Forum posts

No, the env_global should be used to change the state of the multisource. Since the multisource is set as the master of the door, it's state controls whether the door can be opened or not.

So you should trigger the env_global by the switch, so he changes (sets off) the state of that multisource.

EDIT: I've still got an old example map on a site somewhere, here it is. I just hope it's the right one...

Use a multisource and fill it's name in in the master property of the door. Now set up an env_global and set the global state of that multisource off whenever you trigger the env_global. Trigger the env_global by the switch and you can lock the door.
Basically, the state of the doors master (the multisource here) controls whether it can open or not. The same can be applied to buttons and a lot of other entities.

Alternately, you can give the door the name 'locked' so it's always locked (thanks to Muzzle for that one ;)). Not what you need here but just in case, that's the easy way for always locked doors.

Use imageshack or a site like that to upload your picture.

wpoly count can easily go up to 900 or 1200, depending on the systems you're target audience have. 900 is fine these days.

The face splitting you see in wireframe mode is caused by brushes that intersect or touch other brushes, but large faces get split anyway (every 240 units if the texture is scaled at 1).
There are some techniques to counter this, by using entities (when an entity touches a world face, that face doesn't get broken up) or leaving a little space between the brushes (often looks odd). Don't waste time trying to avoid all splitting, just do it for small or oddly shaped objects. Especially circular objects cause a lot of extra face splitting so these are best made into entities.
About the 240 units splitting: you can use a different texture scale to change the splitting amount, but if it's a floor the player walks over, then a large scale often looks bad so that's not a good idea actually. I'd say don't worry too much about this type of face splitting. It's normal.
@Kasperg: you probably refer to HL2's wireframe mode? That one shows triangles, HL doesn't.

epoly's are model polygons. These render roughly 10 times faster than world polygons.

Anyway, the map you've got now probably doesn't need optimization because your poly counts are pretty low and performance is good, isn't it?
Though planning the layout beforehand so area's are blocked from each other (e.g. there's no direct line of sight from one area to another) is an important step. Check 'gl_wireframe 2' to see what's rendered (it's more than you can see as a player).

Oh, and a polygon is a multi-sided shape. Not necessarily a triangle.

Heh, fun idea, Muzzle. But uhm, just HL, or HL2 entries as well?

Not that I'll enter, I'm way too busy for that now, but just a question anyway...

I've used QBasic too. 7 years ago. I'm glad I switched to C++ actually.

// Now to finish that logigram program... gotta write a script parser for the input and a solving alghorithm for, well, the input...

Rules of Life:

If you don't know what it does, DON'T TOUCH IT

Rules of Mapping:

If you don't know what it does, TRY IT OUT.

// Couldn't resist that one, Bratty...
Man, I've spend years of mapping trying things out, experimenting and all. It was fun and it even pays off these days...

I don't see the need for such a tool actually, as Seventh said...

Or are you learning to program and chose an exercise for yourself or such?

I noclipped down in the floor, under the bences to take a interesting screenshot, guess I phailed..

It's best to take screenshots while just walking around rather than noclipping to give a good sense of the proportions of the map, unless there are objects that define the proportions clearly.

What intention do you have with the map? If you're making it for Garry's mod, then that doesn't say a lot... is it intended for some of those weird multiplayer modes, or for comic shooting, or something else?

Anyay, the shots don't show a lot right now. Only just some props in a pitch-black environment. Well, if you're aiming for such a style, maybe that's cool, but it's not clear where you're going with it now.
I don't like the teleport trail, it's too black'n'whitey. Leave those ugly white parts out and it would look a lot better.

Well im not Looking to make a massive sucsessfull MMO, Im just want to make a small comunity, And a few worlds for me and my mates to hang out in... maby do some quests and stuff together...

It's fine to create something for you and your mates, but I suggest you take a look at what it's going to cost you. MMORPG's take a lot, a lot of time to create and if you're unexperienced, then changes are you're going to fail anyway. 99% of all HL mods died, 99% of all custom games failed (check flipcode.com for example, someone compiled a list of all games that were started there, and a list of those that finally got out or were close to a release. 250 against 3 or such)...
So, know what you're doing. It's more fun to finish something than to have wasted your time on things you never were able to play.

Well, not that you won't learn anything from that...

I just cant work with the hl2 textures, it just doesnt match for me

Same here. I do have (or make) some time for creating new ones though. Map creating costs more time now but since I use a more structured approach these days, I waste less time detailing area's that finally don't fit in, so that balances things a bit. Modelling for HL2 is also easy if you're used to do it for HL so although it takes some time to adjust your workflow, it's possible and now that I've done it, it's not so bad.

Heh, I think that's quite true, ZL.

Not to mention the server costs and the insane content creation times... I've seen several teams of script-kiddies and wannabee programmers thinking they could create a MMORPG... I've seen just as many fail. Hey, even whole companies decide to do something else because of the costs...

Besides... if you really want to create a game... why not just learn programming yourself, or team up with a friend programmer?

Well, a long time ago I got a game-maker program as well. I was able to get a sort of Tiberian Sun-alike terrain engine to work, but now that I look back on it... it was horrible coding, a horrible idea and all. Maybe those programs are nice to get people introduced into game-making, but it's nothing serious stuff I think.

Sure it's possible...

Sprites are just materials, usually placed in the materials/sprites folder (though I'd argue any other folder within the materials folder could do). All you need to do is using some different parameters in the .vmt file. For example, this is the blueflare1 .vmt file:

[quote]"Sprite"
{

"$spriteorientation" "vp_parallel"
"$spriteorigin" "[ 0.50 0.50 ]"
"$basetexture" "sprites/blueflare1"
"$additive" "1"

}[/quote]

So instead of using "LightmappedGeneric" for world faces or "VertexLitGeneric" for model faces, you use "Sprite". Well, the rest you can find out yourself...

EDIT: I'm actually surprized there aren't any tutorials or articles about this, at least not on the Valve Wiki. It isn't hard to find out, I just looked at where the browser went when choosing a sprite for an env_sprite and that happened to be looking in the materials/sprites folder, but it's lacking documentation. Ow well...

For low-poly models Milkshape might be a better choice, Elon. Creating models doesn't cost hours - creating good ones (and especially creating good textures for it) can be time-consuming though.

Could do with better smoothing, but it looks good.

Got a wireframe screenshot?

I'd say, filtering out models texture in Hammer would be a smart thing indeed. Why they didn't do it? No idea...

Anyway, that map looks a bit standard lighted. Not really realistic, it's too dim for that and there's no really interesting shadows yet.
Other than that, it looks like a good start and it's not hard to see what theme you chose.
Good luck with this one!

Half-Life 2 is different than Half-Life so HL-type answers do not always apply.

Use a func_tracktrain instead, and set this tracktrain as the parent of the func_doors or func_door_rotatings, so they will move with the func_tracktrain. And yes, you can attach multiple entities to another entity with that parent system...

Just give it a try. I can see no reason why there's a limit of one door to be attached to a func_tracktrain...

You can pick up sounds and broadcast them at another place in HL2. Check out the env_microphone entity for that.

Perhaps using a phys_hinge entity on a door prop_physics is better. So instead of opening with a vast speed and a certain distance, it reacts as a physical object you're running into, going open as you push it. Remember those grate doors in HL2, just after Alyx was captured? That's what I'm thinking of.
Oh, or try the Force Closed option from the func_door_rotating. Perhaps that one is set to 1 so it always closes no matter if you block them?

Maybe you still have the .vmx? It's a backup of the .vmf, automatically created.

And as for Vmex... since it's your own map, how could you possibly steal it? Go ahead and use it if you've lost your maps source.

Hmm, I think it's better to continue with the outline of the next maps rather than detailing these ones. During the modding process you'll get better and when you look back at those first maps, you'll see a lot of things you want to change, so the whole process goes on again.

Plus, people like to play new stuff rather than rehashed old stuff...

At least get rid of the annoying moving white dots on your site. A flickering horizontal scrollbar isn't a nice thing, you know...

As for the mod, how's progress since I last played it?

Learning how to model should do you good. Drawing weapons is nice, but getting them into your favorite game is more fun.

I'd say, check out a site like this for inspiration...

Well, a broad question...

Not all surf and rats maps run well, but when they do, they either use simple architecture, stretched faces (often the case for rats maps) and a good optimized layout.

Check these, and your, maps with gl_wireframe set to 2 and with r_speeds on. This should give you a broad idea of what's going on in most maps.

I would ask myself: is this really worth all the trouble and will the faking actually work correctly, or isn't that ladder so necessary for the gameplay and can I use a simple pushable instead?

Not that experimenting is a bad thing, but sometimes things aren't that necessary as they initially seem to be...

What's the problem with OpenGL? The mipmapping that makes textures afar more blurry? Personally I think not using mipmaps looks worse, but ok. You can tweak these settings somewhat with in-game console commands, something like gl_mipmap or such. I'll do a search for them if you want to use these commands.

Software mode looks like it was built in primarily. DirectX support was thrown in quite late in the process though.

Quoted from the modelling document in the HL SDK:

All of the models in the shipping version of Half-Life were created with 3-D studio Max from Kinetix Inc., and the export tools included with this SDK currently only support Max.

They used Softimage XSI for HL2 though, or at least some modellers at Valve did.

Unemployed? Sounds no fun... Is it so hard to find a job in France now then?

Valve used 3DS-Max for their HL models, actually...

The HL SDK includes max-to-smd exporter plugins for versions 1.2 - 3.1.
Alternately, you can use Milkshape as it comes with a good HL smd exporter by default (plus it offers easy model decompiling and .qc file editing), and it's able to import a lot of different filetypes including .3ds files.

So you've got no reason at all to be so stressed. Relax a bit, things aren't as negative as you think they are. Why not take advice from fellow mappers, mappers that are a load better than you are and that have found out a lot of things through trial and error, and share their experience so you don't have to go through all the trouble? Really, discarding that is going to hold you back a lot. You learned the technical basis from others through tutorials but you can learn a lot from others in the design area as well.

And if you want me to do some modelling work, just send a PM for contact details.

Dave Johnston is the creator of some official CS maps. A recommended read for any CS mapper. You can learn a lot of others who did the same things before you. And, personally, I always find it interesting how other people do their things. It broadens my point of view and often gives me idea's I wouldn't have gotten otherwise.

Anyway, about some mapping workflows: I used to create detailed area's and expand upon that and I often either got stuck, or had to redo a lot because the layout didn't work really well. Now, I just create a layout first, playtest it and only when it plays nice, I continue with the look. This saves me a lot of time and makes sure that every texture and model I create also actually gets used. With increased construction times for nowaday games, that's invaluable.
This method also gives me a first impression of the maps framerate and lets me plan where I can add details, and where I need to hold back to keep it playable. Planning things out didn't come naturally for me, but it does help me a lot now.

About the model, I had no problems at all. You just have a bad exporter if those things happen to you... a really bad one. Aren't there others for 3DSMax?

And for me helping out, I am busy with school since I'm taking extra lessons to do it in a shorter time, but ocassionally I'm able to do some little prop models. PM me if you want to talk about it.

EDIT: 'exporter skills'... ghehe, exporting is easy if you've got a good exporter. Has nothing to do with the person...

I don't see how I could make it any less boxy.

Trims, supporting beams for the wooden ceiling... the problem is not so the squareness of the environment but the detail spreading. Long, boring surfaces make a square map look blocky. Putting some details now and then helps breaking up these surfaces so it feels less boxy.

I think this map also lacks finesse. Those circular pillars are nice, but they're oversized which makes them feel odd. Those planks in the broken ceiling are too thick also. Just some small things that break the immersion somewhat. I see you did pay attention to some nicely made lamps. Just don't forget about the rest.

Oh, also, the inside of the house looks very rich and grandeuse. The outside looks... well, cheap. Most textures don't go well with the theme either, I'd pick a different texture set if I were you.

And light it up somewhat. Nightvision isn't available from the start and it's especially at the start that people need to learn the map. NV doesn't seem to be too popular, either, so it may not be a wise choice to create a dark map.

You probably used model textures. These are vertex lit and so they often flicker between black and normal in-game. Use non-model textures instead. You can recognize model textures by the 'model/' in their pathname.

Dedicated texture artist and modellers
I agree that texture artists and modellers aren't easy to find. Maybe it's good that one of your team-members starts focussing on it if that's possible...

Models
Anyway, for the models: I use Milkshape 3D, a low-poly modeller and it works great for HL models. The fence model took me 15 minutes to redo, giving a lower-poly, smoother version. If you want it, go ahead. I didn't fix the bone structure though and it's certainly not the best quality as I rushed it, but hey... (So, if I have some free time, I could do a model or two now and then...)
If the 3DSMax exporter gives trouble, use something else. It's the end-result what counts, not the in-editor thing. A friend of mine used a plane model I made for him in his game and the textures were all flipped. It turned out to be a problem with the texture format, so he used another format instead.

Spicing up the plane
The plane looks fine the way it is now, but using models for details on it is a nice thing, isn't it? They're not important for collision or anything, just to make it look better...

Gameplay versus visuals... or is it gameplay and visuals?
As for visuals, it's good to focus on gameplay first, but visuals and overall impression are nice things to have at hand when it comes to convincing people to download the map. My brother, a CoD player, likes good-playing maps but he favors the good-looking and good-playing maps. It's a win-win situation, one doesn't substitute the other, they work together.
And hey, didn't you just create some nice terrain there? So you do care about the looks of it...

Getting stuck and generally hard movement
I got stuck in the small crawlspaces between the helicopter and close to the tower. One crawlspace had a 45-degree brush above it so when I tried to enter it (I usually crouch as late as possible to save some time) I often bounced off the brush above it, ending me up next to the crawlspace rather than in it.
Also under the bridge terrain was a bit hard to navigate, the point where you get up the bridge, opposite of the bunkers and coming from the airstrip, also wasn't very easy. I'd rather keep running on a path than having to jump to get further.
The rest of the terrain looks very well done though, good job on that.

Don't let the map be blamed
As for using your brain once, I'd rather see my maps played again...
Sure, vary the 'cost' of a path by making a safer route longer or making a shorter one more vulnerable to enemy fire, but watch out with restrictions. Players must blame themselves for a death due to a bad choice or bad skills, or let them blame the enemy for cheating or just being good. But avoid letting your map take the blame.
Make paths have their cost, but don't forget the gameplay dynamics. CS is a shooter, not a watch-your-steps-game. Players focus on the combat, not on the micro-movement. Speed is key when getting to an area so slow parts can get annoying. Read Dave Johnstons articles if you haven't already. He fine-tuned his player spawn-points up to seconds. A 2 seconds difference meant a total over-balance of the map... speed and timing is key.
Also, narrow area's only allow one or two players at the same time. If you're allowing up to 32 players, then this map is too narrow. It also allows little dodging which might get frustrating.

Oh, and did I already say I'm a perfectionist?

EDIT: I read on your site that you were forced to optimize when the map wasn't even done... how do you guys work? Creating detailed rooms first and expanding the map room for room, or creating a bare-bone basic layout to playtest first and detail later?

Google, Muzzle...

Nice tool. Nice idea, too.

I've played through the map and since I was bored with my homework, I decided to write some stuff about it. It's mainly suggestions for improvements as I like that aspect a lot (unless a map really impresses me, which rarely happens). So don't take it the negative way, I'm just trying to see this thing getting improved. Here goes:

A little on gameplay
Good to hear it plays nicely. I'm not really a full-time CS player so I've got little to comment on that, except that the map feels cramped at some points (the chrouching is annoying and some halls are just wide enough for 1 player) and that I did get stuck on the triangle terrain now and then. Aside from that, the layout looks ok.

A bland look: textures and lighting
The visuals are ok, but nothing really special. The texture set and the lighting are mainly yellowish which makes the map appear very monotone, the rock and grass textures aren't really contrasting with it either. Add the lack of feasible shadows outside and the result is quite bland. Getting a more distinct and contrasting texture set and tweaking the lighting outside to make it look interesting would do the map really good.

Prop models: touching up and spreading them
Using prop models is a great thing, but they don't look very good at the moment. You might want to reskin and redefine the fences (they're quite rough now) and the sandbags especially (the rock looks strange due to the skin as well - usually small rock surfaces aren't that noisy). As for detail, some area's in the map look well detailed while others consist mainly of long boring (yellowish ;)) surfaces. Spread the detail some more, using a few exta low-poly models here and there doesn't hurt.

Sense of place: low rocks and sudden ending airstrip
Also, get rid of the invisible clipping at the airstrip. Add some fallen rocks or more fences to make it feel right. Adding a low-res textured area behind it really does it good too, as now the map feels faked (the low rocks that give too much sight to the skybox have the same sense-of-place-killing effect).

Sprites: more and better fitting
The sprite-based plants are a nice idea, but they look odd against that grass texture (they look a bit odd anyway). It may also be that there's just one or two popping up on a large field that makes it look odd.

Additional model usage
Using models for the plane and the truck really isn't a bad thing. Or at least using some prop models for the finer details of them (like a lamp or antenna for the truck, or a flight panel for the plane). Just some extra touchups that aren't really necessary for the gameplay but I think they do make the map more interesting for a certain kind of players.

Lighting: shadowed lamps?
Oh, some lights in the basement don't seem to lighten the environment really. It's because there's shadow on them. Changing their render mode should do the trick. A glow sprite could be nice too. Better, use models for them as they appear quite blocky now.

Models: technical approach
I also took a look at the models and you can use some improvements there as well. Think about smoothing groups for the fences (they appear blocky now. With proper smoothing, even a 5-sided cylinder looks better). That model also used an exessive amount of bones. One suffices for static props. I also don't know why you should use so many hitboxes, as I assume you're clipping the model manually anyway.

A brothers comment
Oh, and I let my brother play the map. He played CS for a while and plays CoD often, so I think his comments as a player might be usefull as well. He agreed on the bland look, he also noted the absence of background sound and some strange env_sound behaviour (some echoes didn't sound like they would in reality). He didn't like the crawling spaces as they took too much time and got annoying. He liked the lockheed and the helicopter, though he thinks the lockheed is too small and the wheels look odd.
He found it a fair map, but not for a 3-months job by a multiple-membered team.

Conclusion
Overall, I too think it could and should get better on a lot of area's.
It's a good thing, though, to see people working together on a map and putting so much effort in it. That shouldn't go unnoticed...
I would recommend that your team specializes more, though. Get a dedicated texture-artist and a modeller, I'd say.
Good luck!

atom

Atom.

Nice try.

Cleaning out the map vault is a last resort option, methinks.

Anyway, what about hosting off-site to save on bandwidth, and urging people to host maps themselves? Free webspace isn't hard to get so linking to some free sites on the upload map page could be a good idea...

EDIT: I see, it's just page views and forum posting alone that takes up most of the bandwidth. How much do downloads contribute to that?

There was the wadfather site, but that one isn't really alive anymore last time I checked.

I think it's a good idea to have a vault here at TWHL. A model vault would be nice as well. I still have some prop models and a few .wad files lying around and there are some modellers on this site, so maybe it could take off well...

Your browser type will have nothing to do with viewing the website.

I understand, and I was pointing out that was one of the problems with your idea. Few people would download a new program just to view someone's site (let alone to view a program that tries to be a site) while they can visit other sites without having to download anything new as anyone already has a browser.
Why don't you just learn html and make use of that fact?

1. How would a headcrab be able to fit on a bullsquid?

2. Bullsquids are quite the headcrab killers. How would a headcrab even come close?

3. Monsters can be made to attack each other. Somewhere in the gamecode there's a table that lists the relationship between monsters. As far as I know, it isn't too hard to change that (though it may take some time to find that table).

Personally, I'm gone when a site requires a program to install. You won't get my visit then.

The thing about a plain old html website is: everyone with a half-decent browser can view it. It isn't too hard to learn either. Once you get the hang of it, you don't need no templates anymore, either.
Steam, for example, pops up a news-window every now and then. Guess what? It's a html page... no matter how you program, you will still need some sort of markup language to get anywhere, unless you're satisfied with plain text and images.

Hey, why not write a html editor in VB if you want to do something with it anyway?

Vertical noise...

It looks closer to wood than to metal as far as structure goes. Look at some pictures of real metal to compare your texture with, and try to imitate reality.

Don't underestimate texture-arts: it's difficult and it takes some time to create a good texture.

Creating good textures from photo's isn't so easy. Often you'll have to touch them up so they fit well with other textures, and making them seamless isn't easy either.

Creating textures from scratch doesn't require photoshop skills, just use any program or tool you're comfortable with. The textures I made were done with Paint, Wally, IPhoto Plus and Paint Shop Pro.

I found the tools less important than the actual way you create your texture. A noise filter or a sharpening tool isn't magic - it's usefull for certain materials but not everything. What helped me the most was just playing with the various options of the programs I have, and looking for a lot of reference pictures.
Also, a texture often has several stadia it goes through. At least that's how I work. First I do the basic shapes and colors with Paint, often I create a greyscale texture that defines the relief and I apply it as a layer in PSP. I work out the details with some painting or spraybrush tools and that usually costs quite some time. Testing the texture in a map with several different light colors and settings is also usefull, especially the light settings that it's most likely to be seen under.

Basically, there's no truely right or wrong way to do a texture, since each material has it's own look and feel so you'll often use different approaches for different types of materials. Some textures contain multiple materials and then it might be handy to work on each section separately, putting it together and finishing things up in the end.

QB, as in QBasic?

I can't use anything PHP based,, :/

Try Apache. It may take some time and understanding to get it up and running but it allows you to test php on your own system. That also means you can use phpmyadmin on your computer.

The SDK can be downloaded here as the old link on the VERC seems to give problems.

Anyway, knowing a little C is fine for little adjustments like changing HUD colors and positions and other small changes, but adding new monsters gets harder. Some basic understanding of classes and inheritance would be good if you're planning to do anything more complex.

Also, I'd recommend looking at what changes you want to make to the game before 'doing some coding'. It's no fun to throw away big plans when you find out certain things are too difficult to handle half-way down a project.
Good luck though.

Cap'n P: I'm pretty neutral with regards to most of Java. You know what I mean though, old Java stuff.

I don't like Java too much. Most applets take a while to load only to find out that they're useless to the site's function.
Of course, that's bad Java usage. But I don't like the language itself either. It gives slower results than a C++ program, and uses classes everywhere and nowhere. Classes are fine but they're a tool for me, not something to be forced to work with.

Maybe I'm just a C/C++ guy...

Cap'n P? You're an Opera fan?

Nope, I use FireFox and am happy with it. Never really tried Opera.
I was just pointing out that graphs without extra info don't mean anything...
And no, I'm not going to be all fanboy over FireFox. It's just a nice browser, nothing serious in life...

And you're a Java-hater or disliker, Seventh?

The example map wont help him if he doesn't have Hl 1.1.1.0!

It's a good thing to go over to Steam. Always up-to-date HL.

But you've got a point there: anyone with an older HL version won't see those ringbeams. Might be something to put in the readme.

Err... a little search before posting would help, really.

Look, there's a nice example map about ring beams, and a map from base probably refers to competitions that give you a certain map that you have to use as a 'base' for your entry.