April 22nd, 2015
Blog – The Nerd Hat Edition
This blog post is what I'd call The Nerd Hat Edition™, where I put on my geek crown and discuss nerdy topics. It's been a while since the last technical update from behind the scenes of all things TAGAP, so I thought this would be a good moment to have another.
In other words; If you're not either into creating Mods in TAGAP titles or into game programming, you can easily skip to "What am I working on now (the regular edition)?"
What am I doing right now – The Nerd Hat Edition
Still here? Good!
Right now I've been doing two things – crafting cut-scenes and producing a brand new level. At first I didn't mean to produce more than one cut-scene at this very point in time, but since I managed to develop a new, faster work-flow for it, I sort-of got into the groove. With TAGAPs 1 and 2 I used to be weary of the cinematic creation, but with TAGAP 3 I find it a breeze – and that's greatly thanks to the entirely re-programmed system these animations are based on.
Production of the cinematics is no longer the trial-and-error scripting process that it used to be. Since the completely reprogrammed system allows for dozens upon dozens of camera cuts, sweeps, zooms and focuses per single script, you no longer need to cut the cinematic into multiple script files. This means I can basically script the happenings as I would in a case of an in-game cut-scene – in the editor – then just time the cuts in the sequence script.
In other words, I just jump between the editor and in-game view until the sequence is as I intended. Each cut at this state is basically a set of 'delays', executed from a single 'always' trigger field. When I'm building the script, I simply add in-game buttons linked to those trigger fields so I can test the sequences out. When I'm done, I simply remove the in-game testing buttons, copy the level data over to the sequence script and then set the timers to fire the trigger fields in unison with the camera cuts.
In addition to speedier rigging with even the most complex of cut-scenes, there's another benefit for using single script approach. To understand the benefit, you'll need to know that the audio of the cinematics – as well as the music – are audio streams created using the script time line. So, when everything goes as planned and your computer isn't a potato, audio and video are in sync. However...
In the past TAGAPs, most cuts were made by loading a new script – essentially a new map – with new entities and the like. For short sequences, with one or two cuts, it was totally okay, but with longer ones (30+ seconds), each cut contributed to a time shift lag. Each script loaded needs to be processed – collision vectors calculated, floor meshes created and so forth – and no matter how fast your system is, every time this happens in the middle of a cut-scene, the 'time-line time' and the 'content time' drift ever-so-slightly more apart.
And this in turn made creation of the said audio streams difficult, as with the longer cut-scenes you had to make multiple audio streams as well, loaded in tandem with the cuts. And Petja was in trouble, too, as he wasn't able to create a score with punctuations and stingers; The cut-scene music could be up to 500 MS out of sync at the worst of days, meaning those stingers could be up to half-a-second too early!
Now? All that is gone. Heck, if the animation has an audio stream and the cinematic is only one script long, the entire content timeline is slaved to the stream, meaning everything happens in sync with the audio even if for some reason your PC CPU develops hiccups out of nowhere half-way through.
Oh, in case you were wondering why we didn't make the system like this in the first place with the original TAGAP, the answer is simple; It was never supposed to be 'a cinematic system', but a way of displaying simple title cards in the retro-way. As things evolved, more and more stuff got piled into the 'sequences', until it finally WAS a way to create cinematics, albeit in a very messy and unmanageable way. It had great potential, but it had to be re-written from the ground-up, which is what I did with TAGAP 3.
Finally, I have to again mention the skeletal animation system we have now; No longer will you need to create a custom entity for every special animation in these cinematics – most of them you can pull off with dynamically loaded pose scripts. You can create these poses in a text editor, ALT-TAB back to TAGAP 3, hit level reload button and there you have it, a skeletal animation frame, in-game. It's far from ideal for those used to them cool animation programs of the professional world, but for our needs it's pretty perfect.
What am I doing right now – Regular Edition
Fear not, now I've taken the nerd hat – which suspiciously looks like a fez – off. So, also in the works right now is a new level. No spoilers again, but as I've mentioned before, TAGAP 3 is split into episodes consisting of three-to-four big levels. Well, the level I'm working on now is the first one from the final episode. Before you get all excited; No this DOESN'T mean we're almost ready – I'm doing these levels out of sequence and there are a slew of earlier levels I still need to do.
Anyway, this new level is, in a word, enormous. It's likely the biggest darned level in any TAGAP game to date. While the level is relatively linear in the TAGAP 2's set-piece style, it's technologically complex as hell. Do you remember the train level from TAGAP 2? It was split into bazillion little segments, not because of pacing, but because of the technical challenges the constantly shifting environments (in and out of the moving train) brought on. From technological perspective this level is similar – you weave in and out of three different environments and you even have windows from which you can see from one environment to the other!
As if that wasn't enough, the level is chock full of custom assets, textures and – this being on the final stretch of the game – loads of big enemies. This means this level will become the ultimate TAGAP 3 stress test; How much stuff we can throw in one level before things get unmanageable? It might be that I need to split the level in two in order to limit its memory use, time will tell.
I've been working on this level for almost a month and there's still plenty of work to be done. On a plus side, once this beast-of-a-map is finished, we'll have a huge chunk of TAGAP 3 in place :D
In case you are new to our blog features, Backlog adventures is a spot where I highlight some of the games I've come across in my lengthier-than-lengthy backlog of previous generation titles. I've always maintained that in order to make games, you need to play them, preferably with a broad scope when it comes to genres, so each day I dedicate at least an hour to my backlog. The rest of the free time? There is no such thing, it all belongs to Pablo, with the exceptions of a couple of hours on Sundays dedicated to The Doctor. No life like no-life!
It's been a couple of months from the last backlog update and I've blown past five games so far. Apart from one utterly anarchistic title, they fall into the category of treasure hunting adventure.
First and foremost, I played through the second Tomb Raider series for the first time. If you recall, back in when I praised the new reboot Tomb Raider in our GOTY posts, I mentioned that I couldn't believe how much Lara had grown as a proper character, from the cardboard pin-up to a believable person. Well, if that seemed harsh, it's because I had totally skipped the second Tomb Raider series after being burnt to the bone by the infamously horrendous Tomb Raider: The Angel of Darkness. In case you didn't know, the first series consists of Tomb Raiders 1-4, followed by PS2's The Angel of Darkness which was so disastrous it almost killed the franchise.
The second series, consisting of Tomb Raider: Legends, Tomb Raider: Anniversary and Tomb Raider: Underworld, is pretty much the intermediary Tomb Raider experience. It's far more polished and modern than the original five, but is still light-hearted and more game-like than the super-cinematic and character-driven reboots. And discovered that I really liked them.
The weakest link is Legends, which tries to be a cinematic action game, only with the clunky mechanics of the past. It's still a fun ride, as those trademark Tomb Raider climbing and platforming bits more-or-less redeem the awkward shooting. Anniversary is the second part in the trilogy and is essentially the original Tomb Raider remade in the engine of Legends, complete with new puzzles and more fleshed out story. The level design in anniversary is absolutely sublime, with the focus squarely on what works; Lara's acrobatic platforming puzzles. I'd go as far as to say Anniversary alone makes the bargain bin treasure hunting well worth it.
Finally there's Underworld. It is basically the end of the whole classic Tomb Raider saga, the original five included. While it does hint at the other games, if you've played Legends and Anniversary you'll understand most of the plot twists. And it's a great game, too, having better action gameplay than the previous games while retaining most of the clever acrobatic puzzles of Anniversary. The only chink in Lara's armour are the annoying-yet-thankfully-sparse bug-outs that leave a bad taste. Still, all in all, it's a fitting farewell to the classic Lara Croft.
In addition to these Tomb Raiders, I also blasted through similarly-themed Deadfall Adventures and the absolutely insane Deadpool. While Deadfall Adventures was one of those rough-around-the-edges hidden gems that, if given a little more polish, could be something really special, I have to admit Deadpool steals the show. It is basically Meta The Game, as the whole story centres around Deadpool blackmailing the developers, High Moon Studios, to develop a game about him. Of course, this being Deadpool, he doesn't bother to read the script and instead does what the voices in his head tell him to do, blowing the game's budget in the process. It's been a while since I played a game so legitimately funny and clever in satirising something, let alone game development itself. As a game it's an okay action title, but in writing it wins so much it hurts – and it absolutely nails the character of Deadpool.
After playing Deadpool I'm even more gutted knowing these insanely talented folks of High Moon Studios have been either fired or relegated to Activision's Call of Duty port duties. Sigh.
Until next time,
Jouni Lahtinen, the head penguin