Fear for .NET

Started by
8 comments, last by savagerx 21 years, 9 months ago
Is it true that .NET is as lame as Java
The road may be long, wind may be rough. But with a will at heart, all shall begone. ~savage chant
Advertisement
actually, it' much worse. while java now has third-generation hotspot vms that compile the bytecode to native code before/during execution, surpassing programs written in c and assembly in performance, .net can only execute one line at a time, and it is comparable to qbasic 4.5 as far as speed is concerned. plus, you can get excellent java ides for free at sun's website and elsewhere and you have to pay microsoft an outrageous price for their vs.net. avoid .net for as long as you can.

edit: for those of you who thought i was serious.

edit2: can't spell.

---
Come to #directxdev IRC channel on AfterNET


[edited by - niyaw on July 24, 2002 2:29:50 AM]
Do you normally spurt such hidiously inacurate information?

.NET languages compiles initially to IL which is then compiled to byte code either at install time or run time as the situation calls for. This has been in .NET the hole time and hasn''t taken ''three'' generations to reach acceptance.

From my own use of .NET I found it equal and frequently faster than .NET. It is only new so there aren''t as many examples for it as there are for java, the same can be said of java compared to C++ though. As for VS.NET you don''t have to buy it to get the .NET framework or compilers; you can download it or order it on CD, at least there is a standard and useful IDE for .NET that is only provided by third paties for java.

As for cross platform that is a bit of an issue but not as much as it seems at first on because of mono and similar effors, secondly the type of technology the .NET enables is very losely coupled and doesn''t require the use of it exclusivly.
quote:Original post by RegentZ
Do you normally spurt such hidiously inacurate information?

no, i thought the surpassing bit would have made it obvious. here, i'll add a for clarity.

if you insist, however...
quote:
.NET languages compiles initially to IL which is then compiled to byte code either at install time or run time as the situation calls for. This has been in .NET the hole time and hasn't taken 'three' generations to reach acceptance.

i'm aware of that.
quote:
From my own use of .NET I found it equal and frequently faster than .NET.

now i didn't expect .net to be faster itself, mind clarifying this bit?
quote:
It is only new so there aren't as many examples for it as there are for java, the same can be said of java compared to C++ though.

i'm sure msdn has quite a bit of information about .net, and i expect stuff to be available elsewhere, too.
quote:
As for VS.NET you don't have to buy it to get the .NET framework or compilers; you can download it or order it on CD, at least there is a standard and useful IDE for .NET that is only provided by third paties for java.

i was referring to the ides, runtime downloads are always free.

no comment on cross platformity, as it's not an issue to me at this moment.

edit: don't know grammar.

---
Come to #directxdev IRC channel on AfterNET

[edited by - niyaw on July 24, 2002 2:30:11 AM]
quote:Original post by niyaw
no, i thought the surpassing bit would have make it obvious. here, i''ll add a for clarity.


I seem to have slightly overlooked that comment , sorry for the unwaranted flame.

quote:Original post by niyaw
now i didn''t expect .net to be faster itself, mind clarifying this bit?


I have no quantifiable data on this, I just find that GUI .NET programs are faster to java programs of equivalent functionality, as example I notice redraw with java but not .NET. I would expect this to be the case partly due to MS being more capable/knowledgeable of optimizing for there own platform.
From a theoretical standpoint, its impossible to go faster than assembly, and even from a practical standpoint I find it to be quite an audacious claim to say that java is faster than C let alone assembly. Would you happen to have stats to back that claim up?
quote:Original post by Anonymous Poster
From a theoretical standpoint, its impossible to go faster than assembly, and even from a practical standpoint I find it to be quite an audacious claim to say that java is faster than C let alone assembly.

hehe. optimizing compilers and vms are pretty good today, and i'm willing to bet you that they can generate code surpassing poorly hand-written assembly in performance. as for java vs c, jvm can make processor-specific optimizations at runtime that c compiler cannot do. so theorerically, jvms can be faster than both c and asm code.

want benchmarks? go google.

[edited by - niyaw on July 24, 2002 3:26:19 AM]
During my graduation we abused the hotspot vm in very insane ways. For that purpose we had to switch it to debug mode. Since performance was an issue, we benchmarked both classic and hotspot vm in debug and compiled mode with some highly recursive functions. Just out of curiosity I compiled a similar program with the MS C++ compiler that comes with VC6.0 Standard (hence, no optimizations). The results were as follows:

* classic vm is slow, no matter what the mode is
* debug mode is slow on hotspot
* C++ is fast
* hotspot is faster

I never got any deeper into the matter, but you must admit that it IS surprising.

I know that benchmarking a recursive function with a highly optimized vm vs. a nonoptimizing compiler is far from optimal. And I am pretty sure that the "professional" compiler would have done better (let alone the intel compiler).

I think the only thing this benchmark actually did proof is that you can get slow software with any language if you code it badly enough.

btw no offence meant, I prefer C++ to anything else...

Tilmann
quote:Original post by savagerx
Is it true that .NET is as lame as Java


What do you mean by "as lame as Java"? That's an amazing ignorant viewpoint. Either you know next to nothing about programming and Java or else....

**************************
The best way to accelerate a Win9x machine is at 9.81m/s2

[edited by - rohde on July 24, 2002 8:16:41 AM]
"We confess our little faults to persuade people that we have no large ones." -Francois de La Rochefoucauld (1613 - 1680). | My blog
here''s food for thought that relates this topic to this site''s overall reasoning ... if Java is so fast, in fact faster than assembly (a statement that shows you''ve got ALOT of more research to do), why aren''t professional games programmed in Java?

I mean really, since Java is in fact a fairly flexible and powerful language, something I''ll certainly admit to myself, why isn''t it used for programming say... the next Xbox, PS2, or GameCube title?

I''ll tell ya why ... Java is slower than a stoned snail...
laziness is the foundation of efficiency | www.AdrianWalker.info | Adventures in Game Production | @zer0wolf - Twitter

This topic is closed to new replies.

Advertisement