Share this project

Done

Share this project

Done
Humanity no longer exists - unless you go back in time and save everyone in this "Gauntlet meets Terminator" style top-down shooter!
Humanity no longer exists - unless you go back in time and save everyone in this "Gauntlet meets Terminator" style top-down shooter!
Humanity no longer exists - unless you go back in time and save everyone in this "Gauntlet meets Terminator" style top-down shooter!
121 backers pledged $10,320 to help bring this project to life.

The Big Honkin' Update

1 like

So, for those looking for a tl;dr version of this post:  Yep, Crucible is still coming (and we've a got $10,000 incentive to finish and ship this - we're part of the Free The Games Fund, and that's worth quite a bit).  We ran into some interesting problems along the way, re-thought quite a bit, changed engines, and simplified it back to our original scope:  Terminator Meets Gauntlet.

Now, for the long version, for those that want to know all the juicy (and slightly embarrassing) details!

Phase 1:  Failure #1

One of the hardest things for a game developer to do is look at something the team has designed and say "This sucks.  It's boring, it's too predictable, and it's not fun."  A lot of the time that would be "The End" for a project - you'd wrap it up and walk away.

However, I'm rather emotionally and financially invested in the concept of Crucible.  The core concepts were cool.  Terminator Meets Gauntlet - how could this idea go wrong?

There's a concept called "Feature Creep".  Start adding in new, cool ideas, then forget to set your foot down that we can't add anything more.  Just, well, keep adding cool ideas to the basic, simple concept we had.  Eventually, you'll end up with something that no longer matches your original idea (which, honestly, isn't always a bad thing.  Sometimes the original idea is cool, but the expanded idea is much cooler.)  In the case of a product that you're developing in-house, without a promised concept, varying from the original concept is just fine.  

In the case of a Kickstarter, we've got the original idea, and we're restricted from straying too far from that.  If we produced, for instance, a 2.5D Isometric Rouge-like instead of a Gauntlet style game, it might be cool, but it's not what we said we were going to deliver.

We also didn't like the graphics.  We knew from that all the graphics (except possibly the logo) would end up ditched, and we'll be creating new stuff from scratch - hey, that's what part of the Kickstarter budget is all about, so no big deal.

So, we looked at this bloated pig of a design, what we had for graphics (including new graphics we had hired in), and our internal testing told us something important: the game sucked.  It just wasn't right.

Phase 2:  Exploring The History

A completely new tactic came about:  we needed to play Gauntlet, all over again, from beginning to (non)end. (For those who've never play Gauntlet till the end in the arcades, you're not missing anything - there is no end to the game.  It rotates the levels, and plays them again.)  There's only a few iterations of Gauntlet we didn't play.  For instance, we didn't touch Gauntlet: The Third Encounter for the Lynx - that game was actually a whole different game by Epyx that Atari rebranded as Gauntlet to try and increase sales.  Oh, and it sucks, a lot.

That includes playing a number of the more modern games (Gauntlet Legends), but not the most recent Gauntlet game.  We even went as far back as Dandy / Dark Dungeons (Dandy is the direct inspiration for Gauntlet)

Oddly, one of the things we kept coming back to was Gauntlet 4.  While it doesn't have some of the ideas we wanted, and has some things we have no interest in, it had a really good difficulty balance for most stages.  However, the game is actually a bit short - 40 levels.

After playing it all over again, we sat down, and pounded out the core tenants of our design guide, and started picking features that would be put in, and what wouldn't, very, very carefully.  80% of our over-bloated design from the Phase 1 failure ended up on the cutting room floor, so to speak.  We were going back to the true core of Gauntlet, and extending it very careful into the time-travel concepts.

We did lose one feature that I'm sure is going to make it into some other projects some other time:  the idea that death was a setback that you could overcome easily (in fact, it was a way for you to "cheat" and continue to increase your abilities until you could tackle a level that kept killing you.)  It didn't feel right.  Here's the thing: Gauntlet was a quarter eater in the arcades.  There has to be some sort of suspense in the game - in the arcade, if you ran out of health, put in more quarters.  Eventually, you'd start to hit an investment wall:  at what point did you quit putting in quarters to make it even further?  Death wasn't permanent, but that health counter cost you real-world money.  There was suspense.

With the home versions, they made up for it (in some versions) by giving you a quarter count.  But, death was a permanent thing.  Heck, if you just fire up an emulator and play Gauntlet that way, it starts to get boring without some sort of limitation on it.  So (insert evil grin) we put a new and different limitation into the game that uses our time-travel system.  Some of you will hate it, some of you will love it, but it's a good solution to making the player sweat a little bit from time to time ;-)

Phase 3: Failure #2

Then came another heart-aching inducing failure:  We redesigned, changed the pacing, and started building again from the ground up.

The engine couldn't handle it.

This shocked us.  We had already used the engine for Jumpman Forever - there shouldn't be any issues.  But, simply put, when we had a full level, with lots of stuff going, the collision detection system would crap out.  After lots of investigation, we discovered the problem, and well... it wasn't within our scope (or quickly dwindling resources) to rewrite a huge chunk of someone else's engine.

So, fall back again.  Now, this also put us in a second bind - we had planned on standardizing on an engine, and running with it for the next however many games after Jumpman Forever and Crucible.  This sort of kills that plan.  Now, in all honestly, we had ran into some performance issues with Jumpman Forever, but we worked around them here and there.  With Crucible, we just couldn't get a good way of having that happen.

And, to further cause drama, we wrote custom code for Jumpman Forever to get OUYA support (required by OUYA's Free The Games Fund, for obvious reasons), and engine updates never came that integrated full OUYA support, and the person who directs development on the engine we were using declared that OUYA was a "dead platform".  (We don't quite a agree with that - OUYA moved their business model, but uses a lot of the same tech as before.)

Which means, every time there's an engine update, it risks breaking (our currently broken!) OUYA Everywhere support - which makes long-term support for the platform a problem using that particular engine.

We fell back, and decided to look at all the available options to us, and looked at everything from cross-platform support (a must, including OUYA) to performance.  

Eventually, we nailed it down - we rebuilt on top of the Unreal Engine 4 platform, using their 2D setups.

Phase 4:  Moving Right Along

Of course, we've burned damned near every bit of our budget going into a third overhaul on the project.  That sucks.  But, we're not giving up.  At the moment, well, I've got to work some regular gigs to continue development on it, but that's OK.  Not including stuff like game sales later down the road, we've still got $10,000 from OUYA riding on shipping the full game.  But, along the way, other interesting things happened, too - one of them was being approached by a publisher for Jumpman Forever and Crucible.   That further solidifies the need for a very solid product when we launch.

Were are we now?  We started the real work in the new engine about a month ago, and are reworking artwork to better fit the UE4 engine (we actually licensed it before that, but we needed to take some time to get up to speed on a new engine and art pipeline, so that we didn't shoot ourselves in the foot. Again.)  

Of course, there's still more to the story than just this - for instance, the lead developer getting his life turned up-side down due to a divorce, and plenty of other development drama.  But, this gives you a really good idea why we were silent - as we went into Phase I, and started seeing issues, we got quiet over what was going on.  After deciding to rethink it all in Phase II, we didn't want to say "Hey, we just trashed all our work, but this time, we've got it right!" (and then have Phase III happen, and having to say "Uh, well, see..." ;-)

Contact Davis

So, one of the quirks of Kickstarter that developed quite a while back (and I mentioned it on the Jumpman Kickstarter, I think) is that Davis, the lead developer, never gets email from Kickstarter.  No idea why.  So, feel free to contact him directly at daviss@midnightryder.com if you've got any questions - you'll get a whole lot faster response than waiting until someone taps Davis and says "Hey, did you see that comment on...?" ;-)

Thomas R. Melinsky, Jr. likes this update.

Comments

Only backers can post comments. Log In