Forum posts

I'm very unhappy I missed that.
On the other hand. The Grand Canyon is nice this time of year

So what are we looking at in that screenshot? I'm a little lost as far as details go.

Yeah but arguing without making any progress is a waste of time
Video tutorial coming soon HERE - CLICKY because I'm bored.

Relax guys, we're here to help, not argue. I just tested a new texture and found the last color doesn't HAVE to be blue.

Steps i took:
1- Make texture
2- Open in wally
3- Change last color index to pink
4- Paint bucket the transparent areas with pink
5- Re Mip
6- SaveLet me point out however, that when i made the texture in paint.net, i used a single orange color initially.
When imported to wally, there were 5 different shades of orange. Only one of them was the last color index, and there were 4 shades hInging around the middle of the pallette (which i then removed). When i paint-bucketed over the orange spots with pink, i found some rouge orange pixels faffing about that i then penciled over. I would recommend this method to ENSURE you have the actual last index.

You collide with the func_ladder entity though so I recommend making the ladder detail a func_illusionary to save clipnodes.

but its RGB values don't matter

stu is correct, but to expand on his note, RGB values of 255, 0, 0 are the most common.

Open the palette in WALLY and make sure the last color index is your transparent color.

you can decompile, import the mesh to blender, change and re-export.
You shouldn't need to change the skeleton or animations.

thing is you get 3 rockets when equipped. I know you can run custom scripts but i have no idea if you can directly affect held ammunition with an entity-fired script.
I couldn't find anything helpful on the VDC and you apparently can't directly modify variables with scripts (i tried)

Orange you glad it's not banana

I can't see any options for that. It might be a hard-coded event.
There's no other options for ammo counts, or an 'empty' flag.
Perhaps just equip the Launcher later?

Shoot TWHL a link too man we should all work together to support the modding community. I'll check it out!

I have been doing a lot of modelling for GS recently.
What problems exactly are you running into?

The info landmark needs to be in the exact same reference position.
Say your level transition is a hallway with a crate in it. Both levels have the same hallway and the same crate. As an example, place an info_landmark 16 units above the crate's center in both levels and you're good to go. The info_landmarks need to have the same name.
You don't need a crate per-se, but it's a good tangable landmark. The same can be said for walls or doors or some other detail near the level transition; pick an area common to both levels to plop your landmark down near.

I write my levels like a paper.
I write down a one-line main idea for a map, then i break it down into the major sections
each one of those major sections gets details.
once you have the script down, you move to paper.

For example: Hero must avoid capture and launch a missile.

• There's a valley in there me must cross
• There will be heavy resistance in the form of [enemy]
• the missile is located [location]
• the player starts at [location]

then you build each of those ideas and tie them into eachother.
Just basic overhead views with maybe some perspective sketches in there to set the theme.
Once you have your overviews, you just start mapping what you have on paper.
block everything out with basics, add 1 level of detail then start doing scripts and triggers. once those work perfectly, i usually start adding in more detail and messing with lights.

Now you don't have to follow that structure to the letter because i always deviate from it. But a wise man once told me "You always need a plan because you need something to deviate from."

to me, the more creative time i spend doing prep-work, the easier it is to map because i can spend time mapping what i have already created instead of creating something from scratch in-editor.

technically the 'auto'snipers are semi-auto snipers... the terminology stuck after 1.3 ish

I'll be in Arizona next week. Going to visit the famous Grand Canyon. I'll bring my Nikon and my zoom lens.
Expect some mediocre pictures soon!

Well i synced all my files to my laptop so if i find downtime on vacation im going to try to make progress. It's really just setting up all these damn triggers. I always pick ambitious goals.

I might withdraw. I'm out of state for the remainder.

Well there's been a lot of progress in community-coded goldsource compilers. Source actually has smoothing groups and lightmap control that contribute significantly to the look and shading of world brushes.

Not taking anything away from ya, TJB - that looks fantastic!

This was answered yesterday, oddly enough...
REFERENCE THREAD

Check out monster_furniture used with a scripted_sequence

I"m not 100% sure if it'll work in CS though.

I hope to have 90% done tonight.
hope

Having some issues with 2+ bone animations.
Turns out my tutorial was a fluke and only worked because it had only one bone.
well shit.

I'll have to figure this out but i think it has something to do with the SMD converter, everything was working fine with milkshape but im not paying for that

Thanks Dr. Orange. I had no idea

Hopefully I can get this done tonight.
I might make one of those chrome trashcans with the plastic black trim that opens when you walk up to it and step on the foot pedal.but square.

This will be an interesting project to complete because I'll need to create the following assets:

Trash Can Mesh Parts
Bone structure for animation
Idle Animation
Open/Close Animation
Chrome Texture
Plastic Bag Texture (Possibly Chrome to be shiny)
Rubber Texture

The only thing I'm unsure about is controlling the animations in-game. I know you can use env_renders to toggle visibility of entities, however I don't know of a way to specifically start an animation. I might do some research before I start.

Ok so would you guys prefer a real-time narrated video series or a single time-lapse one?

To Update this thread for reference, Open the model with Jeds Model Viewer, Tick the ADDITIVE option, and things become transparent.
Transparency is based on the amount of black in the texture. more black(darker) = more transparent

this is all necessary to get the door from lock state and letting it be unlocked!

False,
Globals are variables stored between levels; unless your button and door are in two separate levels, you don't need that.
Not to sound rude, but it's much much simpler than that

My example will work, just make sure to have the button reset to -1 to it stays pushed.
Optionally, if you want a button that resets, you can trigger a relay into the 'on' state, and have THAT trigger the door.
Simple Example - http://twhl.info/vault.php?map=6046

Also, Since we're using Spirit (for TWHL TOWER), you can follow Urby's method, but i'm pretty sure Spirit is backwards compatible with Vanilla

@Urby, get me some sketches/references/dimensions, and I can
Captain, I figured out model transparency doing the Alien Pods, so grab me some reference material and I'll see what I can do.

@ Ninja - A fog model? probably possible but I would think sprites are still probably better for fog. It might be possible by mapping a transparent fog texture to a bunch of slow moving rotating spheres - but then it might look like a bunch of slow moving rotating transparent spheres... Sorry I don't have a better answer.

@Crypt - Goldsource has a hard entity limit built into the engine - which you probably won't hit with detail models. Your bigger concern is the Epoly drawn at any time (which I have gotten 14K+ without any issues) - I'd say use models wherever you can. Keep in mind that models are lit by their world origin. Source has the advantage with better looking vertex-lit models.

So I can probably do a demo video sometime later this week. I'll just start my screencap software and start modeling away. I'll speed up parts of the process to keep the video between 10-15 minutes, or I can split it into parts.

Yeah no problem I'll whip something up.
Any requests?

Watching these again, I missed some points I wanted to cover, so i'll make an addendum video highlighting some of the finer details I overlooked. One thing I don't cover, and won't really get into is how to model and rig in blender; It's outside the scope of the tutorials. There's countless tutorials on youtube for that, and if you're looking for those I can link to some I found helpful.

@ Captain, Yes that's the plan, I first need to organize everything in a more professional format - linking to a playlist isn't the best way to present the data. I'm also going to make a verbal list of notes that follows along with the videos and maybe expands on some topics that might not be as clear if you're not watching in 1080p

@ Crypt, YES It's possible and very simple once you get the flow down. Again, if there's anything you're not 100% clear about, or if you're running into issues, post here and i'll help you fix it. I monitor TWHL pretty much hourly (if I'm awake)

@ Urby. Thanks! You're too kind. Now I expect to see some kickass models for The Core.

Hey all!
I just finished up a video tutorial series on how to export blender models to goldsource.

I'm hoping this thread can answer any questions people might have or add some additional points of interest that I might have missed.

Enjoy!

http://www.youtube.com/playlist?list=PL9CPpo_1svGNnZI1iPL2peExsuPk3gYWN

jesus that's faced paced. My first thought was "OMG THE BUNNYHOPS"
Seems interesting?
It's like capture the scientist meets natural selection (research and weapons) If i gather correctly

Ninja are you doing this for Source or Goldsource?

Multisource for masters, not multimanagers
Last paragraph

Name your multisource: "Door_master"
The door doesn't need a name
In the door's "Master"field, type in "Door_master"
Have the button trigger "Door_master"

Done.

still though. for a fast-paced game you don't want to get caught on fancy geometry. The best bet is to clip it off anyway

Tet, would the simple action of just using Bevel textures rather than null not massively help on the clipnode side of things?

I haven't the slightest clue. I'll have to make a test map and check the compile log.

Oooh right.
Forgot about clip economy.
Im not sure about detail vs wall

Captain, I recommend reading up how the compile tools actually split the map, and how the engine renders it. If you understand the framework of the engine, it's easier to control.

We ran out of clipnodes - Func_illusionary all the things!
We were sky-high on wpoly - Hint all doorways, and split vertical sections. Func_illusionary-ing complex geometry and covering with basic clip brushes helps this.

But wait, Tetsu0, I thought the new func_detail is a great new addition? What about that?

I have actually found out through testing that passable func_details generate clipnodes and func_illusionaries do not. It was a lesson learned too late but I created a test map to check the differences and was amazed.... All you have to do is look at the compile log to see this. (it tells you what % of clipnodes, models, brushes, etc...)

I also increased all floor and ceiling texture scales to 1.5.
Any un-accessible areas were heavily nulled, func_illusioned and I also increased THOSE texture scales to 1.5-2.0 depending on distance to player.

Our issue stemmed from throwing in massive amounts of detail all at once and expecting the engine to take it. I'm actually impressed that the engine stood up to so much. I'm even more impressed that a handful of Hint brushes took the highest wpoly count about 65% (looking straight across the map). It only took an evening or two (probably 4-6 hours) to get the map running smooth. Sledge's auto visgroup function was ABSOLUTELY ESSENTIAL to my progress there, so Penguinboy, you really REALLY have my love.

I have a great deal of respect for the goldsource engine after making that map - it was also a fantastic learning experience - I hope i have the time and motivation to make a tutorial or a write up on how I optimized that map. I think everyone on this site should have those skills. They also apply to source

I'm woking on this to the tune of one room a night. I"m still on basic details and triggers.
The extension is welcome.

After that last day making the map, I went to sleep and saw hint brushes XD

Woo Hoo!
Thanks all

Interesting read Potatis, thanks.
I'm now cursing my professor

I stand corrected.
This probably isn't the thread for it, but I wonder why you would want to declare "std::" more than once. I'm not very experienced TBH so there's a lot I don't know

Sledge functions very much like Hammer, with a few improvements.
There's nothing stopping you from mapping for TF2, once you get Hammer, everything is set up automatically for you.

Ninja's example should have just included "using namespace std" to simplify the cout instruction.

cout is short for Console out, it just prints information to the console.
It really is what everybody starts with.

http://www.fabricsolutionsonline.com/concave-triangle-sign-purchase
half life 3 confirmed

Zeeba!
So the most basic form of programming is Scripting - which you already do.
Scripting and triggering in Hammer/Sledge for Half-Life is already programming. Don't let anyone tell you it's not. So you already have experience with 'boolean logic' (1's and 0's... On or Off, True or False) Think of a trigger_once being a variable with a value of "False". Once the player walks into it, that variable becomes "True" and things happen; such as turning on a light, or opening a door, or firing a multi_manager.
Expanding on that, A multi_manager can be thought of as a function; Once fired, it triggers different events at specified times.

Multi_sources are boolean conditional statements; they evaluate any number of inputs and are 'false' until all the inputs are 'true' (or 'on' / 'triggered').

The hardest thing about jumping in (in my experience) is overcoming the syntax of a new language - it's even harder when you're learning new concepts like loops, logic, structures and variables.

Once you get a firm grasp on programming as an idea, it's much easier to learn a new language/syntax/whatever.

So If you want to make games with code, you probably want to jump straight into C++. It's one of the most common languages and it's prominent worldwide.

w3schools is good, but might I recommend a more game-orientated approach.
Check out 3DBuzz.Com's C++ intro course. It takes you through not only the C++ syntax, but the general Idea of coding.

Aside from that, why not head to your closest bookstore and drop a few dollars on C++ for dummies?

Regardless of where you go to learn, the road is tough. You'll be thrown into the barebones introductions and you'll learn to print things to console, declare and use basic variables, then you'll probably move onto basic loops and functions. It's all very boring stuff and it takes a while to learn but stick with it. There's nothing like a poorly-written SDK (cough - Half life - cough) to turn someone away from coding.

I code regularly in C, and have a few dozen hours of C++ experience. I'm always here to help.
Good luck!

Looks like something from Spore. I like it!