Why a new sound from sample of a sound isin't equal to the original sound


Why it return false ? Normally, there's the same sound ! (if i listen the new sound, it's like the original sound for my ears)

In my database project, i save the samples of the soung into the browser and i check if the value is saved correctely into the browser: this step fail ! because the 2 lists aren't equal...

Column A is sample of sound, column B id sample of the new sound of the sample of the original sound ! (haha)

see the difference and why it was a difference ?

similar problem with pixels of costume...

are there any numbers in that array that aren't very near zero? Can you just clip the noise from the sample? (If sample < 0.00001 then 0 else sample).

It's easy to imagine why the sound one doesn't work exactly when decoding and then re-encoding as it involves floating pointing operations, but harder to understand why decoding/re-encoding a costume would give different pixels values

Can you publish an example project that has the problem?

media is not stored as Snap lists of float samples, but there is compression involved, we're just letting you use them as if they were. I thought that was obvious.

Remember Obiwan - we are mere padawans :slight_smile:

Yeah me too. I mean, I figured sounds are stored in whatever format the underlying browser sound feature provides, but I've never looked into what format that is.

Thank you, I took another method to perform the verification

Finally, the problem came from elsewhere for the costume...

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.