Just a thank you -
over 8 years ago – Tue, Jul 25, 2017 at 03:35:55 PM

I've said it many times as a closing bit in updates, but it's never been the focus of an update.  I felt it was time that it needed to be.  Thank you to everyone who supported and continues to support this project.  Thank you to the people who have shared their excitement with others and helped build an anticipatory community of people.  Thank you to the people who have voiced their patience time and time again.  Thank you for the critical feedback and to everyone who in some way contributed to making this project better.

The New 8-bit Heroes, and bringing Mystic Searches to life, and developing a suite of tools for others to develop NES games, has been unlike any creative project that I've been part of for my 37 years.  It is directly associated with meeting so many people I now consider close friends.  It is associated with the birth of my first born son.  it is associated with meeting my long time influences in movies and music and writing in addition to game development. 

So thank you.  To all of you.

Back to world building...

Joe

Continuing Design Challenges...Aesthetics versus Narrative versus Gameplay...
over 8 years ago – Mon, Jul 10, 2017 at 09:55:07 AM

The creation of the world for Mystic Searches has not been easy.  Rather, I should say...the *translation* of the world from limitless illustrated imagination to 8-bit, limited color, limited graphical assets realization has not been easy.  It has, again, yielded interesting late-phase design challenges.  

BUILDING THE WORLD

At first, we set out to reconstruct the world as close to the map as possible, with broad strokes, assigning general palette sets to each area.  We could then look at *the map* in a general way, color coded to make sure each area of the world had enough screens, where each geographical part of the world could access other parts of the world (where the forest became grassland, where the mountains edged to the beaches, etc).

This went smoothly, as it was still very conceptual.  We gridded this overworld map to get a loose look of what this world would look like when broken down to 16 screens wide, 16 screens tall (256 overworld screens). 

Then it got to the harder part.  And that was to make the tile art generally suggest the concepts for the areas.  A lot of time went into the aesthetics to give each are the right *look* or *feel*.  Of course, this couldn't be perfect, as the tilesets were terribly limiting.  However, for a NES world, I think we did a fairly good job at translating!  Here are some examples of first passes of some of the areas.

Here is an example of Galbeton, the mining town that straddles the forest and the mountains.  Using both the map version and this detailed illustration, I tried to suggest the forest border, the tiers of height, the wood and stone, the giant Ruhk tree.   

 This is Iohai.  I tried to capture the air of aristorcracy; more intentional architecture, the controlled water features, manicured hedges, sidewalks.  Add in the NPCs and this city does a decent job of suggesting the concepts (the waterfall empties out just to the east of this capture.

The idea of Swynhall was that it was formerly a thriving city as well, subject to decay and neglect.  Poisonous swamps, sunken idols...just ruins of a once proud civilization.  For this, the color palette changed to really denote the poison, and there has been much discussion about attempting to make the trees a bit more *willow* like.  

 This one was a fun challenge - to represent *inverted pyramid* cisterns.  These are the bastions of life in a harsh desert.  While not a perfect translation, this does convey a lot of the architectural suggestion in the concept. In the game art, the temple looks to be inside the inverted pyramid where in the concept, it is out front, but all in all, the concept actually translates nicely.

THE DESIGN PROBLEM

The problem comes down to gameplay and narrative development.  In this first pass of building the world, we've created a sort of sandbox without context.  A general world for the character to run around it and experience these different regions that lined up well with the concept art.  But as we began to actually game-ify, we realized that many of these aesthetic decisions that made the world *look* good or resemble the concept art or make logical sense as far as fantasy world goes weren't necessarily the most fun (or intuitive) to *play* through.

And so the balancing act began of the balance of importance between the aesthetics and the gameplay.  Consider a game like Super Mario Brothers...bricks and question blocks floating in the air...pipes leading places and crossing immense distances of a world over the course of one *underground* screen, cloud platforms and flagpoles and giant pulleys.  None of it makes any sense at all aesthetically.  But it becomes a language that we adapt to, which trumps the need for any level of narrative logic.  The levels are designed within these rules that use defined iconography, and the aesthetic sense carries almost no weight.  It does what it does almost flawlessly, and subverts the expectations in places to reward the player for experimenting (like trying to jump *over* the level and finding a path to a warp zone).  

So the tricky part in designing screens for Mystic Searches is, when designing each screen, how much attention is devoted in service to the immediate playthrough of the screen versus its context in the surrounding world and in terms of narrative and aesthetics? 

Another interesting creative challenge, with no right or wrong answer.  Just thought I'd share some insight on this July Monday.

Follow up on 2.5d Platforming
almost 9 years ago – Mon, Jun 26, 2017 at 10:36:40 PM

The question is...do I cut my losses, or march on?  Do I take the time to perfect this and the few funky things that will likely come as a result that will inevitably need fixing, or do I step backwards a few days and pretend like it never happened....

Creating platforming mechanics is not hard.  Creating top down mechanics is not hard.  Creating more robust platforming mechanics within the top down game engine for Mystic Searches...well...that presents a whole bunch of challenges.  Is it worth it?

What do you think?  Sound off below!

What the beta taught us; late phase design challenges...
almost 9 years ago – Fri, Jun 23, 2017 at 10:32:44 AM

Greetings everyone!  I wanted to make sure to post a good solid post-beta update.  Since it's been a few weeks, there are a lot of things to touch on, so this will get pretty long.  But to those of you who like the BTS of the game development, some of these late phase design challenges should be very interesting to you!

The Film:

First and foremost, let me just say how gratifying it is to find that the film itself has not only been official selection at many festivals across the US, it has begun to win awards, including best video game film at Origins Film Festival, Best Feature (wow!) at Southern Fried Film Festival, and was nominated for best documentary at the Toronto Independent Film Awards (double wow!). 

The world of film sales agents, legal clearance, and distributors is very slow and very obnoxious.  The good news is, the positive accolades the film has been receiving is certainly helping.  Hopefully we'll have some announcements on this part of it soon!

The Beta:

Creating and shipping Mystic Origins was a very important undertaking in the process of the development for Mystic Searches.  First, it allowed us to demonstrate a taste of what Mystic Searches will be, and just how far along we are.  Acknowledging the bugs inherent in it being a beta, the feedback has been unanimously positive and excitement for the final game has only increased.  Our devoted fan base getting to play it without much instruction and reporting on the experience gave us some real insight into the things that are intuitive, that work well, and the things that need tweaking.  But more than that, it also let us evaluate the entire pipeline of having the finished game file, mass-flashing to cart, assembling plastics, applying labels, shipping...actually getting a *finished* NES game in the hands of the people.  We were able to analyze all of the unanticipated issues with this as sort of a *practice run*, and will make the process of getting the final game-in-hands all the more efficient.

Late Phase Design Challenges:

This part is for those that love the behind the scenes discussion about NES development and about our creative choices for Mystic Searches.  It's going to get into abstract thinking a little, but I love being able to share riddles like this with you guys.

Our game faces a difficult graphical challenge.  It is a 2.5d game...that is, it is a game that has a z coordinate where by a character can jump *up*, however it is a top down game where *up* also means north, if you will.  Conveying this visually and mechanically presents extraordinary challenges (especially within the constraints of the system).  

To start the discussion, let's take a look at three games and how the represent themselves graphically for comparison.

 First, we have The Legend of Zelda.  This sort of became *the standard* that most top down games adopted.  Generally regarded as a birds-eye or top down view, it's actually a little bit of an MC Escher type abstraction of a perspective.  We don't actually look at the top of the character's head, he more or less is seen from a side angle (like in Milon's Secret Castle screenshot on the right).  However, even though this seems intuitive to our eyes now, it's actually really weird in his environment...look at the door to the right.  It definitely seems like the perspective is wrong.  That for that sort of door, where we are seeing the room itself in a top down manner, the hero should be presented closer to the Gauntlet perspective in true top down fashion.  We shouldn't even be able to really see the bottom wall considering the perspective we see of our hero - we should just see the lip of the wall, and the player should appear to *go behind* it.  It certainly shouldn't appear *upside down*, as it does in NES Zelda.

But don't get me wrong - it works for Zelda fine.  And it became the standard for *top down* that many games used.  We quickly get used to the rules of the graphics and screen-layouts, and it's perfectly intuitive.  I just wanted to point it out as it's a good starting place for analysis in describing our current design challenge.

Since our game does have that z-coordinate, where the player can *jump* as a core mechanic, the gameplay perspective itself sometimes feels more like games like Battletoads, TMNT, Double Dragon, or other beat-em-ups.

 Notice in each of these games, the player is free to move up, down, left, and right, but also can *jump*, and there is a perceived height.  We know that Michelangelo there isn't *stuck in the wall*, but rather that the shadow beneath him denotes where he is in space in terms up up/down/left/right, whereas his distance from his shadow represents his *jump height*.  We'd expect, then, in that shot to be higher than Leonardo's head.  Similarly, Zitz in the Battletoad picture isn't *stuck in the wall*.  In fact, gravity is pulling him down, and  if he doesn't move to the left, he'll fall into that pit when he reaches the ground.

These games have fairly clear 2.5d perspectives.  Like Zelda, this aesthetic became a standard for 2.5 design that we all are used to, and we understand pretty clearly how it operates.  In these games, you usually have long horizontal levels that have an almost isometric, 30 degree skew, and the world is usually seen in a 3/4 view, where the top portion of the screen is *the wall* (or edge, or whatever).  The player usually has only left and right graphics, even though he can move up and down, which helps with the linear directionality of the mechanics.  Most of you have probably played both types of games...the top down, and the 2.5d, so you likely know what I'm talking about even if you've never analyzed the art style like this.

So let's talk about how this all plays into design challenges for Mystic Searches.  For those that have played the demo (or have been keeping up with the project to any degree), you know that the game generally follows the *Legend of Zelda* top down, 3/4 character view graphic and mechanic style for the most part.  

In this comparison, you can see the perspective by which the character is presented, and you can see an *obstacle* circled in red.  In Mystic Searches, the player has the ability to jump over many obstacles of this size and shape as part of the game's mechanics, which starts to feel a little more like the beat-em-ups above.  We track the jump height by the use of a shadow, like in the TMNT example.

And like in Turtles, it's pretty clear that the player shouldn't be *stuck in the background* (in Turtles, the wall, in our game, the beam).  

It has a few unintuitive quirks, just like Zelda-doors perspective...like from the height you can jump, it seems you'd be able to jump over a tree, but you can't.  Some objects that look like shorter objects represent solid barriers.  But just like with the wonky perspective of Zelda's doors, it becomes intuitive very quickly.

For most cases, this works just fine.  However, our game is NOT laid out like the beat-em-ups.  It's laid out more like Zelda, Star Tropics, or Crystalis.  Thus, the current, late phase design challenge.

Take a look at the two screens below, and then we can start examining the problem.

 If we look at the two Mystic Searches screens, they look like they belong in the same game, even though they sort of combine these two graphical perspectives for the way they'll use mechanics.  

What do you think should happen when you play through each of these areas?  Intuitively, based on the graphics, it would seem that you can walk around the forest (left) in every direction, and possibly jump over the logs, sink in the water, etc.  What about the right?  It looks like it should behave more like a platform game, where if you don't quite make it to the next ledge, you fall to the bottom of the screen.  Or, where you'd have to *hop up* to the next ledge.

And that's where the trouble comes in.

Let's talk about how collision detection works in the game.

Like in Zelda-esque games, our collision detection checks to see if the place in front of the character is free, and if it is, he can move there (easy to see in the image to the right).  Notice, when jumping, it is checking the position *in front of where he would be standing when he falls to the ground*.  In this image, the player will be able to *jump over* the statue face once the player reaches a certain z value.  We don't want the player to be checking the TREE to know whether or not he can jump over the statue face, because obviously it will read that the tree is solid, and we won't be able to jump forward (in fact, we'd be stuck in the beam right now).  From a gameplay perspective, this makes perfect sense and feels natural (for those that have played the beta, this is how it works, and you can probably attest, it is working just fine), but it's important to articulate what is happening in order to demonstrate the design challenge. 

Here's one of the challenges:

All of the sudden, when we find ourselves in a real 2.5-d situation, our rules for the game break down a bit.  If you look at this sample screen and think about it intuitively, it seems that as you're playing, you should be able to *jump up* to the ledge to the right.  That's what feels natural.  However, the collision detection is not factoring in true depth of that cliff ('true depth' doesn't exist...if i were to be standing on that cliff, my z value would be 0, and if i am standing on the lower ground, my z value is 0...it's only faux depth, not true depth where the higher cliff is actually *higher*).  So now, what feels intuitive no longer jives with how our game handles collision detection.

But changing the way our collision detection works to incorporate z would mean...

So the main challenge becomes, without writing two completely separate screen-mechanic methods that could vary from screen to screen (let's build it all as a cohesive system), and without having to create a *depth table* (no room for that in the ROM), how do we handle these very different respective circumstances to work the way that feels intuitive?

There are many methods we can poke at, and I think I've arrived at a concept for one that I like best, but I thought this was a fun challenge to share with everyone to give a little insight into this late phase design process work! 

Hope you guys enjoyed a look at a little process work.  Happy June, and I'll be posting more soon!

Updates to the tool, and what it means for the narrative...
almost 9 years ago – Tue, May 23, 2017 at 09:33:30 PM

Hey everyone!  So, the Mystic Origins carts have been arriving in droves.  I know some are still making their way, but the majority of carts seem to have arrived at their destinations.  So far, the feedback has been great...a few helpful tips and bugs found, but mostly just a lot of excitement and responses of *feeling like a kid again*.  This has been the most gratifying thing to hear as a developer. 

I got to see a few screen captured first-run long-plays of the game, and I watched each one of them to see how players learned and adapted to the game, what confused them, what was intuitive, what was gratifying to find / learn, etc.  This was very helpful to be able to see the game to fresh eyes.  

In the meantime, developing Mystic Searches continues!  We made some major modifications to part of the NESMaker tool to mine for more narrative flexibility in Mystic Searches than we had for the prequel / beta.  Some of you have seen me posting about this (trying to get independent monster / NPC loads for day, night, day triggered, and night triggered, but running against the brick wall of limitations).

Well, it took some time, and we had to break a few things and figure out how to put them back together, but now we have the above set up working just as we want, and accidentally opened up even a few more possibilities we weren't expecting as a result.

Since I haven't done it in a while, I figured I would do a BTS of the tool as it pertains to these updates.  If you're interested in this sort of thing, check this out!

 This is a really rough, sort of unplanned last minute video that I did last night when I had some down time.  Anyone here want to see more about how the tool works?  I might start a small YouTube series that shows off its features one at a time.  Let me know what you think!

As for the game...super proud at how open world it is feeling and how dynamic and interactive it is.  Still lots of rough spots, but...overall, you guys have helped me create something way beyond anything I would have anticipated.  As always, thank you so much for all the support!

In the meantime, while you're waiting for Mystic Searches, check out Twin Dragons, another NES Kickstarter by a good friend of our project!

https://www.kickstarter.com/projects/1454702417/twin-dragons-a-brand-new-game-for-the-nes
You all rock.
Joe