Historical errata for Programming Sound with Pure Data

Para. Controls and Display:
“The “rate” and “depth” labels are comments that can be created by ctl-5".
My version of Pd (0.43.4-extended) allows for adding Labels to the slider controls. Unless you’re just pointing out that comments are available, I’d suggest mentioning the Label property as an alternative to using comments.

The sentence “Now select everything and move it so that the array1 graph neatly fits inside the red box.” is confusing. I got stuck trying to “fit inside the red box”. It seems that pd does not allow the array image to be resized, and the previous instructions to create a 100x100 red box made the box too small. Searching the pd docs was no help, so I decided that the solution must be to make the red box bigger (turns out the canvas needed to be 200 x 140 to fit the array exactly. Maybe this is a platform thing, I’m running pd on Mac OS X.

The image of the adsr_test pach is missing a tilde from the adsr object.

The recording setup on the main patch is missing the amplitude multiplier [*~ 0.25] after [catch~ out] that was created in the wineglass patch. Without it, the sound is recorded full blast and has distortion.

Subtley should be subtly

“The two panes below that are more interesting to us now because they tell us the spectral content of the sound..”
This feels better:
“The two panes below are more interesting to us now because they tell us the spectral content of the sound..”

The right setting for the rate slider should be 200 not 20 to match your comments on page 55 and 59 of the PDF.

The last word in “The pack’s right inlet comes from either the second argument, $2, or the third inlet, attack.” should read “attack speed” in order to distinguish from attack level.

“Then create two messages: 1 and 0 connected to the left inlet of an adsr 1 10 500 10 10 object”
Tilde missing after adsr

Figure 18 is labeled “Varying the Length of the Playback” and should be “Randomizing the Panning”

Fourth paragraph:
“The last thing to note is that there is a quick burst of energy across the whole frequency spectrum at the beginning for 2 three 3 milliseconds.”
Should be:
“[…] for 2 3 milliseconds.”

In Chapter 9 (no idea about the page, reading on Kindle app, it says location 2384 in the status bar) I suggest that the following requirements be added to the iOS example discussion:

Have XCode 5.1 and the iOS 7.1 SDK installed in order to build and run the iOS code examples. Notice also that you can set the target to build at the top of the toolbar (libraries, Tasks, etc).

The second partial is 70% as loud as the sharp attack, not 70% as loud as the fundamental.

Corrected:

“That makes the second 70% as loud as the [sharp attack] and fading over 300 milliseconds, and the third 60% as loud and fading over 200 milliseconds.”

“sends a message to ramp down to 0 in the number of [milli]seconds specified to the duration inlet.”

Change “right” to “left” in the text in order to match the sample sword_fight.pd file. If 1 is sent to the right inlet, as written, no sound is generated. Sending 1 to the left inlet, as in the sample code, works.

“First it initializes the line~ to 1 by sending a 1 to its right inlet,”

Should be:

“First it initializes the line~ to 1 by sending a 1 to its left inlet,”

“Notice that we’re assuming the list~ is always set with the correct previous value.”

Should be:

“Notice that we’re assuming the line~ is always set with the correct previous value.”

Let’s take a look inside the table[s] subwindow

s/unsheathe/unsheath/

The sentence below implies that the idle sound fades out after sheathing is complete, but in reality it fades out when sheathing begins:

“This will fade the idle sound in or out over 100 milliseconds after the unsheathing or sheathing is complete.”

Following is correct?

“This will fade the idle sound in or out over 100 milliseconds after the unsheathing is complete or when the sheathing begins.”

Some of the .so files in thsound-code download are corrupted.

Dump of file: android\\PdCore\\libs\\armeabi\\libpd.so, type: ELF32
Error 2110: COFF file section table corrupt

At run time this results in a “empty/missing DT_HASH” error.

In addition, the build.gradle files are out of date and need some work in order to build the project in Android Studio 1.0 (which is the only “supported” IDE).

I haven’t been able to rebuild the .so files from source yet, but I was able to build the Tasks app in Android Studio and run it on Android after I modified the gradle files and downloaded valid .so files from github.com/libpd/pd-for-android

By the way, I copied the valid, prebuilt .so files to android/TasksProject/Tasks/src/main/jniLibs