dbE2.wav dbE1.wav A2.wav C2.wav A1.wav G2.wav D2bass.wav C3pizz2.wav B1pizz.wav E2pizz.wav Fsharppizz.wav

Hi David, Thanks so much. I've uploaded the sound files, as they're quite small. It'll make it easier as they'll just loadbang into the tables if you have them in the same folder. I really appreciate your looking at that. I'll have a look at those links carefully. -George

Hi again, and thanks for your suggestions.
I tried lowering the gemwin fps from 60 to 20, which I believe is the default for gemwin, but that didn't help. I still got an audio error with crackling.
I also tried slowing down the flashes. Also, no luck.
What I did find out is that the error only occurs when the gem window is positioned on the other monitor; having been dragged with the mouse before I've set the patch in motion. This error also occurs if I drag the gemwin around while my patch is functioning with the metro without any monitor connected. In other words, it's clear that simply having the external monitor connected and functional is not the cause for the error. It seems to be an issue with coordinating the video signal with another monitor, or when the gemwin is being interfered with while the patch is in progress irrespective of the connection of an external monitor. This may not seem like a revelation, but at least I know it's not simply the presence of the second monitor attached to my computer that's causing the audio error. I'll keep trying things. Thanks for offering to run the patch on your own system. I'll upload it, but I don't know how useful it will be as you won't have all the audio that's probably encumbering the system. For that I'd also have to upload the sound files - about 8 cello samples. Anyway, here's the patch, and thanks for your help .Study no 3 draft4.pd

Hi again, Thanks for sending the fix patch, David. However, it unfortunately hasn't resolved the issue. I've tried now using an external sound card, and that hasn't helped either.
Could the issue be with the delay I have coming from the audio sound card? I don't really want to increase this delay, which is currently 80 ms, and the block size is 64.
The patch is being used for a psychology experiment, where flashes on the screen need to be synchronized with beeps; so increasing the delay could conflict with what I'm trying to achieve with the experiment.
The crackle is really distracting, but I'm also concerned that the audio might ultimately fail in the middle of the experiment, rather than just give an error message and crackle.
Thanks in advance for any additional suggestions.
-George

Hi,
Thanks for getting back to me.
I'm using a MacBook Pro with 10.9.4 OSX.
My PD version is 0.43.4 extended.
The monitor I'm using is just an old Dell computer monitor, so not high definition as far as I know.
I connected it into the computer, but nothing new came up on the audio output window. The thing is that the audio is in fact playing, but the PD window shows the audio I/O error message in red, and there are crackling sounds.
-George

Hi, I'm making a patch with gemwin and audio. It works no problem when I use it on my laptop computer screen. As soon as I plug in an external monitor and display the gemwin on it, however, I get 'audio error' in red on the pdwindow, and hear little crackles and glitches in the audio. I'd be grateful for any advice on how to fix this. -Thanks

Will do. Thanks again.

Thanks so much, Sunji. That explains the problem, and I've increased the frame rate.

However, I read somewhere that the frame rate limit for gemwin is 60 fps. Do you know is this true?
Does that mean that the quickest flash event possible is 33 ms?
I was hoping to go down to at least 20 ms per flash, if not lower (7ms according to the literature visual perception of flashes).
I appreciate your help on this.

Hi, I'm using pd to create a flashing light on a computer screen for a psychology study. I simply need a white circle to flash twice in quick succession, or once quickly, and for the participants to tell if it's flashing once or twice.
I used a gemhead with a circle object, and experimented with using a delay to turn on and off a toggle into the gemhead. It seems the system is reliable if I keep the delays at 50 ms for everything, but unreliable if I go lower than 50 ms. I need to go as low as 17 ms delay for the circle on. Is there some kind of latency in terms of the gemhead processing the incoming signal? I can't imagine there's a great deal of computation for a single white circle. Please tell me if there's a computationally less expensive way. Thanks.

I have tried many times and asked quite a few people for advice on this, but I'm still failing to get a patch to open on startup.
I'm kind of desperate for help at this point as I have a show opening on Friday which is running PD and needs to open on startup.
I checked the path on terminal for the patch I want to load and it told me it was:
/Users/georgehiggs/Documents/Evolvaphone/A\ drafts/evolvaphoneperformancenew1.pd
My first question was why there was a backslash after the 'A' on 'A drafts, as there is none on the folder name, but there is a space.
I tried taking out this backslash, then replacing it with an underscore, and still I get this error message:

consistency check failed: sys_decodedialog: \{+open-+_/Users/georgehiggs/Documents/Evolvaphone/A+_drafts/evolvaphoneperformancenew1.pd\}
error: .inputBox.txtInput: no such object

I'd be so grateful for any help, as I've checked the old posts but can't find any help for this specific issue.
Thank you.
-George

This is a simple question, but I am having trouble getting a patch to open on startup, presumably because I am using incorrect commands. As a start up flag, I typed in:
-open georgehiggs/Documents/Evolvaphone/A drafts/draft4.pd

Here is the error I get:
-open georgehiggs/Documents/Evolvaphone/A drafts/draft4.pd: can't load library

I've tried a forward slash before georgehiggs, as well as a tilde plus forward slash.
Can anyone help me here with this simple issue?
Thanks.
-George

Has anyone experienced the CPU load going up gradually in a patch when the dsp is switched off? When my dsp is on, the load metre is around 29-40 percent. When I turn it off, it falls to 4 percent then starts climbing gradually in steps until it can go well into the 370 percent zone, and then crash (that's only if I turn my back for ten minutes). The gradual climb goes up say 3 percent then down 2 percent, and then up 3 again and so on, rather than just a constant climb. I thought it might be some kind of feedback I have in the patch, but can't understand why this wouldn't occur when the dsp is on.

When building a patch with numerous elements in different windows, is it possible to say which is less taxing on the processor: abstractions or subpatches?

I have been getting the repeated error message " multiply defined" when I use multiple sends with the same name. When I remove them, the error message disappears.
However, I was reading in the documentation that you can have as many send and receive objects with the same name as you want.
Can anyone clear this up?

I was wondering if oscillators(osc~) might behave unusually if the processor is being stressed. I have quite a few oscillators going to different outputs, and if I leave the computer, or change something, the pure sine tones become distorted, even getting a dynamic ring modulation sound (rather Cylonesque).

I was wondering if anyone would have advice on how to trim the silences around a recorded sample once the sample is recorded into a patch.
Also, as an alternative, would you have any suggestions on how to set up a adc with tabwrite so that it only records when a certain threshold has been achieved, like a traditional sampler?

I tried to use the back slash, and literally cut and pasted from the terminal window again, but still the error reads:
open- /Users/georgehiggs/Documents/Evolvaphone/A_drafts/evolvaphoneperformancenew1.pd: can't load library
I didn't use the underscore when I entered it, but instead used backslash space, just as in the terminal window.
Can you think of any other suggestions to try?
Thanks.

Hello steffen,
I found the problem with that dsp load increase. Basically I had a bunch of table reading through counters which I were in conflict. It doesn't explain why the load increased when dsp was turned of, but I sorted it out.
Thanks for the offer of uploading the patch, but it was such a mess I was embarrassed to.
-George

I took your advice and went through the patches thoroughly, chopping as much as I could. As well I increased the latency somewhat. This helped tremendously, and the patches are running at about 1/3 the previous load. It seems that the dimensions of GUIs alone presents a challenge for the processor, so even reducing the size of all my sliders and arrays helped. I may be mistaken on the last point, but it seemed to help.
Thanks for your advice.

I think I answered my own question. I believe gaps occur in the audio if the processor falls behind the real time audio demands.

Cool. Thanks.