various suggestions

Request new features or present your ideas.
Post Reply
User avatar
Viech
Project Head
Posts: 2139
Joined: Fri Aug 03, 2012 11:50 pm UTC
Location: Berlin

Re: various suggestions

Post by Viech »

freem wrote:

* alien's wall climbing is hard to master. It could be made easier by having an indication about where is the roof (or the opposite, I do not know the name... earth? land? not sure... but I hope you'll understand). A solution to fix that could be to add an indication on HUD to give the player that information, like an arrow

This has been proposed before and I still find it a good idea. Someone would need to do it, though (our programmers are rather busy).

freem wrote:

* the new building mode is really great

Thank you! :)

freem wrote:

But is also unbalanced.

It probably is, testing and fine tuning requires us to observer a lot of non-bot gameplay. However, balancing isn't our main concern right now. If you think of balance all the time you are hindering yourself in trying out new approaches.

freem wrote:

** avoid aggressive building spam (it does not include barricades, since they are only useful if they are near on many maps) with minimal distance between them.

Such spam is already a very bad idea since as soon as your enemy is on stage 2, they have access to grenades or adv. mara/goon and can do a lot of damage and get stage 3 easily. I see an issue though if stage 1 aliens have to fight a turtle base full of turrets. Not sure how to solve this yet, I guess aliens will need a stage 1 weapon for this.

freem wrote:

** reduce eggs cost: human repetitor cost only 4 and allow to build turrets. Total cost is 12. For aliens, egg+acid tube costs 18... Of course, alien's egg can allow them to spawn, but acid tubes are far weaker that turrets, even with regen. (in my opinion, alien's def need real spam to be of any use before level 2, unlike human's)

I wouldn't reduce egg cost but we might look into removing or changing creep/power requirements for building stuff.

freem wrote:

** improve egg's or leecher's efficiency and/or range so the higher cost of alien's outposts will not be as problematic as it currently is.

I wouldn't try to balance the teams by giving them different attributes in regards to resources or confidence. This will make further balancing very hard and will also just hide the underlying problem. I'm also not sure if there is any. Aliens get a strong forward with an egg and booster only while humans would need a repeater, a medistation, an armory and a node to get the same. (Not accounting for the fact that the egg also slows enemy, so humans would want to have a turret for better defense.)

freem wrote:

** reduce barricade's cost. Seriously... they are useful, but cost as many as a turret!!!

Don't underestimate the use of a barricade. Alien defense is based on slowing humans down, not just making them direct damage. A lot of barricades can, juts like turret spam, make the bases too resistant and take pace from the game. Apart from that, I did make them more useful already, since killing a barricade will give just 10 confidence points to the enemy. So if a barricade protects two acid tubes (15 points each) from a grenade attack and dies, the build points were really worth it.

freem wrote:

take a turret alone, and place an alien in front of it (a dretch or a granger). The alien might be able to destroy it (on training maps, I have destroyed some turrets as a granger) but it will take at least 5minutes.

Destroying a lonely turret with a basilisk or mara takes virtually no time and if you have good movement, you will take no damage at all. A single mara or goon can easily kill even two or three turrets in a few seconds if they are not in a very good layout.

freem wrote:

Now, compare the behavior of the alien with the one of a human (with a blaster or their basic weapon) in front of a barricade alone... The human will destroy alien buildings without a problem, when the alien will risk it's life for that. Distance weapon on a wall, against very weak weapon on an aggressive building, which have the same cost?

A base that isn't defended by someone will almost always die to a single good attacker. I do believe that aliens are the stronger base killers, only compensated by the fact that humans are more frequently in their base. Furthermore, the alien barricade eats a lot of amunition before it goes down, so while an alien regenerates both life and ammunition when attacking a lonely base, humans don't have enough ammo to kill an undefended main base in a single rush.

I'm aware that the game often seems unbalanced to new players, but you will be surprised how often you'll be changing your point of view and think the other team is really stronger. That's often a matter of perspective. My current point of view (for Tremulous GPP) is that aliens have the overall advantage in public games but I've been told by more experienced players that scrims are rather human biased (good aim and coordination makes humans strong there).

I'll comment on the rest later! Thank you already for your good feedback!

Responsible for: Arch Linux package & torrent distribution, Parpax (map), Chameleon (map texture editor), Sloth (material file generator), gameplay design & programming, artistic direction

User avatar
Viech
Project Head
Posts: 2139
Joined: Fri Aug 03, 2012 11:50 pm UTC
Location: Berlin

Re: various suggestions

Post by Viech »

freem wrote:

* build leechers/diggers (not sure about human's building) at startup please! Bots will never build them (not a real problem honestly) and players will have to loose time for the first one...

Mappers have the option to do so but I believe requiring some effort to get a good base to start with is more fun than having a solid base right from the start. If you feel risky, you can attack the enemy at the beginning of a match and fall back to building if you're pushed back and realize you will need more resources. By giving the teams bad initial bases we will also increase variance in base building.

freem wrote:

* aliens builders needs to explicitly evolve to do new buildings, unlike humans. I can understand it on role play term, but a beginner as I was (I'm still one, but I've learn that) will not understand why he can not build new stuff. This could be explained more obviously.

It bugs me a little, too, that humans can always switch to building at any time while aliens often have to die in order to respawn as a granger. I'd like to change that in the future: Aliens should have more freedom to 'morph' into other classes (evolution doesn't really explain why a dretch can morph into a tyrant anyway). Balance is an issue though, aliens shouldn't be able to devolve in order to fit through a vent and evolve to a higher class again for free. I was thinking that any class you pay for gets 'unlocked' while you're alive and you can switch to any unlocked class for a single evo (equals 0.5 tremulous evos).

freem wrote:

* there is a good wiki. Still not complete, but very useful. It could be interesting to integrate some of it's informations in game, like weapon damages or range?

Planned, when we rewrite all the menus.

freem wrote:

* bots are not ready. This is not a problem (I can understand how hard it is to make such kind of work) but it could be interesting to auto-kick/auto-invite them when humans come in game. By example, if a server determine that it is for at least 6 players and there is only 1 human, it will make 5 bots. If 6 humans join (so that they are 7) it will kick all bots except one (to make teams "balanced" in number).

I've been asking for this for a while now but Fuma (hwo wrote the bots) said it isn't trivial to implement. Currently he focuses on making the bots more intelligent and I hope that usability will follow as soon as the bots behavoiur is done.

freem wrote:

Another improvement for bots could be to allow players to artificially change their current goal, so that when they camp (or bug) a player could say them: "go to enemy base and attack all bots on the way" and the bot will "simply" obey.

Yes, this would be cool! Again, there's this pile of TODOs for bots, so I can't tell you whether something like this will be done soon.

freem wrote:

* the previous section of that long post gave me the idea of reproducing a mechanism I've seen in 2 games: neverwinternights and battlefield 2142 (notice: a RPG in 3hd person and a FPS) . In those games, to communicate, there is a "rose" of buttons which spawns around the center of the screen. The central button give access to the most common action, while the others give quick access to options. This could be used, by example, for pre registered "taunts" like "follow me", "defend here", "attack" and many other I can not think about.

Absolutely. We are going to use circle huds for buying weapons and evolving so having one for voice says would just be consistent.

freem wrote:

a single-player mode? I guess it's planned :)

Sure we would like to have one but this is soooo far in the future. As a mapper, I would be very interested in designing maps for a single player campaign but our engine doesn't really give us the tools to design one yet. Tim has been doing great work on entites that can be seen as a step in this direction.

freem wrote:

** maps are not only training maps (there are only 3, am I wrong?)

A split into a PvE (players versus environment) and a regular newbe server has been proposed. I like the idea but it would probably make sense to wait until we have enough newbes to populate both servers.

freem wrote:

** on some training maps, alien never evolves (at least one: when we are humans with unreachable telenodes)

Do you really want to fight an army of goons on tremor? :D

freem wrote:

* make a guide (IG or not) to create player's own training games.

Nice idea. If we have something like a PvE server it would be cool if the training layouts could be player-generated. Not that easy to do though, you would need a website for layout upload as well guidelines and sanity checking of the uploaded layouts. Could be something Anomalous (hwo maintains the training server) would like to look into.

freem wrote:

* create a "puppet AI" mode with dedicated maps were AI are just doing a registered move (being immobile, walking slowly, running, ...) , plus respawning to allow beginners to train their aiming (remember when you first tried the dretch!).

This would be one of the many requirements for having a single player mode, too. And it would also add to Fuma's TODO stack. I don't think something like this will be done soon. :(

freem wrote:

but, I can not really understand why you consider unv as an alpha, honestly... at least not on programming side. 3D models? Textures? Can you not simply consider it as a fork of tremulous and go on?

No, we really want to be a different game, not just a trem mod. I personally would call it a Beta if we replaced all models, have the gameplay in a stable state, got two more maps in the size of yocto/parpax and maybe one more in the size of plat23 and have the new user interface done. There's a lot more important stuff to do before the release but I feel these are most important so that tunvanquished feels like a new game as opposed to a mod.

freem wrote:

But what I noticed quickly is that there is a "development server" and "other servers".

This is how it's done in software development. If you plan major changes you don't just throw them in the main program but create a branch were you can freely play with your ideas without interfering with the progress of anyone else. The dev server resembles such a branch. As soon as we had enough testing and like the new gameplay better than the old one in every aspect, we will merge the changes into the master branch, so that every server will be like the dev server.

freem wrote:

why do you not try to attract more players? More players, more potential for contributions...

We will focus on advertisement as soon as we hit beta. It doesn't really make sense before that: Not many players like to play an unfinished game that constantly changes and we don't want to deter anyone.

Responsible for: Arch Linux package & torrent distribution, Parpax (map), Chameleon (map texture editor), Sloth (material file generator), gameplay design & programming, artistic direction

User avatar
ViruS
Granger
Posts: 1020
Joined: Sun Mar 11, 2012 4:24 am UTC
Location: Antartica - West Australian Post shore
Contact:

Re: various suggestions

Post by ViruS »

The dark one with blue/gray textures is tremor. You spawn as humans and all bots go aliens.

ImageImageYou[TubeImage

User avatar
velociostrich
Dragoon
Posts: 318
Joined: Thu Mar 08, 2012 6:24 pm UTC

Re: various suggestions

Post by velociostrich »

Hey, thanks for taking the time to try and write some organized feedback!

freem wrote:

* there is a good wiki. Still not complete, but very useful.

Good to hear that you think it's good! As for incomplete areas, is there anything you can think of (that's not on the TODO list)?

freem wrote:

* the previous section of that long post gave me the idea of reproducing a mechanism I've seen in 2 games: neverwinternights and battlefield 2142 (notice: a RPG in 3hd person and a FPS) . In those games, to communicate, there is a "rose" of buttons which spawns around the center of the screen. The central button give access to the most common action, while the others give quick access to options. This could be used, by example, for pre registered "taunts" like "follow me", "defend here", "attack" and many other I can not think about.

Viech wrote:

Absolutely. We are going to use circle huds for buying weapons and evolving so having one for voice says would just be consistent.

Just FYI these are what you call "pie menus" and have actually been implemented in quite a few games. Aside those you mentioned, The Legend of Zelda: Twilight Princess, Enemy Territory: Quake Wars (iirc; might be wrong on this one), and Battlefield 2 also had these for different purposes. They can be done good or bad, but yeah, I (and I think most everyone else) agree that they're a much better idea than our current menu + scrollbar UIs (which are due for replacement once libRocket integration is finished).

Post Reply