19
The dependence of David’s software on ALSA was a revelation to us as was the
unavailability of a development environment on the PCM-5896. We had wrongly
assumed that all Linux distributions ( or distros as they are affectionately called ) had the
make and compilation utility. When this view was found to be incomplete and incorrect,
we were left facing a crisis.
7.2 Hardware problems.
The PCM-5896 uses an ATX power supply. A lack of familiarity of this type of hardware
led us to damage our initial board because we were not aware of the fact that there was
always a 5V source present in this type of power supply even when the contact was
broken using a soft switch connected to a jumper. This most probably led us to our
damaging the board without even knowing it because we were supposed to have cut off
the power from the wall supply while working on it. As we found out belatedly, this
ignorance was fatal to our application as it set us back at least a week in having to return
the hardware to the vendor. The vendor was kind enough to send us another board with
an identical configuration. If it hadn’t for EMAC’s willingness to replace the hardware, it
would have been impossible for us to achieve what we have achieved today. However,
this may have been a little bit of a blessing in disguise as it encouraged us to be extra
careful during the latter and most crucial part of our project.
7.3 Lack of familiarity with the Software and substitutes
The software application that we were to use, ClVoip, was developed at GeorgiaTech by
students of an earlier batch of this same. This included David Jeffery, our Lab Teaching
Assistant. Because of the presence of this obvious support, we were hesitant to examine
other software that could have suited our application and our hardware platform better.
We were afraid of the unknown and that fear ended up hurting us temporarily.
Fortunately, we were able to rebound with a better choice in software.
7.4 Our behavior in times of crisis
Although we were beset by so many problems from so many angles, we did not
relinquish our goal. Instead we chose to think and act on our feet. When the solution
based on ALSA did not work, we chose another one. This also had the potential to fail as
it was distributed as source. Given the lack of a compile environment on our Board, it
was a challenge to get it working. We came up with a way to build the executables and
the needed libraries on our development system and then transfer them and their
directory structures in their entirety to the PCM-5896. Although this approach had failed
for the ALSA libraries, we chose to look at the new situation on its own merits. We knew
that the new software had limited tie-ins with the kernel and would be less inclined to be
affected by the difference in the Linux distributions in the development (compiled) and
the application environment. We also took a guess that the new software (Simph323) did
not depend on ALSA as such a dependence was not mentioned anywhere in the
documentation. We also encountered other obstacles such as the failure to record sound
and the wavelan interface not coming up. We refused to be discouraged. Instead we chose
to expand our limited knowledge of these technologies by trying different approaches to
(3 Seiten)
Manymanuals.com
Manymanuals.de
Manymanuals.fr
Manymanuals.it
Manymanuals.pl
Manymanuals.cz
Manymanuals.es
Manymanuals-pt.com
Kommentare zu diesen Handbüchern