lynx Posted August 24, 2015 Share Posted August 24, 2015 Since the last release a month ago, a lot of work went into pst compatibility. As far as I know, it can now be finished with gemrb! So we're looking for testers to find any last blocking issues or rarer scripting/dialog bugs. Chiv made a nice wiki page with the details, but since many things changed, a new playthrough is needed: http://www.gemrb.org/wiki/doku.php?id=pst_quests Caveats: - think of it as story-mode, since some gameplay elements are still wonky or missing - http://www.gemrb.org/wiki/doku.php?id=pst_bugs - that link and our general todo cover a lot of stuff, so no need to report duplicates - I'm not interested in minutiae, but if you're the type to notice them and can verify the difference in the original, go for it — it will come handy some other time It'd be best if you could come to our irc channel when having problems, but noting stuff here is also fine. As always, recent builds are in the Buildbot Binaries folder: https://sourceforge.net/projects/gemrb/files/ Testing of other games is also welcome, of course, as always, as some changes were not game-specific. Thanks and enjoy the g3 revival! Quote Link to comment
lynx Posted August 24, 2015 Author Share Posted August 24, 2015 Alternatively, the quests page link also has some pure researching / digging around tasks that don't require gemrb or playing the game at all. Stuff like Fell's store ( ): http://www.gemrb.org/wiki/doku.php?id=pst_quests#vice_quarter Quote Link to comment
CrevsDaak Posted August 25, 2015 Share Posted August 25, 2015 Stuff like Fell's store ( ):Doesn't that work by setting the dword on the Item for sale's 0x1c offset to the strref containing the trigger (yes, string-based triggers. I'm not even joking)? http://gemrb.org/iesdp/file_formats/ie_formats/sto_v1.1.htm#storv1_1_Sale I would love to go through Ps:T again but I don't have the time Caveats: - think of it as story-mode, since some gameplay elements are still wonky or missing Gameplay in the original was bad anyway =P Quote Link to comment
lynx Posted August 25, 2015 Author Share Posted August 25, 2015 (edited) Nice, you're right. Even DLTCEP shows them. edit: oh, we even support this already, there's just a glitch somewhere, so Razor angel ink tattoo shows up too early. edi2: eh, that item has an extra space in the trigger string, so it was only that ... Edited August 25, 2015 by lynx Quote Link to comment
Ishad Nha Posted August 25, 2015 Share Posted August 25, 2015 You really need someone who played the original PS:T, which I have not. Quote Link to comment
Chilvence Posted September 25, 2015 Share Posted September 25, 2015 Since this post, I did a play through of the game in GemRB start to finish (for the sake of honesty though, I did kill the vast majority of things with cheat keys to save myself a few real world days of boring combat). Using the development branch, it is very near possible to complete the game, only at the very end do things start to break down. Trying to avoid spoilers, but I got as far as just before the point before you return to Sigil to get to the very last stage. The problem was a certain character was not killable. But the other 89% of the game was largely doable. Once the remaining unknowns/bugs with scripting are resolved, you will be able to start and finish a game of Torment in GemRB. There is still a lot of work to do, but that is a huge milestone, and I think you all deserve mega congratulations for getting it to work this well. In summary, here is a list of the greatest annoyances left: *Spells are really going to need a lot of work, because they took a completely different direction to the bg/iwd series *A lot of character/cutscene sounds are not there and music cues don't work as intended *Some creatures have garbled palettes and lots of animation sequences are not implemented *A certain Maze needs to be implemented properly... So that is still a lot of work, but the end is edging closer and closer... I think the fact that I can digest it into 4 bullet points says something in itself. I will try my hardest to diagnose or resolve any of the remaining problems myself to help you guys out, and I will keep chipping away at the little annoyances until they can't annoy anyone else anymore. Quote Link to comment
lynx Posted September 25, 2015 Author Share Posted September 25, 2015 Thanks chiv! Baring any serious wtfery, the remaining blockers should be gone soon. I'll need a save for Trias and the incarnation stuff though. Quote Link to comment
Chilvence Posted September 26, 2015 Share Posted September 26, 2015 https://drive.google.com/file/d/0BzBo2zkBt4guZ3VKSjRfeTVXQjQ/view?usp=sharing These are my last two saves. However, since I posted this, I can't seem to reproduce the problem I had before in 'Deferred Destiny', so it could be a waste of time looking at it. In 'Stalling Spectre' though, I am left in a non responsive cutscene mode every time. I should be honest and say that anything after that final save I have hardly tested, I only concluded the game with one ending by 'jump starting' with one of my saves from Torment.exe I don't see the final level of the game being that hard to test mind, just that I haven't in earnest tested it with GemRB because I couldn't reach it with GemRB, because of this one cutscene issue. Quote Link to comment
Ishad Nha Posted September 26, 2015 Share Posted September 26, 2015 You could check the cutscene with Near Infinity or Infinity Explorer. Locating the cutscene can be done by looking at the commentary GemRB generates in the Command Prompt or by a text-search in the BCS files listed in Near Infinity. Quote Link to comment
Chilvence Posted September 26, 2015 Share Posted September 26, 2015 I already have, but I don't really know the code well at all. Lynx has a knack for spotting the problem straight away, after I've been scratching my head for hours Quote Link to comment
Ishad Nha Posted September 27, 2015 Share Posted September 27, 2015 (edited) The location function shows your party at [Python]: ar0901 [453.204] In Near Infinity look for AR0901 in the list of ARE files. Your cutscene can be ended by a command in the Gem console: EndCutSceneMode() Access the Console by pressing Control + Space If the cutscene is needed to move the party elsewhere, this approach won't work. If not, you are rid of the frozen cutscene. A cutscene is a BCS script. Can't find the help docs for the Gem console?! Edited September 27, 2015 by Ishad Nha Quote Link to comment
lynx Posted September 27, 2015 Author Share Posted September 27, 2015 he knows that Ishad ... The docs are at the bottom (the index one): http://www.gemrb.org/wiki/doku.php?id=documentation Quote Link to comment
Ishad Nha Posted September 27, 2015 Share Posted September 27, 2015 (edited) Now I have PS:T up and running. I will try that cutscene. In 'Stalling Spectre', EndCutSceneMode() does work to end the freeze. When you close the Console, movement is normal. So you can test one dialog option after another. However, the situation is unchanged...Console reports: [GameControl]: Position: ar1203 [718.828]Look at the dialog and see what might be wrong, DDeions.dlg? The actions of the cutscenes can also be effected by the Console.Console gave some commentary:[DLGImporter/WARNING]: Can't compile trigger: incrementglobalonce("lawful_deionarra_3","global","law","global",1)[unknown]: [streams]: Invalid seek position 5892193 in file DLGFiles.bif(limit: 3274108)[unknown]: [streams]: Invalid seek position: 5467759(limit: 20068)[GameScript/ERROR]: Invalid scripting trigger: incrementglobalonce("lawful_deionarra_5","global","law","global",2)[DLGImporter/WARNING]: Can't compile trigger: incrementglobalonce("lawful_deionarra_5","global","law","global",2)[GameControl/ERROR]: Invalid window or position: PortraitPosition:4[ResourceManager]: Found '1203cd11.bcs' in 'chitin.key'.[GameScript/DEBUG]: CutSceneID was called by nameless![GameControl/ERROR]: Invalid window or position: PortraitPosition:4[GameControl/ERROR]: Invalid window or position: PortraitPosition:4[GameControl/ERROR]: Invalid window or position: PortraitPosition:4[GameControl]: KeyRelease:32 - 2[KeyMap]: Looking up key: ( ) Edited September 27, 2015 by Ishad Nha Quote Link to comment
Chilvence Posted September 28, 2015 Share Posted September 28, 2015 Hah, looks like yet more bugs in the really dodgy Torment data set... That one 'IncrementGlobalOnce' action has caused all sorts of headaches, I thought I finally killed it last week and yet here it is again Quote Link to comment
lynx Posted September 28, 2015 Author Share Posted September 28, 2015 it goes wrong before that - it didn't even run its function. Just need to improve the parser — perhaps another case of extraneus spaces (not visible in the output above, but in the dialog itself). We should never fail a compile without an excellent reason. I'm stuck hunting down a memory issue, otherwise I'd look at this already. Quote Link to comment
Recommended Posts
Join the conversation
You are posting as a guest. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.