I know I've asked a couple questions already but I need a bit of help and am trying again to see if anyone can suggest what to do.

I have two samples, one at 112 and another at 142, both from the same song.

I'm alright at getting the samples recorded on a pattern with the right number of measures and the correct BPM, but when I try to play them together the drums are either too fast or too slow for one of them, seeing as the machine makes me keep the pattern select BPM at one set speed.

Is there a way to make two different BPMs play as I recorded them or is this just a wall I have to hit? It seems impossible that I can't play whatever I want considering.

Please help. Much appreciated.

Which sp model are you using?

SP-404sx, sorry forgot to mention.

For what you're trying to do, chances are the answer's no (this time). You'll have to timestretch and chop one loop to match the other.

Be aware of how tempos relate to one another. Halve a tempo & you've got something twice as long. Double the tempo & now it's half as long. Simple maths, really (90bpm becomes 45 or 180). The problem this time is that 112 & 142 bpm don't have much in common. Stick to the double/half rule & don't vary by more than a couple of BPM & you should be OK.

Ignore the rest of this if you don't want to get technical:

As 142/112=1.27 (rounded off), try looking at it the 112 sample as being roughly 4/5 as long as the 142 sample. If you can, chop into quarter note lenghts then get the 142bpm sample looping & try tapping out the fresh chopped 112bpm stuff in time to it. Chances are you'll end up having some mad swing on the 112 chops compared to the 142 sample. There's no reason why you can't do this in reverse.

In future you may also want to mess around with polyrhythms. That's if you continue to work with out of time loops. This is where two different measured phrases play against one another & eventually fall back in time. A lot of technically minded metal bands & IDM producers use them. To make a simple polyrhythm in a DAW, make a simple 1 bar drum pattern in 4/4. Make another simple 1 bar loop in 5/4 (without changing the master tempo). Start the 2 patterns at the same time & copy-paste them over & over (from where the loop ends) until you see that they match up. With 4/4 & 5/4 times it'll take 20 bars for them to match. Hit play on your DAW, notice how they seem to fall further & further out of time then start getting back in time? Its a really interesting technique, but not suitable for all genres.

If you're super patient, try hitting each loop simultaneously & listen to what happens. Eventually they'll match up & start over again. Be careful though, you might be listening a while.

Having only used the SP-606 I know I can quickly sequence polyrhythms into it, not so sure about the 404s (OG & SX) without a bit of work & patience.

Wow, thanks so much. Even though I'll have to learn a bit more to understand the technical side of it it's good to know there are legitimate reasons. Insanely interesting about the polyrhythms, I think I'll mess around with those bright and early tomorrow.

I'm not too sure how the BPM changes in the song ("Deception" off of Birth of the Cool) don't seem to interrupt the flow of it at all on the record, the change is seamless, considering that I thought I could easily replicate it on the 404sx.

You'd think it would be easy to just record patterns separately from one another altogether regardless of each other's time signature. If it sounds wonky it should be the producers fault, I don't think it should be the machine's job to try and fix it. Maybe if I put the sample that's playing at a different rate in another bank it would work?

Anyway, I'll definitely try working with it a bit more before committing it to a pad.

Again, wholehearted thanks. I just started with these and really appreciate the help.

Glad you found that useful! Ah, knew I should've mentioned that polyrhythms also get used a bit in jazz! Should've also mentioned that polyrhythms sound great when applied to a single instrument.

Just listened to that Miles Davis track then, are you referring to around the 1:54 mark when the lead switches from trumpet to trombone? That's not actually changing tempo, timing or anything else much more than the lead player. The reason why it still flows when the leads change is because the whole band's playing the same scale. Davis ends on an octave (I think that was it) & the new player comes in with the scale's root (an octave below Davis' last note). Theoretically speaking, doing this will always be pleasing to the ear.

Listen to the hihat pattern (save for a few fills early on & how the drummer accents the hits) it barely changes through the song. It's all in 8/4 (same as 4/4 but faster) with the drummer playing in triplets to hold it all down. What does change is the length of the motifs they're riffing around. No big deal there, but it does sound good. Keeps things interesting for the listener & makes a fairly simple set of rules sound more complex.

You can do stuff like that with any instrument. Just stop working in 2, 4 or 8 bar blocks on everything. Try making a pattern that goes for 5 bars, switch the melody around & play that for 9 bars & then use a 3 bar bridge/fill before going to an 8 bar outro... BAM! You've just made minature a jazz odyssey!

When the rhythm instruments like bass & drums change what they're doing and you need to change what you're counting to when listening to a song, you've just heard a change in time signature (not always BPM). Not much music deliberately and noticably changes BPM. When it does it tends to be with live performers, but that's all to how the band interacts with one another as opposed to being a composition trick.

If any jazz fiends read this & can call bullshit on what I've just said, please do it! I need to brush on some theory.

No, that wasn't the exact spot, but I can totally see what you mean by listening to it. I've actually gone back and just tried to make some beats to get a grip with the machine without samples and just by reading what you wrote and working to understand how the signatures work has helped tenfold. Can't thank you enough! Now I'm getting places.

Cool, yeah understanding (at least) the basics of how music's made is pretty much essential to making beats that are consistently great. Which is a bit of an obvious thing to say but it sounds like loads of people buy samplers to make music without considering how to make music. I know I did that back in the day. Took months of dicking around on the 606 to start understanding why some of my chops sounded great when others sucked.

Have a go at making tracks for the beat battles here, that'll really help you out in the long run. Having a point of comparison to what others made with the same source is a real eye opener. Can't recommend them enough, even if you don't submit anything for a round or two grab the sample & flip it as hard as you can.

Damn, totally missed this bit the first time. You're right 100% about this. You can sample two phrases & have them match just fine in terms of duration but the tempo, time sig & scale will probably be out, meaning the two phrases won't sit right. Same goes with any combo of those four elements.

The producer needs to make the samples work together, thankfuly we've got a bunch of in the box tools to make that work (chopping, resampling, sequencing). If a beat sucks its often because the producer picked the wrong sample for the job or they didn't follow what the sample is doing in relation to their beat, that includes considering the sample's original context. Finding the right sample can be turned into an art hence why so many awesome producers have immense music collections. But sample finding doesn't have to be an art if you don't want to it to be.

On reflection your orignal question & this discussion was probably best asked in the production techniques section. Hope you don't mind if I get a mod to move it there. Thanks for reading my lecture length answers too. Cheers!

Interesting topic. I think some samples have tempos that will match in their original way when it complements the rhythm. Half the tempo, twice the tempo that kind of stuff. Its the same deal with finding cool echo rhythms that just fit in. There is some logic involved of course but i would try out as much as possible, even with limited time stretching. I have often used a slicer fx to redefine the tempo and while it certainly doesn't always work, it gives a good alternative to trying to beat match more or less incompatible stuff.