Campaignjunkie

Cool stuff floating around in this thread. Albatros: I agree about the normal maps being too strong. In general, I think you're trying to cram too much surface detail on everything; it's all busy, all noisy, all competing for my attention. I like where you're going with it though, it'll be a wonderfully distinct DoD map. Anyway, here's a small personal singleplayer project I'm working on: You're in this guy's brain and you're exploring his memories; each Grecian urn is a part of a story / memory. It's kind of a Psychonauts-style abstraction in that respect. Not much to look at yet, as I'm trying to prototype the gameplay first (for a change!) but I felt like I should be lurking less and posting more often.

After finishing Ep2, I can't wait for the SDK. My mind is buzzing with possibilities. Plus, Portal maps will be a nice change of pace from the back-breaking epic map releases most people are attempting. I kind of agree with Mazy about Alyx's dialog. A little too much positive gratitude now, especially after some fairly trivial puzzles - for example, when you break the wood planks to send the mine cart down the slope and break through the wood barricade to get to the first transmission station, Alyx says something like "Great job Gordon, I can really count on you, you can always find a way through!" and I'm thinking "Wow, it just took me a few seconds and 2 mouse clicks..." Overreact much? I liked everything else though. Vortigaunt wearing a labcoat = awesome. Magnusson and his nearly complete indifference to your fame = awesome. Fighting hunters (which are really cute, by the way) = awesome.

The lady doth protest too much, methinks. (Okay, so it's Hamlet, but I can't remember anything appropriate from Macbeth)

Okay, define "failure." So they didn't become gigantic successes - not many mods do. They still have passionate fan-bases. Join any SourceForts server and try to learn all the advanced building techniques that have evolved. Watching that HL2 CTF video where Valve planted their own cheats in the code was also pretty entertaining. People still play Adrenaline Gamer. I think the problem is that people are doing "release often" but not following up with "iterate." Okay, so the gameplay isn't interesting - the solution isn't a balance change or a new player class, it's a different set of rules and game modes... But then you risk upsetting the community that has stayed on since the beginning. In any case, I agree that none of this is as simple as Garry makes it sound. And a BMS rebuttal (not speaking for the team though): I agree in principle with "release often," but I don't think we'll get much benefit out of iteration. We're basically just doing a facelift on Half-Life 1; there's not much gameplay to iterate and test. Plus, it's single player. People are going to play it once at most, and I don't blame them.

Exactly my thoughts. So what? Its popularity still exceeds any other hl2 mod's out there. Because modding is all about the popularity, right? Funny... I'm doing it because I like to build levels and think about player psychology. :wink: Oh, and that Black Mesa stab he made; ouch.

I'm in love with this game already.

I'm in the same boat as Mr. Skjalg. The maps and all the art look great, but this kind of gameplay isn't for me at all. An idea that you've probably heard before: throw out all this silly realism and make an arcade mode for the rest of us.

Shoveling through The Longest Journey. The graphics are kind of painful to look at, and don't really have the charm of older adventure games like DotT and such, but at least the writing is pretty decent. Steam also roped me into buying Dark Messiah for $20, so I guess I'll see how that goes.

I think the "one puzzle" is the best idea; it keeps the focus on design / gameplay, and you can interpret it to be a labyrinth or a room where you wait 20 minutes or whatever you want. But then it's also kind of general, which defeats the purpose of a contest to place constraints and foster creative solutions. Sentura: Read up on the history of the labyrinth, first. Merk is obviously referring to the classical labyrinths - most commonly Greek / Roman / medieval.

I can't get over how nice the maps look - it's so painterly. Argh!

Screenshots of the kid's map: http://www.flickr.com/photos/8082096@N02/?saved=1 I think it's actually pretty good, conceptually. He has some good scale and shapes going on - but other than that, blech. Pretty ugly and bland. You have to wonder: if stuff like this can get a kid arrested, what would some of the people on these boards get if they create their own schools? The death penalty?

Well, shit. I was actually planning to map out my university.

Bring it on.

While I'm not sure about video memory, uncompressed normal maps definitely take up more disk space - and it can add up very quickly if you're using lots of custom textures. But in my experience, I think the VTF compression is good enough to the point that you can compress most normal maps. Anyone care to do some experiments?

Thief 3 - Shalebridge Cradle, and its cousin Thief 1 - Return to the Cathedral; both being "difficult" through their scariness. But the Cradle... Oh man. That level is sheer genius. In my opinion, every other level in Thief 3 exists to build up momentum towards the Cradle. I think a guy on PC Gamer wrote an essay about how great this level is, but he kind of glazed over the design aspects in favor of the narrative. There are few levels that I would elevate to the pedestal of "masterpiece" - but this is definitely one of them.

http://developer.valvesoftware.com/wiki ... ing_Models http://developer.valvesoftware.com/wiki/GUIStudioMDL You can generally ignore all the junk about paths and setting a VPROJECT path in the Compiling Models link, as long as you use GUIstudioMDL or some equivalent. It's also kind of XSI-specific, but the general pipeline is always the same: export SMDs, write a QC script, run the QC through studiomdl.

Haven't had any work to post for a while - well here's something now! Right now I'm finishing up what is quite possibly my last HL1 project, a Prince of Persia themed singleplayer climb; it's kind of like a poor man's Lost Coast. The only thing stopping me is that I keep hitting the max_engine_entity limit! Anyway... I'm really not happy with how the cliff turned out, it's really uniformly lit, but I don't have any resources to spare on detailing it any further. Oh well. C'est la vie.

I don't think a tablet will make you a better painter per se - if you can't draw or paint without one, then you still won't be able to draw well, even with a tablet... Or at least that's what I realized after I got an intuos2 a few years ago. It's handy for tracing and such, but in the end it's still just a tool.

It looks nice and everything - now you just need to make a level out of it.

Quite good, though the laptop / keypad gameplay got a little stale after a while. There were some very nice setpieces though, my favorite being the low gravity reactor. I'll probably write up a review on Ten Four for it.

Are you already getting requests to cannibalize this map for the custom content?

Oh snap indeed.

Lots of maps are blocky and sloppy, relying on props to unify the level. If anything, there isn't enough brushwork these days.

Looks good, BUT... You're using the stone trim texture to form a lot of 90 degree angles. It results in a very mechanical feel, like a style of video-game architecture that doesn't seem particularly old or realistic. I suggest making some more trim texture variants and mixing some slanted 20 or 60 degree angles into the walls and layout. In the new Prince of Persia's (especially Sands of Time), look at the structure of the environments more closely - they curve a lot of straight edges and stretch some perpendicular corners to get a more "lived-in" feel to the levels. Sometimes it's very subtle, and almost borders on useless, but it really matters in the end.

I think the func_detail is blocking light from reaching the brush underneath - so when it samples the lightmap under the player's feet to light the weapon model, it chooses the dark world brush that's covered in shadow rather than the func_detail. My advice: Turn the street back to a world brush. Why would you func_detail it anyway?