This game reeks of poor quality control (ranty)

Tl;dr: I don't think anyone at Supercell plays this game

If you're not into software or don't know how development cycles work, this may not make a lot of sense. I tried to make most of it English, the nuances will be lost.

I'm a software engineer by trade, and as a result older and more bitter than most people here. I started playing Clash Royale to get my fix of card playing games that I've been trying to fill since I stopped playing Yu-Gi-Oh in high school. Blah blah F2P blah blah learning card interactions blah blah.

Since I'm a bit of an anti-meta guy (played burn/stall in YGO, always trying to work different cards into my decks), I've been much less affected by the update. But this isn't about that. This is about the future of Clash Royale as a product (I may slip up and call it a game).

I'm sure you all know about the recent bugs, most of them related to the Executioner. People here were surprisingly ok with them, but I found it disgusting. I have little tolerance for incompetence, and the Executioner was on the edge of that distinction. These were incredibly obvious problems with the game and somehow no one at Supercell caught them. There are only 3 possible answers:

  • There are no troop interaction unit tests in the code base

  • There was no manual testing done of the new troops

  • Supercell gets a chubby putting out a broken card and calling it a "bug" later

  • Someone rushed the release

Option 4 is one that would make my blood boil. We'll leave it at that.

I choose to not believe the third option, because long term that is terrible for brand reputation and they can't afford multiple occurrences of this. Given the host of other bugs in the game, it seems more than likely that the third option isn't true. To reach the state of the Executioner pre-update, both of the first two options must have occurred.

Maybe they just forgot to implement tests for the newest troops? Then the most recent update should fix that then. Instead we have a host of other problems:

  • New targeting system (this is the nice way of saying "we couldn't be bothered to write a good targeting system to begin with so we'll patch it a little later") makes standard skelly counter useless.

  • Rapid fire Sparky when it's shot doesn't connect with it's original target (related to above)

  • Executioner's blade kills skellies as it comes back as if it were at full extension

  • Bandit challenge displays other person's language instead of yours (if this is a feature, it's a feature that makes me think your game is broken)

  • XBow and Darblin have incorrect hitspeed

And don't forget about other problems from before the update that would have been easy to test (Raged Electro Wizard perma-stuns Bowler and Pekka being the one I remember most vividly).

All of these issues would have been uncovered either (you should have both but either would catch these) having a real regression test or having unit tests for all troop interactions. Unit tests are the best because they're the fastest. Want to nerf Elite Barbarians? Change their damage output and hit speed and see which interactions change based on your failing tests. Always want Zap to stop short of killing Stablins? There's already a test for that and it'll fail when you change something too much. Then you pass that to the QA team for a regression pass to make sure what you changed makes sense while playing. But that's clearly not what's happening.

But my biggest gripe with the game right now is that "standard interactions" just don't happen on the ladder. I have level 8 fire spirits and aam facing a level 8. One spirit should die to tower damage and all the trailing ones should connect, right? Nope. That only start happening at level 9. Those people calling Furnace OP? I had no idea wtf they were talking about until I got one in a draft challenge.

Some other nonstandard things that happen regularly to me on ladder: Zap kills Minions, Fireball kills Ice Wizard. I get it, leveling up cards one at a time makes things inherently unfair. But who in the world thought trophies were the best measure of success? Anyone who plays the game know card levels determine most interactions. So why don't they play a role in matchmaking? This clan chest at 2600ish I ran into a level 12 RG and Barbarians. That should never happen.

I was taught the 5 paragraph essay so I'm supposed to wrap things up now. Based on what I've seen in this game, no one at Supercell plays the game like we normal shmucks do. They may test equal level unit interactions in tournament-style settings, but they don't care what happens to most people on the ladder. But more worryingly to me, and what you all should be worried about long term, is that there doesn't seem to be any troop interaction unit tests. The product will slowly become less and less stable as more patches and troops are throw in. Do yourself a favor Supercell, take a month to make every interaction tested (rage, troop dropped in front, troop taking tower damage, everything). And fix your matchmaking so your level 6 QA doesn't rage facing level 9 Royal Giants every week. If the guy getting paid wants to quit your game, you can bet your butt the gal you're asking to pay isn't going to be around for long.

Hopefully this all is coherent. I wrote this all on my phone.

Thoughts? Disagree? Want more features or more stability?

Leave a Reply

Your email address will not be published. Required fields are marked *