OpenTopography.org is a resource for a limited, but massive, set of LIDAR point data. LIDAR is a LASER-based method to determine topography. It is pretty dang nice when you want to look at the details of topography at a very large scale.
GDAL can process the returned LIDAR data. I've been using GeoTIFFs, since it is lossless, and isn't an ESRI specified format. GDAL is open source.
gdal_contour, gdaldem are two GDAL tools you will want to use if you want to visualize LIDAR data.
This image shows a LIDAR DEM that has been processed using the GDAL toolset. A hillshade has been generated using gdaldem and its defaults. A 5 m contour interval is shown, but there is also a 1 m contour interval in the legend. I turned it off because it obscured the hillshade. Imagine doing a 1 m contour interval on USGS DEM data: not a bright idea! This LIDAR data was downloaded at a scale of 1 m per cell (I haven't verified that is what I actually received, but it certainly looks great at incredibly large scales). I must note what you probably have noted: that is some awful labeling. The labeling engine in QGIS (used to display this data) is still under serious work; patience is necessary. Until then,Inkscape is a good way to manually, and deliberately, label your maps.
Notice at the center of the image there is an offset of two stream channels. Both have an offset at the same time. And if you look closely, you can see a faint trace of the San Andreas Fault. This is Wallace Creek! So if you wanted proof that the San Andreas fault is a right-lateral strike-slip fault, here it is. Please note that this creek has been beheaded by cummulative offset, not by a single event. The old stream channel is towards the west (left).
It is worth noting that the trace of the fault really stands out when you generate a slope map. Distinct breaks in slope are great for finding faults or landslides. At such large scales, you can really understand the geomorphology of a region.