That pretty much summarises it. An ol dprof of mine used to quote:

Make it run,
Make it right,
Make it fast,
Make it small.

Compared to C, C++ makes the first two steps much easier. For low leel optimizaions there's not much in it, but C++ gives you more time for the third step (due to the first two being easier) and also the expressiveness helps for restructuring or choosing different algorithms.

For the last step it's a wash again, but given the same amoun of time overall, C++ again wins due to makeng the earlier steps easier.

Shove DosBOX running Windows 3.11 on top of that and you're golden!

So much bullshit.

In C if you see code written as:

assign(i, multiply(j, 5));

you have equally fuck all idea what it does. In C++,

i=j*5 means this:

i.operator=(operator*(j, 5));

If you don't read i=j*5 as the above then you're incompetent at C++ and just crying because it's not identical to C.

I think the danger of that happening for low level, frequent-running, system code outweighs the flexibility that C++ gives you, and vice versa for app code.

Premaure optimization is the root of all evil.

slower executables;

Nope. std:sort is frequently faster than qsort.

less comprehensible code

Nope.

std:vector a;
a.push_back(1);

so much more comprehensible than a cluserfuck of manual memory management.

Anything you want to do in c++, you could have done in c, the primary difference being that as a c programmer, you would have -- had to have, in fact -- actually understood what it was you were doing.

Now change C++ to C and C to ASM. Now change C to ASM and ASM to machine code. All it is is a lolz newbZ suck argument and it's crap.

c++ is building kits from nasty little pre-molded pieces. c is carving your own pieces, where, if you like, every one can be a work of art and YOU control how they all fit together. You even get to make your own glue. :)

You truly do not understand C++. You can do all that with C++, but there's no need to manage a variable length array AGAIN.

Basically, C is a compiler. C++ is a programmable C compiler. I have never understood why C programmers have a phobia about programming the compiler. It's like you guys just love typing out the same stuff over and over again.

Me? I like automation.

Oh certainly.

My implementation has a matrix being a linear array of vectors.

operator[] on a matrix returns a vector object which has a pointer to the data.

operator[] on the vector indexes the pointer.

It means I can use proper C/C++ style multidimensional array like syntax (yay one of the few correct uses of the C/C++ term!), i.e.

some_matrix[row][col]

The optimizer squashes away the intermediate Vector object and the resulting code is identical to raw access of a multidimensional array, or by-hand multidimensional access of a linear array.

The actual C++ language doesn't require using accessor functions for every coefficient in a transform matrix, that's just what people do.

Indeed, I often write with structs. POD is a very useful concept in C++ and I like it.

Except for matrices where I do use accessor funcions, namely operator[], or did you mean a .get_11(), get_12(), etc...? I've certainly seen code like that before.

There is no risk angle that "drone" brings in here.

Um yeah there is. RC helicopters are fearsomely hard to fly. Drones are trivial to fly. As a result there are now many, many more drones in the air and that massiely increases the risk. They're also flown by much less experienced people which further increases the risk. AND they can be flown BVR which they can be flown into troublesome areas more easily. That increases the risk even further.

Drone is not a "scare word" but that doesn't mean that the drones do not come with increased risks.

Just like with lasers in cockpits. sure in the 80s, someone coud have bought a $15,000 lab laser and invested time to learn how to operate it and shone it at an aircraft. Funily enough no one did until cheap, integrated lasers you could buy ready to operate came available.

This does not mean technology is evil, but making potentially dangerous tech more easily available does increase the risk.

But commercial aircraft have prima donnas for jet engines.

Of course they do. the A10 engine is a rather rough, inefficient thing designed ot deal with foreign objects. A modern jet airliner has an engine with an astonishingly high pressure ratio, which requires very exotic materials and very fine tolerances. They're designed for maximum fuel efficiency, low maintainance and not shedding blades into the cabin whe na birdstrike happens.

It's the high pressure ratio that makes them prima-donna engines, but that and the low maitainance is what makes moderately cheap flights possible.

But anyway, I think it's a pertinent question to ask why "drones" attract "idiots" where the RC aircraft folk have themselves well in hand.

RC things used t obe expensive and hard to fly. As a result of an investment of time and money the operators were somewhat thoughtful, and enthusiastic about flying for its own sake. Now any yahoo can buy a quadcopter with a camera mount and go look at interesting stuff for a few hundred bucks. No knowledge required.

the risk has increased simply due to proliferation

Proliferation and ease which go hand in hand.

I had a few goes at flying an RC helicopter when they were very much part of the RC crowd and available from model shops only, not every random toy store. I also tried flying an early (and also building) an early quadcopter when they were basically the domain of university research groups and rich hobbyists.

My god those things were hard to fly. The quad copter was much easier but still very, very hard and took hours before I could hover the thing, provided it stayed within about 45 degrees of pointing forward relative to me. The helos were much, much harder due to much less linear coupling between the various controls and a substantially higher power to weight ration (early quads were quite marginal).

The reason this matters is you didn't get any random yahoo flying them. You had to be quite seriously dedicated in terms of time and reasonably in terms of money (a new helo could be had for £100 or so, but learning to fly could easily eat up that much in spare parts). Oh and be mechanically adept enough to fix the bugger when you wiped out and wrecked some important part again.

Basically this raised the barrier to entry to really dedicated people.

They generaly fall into the category of "know what they're doing" and have more of a feel for aircrft in generl due to the time spent and I would say are less likely to do something stupid. I think in part this is also because many people also joined local clubs because getting advice on these things was very useful.

Secondly, thwasn't much you could do except fly them aroud and watch. The modern style with a high res camera and remote video link is a new phenomenon and makes people willing and able to fly these things further away.

The combination of cheap, accessible to any yahoo and also interesting for more than flying to its own sake all add up.

I think you may have interpreted the AC. I *think* he was pointing out that the GP's post about rational drone operators fails to take into acconut that they are not in fact rational.

Anyway re: birds.

Geese tend to like airorts, due to the grass.

I'd hazard a guess that RC/Drone fliers ALSO like airports, probably more than geese due to the planes. I mean you have to be a bit of a plane nerd to fly those a lot and big planes are cool.

Another problem is that quads in particular have 4 large motors made out of chunks of hard metal. While birds and geese in particular (grass is tough to grind up) hold a bunch of stones in their gizzard, the drone will have a LOT of large, hard parts.

On the plus side, drones don't tend to flock.

Herons and large raptors have also been ingested into engines.

Aparently a Ruppell's Vulture wsa ingested at 37,000 feet and presumably at a cruising speed of about 600 mph. Given their weight of 8kg that must have come as an astonishing shock to the pilot.

I would also note that almost no one here is actually a scientist, much less a Nobel prize winner.

There are plenty of scientists here. And I don't need t obe a Nobel prize winner to know that for instance ESP does not exist (unlike Brian Josephsson). You don't need to have a Nobel prize to know when a prizewinner is talking crap.

Yes it is, but that still doesn't make it like china.

There are no such rules in writing.

There is one single, immutable rule which there is no way of breaking in a book worth reading: don't waste the reader's time.

The remaining rules of writing are strictly optional. However, a good first order approximation is that you aren't good enough to break them. Rare, sufficiently skilled writers can break all sorts of rules and still write a good book.

I'm not a writer (I certainly am not good enough to break the rules), but I've been learning about it for interest's sake. It gives one a whole knew appreciation of some works when the author has successfully broken the rules and written a good book.

Mostly the rules are more technical. Few people want to read massive wodges of exposition, hackneyed characters speaking in "accents" and featuring a massive mary-sue POV character.

What we need is someone who will do for SF what Tolkien did for fantasy.

Alien elf songs?

Second, most writers still use the novel format, which is around 400 years old in it's current format. This is different from older western forms, which tended to be more spoken word, such as Beowulf You can still buy 400 year old novels such Don Quixote. I would suspect that if one were doing something new, then moving from the novel format, or at least messing with it as Kurt Vonnegut did, would be the minimal requirement.

Another interesting one to read if you can is "The Life and Opinions of Tristram Shandy, Gentleman" from 1769. Or at least look at. I failed to get more than about 1/4 of the way through.

The entire thing makes a complete mockery of the concept of "postmodern" since it did that first hundreds of years earlier and only about 50 years after the novel format realy settled out in its current form (not a series of letters, short stories or poetry). It's also packed full of pop culture references (I mean really stuffed---it's impossible to read without the footnotes which explain what the hell was current in terms of slang, memes and so on circa 1769).

Despite being next to unreadable, many of the things---zanyness, kind of random humour, pop culture refrecing, bizarre, random pictures---are things many people think of as recent but aren't and feel really familiar.

For more entertainment, read the commentary about the book from when it was written. By all accounts it was as almost as unreadable then as it was then. However some people latched on to it as the height of sophistication, so people argued about whether it was good or a total crock.

Basically you could transplant the entire thing to now and it wouldn't make the slightest bit of difference.