Black Friday has come and gone. And with it, some of the biggest deals on TVs. Whether you were in the market for a new flat screen or not, you know the experience… You got the TV store and browse through aisles upon aisles of TVs. There are smart TVs, curved TVs, 4K TVs, LCD, LED, OLED… So many options! How do you compare them all? And what does shopping for TVs have to do with your music?
One of the biggest problems when comparing two TVs is that it’s difficult to know if you’re making a fair comparison. Were both TVs calibrated correctly? Are they being fed identical source signals? Are you being influenced by the brand or sales person? The same types of problems exist when mixing and mastering…
Consider that when you are mastering, you will typically have a mastering chain that consists of compression, limiting and EQ. When you tweak one of these elements, how can you be sure you’ve improved the master? More specifically, how can you put yourself in the best position to make an objective comparison?
Loudness Deception
One critical factor to be aware of when making tweaks to plugins in your mix or master is “loudness deception”. It is very easy to have your judgement thrown off by 1 - 2 dB differences in loudness. What sounds better may just be louder. So, when deciding whether your tweak is an improvement or a regression, it is important to loudness-match the “before” and “after” versions. That is, you should make sure the tweaked version has the same loudness as the original.
How can you do this? You could bounce both versions of the track or master, normalize them and then compare them, but that would be extremely tedious. A quicker approach would be to loudness-match by ear using a gain plugin. (There are likely some other manual approaches, but none that I’m aware of that are fast and accurate.)
The lack of a good solution was actually the inspiration behind Perception, the plugin that we co-created with Ian Shepherd. It lets you instantly loudness-match and bypass an entire plugin chain. NUGEN’s MasterCheck has similar functionality.
Blind Testing
For important mixing and mastering decisions, try performing a “blind” A/B test, or even a series of blind tests. This helps to eliminate confirmation bias: the tendency to seek out evidence for things you believe to be true while ignoring counter points. For example, if you believe that a particular plugin will improve your mix, perhaps because you paid a lot of money for it 
, you may be more likely to seek out evidence that it is better while ignoring evidence to the contrary.
Pepsi Challenge 1983 Commercial
You could get a friend to help you with blind testing, or check out HOFA’s 4U+ BlindTest plugin, which will assist you in comparing multiple tracks by randomizing their order and asking you to rank them.
Repeating a blind test several times and tallying the results will give you a measure of confidence. If you run 10 tests, and prefer version “A” over version “B” nine out of ten times, you can be reasonably confident that there is a noticeable difference. If you only liked version “A” six out of ten times, it’s unlikely that you can tell the difference.
Memory
Finally, recognize that A/B testing audio relies on memory, since you must listen to one source and then the other, in series. To reduce the likelihood of this throwing off your judgement, try to reduce the time it takes for you to switch from one source to another. Several of the plugins mentioned earlier will help you instantly bypass an entire plugin chain, for example.
Perspective
While all of these tools and techniques are great, don’t become bogged down with process to the point that you become robotic and uninspired. Mastering is an art and a science. With that said, I encourage you to experiment with eliminating loudness deception, with blind A/B testing, and being aware of potential biases.
What else do you need to watch out for when A/B testing? Do you have any tips, techniques or tools?
