BlackBoxBETA - Beta Test Community

Make sure you use the SVN link above, and not the download one.. much more uptodate version of the game, with a lot of improvements included already, much more polished

I tried the SVN today for a few hours solo and quite enjoyed it, as an old Q3A player it all seems pretty familiar to me tbh, which isn't a bad thing! It does seem a bit loose in some areas still vs some of the big names in the genre, will do my best to describe what I mean but to some extent it's just a feeling while playing..

Partly I think it's the hitboxes.. especially (and maybe only with) the rifle and shotguns, which seem to give just far too much "fluff" to the hitboxes, especially on the head. No problem once you have your sens dialed in right and your getting used to the way the game moves to just bounce about headshotting things - the zoom is a hinderance if anythng with it's over-kill sensitivity.. just don't need to use it when a shot in the general area of the head garuntee's a kill. I don't notice the same effect with some of the other guns though, the SMG seems almost the opposite - refuses to hit the head unless your right ontop of them spamming away.

Leads onto my other complaint - the speed of things.. I love a fast paced game but some parts of it seem overly slow, and others overly quick in comparison.. I mean things like the projectile rate for some guns vs the jump/dodge rate, I've tried the bots up on a really high level for a while, and once you've got used to how easily they can track your head and kill you (% accuracy always was a load of crap.. bots ftl) you notice how impossible it is to track things properly yourself once they are really bouncing about, let alone then lag shoot sufficiently to actualy hit them with some guns.. I think I'd just say the jump/dodge system is just a bit overpowered as it stands, bit too fast, too repeatable.. perhaps some sort of stamina could be used to limit the jumping and add another element to things: using stamina when it really matters, saving it when it doesn't etc

seems a good start, just doesn't quite all "gel" together quite yet, as I said at the start it's a kind of loose feeling that I think just needs a bit more tweaking to iron out... certainly feels a lot better in the SVN vs the beta file, just not quite there yet.

just my 2cents, hope it helps, kinda new here so don't wanna ruffle any feathers
-craig

Hi hi! One of the devs here.

As of late, the game does seem to be quite a bit faster than it used to be. A lot of this is due to tweaks involving adding new settings to characters(weight etc) in preparation for newer features. One of the nice things that we've done for mappers though is to create a way for the mapper to affect the gameplay significantly during map creation, simply by changing the map defaults. These defaults include gravity, player speeds(this is actually defined on a per map basis), jump strength, double jumping(which can be set anywhere from infinity to essentially allow flying to completely off for pure tactical) and other things.

Player speed is also affected by what weapons the player is carrying. Lugging around big guns slows you down, and small guns will let you move more effectively. Since this is a new idea, feedback on that aspect is really appreciated.

But the speed complaint is actually common - and even I think it's too fast ATM. I will probably dial things down somewhat in the map configurations in the near future.

The scope does zoom way in, and is mostly meant for larger stages or very long corridors.

As far as the hit detection goes, this is something we've inherited from the Cube 2 engine itself. It is a possibility that we could add better detection in the future as opposed to a box with a percentage(between here and here is a torso shot, above here is a head shot, below here is a leg shot).

One thing to keep in mind is that we're currently working towards adding a co-op Player Vs. Other mode, so until we see what kind of network traffic that is going to cost us, I can't say if we'll get to add a complicated hit detection model. Something that's a benefit in multiplayer CTF/DM games may be too costly when it's three or four players against a horde of monsters. In other words, that's a "we'll wait and see".

The weapons are going to need tweaking, but again this comes down to a question of whether or not it is a weapons balance issue or something that is unbalanced in the movement or both.

And yes - the bots are extremely evil. The same AI will be used for the co-op mode with some modifications and using far less network traffic - in other words, monsters will be comparatively dumb compared to bots playing with guns etc.

What are your general feelings on jumping and dodging in FPS games? This is one of those things that is difficult to get right.

Ahh, wasn't aware that some of the basic engine mechanics like hitboxes have of course been inherited from the game engine itself - not the easiest to change then..

As long as monsters are only "comparatively" dumb thats fine, as with all bots you reach a point below which the term dumb becomes an understatement, and actually your facing the equivalent of someone's dog using the PC. Sure this won't be the case here, just expressing a worry when bot AI is then dumbed down to provide monster AI it won't always work.. (played plenty of UT mods/conversion to justify this one btw) or at least takes a bit of extra work with good pathing etc

As far as dodging/jumping in general for FPS, it's always going to be dependent on the type of game, but I guess we can safely assume this is directed at the Q3A/UT/BF style of quick-arena based gameplay rather than something more "serious" like CS or CoD..

As far as this game is concerned, I think the Impulse system (yeah, i did some research.. ) is pretty good as it stands, nicely combines dodging with a rocket-jump size move also, and the control it gives you in the air is great fun and leads to some intresting fights in some wierd places (especially some maps.. some real hax positions) I just feel it needs a little bit of dumbing down somehow, although if you wanted to keep the sheer speed of the game as a feature then the stamina thing i mentioned earlier could be a good method of doing it..

another idea I had, and forgive me if this is just plain impossible to implement, or even just a bit silly/against what your aiming for.. but anyway, how about making it so the impulse-jump requires contact with a surface of some kind to be "allowed".. wall/ceiling/floor etc.. would be a way of essentially explaining how it's all possible, and add a cool element to the map design also. Would still allow on the ground dodging no problem, but would limit the players ability to fly about up in the air for quite long lengths at fairly high speeds to when there's scenery to allow it.. again makes certain areas of some maps really important/cool

have to admit i've been playing mirror's edge lately, so this might just be influenced by that.. felt it could have made a real fun online FPS

-craig

craig wrote:

As far as this game is concerned, I think the Impulse system (yeah, i did some research.. ) is pretty good as it stands, nicely combines dodging with a rocket-jump size move also, and the control it gives you in the air is great fun and leads to some intresting fights in some wierd places (especially some maps.. some real hax positions) I just feel it needs a little bit of dumbing down somehow, although if you wanted to keep the sheer speed of the game as a feature then the stamina thing i mentioned earlier could be a good method of doing it..
-craig

Think I agree with this tbh, it is just a little bit TOO good. Do the bots know how to reach some of the dodgy jump spots on maps also?


another idea I had, and forgive me if this is just plain impossible to implement, or even just a bit silly/against what your aiming for.. but anyway, how about making it so the impulse-jump requires contact with a surface of some kind to be "allowed".. wall/ceiling/floor etc.. would be a way of essentially explaining how it's all possible, and add a cool element to the map design also.
-craig

Add in wall running and it's starting to smell very matrix-y

Nah, in all seriousness this could be a cool way of doing it, but I think it's probably something you would have had to aim for as one of your game styles to actually have a chance of implementing it properly.. maybe i'm wrong, but otherwise it would probably feel like it's been added in late, which is never too great for the overall polish of a game

As is, impulsing is pretty simple to get the hang of, and there aren't too many tricks to it. The maps are usually designed in such a way that it is accommodated well - up to and including some spots that the bot AI can't get to, usually because it requires a combination of impulsing and gun jumping.

Initial thoughts for the idea were much more centered on things like wall jumping, wall running, and Action Quake type moves. Impulsing was a much more simple solution that offered more variety and less constraints, while avoiding potential pitfalls like players working their way up a 100 foot wall be wall jumping in a corner.

In other words - it could be done, but what are the consequences?

Bots with super deadly aim, too-fast movement and some other issues have all been toned down in the last couple of days. It is worth taking a look at the latest SVN version to see how things feel now.

Currently we're ramping up to implement co-op, so from a philosophical/gameplay standpoint I'd like to hear everyone's opinions on that particular direction.

This is the bit I'm looking forward to seeing tbh..

Will do some writeup of my expectations as afar as coop goes later, sure others will chime in too

Co-op is a hard task, and one that I think a load get wrong, and a load don't bother with because it's too much of a challenge to implement - all good reasons to try yourselves in my opinion.

A good co-op game has gotta be my personal favorite, really opens up the multiplayer options without having to limit yourself to typical game styles. Often I'll only experience the single player side of the game if there is some manner of co-op, too often I'm playing with friends and the idea of sitting there and blasting away on your own when you could be in an MP game together is not

Think I'll start siting some examples of where ppl have gone wrong, and where it's been done well.. and what made the difference, then bring it all back together to be relevent to BF directly.

Diablo/Baldur's Gate/The RPG Market

There are one's that consistently get it right, although aided quite dramatically by the simple mechanics behind the multiplayer. i.e. for each extra player increase monster health, damage, droprate and spawnrate by 50%.. rinse and repeat..

Not totally relevent to BF of course, no way to implement such a simple system in a story-driven FPS.. the RPG gamestyle lends itself very easily to a CO-OP game, the FPS genre does not unless you make the right choices.

They still have an important point though, the open-ended nature of the levels/story, and the easy manner in which the rewards system (items and weapons etc in RPG) scales to match the players. With everything relying on random calcs for dropchance and such, it's a very simple thing to increase this figure to match the number of players, and It'll be important for BF to find a similar manner in which things can be scaled quite simply by multiplication - if only to make things more simple in the long run

Left For Dead

I think if you want an example of how to make a coop game not only great - but actually groundbreaking in how well it plays and works, this is the one.

L4D relies very heavily on a couple of simple mechanics which bond the COOP experience into a very consistant one, and this is definately something I think BF should try and match in some manner..

Basically It's the idea that any single player split from the team is garunteed dead meat without help - essentially scripted events in which another player must interact to save the player in trouble. It's important these are not "triggered" events, the player should not be able to predict when something is about to happen, instead the idea is to keep the player on guard as much as possible.

The above "saviour" idea, combined with some clever map design which does not allow backwards movement in places meaning you team members can be trapped, and features some very well scripted "events" (these are the ones the player can expect, a larger scale happening in which preparation is needed to survive - you HAVE to expect these to stand a chance.. add it all together and you achieve something that most games don't come close to, and very few CO-OPs achieve either.. players who don't know each other, and would never normally interact online are forced not only onto the same side, but to work as a team or there's no chance of sucess

(Halo would be another good CO-OP to site here, although I don't feel they managed it anything like as well as L4D has..)


How this relates to BF:

It's hard to relate properly without some idea of your plans for this already, but I'll try and suggest some elements I think could work/could make a great CO-OP experience.

The "Saviour" effect: In L4D certain monster types were used to create this, with abilities that meant the player targetted was "trapped" until another player free'd them.. not a hard job to free people, but easy to not notice in the confusion, and also easy to end up in a place where being free'd becomes impossible because the group has split up.

How to implement this would depend a lot on the monsters we'd be facing in blood frontier.. It might be suitable for some of these to have this sort of ability.. the important part to make it work is both that this happens quite often so the player is forced to fear it somewhat, on edge.. Therefore it's also important that these "attacks" don't become predicatable - random mob spawning perhaps allow the player to be attack from behind?

Map Design/Scripted events: It's all about height in my opinion, creating routes that are only open in one direction, or areas where the players MUST get in a certain place to prepare for an oncoming attack that would overrun them elsewhere. Again it's an aim to introduce a level of fear/reliance on your teammates that means that even the least social of players has no option but to cooperate to stand a chance of winning.

This could include some Halo-style events also.. e.g. bases that must be defended using players mounted in turrets, requiring organisation beforehand, or the "survival" of the team against a massive assault by barricading themselves into a bunker (think AVP here also.. heh)

This bring in the timing of the events also, which again I'll site L4D for.. There are a number of places where a "big" event is due, perhaps the end of this particular level. The players are given both warning that an event is going to start shortly, and the basics of what it is (e.g. were all gonna die if we don't get to xxx. Or we must defend xxx until the chopper arrives etc). The players are then left to trigger the event start themselves, by calling the chopper for instance, allowing time once again for strategy and planning - without which they'd never survive

Would love to hear a little bit about what your thoughts are on the BF story.. is Mars going to be the site for all the action? Who is the enemy? etc etc.. Give me chance to put some of these thoughts into context with BF

Final thought: Would it be designed as CO-OP? Or designed as single player with co-op options? Think this can make a big difference, aiming for coop all along will result in a more slick result, but will also need a lot more work to support players when they are playing alone - bot scripting and such

My current thoughts on it are along a similar vein. One of the things that I think L4D gets wrong is allowing players to actually die during co-op, even if it is slightly difficult to do so. One of the things that I've personally always hated is being punished by having to sit for several minutes while other players try to complete a stage.

To that end, I'm hoping the AI implementation is robust enough to leave the players constantly surrounded - to the point that survival is only possible if they stay together - even if they decide to split up into one or two small groups to do so.

Players already have an infinite amount of ammunition and regenerating health, so the idea is actually a little more similar to gears of war, except there won't be a bleed out. Once a player gets downed, any other player will basically be able to patch them up and get them back into the action. While it doesn't seem much of a penalty for dying, I do believe that it will encourage players to stay together in order to up the amount of firepower that they can bring to bear on the horde.

Monsters will basically generate from spawns that are placed strategically and activated/deactivated via triggers, so that the levels remain populated and attack can come from any direction, with the number of monsters increasing and decreasing relative to the number of players.

The end result, I hope, will be that all of the players can make it through, or all of the players die. The idea of gimmicks that snare players individually to force cooperation is neat, and I think it works well in L4D, but in this particular circumstance the players are all regenerating badass ninja androids, so attempting to cheap on something that can probably just rip the arms off of a bear as an afterthought doesn't seem feasible. The end result will hopefully be that players walk together and rock together just to keep from being overrun.

Weapons that are slightly unbalanced in DM will find new fans. The SMG might be weak, but the sheer amount of bullets and the quick reload times will make it far more effective when you're dealing with a lot of weaker monsters that can be temporarily stopped/slowed by a bullet hit. The shotgun isn't effective at range, but is again almost an instant kill weapon at point blank - and the monsters will be considerably weaker than they players, just far, far more numerous.

Again - this is just the thoughts at the moment. There are times when quin brings out a surprise way of doing things that is simply more fun and more elegant, but we don't want players to be punished too much by being killed.

acord wrote:
One of the things that I think L4D gets wrong is allowing players to actually die during co-op, even if it is slightly difficult to do so. One of the things that I've personally always hated is being punished by having to sit for several minutes while other players try to complete a stage.


Ahh, L4D had the rescue thing where you could go and resuce your dead buddy from a cupboard or other equally obscure place - agree though, it's an annoying way of doing things.. What I liked from it was basically the bit you've got above, the idea that a player can be taken down + revived, and that in the end you could have all 4 on the floor at once and that would be the end of it all..

I still think if you can find some manner of introducing L4D's "ease" of taking players down, I understand with this story it can't be so simple as just being pinned down etc.. but there might be a method still to implement it and this combined with some good scripted events can really draw your players into proper coop.

It's starting to remind me a lot of AVP the original, could play that coop as marine's vs aliens on a map outside where you had to fight through an alien base to a bunker full of weaps. I can't seem to remember there being any point to actually reaching this bunker however, but even so we spent many an hour spawning blowing the crap out of the aliens - there's definately room for the well made blaster just drawing people to play together properly, and i can definately see that fitting well with how BF plays already tbh, fast paced and visceral (might need to add some more explosive stuff - sp only perhaps?)

I'm seeing AVP with a bit of L4D/GoW mixed in, with perhaps a sprinkling of Duke Nukem - alien bashing theme on top to polish things off, and i like it tbh.

how would the cube engine take to vehicles btw?

Never played left 4 dead, think i might give it a try now.. sounds pretty fun.

Gears of war I did play, and the coop in that + halo which is pretty similar are very smooth, assuming they would scale ok to a larger number of people then would be great fun

wondering about a few things though.. how many players at once on co-op? the traditional 4 or something a bit larger scale, think the game would suit a massive battle - would be pretty unique too

and when you say coop, do you mean a single player game thats also got a coop mode, or something like the old half-life coop mod that was designed for coop only?

AVP... never played that one co-op, but we're hoping for something similar to the following comparison:

BF -> Robotron
L4D -> Berserk

Quin's pretty sure that with the netcode we'll be able to do obscene amounts of active monsters on screen at a single time. Currently we're looking at about 50+ active if he's calculating it right over regular connections, so it's a good bet that we'll be offering something different

As far as vehicles go, it's a possibility, but it will rely on the ragdoll system development somewhat, as we're hoping to abuse it for things like frames and shocks etc. Mind you - you can already run 45 - 50 MPH and jump 30 feet high, so is there a need for vehicles?

This will basically be a pure co-op mode, with difficulty scaled for number of players.

As far as how much action it can handle at once... That depends. AI and bots are actually divvied out among the players, so the clients are each responsible for a certain number of bots. Monsters will be similar, with a lot less updating required.

In theory, you could get a full roster of players - up to 32 or 64 depending on the server. It would require an *awful* lot of bandwidth in co-op. Each player would need the bandwidth to handle themselves plus their load of bots, and it couldn't overwhelm the server. So if, on average, someone can host 4 DM level bots on a regular connection, then they should be able to do 10+ monsters since they'll be less intelligent.

So if someone plays the SP mode - they'll be playing local, connected to their own server, and they'll have some of the DM bots along as allies with no requirements for their survival or whatnot. That sort of crap is always irritating, although of course you'll want to keep them kicking just to have the extra firepower along.

I can honestly tell you right now: I have no interest in any "save the guy and keep him from dying" stuff. Escort missions give me a rash on a personal level, and most of the time it is impossible to pull off in a way that makes players feel like they're in full control.

Co-op is our next big push here. I don't know if there will be enough scripting capability to create anything approximating a cut scene for awhile, but machinima type capabilities are planned. We know that they're possible due to a cube 2 mod that implemented a simple version of it.

I also hate escort missions, load of rubbish that always end with frustration at dumb AI - not so bad when your protecting a player, but tis still the problem of "dumb AI"

Defend/Assault however is often quite fun, as in defend/assault a base/vehicle/gun placement/bunny playground. The scale still sounds really fun.. 32 player coop would be overkill anyway, but if you could manage say 10 players you've got a pretty unique style to it, all sounds very cool.

DEFINATELY agree with the single player plans, if you arn't with human players your teammates shouldn't really matter at all.. same as the escort stuff, nothing more annoying than trying to predict what an AI is about todo

acord wrote:

I can honestly tell you right now: I have no interest in any "save the guy and keep him from dying" stuff. Escort missions give me a rash on a personal level, and most of the time it is impossible to pull off in a way that makes players feel like they're in full control.

Sorry, didnt mean escort style stuff, was just describing how L4D had managed to pull things off quite well without following typical routes..
The idea was not that someone would have to be protected, but more than it's quite easy for the player to essentially "die" and need reviving.... L4D achieves this with certain abilities on monsters that are capable of basically killing you in a single hit if you fail to avoid it. This would again promote the idea of players sticking together, as you could quite easily get into a situation where you need reviving but are too far away from the action for anyone to help.

e.g. (L4D examples)

Smoker = Tongue can "constrict" a player, which will drag the player away from the group, and shortly incapacitate them

Hunter = Can pounce on a player, pinning them down and immediately incapacitating them

Tank = Hits like a train, will knock a player flying and incapacitate if it actually manages to connect with the target

Im with Icarus on the idea of around 10-12 players max on coop, anymore will become carnage and impossible to organise tbh! Also think BF could lend itself quite well to a UT style "Assault" mode/missions

blu? wrote:

UT style "Assault" mode/missions

What I was thinking, bring a good balance between/ fragging/ moving/ camping/ surviving

It may be a bit like capture game mode (with longer capture time and radius for camp fraging scene).

Last captured point may become the team respawn point to keep players together
(acting like a checkpoint).


Probably a countdown that end game if fall to 0, every captured point giving a time bonus (remind old school arcade ) (maybe a new entity var for flags that give the time added)

died player may respawn in waves
(all dead respawn at the same time every min , but it sometime give to much interest in dying when respawn timer is around 0 for insta respawning).
Or they may only respawn when a point is captured
(don't need game count down in that case, if every one die, game over)

Better weapons on more advanced Cap points (mapping stuff)

Stronger and stronger monster for each captured point

A level Boss spawning when all points are captured ?
If players have infinite respawn it may be too easy and repetitive (spawn, spam the boss, die, repeat) making dying more advantageous than loosing time trying too survive
(not the case when capturing cause you also lost the distance you made is you die).
In that case, Boss may get HP every time someone die
(even if he sucide or it will become 'ah, Shoot myself before he kill me'

Also, I like the stamina idea (only for impulses), would redux Z axis abuses without changing the gameplay too much.
In quake or ut, being on the ground have its advantage due to low air control, however, with BF air control, it's better being in air in most cases, and Jump Jump is a bit repetitive.
(only my opinion, I anyway love the game like it is specially the last SVN's updates)

Excuse my French for those that don't know me yet

edit : unlogging made me rewrite the full post meh. Bah, now I know I have to copy it before posting

edit 2 : I meant secure the flag game mode, Sauerbraten fault

Like the sound of that a lot, would fit well with the BF style methinks..

especially like the idea of basing things heavily around the cap points, whatever you wanna call em.. bosses, weapon upgrades, chances to heal + res all center around captureing these points, with waves of monsters between of varying difficulty - with a decision to push fast and avoid some waves, or go slow and safe but end up fighting more

keeps things very simple too.. only thing is i'm not sure of is if a storyline is wanted for all this, in which case this sorta setup gets kinda hard to explain (well, without stealing UT's storyline)

also wouldn't put too much into the boss stuff either, too complicated and you loose some of the fast paced and gregarious action due to having to follow set "tactics".. more like an "event" at each cap point, like some defense turrets to kill, or vice versa - to man and defend from a massive wave. Would fit "bosses" too, just think it should be wider than that

also lovin the SVN updates lately, also notice acord has been busy texturing again even if he hasn't posted on here in a bit

sidenote: thought i'd fixed the timeout to long enough to not mess with forum posts, will have a look at it because it's damn annoying when you don't think to copy before postin

For the story line stuff,
You have to activate antidote generator/antivirus/ or something (computers models instead of flag, maybe the overseer map one (recycling is good ))

LuckystrikeRx wrote:
For the story line stuff,
You have to activate antidote generator/antivirus/ or something (computers models instead of flag, maybe the overseer map one (recycling is good ))

It's eco-gaming.. Blood Frontier goes green!

Any news about the co-op stuff? Looking forward to this...

Have to admit I've not updated latelyfrom the SVN, maybe it's there already

No, still no Coop. last new stuff was the arena mutator (no weapon pickup, start with one weapon of your choice)

I guess things get slow because of the lack of modeler, Acord isn't much active lately (busy with real life I think)

I've heard of this RL thingy, nasty stuff!

Hope we see some coop soon, also looking forward to it a bit (although not tired of MP quite yet!)

Hello everyone first post on b3
and in bloodfrontier forums too, I have been a long time fan on sauerbraten and I just recently came across bloodfrontier (I must have been living under a rock to only see it now), frankly this game goes well and above all the things I've seen in sauerbraten cube 2 I love it!! it's perfect I'm getting the svn now, since I've been doing some personal stuff with cube 2 engine adding a few custom models changing some pixmaps etc I've been looking at the sources for bloosfrontier and I couldn't find a point in there where I can change player models, the playermodel command doesn't seem to work, and for adding new weapons too I can't find my way around BF's sources as a lot has been heavily modified, might there be a chance that these aren't included in BF or is it not in for now?. I don't know the advances of the game so I don't know if it meant to be that way
I was just trying it out to check the capabilities of the new addons and modifications, keep up the good work mates I'm loving the story line though very good

Welcome to the forums fella

can't help you too much with the technical side of the source code modifications.. but faily sure accord will drop by somepoint soon to give you some pointers. most of the modelling work so far has been done by him, and I know they'd welcome any help they can get adding further content in there

Also lovin the story, a nice twist for a game like blood frontier that originally seemed to be a pure MP shooter through and through.. and yet pulling off a story based game as well somehow!

the BF guys are always very open with any suggestions or additions, sounds like you've got some thoughts already as to what could be added.. very welcome

spectroplasm wrote:
I can't find my way around BF's sources as a lot has been heavily modified, might there be a chance that these aren't included in BF or is it not in for now?
You can always come see us on IRC to ask questions in realtime. Generally we can't run support on every forum out there (it's basically a 1.5 man project), so third party sites may suffer from a bit of lag on our part.

You'll find most things defined (like the player models) in "src/game/game.h", but yes - the codebase is heavily modified to suit our purposes.