Instant Recording Software Tests

How to analyze recording software like sequencers and other multitrack applications without expensive testing equipment

To avoid bad surprises you should do this simple investigation.
Do it, before you decide to start your project with a certain soft- and hardware combination. It doesn't take long and reliably discovers the most severe problems immediately. Test before you buy!

Now we have our stuff together and can finally start with...

A. Basic technical tests of recording software:
(I would recommend to print this page and take it to your music system. You can use it as a checklist and mark every item, when it passed)

1. Audio Timing:

Every recording system has some delay.

While it's insignificant in the analog domain, the digital world has some bigger problems, also called latency. As some manufacturers don't like to talk about it, you should always check it out before you buy.

Here is a pretty simple and accurate test, that takes only a few minutes (I tested a Pro-Tools MIX system with 888/24 Interface, but you can do it on any other system).

  1. Make yourself a Test-Sample:

    With your Sound Editing program or directly in your Multitrack Recording Programs Editor, make a short sample which consist of silence. Then Zoom in to Sample level around the middle of this sample and draw a single peak with the pencil tool. Ideally, this peak should be only one sample long. Amplitude should be around 80% of the max. Amplitude.
     
  2. Before you play it, turn down the volume of your speakers. You should hear a short tick.

  3. Make a song with about 120 bpm and place the sample into an audio track as quarter notes (about 4 bars long) like this:



    This track serves as a sound source to record other tracks. It makes tick-tick-tick-tick...etc. in quarter notes. We call it Klick-Track for now.

  4. Record a second track through an internal bus

    If your recording software is equipped with internal busses, you can make this test. Route the output of our Klick-Track to an internal bus. Then take this bus as an input for recording another audio track below the first one.

  5. Record a third track through the digital inputs

    If you hardware has digital inputs you should do this test. Route the output of our Klick-Track to a digital output, connect this output via cable with a digital input an record another track using this input.

  6. Record a forth track trough the analogue inputs

    Do the same as in the step above but this time you use the analogue inputs and outputs.

    Now it should look somehow like this:


     
  7. Copy the forth track and cut it randomly (only cut it...do not move the pieces).

    This should test, if the timing stays the same, when you cut a track in pieces. You might consider this a unneccesary test, but there was very well known recording software around that had this problem. As soon as you cutted a sequence it moved the right part to the next midi-position instead of leaving it where he was.

  8. Zoom into Sample level and look, if the peaks of the tracks are at the exactly same position.

    If zooming is not possible this way with your recording software, mix and record (or bounce) tracks and look on the mixed track, if the peaks are at the same sample position.

    I do not show the cut track here, because it should look the same as the one before (In this case the "Analog I/O 81 samples Delay").


    (extreme Zoom)

  9. Check for delay. The tracks should be in sync. In the digital domain that means, that delay should not be greater than one millisecond (about 44 samples).

    If you look at the recording over the analog I/Os, you see that the delay is about 2 milliseconds. That's not too much but If you work with tight, percussive Sounds it can be a problem.
    • If the delay is random, something's totally wrong.

      Check first, that your systems word-clock runs on internal sync and if the answer is yes, ask the manufacturer what's going on.
    • When the delay is constant, but bigger than 1 millisecond, look into the manual of your soft- or hardware to find some information about tweaking the timing. It's often in the hardware setup or driver setup portion of a program, named playback delay, latency or similar names.

  10. If you cannot fix it that way, ask the customer support of this software. If they cannot help you, it becomes cumbersome.

  11. The last things you can try are:
    • to get help from a forum.
    • ask other users, if they noticed the same problem.
    • check for a driver update and try it.
    • try another hardware.
    • skip this recording software and test another one.
    • ignore or live with it (not recommended).


2. MIDI Timing:

Notice, that Midi-Timing can't be as exact as Audio tracks. It's said, that delays of 1 millisecond and up can be a problem for percussive signals.

  1. Load the sample into your sampler and make it play.
     
  2. Make a Midi-track with about 120 bpm and quarter notes, that plays that sample.
     
  3. Record the sampler output on an audio track trough the analogue inputs (if available).
     
  4. Record another track through the digital inputs (if available).
     
  5. Go through steps 5-12 like in the test for audio timing above.
     
  6. Decide for yourself if you can live with it.
     

3. "How many tracks?" - Check:

Don't be shy. This is exactly the right time for a crash test of how many tracks your recording software can record/play simultaneously.

  1. Create a lot of tracks (let's say 32).
     
  2. Record enable them all.
     
  3. Try to record.
     
  4. Listen to the hard-disk.
     
  5. You should hear a rhythmic sound like
    rrrrrrr---rrrrrrr---rrrrrrr---rrrrrrr---
    where the space between the rrrrrr is silence.
     
  6. When this silence is short, you know you are near the limit of this hard-disk.
     
  7. When the rrrrrr is short and plenty of silence in between you know you can add some more tracks.
     
  8. Do that, until it quits.
     
  9. Notice how it quits.
     
    • Best - It tells you a warning before you can even start.
       
    • Not as good - It records a bit and stops with a warning.
       
    • Bad - It stops and crashes the program.
       
    • Worst - It stops by crashing the computer.
       
    • No Comment - It crashes and the audio files you recorded up to that point are lost too.

       
  10. Try to improve the performance by trying to give more buffer memory to the hardware drivers.

    11.  
  11. Make the disk full. You should test, how the system reacts when the disc is nearly or completely full. How does it stop recording?
     
  12. Try how many tracks can be played back. Copy tracks and try to play them back, until your recording software quits.
     
  13. Try how many tracks can be played back. Copy tracks and try to play them back, until the application quits.
  14. Try to improve that with playback buffer settings.
     
  15. Now cut and paste a lot. Cut out little pieces over many tracks and copy/repeat them until your screen is cluttered with regions.

    16.  
  16. Try to play that and notice again, when it quits.
  17. Now you should have a good idea of the real record/playback limits of your system, regarding hard-disk, audio hardware and driver performance.


4. Plug-In Crash test:

This one is short and easy.

  1. Take a plug-in that needs a lot of processing power and insert it in some of the previously recorded tracks (It's the same amount of processing power if the track contains silence or some signal).
     
  2. Do that until it quits.
     
  3. Check again (like in 9 above) how it quits.
     
  4. Play with inserting different Plug ins until quitting.
  5. You probably know by now, how much Processing Power for Plug-ins your system approximately has.

B. Basic handling tests of recording software:

Beside technical performance you should also test the look and feel of a software. If you don't like the handling, making music can become arduous work.

There are certain things, that should fit, because you will use them over and over. So I made kind of a checklist to test that:

  • Midi-Setup
    • Check how you can build your studio setup.
    • How easy is it to get midi in and recorded?
    • How can you select a midi device on a certain port (the output of the midi interface)?

  • Audio Setup
    • Does the software support all the hardware you need now and in the future?
    • Do the drivers for your hardware really work, without having to tweak a lot?
    • Can you change the different inputs and outputs, word-clock settings, sample rate and sync settings from the programs surface, or do you have to dig into the hardware setup all the time?

  • Recording
    • Just try to record a midi and an audio track. How easy is it to do?
    • Is it easy to record enable the track(s)?
    • How many actions do you need until you can actually record? (notice that every unnecessary step multiplies with the number of takes).
    • Check for features you might need like cycle or loop record (every pass makes a separate track), destructive takes (new take erases the old one), punch-in / punch-out, etc...
    • Look how these features are implemented in this recording software and how you like that.

  • Arrangement
    • Make one track, listen to it and find a position, where you want to cut it.
    • Try to spot the exact location with all the tools you can get (zoom, scrubbing, wave-form overview, cursor line, etc...) Was that fast and easy to do?
    • Try to cut out a piece and move it around. Do you like the way to grab it, move it, drop it on a grid, change that grid, etc...).
    • Record more tracks, cut the whole song and try to make a short arrangement. Is that easy to do, or do have missing note-offs, wrong control-pedal, or other unwanted artifacts...).
    • When cutting audio, do you sometimes get a click at the beginning and end, or can you easily prevent that by automatic cross-fades?

  • Automation
    • How many steps do you need, to draw a volume or controller-curve in your automation? (you would be surprised about some recording software).
    • Draw a fade out with your automation in the middle of a track. Play it and then jump to a location before that fade. Does the automation follow correctly or keeps the track volume down?
    • Can you automate all the plug-ins you like and does their automation really work (check every parameter you need automated).
    • How easy is it to record, suspend and tweak automation?
    • How much can you automate without CPU overload?

  • File handling
    • Where are your files stored? Is it easy to setup, keep track and change?
    • How secure are your files? Can you accidentally delete them?
    • Can you export, import, format convert, delete all the files you want?
    • Can you easily name and find your files?

  • Templates, Libraries, Building Blocks
    • Most recording software also works with building blocks (loops or one-shot sounds that you paste into the arrangement). If this is important to you, make a few tests:
    • Does this recording software provide the style of templates, loops or arrangements you like?
    • How easy and flexible is it to arrange and mix these blocks?
    • If this recording software provides pitch shifting and time stretching, listen how it sounds with different loops.
       
    • Is it easy to change the starting point and the timing of the loops?
    • Can you customize templates and create your own, or just use what's there?

  • Musical Intelligence, Styles, Automatic Accompaniment
    • What's left to entertainer-keyboards at the moment will probably become a topic for recording software in the future as processing power grows...
    • Look for algorithms and styles that sound good to you.
    • Can you easily build songs and record, edit, transpose, copy and paste the chords of a song?
    • Is the timing O.K.?
    • Can you make your own styles and is that easy to do?
    • How is the Import / Export of styles?
       

By now you should have a good idea of the technical matters and the overall handling of the tested recording software. You know how it performs and how it matches your working-style.

If you like it you can go for it or keep on testing other recording software. You will notice that if you do that a second time, it will be much faster.

Good luck, and find the right one !