Closed acrymble closed 9 years ago
For each tag (portrait, heraldry, map), we should personally check a small batch of them. These are our golden set. The others that have been tagged by other people are our maybes. The third reel will have a mix of these, with the first two reels already 'held' on images from the golden set.
If a player stops the reel and does not choose to respin, they are asserting that the third reel matches the first two.
Edge cases: assertions when there are no respins left should be taken under caution!
The difficultly of the challenge is varied by changing the proportions of golden:maybe:random-untagged on the reel, always ensuring that there is at least one correct answer on it.
(adding response to gameideas.md)
This ties in with a wider need then to coordinate these various 'golden sets', because I think we've now got the idea that each game should feed into a master dataset that learns from previous contributions.
We should probably then prioritise getting the data, deciding on the variables, and structuring that into a format that all the games can use and feed into.
On Thu, Jun 25, 2015 at 9:34 AM, benosteen notifications@github.com wrote:
(adding response to gameideas.md)
— Reply to this email directly or view it on GitHub https://github.com/BL-Labs/arcadeinterface/issues/25#issuecomment-115162275 .
@benosteen I've added a question about the slot machine game on the gameideas.md file.