Maps for bushwalkers

[GPSGuided]

Thanks SoaB for the expert take on this.

I guess the obvious next question is, for smartphone users, which style of mapping is most energy efficient? One that is obviously a factor when used without readily available recharge power. Or is that more a matter of app design than the dataset used? Do you review your app for power usage during its development? I assume that's just in CPU load, right?

[Son of a Beach]

I haven't compared power usage between vector and raster processing, but I think that it ought to be related mostly to CPU usage on mobile devices. On desktop devices with HDD, a large raster dataset would take more energy to run the disk (although they're usually spinning most of the time anyhow). On a mobile the secondary storage is on flash, SSD, or similar and not on HDD, so that file-sized based reading usage would be minimal.

For online maps, raster could use a lot of radio power, taking more time to download larger datasets. Of course this does not affect offline maps.

However, the single biggest usage of power in my experience (for my own app usage) is the location based hardware (mainly the GPS chip) when recording tracks - and this is unrelated to display of maps (vector or raster). In general, you only view a map for a few seconds but if recording a track, this can take hours. Of course if you're not recording tracks, this is irrelevant.

So for plain map viewing, I would assume that vector rendering would use more battery power than raster drawing simply because it uses more CPU (it certainly gets my iPhone hotter!).

Note that most of this post is just guess work and conjecture.

[GPSGuided]

Thanks! Educated or uneducated guess work and conjecture? I assume educated.

[Son of a Beach]

Well... self-educated by my own app development and by trial and error rather than formal education. Not sure if that counts.

[sthughes]

More important to me is memory usage. My iPhone is only 16GB, the full set of Tassie in just 1:250k, 1:100k, 1:25k and street atlases in raster format will take about half that capacity by itself (unless they manage to compress it more than the current geotiffs). OSM of Tassie (with 20m contours and admittedly with significantly less features mapped) comes in at about 35MB. So about 1/200th of memory usage. I'd love to know how big the data set for the vector Tasmaps used at on LISTmap is.
The other issue with Tasmap rasters is that some are quite poor scans, and have grid lines varying between ADG66 and GDA94 depending on which area you are looking at.

Personally I'm happy with the raster product for most uses. It's only when it comes to trying to load them on a Garmin GPS, or in the very old and/or poorly scanned areas that it would be nice to have access to the vector data.

By the way, Apple & Google Maps are both vector maps now (on iOS devices at least).