emusonacid! legacy

One technique I have developed over the past year or so with my favorite source, the 'Sstart, is the ability to "cut up" difficult samples. I found that certian samples, like guitar riffs or instrumental loops can be very difficult to cut up.

So I came up with the idea to set up the Mod Wheel or whatever cc I wanted to use with the 'Sstart and draw in the cc values in my sequencer. This, however, is a bit time consuming and also makes it difficult to do stutter type effects you could do with a cut up break. But it yeilds very good results for your basic resequecing and as such, I avoid timestreching the sample.

So... what if you could save time and not have to cut up a break but just use the above cord? That is the original idea behind this new developement. Oh.. and it failed. You would have to draw in all the cc's and such. You could spend just as many hours drawing CC's in as you would cutting the break up and it would also not sound as tight as the cut up break due to the restricted resolution of the CC values. When you cut up a break you can adjust upto 9 samples. This cord is dependant on the lenght of the sample of it's resolution. If you have a long sample, 1 value can be 100 samples instead of just 3.

But, in my experiments to find a quicker alternative I did find a surprising short cut. It may not help with the breaks but it will for sure help when it comes to tweaking the hell out of a guitar riff.

I actually don't remember how I came across this, it doesn't actually make sense. I suppose I just ran through all the different sources or destinations until I found something that worked. And so comes the title of this thread.

Gates. That is what worked. By setting the matrix mod like this:

Gate>'Sstart>XX%

I found that I could start the sample at any given point. At 50 percent the sample started in the middle. At 25 percent, the sample started a fourth in.

See, what's funny about this is that our difinition of a Gate wouldn't really yeild to a control fundemental that was exposed the other night. I would never have dreamed that Gate would have been the choice. It's just fucking weird.

Basicually, what you do to get the quick and easy edit type control over a guitar riff or whatever is to treat it like a cut up break. You copy the same voice over and over again. Maintain the same sample, however, place it on different keys, just like you would for a piece of the break. Then go in and set the percentage to the desired value. What is so so cool about this is, if you need to make an adjustment you don't have to fight with the CC in the sequencer. Just go into the offending piece and adjust the percentage until it sounds good.

There might even be a way to increase resolution response with the multiple sources... maybe. I haven't tired it yet.

Hope this helps you guys out. It allows me to speed up a bassline in a remix I am doing by basically playing the rhythm of the notes but set each sample to play at the start of the note in the sequence. So so nice.

snh

Hey, SNH, how much do you love talking about 'sstart - probably as much as I do about realtime crossfading We have chatted briefly before about the pros and cons of offsetting in cords here - viewtopic.php?t=492 - crikey, almost 2 years ago

Nice discovery about the gate as a more accurate source then using a CC. You will get exactly the same results by putting DC in as a modulation source. The 'start offset' in the 'Voice Modifiers' page is completely sample accurate but your way has the advantage of being easier to visulaize as it is displayed as a percentage.

Get a break or any sample come to think of it, make sure it doesnt swing (straight 16ths) and make sure it has a regular length (4beats, 8 beats etc) then try the below.

06 MidA C07Amt +100%
07 DC Quantize +0%
08 Quantize 'SStart +100

If you trigger the voice every 16th or 8th notes you'll get the idea. Play around with all the values, especially the MidA to C07Amt. The quantize will cut the sample up into 16ths, or less if you adjust the MidA C07 amount.

I emailed Emu about the 'SStart not being sample accurate in the cords but guess what, I never recieved a reply. Luckily we do have a completely sample accurate offset in the 'Voice Modifiers' page. It's a shame because of they could have fixed the cords bug and added much more resolution then we would have something that would have kicked arse. However, we still have a pretty powerful feature, it's just that it needs a bit of bodging to get the right results.

Andrew

I find it quite easy and swift to cut up breaks using 'sstart in the voice modifiers page, as illinformed notes. However, I find using envelopes to control the length of the "cut" can be (relatively) time-consuming. Is there anyway of using something like the Gate modifier to set an envelopes hold time? I just want to hit a note and have a cut play back, without having to use the release phase of the envelope.

I've been thinking about ways of doing this, and was going to experiment with Gates. However, i always took Gate to be similar to gates in synthesis. From what you are saying this appears not to be the case. Could you clarify the way Gates work in EOS?

Mark

sixtysixnorth wrote:I just want to hit a note and have a cut play back, without having to use the release phase of the envelope.

Not sure if I know what you mean but this might help. I usually let the sequencer control where the note will cut but there is a nice way using a midi CC. In the 'Amp Envelope' Screen change Dcy1 to rate 127 and level% 0 (the default for a new voice is rate 0 and level% 99). Then in cords assign MidiA (or whatever) to VEnvDcy by +81% (any higher and it doesnt work). Now when you play the break you can control the length via MidiA.

Gate is a Source that gives a value when the key is pressed on. A way to show this is to get a synth noise with a looped end. Set the Rls1 to a rate of 71 in the 'Amp Envelope' screen so you get a couple of seconds of the sample when you let go of the key. Now in the Cords assign Gate to Pitch at -38%. You will find that the Pitch plays roughly an octave below when you hit the key but goes back to normal when you release.
I like to have 2 voices stacked - V1 has it's volume turned down to nothing but Gate makes it louder, and V2 at a normal volume but with Gate making it quiet. When you hit the key, you hear the on sound of V1 but the release sound of V2 when you let go.

What I'm really trying to do is make a sound last for a specific length of time before it enters the release phase of the envelope, regardless of the lenght of the trigger note. Then i could cut a snare from a loop, have it play the hit then enter a short release to smooth the end off a little, without having to be too accurate about how long i hit the key for.

I thought the Gate source might help with this, but now I don't think so.

...sounds great and very interesting, but...i didn?t understand why you (all) don?t work with Steinberg`s "Recycled" or Logic Audio "StripSilence...with EXS24 and Touchtracks ". I mean you spend so much time with drum Programing ...and this tools are much faster.


For me is the Emu a "Pad-String Maschine" ( Fl?chen-Maschine ). Here the Emu plays its strengths out , in combination with the whole Chords and Voice Matrix.

I mean specially in Live Act , " Ableton " do fundamatel works on this subjekt.
In all this Combinations ( Ablon/ReCy/LAM etc )you can realize your drum (s)- Fill?s- Break etc quasi in Realtime. You can drink you beer much earlier at the bar as the others.

Please let me listen to what you do with you Drum Work ( Music)

Michael

I perfer to cut my breaks up by hand. See, if you read the response I did in the knowledge section in a thread called "New to Breaks" you will see that I am actually performing some somewhat complicated stuff to maintain drum fidelity. Like reversing parts to maintain length of the sample while move stuff around so it fits in the grid better. That kind of thing. Also, cutting a break up by hand allows me to hear the breaks make up. Sort of like looking at dna of a life form, I learn what makes up the break and as such I learn how I can manipulate the break and come up with natural or not so natural sounding rhythms. I also have a slight photographic memory so when I cut a break up into it's 50 plus pieces, I will actualy, to a degree, memorize all the parts and as such select that parts that I want to use. I have actually made breaks by jumping all around the length of the break and mix and matching various parts that are very far from each other.

I also don't trust machines to do things to my specifications. I cut breaks and instead of trying to figure out the correct thresold I want for the machine to cut the break, I do it myself using several tests per cut to make sure that the actual start of the hit starts at 0 instead of a few samples later. I found that I started to hear the latencey in breaks if they where cut with spaces ahead of the hit. See the thing is, if you play a hit at origin, it may sound tight and all, but if you play it several octaves down from origin (this is one of my tests) you will actually find it's 10 to 30 samples late. And when you have a break with cuts that all have these spaces a head of the hits, the break sounds muddy or slow because everything is falling slightly behind the grid. So you get a drug-y kind of feel to the break. I like mine tight. Oh... and I also gadged how sensitive my ears are to delays, on my cd players I use to mix. I found I can hear a delay (where the machine cues the song a little bit ahead of it's actual starting point) upto 2/75ths of a second.

I am a perfectionist. I also have a rule that when I cut breaks up, each piece must be with in 0.01 seconds of each other. One can be 0.12, the next can be 0.11 or 0.13 and so on, anything higher then that is unacceptable.

That is me anyway. I have actually thought about using recycle... but it just creeps me out for some reason. I am afriad I wont get to know the break personally and it will do stupid cuts. But if it works for you, then great. I suppose it does do well. I just don't trust machines that much. I have turned just about every auto function off on my emu. The tapering and turnicating to normalizing. I have had the emu do too many stupid things that caused me alot of rework.

Illinformed.. I will mess about with your cords and see what I can get. It looks good.

snh

I'm in agreement with SNH there.

Ableton is great 'out of the box'. I've messed with it a load of times and it's very easy to get some complex beats quite quickly, it's a great jamming tool. However, when I want to fine tune the breaks to something a bit more like the idea I have in my brain, then it's not as fast as I am on the Emu. The reason for this is that I have already done most of the hard work in the Emu.

I have lot's of template Break/Beat Presets for my emu. Bass drums, snares, percussion bit's etc all go in the same place. The first 30 voices of my Preset Templates are dedicated to beats. The first 6 voices, C0 to F0 = bass drums, next 6 from F#0 are snares etc. Believe it or not I use an rm1x for my beat programming. In the Rm1x, I have a load of template midi sequnces that corresond to my template Presets. It's as easy as turning a dial in realtime to play the next beat sequence. I've got another fader setup to change the Presets on the Emu (via sysex because it sounds much smoother than crossfading).

My favourite setup is one of the Amen Break (I know it's overused but I still love it). I've got a midi fader that is assigned to Preset change on midi channel 1. I've also got 60 Amen Break Presets that it will swap between. Some are pitched differently, some samples are put in the wrong place (ie snare where bass drum should be). Here's a silly example me changing Presets via a sysex sweep click here The rm1x was only playing 10 midi notes each bar. The track is very silly and it was for a mates ring tone - it's his voice lol. It's very easy to get squarepusher/aphex programmed beats like this.

The only time consuming bit is chopping up an original sample and I use an ancient version of soundforge on the PC to do this. By now I probably come across about 1 original break per week so it isn't all that time consuming. The problem is that if I were to start from scratch then it would be very time consuming.

Lastly, play a break in Ableton, now play the same one in ther Emu - I know which one sounds better in my ears

Thank you for answering.
It?s very interesting for me, completly an other Thinking.
I Think i have to argue more with that Stuff. ( Drum Programming with/in Emu ).

So , so i said , for me is the Emu more a SurfacesModularSythem. The most Drum s comes from the EXS24 in Logic. the rest is completely the Emu. Listen here

Michael