Porting my Gloom-inspired mod to Unv?

[wepy]

Hi everyone:
I've been porting Gloom to the Source Engine for the past year under the project name, Darkness: Source, and I would like to port it to Unvanquished now. Since most of the developers on IRC apparently haven't played Gloom before, I was wondering if we could try my mod together before I start porting.
Here's what I want to know from Unvanquished players/devs:

• - What features, if any, would you want to put into mainline Unvanquished?

  • What features would not be easily implemented in Daemon?

  • Can we agree on features for a single game that brings in fun aspects of both gameplay styles, or would Gloom be better off as a separate mod?

  • If the games are too different and nobody really likes Gloom-style gameplay, I'd like to discuss potential ways to make the mods more compatible, perhaps instead of a separate VM, we could achieve the gameplay modes with conditionals...

  • You can download it on moddb or at the main site: http://darkness.dnbrown.net (mac/linux/windows: mac version of Source is a little broken, you won't be able to connect to the public server; linux = i don't know; Windows = works best). If you want to try all of the classes in a local game: map spacewreck; sv_cheats 1; join a team/pick a class; impulse 102 for 10 frags/points.. respawn to try other classes. dk_team_balance 0 to disable team balance enforcement. Can't claim that anyone would be playing on the one available server.

  • If I make it a mod for Unvanquished, how should I make it better?

Anyway, I promised honest feedback for Unvanquished as well. Please don't take this personally if you implemented these features, I like the game, I want to work with you. Also, feel free to tell me how much my mod sucks in return. Here goes:

• - I couldn't get into the game.. wasn't fun enough. I know that's a complex and highly subjective comment, maybe it's due to my Gloom preference, but maybe there are some gameplay-level things that we can improve too.

  • The Engine doesn't look as good as Source: Looking forward to seeing the replacements for the Tremulous sounds/sprites. Also, circle shadows under players look funny

  • The netcode isn't predicted like the Source Engine, which feels different from most new games I've played (shoot a gun, don't see hit sprites until server response is received). Here's a description of how they might differ: https://developer.valvesoftware.com/wik ... Networking

  • Sticking to walls is really hard to do as a spider, the normals change too fast. As I've mentioned before, this is one feature I'd rather replace with a grappling hook-like functionality as you'd see in Darkness/Gloom. Grappling is a lot more fun.

  • Generally, not a fan of the NS style team thing, where you can't buy X unless your team is working well (which happens so rarely in online games). Much prefer the Gloom style.. Get a frag and you can spend it on a new class.. save up 5 frags to get the best class on either team.

  • Movement was really fast... it was really hard to hit the little bugs, even with the automatic rifle at range, where I should have the advantage as a human (the small ones should rely on stealth, no?)

  • Maps too bright

  • Spider/bug entities were not straight forward in use.. really have't seen most of them I suppose. Gloom had obstacles (smaller than Unv's, and stackable), spikers (easy to undertstand just by looking at them), gassers (also easy to understand), healers, and eggs.

  • Weapons not satisfying/fun for humans (except maybe the shotgun?). The lucifer canon's effects didn't make it feel dangerous.. charge was too fast, didn't do enough damage to friendly base structures/nearby units. Rifle didn't work well against even smallest spiders...

  • Player death was awkward looking/sounding for some reason.. I think I'm used to ragdolls in Source.

  • No flashlight/darkness game going on.. spiders have bug vision, but don't even need it in the maps i saw because there aren't many truly dark areas

  • Low framerate with the Rocket menus.. some menu glitchyness/unresponsiveness. UI was better than what I had in Source though.

Thanks!

[kharnov]

What features, if any, would you want to put into mainline Unvanquished?

Many. Check out all of the articles that I've posted onto the main site. You should also probably read the game design document that Viech wrote, considering that our gameplay is in a constant state of flux and is moving towards the goals stated there.

What features would not be easily implemented in Daemon?

I don't know. That depends entirely on the skillset of the person writing the code. Renderer features tend to be very tricky. Dealing with our inherited gamelogic has also been a mess.

Can we agree on features for a single game that brings in fun aspects of both gameplay styles, or would Gloom be better off as a separate mod?

A separate mod, considering that we already have something in mind that we are moving towards. You are free to comment upon it and give suggestions, though.

I couldn't get into the game.. wasn't fun enough. I know that's a complex and highly subjective comment, maybe it's due to my Gloom preference, but maybe there are some gameplay-level things that we can improve too.

Subjective and due to your Gloom preference, as you've said.

The Engine doesn't look as good as Source: Looking forward to seeing the replacements for the Tremulous sounds/sprites. Also, circle shadows under players look funny

All of our shadows are placeholders until we've redone the lighting system to have proper realtime lighting. For the meantime you can play with the shadows in the options menu, we have more than just the circle blobs. There are a lot of things we are doing to make the engine look better, and we've come a long way considering that the game didn't look particularly attractive in our early releases.

Generally, not a fan of the NS style team thing, where you can't buy X unless your team is working well (which happens so rarely in online games). Much prefer the Gloom style.. Get a frag and you can spend it on a new class.. save up 5 frags to get the best class on either team.

That's a difference in styles, then. Our system is actually a bit less rigid than the stage system present in Tremulous. The unlocking of upgrades is supposed to represent your momentum as a team. When you start doing well, you can start spending on the more hardcore things to take the enemies down with. On the contrary, as your team starts losing for too long, you actually start locking upgrades again, so that the game doesn't end up being a stalemate.

Movement was really fast... it was really hard to hit the little bugs, even with the automatic rifle at range, where I should have the advantage as a human (the small ones should rely on stealth, no?)

That's the point. A dretch has very low health, you don't need many hits to take it down. Blindly spraying ammo everywhere is only going to hurt you. Yes, they rely on stealth because they typically need to make every hit count. They don't make noise when they walk. You are supposed to use the wall-walk ability to creep up on lone humans and damage them. If you don't kill them in one or two hits, you are fast enough to evade most of their shots.

Maps too bright

Hilarious. Most people have complained about our maps being too dark.

Spider/bug entities were not straight forward in use.. really have't seen most of them I suppose. Gloom had obstacles (smaller than Unv's, and stackable), spikers (easy to undertstand just by looking at them), gassers (also easy to understand), healers, and eggs.

There are descriptions for all of them in the circle menu. I made a point of trying to emphasize their uses when writing them. A tutorial mode should cover the remainder, when we get to it.

Weapons not satisfying/fun for humans (except maybe the shotgun?). The lucifer canon's effects didn't make it feel dangerous.. charge was too fast, didn't do enough damage to friendly base structures/nearby units. Rifle didn't work well against even smallest spiders...

Do not look at the weapons in their present states, they are all subject to heavy mutation down the line. As you may have noticed, we are turning the lasgun into a beam rifle, for instance. The weapons in their present state have largely been inherited from Tremulous and are in the process of being redone. Regardless, I do not really see a point in increasing friendly fire damage.

Player death was awkward looking/sounding for some reason.. I think I'm used to ragdolls in Source.

The addition of Bullet physics should give us access to ragdolls. Until then, we have death animations for things.

No flashlight/darkness game going on.. spiders have bug vision, but don't even need it in the maps i saw because there aren't many truly dark areas

We aren't going to deal with light and dark, but rather various forms of radar. The reason for this is that if you make the game rely too much on darkness as a gameplay mechanic, people are just going to turn up the brightness/gamma settings to their maximum values.

Low framerate with the Rocket menus.. some menu glitchyness/unresponsiveness. UI was better than what I had in Source though.

The libRocket implementation is very recent, literally as of last month. Give it time to fully mature.

[enneract]

- I couldn't get into the game.. wasn't fun enough.

- Weapons not satisfying/fun for humans

Could you make an (as detailed as possible) list of things that feel wrong or not satisfying?

- The Engine doesn't look as good as Source

How so? To me Source (especially Half-Life 2) looks like Quake 3 with higher quality models and textures and occasional dynamic lighting and fancy shaders.

- Sticking to walls is really hard to do as a spider, the normals change too fast.

Increasing smoothing time might help.

- The netcode isn't predicted like the Source Engine, which feels different from most new games I've played (shoot a gun, don't see hit sprites until server response is received).

It's unnecessary for hitscan weapons (doesn't distract on low ping servers) and missiles aren't unlagged (so predicting them would be an error).
I'm currently working on unlagged and predicted missiles. Since predicting missiles is complex, I'll probably create a flexible client-side prediction framework (so, as a side effect, predicting hitscan guns could be done painlessly).

[wepy]

- The netcode isn't predicted like the Source Engine, which feels different from most new games I've played (shoot a gun, don't see hit sprites until server response is received).

It's unnecessary for hitscan weapons (doesn't distract on low ping servers) and missiles aren't unlagged (so predicting them would be an error).
I'm currently working on unlagged and predicted missiles. Since predicting missiles is complex, I'll probably create a flexible client-side prediction framework (so, as a side effect, predicting hitscan guns could be done painlessly).

Every other game I play has used the Valve model of prediction since it makes the game feel more responsive (which should mean more immersive). I can feel the difference even with <50ms ping... If you had asked me in the early 2000's which model was better, I might have said Q3...

Maybe it's not worth the extra complexity though..

[Spiney]

Hey, I'm just curious, why do you want to switch from Source to some open source engine?

[kangz]

Because our engine is better than Source obviously.