Serpentine

Sounds like it... will have to take a closer look, might be a missing texture or something causing the readable handling to get strange.

The first time I recall an elaborate setup with some moveable objects, trying to cast shadows which would get me to the yard. Half way into scavenging the items needed to make it all work, I found the alternate path. The alternative path is what makes that mission so great, it just exudes a feeling of quality workmanship imho.

Semi-troll: Interesting how many of these issues seem to happen on nVidia hardware, almost like their OpenGL drivers arnt quite as good as people seem to think! (That said, I have _no_ idea what's going on with this one, seems very strange)

Some of the water effects are a bit heavier than others, this hasnt changed as far as I can tell. Thanks very much for the report tho!

That's my plan Just remember to shout if you need help with this, I'm pretty used to the training mission map

Much agreed We've kept the MP code in the engine, but even before gpl, there are some pretty major issues with it, some bitrot and dependency issues. Things like the lightgem would also become... tricky.

What I mean is expectations, it sure as hell would give rise to things like that. Not to mention 'BUT MY STEAMBOX CONTROLLER!'. Edit: And today the guys over fron Xonotic covered this topic. The steamworks bit is a pretty big point.

noTarget, it'll make the game into just what you're looking for!

I have read 0 posts in this thread... ok like half of the last one above this. I just think there are some major quality bits to be fixed, and portability issues to be fixed before something like this is even through of. It would be great for abusing their bandwidth tho. But how long until 'we need achievements', 'why cant I save onto steams savegames storage?' etc etc.

I think the feature T4 will be sorely missing, which Shadow Warrior has lead the way with, is fully destructible bicycles. When in 2004 we saw what that spindly red bicycle could do for Half-Life 2, we wanted more. Years passed and still they were unwilling to yield to the mangling of a games physics engine. That's all changed now, and there's no putting this genie back in the bottle. What T4 will do to pick up this gauntlet will ultimately define the success or failure of its release.

The engine and the mod work in lock step, this topic needs to be handled internally first so that we can get a better idea of the path forward in that regard. I have some plans on what I want to do to the engine asap, but they are pretty major and require people be happy with them. If they even go ahead at all. But yeah, first I think we'll talk it over as a team. Also the engine doesn't really have a name, and its almost certainly not 'Dark Engine', since well, that's the Thief engine name.

I would pay good money to see a meeting at which Squeenix ask for a public demo to be released a month before release. I can just imagine the look on the developer sides faces. Terror. But hey, PREORDER AND GET A FREE MISSION!111

I have noticed that the information pop-ups in the stealth area seem to only show up once, and have a very short expiry. I remember extending most of them and making them repeat. I have had another new player comment that it was tricky to read them as a result. The builder at his desk, right as you enter the stealth/sneak training area. I didn't quite understand the point of him being there. Admittedly I did not look around too much to figure out if there was a point to that. It just seemed strange, the old area had the carpet which made you aware of your footsteps, this just seemed to have a metal grate which he didnt care too much about when I made a noise walking on it. I think I missed something there!

I thought we could do with a little thread about this most classic of maps. Bikerdude and others did a great job of improving and expanding the map for the TDM 2.0 release. Issues will be added to this OP as the thread continues. Basic questions (feel free to point out other things!): Was there anything we missed? Which areas did you find confusing? Which areas could do with some added love? Did you find the written portions to be easily understandable? Get stuck in a specific spot? Trapped in a barrel? Spot anything buggy going on? Known issues : Melee training AI get stuck after taking a swing/being hit. (Likely script issue) Performance in the melee training area is low. (Needs to drop some AI, replace others with dumb versions) Light going through walls near flashbomb/body drop area (Likely leaks via statics) A slight clipping problem below the rope arrow quiver passage (image) Table in archery section is floating slightly Spade in archery barn falls through the floor when nudged (Likely a bit under the patches) Frob-block the chalice atop of the pillar. Typo in tiled floor popup : "Go through the partially door behind you," Candidates for improvement: Sections for: listening throught doors, dealing with the undead. Better explain the terms 'frob' and 'use' (Need to verify, but this is fairly important) Try to avoid people picking up the book in the item-use area without knowing how to deal with items in inventory and the use thereof. Lock-picking instructions might be improved with some rewording (Check if this is the case) General comments are fine so long as they are somewhat relevant. 'I love the changes' is not really (we all do ) Thanks!

I was replying to a previous post asking for more flexibility in what difficulty could modify, seemingly on a per-AI basis. Edit : Edited the post to make it show that it was replying to someone, reduce confusion

Thats the thing, a lot of this is not criticism of the game for existing -- but rather how badly it has been managed. This is RAGE all over again, a project which can never satisfy expectations that were made by the creators. A different name and this whole thing would never have happened. It is the inability of the PR and marketing guys to understand expectations management. Perhaps the financial guys for not understanding the game itself, and likely pressuring everyone. The lack of acceptance of how important it is to take into consideration the PC market, because of its ability to make or break public perception, since news sites, social media and blogs etc are generally PC heavy. What people are enjoying with this saga is that they simply are not learning. Each time they talk to interviewers, they say conflicting things. Each trailer that has been prepared months in advance, shows the same issues that people have been worried about. It all just goes to reinforce perceptions that its got problems. And their reaction now? to do more interviews and more canned trailers. If they addressed the issues in a formal trailer that actually showed people gameplay that disproves common perceptions, and did not just keep on talking to the media on loose ground; I'm sure they would be looking way better. We can promise the world and miss, because we're not being paid to keep to our word, or even to please anyone. At the end of the day, everyone makes mistakes, but to keep making the same ones just results in a spectacle of stupidity.

Might as well turn noTarget on and just walk around a map if you're not interested in playing something.

For some completely unknown reason people thought they were using CD (cough ttlg), but it was evident that they were using UE3 from the start. The use was mostly because they were going to move part of the DX:HR team onto T4 once it was done, however quite early on in DX:HR they figured they would outsource a lot of asset work and some level design etc. As far as I know this pushed back the T4 timeline since they would not get people with experience directly. But it was back during the ol 08 shitfest and people played it safe and just used less staff.

Their UEd evidently (little birds) had a <5 minute autosave timer, due to it crashing so frequently that staff were asking to be moved to other projects/offices. When your editor is that unstable for an extended period of time. I have a feeling they will be pulling in vendor support in the 11th hour and scraping past with stability at the cost of other things.

Shouldnt be a problem, its only a problem if it shows 0 or NULL... or just freaks out when getting it

Don't wanna give it away for mapper contests, I want new assets from people who might not be involved in TDM. So I will make a little contest to make stuff you'd find in a cave. I'll post about it in the next few days. I expect it to fail horribly.

(This post isn't meant to offend, but is more my take of the balance of options/variables and a 'good solution') It's not even down to the code. These changes are nearly invisible to the player. How many HP does the AI have? Who cares - it's not dead yet. How far can he hear? Why does that guard hear better than that other guard? Why do the guards on this map see me all the time, but on other maps I was fine? Why does this specific AI not die, and I've killed plenty in the past with less. This isn't a shooter with simple "I've been seen now I shoot him until he's dead while he runs in a straight line towards me". But we can use it as an example of sorts. In general : The player has certain expectations and changing them constantly, specially when it is invisible (consider people who are challenged to notice different types of helmets!). A lot of what makes a game enjoyable comes from both the challenge, but also understanding what level that challenge was from your previous experience, and then beating that. When in the shooter you manage to kill what you know is a significantly more tricky opponent, at the same time as normal gameplay, it feels way better than just mowing down a horde of the same peons. If the tricky AI was replaced with a normal peon, you lose that. If all of the peons are replaced with the tricky AI, you lose normal players. The balance in this case is explicit due to the visible difference and beating expectations around that. If you think about Painkiller - The fights were well staged, they mixed up the AI types and always gave you lots of weak AI to pad it out and make it hectic, they didn't just give you the tough guys, and if they gave you a bunch of peons only, it was sort of the 'bonus round' or intro to something bigger. It's also why I think the real take-away portion of a book like Donald A. Normal - The Design of Everyday Things is something that crosses a lot of everyday topics, and shows that problems are often designer created. In my view : by trying to appeal to the 'everything is possible', but without balancing it out with what are realistic use-cases (open ended on purpose). Then again ~game devs~ like to pick apart books like that for the tiny semantics. In all : Difficulty changes require visibility to be appreciated. The mapper controls the environment, the most visible of all, use of which is better than twiddling spawn args.

Oh hell no, two map formats in one engine is just asking for trouble. Not to mention that his changes are only of much help to him since he's using the build in editors etc. His usecase is simple, ours is really a lot more complex. Fixing the actual bugs in the geometry generation would be a faaaar better solution. I have a nasty suspicion a lot of this actually lies in the vertex lerping stage(s), breaking it into two big areas - should the geometry be generated like that? should it be adjusted at runtime? if so what are the problems in both cases. Specially with faces that are removed from narrow cones. That said, I did some tests dmap ages ago, uncommenting some of the ID code in there which looked (this was not exactly scientific) more correct but had comments about problems. In two cases it fixed void holes being rendered, and closed up a few leaks without any new visible/performance problems. However on complex maps dmap ran even slower - and in the case of two of the campaign maps, failed to ever finish. So I think there is some case for looking at the geom generation first.

So many variables will/do come back to bite us all in the ass. AI is way too complex as is - code wise (not badly written, just... wow its pretty insane in there)

Good to hear it turned out ok. I would be interested in a console dump of it under Bootcamp. So if you have a moment, please (I am growing a collection to help spot problems, I really should have a little service online that TDM could post them to directly with some command) Edit : now with added spelling