Connexion rapide:  

Forum: Wishes and new features

Sujet Dreaming of Fluid Beatgrid - Page: 1
chirroHome userMember since 2020
I sincerely believe that Virtual DJ is one of the best DJ programs available today. None of them are perfect, but they all should learn from the competition.

Recently, there have been two things that, in my view, have revolutionized the DJ world – what we call 'game changers.' One is the ability to work with Stems, and the other is the availability of beatgrid detection technology using artificial intelligence, which Algoriddim Djay refers to as 'Fluid Beatgrid.'

In terms of Stems, Virtual DJ is excellently positioned, but when it comes to beatgrid detection accuracy, it is one of the worst programs I know.

Since this is one of the most important things for a DJ, I believe you should focus more on it.

If Virtual DJ had a beatgrid detection algorithm as good as Fluid Beatgrid, you would be number one, without competition.

I hope to see this soon.

Best regards.
 

Posté Thu 07 Nov 24 @ 3:51 pm
Quite a few requests for this lately. Hers hoping :)
 

Posté Thu 07 Nov 24 @ 3:54 pm
Not just lately - for almost a year now, and that's only since Algoriddim's system.

Prior to that, for many years we've been asking for a smarter detection system for variable BPMs, or Ableton Live style warping (straightening out the tempo changes) which Algoriddim are also doing in their Neural Mix Pro software.

Manually setting anchors is so 1990s.
 

Posté Thu 07 Nov 24 @ 4:06 pm
I agree with the request and really hope that it arrives eventually but I really have to ask...how many modern songs have you really had to grid manually that this really stood out to you?

The only songs I have to do this for are older songs (which had bands playing live instruments) and occassionally transition tracks (which are kind of obsolete now with beatlock/auto bpm transition/stems usage.

I guess I'm really trying to understand how much the feature is actually benefitting DJs currently outside of the obvious. Based on what I play (which is mostly newer tracks with occasional older ones), I barely touch beatgrids manually, and if it's an older track, I'd be paying more attention with my ears.

I also hope they get it so that it is easily correctable/respects what is already done. I see lots of praises for it's presence in DJay but I also saw reports where it's very wrong/trampling what was already done.
 

Posté Thu 07 Nov 24 @ 4:20 pm
DJ VinylTouch wrote :
how many modern songs

Where did anyone say they needed this for modern songs?

Consider the big picture. There's a huge back catalogue of music that was recorded by humans actually playing instruments live. Before drum machines. Before sequencers or click tracks.

Even when drum machines and click tracks were being used, they weren't flawless, and things were still being recorded and mastered to tape - a mechanical process where fluctuations could occur.

Having a reliable accurate beat grid is important for any DJ that doesn't just play "modern music".
 

Posté Thu 07 Nov 24 @ 5:37 pm
I agree, but I ask:


  • How many of the DJs you know are actually playing these older songs as their mainstream set vs well quantized, DAW, produced songs?

    The ppl around that I know who use DJay and are commenting about the fluid beatgrids are mostly playing EDM tracks...there is no real gain from fluid beatgrids there. If your scene involves mostly older, unquantized tracks, I get it.

  • For the people who played the unquantized, older songs, what did you do before?

    I know ppl made warped tracks or beatmatched by ear. That kinda doesn't change, although you get welcomed assistance now


I think it's an excellent feature and I do want it too. But I'm pretty sure the regular club/lounge DJ doesn't see the immediate benefit, if they are playing current tracks. With old tracks I understand, but you can still work with them.

OP also claimed VirtualDJ has one of the worst BPM detection algorithms. Not trying to start an argument but I would want to see proof of that...I've used Serato, Traktor, VirtualDJ, Rekorbox and even DJay (before Fluid Beatgrids and after) with various styles of music and I've never really seen it do terribly on the detection with well quantized music compared to the others (I do have to move the grid a bit though sometimes).
 

Posté Thu 07 Nov 24 @ 6:58 pm
There are a lot more function and "wedding" based DJs using VDJ than you probably think.

In all fairness the the quantized edits from Mastermix and such sound absolutely horrible.

I currently just "chop" mix disco songs and the likes but a "fluid" beatgrid where you could choose two older tracks and have them iix seamlessly would be a bigger hoily grail than stems.
 

Posté Thu 07 Nov 24 @ 7:15 pm
I do weddings too lol. For the older tracks I would play, they are either already manually gridded correctly, (if they are more popular) a quantized edit and if it's really one off request, them I either drop one one or try to ride it (if not too terrible).

Fluid Beatgrids by itself is just the track playing (which VirtualDJ can already do). Fluid beatgrids + sync is similar to a quantized edit that's happening in realtime, with the master deck tempo serving as the project BPM. It can sound the same, better or worse depending on the fluctuations of the tempo of the master deck.

I get that the feature us useful...I use DJay too and would readily welcome/use the feature here too. I don't think I agree with the stems comparison, but I do respect that fluid beatgridding matters that much to you. I just don't think it's the end of the world the way the beatgridding exists currently, but I guess I don't use the sync functionality enough to feel that strongly about it.
 

Posté Thu 07 Nov 24 @ 7:33 pm
Yes I agree it's not the end of the world as it stands however as we know anything to make life easier would be more than welcome, and as this feature alread exists in other software surely it should be on the list of VDJ features for the future.
 

Posté Thu 07 Nov 24 @ 8:02 pm
I agree...it would be a welcome addition. Hopefully the devs make it a reality in the near term.
 

Posté Thu 07 Nov 24 @ 8:56 pm
chirroHome userMember since 2020
"There are a lot more function and "wedding" based DJs using VDJ than you probably think.

In all fairness the quantized edits from Mastermix and such sound absolutely horrible.

I currently just "chop" mix disco songs and the likes but a "fluid" beatgrid where you could choose two older tracks and have them mix seamlessly would be a bigger holly grail than stems."


Completely agree with this.
 

Posté Fri 08 Nov 24 @ 7:49 am
What would set any forthcoming VDJ system apart from the competition, is the ability to choose between having the grid follow the track (djay Pro style) or the track follow the grid (Ableton Live style).

That way, you can either keep the loose feel of the track, or you can tighten it in line with modern tracks.

Maybe something by the end of the year as a festive gift, Atomix?
 

Posté Fri 08 Nov 24 @ 11:10 am
May I ask a question please ?

You have a track with fluctuating BPM (let's say between 120 and 128 BPM) and a perfectly steady track of 126 BPM that you want to mix together.
And you don't want to do it the "easy" way (with a short mix or by using a short loop on the fluctuating track) but you want to mix them together for 32 beats or even more..

The question is simple:
What should the software do ?
1) Try to keep the fluctuating track "steady" at 126 BPM ? That would produce the "unwanted" quantized audio you described above, that really a lot of people don't want.
2) Try to match the "steady" song with the fluctuating one ? In other words, it should "fluctuate" a perfectly stable song ? Won't that introduce another kind of unwanted audio ?

I genuinely want an answer on this question, and most importantly WHY the software should behave like this.

It's easy to forget, but you're mixing TWO songs together, you're not playing just one song at a given (steady or not) BPM.
 

Posté Fri 08 Nov 24 @ 2:20 pm
Answer: It depends whick deck is set to "master".

i.e.
If the fluctuating track is in charge, then the steady track is synced to that.
If the steady track is in charge, then the fluctuating track is synced to that.

Obviously there is going to be some audio manipulation happening, but surely no more than if you should attempt either manually?
 

Posté Fri 08 Nov 24 @ 2:43 pm
And obviously not for tracks with huge multiple variable BPM differences either, more for older tracks that have a steady enough beat but just drift a bit.
 

Posté Fri 08 Nov 24 @ 3:07 pm
@PhantomDeejay this is exactly what I was getting at when I said this:

Quote :

Fluid Beatgrids by itself is just the track playing (which VirtualDJ can already do). Fluid beatgrids + sync is similar to a quantized edit that's happening in realtime, with the master deck tempo serving as the project BPM. It can sound the same, better or worse depending on the fluctuations of the tempo of the master deck.


I mean I understand it's value, but I can see a few downfalls, especially if the master deck is playing a variable BPM track. It is a natural progression of multi bpm analysis though, so I definitely would vote for it being done...I just wonder if there is a new type of trainwreck being created/enabled here (weird sounding/warped mixes) from its (ab)use with sync in certain cases.
 

Posté Fri 08 Nov 24 @ 3:15 pm
You guys seem to be assuming that if it's done by the software, it will sound worse than if the DJ does it live. How so?
 

Posté Fri 08 Nov 24 @ 3:18 pm
groovindj wrote :
Obviously there is going to be some audio manipulation happening, but surely no more than if you should attempt either manually?

Now, would you manually attempt to beatsync these two theoretical songs in an area with many/big fluctuations ?

kradcliffe wrote :
And obviously not for tracks with huge multiple variable BPM differences either, more for older tracks that have a steady enough beat but just drift a bit.


Well, if you introduce a function, you have no control over how users are going to use it. So you must assume they will use it not only on cases with a small drift, but on cases with large fluctuations as well..

I'm trying to understand where and how this "fluid" thing would be used, while at the same time I'm trying to make a point.

For the moment I'm leaving "small drifts" outside the question, because let's say that if we decide what we want to do with big ones, small ones are the easy case..
 

Posté Fri 08 Nov 24 @ 3:20 pm
groovindj wrote :
You guys seem to be assuming that if it's done by the software, it will sound worse than if the DJ does it live. How so?

I'm actually saying that a DJ live, would not either TRY to beatmix in areas where there's no stable tempo.

If we strip all the technology away and we return on 2 turntables and mixing "by ear", a DJ would have to consider that the two areas he's about to mix, have steady tempo. Yes, the tempo on the area he's about to do his mix is not the same as it was on the beginning of the track, or yes, the beat may have drifted position, but it doesn't really matter. But what DOES matter, is that the next 8/16/32 beats provide a stable enough tempo to mix with the next song.

So, mixing a fluctuating part of a song with a steady part of another, is NOT something a DJ would do, and I'm not sure if we want a software to do either..

But I'm open to suggestions..
 

Posté Fri 08 Nov 24 @ 3:28 pm
PhantomDeejay wrote :
mixing a fluctuating part of a song with a steady part of another, is NOT something a DJ would do


You're making this decision for all of us?!

It's something that DJs have had no choice but to do in years gone by. If you have an old track and a new track (or two old tracks), then you have to do it - by whatever means necessary.

As Keith says, we're referring to tracks that have a slowly drifting BPM, not ones that change suddenly from one BPM to another.

Having a fluid beatgrid type system would make it easier.

George, have you tried using the Algoriddim system?

 

Posté Fri 08 Nov 24 @ 3:44 pm
63%