Think of this... under the "sound" menu there could be an item called "shuffle blend" or "smooth shuffle" or something of that sort... it could be toggled just like the current "shuffle" feature is currently... below the menu entry there would appear only one text box - the current status of the switch; drilling down to it would toggle the status.

This feature, if enabled from the sound menu (above), could fade in and fade out the beginning and ending two seconds of tracks when the empeg is playing in shuffle mode.

i know that when i shuffle-play lots of music, invariably i have some live concerts in there with crowd noise (Dave Matthews at Soldier's Field & Pepsi Arena and the Floyd live at Fort Worth in '77 to mention just a few). These tracks all blend smoothly together when played in order, but the vast differences in crowd sounds makes shuffle play very herky-jerky in between tracks.

If the empeg could fade in and fade out of tracks while on shuffle mode, such situations would be averted. Maybe the next version of emplode could even make the duration of the fade period a user-configurable value. (1 to 10 seconds or whatever the person likes, really.)

If the fader was only used during shuffle play, it would be fine to keep the toggle buried as a sub-menu of "sound" since you wouldn't need to fuss with it a lot.

Any thoughts?


- Dixieland
MK1 : 6 GB : s/n 00246
MK2 : 16 GB : s/n 080000348

My problem with such an option is that certain songs would suck when faded-in. "Tom Sawyer" must be played at maximum volume from the first microsecond.

If you have specific songs that you want faded in/out, you can do that to the MP3 files themselves with MP3trim. I realize that wouldn't sound great if listening to a concert album sequentially, though.

It'd be neat if there were a check-box for each song saying "this is continuous-track material". Then it could fade just those songs when shuffling. That would be awesome.

___________
Tony Fabris

hehe... agreed. but you see, that's why people who had a lot of songs like "Tom Sawyer" or "Roadhouse Blues" could use the toggle function right on the empeg's "sound" menu. it's a win/win

(definitely need empeg at 0.00 dB or more for the start of certain songs)

heh, re-encoding the concert tracks would certainly suck... but then again.... manually adding two seconds of silence on the tracks that should never fade would remedy that issue nicely.


- Dixieland
MK1 : 6 GB : s/n 00246
MK2 : 16 GB : s/n 080000348

I second this idea as i have some tracks that fade out like ones frm the end of an album and then you shuffle over to something real loud and i have to change my under wear

heh, re-encoding the concert tracks would certainly suck...

MP3Trim will add fade in/outs without re-encoding the tracks. It does this by editing the global volume flag which is a part of each frame.

It can also normalize MP3s without re-encoding them. It's a pretty amazing little program.

___________
Tony Fabris

so would that lead me to believe that empeg Ltd does not see much need in the near future for users to have a "shuffle blend" function?

(see... the problem with editing the MP3s at all is that you'll fubar the sound of the concert when played normally - from start to finish. only a non-editing solution would allow users to listen to the tracks normally but also smooth the transitions on shuffle play)

- Dixieland
MK1 : 6 GB : s/n 00246
MK2 : 16 GB : s/n 080000348

so would that lead me to believe that empeg Ltd does not see much need in the near future for users to have a "shuffle blend" function?

I don't think anyone from empeg has said anything either way.

___________
Tony Fabris

i guess i'll just keep my fingers crossed.

- Dixieland
MK1 : 6 GB : s/n 00246
MK2 : 16 GB : s/n 080000348

As it happens Toby has been experimenting with something similar to but rather more advanced than this. It's really just a concept right now, and will certainly be post-1.1 (if it ever happens at all), but to drop a couple of hints he now has a reliable BPM and beat phase detection routine.



Rob

wow DJ in a box that will be cool. But where do we put the M&Ms

i have about 4500 songs on my empeg and no interest in editting them individually... a crossfade for root playlist shuffle would be a dream come true...

Is the BPM feature going to be called the Oakenfold switch? ;)

32GB (JUST UPGRADED!) Mk. II BLUE
Detroit, MI USA
www.PfeifferBeer.com

BPM counting? Cool!

I imagine Toby would be working on seamless crossfades across techno tracks as live DJs try to do. I'd personally just like the empeg to put up the BPM count somewhere on the display screen as some of the pro DJ equipment tries to do. That would be hugely useful to me for my own DJing. Unfortunately, when I've tried automatic BPM tools before, I've found them to only be reliable for techno. These things gets very, very confused on jazz music. Any chance Toby might be experimenting with jazz?

I believe Toby has looked at a wide range of music, and considered many aspects - such as middle passages, slow starts etc. Although the algorithm displays a real time approximation, it must "see" the whole track before offering a definitive BPM.

I'd try it with some Jazz, but Toby is on holiday and we compiled out his BPM stuff in his absense, because it scares us

I'm sure you'll see lots more of this feature soon.

Rob

IMHO... The only time a BPM count would be useful is when you're dealing with dance/techno music. And since the BPM counter works well with that kind of music, I don't see a reason to worry about whether or not it works for jazz...

___________
Tony Fabris

Well... maybe Toby has some ideas for music selection which go a little beyond what most of us are used to...

Rob

It's might be useful to think... could YOU crossfade these two tracks? Some of them just don't make sense. So perhaps if the BPM code couldn't find definite enough beats near the end of a track, it wouldn't crossfade at all. How do you beat match a dance track against some classical piece with lots of rubato?

You probably wouldn't want to do big long crossfades, either. I've heard some pretty dodgy examples of those done by humans. It might be a bit much to expect some software with a couple of meg of memory to handle it! :)

Hmmm, it could be very nice. If you drop largish chunks of samples (rather than dropping every nth sample) you could speed up the music without changing the pitch. Coooool! (Presumably this is what Toby has in mind.)

Richard.

If you drop largish chunks of samples (rather than dropping every nth sample) you could speed up the music without changing the pitch. Coooool!

As MP3 encoding is frequency-domain-based, could you get that effect even more simply, by just converting back to a different time-domain sample rate?

Peter

IMHO... The only time a BPM count would be useful is when you're dealing with dance/techno music. And since the BPM counter works well with that kind of music, I don't see a reason to worry about whether or not it works for jazz...

In my spare time, I spin jazz music for swing dancers. I count BPMs of everything I play and have it all written down on the side. This helps when you're DJing live and you decide "I want to slow things down from 180 to 140, now what have I got near 140, hmm..." This practice is quite common among swing DJs and is a regular topic of discussion on the SwingDJs mailing list.

Part of why I love the empeg is that I have it stuffed full of swing music, playing at random, and it will often help me "discover" gems hidden within my own collection. If it displayed an accurate BPM count, that would mean I wouldn't have to measure it for my notes but could just write down the number.

(And, if 1.1 finally allows for skipless playlist editing, then I could consider using the emepg for live DJing as well.)

Fascinating! You learn something new every day.

___________
Tony Fabris

As MP3 encoding is frequency-domain-based, could you get that effect even more simply, by just converting back to a different time-domain sample rate?

I don't think so. In that case you would have to play the samples at the new rate or your pitch will change. But our rate is fixed. What you maybe could do is figure out the sample rate you need, and then scale in the frequency domain before the conversion to compensate. It would seem that would be a quite subtle way of changing the speed. (At least, compared to just dropping samples!!). It also makes sense to slow things down with this method, which could be hard the other way, because you have to figure out where and which samples to add.

Or am I wrong?

Richard.