Entra:     

Forum: General Discussion

Topic: Song Spectral Balancing
I'm a live sound engineer for a high-profile wedding reception band and a beginner DJ.

At my gig over the weekend, I was using VDJ in AutoMix mode to play a 20-minute playlist while the band was on break. I wanted to hear the transitions thru the PA system. The playlist had some rather old music from the 1970's and mixed alone with current music. After listening to the entire playlist, I noticed that the spectral balance between a few songs were quite drastic. I was wondering if VDJ had anything in its algorithms that could partially correct the spectral balance of a few of these older songs?

Unfortunately, I did not have Soothe3 VST3 plugin installed on my laptop that I was using on this gig or I would have used it.
 
Inviato Mon 07 Sep 26 @ 2:53 pm
Tbh (just my opinion, the @devs and others may have a different viewpoint), this is out of scope for DJ software.

The point of DJ software is to be as true to the original sound as possible, and only provide the modifications needed to blend music together for continuous playback, live, and as quickly/efficiently as possible and only when explicitly used by the DJ. The requested tuning requires precision and time, and would, in most live DJing circumstances, not be discernable by the audience unless the track is horribly mastered relative to other songs you playback at the event (then you should get/make a new version).

Your best bet is probably a VST that does the job, or doing the adjustment in a DAW, exporting the result, then using that result for DJing.
 
Thanks for that info. Since I’m new to DJing I wasn’t 100% sure how it’s treated only what I heard thru the 10,000w PA.
 
This is one reason why so many old tracks are available in X number of remastered versions. At the time of release they were mastered to suit the typical playback equipment of that period.

Ideally a DJ would maintain their library and ensure the tracks sounded well balanced beforehand, but in a live scenario then EQ would be adjusted on-the-fly as required, via the three band EQ. Even using stem separation, tracks can be rebalanced in ways previously not possible.

If there's no DJ in control, then the results were to be expected. 🤐
 
Thanks—this is close to the workflow I’ve been considering.

My thought is to analyze and spectrally balance older tracks beforehand, then save reception-ready versions rather than depend on broad three-band EQ adjustments during the performance.

I’m not trying to make older recordings sound modern or erase their character. The objective is simply to reduce distracting differences in low-frequency extension, tonal density and high-frequency balance when older and newer productions are played consecutively through the same PA.

Live EQ would still be available for venue-dependent correction, but the detailed per-song work would already be completed.
 
There is also the question of monitoring accuracy. If the DJ is positioned onstage or behind the main loudspeakers, how can they accurately determine the spectral balance being experienced throughout the dance floor and venue?

Unless the DJ can move continuously around the room, broad three-band EQ changes are being made from a listening position that may not represent the audience’s experience.

That is another reason I’m considering detailed per-song preparation separately from venue-dependent system EQ. The PA should be tuned for the room, while known spectral differences between recordings can be addressed beforehand.
 
Besides the audio engineer also ensuring that the audio system is properly EQ'ed for the venue:


  • The songs being used should be vetted/prepared before reaching the venue.
  • The DJ should be equipped with a good pair of headphones (preferrably with flat frequency response) and probably be backed by on stage/booth monitors to aid in doing adjustments live, before it is played out to the audience and during.


All these are what makes DJing a profession and not just a hobby.
However, it's not the job of the DJ to ensure the spectral balance is in order to that level of detail - that is the job of the mixing and mastering engineers. If there is a problem with the older master, then source a resmastered copy, or get a lossless copy of the master copy, remaster accordingly yourself in a DAW, then export and use that. You should not need to make aggressive adjustments at the live DJ playback level (although some adjustments may be required), come performance time - you most likely would never get it perfect at the event itself, and is more indicative of a badly prepared song.
 
Basically I'd say you need to know your system. Know what the main speakers sound like, both from the front and from the back. Use a small monitor speaker to make up for lack of highs and mids when behind the system, or wear headphones or IEMs that you also know well.

That way you will learn that when it sounds like X in your ears, then it sounds like X from the dancefloor too.

Over the decades I've done many tweaks to adjust songs to my preference. Mr Brightside is notoriously squawky, so I've got my own mix. Dancing Queen likewise, but it also has a great bassline if you can pull it forward.
 
Thanks for the information and perspectives.

I’m still learning where that fine line falls between the responsibilities of the DJ and those of the audio engineer—especially when preparing music from different eras for playback through a modern PA system.
 
FYI, I genuinely wish I had taken an interest in DJing years ago. Many of its concepts and principles would have benefited both my musicianship and my work as an audio engineer. I’m discovering that these disciplines overlap far more than I ever realized.
 
Tbh, a lot of DJs that I've encountered are not actively aware of music to that level (they barely even pay attention to proper gain staging) and are much more focused on emulating what they see their favorite DJ doing (scratching and effect techniques) and using (software or hardware).

You are in a good position to understand the reason why certain things are done, rather just doing them because you saw someone else do it and it was cool - just always keep the why in mind, even if it's a response give by a member of the forum or the devs.
 
Absolutely. Understanding the “why” has always been central to how I approach music and audio.

Thanks for the perspective.
 
Just my 2 cents:

People expect to hear an old popular track, the way they are used hearing it for the last 30/40/50 years.
There's no point IMHO to try to mess around with the sound coloring of old popular tracks.
It will be either too much effort for too little gain (there's only some eq-ing that can be done before the track "sounds/feels" different) or it will be too drastic and people will dislike it without (usually) knowing why.

As a sound engineer you can mess around a little more with the dynamics. Not on an individual (per track) level, but globally.
Most likely you have/use a digital mixing desk (Yamaha, DiGiCo, Avid, Allen & Heath, e.t.c.) and all those desks offer a basic gate/compressor/limiter per source input.
Therefore if you wish, you can use a compressor there to keep dynamics a little closer as people don't usually notice dynamics alteration as much as coloring.
Yes, they will notice heavy compression, but the margins are much bigger than eq-ing.

So, for me:
Yes, to using a light compression just to tight up the mix a little.
No, to messing around with EQ on a (per track) basis as it's pointless.

PS: Or just don't mix old with new songs one after another..
When I need to have a playlist play by it's own, I tend to place songs "gradually" in the list.
Yes, the list may contain a track from the 70s and a track from the 90's, but between them I will try to squeeze a track from the 80s so that the difference is not as drastic when going from one song to another.
 
Izotope Ozone is excellent at bringing a dull track to life...(pre processed)

You might even run it as a vst in VDJ??
But latency would likely be an issue
 
I've used ozone as a VST in VDJ.

Using just maximiser in low latency mode (one of the other settings also increases latency, can't remember which), and it works fine. You just might notice the phase difference if you have the master and pfl playing in headphones together.