Gamestudio Links
Zorro Links
Newest Posts
Executing Trades on Next Bar Open
by Zheka. 06/20/24 14:26
Lapsa's very own thread
by rki. 06/19/24 11:27
A simple game ...
by VoroneTZ. 06/18/24 10:50
Face player all the time ...
by bbn1982. 06/18/24 10:25
Zorro Beta 2.61: PyTorch
by jcl. 06/10/24 14:42
New FXCM FIX Plugin
by flink. 06/04/24 07:30
AUM Magazine
Latest Screens
The Bible Game
A psychological thriller game
SHADOW (2014)
DEAD TASTE
Who's Online Now
1 registered members (AndrewAMD), 1,320 guests, and 1 spider.
Key: Admin, Global Mod, Mod
Newest Members
Mino, squik, AemStones, LucasJoshua, Baklazhan
19061 Registered Users
Previous Thread
Next Thread
Print Thread
Rate Thread
Page 1 of 2 1 2
Performance benchmark of Lite-C #398881
04/08/12 13:17
04/08/12 13:17
Joined: Mar 2012
Posts: 43
J
Juggernaut Offline OP
Newbie
Juggernaut  Offline OP
Newbie
J

Joined: Mar 2012
Posts: 43
Hello,

Is the executable produced by the Lite-C code as fast as C/C++ code ? If not, is there any performance comparison available ?

Thanks,

Re: Performance benchmark of Lite-C [Re: Juggernaut] #398935
04/08/12 23:48
04/08/12 23:48
Joined: Jun 2004
Posts: 2,234
Wisconsin USA
FoxHound Offline
Expert
FoxHound  Offline
Expert

Joined: Jun 2004
Posts: 2,234
Wisconsin USA
Lite C = "mom's car"

C++ = Mustang GT500


---------------------
There is no signature here.


QUIT LOOKING FOR ONE!
Re: Performance benchmark of Lite-C [Re: FoxHound] #398949
04/09/12 09:20
04/09/12 09:20
Joined: Mar 2012
Posts: 43
J
Juggernaut Offline OP
Newbie
Juggernaut  Offline OP
Newbie
J

Joined: Mar 2012
Posts: 43
Hello Foxhound,

Thank you for your reply. So what do you suggest for commercial grade games like the "Supercan" - use the engine from C++ or use the engine using Lite-C to create the game ?

Thanks,

Re: Performance benchmark of Lite-C [Re: Juggernaut] #398958
04/09/12 11:33
04/09/12 11:33
Joined: Dec 2003
Posts: 1,225
germany
gri Offline
Serious User
gri  Offline
Serious User

Joined: Dec 2003
Posts: 1,225
germany

C++ doesnt speed up a game. Huge texturesizes and heavy use of shaders can drop your fps to an slideshow even if the game is coded in Assembler.


"Make a great game or kill it early" (Bruce Shelley, Ensemble Studios)
Re: Performance benchmark of Lite-C [Re: gri] #398961
04/09/12 11:40
04/09/12 11:40
Joined: Feb 2009
Posts: 3,207
Germany, Magdeburg
Rei_Ayanami Offline
Expert
Rei_Ayanami  Offline
Expert

Joined: Feb 2009
Posts: 3,207
Germany, Magdeburg
>>implying anyone could write better optimized assembler code than the compiler

It does not matter if you use C++ or LiteC if you are not able to code effective source code. You can ultra-slow C++ code, and really fast LiteC code if you do a lot of mistakes.

However - take this tip: "Do not optimize (yet)" wink

Re: Performance benchmark of Lite-C [Re: Juggernaut] #398962
04/09/12 11:45
04/09/12 11:45
Joined: Jul 2001
Posts: 6,904
H
HeelX Offline
Senior Expert
HeelX  Offline
Senior Expert
H

Joined: Jul 2001
Posts: 6,904
Don't take it personal FoxHound, but comparing two language specifications by speed is so senseless. Saying "just because you use C++, your game is 200% faster", is severely stupid. It depends on the compiler.

And yes, the Lite-C compiler is purely functional - it is not speed optimized. So, if you take one Lite-C function and throw it into a DLL compiled in Visual Studio against the Gamestudio SDK, it will execute most certainly faster.

But imho, most speedups can be easily achieved by identifying general bottlenecks in your game architecture and choosing of datastructures and general algorithms - and are not so related to micro-optimizations performed by the compiler. Plus, if you multithread your game, you can get also a significant boost.

If you are seeking for code-related speed, my advise would be to write your game in Lite-C using the wait(1)-paradigm, using a plugin DLL that utilizes C++ code. Multithread, where possible and use e.g. the standard libraries or Boost or so, if it comes to standardized algorithms and datastructures.

Re: Performance benchmark of Lite-C [Re: HeelX] #398989
04/09/12 21:59
04/09/12 21:59
Joined: Jun 2004
Posts: 2,234
Wisconsin USA
FoxHound Offline
Expert
FoxHound  Offline
Expert

Joined: Jun 2004
Posts: 2,234
Wisconsin USA
Before I respond back can you make sure you are referencing thing I actually said? This thread was about which was faster not which was better or easier to use or a personal choice. So a car anology works fine when referencing speed.


---------------------
There is no signature here.


QUIT LOOKING FOR ONE!
Re: Performance benchmark of Lite-C [Re: FoxHound] #398990
04/09/12 22:09
04/09/12 22:09
Joined: Jul 2001
Posts: 6,904
H
HeelX Offline
Senior Expert
HeelX  Offline
Senior Expert
H

Joined: Jul 2001
Posts: 6,904
If you take the whole picture into account and everyone agrees on the boundary conditions, then yes, a car anology could work it out.

Re: Performance benchmark of Lite-C [Re: Juggernaut] #399004
04/10/12 07:41
04/10/12 07:41
Joined: Mar 2012
Posts: 927
cyberspace
W
Wjbender Offline
User
Wjbender  Offline
User
W

Joined: Mar 2012
Posts: 927
cyberspace
Originally Posted By: Juggernaut
Hello,

Is the executable produced by the Lite-C code as fast as C/C++ code ? If not, is there any performance comparison available ?

Thanks,


well.... c++ has so many compiler options that a executable can be produced
that's faster yes but i do not think that gamestudio compiler will ever give
you that much compiler options ,so looking at it that way i would say yes
c++ rules just because you were talking about "executable produced" and in
my opinion that's a compiler's job we are comparing here.. now for every thing
else speed wise im telling you now mate no ones gonna agree on anything here
,this is much more of a personal preference battle than anything else.

the only way out here would be if you made a test for both ...
then simply choose for yourself!

dont wanna make a fool of myself but gamestudio had to be coded by a language
i would IMAGINE that language was/is c++ ,now just remember c++ is a proffesional
language ! but ,ease of use? thats why we use lite-c ,people dont want to code
a game in lower level language the whole idea is to stick to the higher lever
of coding when it comes to games .. having a comparison that leads one to
start coding in c++ or lite-c for reasons of speed alone i would think
may only complicate things more than it needs to be ,the languages
differ so much in what can be done with each that it remains a personal preference
and seeing that c++ can be used to achieve far more complicated stuff(language wise)
those more complicated stuff may very well be the downfall of a coder when starting
to experience complications in how to use the engine functions properly to
achieve what coder wanted to achieve,in MY opinion simply because of
EASE OF USE i will stick to lite-c (even though i am a c++ guy) what
i experienced in using this engine in c++ is just plain NIGHTMARISH
it wasnt tailored for c++ its HORRIBLE in c++(not c++ itself but the engine coding in c++)


Last edited by Wjbender; 04/10/12 08:00.

Compulsive compiler
Re: Performance benchmark of Lite-C [Re: Wjbender] #399018
04/10/12 11:13
04/10/12 11:13
Joined: Mar 2012
Posts: 43
J
Juggernaut Offline OP
Newbie
Juggernaut  Offline OP
Newbie
J

Joined: Mar 2012
Posts: 43
Thank you guys for giving your viewpoints. I guess, I will stick with Lite-C for the time being while I am learning the ins and outs of the 3D Game Studio A8 Engine. After I have mastered all the functions of the engine - then will be the right time to use the engine from C++ ..... laugh

Thank you ..... all of you ..... for giving both time and effort to my thread.

Thanks,

Page 1 of 2 1 2

Moderated by  HeelX, Lukas, rayp, Rei_Ayanami, Superku, Tobias, TWO, VeT 

Gamestudio download | chip programmers | Zorro platform | shop | Data Protection Policy

oP group Germany GmbH | Birkenstr. 25-27 | 63549 Ronneburg / Germany | info (at) opgroup.de

Powered by UBB.threads™ PHP Forum Software 7.7.1