掲示板 Forums - Need volunteers for Shiritori Cat improvements
Top > renshuu.org > Feature Requests/Improvements Getting the posts
Top > renshuu.org > Feature Requests/Improvements
Looking to make some internal speed improvements to Shiritori Cat, but I need some volunteers (who play it regularly) to test the new code. Let me know if you're interested!
There will be no special requirements in playing except to just a)play normally, b)let me know if it seems faster, and c)report any bugs.
Hmmm, you can say I play regularly ( Not the kanji one but the hiragana one ).
I haven't played Shiritori Cat as often lately, but I'd gladly return to playing it regularly for a while if I could contribute data. Count me in if needed!
What about a new Friend System ( just a recommendation ) : Allow players to add friends and challenge them directly. Creating a social community around the game can increase retention.
This already exists for Shiritori. There are the public games under Games -> Shiritori (not to be confused with Shiritori Cat which is for solo play), and you can also make your own private Shiritori games and invite friends.
I went ahead and opened up an experimental setting for this to keep it easy. Menu > Settings > Experimental.
Please report any issues!
I would be happy to help! I’ll try to play (semi) regularly. Should I say any problems I find here, or on a different thread?
Should we activate it?
I activated it and I could see it has definitely increased the speed of the bot.
Hmm, so far no change for me after flipping the experimental switch. But I'll try again on the European server, Renshuu is often sluggish for me anyway, particularly on the app.
Didn't you notice any speed in shiritori cat? I took a timer test .
With enabling : 14 words - 2 min 10 sec ( with four hits )
without enabling : 14 words - 2 min 30 sec ( with four hits )
Running speed tests on your end is probably not going to be ideal: ever game is different, so two games cannot be compared in that way.
Rather, the older version was more variable in the speeds. Sometimes, it would take (this is just an example) .01 seconds to generate an answer for the cat, while other times it would take .5 seconds. I believe this new one brings them all closer to the lower amount, so that the "odd" ones are less frequent.
As long as it is not slower than the old/current version, then I can switch to it and more easily make speed improvements to that one.
I see where you are going, you're right .
This might result in a slowdown just for the period of testing, but I run both versions (old and new) one after another, then log both so I can get a better comparison between the two of them.
With that being said, the only thing I'll need to see here is if anyone has any functional issues (bugs, etc.) with it.
The new logging is in place, so it will be slower than before, but this gives me much better data to look into for speeding things up.