[Mono-list] Array benchmark
On a game development forum, someone asked what method of storing 2D arrays is faster. On .NET, there are three possibilities: 1. [y*width+x] 2. [x,y] 3. [x][y] I was wondering too, so I wrote a benchmarker. I attached the source and the results (as pdf). The results are remarkable! With all methods with all resolutions, Mono -is- slowest. Until.. maximum optimalization is used. Then the Mono results are about 5 times as good as with Microsoft.NET. Another weird thing is that normally, method 1 is way faster, but when Mono is used with maximum optimalization, method 2 is faster. And when the version with a resolution of a power of two is ran, both method 2 and 3 are faster. Just take a look at the pdf :) Could someone explain this odd behaviour? And why the speed difference between non-optimized and optimized mono is so big? And why such optimalization is not used by default? -- 'May the Lord bless you and protect you. May the Lord smile on you and be gracious to you. May the Lord show you his favor and give you his peace.' 'De Heer zegent je, en Hij bewaart je. De Heer kijkt met liefde naar je, en Hij is je genadig. De Heer bedenkt het goede voor je, en geeft je vrede.' Sijmen Mulder ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
Re: [Mono-list] Array benchmark
On Sat, Oct 09, 2004 at 11:29:21AM +0200, Sijmen Mulder wrote: On a game development forum, someone asked what method of storing 2D arrays is faster. On .NET, there are three possibilities: 1. [y*width+x] 2. [x,y] 3. [x][y] I was wondering too, so I wrote a benchmarker. I attached the source and the results (as pdf). The results are remarkable! Forgot the attachment? With all methods with all resolutions, Mono -is- slowest. Until.. maximum optimalization is used. Then the Mono results are about 5 times as good as with Microsoft.NET. Another weird thing is that normally, method 1 is way faster, but when Mono is used with maximum optimalization, method 2 is faster. And when the version with a resolution of a power of two is ran, both method 2 and 3 are faster. Just take a look at the pdf :) Even without the attachment -- array bounds checks. In the first case you need one check and in the second case you need two. But when array bounds checks are removed (with -O=all), the second version can use some internal magic to get it done faster. Bear in mind that it may not be possible to always remove array bounds checks. -- : Michal Moskal :: http://www.kernel.pl/~malekith :: GCS !tv h e+++ b++ : ::: Logic is a nice contrast to the Real World. :: UL$ C++ E--- a? ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
Re: [Mono-list] Array benchmark
Forgot the attachment?
Here they are. The exported txt file doesn't look to good, though, but
it'll be clear.
--
'May the Lord bless you and protect you. May the Lord smile on you
and be gracious to you. May the Lord show you his favor and give you
his peace.'
'De Heer zegent je, en Hij bewaart je. De Heer kijkt met liefde naar
je, en Hij is je genadig. De Heer bedenkt het goede voor je, en geeft
je vrede.'
Sijmen Mulder
using System;
using Microsoft.Win32;
namespace ArrayBench
{
delegate float TestMethod();
class Program1000
{
const int size = 1000;
long endTime;
TestMethod[] tests;
void start()
{
endTime = DateTime.Now.Ticks+1;
}
bool stop()
{
return DateTime.Now.Ticks=endTime;
}
float linear()
{
int[] array = new int[size*size];
int count = 0;
int temp;
start();
while(!stop())
{
count++;
for(int y=0; ysize; y++)
for(int x=0; xsize; x++)
temp = array[y*size+x];
}
return count;
}
float linmatrix()
{
int[,] array = new int[size,size];
intcount = 0;
inttemp;
start();
while(!stop())
{
count++;
for(int y=0; ysize; y++)
for(int x=0; xsize; x++)
temp = array[x,y];
}
return count;
}
float matrix()
{
int[][] array = new int[size][];
int count = 0;
int temp;
for(int i=0; isize; i++)
array[i] = new int[size];
start();
while(!stop())
{
count++;
for(int y=0; ysize; y++)
for(int x=0; xsize; x++)
temp = array[x][y];
}
return count;
}
public Program1000()
{
tests = new TestMethod[]
{
new TestMethod(linear),
new TestMethod(linmatrix),
new TestMethod(matrix)
};
}
public void Run()
{
RegistryKey key = Registry.LocalMachine.OpenSubKey(
HARDWARE\\DESCRIPTION\\System\\CentralProcessor\\0);
string name = (string)key.GetValue(Identifier);
intspeed = (int)key.GetValue(~MHz);
Console.WriteLine({0} at {1} MHz, name, speed);
Console.WriteLine();
Console.WriteLine(Testing {0}*{0}, size);
foreach(TestMethod test in tests)
{
Console.Write(Running {0}: , test.Method.Name);
Console.WriteLine({0}, test());
}
Console.WriteLine();
}
}
class Program1024
{
const int size = 1024;
long endTime;
TestMethod[] tests;
void start()
{
endTime = DateTime.Now.Ticks+1;
}
bool stop()
{
return DateTime.Now.Ticks=endTime;
}
float linear()
{
int[] array = new int[size*size];
int count = 0;
int temp;
start();
while(!stop())
{
count++;
for(int
Re: [Mono-list] Array benchmark
OFF-TOPIC -- Sijmen Mulder wrote: On a game development forum, someone asked what method of storing 2D arrays is faster. On .NET, there are three possibilities: 1. [y*width+x] 2. [x,y] 3. [x][y] 4. [offset + x] offset = columOffset[y] cellContent = map[ offset + x ] will be need to populate columoffset with values: columOffset[i]= (i - 1) * width I cheat because I dont use a 2D array. I know is Ok to cheat coding videogames, anyway. You dont really need to draw 2 zimbillions of trigs If you can cheat and draw only 200 and gamers will not notice. I was wondering too, so I wrote a benchmarker. cool. I attached the source and the results (as pdf). The results are remarkable! where? maybe this mail list remove attchs. With all methods with all resolutions, Mono -is- slowest. Until.. maximum optimalization is used. Then the Mono results are about 5 times as good as with Microsoft.NET. Another weird thing is that normally, method 1 is way faster, but when Mono is used with maximum optimalization, method 2 is faster. And when the version with a resolution of a power of two is ran, both method 2 and 3 are faster. Just take a look at the pdf :) Looks like you may need to force il code and compare what code is generated. You are optimizing very tigth loops, so its critical the compiler optimizations. Its Mono cool enough to temporaly disable optimizations??? #pragma disable inline expansion code that work better with inline disabled #pragma enable inlince expansion Even gamercoders get more speed switching to better algoritms. You can't optimize bubblesort beyond some limit. Better algoritms exist out here for most stuff... Compiler oriented optimizations may result slow with a different compiler, version, client hardware, etc.. --Tei ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
RE: [Mono-list] VB? Or just C#?
Donahue It would help us a lot if you could send across the error page returned by the server or better still your aspx pages. So I can only make a guess what the problem could be. Error message: (0,0) : error : THIS IS STILL ALPHA AND UNSUPPORTED SOFTWARE, USE AT YOUR OWN RISK. VBCodeCompiler used to consider the above ALPHA notice as an error. This was fixed in VBCodeCompiler.cs (part of System.dll) sometime back. To rule out this spurious error you can either 1) compile System.dll from cvs HEAD and install the same or 2) Upgrade to the latest mono 1.0.1 packages. Regards, Jambunathan K. David P. Donahue [EMAIL PROTECTED] 10/09/04 6:21 AM I managed to make the same small test page on a machine running Visual Studio .NET 2003, just to test the updated Microsoft.VisualBasic.dll and see if that solved the problem. Now when I try to browse to the page (which is just a button and a label) I get the following error page: Server Error in '/gwcc/test' Application Compilation Error Description: Error compiling a resource required to service this request. Review your source file and modify it to fix this error. Error message: (0,0) : error : THIS IS STILL ALPHA AND UNSUPPORTED SOFTWARE, USE AT YOUR OWN RISK. File name: /home/www/.www/gwcc/test/Global.asax Source File: Line 1: No assembly returned after compilation!? Also, I don't know if I mentioned this before or if it makes a difference, but I'm running mod_mono in Apache 1.3.31 on Slackware 10.0. What could be causing this? Regards, David P. Donahue [EMAIL PROTECTED] -Original Message- From: B Anirban [mailto:[EMAIL PROTECTED] Sent: Monday, October 04, 2004 5:25 AM To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: [Mono-list] VB? Or just C#? Hi David, This problem is occurring since you have an older version of Microsoft.VisualBasic dll. And mono is not yet equipped for remapping this assembly to the newer version automatically. Please report bug in bugzilla (bugzilla.ximian.com) on this against mbas. Thanks. Anirban. David P. Donahue [EMAIL PROTECTED] 04-Oct-04 12:45:52 AM I just wrote a very, very simple ASP .NET project in using VB .NET in Visual Studio .NET and tried to run it on mono. The project is just one page with a button that, when pressed, changes a label. I created the page on the localhost on my Windows XP machine, build the solution in Visual Studio, then copied all the files to a folder shared by my Apache server (running mono) on my Linux box. When I attempt to access the page, I get a stack trace that ends with a file not found error for the .dll (even though it is there) and see the following in my Apache error log: ** (/usr/local/bin/mod-mono-server.exe:10038): WARNING **: Could not find assembly Microsoft.VisualBasic, references from /home/www/.www/gwcc/test/bin/gwcc.dll (assemblyref_index=1) Major/Minor: 7,0 Build: 3300,0 Token: b03f5f7f11d50a3a Should this be happening, or am I doing something wrong? Regards, David P. Donahue [EMAIL PROTECTED] -Original Message- From: Juan Cristóbal Olivares [mailto:[EMAIL PROTECTED] Sent: Sunday, September 12, 2004 4:16 PM To: David P. Donahue; [EMAIL PROTECTED] Subject: Re: [Mono-list] VB? Or just C#? Mono does not have a VB.NET compiller. You could compile your VB.NET site with .NET (Windows) and the run it with Mono. Juan Cristóbal Olivares - Original Message - From: David P. Donahue [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Saturday, September 11, 2004 7:49 PM Subject: [Mono-list] VB? Or just C#? Can Mono compile ASP .NET websites written in VB, or is it limited to only C#? If it can, I'm having trouble finding documentation on how... David P. Donahue ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
RE: [Mono-list] How many strings do I have?
On Fri, 2004-10-08 at 04:39, Jeroen Frijters wrote: Jonathan Gilbert wrote: Once a string is in the pool, it can NEVER be removed (until the AppDomain is disposed). In Java the string intern pool uses weak references. I don't know about .NET, but I'd hope it does the same. I believe it doesn't. From [1]: Interning happens two ways in the CLR. ... 2) It happens automatically, when you load an assembly. All the string literals in the assembly are interned. This is expensive and in retrospect may have been a mistake. In the future we might consider allowing individual assemblies to opt-in or opt-out - Jon [1] http://blogs.msdn.com/cbrumme/archive/2003/04/22/51371.aspx ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
RE: [Mono-list] VB? Or just C#?
It would help us a lot if you could send across the error page returned by the server or better still your aspx pages. The error page returned reads as follows: -- Server Error in '/gwcc/test' Application Compilation Error Description: Error compiling a resource required to service this request. Review your source file and modify it to fix this error. Error message: (0,0) : error : THIS IS STILL ALPHA AND UNSUPPORTED SOFTWARE, USE AT YOUR OWN RISK. File name: /home/www/.www/gwcc/test/Global.asax Source File: Line 1: No assembly returned after compilation!? -- So I can only make a guess what the problem could be. Error message: (0,0) : error : THIS IS STILL ALPHA AND UNSUPPORTED SOFTWARE, USE AT YOUR OWN RISK. VBCodeCompiler used to consider the above ALPHA notice as an error. This was fixed in VBCodeCompiler.cs (part of System.dll) sometime back. To rule out this spurious error you can either 1) compile System.dll from cvs HEAD and install the same or 2) Upgrade to the latest mono 1.0.1 packages. I was using mono 1.0.1, sorry I didn't specify that earlier. I just installed mono 1.0.2 but am having the same problem. Should I push into 1.1.1 (development version) or is that not recommended? Regards, David P. Donahue [EMAIL PROTECTED] Regards, Jambunathan K. David P. Donahue [EMAIL PROTECTED] 10/09/04 6:21 AM I managed to make the same small test page on a machine running Visual Studio .NET 2003, just to test the updated Microsoft.VisualBasic.dll and see if that solved the problem. Now when I try to browse to the page (which is just a button and a label) I get the following error page: -- -- Server Error in '/gwcc/test' Application Compilation Error Description: Error compiling a resource required to service this request. Review your source file and modify it to fix this error. Error message: (0,0) : error : THIS IS STILL ALPHA AND UNSUPPORTED SOFTWARE, USE AT YOUR OWN RISK. File name: /home/www/.www/gwcc/test/Global.asax Source File: Line 1: No assembly returned after compilation!? -- -- Also, I don't know if I mentioned this before or if it makes a difference, but I'm running mod_mono in Apache 1.3.31 on Slackware 10.0. What could be causing this? Regards, David P. Donahue [EMAIL PROTECTED] -Original Message- From: B Anirban [mailto:[EMAIL PROTECTED] Sent: Monday, October 04, 2004 5:25 AM To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: RE: [Mono-list] VB? Or just C#? Hi David, This problem is occurring since you have an older version of Microsoft.VisualBasic dll. And mono is not yet equipped for remapping this assembly to the newer version automatically. Please report bug in bugzilla (bugzilla.ximian.com) on this against mbas. Thanks. Anirban. David P. Donahue [EMAIL PROTECTED] 04-Oct-04 12:45:52 AM I just wrote a very, very simple ASP .NET project in using VB .NET in Visual Studio .NET and tried to run it on mono. The project is just one page with a button that, when pressed, changes a label. I created the page on the localhost on my Windows XP machine, build the solution in Visual Studio, then copied all the files to a folder shared by my Apache server (running mono) on my Linux box. When I attempt to access the page, I get a stack trace that ends with a file not found error for the .dll (even though it is there) and see the following in my Apache error log: ** (/usr/local/bin/mod-mono-server.exe:10038): WARNING **: Could not find assembly Microsoft.VisualBasic, references from /home/www/.www/gwcc/test/bin/gwcc.dll (assemblyref_index=1) Major/Minor: 7,0 Build: 3300,0 Token: b03f5f7f11d50a3a Should this be happening, or am I doing something wrong? Regards, David P. Donahue [EMAIL PROTECTED] -Original Message- From: Juan Cristóbal Olivares [mailto:[EMAIL PROTECTED] Sent: Sunday, September 12, 2004 4:16 PM To: David P. Donahue; [EMAIL PROTECTED] Subject: Re: [Mono-list] VB? Or just C#? Mono does not have a VB.NET compiller. You could compile your VB.NET site with .NET (Windows) and the run it with Mono. Juan Cristóbal Olivares - Original Message - From: David P. Donahue [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Saturday, September 11, 2004 7:49 PM Subject: [Mono-list] VB? Or just C#? Can Mono compile ASP .NET websites written in VB, or is it limited to only C#? If it can, I'm having trouble finding documentation on how... David P. Donahue
Re: [Mono-list] EntryPointNotFoundException: map_Mono_Posix_PollEvents
On Fri, 2004-10-08 at 11:21, Shaun Jackman wrote: I am not completely familiar with the C# platform. Are the .exe files that mono generates cross-platform? For example, can I expect the same exe to run on Linux, Win32 and Pocket PC platforms? You can, with some limitations. Basically, the code the .exe contains/uses must be present on all platforms you try to run it on. :-) So if you use (e.g.) System.EnterpriseServices, it'll likely only run on .NET, and not Rotor, PNet, Mono, or Pocket PC. While System.Collections should exist on all those platforms. Another exception is mixed-code assemblies -- assemblies that contain both native and managed code. Currently, only Managed Extensions to C++ can generate such assemblies, and such assemblies CANNOT be run on multiple platforms (nor should you expect them to, as a PowerPC can't run x86 code without introducing an emulator...). - Jon ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
[Mono-list] Using the mono jay parser
Hi, We are developing an application using mono which have to parse molecular definitions written in Smiles format. We have already use bison and flex tools to produce scanners and parsers before, so we want to keep doing the same ;). We have started to figure out how mono uses the jay modified version to produce its c# parser but we are having some problems. One problem is that we don't find how mono produces its scanner. What lexer is mono using?. The other problem is that we haven't found much documentation on how to use jay (not how to use lex and yacc languages) to produce a parser. It would be great to have some guidelines to start. Anything will be appreciated: files to look, ideas, concepts on how to use your modified jay... Cheers, -- Francis Brosnan Blázquez [EMAIL PROTECTED] Advanced Software Production Line, S.L. ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
Re: [Mono-list] Client side validation
sarvesh wrote: Now the bad news. 1. None of the client side scripts work on non-Microsoft browsers. Now i haven't been able find out any details about this. Can the developers throw some light on this. Will client validation not work on non-MS browsers only if they are ported from visual studio? i don't know and i haven't been able to figure out. There are other, 3rd party validation controls, and you can build your own, that could support other browsers... Unfortunately the stock ones don't work with other than IE. -- Michael J. Ryan - tracker1(at)theroughnecks(dot)com - www.theroughnecks.net icq: 4935386 - AIM/AOL: azTracker1 - Y!: azTracker1 - MSN/Win: (email) ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
[Mono-list] FC2 yum repository for 1.1.1
I don't see gtk-sharp or related packages (monodoc, monodevelop, etc) anywhere in the fc2 yum repository for 1.1.1 (http://www.go-mono.com/archive/1.1.1/fedora-2-i386/). Also, trying to install the mono-complete rpm doesn't pull these in as dependencies. Are these not supported in 1.1.x yet? Can I install from source if I need them? Is it possible to mix the 1.0.x versions of these packages with the rest of the 1.1.x packages? Thanks, Mark ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
Re: [Mono-list] FC2 yum repository for 1.1.1
On Sat, 2004-10-09 at 14:05 -0700, Mark Lewis wrote: I don't see gtk-sharp or related packages (monodoc, monodevelop, etc) anywhere in the fc2 yum repository for 1.1.1 (http://www.go-mono.com/archive/1.1.1/fedora-2-i386/). Also, trying to install the mono-complete rpm doesn't pull these in as dependencies. There is no gtk-sharp 1.1 release. To get gtk-sharp packages, please get it from the 1.0.x repository. As for mono-complete, it's a meta-package to install all of the mono-* packages. It is possible that we will ship something like mono-complete-environment or some such that will include everything we distribute. Finally, if you want gtk-sharp, you should then install gtk-sharp. The dependency tree for gtk-sharp will (if not, please report the bug) pull in all of its mono requirements. -- Duncan Mak [EMAIL PROTECTED] ___ Mono-list maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/mono-list
