Issue Nr. 5
Feb. 20th 2001.
Contents:
Starting yesterday (Jan. 19. 2001), users of Rayfront can download
an upgrade to the newest version 1.0.1, which fixes a number
of issues that became apparent in the use of 1.0. There are two
types of changes in this update. One includes actual bugs in
Rayfront, that have their cause in errors from my side. Those
usually became manifest when working with the Rayfront user
interface, or when interacting with AutoCAD, and would produce
strange error messages or other funny effects. All reported
problems that I could reproduce or otherwise identify are fixed in
the update.
The second type of changes are fixes to Radiance itself. As a
convenience for users working with both Windows and unix systems in
parallel, the included Radiance binaries for unix now accept input
files with DOS style line endings in all situations. This means
that you now can edit all files with a normal text editor on
Windows, if you have the need to do so, and nothing will break
because of that. Obviously, this possibility is something that is
of most relevant to expert users, likely with previous Radiance
experience.
The biggest chunk of changes, however, is in the rview.exe
previewer program for Windows. I had already tried to fix the
most obvious bugs in this program for Rayfront 1.0, but introduced
some unnecessary restrictions on it's use while doing so. In the
mean time, I found that the basic architecture of the program was
fundamentally flawed. There are some technical details involved,
but the core issue is that an interactive program on Windows may
only trigger long computations in very specific places in the code.
The previous version of rview (as the one included with Desktop
Radiance) called the code sections to execute the actual
simulations from non-standard points in the program flow. This
caused a number of strange effects, and a significant waste of
memory. I managed to fix those problem for this update, with the
result that rview.exe has become more robust and reliable than
ever.
Find the download for the free update to Rayfront 1.0.1
here.
With the same update, Rayfront now also supports IntelliCAD
2000 in exactly the same way as already was the case for
Autocad R14/2000. IntelliCAD is a low-cost CAD alternative, which
was developed with the explicit goal of maximum AutoCAD
compatibility. This starts with using the identical file formats,
and doesn't end with a very similar user interface. For more
details, visit the Intellicad Technology Consortium at
http://www.intellicad.org/, where you'll also find a list of
members, most of which sell the Program on their sites.
The advantage of supporting IntelliCAD is very easy to see. Of
course, every supported system is an advantage in itself, but there
is much more to it. IntelliCAD is avaliable at a very reasonable
(not to say "cheap") price. The software is developed in a joint
effort by many consortium members, at least partly following the
open source concept. Someone very adventurous could even get the
source code to it and implement any custom features themselves,
though that may not be an option for the normal user.
It is interesting to find that a supported CAD system now actually
costs a lot less than Rayfront, which is in strong contrast to the
situation with AutoCAD. Even if you create your geometry with a
completely different program, it might be worth to get IntelliCAD
just for working with Rayfront, as this simplifies the data
transfer quite noticeably. Despite it's very affordable price,
IntelliCAD is quite close in functionality to AutoCAD for
most practical purposes. What is currently still missing are object
oriented modules like the ACIS solid modelling kernel, and a number
of the more advanced application programming interfaces.
This is almost becoming a "running gag" in this newsletter, but I
hope that I can finally bring it to a resolution in the near
future. There have been long standing plans to create a plug-in for
simulating daylight redirection systems with Rayfront. The
technology to do this was developed by Thomas Schmidt of
http://www.exergia.de/, and has been demonstrated to work very
reliably within a useful range of accuracy. Unfortunately, Thomas
had other work to do, which so far kept him from turning this
technology into a marketable product.
We are currently discussing different concepts of how to finish
that work. The most likely scenario is that I will do it, now that
Rayfront is finished and the first round of bugs fixed. The biggest
part of this will be to adapt the interfaces between the two
systems. We already had a prototype of this working quite some
time ago, but will follow slightly different concepts in a
final version. Negotiations on the exact proceedings are still
going on, and may in part depend on the the result of the next
paragraph.
To get a better picture of the potential of a tool like
this, I would like some input from my readers at this point. If
anyone of you can imagine simulating daylight redirection systems
with Rayfront in their practical work, please tell me about it! You
can just reply to this message (but please remove the text of the
newsletter in your reply). The information I am interested in
would be the amount of demand for such a solution, and which types
of environments that you would use it in. Please tell me if your
interest is for commercial use in actual projects, as a research
tool for evaluating those systems, or even for developing such
systems. Of course, the results of this poll will not give me any
hard figures about how many copies we might sell, but a general
indicator that there *is* interest in such a product would already
be tremendously helpful.
Note that the specific transmission data for all reflective and
refractive profile systems can currently be prepared
"synthetically", by another program. For systems with diffuse
surfaces or other geometry types (non-translation-invariant), the
same data can be found through measurements in a goniophotometer.
If there is any demand, then I could coordinate such measurements
in a scientific institution in germany, that also just began to use
Rayfront. Of course, data for the most common systems currently
available on the market would be included with the software, so
that a typical users will not have to worry about such issues.
A number of systems, most of which could be included in simulations
right away, are
documented here:
Please tell me about your thoughts and wishes in this area by
sending a short message to
schorsch@schorsch.com
and I'll report
more on any progress in the next issue of this newsletter.
Website Reorganization
I have reorganized the web site a bit, so that there is now an
extra area for all specifically Rayfront related information.
The support and manual sections have previously been part of the
Knowledgebase, which was a mix of general lighting related
information and data about Rayfront. Now there is a clear
distinction between those areas, which will hopefully make it
easier for everybody to find what you're looking for.
The new Rayfront product pages.
Support Mailing List
Now that the support mailing list starts populating, I'll have to
hurry up on implementing the message archives as well. This will be
a significant part of the Rayfront support area on the web site,
where you can check if any of your problems with the software has
already been encountered (and possibly solved) by another user.
Last Minute News
As just announced by Charles Ehrlich on the Desktop Radiance
mailing list, another comparison study between several lighting
simulation programs has recently been published:
ACEE Daylighting.
Ubbelohde, S. and Humann, C. 1998. Comparative Evaluation of Four
Daylighting Software Programs. 1998 ACEEE Summer Study on Energy
Efficiency in Buildings Proceedings. American Council for an
Energy-Efficient Economy.
As to be expected, the study finds that Radiance gives the most
accurate results for daylighting simulations in non-trivial
architectural environments. Still, very interesting reading.
Issue 08 (2002 Sep. 9.)
Issue 07 (2002 Apr. 29.)
Issue 06 (2001 Jul. 10.)
Issue 05 (2001 Feb. 20.)
Issue 04 (2000 Dec. 5.)
Issue 03 (2000 Aug. 31.)
Issue 02 (2000 May 15.)
Issue 01 (2000 Mar. 19.)
| 