next »

[raxter]

6x Mass Production



Info
6x Mass Production is a puzzle game about constructing things out of hexagonal blocks.

My intention with this game is for it to be a truly challenging but fun puzzle game. I've always loved games like SpaceChem and Codex of Alchemical Engineering and this is my attempt to create something like that experience.

Currently the majority of the levels introduce the mechanics and the last 5 or so levels are super challenging (this is the actual meat of the game). I'd love to make more challenging levels if I think enough people would enjoy playing them.

F1 will display debug text in the top left with the level name and number if you need it.

Feedback I am looking for
I will take any and all critiques and comments so anything that you notice, please say!
 
I specifically want to know if the game is enjoyable to get into and if the learning curve is not too slow or too fast. Or if anything or any particular level was confusing. The last 5 levels are the challenge levels (the rest are basically an in-game tutorial) and I would like to know if those sorts of levels are fun to play or if they are a bit much.

The other burning question I have is a little more difficult to answer. I've been working on this game for far too long a little while now and it's time for it to sink or swim. I would like to know in your honest opinion if the game is worth taking further or if it's not strong enough (and why).

I'd also like to find out how to find an audience for this game. Is there a puzzle game forum I can go to to get this game name out there? This isn't a particularly casual puzzle experience and the feedback I've gotten so far is that it would suit a great niece audience (same crowd who would like SpaceChem and other diabolically hard puzzle games). Just... where would I find them and how would I go about building a community?

Webplayer version here: http://www.raxterworks.com/6xMassProduction/6x%20Mass%20Production.html

Many thanks!

p.s.
Windows download: http://www.raxterworks.com/6xMassProduction/6x%20Mass%20Production%20WIN.zip
MacOS download: http://www.raxterworks.com/6xMassProduction/6x%20Mass%20Production.app.zip
Linux download: http://www.raxterworks.com/6xMassProduction/6x_Mass_Production_LINUX.zip

--------
www.raxterworks.com
@raxterbaxter

[jgrams]

F1 will display debug text in the top left with the level name and number if you need it.

Except that it appears to be zero based, where the level select display is one-based...

I specifically want to know if the game is enjoyable to get into and if the learning curve is not too slow or too fast.

It seems quite slow to me...e.g. levels 7-10 could easily be compressed into 1. Similarly probably some of the earlier ones?

• Level 7 (180 or 120 degrees) has help text saying it can be solved with 5 instructions, but you only need 4. This is bad because it makes me not trust the level designer...
• The delay between levels is annoyingly long. Is that a technical limitation (loading time), or can you eliminate it?
• It's annoying that you can't add instructions at arbitrary positions. Don't force me to work left-to-right, just fill any undefined ones with skips. For instance, on level 13 (but you'll have to work for it, the first level which was even slightly interesting) I wanted to start by putting the "close" instruction for the second arm in a matching position with the "open" instruction for the other arm and then fill in the rest from there.
• It would be nice if I could just drag the arms to rotate (edit: and extend) them instead of having to hit the tiny rotate arrows.
• If you're going to lock things, you should lock them. For instance, it seems like you can't modify the instructions in Paired Up, Paired Off, and Paired Out, but you can still drag them around, so you can solve all three exactly the same way: no need to think.

That's all I have time for ATM, I'll mess with it more later.


OK, finished the tutorial levels and did the first three challenges. That's kinda fun. I don't know if I'd keep playing though: I think I'd rather write real programs than play around with a limited world like this...

I got:

Challenge 2: 4 grabbers, 22 instructions, 367 steps (I wondered if this could be done better? edit: Oh, of course it can. 3 grabbers, 15 instructions, 303 steps)
Challenge 3: 4 grabbers, 20 instructions, 365 steps (pretty happy with this one. edit: or not. 3 grabbers, 15 instructions, 364 steps)

Found a few more things to comment on:

• It's odd that you have this excruciatingly long set of tutorial levels but don't cover the extend/retract instructions.
• It took me a while to figure out that you can just mouseover the grabber and use delete and the hotkeys to edit the program straight from the keyboard. Maybe you should mention that in the tutorial somewhere? Before I noticed that, I kept getting annoyed at how clunky the "move mouse, hit key, repeat" loop was and wanted keys to move around through the program. Might be nice anyway...
• It would be really nice to have hotkeys to control the simulation.
• The extension arrows are using the tool-tips for rotation.
• The "rotate part" command icons are hard to read. It looks to me like they are just pixelated?
• The multiple arrows when dragging an instruction are cute, but they're more confusing than helpful. I'd stick with only one.
• It would be nice if hitting Escape closed the instruction menu: I did eventually realize that I could just click on the white ring again, but at first I didn't think there was any way to close it without selecting an instruction.
• When paused, you can modify the programs by using Delete and the hotkeys.
• In challenge 3 where you're creating a ring, there's no welding effect when you close the ring. The join indicator appears, but there's no sound effect or sparks. I assume it's checking whether they're part of the same piece rather than whether there is a join between those two parts...?

[raxter]

Thanks jgrams for the details feedback. It's really, really useful!

I'll certainly be sifting through the specific comments, fixing the bugs up and addressing the concerns. Thanks so much!

One of the general things that I'm finding from this round of feedback (and you mentioned too) to is that the level progression is quite strained for some but ok for others, but generally quite strained. I'm thinking about a non-linear level progression where expert players who get it can progress very fast.

Thanks again this is going to go a long way to making the game a better one

[jgrams]

OK...maybe I would keep playing. That sixth challenge is a killer. Took me way too long to solve it.

I wonder about the level progression...I think maybe you're focusing too much on the mechanics (which are fairly straightforward) and not enough on the strategy (which is the tricky part). It seems silly to have levels where all you have to do is hit play, but you probably need more cues for what to do at first. The first level where you have to choose your own instructions, I sat there blankly for a minute. So maybe you could have levels where there's one arm doing something, and you have to program another arm to pick them up and move them. The mauve hex showing where to place the arm in that one level was a nice touch.

Or having levels where there's an existing program which *almost* does the right thing and you have to fix it. So you would learn that when you drag an arrow from the end to the start, you have to rotate the start position in the opposite direction. Also teaching the "prime factorization" thing like in Challenge 4 (?) where the easy way to do part of the construction is in 5-step programs, but then you want a 25-step program, and you can't do that, so you have to make it all in 6-step programs so you can have a 5*6 = 30-step repeat on the final two arms.

And teaching how to trigger on things where a part is only in a certain place when it's finished. But doing it all in very small levels where it's easier to see and then working up to more complex puzzles.

Prismatica did this really well when Thordur was playtesting it...I'm usually rubbish at Rubik's cube-style puzzles, but it introduced things with very simple puzzles to start, and I felt like I really got it.


Also I noticed that when dragging an instruction to the very end of a program, there's no arrow to show where it goes.

[raxter]

Hello!

Sorry for the late reply.

Awesome! Yeah I really hear you on the level progression, it's something that's been in the back of my mind for a while but having it pointed out really solidifies it for me. I'm trying to toe a line between so many players I think I might have to come up with something really unique to get over this problem (or just tailor it to a more 'expert' audience and not worry about people not getting it :p).

I really like the idea of completing the last bit of a complex-but-not-overwhelming level. It would also allow the player to play around if they feel like it (though maybe giving too much freedom early on might be confusing - *throws hands up in air with deflated not-knowingness*)

Will ceratinly have a look at Prismatica! Looks a great resource!

Generally what I am fearing (and what is becoming more apparent) is that I don't have a good understanding of the mental steps that people are going through when they are learning the game (because I'm playing it at such a developed meta level I don't see them anymore). As a result I'm makking tutorials that are missing the mark in most cases (rather than hitting the mark in most cases with my intended audience).

On an interesting side note: I recently read an article by Zach Gage called Designing For Problem-Solvers about forming mental models that really cleared up a lot of things for me in this area of how problem solving people learning.

[jgrams]

That's an awesome article! In that vein, I bet extending the "player" to be a debugger would go a long way toward giving people that sort of sandbox...when I was working out my solution to "Probably Impossible" one of the most tedious things was how long it took to get to the problem spot. So I'd run it fast, and then I'd overshoot. Or I'd stop at the right place, but then not be looking at the right thing when I ran the critical steps. Or...I thought I had the construction steps correct, but once I added the "move it to the finish area" arm I found that it didn't construct the second piece correctly.

So if you could debug it on the fly: run it, back it up, move things, control the arms manually, add/destroy blocks/pieces... Maybe even the ability to edit the starting layout while leaving the current debugger state as a transparent overlay as a guide? You could just make it so that using any debugger features disabled the ability to win the level until you reset the simulation.

Being able to undo changes to the layout would also help with that feeling of freedom to experiment, I think. There were times that I wanted to make drastic changes to try things, but then how do you get back if it doesn't work out?

OK, I'll stop pestering you now.