ge 
tial 
ding 
ng 
for 
ere 
eloped 
ould 
2). 
ided 
with 
e 
divided into parts containing about 50 units and the coordinates of one 
randomly selected house were determined. 
The measurements were recorded on punch tape and the development 
densities of all areas were computed (Fig 5). 
Building heights were also measured to make it possible to separate low- 
rise and high-rise slab blocks - it is often difficult to count storeys 
in the stereomodel. 
Areas of different character required a varying amount of time, both for 
the use studies and the density measurements. We needed three stereo- 
models to cover the town. The compilation of basic data for the three 
maps took about 20 hours. More than half of this time was spent on the 
city center and its near surroundings. 
Unfortunately, it is difficult to check the accuracy of our maps. A 
photograph is a picture of one existing situation - town maps and plans 
are not. They are amalgamations of actual pictures of varying age and 
accuracy and sometimes of pictures that only existed in the planner's 
mind. 
The development density map; however; has been checked with manual 
measurements in about 20 areas. The maximum deviation was two per cent; 
and the average deviation was less than one per cent. 
Later, we used the same stereomodels to measure building volumes. The 
measurements were needed for a statistical study of energy consumption 
in buildings. Building volumes on 88 properties were measured. We used 
a Wild A8 stereoautograph with a Kongsberg photogrammetric digitizing 
system. Computations from the photogrammetric data were performed on a 
gpc system. 
The properties were listed and had to be located at first on a map and 
then in the stereomodel. which presented some difficulties in areas where 
the property borders were indistinct.