Sorry, only 4/4 grid timing is supported within rekordbox currently.
I would like a way to accurately grid songs with various time signatures so the downbeats can always be accurate.
Sorry, only 4/4 grid timing is supported within rekordbox currently.
I also would like to be able to change time signature. A lot of Latin music use different time signatures. Is there anything in the works on this?
You can, in fact, accommodate complex and changing time signatures in Rekordbox. To do so, you will need to modify the rekordbox.xml file, and to do that, you will have to understand it's structure. See the following from Pioneer.
This is an incredibly powerful feature for anyone spinning many "world music" (how I hate that phrase) genres including Latin, Afro-beat, middle eastern, anything Asian. And much, much more.
As far as I know, no one else supports this feature. Not Serato, clearly not Traktor. Denon DJ Engine Prime may, through importing a Rekordbox library, but I have no way of knowing.
I don't understand why, instead of denying this feature's existence, Pioneer doesn't tout it, and upgrade their rather pathetic beat grid editor to something useful.
It frustrates me to no end that such a powerful platform is hobbled by weak editing software. A good developer worth his/her salt should be able to create a standalone editor, although this is not ideal, unless Pioneer agrees to continue to support the current structure.
If you want to play with this feature, pick a song that tends to slip or stuff beats. Gridmark it using the standard editor, and carefully note the spots where the ones try to get away from you. Setting cue points is an easy way to do that, since it will tell exactly where they are, and you can comment them with what needs to be done. Your notes will be right there in the XML that you are already editing.
I am away from home right now, in Colombia 😁 to listen to cumbia, pig out on unhealthy, delicious food, and otherwise just chill. When I have a chance I will post an example with before and after snippets, with a link to the underlying song.
Pioneer, if you're listening, I would be happy to produce a spec for an enhanced grid editor, with comments and justification for each feature change or addition. The current product, with a useful editor, would blow anything else out of the water in the world music sphere, and make life so much easier for anyone mixing music that gas not been put down over a drum machine or click track. And if we only mix stuff with a fixed tempo we'd never need anything beyond Traktor.
@James > This is true; with the XML editing you could create any kind of grid to suit your needs, however, as you pointed out, this is incredibly tedious and difficult, even for users who know what they're doing. A stand-alone visual editor would be amazing, but having used the XML files on large collections for testing purposes, I also know this is a very slow option when needing to bring that data back and forth.
I agree with you that there is much potential to make the grid editor more powerful, and considering a large percentage of the world's music is not 4/4, it would make sense to have the capability of handling it, especially when others are not.
It's more about the time signature than the variable nature of the beat; rekordbox's dynamic analysis mode can cope with imperfectly timed tempos, but a 3/3 or 3/4 signature is downright nasty for any DJ software to handle.
Your feedback is certainly appreciated and I'll be sure to include that in my notes to the product teams.
Thanks so much for your speedy respnse. Yes it is a royal PITA to do manually. I know, because I've done about fifty, and have about a hundred more in the queue. I need to do them, but somehow never "now".
As to the the notion that a standalone editor would not be able to effectively handle a very large collection I can tell you that I have built systems using XML to handle years of trading data, and if done properly even a massive collection should be able be handled without notcable lag on a computer capable of running Rekordbox in the first place. The more desirable option, to be sure, would be to incorporate the handling in the existing editor.
Rekordbox's handling of music created by live musicians is problamatic in any case, even with strictly 4/4. That is not a knock on Rekkrdbox. How can a program say definitively where the beat comes down when even the musicians playing can't agree. A couple features that would make the editor come to mind, simply because they have been chronic annoyances to me, as someone who can be a bit sloppy.
1 - Allow the user to limit changes to a specific range, with both a bigin marker (already there) and a user defined end marker.
2 - Allow a user to tweak just one mark, leaving the others alone (a special use case of one, above).
Anyone who has carfully gridded a difficult song, only to find some mess back at the beginning, would greatly appreciate those features. Another use case is tweaking a grid produced automatically.
I hate to sound arrogant (my wife might disagree) but I can't imagine that anyone managing a collection of live music (as opposed to music proced with a click track or, say, a Roland 808) would not tell you any different.
If a developer working for me told me that that would be difficult, or that they didn't add sufficient vale to be worth it, I would either jump in with them to make it happen or find another person for that role.
Let's say that those two features were added. All one would need is to be able to, through some user interface tweak that didn't get in the way of existing features, simply define the time signature at some point, as is already defined in the XML spec and respected by Rekordbox. Those three features would, for an admittedly limited set of users, create something that no one else has, and would make things much more attractive to DJs who, for reasons of practicality as much a the look of the thing, use vinyl and a standalone mixer without software.
And as long as I'm wishing, fix up smart sync so that two properly beat marked tracks stsy lock and I would be in heaven. I say that as a new DJ with far more ambition than talent whose interest is in mixing some of the more challenging material. I've tried accomplishing the same thing with Ableton to normalize my tracks, but that sounds to me way too much like autotune for rythm.
These things have obviously been on my mind for quite a while. As a retired software engineer and systems architect (and a really good one) and a DJ specializing in cumbia (and a god awful one) I'd be happy to kick in my time to work with you under NDA. If this is at all interesting, you have my email address.
Jim Siler (AKA DJ L. Viejo).
Again, thanks for your insight. I've created a ticket for our engineers to read your comments, and if they have any questions, we'll be in contact. Cheers!
Well over two years have gone by since this topic was started, and since the last comment (prior to this one, from Pulse). Other than the brief(ish) abandonment in Rekordbox 6 of the rekordbox.xml export and import (and thanks for those powers who reinstated this critical functionality) nothing has happened with the user interface to improve beat map editing. That's a shame, because the internal architecture of Rekordbox is incredibly powerful in this regard, but it is virtually impossible to utilize some of its most usefull features.
I understand only too well the difficulty of prioritizing features that are usefull to a relatively small number of users, but given the massive improvement of competing software (e.g., Engine Prime) and the fact that the heavy lifting was done ages ago when the architecture was developed and the rekordbox.xml spec was publicly released I am dissapointed that Pioneer DJ has chosen not to leverage the power of the architecture. Once again, I'll try to come up with a reasonable wish list.
1) Add an option to the Analysis Setting dialog to allow the user to specify "Normal" or "Dynamic" analysis on a one shot basis. This should be a trivial modification, have no effect on the current workflow, and be tremendously useful to anyone mixing old school and modern tracks (think reggae).
2) Supplement the start marker for the modification region with an end marker. It is incredibly frustrating to spend a long time editing a beat map only to discover that there are some issues early on and no way to fix them without messing up the entire edit.
2b) Allow an individual beat mark to be moved. This would go a long way towards resolving the problem I describe in 2, above.
3) Allow a new downbeat to be set going forward without changing the preceeding structure. This is especially useful in a lot of latin tracks (and others as well) that stuff extra beats or set a new downbeat. This occurs a lot, and not just in old school tracks. Many modern fixed tempo dance tracks do this, especially in latin music.
4) For those who need it, allow the user to specify the time signature going forward from a given beat. Probably not useful for most cases but incredibly useful if you need it.
I an earlier comment (March 20, 2019 14:22) Pulse says "It's more about the time signature than the variable nature of the beat; rekordbox's dynamic analysis mode can cope with imperfectly timed tempos, but a 3/3 or 3/4 signature is downright nasty for any DJ software to handle.' That's clearly true for automated analysis but Rekordbox already does handle variable time signatures, and quite well. I know this because of a number of tracks for which I've manually made the changes in rekordbox.xml.
Again, I know that it is difficult to prioritize less popular features, especially when they may be less than trivial to implement. Not knowing the internals of Rekordbox I can't say which of these requests (other than 1) is trivial and which are a bear. That said, given the facts that a) Rekordbox has the internal architecure to handle these things and b) that the competition is nipping at Rekordbox's heels doesn't it make sense to leverage these features and differentiate Rekordbox in a way that would be quite difficult for any product not built from the ground with a structurally aware engine to match?
These requests still sit with the rekordbox team, however I know that some are low priority, and as such, are left on the back burner until there are fewer priority issues to address.
I'll point out that item 3 is already in the software, even though it's not explicitly noted. See my video on gridding a transitional BPM track here.
The problem with a different time signature (other than 4/4) is also in sync, in determining how to synchronize the two tracks, not just for beat (it's easy to match BPMs), but also quantize and phrasing. These requests may be monumental in terms of engineering hours required to make it happen for very little demand (I think you and maybe a handful of other users are asking for it, but that's really it).
I'm sorry that I can't give you a more definitive answer or timeline.