ASM > everything else

Am I the only one who hates coding in high level languages? I can write all kinds of games and programs in 68k assembler, but I just choke all over C. It just makes more sense to deal with actual memory spaces than to mess around with pointers and libraries and shit. I don't like having to use any code that I didn't write myself.

Abstraction. /thread over

trying writing a full scale TURNKEY ENTERPRISE SOLUTION for a LARGE MULTI-NATIONAL  in ASM. You'll come begging for a garbage-collected object-oriented multi-paradigm language that allows one to follow ENTERPRISE BEST-PRACTICES.

>>1
Yes, you are the only one who hates coding in high level languages.

OP is under 15 and some hacker-wannabes told him that asm is faster than seeples

>>5
They were right, although not for the reasons they think

Asm is the fastest indeed, but compilation speeds are glacial, as well it requires a human compiler.

>>7
Somebody should come up with a way to computerize compilation to speed development time. Then we could let the computer handle parts that weren't speed-critical.

>>3
TURKEY ENTERPRISE SOLUTION
FTFY

asm written by humans is slower than compiler generated asm unless you are some kind of superhuman asm grandmaster (PROTIP: you are not).
the whole "write your programs in assembly to make them super fast lol XDDDDD" thing is just a myth spread by people who don't know any better

http://wisdom.sakura.ne.jp/programming/asm/index.html

>>10
That's why John the Ripper's tripcracker isn't twice as fast as the next fastest implementation. Oh wait.

>>12
i didn't say it's impossible to write faster assembly than a compiler, i just said that it's impossible for you.
lrn2read please

>>10
Some compilers are very good, but don't underestimate human compilers, they can produce fairly decent code too if they know the CPU architecture and instruction set good enough. You know that producing 'the best' optimizing compiler is at least an NP complete problem, so you can almost always find better optimizations. Of course, few compilers produce good quality code anyway, most produce just average quality code.

P.S.: I'm not advocating people code anything in ASM, except for:
1) parts of a compiler's code generation and standard library
2) Platform specific code (such as those used in OS'es, some device drivers and so on)
3) Programs which require very high performance, such as media encoders or password/crypto crackers and other similar things.
For most other applications, not using a high-level language is mostly a waste of time.

>>12
No, it's because it uses a different implementation.

i love you, /prog/ :3
you're so much fun to argue with over meaningless things

>>2
Abstraction is overrated.

>>1
You've basically learnt how to program in Basic, that's why higher level languages seem hard.

>>18
ENTERPRISE BASIC

Just to extend my post, I always find it strange that some people, typically Computer Science undergraduates, always seem to point out something like "Oh, I started programming in <insert CPU architecture> assembly." Is this supposed to impress us? You've basically been programming in a form of unstructured Basic; the most primitive type of language next to machine code.

I would be more impressed if you used and understood closures/lambda functions in a higher-level language, or used OOP to correctly model a project.

>>20
What's wrong with writing software in the processor's native tongue? Not academic enough for you?

>>21
There's nothing wrong with it, just like there's nothing wrong with building a rat bike[1] to ride to work, even when there are more suitable modes of transportation.

[1] - http://en.wikipedia.org/wiki/Rat_bike

>>21
Not academic enough for you?
What a strange thing to say. I'm trying to promote an easier way of doing things. There are easier ways to create games, for instance, than hand-coding things in assembly.

It is I who should be asking you if our practical, high-level languages "aren't academic enough for you."

>>22

I program in assembly, and now that I've read that wikipedia article, I like the aesthetic of rat bikes as well. Different strokes for different folks, I guess.

>>23
High level languages require an enormous amount of infrastructure to work effectively; think GCC/MSVC. Assembly, on the other hand, requires only an assembler, which can be made in a few kilobytes of source in a few days' time, and no greater toolchain is required. Depending on the situation, assembly is the easier way of doing things. Plus, it's just nice to know you're actually working with the hardware, telling it what to do and such; high level languages just feel too sterile to me.

>>25
I don't feel the same.
I've known ASM way before I knew high level languages.
I find its value for doing certain things, which just can't be done without access to the hardware, but for the rest, it would be just a huge waste of time and it would mean disregarding the progress made in Computer Science in the past 50 years or so. A lot of high-level languages either do have inline assembler, or allow linking of calling machine code directly, or can be made to allow direct access to the hardware by tinkering with the actual implementation, which means I can use ASM anytime I really need it, and frankly, it's very often that I need to use it for normal coding tasks.

As for the toolchain "problem"? You need an OS to operate a computer. You need an assembler to assemble some code (unless you want to use the pen and paper aproach to writing and assembling code by hand). Why is it a problem to use a compiler or an implementation/runtime for a high-level language?

s/very often/very rare

HOLY SHIT ARE YOU SERIOUS?  This is the most obvious troll and still got 30 replies.  Fucking students flooding /prog/.

>>26
I guess it's just the minimalist aesthetic of it; for an operating system, I loved DOS, and for a programming language, I love assembly. Not too fond of computer science, I just like programming really.

>>25
One could argue that assembly language is simply high level microcode, and you aren't really any closer to hardware than you are by programming in, say, C. If the distance between the earth and the sun represents how close you are to hardware from C, then the distance between the moon and the sun would be assembly. A rather fitting analogy too, since programming in higher level languages is more down to earth than you ASM astronauts :-)

>>29
You might like to try Verilong or VHDL then, as >>30 said.
However, I bet that is too high-level for you, so you might end up just using schematics in the end, or even doing full-custom designs (instead of piecing components together, you would design the components directly (at the transistor level)).

*Verilog

>>30
Haha, no. Assembly is the closest that software can get to the processor. What's within the processor is outside of the realm of software; you can't exactly reprogram the microcode of your machine--well I guess you could, there have been microcode patches, but even then the microcode ROMs are hardly enough space to implement anything useful, and doing so would disallow any other software to run besides your customized ISA. This is completely negating the fact that some architectures don't even have microcode. Basically, software-wise, assembly is the lowest practical level, other than drawing up some schematics for your own processor design (which is fun too, but not within the scope of programming).

>>33
Unless you program kernels and drivers for a living, then there is no reason to use Assembly. "Close to hardware" is a myth perpetuated to make proponents of assembly language feel good about themselves; you can access hardware just as easily in C, via system calls provided by the kernel and memory mapped I/O.

I submit to you that assembly is nothing more than the illusion of being close to hardware. It's a language which gets transformed into a set of numbers in a specific way; nothing more. C is a more expressive and equally powerful language which also gets transformed into a set of numbers.

What's a good 16-bit debugger/emulator for Windows?  I use emu8086 but it expired and I can't find a serial for the newest version so I am fucked unless I want to risk downloading some malware by trying to get a cracked, older version of the program.

>>35
implies to have some knowledge of x86 assembler
but, can't crack software or do a little bit of malware analisys

>>36
implies to have some knowledge of ^B_BCode
but, can't multi-line quote

>>34
Close to the hardware is the truth, though; when programming in assembly, you know everything about your program--how large it is, how many cycles certain parts will execute in, which parts of the CPU are in use, and when exactly everything happens. With C you lose all of this. Most importantly though, programming in assembly is more fun.

>>38
YHBT

>>39
IDTIHBT
HYBT?