To latest comment
The comment you are looking for have either been moved to a new discussion or been deleted.

Put a limit to the minimal distance needed for 5000 points, and display distances with .01 precision

Hello and merry Christmas for everyone celebrating!

When checking the best  scores on some official maps, I noticed the maps covering a town (London, Paris, New York, San Fransisco, Philadelpha maps) don't have any player with a perfect score. The reason is simple: to get 5000 points you need to get very precise. TOO much precise. In Paris, you lost a point with less than 1 meter, which is insane. Even with cheating, you can't get a perfect!

Here are two examples:
• Paris map, the current leader made a guess 0.6m away, and still lost a point: https://geoguessr.com/results/8go59F8f98gbSGnY
• London map is way worse: 0.1m off, and still losing a point! 10cm isn't enough! https://geoguessr.com/results/lz62yrJHkWk49dHP

This is just unplayable, because Google Maps itself isn't enough precise to get that close. 10 cm is impossible to see on Maps, it's already a matter of luck to be under 5m in most of cases. This makes small maps being so random as the luck takes over the skill in most of cases.

So I suggest to put a minimal distance required to get 5000 on all the maps (official and non-official). I think 5 or 10 meters should be enough to score 5000. Even 5 meters may be hard, as sometimes the result place is so misplaced (I once got a place that was like 40 meters off the actual road, when the blue Streetview line was actually on the road on Google Maps). I personally would go for 10 meters, and eventually increase / reduce it a bit in case it makes things too easy / still not enough reasonable.

I add another request: put the distances with 0.01 precision. You can see on the London high score that the player did two rounds with 0.1m off, and one round scored 4999 while the other one scored 5000. It's because the actual distance was more like 0.12m vs 0.18m or something similar, but we don't see the .01 number. This will let us getting more detailed stats, especially on such precise guesses.

Hope you'll consider it!


KillerMapper Report inappropriate content

Comments

  • Hi KillerMapper

    This is an interesting idea. As you probably noticed we're currently using a scoring algorithm that calculates the "max-scoring-distance" based on the other locations of the map (for example you have to be closer on the Paris map than the world map). The main idea is that we want to have a singular scoring algorithm that works on all different kind of maps. However, I totally see your point with adding a sort of "grace distance" that is rewarded with maximal points (as you said Google maps often have an error). I think we have to investigate this a bit as it will affect such a basic function in the game, but we'll definitely consider it!

    Regarding the extra decimal it should be a simple change but we'll need to test this a bit to make sure that the result screen doesn't become to cluttered for the more casual users!

    Thanks for the feedback!
    Regards,
    Anton
  • This is a thing I find really problematic with the game, because the difficulty of pinpointing the exact location should be independent of the map you are playing. You say "you have to be closer on the Paris map than the world map". May I ask why? The difference between the two maps is that it takes some time on the world map before you realize that you are in Paris, but from then on, they play exactly the same, so why must I be more precise on the Paris map than on the world map, when I start from the same state (knowing that I am in Paris)?

    I think the algorithm should be made such that it scales from max points at the same distance from the location on all maps, to zero points at a certain percentage distance away from the location depending on the size of the map.
    Frank Bertelsen

Add your comment or create a new post

Your name and post can be seen by everyone.Your e-mail will never be shown publicly.