Please ask your basic (newbie) questions in here.

[Flux]

It's not about being paranoid. I do clone myself, keep in my mind which actors I do. However, back in the days, for any unreal engine, I've seen people banging their head on the keyboard for a long time just because of this copy-paste thing. The chances of this ridiculous bug happening is low, but I see nothing wrong with this kind of seemingly implausible warning since people are quick to jump into conclusion "omg this editor is so buggy, I'll stick with..."

You know, so many people pop their head in, ask a few questions then never release anything...

Also, hence the rank of "resident know-it-all", I'd like to hear your comments about this
Reply in a separate thread if you wish, I'd prefer that actually. I remember discussing about lights/level limits and missions being playable on low-end rigs some time ago with Judith, you and me.

[Beleg Cúthalion]

@AI running into each other: I solved this by placing one guard further away from the first patrol marker...rather simple, but I'm eying some "following script" anyway. Just simpler and bullet-proof. Oh...at least I hope so.

@havoc: If you're indeed such a technical freak; T3Edding still offers a broad field of exploring unlike DromEd for instance. Approximately two years ago there must have been a lot of nerds around here digging deep and creating rough drafts of rope arrows and swimmable water and then most of them left. Now it seems to me that people would have to start again where others left years ago. So, if you feel underemployed...

Oh, and about the story (and I fear my technical skills are even lesser than what I think about stories or gameplay) – this Garrett-has-been-trapped-and-imprisoned story has already been subject of a couple of TDS FMs (and there are few!), so right after the story about that mystical artifact this is the most often used type until now. If you don't mind me meddling in your affairs, I'd rather suggest something like breaking into a prison like you did in Cragscleft. Simply because I haven't played a lot of FMs (at all and) like that yet.

[Ziemanskye]

Performance things are a bit of a pain in this engine, because most of the people who are making missions have higher specs than the people playing, as far as I can make out. I'm on a fairly weak system (as I tend to harp on about), which means I'm rather more paranoid than most people are: I genuninely can't play quite a few of the FMs that have been released just because of how they've been built for a bigger rig. It's been a bit of a stick I've been beating the other Cabal guys with for a while now, but they're learning how to appease/distract me

His specific problem, with it only happening around certain flames, probably just means that it's either a light too far, or more likely a weakness in the graphics card/chip itself with regards to particle effects - my previous system was a bit like that; would crap out over things that shouldn't cost much (that was a GF5500, I'm on a GF6150go now). The type of thing where shaving a MaxParticleCount property from a 10 to a 7 in all the emiiters in the scene would probably avoid the slowdown just because there's a little too much overdraw for that setup to cope, and would barely affect how it looks either.

Best way to test things is to get someone with a weaker system to have a poke about for you and see where it sticks/stops/chugs/whatever. Keep him on a list somewhere for beta-testing your next map

[Flux]

Thanks for the detailed explanation but I was actually referring to the fact that even someone with geforce3 card can say "I enjoyed this level", and that level happens to be largest one up to now and its diorama of "toggle lights that touches too many objects" is completely red. And I can also say it is relatively complex with regards to "number of visible shadow casting lights per scene, number of ai & patrol points"

Seriously, when I shipped first beta version of Valley, I knew all my beta-testers rigs and if someone told me by then a player like Sting could play it one way or another, I wouldn't believe...

Of course he'd have problem around flames, everyone does. What surprises me he can still play it. What really puzzles me, you have gf6 and he has gf3... And you say you can't play some missions...where he can...His cpu isn't bright either.

Maybe the Cabal guys should start a uproar against your stick.

Let's see how different players will respond to my query of "performance feedback"...Not that one level should be a benchmark but the case piques my
interest if that level almost does what people considered as blasphemy regarding optimization before. Add to that, I have ambient lights with 200-255 radius in every scene, which are not "ambient-vertex". Yes, I'm a normal map junkie.

[Judith]

Maybe the Cabal guys should start a uproar against your stick.

No worries, the revenge will be sweet once he gives us playable version of his map for beta tests

Add to that, I have ambient lights with 200-255 radius in every scene, which are not "ambient-vertex". Yes, I'm a normal map junkie.

It's hard to imagine how your layout looks like with "lights that touch too many objects" filter turned on.. Is there a scale for that?

[Flux]

Heh, looks like doom3 hell level.

http://images.bittersense.com/valleyred4.jpg

More ambient lights with less radius didn't change anything the on framerates at all, so I left it that way. I know hard to believe, in many areas setting them to ambient vertex only effected 1-2 fps and that was on only some obscure angles. So they are all omni-no shadow.

[havoc211]

@AI running into each other: I solved this by placing one guard further away from the first patrol marker...rather simple, but I'm eying some "following script" anyway. Just simpler and bullet-proof. Oh...at least I hope so.

@havoc: If you're indeed such a technical freak; T3Edding still offers a broad field of exploring unlike DromEd for instance. Approximately two years ago there must have been a lot of nerds around here digging deep and creating rough drafts of rope arrows and swimmable water and then most of them left. Now it seems to me that people would have to start again where others left years ago. So, if you feel underemployed...

I largely just love problem solving. I have a story to tell and the engine is how I go about telling it. So if I need something, I'll stand the engine on it's ear if need be to get it done. If it can't be done, I'll punt. lol.

Oh, and about the story (and I fear my technical skills are even lesser than what I think about stories or gameplay) – this Garrett-has-been-trapped-and-imprisoned story has already been subject of a couple of TDS FMs (and there are few!), so right after the story about that mystical artifact this is the most often used type until now. If you don't mind me meddling in your affairs, I'd rather suggest something like breaking into a prison like you did in Cragscleft. Simply because I haven't played a lot of FMs (at all and) like that yet.

Well, I have 3 dvds full of FMs that I haven't touched yet. I don't know how they all implemented their breakouts. Mine isn't so much about breaking out as it is about Garrett finding the plot on the way. It's just a convenient means to an end the which idea I liked. Has a lot of room for some laughing that I expect to plumb. To that end, it's kindof like westerns. They all tend to be alike in many ways; but, who would turn their nose up at another 10 episodes of Deadwood.

To me, it's not what the plot is, it's how it's executed and how well the story is told. If it's compelling and well constructed, it'll be worth it. If not, it'll deserve feeding to wu's pigs and perhaps the recounting of that tale by Benny.. after a few pints of course.

[Abru]

@havoc: If you're indeed such a technical freak; T3Edding still offers a broad field of exploring unlike DromEd for instance. Approximately two years ago there must have been a lot of nerds around here digging deep and creating rough drafts of rope arrows and swimmable water and then most of them left. Now it seems to me that people would have to start again where others left years ago. So, if you feel underemployed...

I think the Most important thing for us is 3ds-->Tim Converter

http://www.ttlg.com/forums/showthread.php?t=96152
http://www.ttlg.com/forums/showthread.php?t=94544

… and simply creating really good missions.

[Beleg Cúthalion]

...and a custom tool to create matlibs without 3D Studio Max, if that's in fact still on the agenda.

[havoc211]

I think the Most important thing for us is 3ds-->Tim Converter

http://www.ttlg.com/forums/showthread.php?t=96152
http://www.ttlg.com/forums/showthread.php?t=94544

… and simply creating really good missions.

lol. Got my photoshop manuals dug out this afternoon and got it installed again. I think my old version of 3dsMax is only 5.0. I'll have to install it to check if the updates are applied for 5.1, I don't remember and the disc faces are soot damaged. They survived a house fire.

Is there a list somewhere of all the plugins needed for these two apps?

Oh, and got your tools this afternoon too, thanks

[havoc211]

Ok, all is well in mudville after 2 days of tearing the house apart looking for my install sets, running imagecfg on everything that moves, installing everything under the sun.
Two days off and most of it so far has been to accomplish this much.

I found my 3ds install set.. it was 5.0. Luckily, Autodesk had the 5.1 update and the SP1 upgrade both still on thier site. Been to Nvidia and now have the tshirt for that set of install woes from them.

Photoshop did not want to take the dds plugin. After a stern beating.. she's working. Exported my first DDS file a little bit ago to a huge sigh of relief. Now just have to figure out how to dump it into a lib file and I think I'll have found my stride.

Did a lot of reading overnight too.. I may be approaching knowing 1/2 of 1 percent of what I need to; but, should make 1 percent soon

[Ziemanskye]

There's a lot less to this than you might think, unless you do actually do try to make the conversion tools we could really do with having.

And you don't strictly need Photoshop for this, there's DDS tools for most things, and command line versions in a few places, but currently, Max 5/5.1 is sort of non-negotiable if you want to make your own custom content.

[Beleg Cúthalion]

DDS-Converter for instance.

Oh, and I had one more freeze yesterday when I was supposed to be in bed already, far away from any suspicious patrol. But left aside that it only makes the patrol issue an indirect cause, it told me...nothing.

[havoc211]

There's a lot less to this than you might think, unless you do actually do try to make the conversion tools we could really do with having.

And you don't strictly need Photoshop for this, there's DDS tools for most things, and command line versions in a few places, but currently, Max 5/5.1 is sort of non-negotiable if you want to make your own custom content.

Well, now that I have ps installed, you tell me lol.. just kidding. I used to use ps quite a bit; but, haven't had a real need for it, so it's stayed in the box collecting dust since my old place burnt down. That's been a little bit.

3ds is givng me fits though. Been fighting with it for hours and think I finally localized the problem. Imagecfg apparently is tripping the licence manager and then won't allow the app to run at all. I played with it and it worked fine until I ran the affinity tool then it exploded on me again. After that, Windows decided that moving to a dual core processor and main board was too much of a recent change or something so now I have to reg windows ... AGAIN. I keep asking myself why I haven't installed Fedora Core. MS SUCKS. Anyway, gotta reg and try a reinstall again and see if I can limit affinity through taskmanager without crashing max. If not, I'm pretty well sunk there. Only have one machine in the house that isn't a 5600 dualcore and it's a bleatin junk intel celeron with shared graphics mem.. ok, I got it because it was practically new and the guy wanted 50 bucks for a 2.5 ghz machine. Good web browsing box. Lame otherwise lol.

What was that about building utilities? rofl. Maybe I should just break out my old borland 5 package and punt?

[havoc211]

Ok, see what you think. Question involved.

Counter-1
Counter-2


..This is my kitchen counter. It's functional; but, it's expensive to the map size. I started with a map size of 486k
This is the third attempt at construction.

My first go round approached it using individual planes added
and cutout to fit together. That pushed my mapsize to almost 1800k. This go round I used a single poly with two
cutouts in the sink area, another added poly for the counter
top, and another for the sink. The counter has one cutout. The sink has two. With the four doors added in, Map size is currently 1030k.

Is this typical of the system? Some of the static meshes are far more involved, seemingly. Should I use a mesh to do this? I'm using 3 textures right now.

Anyway, off to take another more slimmed down poke at this.

Update:

Counter-3

This is the same layout without the cutouts for the space under the sink.
It uses a single poly for the main counter, 2 poly cutouts for the countertop lip, 1 poly for the sink with 2 more cutouts.
And finally six poly doors added back to the cutout area in front.

This gets my map size down to 978 but leaves a noticeable floor and wall seam.

The plumbing not being there is part of the story line, so; I'm not terribly worried about it. But those seems I need to deal with
or find another approach.

My initial thought was to kick it back from the wall front and back by 1 grid. I actually did that.
It increases my map size to 1005. I also tried using built in meshes in odd situation and adding
a countertop to them and sink. I lose the seams but, I still end up at 979 on my map size. And
I inherit a problem of doors that don't match. So, looks like I've reached the end of work on this
for the moment.

[Ziemanskye]

Map filesize (especially the unr file) really isn't something to worry too much about in this. A compiled *.ibt file for the level will be 50-100Mb after all, and the *.gmp file should be smaller than the *.unr you're building from anyway, since it's the same thing but without all the overhead of the editor-readable gunk

But I have to ask - is that all BSP work? BSP is expensive to the file size, where a static mesh is basically a point in space and a name for the object to place there. If you want to build it with BSP though, I'd recommend doing this: build it in an empty box area (not touching anything), texture it as you want, then make the builder brush bigger than the whole thing and Intersect. Put the funky shaped builder brush in at the place you want the object, and Add Special - and set it for additive Semi-Solid. That'll stop it from ripping up the space quite as much, so it should look a little neater, and cause fewer problems later on.

It's also a *really* bad idea to do complicated things like that with BSP - after a while you'll find yourself with dead-space areas you just can't build in, missing faces, BSP-holes and other ickiness: the flesh renderer is really not so hot at BSP work, but smeshes, well, they don't cause holes in the world. And you can have upto about 10,000 triangles per mesh (I'm currently making entire building exteriors in about 2000-3000, so that's plenty for most things), and you can fill the screen with them if you want. Still cost's you elsewhere, but it causes less problems overall.

[havoc211]

Map filesize (especially the unr file) really isn't something to worry too much about in this. A compiled *.ibt file for the level will be 50-100Mb after all, and the *.gmp file should be smaller than the *.unr you're building from anyway, since it's the same thing but without all the overhead of the editor-readable gunk

But I have to ask - is that all BSP work? BSP is expensive to the file size, where a static mesh is basically a point in space and a name for the object to place there. If you want to build it with BSP though, I'd recommend doing this: build it in an empty box area (not touching anything), texture it as you want, then make the builder brush bigger than the whole thing and Intersect. Put the funky shaped builder brush in at the place you want the object, and Add Special - and set it for additive Semi-Solid. That'll stop it from ripping up the space quite as much, so it should look a little neater, and cause fewer problems later on.

It's also a *really* bad idea to do complicated things like that with BSP - after a while you'll find yourself with dead-space areas you just can't build in, missing faces, BSP-holes and other ickiness: the flesh renderer is really not so hot at BSP work, but smeshes, well, they don't cause holes in the world. And you can have upto about 10,000 triangles per mesh (I'm currently making entire building exteriors in about 2000-3000, so that's plenty for most things), and you can fill the screen with them if you want. Still cost's you elsewhere, but it causes less problems overall.

Thanks for that. That's exactly what I needed to know.
If space isn't a huge issue, then I'm good for the moment with the bsp build. I wouldn't mind a mesh to dress it up a little more; but, I'm a little burnt on it for the moment lol.

Someone mentioned rope arrows recently. I'm a little puzzled. Has anyone noticed that the class is defined in the
gamesys file? It looks to have had at least part of its data removed (mesh?), but:

classname:
Class: D_79

...looks like there's more data in there to collect. I didn't see a helptext string defined; but, I ran through pretty quickly. There also seems to be scripting support for it.

How long has it been since anyone looked at it?

[Beleg Cúthalion]

Two years ago as far as I know. What came out in Krellek's Labyrinth looked like that (minor spoilers, I made one or two cuts because of the demo effect – these invisible ladders are difficult to climb; server may be offline during the night in Europe). Just ask the forum for rope arrows and it will show you some of the earlier threads. My latest idea about that, by the way, was to make the ladder rope-textured to create a rope ladder which would be easier to get on.

[havoc211]

Two years ago as far as I know. What came out in Krellek's Labyrinth looked like that (minor spoilers, I made one or two cuts because of the demo effect – these invisible ladders are difficult to climb; server may be offline during the night in Europe). Just ask the forum for rope arrows and it will show you some of the earlier threads. My latest idea about that, by the way, was to make the ladder rope-textured to create a rope ladder which would be easier to get on.

Gotcha. I did a search through the binary with a hexeditor. Looks like there was an easier way to find the info.

Hmm. Read through both threads. I didn't see it mentioned, but did anyone think to make circular rungs attached at the centermast and showing only the center mast? Would give you the rung element, make it frobable from 360 degrees... just a thought. I'm new to 3d construction; but, critical thinking is my strong suite.

Doesn't look like they were far from solving the problem. Just seems they gave up too quickly. Some of the discussion seemed to portray to me a lack of *grasp* if you will. Everyone seemed so busy with the physical object "rope" that they didn't *seem* to stop and consider that the physical presentation of the rope was immaterial or secondary. Feel like I'm a kid in the matrix; but, you need an invisible volume that is climbable allowing a 360 degree rotation around the cylindrical "solid" with a visible meshed "rope" running through it. The rope is for appearance. The construct is the issue.
The best thinking that I saw was the person that remarked on making it a climbable surface.

To me, if you could give Garrett the proberties of the climbing gloves (if it's not just a prop ..haven't looked) You could dispense with wall climbing or simply turn the gloves into a prop that must be worn (if they do have properties applicable to task), and limit to rope climbing and using the climbable surface on a solid invisible cylindrical wall mesh.. with a rope running through it of course.. lol

Again, talking in terms where I'm not fully capable. It's been a long time since I did any modelling and that only a whim. I'm an artist, not a 3d modeller. Any of this make sense to you? Is it practical? I feel a bit in the dark on the
modelling aspects; but, talking through it might help bring some more ideas to light .

To me it would seem the starting point might be to find the smallest surface Garret can climb (width), rotate that into an 8 plus sided column and see if he can climb up and around it. If that works, the next step would seem to be to find out how tall a solid garrett can climb or if he can climb multiple short solids linked together like a chain of pearls

---[]-[]-[]-[]-[]-[]-[]-[]----

That's kindof crude and on the horizontal; but, should illustrate the concept. For sake of animation, garret would seem to hold the center "rope" and it would be modeled that way.
The solids would be transparent and "climbable". Not sure on the scripting entirely; but, I'm
also not entirely pleased with the script system. Poor design IMHO. What kind of system gives
you logical operations with no "if/else" option. The power of the script language would increase
monumentally with that one feature added. But, what do I know rofl.

Also, there is a rope hanging from the bell in the hammerite Cathedral that could be looked at for some practical experimentation. And is there a weighting system in the physics engine that might employ gravity to pull a rope down vertical by means of attaching a weight to the off end? Don't know what all was tried and monkeyed with; but, ideas just keep coming..

Anyway, perhaps this could provide a fresh set of eyes and direction? Any comments? Ya'll have my leave to be brutal. I'm just plumbing the depths here for something we can use. There's always a way.. unless there isn't one lol. [/QUOTE]

[Ziemanskye]

Okay, I'll start by admitting I didn't read all of that.

But first here's what I understand of why the rope arrows were never "complete": the climbing code works based on a static mesh - it doesn't really care what the mesh is, but it does care about the origin (and consequently, what direction the mesh is "facing"), and it's outer collision area. Hit the mesh from the front/back, and you go into climbing mode, hit it from the sides and you don't, walk into the top of it and you transition down into climbing mode. If you can climb it or not is a property you can add to a smesh.

One (relatively) simple bit of unrealscript (and some slightly icky math) could have made the rope arrows work, if we had the ability to use it: just make the "rope" mesh always rotate to face the player's position (on the right axis, of course). But we can't do that, and I don't think we have trigger-script actions to do it with the scripting tools we do have. I will admit that I never actually considered the lack of rope arrows that much of a loss, so I genuinely haven't looked.

Second - yes, the scripting system is a bit borked. The lack of If/else is kind of false though - there's loads of "if", and you just need two different scripts for "else". I'm not really arguing this with you though - It would have been nice to not have to do it the long way and involve a bajillion extra variables/flags/messages between objects. I'm not a good programmer, but I can do it, so the lack of things like that seemed kind of dumb/weird to me too. Just thought I'd say: if you want to do that kind of thing, you do it the hard way, but you can still do it. (I think one of the devs showed up around here once to say the lack of UScript was attributed to the extra overhead the interpreter needed on the Xbox, and they were already out of memory/processor and cutting every corner they could find)

Third, the climbing gloves hack you suggest would be a huge thing to do. New Garrett/animations for a start, but you've also got to mess with a lot of the materials to take the Climbable properties off. As for "smallest" surface he can climb, that's probably going to be a 32x32 face given that's just under how wide Garrett is (he's 33 units, in my experiments), but that's just a guess: I didn't lament the loss of the ropes, and the climbing gloves weren't much more of a feature, yay or nay, to me.

Finally, as for pulling the rope down, well, I suppose you could, though you might have trouble stopping it again, but as far as I remember, Ascottk's rope arrows did actually solve the issue of how long to make the rope. Again, that was a while ago and not something that interested me much, so could be misremembered.

[havoc211]

Ok, that puts more perspective into it.

So, they hobbled the capabilities of the game engine to get it onto an inferior platform and rendered the pc market as useless as the console. I'm so mad right now I could ____.

I spent six hours doing climbing tests the other day, at any rate. Wasn't a complete loss; but, it didn't give me great confidence. I built a rope, attached it and tested and was able to climb it.. only to find that I could only climb in one direction. That's bad. It's cludgy too.

I also setup differing sized volumes with appropriate climbing textures and tested them. The smallest I got down to before quitting for the night was a steam pipe
mesh. The intriguing thing was that if you skin it with a
climbable texture, you can climb up and around it the same way the old rope worked.

I also tested the invisible volume around a rope theory.. If you set the property of the volume to "invisible" you can't
frob it. On the one hand, that's bad. On the other hand, there's also no collision or no result of collision as long as
the volume is invisible. So, my next thought is to use transparency and modify collision aspects. Maybe setup a transparent texture with climbing capabilities.. I still need
to look at that steampipe though.. need to find out how much smaller we can go and still get that effect.

I tried texturing a wall with a climbing texture and could do nothing with it in the engine. So I may be sunk for testing ability myself for a bit. But I was thinking of exporting the rope texture and reimporting it with climbing attached to it.
May have to redefine the material; but, ..

Another thought was to import the new sectional pipe meshes and play with them. The ladder approach doesn't appeal to me at all. It just doesn't give the feel of the original arrows or the functionality. I tried ladder combinations myself in my testing and hated it.

Gotta go to work, there were some other things that were coming to mind; but, being pressed for time, they don't seem to want to out for me at the moment. I'll check back in tomorrow. But that steam pipe showed a lot of promise. Had to have the climbing gloves to use that idea; but, there might be a way around that in scripting.

[Beleg Cúthalion]

And now we'd need some moderator to get these things moved.

I have a little problem with a phantom fire. I once placed an oven behind some arches to check if I would get the desired fire shining through to make it look cool etc. ect.. However, after deleting this thing again, the light remained although there is nothing to mark and delete left. I already switched all the actors to "off", built all, switched them on again and built again, but this light was still there (no pulsing or flickering, though). It's not that it's completely out of place (since I wanted this kind of effect anyway), but for order's sake I want to get rid of it, furthermore since it is in the largest area of the whole mission.

[Flux]

No, we don't need a moderator for this, simply overcome your anxiety about posting new threads...

Ok, I don't remember what this phantom fire is, is it one of those emitter actors?
Anyway, when the same things happens to me, I add another same actor, then select all matching actors, then delete them all. When you do this this "invisible" actor gets selected also even if it doesn't show up there.

Hopefully, you don't have other phantom fires in your level that you don't want to get rid of.

Sometimes, saving and re-opening the level, helps you get rid of inivisible actors.

[Beleg Cúthalion]

With "phantom fire" I just meant that I cannot catch it but it's still there. But I'll try what you suggested. There is only one more oven like that in the mission; I guess I can overcome that. Or reload.

[Flux]

Just remembered. You can also use "search for actors" thingy. That binocular icon next to undo-redo.