Quick Sign In:  

Forum: General Discussion

Topic: Virtual DJ 8 BPM issue - Page: 5

This part of topic is old and might contain outdated or incorrect information

VDJ RonPRO InfinityMember since 2010
Good stuff !
Thanks

I've got Ableton standard..but it's extremely dusty.
More or less all I've been interested in over the last year is video.
 

Posted Mon 01 May 17 @ 5:44 pm
groovindj wrote :
Andy7689 wrote :
old Stax and Motown songs
you are never going to be trying to beat match them

Actually it's not beyond reach!

As far as I know, no DJ software does it yet, but Ableton Live can "warp" old tracks like that and correct (straighten) the BPM, making it possible to beat mix them, add your own touches (a bit more percussion or a more modern beat) and so on.

It's a two stage process if you want to play your warped tracks in VDJ - correct them in Live first then add them to VDJ - but it's better than nothing.

Ideally it would be possible to do this in VDJ (now that would be a killer feature that would leave the competition standing) but we're not there yet.



Im pretty sure Maudio torq used to do this...but thats outta here. Edit well its something like warping or am i misunderstanding
 

Posted Mon 01 May 17 @ 6:57 pm
No, that's not it.

What the video shows is similar to the system VDJ has now, where you can add anchors which allow the CBG to run at a different BPM over certain sections.

In other words, the grid is what moves. In Ableton Live it's the waveform (the track itself) that is adjusted, to match the grid.

The result is like what would happen if you could adjust the pitch fader live, as the track played, to keep everything in time - with master tempo (key lock) switched on.

You don't / won't hear any key changes, but the slower sections of the track would be made faster, and the faster parts made slower, to keep it bang on beat.
 

Posted Mon 01 May 17 @ 7:16 pm
Cam i just add didi anyone use a time code in V8 and V7 I got mine in 2009 with V5 i upgraded to V7 because it said would sort the newly developed BPM problem out. I thought, why upgrade it worked perfectly, locked together so you could run the double mix perfectly also in the automatic choosen professional mode not reduced level on consummer equipment. While i play a live synth piano along a bit like running brothers and sisters with trible church remix joe inferno/ dye witness and record the sessions. But could never do. I could only play the one song and play piano.
I tried a doulble mix just like the technics 1210s, keep them in time manual. I am muscian and was a DJ Full Moon, Freedom, ritual, blackmoon, halfmoon Party DJ on Koh Samui 3 years Technics 1200... I move to CD decks pioneer MK2 and hey they locked perfect that left you free to make the mix night better. Heads up I was using software to make my CDs and used only the wav. format CD. The othe DJs where doing same, but ran wav song on one deck and MP3 or AFF on the other and even thou the BPM was same they would not sync. It was like talking to a monkey. You need a time code stripe similar to stripe the tape on you 4 track, so evrything when played, will be in time. Now obviously when i bought mine it was automaticlly programmed with the timecode. After all this time, because i belived in this procuct, every holiday and my spare time, ive tried to sort this problem out, and seems to be exploitation of the software, basiclly on a timer to get you to buy the upgrade. And now you have to go online and send all your mixes to them. OK if your not proffessional or even if your not, what idea you recored and forgot, someone has had a look to see if theres anything worthy of remixing. Proberly comes out on a CD and you think done that the other week?!!!HUmmm. Ok pro dj do the same and get big bucks and has sent everyt5hin to vdj don't think so. With V8 t they tell everybody and think its great to have that facillity but with other versions it set that same root out when you set up the software making sure you ate them cookies!! All other midi equipmenthas a is code is locked in when you lock this in on VDJ it changes so this is an unsable platform that needs to be stablised. As im lossing my stabillity sorting this out its a board decision and if bad practice is the game. I am one person with a lot of tecnh back ground obviously not enough to sort out. with this VDJ and alot of my ideas and inspirations have been lost because4 i have had (wanted) to try make this work. I paid for mine and it worked and i paid for the pro infinity licence which means i get the uptodate controller firmeware and driver from hercules & guillimote, and Virtual DJ software. Ive heared of ATOMIC mix but i think that is the problem. There putting software out that works on its own and other hardware there partners with. Tied now missed half the mat6ch typing this. Check out the mixes i did manual BPM. going t6o put the piano overlay later. DJ stevie b merseyside 90 remix 2015. Good luck it took me 8 years and alot of tears trying to sort this out.
 

Posted Mon 01 May 17 @ 7:22 pm
groovindj wrote :
No, that's not it.

What the video shows is similar to the system VDJ has now, where you can add anchors which allow the CBG to run at a different BPM over certain sections.

In other words, the grid is what moves. In Ableton Live it's the waveform (the track itself) that is adjusted, to match the grid.

The result is like what would happen if you could adjust the pitch fader live, as the track played, to keep everything in time - with master tempo (key lock) switched on.

You don't / won't hear any key changes, but the slower sections of the track would be made faster, and the faster parts made slower, to keep it bang on beat.


I actually thought it was warping. Does ableton bridge or link allow this.? If not...Im saying if VDJ did that it would be the topping on the cake.
 

Posted Mon 01 May 17 @ 8:09 pm
Djratedxxx919 wrote :
Does ableton bridge or link allow this.?


The Bridge is dead. People have been asking why/where it's gone for years, and at one point some Serato guys did say it would come back - but it never did.

Abelton's new Link seems to be much more popular. It's not tied to Serato like The Bridge was. It's freely available to developers, and lots of software is using it - including Serato DJ and Traktor.

The thing is, VDJ insists on being the master. Link has no master/slave - any app can adjust the tempo.

I've not tried Link yet, on anything. It'd be interesting to see how it works with Serato DJ and Traktor.

I suspect that tracks would need to be perfectly aligned to the grid, and old variable BPM tracks would still need warping first. Link won't magically warp tracks, in any of the DJ software. It's just a tempo sync tool.
 

Posted Mon 01 May 17 @ 8:31 pm
Have you tried converting all mp3s to wav in itunes i know apple. then as long ass you got bthe time code working they should all run at same speed specially. try with two tunes the same song same formate same sample rate, cue them up start manuallly synch and wait shoulkd be a little fluctuation. Then try locking together auto. If that dont sort it out make sure that it work when you bought it with the hardware or is it a download coz each time you upgrade you have to install the firmeware. Also i noticed the the address string goes into an unerscore then continues with the rest of the address, whats strange is i installed the software download and its atomic mix with a crack file. Were they not the hacking team that hacked the soaftware in the first place. I read that they are partners with virtual dj, well that answers the last post i sent theres the boardroom bad practice charge everyone as much as they can by breaking up the midi programming and charging. They probeley got the terms and conditions well sorted but that still leaves me with hours of waisted valuble time. Thanks so much vdj i thought there would have been some empathy, hay, there hackers MF with a parnership. Seems to be happening with everything and everyone..
 

Posted Mon 01 May 17 @ 10:04 pm
Think i better finish off with the solution. You buy the harware and software together making sure everthing does what you want it to do. Dont let them tell you you caqn sort it out in the editor control syntex code, I only have a Masters in music music technology not a PHD, with the problems you got this will at first ok i can do this, dont believe it. It will take all your spare time and more even pulling you hair out taking you to the end of your tether, There hacke3rs arnt they with a parnership. It can be sorted and i had v5 worked locked perfect thats what all the advertising was about at the time. So if they were nearly going out of business throu hackers why not join forces best thing they ever did but put bad practice in place so you carnt het an answer coz it take a real tech to know whats happening and with a few right prceedures and protocol everthing works thas what hercules have t5old me it shoul work. as i said idown loaded v8 and it had a crack file. Ive downloaded this from virtual dj site and it wasent a spoof page . All the new security i swich of to download and install as they just cause more problems. I think ill have look at tractor. They pob do the same as shit like that gets around the big companies they dont care how many hours they steel from yougsters that are curious about become a dj to be give a monumental task of sorting problems out with software and hardware firmware issues they just want to mix like onh the anolgue decks. Only they said this was a better system coz you can lock th edeck perfectly, yes they did i saw i saw i saw i saw what isaw what i saw. as a doble sometime 3 -4 remixer with years of DJing it would leave time to implinebt some more inbto the nights mix. I mean you start off with one start one finish as house dj bedroom mixer. But when your doulb mixing manual your working every beat all night yea thats a lot of consitration and crystal. Why cross over when you are wishing you could keep that middle bit cross over going all night. Well I did, but i always wanted to play the piano along on the spare of the moment as what come out is pure inspirational off the cuff and i wanted to record it. VDJ sort this out.
 

Posted Mon 01 May 17 @ 10:37 pm
SerinePRO InfinityMember since 2013
I don't know when you downloaded this but the most recent update hugely fixes all the problems, especially on newer tracks, the BPM and grid detection is superb
 

Posted Tue 09 May 17 @ 9:39 pm
Yeah, the BPM detection is a million times better now.
 

Posted Tue 09 May 17 @ 10:16 pm
lincol2PRO InfinityMember since 2011
Andy7689 wrote :
Yeah, the BPM detection is a million times better now.


Are you speaking of?:

 

Posted Wed 10 May 17 @ 1:28 am
SerinePRO InfinityMember since 2013
Yes, although most of the changes are not "experimental" in the latest release and have been made a part of the standard BPM analyzer.

Personally I always found the BPM detection superb, what was lacking was the first beat detection, either detecting beat 2,3,4 or missing the beat completely. And yeah, I think my library has gone from 25% bad detections to maybe 2% bad detections for modern tracks.
 

Posted Wed 10 May 17 @ 9:41 am
groovindj wrote :
I've just encountered a problem with VDJ 8 (3710) analyzing the BPM of Be My Lady by Kool & The Gang.

It's supposed to be around 125 BPM (and Mixed In Key gets it right) but VDJ's first attempt tells me it's 83.1 BPM.

If I switch to the new experimental analyzer it says 62.3 (which I guess is half the correct tempo at least).

Another example - Cosmic Lust by Mass Production. Again VDJ says it's 84.2 when it's actually around 128.

Anton Powers & Pixie Lott - Baby (VDJ says 81.3 and it should be 122).

[later]
Now this is strange......the above results were from my laptop (Windows 10 64 bit). On my older desktop PC (Windows 7 32 bit), VDJ gets them all right first time. Why should that happen?


This is still happening. Don't You Feel It by Sub Focus analyzes as 133 (wrong) on my laptop and 100 (right) on my desktop. Same build on both systems.

Interesting - Mixed In Key on my desktop also gets it wrong (133) where VDJ gets it right....



 

Posted Sat 27 May 17 @ 9:00 am
AdionPRO InfinityCTOMember since 2006
Looks like it's quite certain of 100bpm when experimentalBeatAnalyzer is on.
This algorithm will be default in the next update.

Note that if the algorithm is not completely certain which one is correct, there will be a slight edge toward the bpm that was already set earlier, so after changing the bpm manually to 100 it's possible that a re-analyze keeps that value.
 

Posted Sat 27 May 17 @ 9:27 am
SerinePRO InfinityMember since 2013
This seems to be having problems again. I think it is still better than it used to be but I am still getting tracks where the first beat is detected before the track starts or in a period of silence.

First beat before start of track:

10 - Sigala - Give Me Your Love (feat. John Newman & Nile Rodgers) (Andy C Remix).mp3 from RAM Drum & Bass Annual 2017 4m14.7s

Other more general bad detections:

Luca Debonaire - Walking on Clouds.m4a 03m43s

Shapeshifter NZ - Stars (Hugh Hardie Remix) (Hugh Hardie Remix).mp3 04m51s

Fred V & Grafix - Hurricanes (Wild Love) (Original).mp3 03m52s

S.P.Y - Love & Hate.mp3 05m38s
 

Posted Wed 16 May 18 @ 11:51 am
SerinePRO InfinityMember since 2013
RAMM300D-022-Frankee-Drop_It_Low.mp3 4m03s also detects two beats out of phase before the start of the track.
 

Posted Thu 17 May 18 @ 9:57 am
AdionPRO InfinityCTOMember since 2006
I've found the songs on Deezer, and for me

Sigala - Give Me Your Love (feat. John Newman & Nile Rodgers) (Andy C Remix)
Fred V & Grafix - Hurricanes (Wild Love)
S.P.Y - Love & Hate
look correctly detected

Frankee-Drop_It_Low
is one beat off

Luca Debonaire - Walking on Clouds
a little-bit off-beat

Shapeshifter NZ - Stars (Hugh Hardie Remix)
Indeed 1/4th beat off or so
 

Posted Thu 17 May 18 @ 11:03 am
SerinePRO InfinityMember since 2013
This is what I get for the first three. The first beat should be right at the first hint of waveform for all of these. Love & Hate isn't so bad (but is audibly off) and the other two are at least a beat out.





So maybe the question is more, why are songs which are presumably the same giving completely different analysis results?

Maybe I can email you my mp3 files - these three are all purchased directly from the record label.


I am using V8.2 b4291
 

Posted Wed 23 May 18 @ 11:03 am
SerinePRO InfinityMember since 2013
RAMMLP14D-017-D_Kay_and_Lee-Tuning_VIP.mp3

D Kay and Lee - Tuning VIP (05:35) from Andy C Nightlife 5

Is also analysing with the first beat detected off beat, before the file starts.
 

Posted Wed 30 May 18 @ 9:19 am
Here are some notes about fractional BPM values.

When you are looking at BPM values, you may have certain expectations what they should be for various reasons like:

o - someone or something told you what it should be
o - one or more programs says it this and since it met your expectation it must be right but you can't assume that.

I do BPM analysis off and on. Pain in the ass. I wanted to know some absolutes so I began using metronome files with no noise etc. and with just constant audio markers at specific locations. I downloaded some prepared files for this. My assumption was that these are computer generated files and should be accurate. They also should represent the easiest possible case for beat detection. With my own beat detection software I started to notice fractional files where they should not be which was a surprise. But then I noticed that the beat markers lined up with the audio values exactly when using the fractional values but not when using the expected integer BPM values. My thought was that these files must have had been generated incorrectly. Most likely due to some round off condition. I left it at that for awhile. Recently I looked it at more closely. There are several examples but chose an mp3 file that is supposed to be exactly 161 BPM. With my calculations and also VDJ the actual BPM value is 160.998 and lines up perfectly. Then I created my own 161 BPM wav file so I would know it would be exact. I used wav since I did not want any compression getting in the way. With my own code it is detected as exactly 161 BPM and it is exactly that. I have a lot of information to show this but trying to be somewhat short right now. Oddly VDJ detects this exact 161 file as 160.998. VDJ does a pretty good job usually compared to other programs I have tested. I can't consider any program I have tested to be good at BPM detection. That means a lot of things. It depends on what you consider good, how well it meets your expectation and other things. None of it means that it is right.

Some programs will detect whole number BPM values when it is not. That is easier than trying to come up with actual BPM values. Since those programs met your expectation you might assume they are correct and if a program came up with something different then it must be wrong. Again, you can't assume that. In my option, you must check every computer BPM generated value. In some cases it will be close. In some cases you might not agree. For me the only way I can tell is to see how well it lines up with the audio. Depending on how long the audio is you might not be able to even see it and so you have to be very careful. You also have to ask yourself how close is close enough. I don't know. You just can't assume any program gets it exactly right. I have found that to be true after testing many programs and a lot of code and comparing the results. I have files and images for further testing and comparison if any one is interested.

I spent just a little time to see why VDJ might come up with 160.998 for the exact 161 file. There are many ways to detect BPM and I have no idea what VDJ uses. I can only use my own software to come up with a guess. The higher the resolution you are trying to get, the more tricky it can be. Internally to the software, these high res values can become very close and make it difficult to come up with an exact number. I modified just one place in my code to see what would happen. A place that is almost insignificant. Then my own code also came up with 160.998 which is not correct for the exact 161 file. Changed that one little almost insignificant value back and again detecting exactly 161.
 

Posted Wed 06 Jun 18 @ 1:40 am
99%