| 51 |
|
vector or matrix, which may be read from the standard input if unspecified. |
| 52 |
|
The daylight coefficients are multiplied against these sky values |
| 53 |
|
and the results are written to the standard output. |
| 54 |
< |
This may be a list of color values or a combined Radiance image, |
| 54 |
> |
This may be a list of color values or a combined Radiance picture, |
| 55 |
|
as explained below. |
| 56 |
|
.PP |
| 57 |
|
In the second form, |
| 58 |
|
.I dctimestep |
| 59 |
|
takes four input files, forming a matrix expression. |
| 60 |
< |
The first argument is the View matrix file that specifies how window output |
| 60 |
> |
The first argument is the View matrix that specifies how window output |
| 61 |
|
directions are related to some set of measured values, such as an array of |
| 62 |
< |
illuminance points or images. |
| 62 |
> |
illuminance points or pictures. |
| 63 |
|
This matrix is usually computed by |
| 64 |
|
.I rfluxmtx(1) |
| 65 |
|
or |
| 126 |
|
will be replaced by the time step index, starting from 0. |
| 127 |
|
In this way, multiple output pictures may be produced, |
| 128 |
|
or separate result vectors (one per time step). |
| 129 |
< |
If input is a matrix rather than a set of images, the |
| 129 |
> |
If input is a matrix rather than a set of pictures, the |
| 130 |
|
.I \-x |
| 131 |
|
and/or |
| 132 |
|
.I \-y |
| 133 |
< |
options may be necessary to set the output image size. |
| 133 |
> |
options may be necessary to set the output picture size. |
| 134 |
|
If only one dimension is specified, the other is computed based |
| 135 |
|
on the number of rows in the result vectors. |
| 136 |
|
.PP |
| 149 |
|
.IP "" .2i |
| 150 |
|
gensky 2 10 15:30 | genskyvec | dctimestep workplaneDC.dmx > Ill_02-10-1530.dat |
| 151 |
|
.PP |
| 152 |
< |
To compute an image at 10am on the equinox from a set of component images: |
| 152 |
> |
To compute a picture at 10am on the equinox from a set of component pictures: |
| 153 |
|
.IP "" .2i |
| 154 |
|
gensky 3 21 10 | genskyvec | dctimestep dcomp%03d.hdr > view_03-21-10.hdr |
| 155 |
|
.PP |
| 182 |
|
To multiply two matrices into a IEEE-float result with header: |
| 183 |
|
.IP "" .2i |
| 184 |
|
dctimestep -of Inp1.fmx Inp2.fmx > Inp1xInp2.fmx |
| 185 |
+ |
.SH NOTES |
| 186 |
+ |
.I Dctimestep |
| 187 |
+ |
optimizes its matrix concatenation by checking for all-zero rows |
| 188 |
+ |
or columns, thus avoiding unnecessary vector multiplications. |
| 189 |
+ |
This can improve performance when a daylight matrix contains |
| 190 |
+ |
zero-filled column vectors corresponding to hours of darkness. |
| 191 |
+ |
.PP |
| 192 |
+ |
It rarely makes sense to specify the |
| 193 |
+ |
.I \-od |
| 194 |
+ |
output option with |
| 195 |
+ |
.I dctimestep, |
| 196 |
+ |
since matrix operations are carried out using 32-bit "float" values. |
| 197 |
+ |
This take less memory, but can also be less accurate than an |
| 198 |
+ |
equivalent invocation of |
| 199 |
+ |
.I rmtxop(1), |
| 200 |
+ |
which performs all operations on 64-bit "double" values. |
| 201 |
|
.SH AUTHOR |
| 202 |
|
Greg Ward |
| 203 |
|
.SH "SEE ALSO" |
| 204 |
|
gendaymtx(1), genskyvec(1), getinfo(1), |
| 205 |
< |
mkillum(1), rcollate(1), rcontrib(1), |
| 205 |
> |
mkillum(1), rcollate(1), rcontrib(1), rcrop(1), |
| 206 |
|
rfluxmtx(1), rmtxop(1), rtrace(1), vwrays(1) |
