS. The way reflections are implemented causes the player to see a reflection of themself (as well as everything around them) under water if they look up. This causes obvious visibility problems and just looks out of place as well. In my opinion, there needs to be to an implementation that disables the water reflections whenever the player's view goes under water.



Not necessary, but nice (and easy) touches that can be made before 1.0. These are mostly tweaking the defaults for configuration settings that are only accessible through the console or advanced options (beneficial to deal with because most players don't snoop around and see what's really available to them).
----------------

Things under this category I recommend:
---------
-Setting cl_decals_time to 1 and tying the decal time option in the menu to cl_decals_fadetime. This allows for a more visually subtle way of removing decals.
-Setting cl_projectiles sloppy to 1 since it increases performance at no noticeable visual trade off
-Changing cl_gibs_lifetime to something (I recommend 2) that allows the gibs fly out but also disappear before they can sit still (removes unnecessary visual clutter and keeps the player from noticing any flaws that may exist in them)
-Changing r_water_refractdistort to something that better shows off it's effect, but without screwing with visibility (I recommend 0.019)
-Tweaking the motion blur settings to something isn't ugly and obnoxious (I recommend the following: r_motionblur 0.007000, r_motionblur_maxblur 2, r_motionblur_bmin -0.5, r_motionblur_randomize 0) and then ditching the slider for a checkbox
-Setting cl_playerdetailreduction to 5 since it decently increases game performance while not really being noticeable (at a stand still, I couldn't notice a visual difference)
-Adding a few typically user-defined binds to the input menu as defaults, in particular toggle chase_active (bound to x by default in Tribes 4) and ADD A BIND FOR USE SO THAT WE CAN EXIT VEHICLES.



master[mind]'s relief mapping thread from four years ago:
http://forums.xonotic.org/showthread.php?tid=1271

The config changes look good, I'll make a branch for them after a bit of messing around with values.
The default bind for use (vehicle exit) is F by the way.

As for actual code issues: these are all huge projects, and developers are spread very thin (i find myself working all day and not getting very far lately).
But, I do think some of these need doing before 1.0.

Rebalancing the vehicles will be a near future project of mine (probably after 0.8 and the update madness), along with making some of the gamemodes more stable (assault, clan arena, last man standing...)

The campaign is indeed in need of a major overhaul, perhaps we could have a mapping contest for this? As far as I can tell, not much actual code would be required for something good.
The bot AI is something that has needed rewriting since they were added, but no-one has time to do this, unfortunately.

Someone with engine knowhow would need to take a look at the cockpit FOV issue, as it would need to be fixed there.

I will also take a look at master[mind]'s efforts and maybe include them if they are truly an improvement.

I come to conclusion that non of this matters. What needs to be done is simply start advertise the game and get a initial pub scene (preferably CA, CTF and FFA since that actually lets new players improve in contrast to minsta where they will never learn the fundamentals). When this initial pub scene is established the chances of new players will be more likely, since new people actually can join a server and start playing. What makes people learn the most is actually playing the game after all however the good tutorial can be made.

I come to conclusion that non of this matters. What needs to be done is simply start advertise the game and get a initial pub scene (preferably CA, CTF and FFA since that actually lets new players improve in contrast to minsta where they will never learn the fundamentals). When this initial pub scene is established the chances of new players will be more likely, since new people actually can join a server and start playing. What makes people learn the most is actually playing the game after all however the good tutorial can be made.

There has been many discussions about advertising the game, most of them coming to the conclusion that the game isn't ready for the masses, and we would only get one shot at first impressions, then the game becomes another of the many failed FPS attempts coming out lately.

But... at the current rate of core development (that is, the important stuff that needs fixing and improving), things might not change much from now to release, unless magic happens and developers start returning, now that Samual is effectively out of the picture.

Added something I forgot to list in the initial post.

I come to conclusion that non of this matters. What needs to be done is simply start advertise the game and get a initial pub scene (preferably CA, CTF and FFA since that actually lets new players improve in contrast to minsta where they will never learn the fundamentals). When this initial pub scene is established the chances of new players will be more likely, since new people actually can join a server and start playing. What makes people learn the most is actually playing the game after all however the good tutorial can be made.

There has been many discussions about advertising the game, most of them coming to the conclusion that the game isn't ready for the masses, and we would only get one shot at first impressions, then the game becomes another of the many failed FPS attempts coming out lately.

Exactly, there's too many rough edges to advertise this as a product that can put up a fight with the other games that are out there. At it's core it already does, but there's a lot of things that need to either be fixed (ammo distribution and weapons, weapon model and cockpit issue) or cut out entirely (bot mode campaign for example should just be ditched) to prevent players from simply dismissing it. Honestly it would be better to wait another two years or so until this is ready before trying to pit it head to head with Unreal Tournament 2015 and Toxikk. By then the game will be where it needs to be as a product that can hold it's own, will likely run on just about everything with the settings maxed, and will have been released when people are thinking about leaving the lesser games released in the wave of arena-styled shooters. Not only that, it's history (how it was originally Nexuiz and the fact that it was an underground game) would also be a major selling point. If it goes that long though, I would really recommend a re-launch of sorts.

But... at the current rate of core development (that is, the important stuff that needs fixing and improving), things might not change much from now to release, unless magic happens and developers start returning, now that Samual is effectively out of the picture.

This is the biggest problem with the project is lack of developers who have time to work on it as well lack of anybody who is able to learn everything needed to help out. There's also a serious lack of activity from the developers on this forum as well I've noticed, which has me worried about the state of the game. Pretty sad to see this after being around since Nexuiz 2.3 and seeing the amount of activity that existed on the old forums not to mention the "We'll show them!" attitude that existed here when the initial fork happened. It's like everyone is losing all their fight.

Oh yes the fov option should DEFINETELY go somewhere where you can directly see it.I have been playing xon about 1 year with fov 90 until i found the fov slider.The menu in general is a lil bit complex and mysterius for beginners.

Let's not forget the whole gods vs noobs division of Xonotic's players.

Let's not forget the whole gods vs noobs division of Xonotic's players.

I'm thinking, for this problem, we can probably display the player's ranking on xonstat in the game so at least we know the level of the players we are against. This information can probably be used for in-game auto team balancing (the algorithm to be designed). We do need to consult the xonstat admins to see whether there are any load issues, so for now, just a thought.

That said, please look at this rough observation. Between 'noobs' and 'gods', we actually have all skill levels. It's not polarized, as people may perceive.

Another issue I noticed with this game is that people seem to hate variety in player models as well as custom player models. A lot. You see, Garry's Mod, Half Life 2 Deathmatch, and Left 4 Dead all have variety in player models as well as many custom player models. Yes, people still use them in those games even unto this very day. However, games that are open source (like Cube 2 Sauerbraten, Red Eclipse, and Xonotic) almost never get any custom player models nor is variety in player models welcomed in them. I guess in some cases people fear that if custom player models based on popular video game characters are made and uploaded, companies like Nintendo, Sega, and EA might get mad and sue and shut down whatever open source games that they were made to be used in as custom content. Never forget the old saying that states that business is war.

Another issue I noticed with this game is that people seem to hate variety in player models as well as custom player models. A lot.

Some of my favorite servers are full of ponies and mario-esque quake grunts.

However, games that are open source [...] almost never get any custom player models nor is variety in player models welcomed in them.

That's complete nonsense. On popular servers like Jeff's, there are countless custom models like MLP, Mario and Yoshi, Homer Simpson, half-naked ladies, androids and so on. You can even wear hats and masks.

However, games that are open source [...] almost never get any custom player models nor is variety in player models welcomed in them.

That's complete nonsense. On popular servers like Jeff's, there are countless custom models like MLP, Mario and Yoshi, Homer Simpson, half-naked ladies, androids and so on. You can even wear hats and masks.

Yeah, but that's just his servers. The SMB servers used to have custom player models as well, but I was told that people complained about them, so they were removed. Now, everybody is Erebus on them, and that is not very fun.

It's not about fun. Most people forces the enemy models to Mega Erebus for visibility reasons, not because it's exceptionally cool or anything, although there is ofc people who think that way too.

I've never seen a 'Pro vs noob' flame war ever in xonotic. I haven't been around as long as some of the other guys but usually Xonotic is very friendly at all skill levels.

Xonstats is super cool. I love being able to look at old games and what not and track my progression. Not many other Arena FPS games have an integrated system like that. Besides, no one really cares that much about elo here. Its all good and fun.

As for the engine, I can't really comment on that other than the game feels great to me 3600 games in and I have few compliants. I just think Xonotic needs big support and media outreach and we can do very well.

All ranking systems have their flaws. Surprise, surprise! You shouldn't take any of them too seriously, including my own! I built stats so *individuals* can track their progress and improve, and I think it serves that purpose just fine. Along the way I found it more useful than not to also include Elo ranking, but that doesn't mean it is perfect. I'm completely fine with the warts XonStat has today (and yes, there are many of them).

The ancient engine is holding both development and new players back. The pipeline for creating maps and content is awfully oudated, full of constrains. And there are lots of problems with gameplay, netcode and such.

Asset creation pipeline is really atrocious. Stuff that takes 30min in unity3d or 45 min in unreal4 takes sometimes 5h in darkplaces and still something is going to not work right the way you want it.

Oh yes the fov option should DEFINETELY go somewhere where you can directly see it.I have been playing xon about 1 year with fov 90 until i found the fov slider.The menu in general is a lil bit complex and mysterius for beginners.

That's just sick.. I personally found the sliding bar after 2min when I first installed Xonotic (ran Xonotic).

It took me 10 seconds to think and write "fov 105" to console, it's a common console command among Quake-engine related games.

I've never seen a 'Pro vs noob' flame war ever in xonotic. I haven't been around as long as some of the other guys but usually Xonotic is very friendly at all skill levels.

There used to be a "War" against "Pros vs Noobs" back long time ago during 0.6.