Synths do not always sound what the interface claims

[xoxos]

i'll (reiterate and) build on aciddose's point..

if an envelope has a release time of 400ms, chances are you're thinking this means 400ms to 0 amplitude. and, the retailer is probably happy to let you think that.

easy, if it's linear. if it's exponential, then it's 400ms to.... -80dB? -120dB?? who can say...


"it would be great" if we could cultivate public awareness of *how* this funky stuff works, so *you'd know* that about envelopes.

say your waveform looks like the envelope is shorter, try expanding the vertical view! normalise the tail to 0dB, normalise again, and again, until you can see the bit depth, and you'll find the end of the envelope is way past the vertical resolution in pixels of your waveform display.



to create an "engaging" parallel, we are not the first generation whose memetic environment is greatly composed of advertisements, and the inflated sense of expectation we have from commercial promises. it is like, we have inbred ourselves to select partners who are really good at sex. their value as parents is immaterial, we just want something that appeals to our cocks.

and the public sucks it right up, to the point where advocates who attempt to edify consumers are ridiculed and disenfranchised by "parties with interests in maintaining the state of commercial appeal instead of consumer discretion". you've grown up with the lies, you want the lies. reality is valueless. tell me sweet "little" lies (oh ye littles) like they do it in the factory or balls to you. you can't even move erudition unless you learn to act like a liar to make it seem appealing.


just sell me something that does it all.

it's my right as a consumer.

i know it exists, because advertisements have shown me that a synth that does it all exists.

i don't have time to learn synthesis, that's whay buying synths is for, to do it for me!

[cron]

AFAIK time measurements in synths/effects (e.g. compression) usually refer to 'rates of change' rather than absolute time values - e.g. (x)dB per (x)ms. It'd sound kind of odd if a very quiet note took the exact same time to fade out as a very loud one - it makes more sense that the rate at which the volume changes remains constant.

[spacekid]

Samplers are the same way. I want to use FL Studio's built in (simple) sampler module. But I can never get it to pass through a pure sample. The ADSR doesn't seem to work fully...and messing with everything else it offers only gets me 80% back to the original sound. --Not always bad results, but I like to start from the original recording---...so I'm looking for something simple to load a sample/maybe two at a time blended into at the moment, for use in FL Studio..sadly, most that fit the description have locked sample sets or one I found uses SFZ, and taking the time to crack the locked format is just a last resort if I can't find anything else. I did it to one player earlier today, and it wouldn't take too much time to do more in bulk for it, but I'd rather not unless I give up the search. That one displays pictures with the samples also, a nice touch

[Mutant]

If you do sound design by listening you will not always get the desired result. Sometimes you need to have highly accurate values.

Which times would those be, then?

100% Matching a sound from a different synth for example.

You're asserting that that's not possible unless you have calibrated parameter values?

I am asserting that it is way faster to see 100ms on the target synth and set 100ms on the cloning synth (1 look, 1 LMB hold and drag), than to use a wave editor or a real time frozen wave display plugin to do the same (after some trial and error and Satan knows how many clicks).

[whyterabbyt]

If you do sound design by listening you will not always get the desired result. Sometimes you need to have highly accurate values.

Which times would those be, then?

100% Matching a sound from a different synth for example.

You're asserting that that's not possible unless you have calibrated parameter values?

I am asserting that it is way faster to see 100ms on the target synth and set 100ms on the cloning synth (1 look, 1 LMB hold and drag), than to use a wave editor or a real time frozen wave display plugin to do the same (after some trial and error and Satan knows how many clicks).

So you dont 'need' them then, which is what I suspected. And they'd only be of relevance when attempting to '100% match' a sound from a different synth where it actually has calibrated parameters, where the timing is consistent (let your analog synth warm up, Mrs Jones) where the parameters on both synths have the same slope/response curves, and are calibrating the exact same thing (as per aciddose's comments).

Carry on then.

[fmr]

This is like saying you only want "pure water"...

[Mutant]

So you dont 'need' them then

I often copy sounds between synths and i don't even attempt it if the synths are too different, don't assume things about me.

If you really want your synths to have strange and large differences between what you set and what you get, go to the dev forum and make a general request.

[exmatproton]

So you dont 'need' them then

I often copy sounds between synths, don't assume things about me.

If you really want your synths to have strange and large differences between what you set and what you get, go to the dev forum and make a general request.

There is a difference between what we want (good read-outs and the same value display as what the signal does), and what a dev has done in a synth.
Also, in a lot of cases, making sounds, playing synths and trying to "recreate" a patch from another synth, is much more fun when you have to work to able to get there (IMHO(!!!)).

Sure, i'd rather see a synth with good read-outs. On the other hand, it is part of character of a synth to do stuff the way it does. I really don't care about inaccurate ms's display, as long as clockspeed timings are right. a 1/4 speed, must be 1/4 speed. Other then that, just make some sounds!

[Mutant]

My posts in this thread were an indirect reply to what the OP said about his experiences with Soundblaster and ES2.
I understand up to 10% error, but not more than 50%.

[whyterabbyt]

So you dont 'need' them then

I often copy sounds between synths and i don't even attempt it if the synths are too different, don't assume things about me.

What exactly are you claiming that Im assuming about you?

You do know that someone claimed it was needed, dont you? And you do know that 'you' can be used generically, dont you?

https://en.wikipedia.org/wiki/Generic_you

I think you've confused yourself somewhat; I asked someone else to tell me why something was needed; your decision to answer that means that you were underwriting that 'need', so dont complain when the response continues on that exact same path, especially when you assume Im talking about you specifically despite not having made any such indication.
Its somewhat ironic that its your assumption that confused you here.

If you really want your synths to have strange and large differences[ between what you set and what you get

Which would, of course be absolutely nothing to do with what we're actually talking about, which is the claim made that calibrated parameters are is the only valid methods of comparison. So could you stick to the point please, instead of introducing random strawmen? Cheers.

[kinwie]

Just found out some synths I frequently use do not do/sound as the interface claims. Probably doesn't matter to everybody, but if you are into sound-design then it definitely matters. For this post it is about volume ADSR envelopes only.

SoundBlaster:

• I set an attack of 1000 ms. I measure it and it is 927 ms (linear). A difference but no big deal;
• I set a release of 1000 ms. I measure 419 ms (exponential). This is just a lot.
• I set an LFO rate of 1.00 Hz. I measure 1.05 Hz.

I assume you are talking about creative soundcard, right? Thus what you mean is soundfont?
As far as my experience goes, your best option to replicate the same result of your sf2 source is sfz with Plogue sforzando, 99.9% accurate! You will only miss the chorus and reverb settings cause that not available in sforzando by default, but you can add the included mda Detune and Ambience to replace them if you want to.
Otherwise, stick to the the sf2 format and these 2 players can play them almost accurately as SB cards: Cakewalk sfz+ and Props NN-XT

[Dúnedain]

Just found out some synths I frequently use do not do/sound as the interface claims. Probably doesn't matter to everybody, but if you are into sound-design then it definitely matters. For this post it is about volume ADSR envelopes only.

SoundBlaster:

• I set an attack of 1000 ms. I measure it and it is 927 ms (linear). A difference but no big deal;
• I set a release of 1000 ms. I measure 419 ms (exponential). This is just a lot.
• I set an LFO rate of 1.00 Hz. I measure 1.05 Hz.

I assume you are talking about creative soundcard, right? Thus what you mean is soundfont?
As far as my experience goes, your best option to replicate the same result of your sf2 source is sfz with Plogue sforzando, 99.9% accurate! You will only miss the chorus and reverb settings cause that not available in sforzando by default, but you can add the included mda Detune and Ambience to replace them if you want to.
Otherwise, stick to the the sf2 format and these 2 players can play them almost accurately as SB cards: Cakewalk sfz+ and Props NN-XT

Yes thank you, but I run Logix Pro X so I'll use Falcon for this job I think. But when working on it it is good to know the values you see in Vienna are not always accurate. This probably explains why some other recreations by me using EXS24 didn't sound as close as I wanted.

[.jon]

Samplers are the same way. I want to use FL Studio's built in (simple) sampler module. But I can never get it to pass through a pure sample. The ADSR doesn't seem to work fully...and messing with everything else it offers only gets me 80% back to the original sound. --Not always bad results, but I like to start from the original recording---...so I'm looking for something simple to load a sample/maybe two at a time blended into at the moment, for use in FL Studio..sadly, most that fit the description have locked sample sets or one I found uses SFZ, and taking the time to crack the locked format is just a last resort if I can't find anything else. I did it to one player earlier today, and it wouldn't take too much time to do more in bulk for it, but I'd rather not unless I give up the search. That one displays pictures with the samples also, a nice touch

Have you tried Grace?

https://www.onesmallclue.com/plugin/grace/

[kinwie]

Yes thank you, but I run Logix Pro X so I'll use Falcon for this job I think. But when working on it it is good to know the values you see in Vienna are not always accurate. This probably explains why some other recreations by me using EXS24 didn't sound as close as I wanted.

So you can run Vienna in Mac OS???
Just kiddinggg
Btw, if you need a newer sf2 editor, there is Polyphone :
http://polyphone-soundfonts.com/en/
It's free and available for Win, Mac and Linux

Good luck with Falcon, btw

[spacekid]

Samplers are the same way. I want to use FL Studio's built in (simple) sampler module. But I can never get it to pass through a pure sample. The ADSR doesn't seem to work fully...and messing with everything else it offers only gets me 80% back to the original sound. --Not always bad results, but I like to start from the original recording---...so I'm looking for something simple to load a sample/maybe two at a time blended into at the moment, for use in FL Studio..sadly, most that fit the description have locked sample sets or one I found uses SFZ, and taking the time to crack the locked format is just a last resort if I can't find anything else. I did it to one player earlier today, and it wouldn't take too much time to do more in bulk for it, but I'd rather not unless I give up the search. That one displays pictures with the samples also, a nice touch

Have you tried Grace?

https://www.onesmallclue.com/plugin/grace/

Yep, I like it. It's in my 'good' pile that I'm narrowing down. I'm still ideally looking for one with maybe 1 to 4 pad assignments per kit. something really simple, but Grace is a good one.