The website hasn’t had an update for a couple of weeks, so I decided that something had to go up.  The main difference to the website right now is that all of my information got updated to reflect my official graduation from the Master’s Degree.  With distinction as well!  Our class is made up of the guys with the green hoods in the picture above.  Not all of us were able to make it back for the ceremony, but it was good to see everyone again as we all head out to find our positions in the gaming industry!

Graduation was a busy time, so that partially explains the inactivity on the site, but much of it is gaming related.

I’ve recently bought even more games (some deals are just too good to pass up) and Batman: AA took up all of my time for a couple days while I finished it.  Truly a great game.  Since I am tearing my way through several games right now I’ve been thinking about doing another edition of Random Game Thoughts, so this could prove to be quite productive.  My recent gaming spree was amplified by my Xbox getting a red ring of death, depriving me of gaming for a little while.  I have to say the repair went extremely well though and I had it back within five business days.  It was still torture though since I acquired seven games while it was gone.

I also have a large drawing I’m trying to get done, which has displaced my Torchlight level for a little while.  Which is only fair, since Torchlight editing had displaced the drawing for quite a while longer.  Speaking of the level, it is still coming along.  It is actually more complete than my forum topic currently suggests, I just can’t justify an update there quite yet.  It should be a decent finale though, I’ve got at least one secret planned a final battle quite a bit different than the rest of the dungeon.

So here is a small screen of the cliffs and hopefully I will have a game-related update in the near future.

It does exist! There is an end!

The time to reflect on part three is upon us.  I had predicted that this section would get completed much faster than the previous one, the type of prediction that usually curses a project.  In most ways it did go faster than the previous level, even though the overall time I put in was probably more.  This level is both larger and more complex, so my previous experience helped keep development from getting out of hand (I will talk a little about that later).

I have to admit that it wasn’t as smooth as I had originally hoped though.  Most of this is due to SunkenTemple tileset, which comprises most of the level.  I thought the transition to this tileset was going to be a straightforward as was to move from the Crypt set to the Cave one.  It didn’t really work out that way.

The original plan for the Jungle and Village, with the ending incomplete.

The first problem is how the floor tiles fit together.  The trim pieces for the Cave set are really simple to work with and I had gotten used to planning a level with those pieces in mind.  The SunkenTemple ones do not fit the same way, so the first parts of the plan I made could not be perfectly recreated.  The second problem is that the floor pieces are poorly organized, so not all of them show up when “floor” is checked in the editor palette.  I also recall some pieces not having a collision mesh assigned to them when I started.  These are the type of little things that add up to a larger pain and it slowed me down near the beginning.

Overview of the final level, much different from the last one.

There are other irritating things about how the walls fit together (which are organized in full, half, and quarter sizes) that make transition not as smooth as I would have wanted.  This complaint may have been amplified by some of the things I did with the walls, so I could have made it a bit easier on myself.  My last complaint (getting too ranty here) is that the diagonal walls don’t display correctly on the map, instead it shows the opposite diagonal which makes the map look terrible.  I’m still looking for a fix, but I don’t know if I’ll find one before I finish the dungeon.

A closer look at the Jungle Shrine area.

I mentioned before about keeping development from getting out of hand, this mostly concerns the organization of the level.  Well labelled groups are a must when making a layout of this size.  I have dozens of them breaking the level down into areas (8 of them) and type of piece.  Since the editor does not have a way of locking a group like one would in photoshop, it pays to have everything properly organized so that visibility can be toggled.  It is also very helpful when cleaning up the map to have pieces split into groups that will or won’t show on the map.  I had organized the Crypt and Cave sections with groups before, but it took much more to keep from getting lost in this level.

So I am moving on to designing the last section of the dungeon.  I only hope that this one hasn’t gotten too large, the load time in the editor is getting a little long.  Still, it is worth it because the result looks really good.  I have also been updating my progress on the Runic forums little by little, so check that out for additional screenshots.  I will make a gallery on this site soon so that there is a more complete collection.  Finally, here is one of the videos I uploaded to my Torchlight Development Playlist on YouTube for this level:

I was also able to put the planning stages to better use during this section of the dungeon.  The cave section started out as an incredibly small sketch, one that I hadn’t given very much thought to when dreaming up the overall dungeon.  I had expected it to not be much longer than the crypt section, mostly just a way to get to the outside areas of the level.  It turned out to be much more.

The original sketch. There is more there than there appears to be.

So starting with the very rough sketch I brought out the graph paper and elaborated on the general shape I had drawn.  I left out the enemy, lights, and sounds details that were shown on the crypt plans.  This allowed me to concentrate on the basic layout and the key moments I already had in mind.  It ended up being split up into three sections; the lower, mid and upper cave.

The entire crypt/cave plan.

After the sections were drawn out and the basic floor and wall tiles had been placed I was left with quite a bit more room than I had originally intended.  So I went through the level and planned out each area in a little more detail in order to keep it interesting all the way through the level.  As I needed to do more planning I went back to the diagrams I had created and sketched in what I wanted to add. (Like secrets!)

The Lower Cave plan overlaid on the actual level.

Details like lights and sounds still haven’t been put on the plans and probably won’t in the future either because there are so many of them.  As the level develops these details are added as I see fit and do not provide any useful information when on the plan.  The crypt/cave section of the level is nearly complete, but needs enemies.  These will be added to the plans because that helps to properly space them and give the level a certain amount of balance.  After that I will be starting the next floor of the dungeon, which takes place outside of the cave.

To see the level in action check out my YouTube page or the player below.  The cave section is spread across three videos.

I often have random game thoughts, some more insightful than others.  A couple of times I’ve considered writing them down, but I’m finally getting around to doing it for once.  There are a couple older ideas I’ve wanted to elaborate on, but for now I think I will just stick to a couple of game thoughts I have had recently.

As the picture shows, the first of the two thoughts is about Super Paper Mario for the Wii.  It is actually more of a complaint.  The Paper Mario games are the slightly more RPG-like games in the Mario universe, though it hardly goes deeper than having more than one hit point.  One thing this game does have in common with many RPGs is a lot of writing.  Writing is fine, especially when there is quality humour to be found in the script, and I don’t mind reading in place of voice-overs.

What I can’t stand is the sound the text makes as it is written out.  It is basically a cross between a squeaky marker and nails on a chalkboard.  You can skip the text after the first couple squeaks, but there is so much text that the player ends up doing this enough to make even small amount very irritating.  It is honestly amazing that play testers were able to endure the torture.  I have a similar complaint about the “low health” sound in Twilight Princess, but that is much less annoying and not nearly as frequent.

The two fit so well, but I can't see it coming true...

The other thought I had concerns the Lego video games.  Playing the games always ignites conversations about what other franchises would work well with the Lego formula (Building + Breaking = Winning).  The one that came to mind the other day was the James Bond franchise and I’m not the only one who thinks so.  It has multiple movies, tons of characters to recreate, and plenty of opportunities for those Lego vehicle levels.  Of course the only problem is having to fudge some of those more questionable names and situations in the movies, but the Lego games have always been good at that.  I would definitely play those games (there would have to be more than one), but as well as the two franchises might fit together, something inside tells me that it probably won’t happen.

So that’s it for these random thoughts, but I’m sure there will be more.

So after a couple video tutorials and some plain old messing around I cobbled together a sketch of a level idea and dove in.  These short blog entries will briefly chronicle my progress over the next little while; I don’t expect this level to take too long to create.  I don’t plan on doing any sort of TorchED tutorial posts, these are more of a development diary for the mod.  Posted after the break are the plans and screenshots for the first section of the level.

Layout plan for the Crypt section of the level.

For each section of the level (I suppose I should be saying dungeon) I first draw up a diagram of the layout.  This one was fudged a bit, since I had already built part of the level before deciding on the rest, but it follows the same organization as the other plans.  Blue tiles are water, green areas spawn enemies, orange (kind of) boxes are triggers, red circles are lights, and yellow circles are sounds.  Some objects come with lights and sounds attached to them, but additional ones are added as needed.

The first section of the level is crypt-themed and is very short.  It is the first of four main sections: the crypt, cave, forest, and mansion.  Each of these are subdivided for organization purposes and will be detailed more when I get to those sections.

The crypt section in TorchED.

My custom level isn’t that similar to how levels are done in the actual Torchlight game.  In Torchlight dungeons are composed of premade chunks that are semi-randomly pieced together based on their connection points.  Each chunk is developed within a single theme (as far as I can tell) so that when they are pieced together the look is consistent.  Each section of my level is making use of pieces from more than one theme and is meant to be traversed in a particular order.  There are some levels in Torchlight that are created in this manner, but not all of them.

I don’t have too much more to say about this section since it is so short, but below is a video of the crypt section in a fairly complete stage of development.  I have the cave section sketched out already, so that update should come fairly quickly despite it being much larger and more complicated.  Being more accustomed to the editor helps as well!

Alright, well it’s been just over a week since the final version of my paper was submitted for grading.  With the end in sight, I suppose it is time for another update.  This update does presume some familiarity with the project, so reading the previous update and the project page is recommended.

The testing went reasonably well, many participants who played through the demonstration preferred the adaptive music to a more traditional passive approach.  The application also ran correctly in all cases; the adaptive music was consistently successful at matching game events with specific elements of the music.  The seamless transitions between sections of the music appeared to go unnoticed by the participants (as they should, since they are supposed to be seamless) and there were a wide range of positive responses to this.

Still, the testing revealed some weaknesses in the testing process.  Many revolved around the questionnaire, which had questions that may have unintentionally biased some participants and others that could have been more direct.  The possible bias was taken into consideration when writing the project’s conclusions and that participants could have been biased in either direction depending on how they interpreted the labels of the musical implementations.  Other questions, which were looking for participants to acknowledge the synchronization of game events with the music were too vague and received suitably vague responses.

These problems meant that the more useful results from the testing were drawn from the conversations with participants following the hands-on demonstration and questionnaire.  Some people found it much easier to describe verbally what they felt the difference between the two musical implementations was.  These results were more interesting.

The final pdf! Gotta love LaTeX.

For instance, although few people wrote about the connection between the colour changes (game events) and the music, most participants noticed it.  Many did not think to include it in their answer based on the wording of the questions.  These responses also reinforced the positive reaction to the adaptive music, whether or not the participants were aware of what the difference was.  It seemed that since the events were playing out during the music that was designed for them, participants felt some sort of subconscious enhancement to the game.  Some felt it was “faster”, “more fun”, and “more epic”.  So although they weren’t the expected responses, which may have been too technical in nature, these answers which concerned a more intangible feeling were valuable.

The colour changing events may have also been too similar to the player’s goal, which was to change the colour of the “goal cubes”, although this was never mentioned by a participant.  More variety, or no goal cubes, could have brought more attention to the connection between the events and music.

Anyway, the rest of the conclusions are in the paper and are too much to review here.  I have to make more changes to the project page to outline the project more clearly.  A great deal of future directions were proposed at the end of the paper, so those will be included as well.  Now, to wait for a response to my submission.