NCG_Mike

Right On Commander!

I assume they had a scaled up integer sin and tan tables and used matrices for scale, rotation and projection?

I do remember seeing this in 68K.

I'd suggest that someone who thinks that UTs are over-rated are over-rating their ability. I'd also suggest that those people are the exact people that should have extra testing done on their code.

Good. The NHS has a terrible history of software development.

I had it too ;-)

A couple of things a C++ developer can do is to put template instantiation code into a .cpp file, where possible.

"#pragma once" in the header files helps as does using a pre-compiled header file.

Obviously, removing header files that aren't needed makes a difference too.

Ta.

I remember Dr Dobbs, way back when, having an article on the search algorithm. I think they had C code for it.

I did a version in 68K on the Mac (MPW Shell days) for a SGML editor I was working on.

If you hit shift+run stop at the end of the command it'd load and run.

Wow. Going back a long time. I was in the MPW Shell camp but did use ThinkC With Objects and it was so much faster than the C compiler in MPW.

Requires some kind of sign-in to read the article. Not going to happen.

The way I built MacOS applications in Obj-C was to use a dictionary with key/value pairs held inside either a NSDocument or a controller of some kind. I'd bind the views to the model, via the controller. No sub-classes of views.

It also had the benefit of been able to add undo/redo with a few lines of code. This even would undo radio button clicks as those controls were also bound to the model.

Saving the model was just a case of saving the dictionary or converting the dictionary in save and load from some other format, say ICC Color Profiles.

I'm not surprised to read that less experienced developers bung all their controller and model code in a sub-class of some view. It was a common thing to see in the MFC or PowerPlant days.

I know them as "roger beeps" from CB radio terminology.

Smog pollution was a big issue back then. You'll have seen the old movies where London is "foggy".

Not sure but there may be two bugs in the code. The use of notify on the condition outside of the mutex lock.

I know this would cause problems with boost, not sure about std::.