emusonacid! legacy

I want to use my emu as a 'rompler' and use the test tones as basic waveforms and the samplers looping, envelopes and filters in a synth type way. How can I assign the test tones to samples/voices/presets? Thanks

I believe you gotta sample them then. I use my emu as a rompler mainly, using the e-synth card and other libraries.

does this mean i need to have 2 outputs going back in to the 2 inputs on the sampler, or can I sample them internally without having to use audio cables?

Well, first. You may not want to use the test tones from the emu. They are a bit noisey. I would get wave forms from Wavelab, as I did so many years ago.

You'll also want to sample them in mono. Which I mean you'll sample in stereo, but have only one input receiving signal. You want to unplug the cable from the emu not the from the source. A connection left un used can cause bio or environmental feedback. So you want nothing plugged into the emu that you don't intend on sampling. Once you have your stereo sample that happens to have only the right or left channel with a signal, you'll want to convert to mono. When you convert to mono, the sample will sound stereo. They will infact go through both speakers instead of just the left. You will have a cleaner image if you do it this way. You will also save polphony. And you will have stable filter responses.

Image. When sampling, and this can go towards all samples, not just test tones, if you take a stereo sample, both channels with signal and convert it to mono, what the sampler will do is sum the two channels together. Basicually it lays the left on right and vice versa. What will occur is destructive and constructive interference. More so that you will get destructive interference. The sample will lose it's high end due to phase cancelation in essence. I had a break that sounded muffled after converting to mono, so my sampling method is of this type. One signal into stereo mode, convert to mono to get a stereo sound. See?

Polphony. The emu counts stereo samples as two voices. So if you have poly up to 64, you can only have 32 stereo samples playing at once. But what happens with this occures is the next issue...

Filter stability. When you have a high poly count, more so with alot of stereo samples, the filters, more specifically the lpf, will cause strange distortion bursts. Loud snapping and popping. The stuff people the house over will hear. It's very bad. I am not quite sure what causes this, some kind of over load of polyphony or perhaps the lpf's curve (it bells at about 100hz, in other words, it actually gets louder at that frequency, whether that is because the various sine waves are combining as they become a signulairity, or that the filter was designed to have a slight Q response at the range is hard to tell) which when coupled with eight stereo voices may cause a quick spike in signal that causes the sampler to distort. But this phenomina is not isolated to just filter sweeps, it usually occures at the attack stage of the sound. Just something to avoid.

The emu also does something kind of odd. At 68 samples, I found that I could play all those test tones in a preset at once with each of them detuned slightly and I still had polphony left, yet I should only have 64 so I think there might be something else the emu is doing when samples are mono. I think it might drop voices as well. but I doubt you'll go so far as to create a 68 voice patch.

Any questions, just shout. I check this from time to time.

snh

sampleandhold wrote:You'll also want to sample them in mono. Which I mean you'll sample in stereo, but have only one input receiving signal. You want to unplug the cable from the emu not the from the source. A connection left un used can cause bio or environmental feedback. So you want nothing plugged into the emu that you don't intend on sampling. Once you have your stereo sample that happens to have only the right or left channel with a signal, you'll want to convert to mono. When you convert to mono, the sample will sound stereo. They will infact go through both speakers instead of just the left. You will have a cleaner image if you do it this way. You will also save polphony. And you will have stable filter responses.

Image. When sampling, and this can go towards all samples, not just test tones, if you take a stereo sample, both channels with signal and convert it to mono, what the sampler will do is sum the two channels together. Basicually it lays the left on right and vice versa. What will occur is destructive and constructive interference. More so that you will get destructive interference. The sample will lose it's high end due to phase cancelation in essence. I had a break that sounded muffled after converting to mono, so my sampling method is of this type. One signal into stereo mode, convert to mono to get a stereo sound. See?

Why not just select left or right as the input on create sample page? I do not see the point of sampling a silent channel and then converting to mono when you can just sample one channel which is mono anyway. Plus you do not have to go through all the hassle of unplugging cables etc.

Also for converting a stereo sample to mono:
1.Copy the sample so that you have 2 versions
2.Sample manage>utilities>erase (choose the left side on one and right on the other)
3.Compare and see which side you like best.

I always use this method for converting samples to mono, like sampleandhold said you lose quality when converting using the stereo to mono option in sample edit mode.

When you set the input on left or right like your suggesting, what you are in effect doing is the same as converting a stereo sample to mono. I have gotten the same results from sampling just the left side or right and taking a stereo sample and converting it. I still had massive loss to the highend due to phase cancellation doing stereo to mono conversion and just sampling the left side which seems to be just a way of bypassing the extra step of doing the conversion later. The cleanest conversions I have gotten where from pulling the plug. Plus, I think the emu only allows you to sample one side or both. I don't think it allows you to select the left or right. So, if you were sampling the amen break, the two sides are very different, so by pulling the plug, you can get either or.

My breaks are what caused me to start doing this activity. I couldn't figure out why I lost so much when I converted to mono or just sampled one side of the break. It's because both processes sum the channels and thus, you lose stuff. So by forcing the dead channel to sum with the active channel, you don't lose hardly anything because there is nothing there to cancel things out. Because of this, I no longer have to shelf my breaks and the details stay very clean and clear.

snh

I have to disagree, choosing left or right as the input is totally different to converting to mono in the sample edit tools. When you convert to mono it sums the two sides together. Choosing left/right or both in the create sample screen is totally different, its just sampling the selected side, no summing is going on.

Like a said previously you could also just sample a stereo version copy it over a couple of times and then goto sample manage>utils>erase(if the sample is stereo then it will give you the option of eraseing left/right or both sides). Erase the left on one version and the right on the other.

There is absolutely no difference in sound quality using any of these methods when comparing the results to sampling whilst unplugging one of the sample inputs, I have compared them.

Anyway I'm not here to start an argument, imo you're a valuable member of the community here, I'm just trying to point out an easier way of doing things.

We'll just have to agree to disagree.

What EOS are you running. I am running one of the older ones, so if you are running EOS 4.7 or something, perhaps the emu acts differently. I know I have tried multiple ways and only found the way I do it to work best, but I am running 4.1 or whatever it is.

snh

sampleandhold wrote:What EOS are you running. I am running one of the older ones, so if you are running EOS 4.7 or something, perhaps the emu acts differently. I know I have tried multiple ways and only found the way I do it to work best, but I am running 4.1 or whatever it is.

snh

E4XT Ultra running 4.7.

yeah, that is probably the difference. perhaps the summing I got from the older EOS is a glitch kind of thing they fixed in 4.7... Interesting.

snh

sampleandhold wrote:Well, first. You may not want to use the test tones from the emu. They are a bit noisey. I would get wave forms from Wavelab, as I did so many years ago.

You'll also want to sample them in mono. Which I mean you'll sample in stereo, but have only one input receiving signal. You want to unplug the cable from the emu not the from the source. A connection left un used can cause bio or environmental feedback. So you want nothing plugged into the emu that you don't intend on sampling. Once you have your stereo sample that happens to have only the right or left channel with a signal, you'll want to convert to mono. When you convert to mono, the sample will sound stereo. They will infact go through both speakers instead of just the left. You will have a cleaner image if you do it this way. You will also save polphony. And you will have stable filter responses.

I cannot thank you enough for this tip. This is working great.


Big ups man.