Forum Replies Created
-
Posts
-
Earlier this year I signed up for a membership at ACCU, which includes subscriptions to two “good old” printed magazines: https://accu.org/menu-overviews/journals-overview/
I don’t believe there is any performance or resource aspect to the preference of bool b{true}. It is however less error-prone since it for example does not allow narrowing conversions. In case you replace true in the above example with a value 2, the second initialisation will just work, while the initialisation with curly braces will give a compiler error
I am working as a software development contractor in Automotive Industry, currently at an OEM and before this at a supplier. I mostly work with features related to Autonomous Driving as well as Advanced Driver Assistance, for example automatic braking functionality based on the identification of threats by radar, ultrasonic and/or cameras. Even though I have been working in this area for more than 20 years, I must admit that during a large part of it, I was working entirely model-based. Meaning: the functionality was developed in Matlab/Simulink after which C-code is autogenerated and often by others integrated in a certain target platform. Even though one had to inspect the generated c-code every now and then for debugging, I did not write any code from scratch myself. Even though this approach has it advantages, the abstraction often comes with a cost in terms of performance or resource consumption in general. Also, I personally disliked that a lot was happening “under the hood” for which there was very little understanding in the development teams. Since a bit more than 2 years I am more and more involved in projects/assignments where C++ is the main language, partially because more companies are trending away from this purely model-based approach but also because I am getting more interested and involved in the so-called middle-ware layer, connecting features with each other as well as with hardware in performant ways.
