Question about Harmonic Match suggestions (major ↔ unrelated minor)
In djay playslist the colored Harmonic Match suggests combinations like:
– C major (8B) → D minor (7A)
– D major (10B) → E minor (9A)
– and so on
I’m trying to understand the reasoning behind these suggestions. I can’t place them within commonly used harmonic mixing theory, and I don’t see similar behavior in other DJ software. Also it doesn’t sound harmonic at all in my opinion..
Is this an intentional design choice based on a specific theoretical model, or could this indicate an issue in the matching logic?
I’m asking because I rely on Harmonic Match in practice and would like to better understand its behavior, or help improve it if needed.
No, it’s on ‘Exact’. But in fact changing this setting doesn’t change the result. In both cases D minor (7A) is shown. And also E minor (9A) is shown for C major (8B).
I think ultimately it should only ever be treat as a suggestion, there will be perfectly matched tracks that sound horrible together, and completely unmatched tracks that work.
Also bear in mind on average 25% (1 in every 4 tracks) are analysed incorrectly in DJ software, so it may not be right key at all and without a musical ear you wont be able to manually correct this. Then there is the fact keys will shift when you adjust the pitch (unless you have keylock on which impacts sound quality), and also producers can and will have different elements of their music in different keys, even within the same track.
TLDR, you have to rely on your ears ultimately if you want to be 100% sure something sounds right.
P.S. this is how all DJ software works with key matching as far as im aware, i dont fully understand it myself to know if it is right or wrong, but the ‘camelot wheel’ works like this, where the corresponding major and minor key are suggested.
Yes, I am aware of this and other resources, and I studied them thoroughly. That’s exactly why I raised this question.
I agree that key detection is not 100% accurate. However, that does not justify errors in the matching logic itself.
If imperfect analysis were a valid excuse for inconsistent matching, then the question becomes: what is the actual value of Harmonic Match as a feature?
It’s a bit like saying: “Some of the instruments in the cockpit are unreliable, but don’t worry, just fly on gut feeling.” In practice, instruments are still expected to be internally consistent, even if they’re not perfect.
So the issue here isn’t whether DJs should use their ears. Of course they should. The issue is whether the software’s matching logic is coherent and based on a clearly defined musical model.
Have they not just implemented the same system that all other DJ software have? and therefore shouldnt it be down to the person who invented the system to explain it if you think its wrong?
I dont think these people have anything to do with the logic that sits behind it at all, they just put in a system developed by someone else.
Obviously if the you have it set to ‘exact’ and its still suggesting what it considers as ‘fuzzy’ then there is a coding issue that needs to be resolved, but if you have it on fuzzy and its suggesting 7B as a compatible song to 6A then its just doing its job, correct or incorrect.
Edit: it seems that ‘fuzzy’ key matching and its introduction is what has muddied the waters for the camelot wheel, by suggesting the next track based on mood rather than a matched key. In whcih case its probably best to take this up with Phil Morse who devised it. It says in the guide below that its not suggesting the next track solely based on harmonics, but if it will theoretically increase or decrease the mood, which i would say is about as subjective as it gets, and really up to the DJ themselves to decide if that will work or not.
What about the Show Keys in Different Colors setting? Are you using:
Off
Color All Keys
Color Matching Keys
Color Matching/Adjacent Keys
Regarding the Match Setting, sorry, I forgot this is only used when actually Matching the Keys of 2 songs not when showing Matching Keys in the library
