Warning - Opening old project file

XP SP3 Audacity 2.0.6

2.0.6 is working fine for new project but I went back to work on something from 2011 and get the “Warning - Opening old project file” window saying that the format has changed from 1.2.6 and that it likely won’t work (and it doesn’t).

So what do I do? Reinstall the old version? Have both versions running? Is there a “convert” routine.

I’m not a heavy or frequent user and I’ve searched for the subject message but must have missed the obvious.


I don’t recognise that warning message. Could you quote the warning word for word?

This is the message:
and if I continue to open the project this is the result:
I’ve got stacks of old projects on my “to do” list but surely this must have been a common issue?

Audacity 2.0.6 has opened the project correctly.
That image shows that the audio in the project is hitting 0 dB and so will probably be distorted.
You can turn off the red “clip indicators” in the View menu. See: http://manual.audacityteam.org/o/man/view_menu.html#show_clipping

Also, please be careful with Windows XP. Microsoft are no longer officially supporting it, so it will be unpatched against any new security vulnerabilities that appear.

Ensure you have a good third-party anti-virus application. Be aware though that this will not give you complete protection against a previously unknown attack.

For some extra protection you can install http://technet.microsoft.com/en-us/security/jj653751 .

If it is not practicable for you to update to the much more secure Windows 7 or Windows 8, you can consider installing a Linux operating system instead. Most versions are free and all are very secure.


Yes I thought the distortion was due to the new version.

Thanks and my apologies. I’ve now (after backing up) opened a couple of other older projects and though I obviously get the warning they are fine. Yes I know about the clipping and I’m astounded that I over cooked that project which was the first old one that I looked at with the new version and hence my panic.

No, that will have been in the original project.

Easily done.
If you keep the tracks as 32 bit float format, then if over 0 dB is caused by processing, it can be corrected by amplifying (either the Amplify or Normalize effects) back down below 0 dB. You can’t do that with 16 bit or 24 bit tracks because those formats do not support over 0 dB.

That’s useful to know (I think). I’m not sure what you mean by ‘caused by processing’.

Most recording are for a specific purpose using a mic and we know the settings needed but every now and again I’ll pick something up from an internet source or from my stereo system and I think if the internet source is live there is little opportunity to get levels right as they change from music to speech etc so the 32 bit float flexibility will be helpful. For the others there is usually opportunity to do a level test beforehand to ensure that peaking doesn’t occur.

Are there any downsides to using the 32 bit float format?

Eh eh. Well here isn’t the right place to get involved in a long discussion but XP has been insecure since its release and in nearly 15 yrs they still didn’t fix it so I’ve used an insecure OS for all that time. I do have other layers of protection and EMET looks interesting though the full Framework requirement is somewhat off-putting. I’m hoping to hang on to this machine until I go Windows 10 (I’ve always trust the odd numbers and it really is 9).

If you set your recording levels too high and the recording is clipped (distorted), then that is permanent - time to reshoot.

If you have a good, non-distorted recording and you apply an effect that results in the waveform getting bigger vertically to the extent that it smashes into the top/bottom of the track, then:

a) If the track is 16 bit or 24 bit, (see the info panel on the left end of the track), then the damage is permanent - time to reshoot.
b) If the track is 32 bit float, then the waveform itself is not damaged - it will still distort if you play it, but it is recoverable - all you need to do is to amplify (“Amplify” or “Normalize” effects) to below 0 dB, then it will be good as new.


  1. Generate a tone (Generate menu)
  2. Ensure that the track is 32 bit float. If not, convert it to 32 bit float (see “Sample format”: http://manual.audacityteam.org/o/man/track_drop_down_menu.html#format)
  3. Apply the Amplify effect, enable the “allow clipping” option and set the “New peak amplitude” to 10 dB.
    Note that playing the tone will be very loud and distorted.
  4. Apply the Amplify effect and set the “New peak amplitude” to -3 dB (minus three).
    Note that the waveform is a nice smooth sine wave and will play as a clear undistorted tone.

Repeat the above but in step 2 set the track sample format to 16 bit.
Note that after step 4, the waveform has the tops and bottoms of the waveform missing and when played sounds very raspy.

The default sample format is set in “Edit menu > Preferences > Quality”. That should be set to 32 bit float.

File size.
32 bit float requires double the disk space of 16 bit.

Windows 10 will be a free upgrade for Windows 7 and Windows 8.1 users for a year after release of Windows 10.

So if you were to upgrade to Windows 7 now (if you can still find a machine that offers it) then you would only have to buy a new OS once.


OK - just a lack familiarity with terminology on my behalf.

Thank you for all the help on 32 bit float which is now set as my default which I trust doesn’t have any downside.