Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[KrunkSplein]

[edit - removed for simplicities sake]

[MaDeR Levap]

The whole reason this argument started is because someone had to say that you shouldn't use Java because it is slow.

Erm, no. Java is slow. Queston about using Java in development of game like DF is different matter. Slowness of Java can be considered a "con" against this idea, but it is entirely possible that exists various "pro" arguments (like speed of development or whatever) that can be used to prove that it is good idea to use Java to make such game.

I did not analyze it; my comment was a sideline friendly jab against language that I use in daily work. It was not intended to taken very seriously.

[devek]

There were computer programmers long before there were functional computers(seriously), and when they wrote to each other they were dicks.

There is a reason for it, but an outsider would never understand.

That reminds me of this... for some reason too. http://www.openbsd.org/lyrics.html#43

Nothing like developers writing songs just to talk shit on other developers, we're cool like that.

[ZCM]

because doing so builds a false dependency on EBX.

Register renaming.

[KrunkSplein]

There is a reason for it, but an outsider would never understand.

Thank you, Devek, for totally validating my point.  How dare I, your assumed lesser, attempt to fathom the intangible depths of the programmer's brain.

Oh wait, that's right!  I have my masters in computer engineering and do development for a multinational technology company.  How'd that slip my mind?

I'm just tired of a few loud assholes making all of us look bad.

[Thief^]

Actually this was a fairly clean and tidy argument as far as they go. No name calling or ad hominem or that lot.
I don't see the problem. Discussion of alternate points of view is one of the best ways to learn.

EDIT: Wait, I'm now arguing about arguing. Dear god.

[Veroule]

Its all fun and games until someone NULL's a valid pointer.
Spacing only matters as far as what doesn't hurt my eyes.  I find huge indentations painful, and { on its own line ugly.  More parenthesis are always good good in my book.  It is better to type a few extra things then spend an hour debugging because you got an order of operations wrong.

Code: [Select]

if (test() && ((a+b)/d>c)) {
 code;
 //egads! 1 liner shouldn't have braces and is own line.  I would only put it here for comment space.
 //horizontal scrolling is bad, why the frak haven't they made a mouse with a scroll ball yet.
}6 spaces, counting the cr/lf as spaces. 41 spaces if you count the ranting comments, and they are often present in my private code.
The whole debate over break, continue, if..else, and goto always makes me laugh.  The first time I heard someone say goto was totally unnecessary was when Pascal had a solid footing and C was gaining steam.  Some young kid thought he knew what he was talking about because he knew both languages.  I shook my head and laughed since I knew they all got compiled to a jump/branch class of instruction.  Then I learned Pascal; and thought well it is nice that the compiler handles most of it, and there aren't labels all over the place.  Shortly followed by screaming, "There is no instruction to pop out of a loop. I have to set the control variable, WTF!"
I am not a huge fan of "while(1)", and prefer to use "for(;".  I have seen compilers reserve a register and test it when using the while.  Probably another of those ancient things that finally got fixed.

Conversely I don't much like using "for" at any other time, and prefer to write out the loop parts.  This is because "for" does its limit condition at the start; and just a little thought can determine if the loop will always run at least one pass. If it will make at least one pass then it makes more sense to use "do {} while;"  It isn't just the 3 clock cycle gain, it is a logic hint for me when I have to look at that code again 5 years from now.
I have yet to see function, loop, or even memory alignment make a significant speed difference.  If any one can point me to benchmarks demonstrating measured gains from alignment with the assembly being the exact same between the aligned and nonaligned tests; I would appreciate it.

[KrunkSplein]

Okay.  Now that I've had a nap, I see that I may have gotten slightly hyperbolic.  I was reacting more to implied tone than any specific word choices.  Some examples:

My opinion: you know little about performance, assembly, or java.

You're just making excuses the fact that you find assembly hard.

There is a reason for it, but an outsider would never understand.

The tone of those is unmistakable arrogance, for which I have an extremely low tolerance.  Nonetheless, I jumped the gun, and for that I apologize for sounding the alarm.

Wait, I'm now arguing about arguing. Dear god.

We've gone... META!

[Normandy]

Actually, I have to take that last part back...

There is some debate about that regarding memory usage... The java and c++ argue about which language fragments the heap worse Both problems that can be avoided with proper programing of both languages, so its moot in my boot.

I love this statement.


I must say, this conversation has gone quite over my head. I don't suppose growing up during the era of Moore's Law has helped. But I'm kind of curious, this is all just experience on your part? I'm kind of curious where one picks up all this.

[Setharnas]

I have yet to see function, loop, or even memory alignment make a significant speed difference.  If any one can point me to benchmarks demonstrating measured gains from alignment with the assembly being the exact same between the aligned and nonaligned tests; I would appreciate it.

Perhaps not 100% what you were looking for, but I found this nice.


And regardless of tone, I always find this kind of discussion worthy of /popcorn. There's bound to be the odd tidbit of useful information being thrown around as disputees do the "look, my nerd-fu is longer than yours, so nya" routine, so I consider it a modern form of educational entertainment. 

[devek]

I must say, this conversation has gone quite over my head. I don't suppose growing up during the era of Moore's Law has helped. But I'm kind of curious, this is all just experience on your part? I'm kind of curious where one picks up all this.

You pretty much hit the nail on the head.

I don't really know the right answer to the problem. It really all boils down to your curiosity and your ability to find out information on your own. My assignment to you is this.

1) Write a functional ELF or PE program using nothing but a hex editor.
2) Write a debugger to single step the instructions in that program.
3) Do it without any help.

Total waste of a time, but at the end of the day(week, months, years?) your journey through this will teach you more about how a computer works than years of reading about them.

[ZCM]

I must say, this conversation has gone quite over my head. I don't suppose growing up during the era of Moore's Law has helped. But I'm kind of curious, this is all just experience on your part? I'm kind of curious where one picks up all this.

You pretty much hit the nail on the head.

I don't really know the right answer to the problem. It really all boils down to your curiosity and your ability to find out information on your own. My assignment to you is this.

1) Write a functional ELF or PE program using nothing but a hex editor.
2) Write a debugger to single step the instructions in that program.
3) Do it without any help.

Total waste of a time, but at the end of the day(week, months, years?) your journey through this will teach you more about how a computer works than years of reading about them.

I out-nerd you: I've designed and built my own CPU, and entered microcode using dip switches.

[devek]

Ha! Electrical processing is for wimps.

[Thief^]

I've been trying to design a TTL chip cpu for a while, but I keep getting distracted by software stuff. It was going to be a one-instruction design, because it massively simplifies the instruction decoder. Needs a lot of registers though...

[Veroule]

I have yet to see function, loop, or even memory alignment make a significant speed difference.  If any one can point me to benchmarks demonstrating measured gains from alignment with the assembly being the exact same between the aligned and nonaligned tests; I would appreciate it.

Perhaps not 100% what you were looking for, but I found this nice.

Not even close to what I was looking for, and I would give it a relavance of about 5%. It was a VERY interesting and enlightening read, and  is a a subject that anyone concerned with optomizations should look at.