Questions with the gps api

I decided to set up a gps tower and I'm having trouble getting a computer to get its position from the hosting cimputers

I cannot get a computer to get its position with "gps locate" I have 4 computers with routers up at y250 all running "gps x y z" with their specific locations. I have 3computers at the same y level and a 4th about 8m below the other 3. Is there something I'm doing wrong?

I've also tried having 2 in a line on that same y, and then 2more in a line 8m above the other two. The test computer still couldn't get its position.

Also made 3 make a right triangle on the same y with one 8m above one of the lower computers test computer couldn't get its location.

You can't just place computers randomly. Read http://www.computercraft.info/forums2/index.php?/topic/3088-how-to-guide-gps-global-position-system/ to learn how to make a correct GPS cluster.

Dlcruz129, on 05 April 2013 - 06:32 PM said:

You can't just place computers randomly.

Well you can. Trilateration doesn't state that the points must be in a cluster, if they did then our real world GPS system wouldn't work. To use trilateration all that is needed is 4 points at different locations to be able to work out its location, the more points the more accurate the location. That being said using the clusters shown in that tutorial does improve the reliability and accuracy of the GPS request AT CLOSE RANGE, if you get too far away from the cluster it starts being inaccurate, so you should always have multiple clusters. however a nice spread of GPS 'satellites' (i.e. a computer at world height) would do the same job as long as the node can get a response from at least 4 'satellites'. It should also be noted that for best accuracy with GPS in ComputerCraft you should have some computers serving GPS requests below the node as well, or at the very least different heights than other clusters… from my own experiences having all the hosts at the same height, and/or above the node, can sometimes cause the y position of the node to be calculated incorrectly.

I've never really read that article; it scares me. :P/>

I've read that post about GPS, but something isn't working for me. These are some setups I've tried.

didn't work - http://puu.sh/2uiy3
didn't work - http://puu.sh/2ujJ4

None of the gps turtles could find their positions.

Make sure that all the computers have a modem, you might have forgotten one :P/>

I've never worked with GPS's but don't they have to be high up?

remiX, on 06 April 2013 - 04:33 AM said:

Make sure that all the computers have a modem, you might have forgotten one :P/>

I've never worked with GPS's but don't they have to be high up?

They don't have to be up high, being up high just increases their range. Also I made sure each and every computer/turtle has a modem and is running its program. I even placed multiple wireless turtles all around the cluster in hope if one of them finding their position with no luck.

Ok this may sound like a stupid question, but heaps of people do make this mistake…. are you using the GPS api (not program api)? If using the api are you opening the modem on the computer you are attempting to get the location of.

Ok, I got it to work. The problem was I was using a gps program and for some reason the computers just did not communicate. I found out the computers already have a built in gps program that works perfectly.