Triangular Wave

[Jody Monahan]

Hi all,

I am having a problem creating a triangular wave. Essentially I would like a wave that oscillates from -0.5 to 0.5 with increments of 0.005. I have been able to do this for 1 iteration i.e it oscillates from -0.5 to 0.5 and back down to -0.5, however I would like this to be periodic, and repeat itself so that there is a certain number of samples left in the array (832267 samples to be exact). The following is the code I have currently come up with:

float[] in_wave = new float[832662];
    	
// Create triangle wave   
float delta = (float) 0.005;
float minf = (float) -0.5;
float maxf = (float) 0.5;	
float trem_length = (maxf-minf)/delta;
// make trem_length an integer value
int trem_length1 = (int) trem_length;
float[] trem1 = new float [trem_length1];
float[] trem = new float [in_wave.length];

for (int i = 0; i < trem1.length; i++)   
{       
	trem[i] = minf + i * delta;		
}
			
for (int i = 0; i < trem1.length; i++)   
{       
	trem[trem1 * i] = maxf - i * delta;		          
}

Thanks in advance.

[David Newton]

So you basically want to repeat what you already have for as many times as you want samples (with the caveat that what you have there already fills a certain number of those samples), yes?

What might that look like?

[Jody Monahan]

Yes thats exactly what I want to do, just repeat what code I have previously. It should look like a periodic triangular waveform of length 832267.

Note: the significance of the number 832267 is the number of samples of a wave file that I am trying to manipulate.

Thanks

[David Newton]

Right, and I'm asking you to make an attempt at that code.

[Jody Monahan]

Oh right, sorry. I have made an attempt just by putting it all in another for loop, but the way I have gone about it isn't exactly correct because I need the array to be exactly 832267 samples in length.

Attempt:

 

float[] in_wave = new float[832662];
						
// Create triangle wave LFO   
float delta = (float) 0.005;
float minf = (float) -0.5;
float maxf = (float) 0.5;	
float trem_length = (maxf-minf)/delta;
int trem_length1 = (int) trem_length;
float[] trem1 = new float [trem_length1];
float[] trem = new float [in_wave.length];
System.out.println(trem_length);
			
			
// Create triangluar wave LFO
for (int j = 0; j < 2080; j++)
{
	for (int i = 0; i < trem1.length; i++)   
	{       
		trem[i + (j * 2 * trem_length1)] = minf + i * delta;
	}
				
	for (int i = 0; i < trem1.length; i++)   
	{ 
		trem[i + trem_length1 + (j * 2 * trem_length1)] = maxf - i * delta;		          
	}
}
			

[Darryl Burke]

edit Cross posting allegation posted in error. The OP isn't a cross poster.

[Maneesh Godbole]

...
[removed unnecessary link caused by a misunderstanding over cross posting]

[Jody Monahan]

Hi Maneesh and Daryl, I hope you realise that that wasn't me posting a similar topic across different forums. Maneesh, I am fully aware of the rules and regulations of the site. If you read my posts I have asked a specific question about my code, and that question still stands if anyone can provide some help.

Thanks.

[David Newton]

Don't worry, Jody, it appears Daryl was being a bit hasty, as the other posts are regarding graphs, and yours isn't. Plus the wording on all the other posts is identical, and this one isn't.

[Jody Monahan]

Thanks for clearing that up David. Any way, I have spent another few hours looking at the code and have come up with the following solution:

    

float[] in_wave = new float[832662];
			
// Create triangle wave LFO   
float delta = (float) 0.005;
float minf = (float) -0.5;
float maxf = (float) 0.5;	
float trem_length = (maxf-minf)/delta;
int trem_length1 = (int) trem_length;
float[] trem1 = new float [trem_length1];
float[] trem = new float [in_wave.length];			
			
int tilde = in_wave.length / trem_length1 / 2;

for (int j = 0; j < tilde; j++)
{
	for (int i = 0; i < trem1.length; i++)   
	{       
		trem[i + (j * 2 * trem_length1)] = minf + i * delta;
	}
				
	for (int i = 0; i < trem1.length; i++)   
	{ 
		trem[i + trem_length1 + (j * 2 * trem_length1)] = maxf - i * delta;		          
	}
}

It doesn't look great but it does the trick. Thanks to anyone that looked at it.

[David Newton]

Have you thought of any other possible approaches? Sometimes just posting "out loud" is enough to do the trick ("rubber ducking" it).

[Darryl Burke]

Jody Monahan wrote:Hi Maneesh and Daryl, I hope you realise that that wasn't me posting a similar topic across different forums. Maneesh, I am fully aware of the rules and regulations of the site. If you read my posts I have asked a specific question about my code, and that question still stands if anyone can provide some help.

Thanks.

Jody, my sincerest apologies to you, Maneesh and all forum members. I should have read the post, not just the subject line.

[David Newton]

@Darryl: just be careful when making accusations. Fortunately, I've learned this: Доверяй, но проверяй. (And am getting good pronunciation tips from my current boss, who's Russian.)

[Maneesh Godbole]

Jody,
My sincerest apologies too. I should not have acted in haste.

[Mike Simmons]

David Newton wrote:@Darryl: just be careful when making accusations. Fortunately, I've learned this: Доверяй, но проверяй. (And am getting good pronunciation tips from my current boss, who's Russian.)

Мы вас похороним!!!

Sorry, got carried away.

So, are we repealing the long-standing rule that this is an English-speaking forum, and that posts are expected to be in English? I kind of like being able to come here and read what people are saying without having to resort to Google or babelfish. I hope that can continue.

[Rob Spoor]

I agree. So here are the translations from Babelfish:
David's text: Entrust, but check
Mike's text: We will bury you

[David Newton]

That one's better in Russian, though, 'cuz it rhymes.

[Rob Spoor]

Not speaking any Russian I'll take your word for it. Now back on-topic!