Jump to content

  • Log In with Google      Sign In   
  • Create Account

We're offering banner ads on our site from just $5!

1. Details HERE. 2. GDNet+ Subscriptions HERE. 3. Ad upload HERE.


#ActualNorman Barrows

Posted 21 February 2013 - 09:59 PM

having map making skill affect automap range for terrain where it makes sense, and affect detail level of whats automapped both sounds pretty good.

That whole getting lost thing is the only problem.

I was figuring, if the player had their own map that could be inaccurate, if they automapped a cave to be 500 feet away from its actual location, they could just wander around til they found it. This would be very realistic. The player would follow the map to the general area, even to the point of being right at "x marks the spot", but no cave! the map's not so good, so some wandering about is required to actually locate the cave.

As the player moves about the world, things constantly get re-automapped at what ever their current map making skill level is. so with a separate player map, it would automatically become more accurate as their map making skill increased and they re-visited areas. if the automap does a "base location + random factor / map making" type thing, and each time it gets automaped, it can only become more accurate, not less, then places visited frequently (like your cave) would soon get zeroed-in on your map, even with a low skill level. that way both familiarity with the area, and map making would affect the accuracy of the player's map.

So should i just bite the bullet and implement this?

The big problem is the size of the world and having to page stuff off disk during play. Right now, the game has no loading of levels or anything like that during game play. just load everything at program start (10 seconds). savegame with automatic backup takes about 3 seconds. local player's map (explored bitmap mask basically) IS loaded and save during gameplay as needed, but its only 264x264 bytes, binary, with no lock on the write, so its not even a blip on the hard drive light. I have yet to even notice it, but the files are there, and the maps show up in the game, so it must be doing it. it uses a cache of 10 local maps with LRU paging.

on the world map, the location of caves etc in a map square is a constant. these would need to be stored as a sparse matrix for each map square, to allow the x,z values to be inaccurate. same thing for each local map. thus the concern with paging.

#8Norman Barrows

Posted 21 February 2013 - 09:59 PM

having map making skill affect automap range for terrain where it makes sense, and affect detail level of whats automapped both sounds pretty good.

That whole getting lost thing is the only problem.

I was figuring, if the player had their own map that could be inaccurate, if they automapped a cave to be 500 feet away from its actual location, they could just wander around til they found it. This would be very realistic. The player would follow the map to the general area, even to the point of being right at "x marks the spot", but no cave! the map's not so good, so some wandering about is required to actually locate the cave.

As the player moves about the world, things constantly get re-automapped at what ever their current map making skill level is. so with a separate player map, it would automatically become more accurate as their map making skill increased and they re-visited areas. if the automap does a "base location + random factor / map making" type thing, and each time it gets automaped, it can only become more accurate, not less, then places visited frequently (like your cave) would soon get zeroed-in on your map, even with a low skill level. that way both familiarity with the area, and map making would affect the accuracy of the player's map.

So should i just bite the bullet and implement this?

The big problem is the size of the world and having to page stuff off disk during play. Right now, the game has no loading of levels or anything like that during game play. just load everything at program start (10 seconds). savegame with automatic backup takes about 3 seconds. local player's map (explored bitmap mask basically) IS loaded and save during gameplay as needed, but its only 264x264 bytes, binary, with no lock on the write, so its not even a blip on the hard drive light. I have yet to even notice it, but the files are there, and the maps show up in the game, so it must be doing it. it uses a cache of 10 local maps with LRU paging.

on the world map, the location of caves etc in a map square is a constant. these would need to be stored as a sparse matrix for each map square, to allow the x,z values to be inaccurate. same thing for each local map. thus the concern with paging.

#7Norman Barrows

Posted 21 February 2013 - 09:59 PM

having map making skill affect automap range for terrain where it makes sense, and affect detail level of whats automapped both sounds pretty good.

That whole getting lost thing is the only problem.

I was figuring, if the player had their own map that could be inaccurate, if they automapped a cave to be 500 feet away from its actual location, they could just wander around til they found it. This would be very realistic. The player would follow the map to the general area, even to the point of being right at "x marks the spot", but no cave! the map's not so good, so some wandering about is required to actually locate the cave.

As the player moves about the world, things constantly get re-automapped at what ever their current map making skill level is. so with a separate player map, it would automatically become more accurate as their map making skill increased and they re-visited areas. if the automap does a "base location + random factor / map making" type thing, and each time it gets automaped, it can only become more accurate, not less, then places visited frequently (like your cave) would soon get zeroed-in on your map, even with a low skill level. that way both familiarity with the area, and map making would affect the accuracy of the player's map.

So should i just bite the bullet and implement this?

The big problem is the size of the world and having to page stuff off disk during play. Right now, the game has no loading of levels or anything like that during game play. just load everything at program start (10 seconds). savegame with automatic backup takes about 3 seconds. local player's map (explored bitmap mask basically) IS loaded and save during gameplay as needed, but its only 264x264 bytes, binary, with no lock on the write, so its not even a blip on the hard drive light. I have yet to even notice it, but the files are there, and the maps show up in the game, so it must be doing it. it uses a cache of 10 local maps with LRU paging.

on the world map, the location of caves etc in a map square is a constant. these would need to be stored as a sparse matrix for each map square, to allow the x,z values to be inaccurate. same thing for each local map. thus the concern with paging.

#6Norman Barrows

Posted 21 February 2013 - 09:58 PM

having map making skill affect automap range for terrain where it makes sense, and affect detail level of whats automapped both sounds pretty good.

That whole getting lost thing is the only problem.

I was figuring, if the player had their own map that could be inaccurate, if they automapped a cave to be 500 feet away from its actual location, they could just wander around til they found it. This would be very realistic. The player would follow the map to the general area, even to the point of being right at "x marks the spot", but no cave! the map's not so good, so some wandering about is required to actually locate the cave.

As the player moves about the world, things constantly get re-automapped at what ever their current map making skill level is. so with a separate player map, it would automatically become more accurate as their map making skill increased and they re-visited areas. if the automap does a "base location + random factor / map making" type thing, and each time it gets automaped, it can only become more accurate, not less, then places visited frequently (like your cave) would soon get zeroed-in on your map, even with a low skill level. that way both familiarity with the area, and map making would affect the accuracy of the player's map.

So should i just bite the bullet and implement this?

The big problem is the size of the world and having to page stuff off disk during play. Right now, the game has no loading of levels or anything like that during game play. just load everything at program start (10 seconds). savegame with automatic backup takes about 3 seconds. local player's map (explored bitmap mask basically) IS loaded and save during gameplay as needed, but its only 264x264 bytes, binary, with no lock on the write, so its not even a blip on the hard drive light. I have yet to even notice it, but the files are there, and the maps show up in the game, so it must be doing it. it uses a cache of 10 local maps with LRU paging.

on the world map, the location of caves etc in a map square is a constant. these would need to be stored as a sparse matrix for each map square, to allow the x,z values to be inaccurate. same thing for each local map. thus the concern with paging.

#5Norman Barrows

Posted 21 February 2013 - 09:58 PM

having map making skill affect automap range for terrain where it makes sense, and affect detail level of whats automapped both sounds pretty good.

That whole getting lost thing is the only problem.

I was figuring, if the player had their own map that could be inaccurate, if they automapped a cave to be 500 feet away from its actual location, they could just wander around til they found it. This would be very realistic. The player would follow the map to the general area, even to the point of being right at "x marks the spot", but no cave! the map's not so good, so some wandering about is required to actually locate the cave.

As the player moves about the world, things constantly get re-automapped at what ever their current map making skill level is. so with a separate player map, it would automatically become more accurate as their map making skill increased and they re-visited areas. if the automap does a "base location + random factor / map making" type thing, and each time it gets automaped, it can only become more accurate, not less, then places visited frequently (like your cave) would soon get zeroed-in on your map, even with a low skill level. that way both familiarity with the area, and map making would affect the accuracy of the player's map.

So should i just bite the bullet and implement this?

The big problem is the size of the world and having to page stuff off disk during play. Right now, the game has no loading of levels or anything like that during game play. just load everything at program start (10 seconds). savegame with automatic backup takes about 3 seconds. local player's map (explored bitmap mask basically) IS loaded and save during gameplay as needed, but its only 264x264 bytes, binary, with no lock on the write, so its not even a blip on the hard drive light. I have yet to even notice it, but the files are there, and the maps show up in the game, so it must be doing it. it uses a cache of 10 local maps with LRU paging.

on the world map, the location of caves etc in a map square is a constant. these would need to be stored as a sparse matrix for each map square, to allow the x,z values to be inaccurate. same thing for each local map. thus the concern with paging.

#4Norman Barrows

Posted 21 February 2013 - 09:58 PM

having map making skill affect automap range for terrain where it makes sense, and affect detail level of whats automapped both sounds pretty good.

That whole getting lost thing is the only problem.

I was figuring, if the player had their own map that could be inaccurate, if they automapped a cave to be 500 feet away from its actual location, they could just wander around til they found it. This would be very realistic. The player would follow the map to the general area, even to the point of being right at "x marks the spot", but no cave! the map's not so good, so some wandering about is required to actually locate the cave.

As the player moves about the world, things constantly get re-automapped at what ever their current map making skill level is. so with a separate player map, it would automatically become more accurate as their map making skill increased and they re-visited areas. if the automap does a "base location + random factor / map making" type thing, and each time it gets automaped, it can only become more accurate, not less, then places visited frequently (like your cave) would soon get zeroed-in on your map, even with a low skill level. that way both familiarity with the area, and map making would affect the accuracy of the player's map.

So should i just bite the bullet and implement this?

The big problem is the size of the world and having to page stuff off disk during play. Right now, the game has no loading of levels or anything like that during game play. just load everything at program start (10 seconds). savegame with automatic backup takes about 3 seconds. local player's map (explored bitmap mask basically) IS loaded and save during gameplay as needed, but its only 264x264 bytes, binary, with no lock on the write, so its not even a blip on the hard drive light. I have yet to even notice it, but the files are there, and the maps show up in the game, so it must be doing it. it uses a cache of 10 local maps with LRU paging.

on the world map, the location of caves etc in a map square is a constant. these would need to be stored as a sparse matrix for each map square, to allow the x,z values to be inaccurate. same thing for each local map. thus the concern with paging.

PARTNERS