TumbleSeed Postmortem: what went wrong | Hacker News Reader

TumbleSeed Postmortem: what went wrong

aeiowu.com

97 pointsdhotson30 comments

I really appreciated this write-up, and I can 100% empathize with the struggle of the game designer [1].

I'd like to give the author one piece of feedback: please refine your color palette to use contrast and hue to make it easier for players to identify the various game entities! Looking at the screenshots and watching the gameplay videos, it's really hard for me to pick out the what's important and what's just background. Everything blends together and just looks like noise.

Take a look at other games, e.g. Raptor: Call of the Shadows, which make it really easy to scan the screen and identify/differentiate background terrain, power-ups, enemies, and bullets. When it's hard to tell what's going on, you add an unnecessary barrier to entry.

That said, keep making games!! Seems like you're good at it. :)

[1] http://www.atomicarmies.com/

As an aside does anyone else find their landing page really poorly designed ?. It scrolls up instead of down and shows a bunch of static screenshots where I wasn't entirely certain where the playable character was. it was all quite confusing until I finally landed on the explanation video which should have been the first thing they showed.

http://tumbleseed.com/

Yeah, that landing page is not doing the game any favors. I closed the tab halfway through scrolling up the page.

I was about to do the same. I think this and their playability issues could have been caught very early on with simple A/B testing.

For a simple AB testing they would need thousands of unique visitors and players for just one test. Just not possible for an indie game I believe

I couldn't understand what you meant until I saw the scripts the browser was blocking. That does sound like a very unintuitive experience, even if it does fit the game's aesthetic. I can't even think of a language system that reads upwards.

Totally agree. I think the upward scrolling plays into a theme of the game itself: "JOURNEY UPWARD TO THE SUMMIT". But yeah, not very intuitive.

It also takes a long time before you get to the video. I was just about the give up scrolling and thinking to myself, just show me a video when the video was there. Then I stopped watching the video half way through because it all looks a bit the same.

I really like the graphics style, but like others have said. It needs contrast to differentiate between playable and non-playable elements. It all seems to blend into one right now.

Nice write-up. It's heart-rending to read an article about coming to terms with the fact that project you poured your soul into didn't work out (I've certainly been there).

The control system would doom this game for me. Squishy, indirect controls means a lot of frustration. The fact it is the central idea around which the entire game is designed doesn't help.

I think that's exacerbated by the visual design. It looks good, but a critical part of the mechanic is the ball ("seed"), is being rolled up a wall. You should intuitively see and understand the danger of a hole, the presence of obstacles, things you can roll up or roll down or over or through. But everything is flat and has almost uniform visual weight. There are some 3D cues, but they aren't consistent and don't seem to directly match the mechanical geometry.

It's tough to read the environment. This makes the indirect unresponsive controls even more frustrating: even when you can get the ball to roll where you want, you have a hard to seeing what you want to roll toward or avoid.

A crucial element that wasn't mentioned here was the lack of play-testing. Having a dozen people play the game and give feedback would have informed the makers that it was too hard, and where.

It ties in with the lack of time and resources too -- if you don't structure your development to have time to do user testing, you won't magically find that you have time to do it at the end, either.

My first thoughts too, although my guess is that they did let other people play the game to find out if it's fun, but there's def a difference between that and a structured blind playtest with randos. It's super crucial early & late in development.

And if they used the same people, they get better at playing the game, it gives the appearance of a game that is easier than it actually is.

That can be an asset though, and ideally you want to have both "longitudinal" testing with the same group of testers as well as frequently testing with people who are fresh to the game. The former can reveal trends that the development team is otherwise blind to. It can also help to exercise subtleties of the gameplay that a new player wouldn't discover, or aspects of the "elder game" if the game has a linear progression that takes significant time to unlock.

I agree with the value you're talking about, but they don't help evaluate how hard a game is.

They give an indirect nod to friends play-testing in the "Time" section:

> "Friends around us were saying “it’s too difficult” right to our faces, but amidst our nonstop-no-days-or-time-off working, we were too dull to let it inspire action."

So I guess they did have some people play the game -- but they didn't build in the time they needed to properly use the feedback, so it was almost the same as not play-testing.

My opinions:

More play testing would have helped a great deal, IMHO.

I played it for about 20 minutes. It seemed an okay game with a really awful user interface, and if they had ditched that mechanic they might have found themselves making something a lot more fun. I would up rage quitting in frustration over how artificially hard it was to control, thinking "These game designers hate their players, and it's a shame because they have something interesting hiding here."

Maybe I quit two minutes before it became a glorious game. That doesn't excuse the terrible experience I had at the beginning, though.

Decent read. I bought this game on the switch as soon it was release, and stopped playing probably 60 minutes into it because the difficulty really did my it less fun. I'll definitely play again once the patch comes out.

Overall the game has an amazing and original design, and I hope the patch improves sales.

Indie game development is pretty cruel compared to starting a business software company. No mercy, even early adopters are much more demanding and less forgiving, the launching it is a major part of the success (and it is much less lean as it is not just a CRUD, but a fully featured software published on another platform(s)), almost no chance to correct more fundamental flaws and pivot.

The first version of a SaaS is responsible for 0.0001% of its lifetime revenue. For an indie game it is what, 50%?

And a new SaaS is compared to your a home made Excel sheet. A new game is compared to Zelda, Mario and games built by thousands of people with millions of dollars.

For sure. The challenges are different but paying the bills with a boring business SaaS is way easier.

It seems casual styled platform-ish games are a tough thing to sell without some sort of catch, or interesting premise. I follow a fairly large number of indie games and looked at TS when it came out, but it was pretty easy to pass up.

It's a game in the purest sense of the word "game". It doesn't seem to have a catchy style, and it covers up the lack of fulfilling gameplay with the "rouge-like" descriptor which is so overused as to be meaningless. If it were on the phone, or I owned a switch maybe I could give it a try. On the PC, I expect a little bit more out of the video games I choose to spend time with.

To be honest, the game itself just _looks_ confusing. I can't tell up from down, where the character is, or anything useful from the website at all.

Wow, this article kind of makes we WANT to play it. I had read about it (before it was released, so didn't see a wall of complaints about difficulty) and watched the trailer, but the aesthetic and the core game mechanic led to a false assumption that it was a really simple game. I honestly almost bought it for my kids (who would probably have found it daunting). I'll pick it up on Switch when the patch is out.

I picked it up on switch and was majorly disappointed to discover that the controls don't take advantage of the joycons at all. It would make the game even harder but it would be fun to control it by juggling the joycons.

I also gave up pretty quickly because it is pretty hard

Good read, and perfectly captures the things that frustrated me about the game.

Looking forward to playing the updated version on the Switch!

Perhaps a dumb question, but in this case, could the core game mechanic have been testable by external players/users with minimal development to see if it is fun or not?

And a second question.... if it's too hard why not make it easier and launch again?

Article makes clear they just pushed a big update to Steam

I founded a startup called PlaytestCloud a while back to tackle this problem for mobile games. We have quite a few customers who build really basic prototypes and use our product to test that with outside players from early on. (It‘s important to test with outsiders because your friends might be too kind to you)

Interestingly in my sphere of critical game review this game did quite well. Everyone seemed to like it and it surprises me to see a postmortem like this.

They might be throwing in the towel too early, games these days seem to have long tails.

I think this 'throwing the towel' article is more of a PR stunt to get the word out and inform that a new patch is coming.

Their website is confusing and doesn't explain the game. The first video from the article's tweet is awful at explaining the game. Jump cuts every few seconds, talks about powers that it doesn't explain, gameplay is totally unexplained.

These guys need a non-artist on the team. No one is limiting them and making sure they understand that the game is more important than just the art.

Why not just make it easier instead of all the drama?

Reply on news.ycombinator.com