demagogue

A blackjack-style KO arrow might be interesting if it follows the limits of blackjacking, as in it only works when it hits a specific radius on the back of the head. So there would still be some skill involved in hitting a smallish area from a distance, and if it doesn't hit it but somewhere else or it makes a noise nearby, all you've done is alerted and angered them.

The crouch to slide down a rope or ladder was something to team debated. It was most definitely an intentional feature added to get quickly down a rope or ladder without just falling (using the crouch button, which is the logical one, as opposed to another key for it). I don't recall all the details, but I think the idea was being crouched didn't change the state of the player on the ladder, and I guess they didn't worry about the player coming off of it not crouched since you can immediately crouch when you get off. Is that your main worry, coming off not crouched when you want to be? I don't want to express any strong opinion about it without play testing the two options to compare them in practice.

I'm just looking at the language of the TDM license. (I'm quoting it above, so here I'll just link to it.) I'll change my above post to point to this, since this is the baseline one should start with. There are still some complications one could talk about, but this is the starting point. I'll try to think through it in my above post. Long story short you're right, and I'll change my whole post above to reflect that.

Edit: Given the challenge from Wellington Crab below and looking at the license, I'm going to change this post. I'm putting my original post, which I'm now deleting, on a Google Doc here for reference, so you know what he's talking about. But I'll disown it now under scrutiny. I think some of the same points will come through, but they have to be put in the correct context with the TDM license as the baseline. The main question is what to do as a default when the potential re-user doesn't know the status of an asset. But to talk about that, you first have to talk about the baseline of the license. So here's the relevant TDM license text. So clearly if the author makes a different license explicit in their FM, then that other explicit license will apply. That's the baseline. It also adds this extra clause which goes beyond CC, "Without explicit permission from the mission authors, you may not redistribute, modify or reuse (parts of) these missions or the assets contained within them." This means FM assets are under a stricter license than the CC license mentioned above, with this extra clause. That also creates the default, if you have no information or unclear information, don't use the assets. This may add some practical complications in some cases when, e.g., if the author is using assets from a 3rd source and they mistakenly misrepresent the license (they think it's redistributable when it's really not), or if the author has made statements where it's a little ambiguous about when it can be redistributed, it's a partial statement (the default should probably be assume it's no unless very explicit though). I would say practically speaking, a lot of these issues can be addressed by asking the author directly for permission and getting explicit permission, or if they've made some explicit statement in their readme or a post or somewhere giving explicit permission, and if you can't find that explicit statement, it's all around best to not use the asset and just use some replacement that you know 100% is in the clear. The catch still hanging is when the asset is represented as redistributable when it really isn't. (The original mapper thought it was, but they made a mistake.) In that case, I think it's worth the mapper that wants to use it to do a little of their own due diligence to confirm that it's really redistributable, and again if it's not 100% clear, if there's any hints it might not be, it's best to not use it and use one you know is 100% clear to redistribute. Edit: Okay I guess there are a few more little technicalities. Because of the new clause, one could argue if the CC additions also still attach, especially Share Alike and Attribution (SA: you shouldn't change the license when you redistribute it, and A: you should give attribution in your readme). I think these may be implied, but even if it's unclear, it's common courtesy to do that anyway, and Share Alike probably includes that you have to get permission from the original author, not the redistributor, or if they modify it, both. Just try to be in good faith & respect everyone involved. Another possible little technicality is one might try to distinguish old FMs before the new clause and new ones after (do the old FMs also get the new clause), but anyway aside from the license, it's also probably fair to say it's TDM policy, so it's good to apply it to all FMs no matter what.

I like to talk about it, but yeah it's probably off topic to this thread to get into it. But yes, it has lots of great gameplay too, so I used the entirely wrong phrasing. For the record, maybe the way to better put it is that it's more cinematic, like with the mise en scene and plot flow design, it's telling a story, which I also like. But I also really like ones that are simulationist, like you're just dropped in an uncaring world that doesn't care where you go & isn't really telling a story, but it has things happening in it that you can observe. This is all nitpicking nuance though. I still agree it's objectively great and I was happy it won & got recognized for how great it is.

Right! Great FM too, although I tend to bias on the side of gameplay-heavy FMs over storytelling-heavy ones, and it was definitely the latter. But it was still objectively great & I still really loved it.

This is referring to TDM's collaboration with Square Enix North America at the time Eidos-Montréal was making the game Thief (Thief 4). SE held a contest of fan material. Basically they wanted to have a contest of fan missions for past "Thief" games and fan art. They did end up having a contest. There were a few TDM entries. Some of us on the team got a free copy of Thief 4 for "participating". Well you can see just from that set up that these were the marketing people that really had no idea what Thief really was, and they definitely weren't going to understand the relationship of TDM to Thief IP, although you could clearly tell the moment when the lawyers probably got involved and they changed the rules to be only Thief inspired and they really watered it down since the link to Thief IP itself was lost, and they couldn't be in any way linked to the TDM/idTech4 engine. It probably would have been great if it'd been the Eidos-Montréal devs we were working with, since those would have been actual gamers that understand the game & us, our kind of people. But it was the marketing people, the Square Enix NA marketing people at that, and they were just in an entirely different universe. I mean completely detached from the reality of this game that they're working on. They were clearly only interested in the business side and thought of fans as shallow entertainment consumers that they were trying to hype up, like we were the core of Thief 4's target audience. They had this really unreal positivity about this contest while being so clueless about what it was really about. Our communications with them were just bizarre and otherworldly. There's another story I can tell about working with a team of lawyers that were pro bono vetting our Version 2.0 release to make sure it cleared id's & Eidos's IP (technically I guess also the asset makers of the assets we were using to replace id's) when we went standalone, or anyway they gave us a memo of all the considerations to look out for in doing it. I made a thread about it, but practically I don't know if it changed that much of what we were doing (the problem kind of solves itself just by our gamers' intuition), and obviously we released it standalone and nothing happened, so it worked out. Actually that Square Enix contest did have one kind of positive result for us, which is we got language from them in writing that seemed to assume we were clear of any claim for Thief IP violations as far as they were concerned. I've kept that message on hand in case we ever do get pushback, but fortunately we never have. Edit: I'm looking back on this message and thinking about the NoClip documentary on the making of Disco Elysium, which is still coming out these days, and I recall that NoClip contacted us once about making a documentary on our making. Or somebody did anyway. But I think the inside story of the making of TDM would be super cool if anybody did decide to make it. There are so many interesting stories like this. Well I think they're interesting. Maybe one of us should make it.

It's funny, I was just watching this video ("How Modern Game Engines Degraded — And Who’s to Blame?") and then I saw this post. That's some cosmic timing. But anyway, what you're saying is making sense. It's not even that surprising. Anybody working with Unreal 5 that was working with these old engines back in the day can see the differences in optimization smacking them in the face. For the record, the team has done serious optimization work compared to what we started with, and the examples you mention are mapper optimizations, not engine. The team is pretty small, and I don't know if they have ambitions to take the engine other places. But anyway the engine is under GPL3 and nothing it stopping people from taking it and running with it. And people have. Blendo's game Skin Deep uses it, or a version of it (as I understand it; we're in the credits), and you can find people posting about their own projects on it. I think people would encourage and help out any big project doing the kinds of things you're talking about. Like most everything, it mostly comes down to who's gonna be a champion for it. Edit: Another funny coincidence, that video I posted at the top is talking about Cryengine as its example of the better optimized older engine, which is the engine that started this thread.

Things are slow enough around here that I didn't really notice , but welcome back anyway!

Well I think any outreach is good. Most users won't be that active, but a few will, and that's a few you wouldn't get otherwise. So in principle it's a good idea. But again, the issue is that despite the name, it's not actually a Doom3 mod anymore, not even really a total conversion. It's a separate standalone game that happens to use its engine as a base, and even that has been turned into essentially a new engine too. Edit: In the big picture, the issue is that we've already been in touch with both Steam and GOG, and there were issues getting TDM as a standalone game on both of them, about which the team basically voted we didn't want to go through the trouble of fixing, at least for the issues involved at that time. We're not really the kind of team that wants to formalize into a corporation to get on Steam, and I forget the specifics of GOG now, but IIRC it was them turning us down, or anyway stopping replying to us, more than the other way around. But again in principle, or at least I personally think getting onto GOG is great if it's ever possible.

Yeah, we get this kind of post every few years, about porting to whatever shiny new other engine is out there. For reasons already discussed, "porting" isn't even a good term for it, since you'd just be making an entirely new game practically from scratch. Just look at the size of our source code. The reason we moved to the Doom3 engine it in the first place was only because Thief Deadly Shadows didn't allow us to properly make good FMs. But now we can make FMs and it's open source. If there is some cool tech that the CryEngine or Unreal engines have, it's gonna be about infinitely easier to just port that cool tech into the TDM engine than vice versa, and in fact we've done that for a lot of cool new tech like soft shadows and ambient occlusion and the like. So that's really the way to go if you want to update the tech. And we're open source so you can do that. Anyway, all that said, one can still make a cool new TDM-inspired game in CryEngine, and I'd even encourage that to someone up to the challenge. For example I'd love to see a cyberpunk or modern stealth game, and that'd be good project for a new engine. I just think if they're making an entirely new game already, why not just make an entirely new game. Well one thing that might be cool is if someone made a game in another engine that could parse a TDM .pk4 file and open up as a map in itself. So then once you've made it, it already as all of our maps ready to play in it. I have an idea that that might not work well though, especially for custom things people put in the map. Also again, we can already play those maps now. If you're going to all that effort, let's have a new cool game with new cool maps!

I think practically speaking the stairs approach is something mappers can do that'll just work. The AI should follow the player in tandem, so I think if they weren't specifically directed to go opposite directions on the stair by path nodes, then I don't think that problem will happen too often if at all in practice.

Just jumping in to applaud Arcturus for coming up with that. That's exactly the kind of thing I was thinking about in my post, thinking outside the box, and looking back on it, I'm almost surprised I didn't think of it myself, although I was hilariously close to it. But anyway, it's a bit buggy, but that can be tweaked, but probably more important is that it's better than what we have now (no climbing at all). The big issue that I see is that it's more of a solution for mappers to do themselves, rather than fit for a general solution for the core package, which I guess is symptomatic of using a hacky approach. But as a hacky solution, it does the job pretty well! I think it'd prove pretty robust, won't it? Wherever the elevator is (if it falls off mid-way like in the 2nd video), the AI will always call it up or down to themselves, it's just a bit more delay. And the mapper should take care of the geometry, in terms of being able to step on and off properly. As for being oriented the wrong way, you know a button can also trigger a script, and I think such a script can just turn the AI pushing it by fiat. (Now I have to remember if the AI ID passes to the script when you push a button. I remember talking about that ages ago, and I think it was dealt with in some way.)

I think it's similar with the problem of flying AI. It'd be a fundamental rewiring of pathfinding to add z-axis motion completely separate from the x-y plane, as in they're not walking on a surface. If I were going to try to rig it though, I'd probably do it in a hack-y way, like you just tweak the pathfinding a little so that the nodes at the top and bottom of a ladder connect, and when the AI wants to traverse it, it approaches the bottom, "triggers" the ladder like a switch, and it transports him already to the top, then throws in some stock climbing animation in between. Maybe not too different to the way elevators work. Actually I wouldn't trust the normal pathfinding to do it (not least b/c of edge cases like AH said), but it'd be even better if the mapper designated the bottom and top with some kind of marker object, and let the code take it from there. Anyway, I think there are some ways it could be rigged, but it'd still be a lot of work & would need a lot of testing and tweaking. But it really comes down to a very motivated & competent coder actually making the thing.