randomtest stop: not on the same line

Game Development ChoiceScript Help
1

(long time lurker, first time poster)
I am working on my first game. The first two chapters went fine, passed Quicktest and Randomtest after working through various little programming bugs, as one does. I am now testing the game through chapter 4. Suddenly, Randomtest is just…stopping, often in the middle of a line, when I try to run starting with my standard MO (10K iterations, seed 0). It looks like it’s just not rendering any more. It runs fine if I change the parameters (ie, 1K, seed 0, or 500 iterations, seed 997, etc).

I have successfully completed my standard run by just chunking through the parameters (fewer iterations, with different starting seeds). I have downloaded Choicescript, so I’m not running it through a browser.

So, I can get the whole game tested this way, but it’s annoying, and I imagine (assuming the problem persists) that it will only get more annoying as the game gets longer. I have read the threads here about randomtest stopping, but none of those circumstances seem to apply (wrong browser, loop in programming, stopping at the same spot).

Can you help? Thanks!

Screenshot 2021-05-29 12.08.21
Screenshot 2021-05-29 12.08.211920×1080 576 KB

2

This might be an issue in CSIDE. Did you say you’ve tried running randomtest from the plain ChoiceScript zip? Does that exhibit the same problem?

1 Like
3

I have never got more than 500 iterations of seeds done in one go in cside. As you say, it just stops. Do it in smaller batches, that should still get the job done.

2 Likes
4

I’m not sure what you mean about running it from the plain zip. Can you say more?

5

The version of Choicescript that you download straight from Github.

6

The seed value most likely isn’t the issue. Probably the game length and increased branching became too taxing for Randomtest to run so many iterations on. Like @malinryden, I also typically run for a few hundred iterations. It still catches most of the bugs that Randomtest is designed to catch (infinite loops and so on).

Although, if you run Randomtest out of the zip as suggested above, and it does work, do please let us know. I’d be interested to know how much of a factor CSIDE is.