C/C++ Programming Help Thread

[hallo007]

O yes, and MAKE SURE that you put

#include </usr/include/sys/stat.h>

up were the rest of your #include's are let me know if this works for you.

#include <sys/stat.h>

will work aswell;)

[Raphael]

Did any one ever experience a problem with Start?
even if I press it once, it acts as if the button's pressed twice, breaking the loop!
OR is my button faulty?

I bet it comes from how you poll the button states.
Some facts about that (a lecture about button state polling and logical ops):
A call to sceCtrlPeek/Read* will only return the *current* button state and does not return changed button states or whatever. Therefore, calling the function twice will return the same button state twice (unless the user pressed a button in the timeframe between the two calls, which is unprobable because it's only a hand full of clock cycles and thats clock cycles/222000000 seconds of time, supposing your PSP runs at stock speed).
Therefore you need to recognize if a button that was pressed previously, was released once before you check again.
That's what the Peek/ReadBufferNegative does. So call it once before your Positive read. Then check if the buttons that were previously pressed, were released again by comparing the last_button_positive bitmasks to the button_negative bitmask.

sceCtrl last_buttons_positive;
sceCtrl buttons_positive;
sceCtrl buttons_negative;
while(blubb) {
   sceCtrlReadBufferNegative( &buttons_negative );
   sceCtrlReadBufferPositive( &buttons_positive );
   if (~(last_buttons_positive.buttons & (last_buttons_positive.buttons ^ buttons_negative.buttons)) & buttons_positive.buttons & BUTTON_MASK) {
     // the button that was previously pressed, was released before, so this is
     // a new "press" event. Same if the button wasn't pressed before.
   }
   last_buttons_positive = buttons_positive;
}

That's what logical ops are for. If you have problems understanding the if statement, read up what & (AND), ~ (NOT) and ^ (XOR) does. It's nothing more then memorizing a 2x2 table of single digits. Then simplify the case for a 1bit button mask and go through the possible cases when you want to check for a button that is currently pressed (button_positive flag is 1):
last_button was pressed (the flag is 1):
- button was released (button_negative flag is 1)
- button was not released (button_negative flag is 0)
last_button was not pressed (the flag is 0):
- see above two subsequent cases

You'll notice that the masks evaluate to 0 when the button was pressed before and wasn't released (or when the button isn't pressed at all currently), and to 1 else.

PS: Alternatively you could use the famous "last_button" method, though that will also recognize button releases as presses (ie you will still have two recognized button events for one down&release event), unless you create a negative mask through that yourself.

[BlackShark]

how would i implement the anolog stick? like what function could i do to use the anolog stick to move my player?

[Glynnder]

Put this before your while(1) loop

		  	sceCtrlSetSamplingCycle(0);
	sceCtrlSetSamplingMode(PSP_CTRL_MODE_ANALOG);

Then use this:

                  sceCtrlReadBufferPositive(&pad, 1);                    
if(pad.Lx==whatever)
{
     do whatever
}
if(pad.Ly==whatever)
{
     do whatever
}

[BlackShark]

Awsome thanks
-= Double Post =-
Ok im having problems with this, it sorta works, problem is, at the start the player moves diagonaly down and to the right and off the screen, here is the code,

//Analog stick                            
             if(pad.Lx >= 40 && playerX < 480) {
                       playerX = playerX + 2;
             }
             if(pad.Lx <= -40 && playerX > 0) {
                       playerX = playerX - 2;
             }
             
             if(pad.Ly >= 40 && playerY < 272) {
                       playerY = playerY + 2;
             }
             if(pad.Ly <= -40 && playerY > 0) {
                       playerY = playerY - 2;
             }

and i did define the statements above the while loop like,

sceCtrlSetSamplingCycle(0);
	sceCtrlSetSamplingMode(PSP_CTRL_MODE_ANALOG);

It compiles HOWEVER it does give me two warnings,

[Warning] comparison is always false due to limited range of data type 
[Warning] comparison is always false due to limited range of data type

My guess is that I can't have negatives when comparing the analog stick? if not, how do i make it so that if up is pressed, player goes up, down is pressed, player goes down, right is pressed player goes right, left is pres...etc...? as in what do i compare it to?

[Glynnder]

analog stick can't go negative
this is the code I used for iMockPod's going up and down:

if(pad.Ly>230){}
//down
if(pad.Ly<30){}
//up

[BlackShark]

ahh, i though so, thanks again!

[psphacker12.]

How do you read from ID3 tags ?

[hallo007]

is there a source code from a prx that take screenshots , because i have problems using the graphics.c+prx

[SG57]

Ill post some screenshots of my current game build. Its a first-person sword fighter in 3D, and i have it all working. Running, standing and attacking animation, each for there respective button presses. Its really nice right now. I should be able to finish the game within 2 weeks.

[MasterChief117]

Ill post some screenshots of my current game build. Its a first-person sword fighter in 3D, and i have it all working. Running, standing and attacking animation, each for there respective button presses. Its really nice right now. I should be able to finish the game within 2 weeks.

cool!

[JaSo PsP]

thanks very much for your help (also i was going to change the STATE names but thats what they really were, STATEs, so i just decided to keep them, but looks like i won't need alot of them much longer, thanks again!

EDIT: alright this should be about the last question I but you with for now,
but how would i go about making a 4 direction bullet system and how do I implement the analogue stick?

-= Double Post =-



Try defining this above your while loop...

int GetRandomNum(int lo, int hi) {
 SceKernelUtilsMt19937Context ctx;
 sceKernelUtilsMt19937Init(&ctx, time(NULL)); //SEED TO TIME
 u32 rand_val = sceKernelUtilsMt19937UInt(&ctx);
 rand_val = lo + rand_val % hi;
 return (int)rand_val;
}

Then call it like,

GetRandomNum(*INSERT # HERE* , *INSERT # HERE*)

O yes, and MAKE SURE that you put

#include </usr/include/sys/stat.h>

up were the rest of your #include's are let me know if this works for you.

It does work man, thanx alot, but there is a big problem. My game needs random numbers to be chosen quicker than once every second. Is there any way to make it generate a number qucker?

[Mr305]

I bet it comes from how you poll the button states.
Some facts about that (a lecture about button state polling and logical ops):
A call to sceCtrlPeek/Read* will only return the *current* button state and does not return changed button states or whatever. Therefore, calling the function twice will return the same button state twice (unless the user pressed a button in the timeframe between the two calls, which is unprobable because it's only a hand full of clock cycles and thats clock cycles/222000000 seconds of time, supposing your PSP runs at stock speed).
Therefore you need to recognize if a button that was pressed previously, was released once before you check again.
That's what the Peek/ReadBufferNegative does. So call it once before your Positive read. Then check if the buttons that were previously pressed, were released again by comparing the last_button_positive bitmasks to the button_negative bitmask.
That's what logical ops are for. If you have problems understanding the if statement, read up what & (AND), ~ (NOT) and ^ (XOR) does. It's nothing more then memorizing a 2x2 table of single digits. Then simplify the case for a 1bit button mask and go through the possible cases when you want to check for a button that is currently pressed (button_positive flag is 1):
last_button was pressed (the flag is 1):
- button was released (button_negative flag is 1)
- button was not released (button_negative flag is 0)
last_button was not pressed (the flag is 0):
- see above two subsequent cases

You'll notice that the masks evaluate to 0 when the button was pressed before and wasn't released (or when the button isn't pressed at all currently), and to 1 else.

PS: Alternatively you could use the famous "last_button" method, though that will also recognize button releases as presses (ie you will still have two recognized button events for one down&release event), unless you create a negative mask through that yourself.

All these Days, was wondering what Positive, negative -- and also that they were synonymous(the same)!
It's all now Clear! Thanx.
What Does Peek do then (when Negative means, the buttons is pressed down currently)?
[I remember using it once & getting all wierd results! ]

[jsharrad]

What Does Peek do then (when Negative means, the buttons is pressed down currently)?

sceCtrlPeekBufferPositive () just skips the vblank wait

[psphacker12.]

can anyone here define the functions "UnmapViewOfFile" and "CloseHandle" ?
I need them...
I have searched GOOGLE(most of them arent general C, WIN32)

[Insert_Witty_Name]

Sure.

int UnmapViewOfFile()
{
return 0;
}

int CloseHandle()
{
return 0;
}

</sarcasm>

[Raphael]

That's IWN's way to say: "What the **** are you talking about?" ;P

[Mr305]

 Outer while 
{
..............
sceCtrlPeekBufferPositive (&positive, 1);
sceCtrlPeekBufferNegative (&negative, 1);
if (~(lastpositive.Buttons & (lastpositive.Buttons ^ negative.Buttons)) & positive.Buttons & 0x000008){
clearScreen(0x000000);
printTextScreen(150,136,statuspause,RGB(0,0,255));
flipScreen();
lastpositive = positive;
do
{
sceCtrlPeekBufferPositive (&positive, 1);
sceCtrlPeekBufferNegative (&negative, 1);
if (~(lastpositive.Buttons & (lastpositive.Buttons ^ negative.Buttons)) & positive.Buttons & 0x000008)
{ 
lastpositive=positive;
break; 
}
}while (1);
}
.................
}

After sending the EBOOT 19th time for testing[after trying several variations], I am posting this...
It works if it is used once in one loop, but above it doesn't seem to work!

[psphacker12.]

what would be the best way to read id3 tags in C?

[SG57]

Uh, use your the logical - thinking lobe of your brain to google, or use that library for what your saying. Maybe... search?

[Grimfate126]

what would be the best way to read id3 tags in C?

google for id3 tag specs. find them, write your own lib.

[SG57]

First thing that popped up in google:

http://dl.qj.net/MP3-ID3-v0.01-Beta-...5445/catid/151

Consult the author for his library (i believe he released it...)

[psphacker12.]

couldn't I just do...
fseek (pFile , -128, SEEK_END);

[PSPJunkie_]

That's what writing your own lib for it means.

[nataku92]

couldn't I just do...
fseek (pFile , -128, SEEK_END);

yeah that would be fine for ID3v1. ID3v2 is a bit different though. Just follow everyone else's advice and google for the formats.

[slicer4ever]

ok i have a prob with my 3d object

Spoiler for code of rotation:

	float inc_rate = 0.1f;
	droney[1] += inc_rate;
	float angle = droney[1]*(PI/180.0f);

	
        dronexrot = (float)sin((angle+90.0f)*(PI/180.0f))*(0.02f);
	dronezrot = (float)cos((angle+90.0f)*(PI/180.0f))*(0.02f)*-1;
	dronex[1] = dronexrot;
	dronez[1] = dronezrot;

	float xtrans = -10-xpos+dronex[1];
	float ztrans = -10-zpos+dronez[1];
	float ytrans = -walkbias-0.45f-ypos;
	float sceneroty = ((360.0f)*(PI/180.0f)) - ((yrot)*(PI/180.0f));
	sceGumLoadIdentity();
        {
		ScePspFVector3 move = {xtrans,ytrans, ztrans};
		sceGumRotateX(lookupdown);
		sceGumRotateY(sceneroty);
		sceGumTranslate( &move );
		sceGumRotateY(angle);

	}

thats just for rotation i got it to rotate in degrees instead of radians but now i can't get the object to move in the direction it is faceing and i'm not sure why i believe that code should work but it won't and it's makeing mad i've made countless modifications 2 it yet it still won't work

[Raphael]

After sending the EBOOT 19th time for testing[after trying several variations], I am posting this...
It works if it is used once in one loop, but above it doesn't seem to work!

I tried the code and it seemed to work for me (maybe your Start button really is just pretty sensitive? in that case you should try adding delays into the button polling to make it poll only ~5-10 times per second - see bottom of post for the delay function), however, I noticed that BufferNegative behaves something different from what I assumed. That made me test it a bit more and the logic seems to be 90% a simple !positive, but in some cases it doesn't recognize button events while the positive does. Well, anyhow I need to revise my sample for propper Button event polling (though it worked, but the logic behind it was not the logic I was trying to show):

	SceCtrlData positive, lastpositive, presses, releases;
	lastpositive.Buttons = 0;
	while(outer_loop) {
		sceCtrlPeekBufferPositive (&positive, 1);
		presses.Buttons = ~lastpositive.Buttons & positive.Buttons;
		releases.Buttons = lastpositive.Buttons & ~positive.Buttons;
		lastpositive = positive;

		if (presses.Buttons & BUTTON_MASK)
		{
			// do what should happen if button was pressed (not released!)
			// for example, wait for another button press:
			while (1) {
				sceCtrlPeekBufferPositive (&positive, 1);
				presses.Buttons = ~lastpositive.Buttons & positive.Buttons;
				releases.Buttons = lastpositive.Buttons & ~positive.Buttons;
				lastpositive = positive;

				if (presses.Buttons & BUTTON_MASK)
						break;
			}
		}
	}

That code works same as my previous version did, but it's easier to understand and easier to handle the button down (presses) and button up (releases) events. You maybe could put the Button masking into a seperate function that updates the global presses and releases flags, just to make the code more readable. Or put the inner loop into it's own "WaitForButtonPress( int ButtonMask );" function and insert a sceKernelDelayThread(1000 000) if you're gonna use it inside XMB/multithreaded application or just want to limit the number of button events received per second (a delay of 100000 means at most 10 polls per second - 200000 means 5 polls per second).

[JaSo PsP]

It does work man, thanx alot, but there is a big problem. My game needs random numbers to be chosen quicker than once every second. Is there any way to make it generate a number qucker?

Well? This has been completely ignored....

[Waterbottle]

Well? This has been completely ignored....

Can't you just call rand() every time you want a new random value...?

[Mr305]

I tried the code and it seemed to work for me (maybe your Start button really is just pretty sensitive? in that case you should try adding delays into the button polling to make it poll only ~5-10 times per second - see bottom of post for the delay function), however, I noticed that BufferNegative behaves something different from what I assumed. That made me test it a bit more and the logic seems to be 90% a simple !positive, but in some cases it doesn't recognize button events while the positive does. Well, anyhow I need to revise my sample for propper Button event polling (though it worked, but the logic behind it was not the logic I was trying to show):

	SceCtrlData positive, lastpositive, presses, releases;
	lastpositive.Buttons = 0;
	while(outer_loop) {
		sceCtrlPeekBufferPositive (&positive, 1);
		presses.Buttons = ~lastpositive.Buttons & positive.Buttons;
		releases.Buttons = lastpositive.Buttons & ~positive.Buttons;
		lastpositive = positive;

		if (presses.Buttons & BUTTON_MASK)
		{
			// do what should happen if button was pressed (not released!)
			// for example, wait for another button press:
			while (1) {
				sceCtrlPeekBufferPositive (&positive, 1);
				presses.Buttons = ~lastpositive.Buttons & positive.Buttons;
				releases.Buttons = lastpositive.Buttons & ~positive.Buttons;
				lastpositive = positive;

				if (presses.Buttons & BUTTON_MASK)
						break;
			}
		}
	}

That code works same as my previous version did, but it's easier to understand and easier to handle the button down (presses) and button up (releases) events. You maybe could put the Button masking into a seperate function that updates the global presses and releases flags, just to make the code more readable. Or put the inner loop into it's own "WaitForButtonPress( int ButtonMask );" function and insert a sceKernelDelayThread(1000 000) if you're gonna use it inside XMB/multithreaded application or just want to limit the number of button events received per second (a delay of 100000 means at most 10 polls per second - 200000 means 5 polls per second).

danke sehr [I ain't no german!]
Works like a charm!