11-12-14, 14:23 | #1051 |
Member
Joined: Jun 2006
Posts: 2,392
|
That's great, thanks.
Health crystals are rendered now, didn't notice this before. Nice |
15-12-14, 17:45 | #1052 |
Member
Joined: Apr 2013
Posts: 343
|
The last snapshot won't generate in source forge
OK, problem has been solved Last edited by Ado Croft; 15-12-14 at 19:40. |
22-12-14, 00:33 | #1053 |
Member
Joined: May 2010
Posts: 256
|
Anybody happens to know how to compile it under Windows? I've tried 4 times with 2 different IDE/compilers, 2 different forks and also tried this guide, but I just can't get it to work, everytime's a different error.
The only one I've managed to compile was an OpenRaider fork, but of course it got me so many runtime errors afterwards I had to abandon it. EDIT: If possible, I'd really, really appreciate someone having a bit of time to help me on this, it might take a bit to figure out all the friggin' problems I've encountered. If anyone's open to waste some time on a poor noob like me out of Christmas spirit, I welcome them! Feel free to PM as well if you're feeling particularly generous T_T And also, Happy Holidays! Last edited by Nickotte; 22-12-14 at 03:06. |
22-12-14, 18:11 | #1054 | |
Member
Joined: Jan 2013
Posts: 195
|
build engine
Quote:
You may use latest binary release with included code::blocks project. I use relative patches in projects, so number of problems is less (no need to fix global patches in project properties). But You need compiled and installed SDL2 library (+ header for SDL image library) and zlib + GCC TDM v4.7.2 (do not use newest version: it has bugs that do not allow to compile bullet library). To install that libraries You need msys. I will try to create screenshots with information about files structure of GCC + msys + libraries. |
|
22-12-14, 20:54 | #1055 |
Member
Joined: May 2010
Posts: 256
|
Wait... the last binary release includes a C::B PROJECT??!?
HOW DID I NOT SEE THIS. Anyway, thank you so much! I'll try it ASAP, and if you're really gonna post some screenshots and info I'll marry you! I'll report back for eventual major problems. EDIT: Trying to run it right out of the box it pretty much compiles fine, up to the same problem I've encountered with the snapshot I used following that old guide, which is the pthread library not found by C::B ("cannot find -lpthread"). If this can be solved with this "msys" you're talking about - or you're going to address it with the possibly upcoming screenshots - then I'll wait for more info on the question. In the meantime, I'll still try to look for a way to fix it myself, so that I can see if it isn't anything big of a deal. EDIT2: Using pthread-w32's x86 library (renamed from ibpthreadGC2.a to libpthread.a) works for now. And by the way, want me to report concerns or bugs/issues here on the forum thread or would you prefer PM/email/Sourceforge? I'm really interested in this project. Last edited by Nickotte; 23-12-14 at 02:50. |
23-12-14, 16:29 | #1056 | |
Member
Joined: Jan 2013
Posts: 195
|
c++ build
Quote:
about msys it is needed for libraries building with autoconfig (I hate autoconfig) or netbeans IDE using. I use netbeans as a main IDE (it highlighting mercurial changes), code::blocks for quick rebuilding (+it highlight LUA scripts!) on old PC and eclipse (faster than netbeans but did not highlight mercurial changes and there are some problems with compiler settings convenience). About engine building: we can discuss engine coding here, but talking about C++ specifies, GCC, IDE's, codes, scripting, a.t.c. in general may be better in gmail (C / C++ programming is GOOD ). P.S. I become to add callbacks generation to the entities + realize pushable doors (TR4, map library.tr4)... see soon in SF. |
|
24-12-14, 08:06 | #1057 |
Member
Joined: Apr 2013
Posts: 343
|
OMG, wau I gotta feeling that something great is going to happen, maybe interaction programming of traps? Great job again and Merry Christmas
Last edited by Ado Croft; 24-12-14 at 08:11. |
28-12-14, 00:28 | #1058 |
Member
Joined: Aug 2010
Posts: 1,810
|
Hi, merry Christmas to all, I was away for some time again - mostly because I had to finish some business before new year, plus we have some financial problems in our country right now, maybe you already heard that...
You should have already noticed by now that TeslaRus fixed alternate room portal swapping, so right now we have fully functional and bug-free alternate room collisions, which provoked me to finally implement ALTERNATE_ON and ALTERNATE_OFF triggers functionality. One thing I wanted to ask is how exactly ALTERNATE_FLAG works in TR1-3? Dxtre3d manual description is a bit hard to understand. By the way, I also realized that somehow, both Popov's TRosettaStone version and any other file format documentation completely lacks info on alternate group parameter! Never mind, already found that in bypassed 8-bit field just after reverb_info! And, this uncovers final unknown in original VT loader for TR4. |
28-12-14, 01:02 | #1059 |
Member
Joined: Jan 2004
Posts: 111
|
Check with fexinspect file level3b.phd, room #19, there is an example how to use those commands.
Alternate_flag is like a PUSH then POP flipmap state, it is like a workaround to a situation where finally in tr4 engine is resolved using what i called "alternate groups" |
28-12-14, 02:39 | #1060 |
Member
Joined: Aug 2010
Posts: 1,810
|
Ok, I checked it, but whole concept of this setup in Tomb of Qualopec doesn't make sense to me.
First off, Lara already starts on a sector with ALTERNATE_ON(0) command. Effectively it means, that all rooms are flipped on a start-up, so if somehow Lara could teleport right to the main tomb of Qualopec hall from that point, we would see "damaged" hall and be able to trigger collapsing ceiling etc., right? Also, there are bunch of alternate rooms on the way back (Lost Valley leftovers), but they are blocked by the stone door and there is no way to reach them without opening that door - but if somehow Lara could bypass that door, we would see deactivated waterfall (state in which Lara left previous level before). Ok, but all that doesn't happen, and Lara have to go through sectors with ALTERNATE_FLAG(0), ALTERNATE_OFF(0) and ALTERNATE_FLAG(1) triggers. So effectively, flag is set to 0, then all rooms get flipped off, and flag is set to 1. In game, it means that Lara reaches "normal" hall without collapsing ceiling. Next, things got weird. There is an ALTERNATE_ON(1) trigger right where Lara picks up Scion - so earthquake happens, hall transforms to damaged, there is falling ceiling, etc. But at the same time, door opens in the beginning of the level, unlocking a way back to pool and Larson. But THAT DOESN'T MAKE SENSE, as this door was always closed before, so Lara couldn't reach that part of level before! Only idea that I can come up with is that this weird set-up was implemented because Core wanted to make Lost Valley and Tomb of Qualopec as a single level, and later they split them up because of the memory issues, but the setup was left untouched. All in all, seems I got the idea of this weird command set-up. So, the purpose of ALTERNATE_FLAG command is to restore ALL alternate rooms state regarding the value set by corresponding ALTERNATE_ON/OFF commands? I mean, if I activate alternate rooms with ALTERNATE_ON(1), then deactivate it with ALTERNATE_OFF(0) command, then step onto ALTERNATE_FLAG(1) trigger, rooms will be back to alternate? But if I'll step onto ALTERNATE_FLAG(0) trigger afterwards, they will be back to normal again? If so, how big is that ALTERNATE flag array? I can assume they used 32-bit field for that (at most), so the capacity is 32 different alternate room states. And maybe you remember some other examples of this setup across TR1-3 to make things more clear... Last edited by Lwmte; 28-12-14 at 02:50. |
Thread Tools | |
|
|