Score:1

Audio Distortion Ubuntu Studio using Jack Audio

ht flag

I just installed Ubuntu Studio on a newly purchased laptop. A Lenovo flex 5 14. There is an issue when using programs like hydrogen or ardour. The audio coming from the mixdown is very distorted. I have had this problem in the past several years ago. It was a mixer setting too high. However here I cannot find the problem. Using alsamixer, or qasmixer on multiple sliders does not seem to affect the quality very much on the output.

Any ideas to trouble shoot. It worked fine on my previous laptop, on that one I had a usb sound blaster card.

Score:0
ht flag

Thanks for the extensive and contemplative understanding. I hope to follow more of what you are describing conceptually, however not fully this moment. Yet I wanted to post the solution which I used to solve the problem.

The period and frame buffers setting in the ubuntu studio audio setup was not set correctly. It was defaulted after install to a value which is not suitable for my laptop. I changed it to 256 and 2. However this change did not effect the operation at once. To properly have the settings changed, I used the qjackcontrol to set it up. After that, it seems that the ubuntu studio setup dialog functioned.

jack period buffers 256 buffers 2

Nate T avatar
it flag
Glad you solved it. One way to check whether these are correct is to check for `xruns`. These will be noted in the JACK logs. If you are using `qjackctl` or `Carla`, there will be a resettable counter telling you how many have occurred since the last reset. You want the smallest setting that keeps that counter at 0. Also, the more plugins you run simultaneously, the more likely you are to get an `xrun`.
Score:0
it flag

By the description you give of the issue, I will assume that the sound you are noticing is clipping. This is a term used by people in the pro-audio industry to discribe non-harmonic distortion of your output. In other words, certain frequencies are exaggerated, and not the good ones. The result will be an immediately recognizable timbre.

We all know what a "blown" speaker sounds like. Clipping inside a DAW will sound somewhere between that an the sound effect you hear in music when someone leans in extremely close to the mic and screams.

If it is just clipping, this is great news, because it is an easy fix. There is no troubleshooting to narrow down the source, because the source doesn't really matter to us...

If a fader is too far open causing clipping, that does not mean you need to turn down that same fader. To understand this, we need a certain understanding of audio signals.

A signal, at any one point in time, can be thought of as actually being an infinite amount of separate signals. Each of these signals goes on to reproduce a different frequency. The volume of a subsignal does not matter; only the volumes relative to each other matter. If A is twice the volume that B is, you need it to stay that way unless you explicitly change it using something like a guitar amp.

Thankfully, with the proper settings the mapping of relative VUs stays pretty consistent. At lower volumes it is ok as well, although with Digital audio, it is harder to accurately record the relative to each other. This is because digital signals are not continuous like analogue. The explanation is out of scope here, but I recommend that you look it up. It is fairly interesting, imo.

Anyway, If the signal is a bit 'left of center' at one point in your chain, you can usually fix it at the next point, so long as those relative levels do not move. Once they are skewed, however, you will need special software to fix, and that only masks. This last part only really affects audio recorded at a clip, so you are good ther.

Anyway, if I have made my intended point, you likely already see the solution. If you take away some of the signal at a point before the corruption, you will counteract it without having to locate it.

mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.