| 1 |
Here is a list of compile switches, used to customize Radiance code |
| 2 |
for specific machines and users: |
| 3 |
|
| 4 |
-DMC If set, switches from default low-discrepency sequence |
| 5 |
sampling to true (pseudorandom) Monte Carlo. Use if |
| 6 |
the "brushed" appearance of specular highlights and |
| 7 |
penumbras bothers you. |
| 8 |
|
| 9 |
-DALIGN=(type) Alignment type, machine-dependent. Most RISC |
| 10 |
architectures align on 8-word boundaries (double). |
| 11 |
The default alignment type is int. |
| 12 |
|
| 13 |
-DSPEED=(MIPS) Millions of instructions per second for this |
| 14 |
processor (approximate). This is used to decide |
| 15 |
certain unimportant timing issues such as how many |
| 16 |
rays to trace before checking input in rvu and |
| 17 |
whether or not to optimize the color table in ximage |
| 18 |
on 8-bit displays. |
| 19 |
|
| 20 |
-DWFLUSH=(rays) Override for number of rays before flush in rvu. |
| 21 |
|
| 22 |
-DBSD Operating system has a strong Berkeley flavor, meaning |
| 23 |
that bcopy() and bzero() are present but maybe memcpy() |
| 24 |
and memset() are not. (See common/standard.h for other |
| 25 |
things this flag affects.) Also affects certain system |
| 26 |
calls, such as signal handling and resource tracking. |
| 27 |
|
| 28 |
-DSMLMEM The system has little RAM available, so size hash |
| 29 |
tables and the like accordingly. Only allows for |
| 30 |
small scene descriptions (32,256 primitives). |
| 31 |
|
| 32 |
-DSMLFLT This setting tells Radiance to use short floats |
| 33 |
(32 bits) throughout, which saves lots of memory |
| 34 |
but can cause calculation inaccuracies in many |
| 35 |
cases. Its use has been discontinued for this reason. |
| 36 |
|
| 37 |
-DSHADCACHE=N The array size to use for caching occluders. |
| 38 |
Setting this to 0 turns off this optimization. |
| 39 |
Values greater than 20 use over 4 Kbytes per |
| 40 |
light source, which can add up to a lot of |
| 41 |
memory in scenes with many sources. |
