KVR Audio

ATN69 · Post by **ATN69** » Fri Jun 28, 2019 7:21 am

Aliasing, filter characteristics, etc. I bet The Beatles, Rolling stones, and others, didn't have these kind of huge problems that would have prevented them from making and recording all their great songs

DJ Warmonger · Post by **DJ Warmonger** » Fri Jun 28, 2019 7:27 am

ATN69 wrote: Fri Jun 28, 2019 7:21 am Aliasing, filter characteristics, etc. I bet The Beatles, Rolling stones, and others, didn't have these kind of huge problems that would have prevented them from making and recording all their great songs

They weren't into synths afaik and they didn't use digital recordings at all - so aliasing didn't apply.

Also, nowdays these recordings sound very dated.

sadowickproduction · Post by **sadowickproduction** » Fri Jun 28, 2019 7:27 am

EvilDragon wrote: Fri Jun 28, 2019 7:18 am
sadowickproduction wrote: Fri Jun 28, 2019 7:11 amDude, you're a moderator on the NI forums so what you say is very suspect. The oscillators alias with frequency sweeps close to nyquist and reflect consistent to a poorly optimized band-limiting filter. Its not "beefing up" the sound.

Serum and Phase Plant do not alias like this. Address that.
Again, use the Filter knob if you want to have a pure sine. The image is shown above. There is no aliasing in frequency sweeps close to Nyquist then (if you could even call that aliasing rather than FFT noise when input changes too fast for the analyzer to react).

Serum and Phase Plant don't sound interesting (not to me at least) due to their overly clinical nature. If you want that, by all means, use them instead of MX.

Anyone can take an aliased sinewave at 500hz and filter it. What are you even saying? be objective NI employee. Also tell the team to do their job.

ghettosynth · Post by **ghettosynth** » Fri Jun 28, 2019 7:29 am

This is the aliasing that I observe, and it is at -134 dB. You can do this yourself, I'm not going to record a video but ED is right, the distortion at low frequencies does not fold back. In fact, if you roll back the filter at low frequencies these (aliasing) frequencies appear (at the same -134 dB level). At higher frequencies the osc filter frequency has no impact on the aliasing that you see here.

So there you go, that's my test, I'm not a moderator anywhere. I might be doing something wrong, feel free to challenge me on that, but bring your own game if you're going to do so.

Here you go, this is at a lower frequency with the (OSC) filter rolled all the way down. I even went to the trouble of doing finger gymnastics to capture the measurement on the aliasing component.

And here's with the filter open. Note, there is distortion of the sin wave, but that is not aliasing and the aliased component is now much much lower.

EvilDragon · Post by **EvilDragon** » Fri Jun 28, 2019 7:31 am

Thanks ghetto, cutting right through the bullshit.

Sadowick is blowing this way out of proportion.

ATN69 · Post by **ATN69** » Fri Jun 28, 2019 7:33 am

Well, these so called dated sounding recordings has become classics for a reason and there is also a reason why developers try to make modern synth sound dated in the name of emulation. Dated is still popular obviously. My point however was that people seems very fixated on measurements and analyzing sound instead of using their ears. Massive X sounds nice in my opinion but I know it's subjective and it might not be for everyone.

Tj Shredder · Post by **Tj Shredder** » Fri Jun 28, 2019 7:35 am

sadowickproduction wrote: Fri Jun 28, 2019 7:27 am Anyone can take an aliased sinewave at 500hz and filter it. What are you even saying? be objective NI employee. Also tell the team to do their job.

Sine waves do not alias at all below nyquist. If you want to test aliasing use something with harmonics! The question is not how clean a sine wave is, or if that tells anything about the quality of a synth.
The only relevant question is: Do you hear a difference in a musical context?

ghettosynth · Post by **ghettosynth** » Fri Jun 28, 2019 7:39 am

Tj Shredder wrote: Fri Jun 28, 2019 7:35 am Sine waves do not alias at all below nyquist. If you want to test aliasing use something with harmonics! The question is not how clean a sine wave is, or if that tells anything about the quality of a synth.
The only relevant question is: Do you hear a difference in a musical context?

But if the sin wave has distortion then that will alias. However, this is the claim of sadowick, that the distortion of the sine wave is what's leading to aliasing. The peaks above exhibit the classic aliasing pattern where as frequency is increased, the peaks move to the left.

In fact, if you distort the sine wave (ever so) slightly the aliasing components are really clear, again, at very low levels. This is aliasing of the distortion components on the sine wave, roll back the filter and they disappear.

Reefius · Post by **Reefius** » Fri Jun 28, 2019 7:41 am

Hey Sadowick, I've been following your Youtube channel long enough to know that you're a fan of digital synths, which is probably why you like Serum so much. To me, Serum sounds lifeless and sterile just because of this digital perfection.

I don't think MX is a synth for you if you expect it to sound digitally perfect.

sadowickproduction · Post by **sadowickproduction** » Fri Jun 28, 2019 7:43 am

Ill make a follow up video in the morning addressing this. I think square waves would be more obvious as you cant simply filter out the aliasing within the oscillator close to nyquist. Ill also compare the behavior of frequency sweeps in Massive X with Serum. It will be more obvious then.

EvilDragon · Post by **EvilDragon** » Fri Jun 28, 2019 7:44 am

Because it's really important to make music with frequency sweeps that are only around Nyquist, yep.

Reefius wrote: Fri Jun 28, 2019 7:41 amTo me, Serum sounds lifeless and sterile just because of this digital perfection.

Yep. Plus it's heavier on the CPU than MX, while not necessarily sounding better.

AnX · Post by **AnX** » Fri Jun 28, 2019 7:47 am

where is Nyquist, is it in Cornwall?

Mushy Mushy · Post by **Mushy Mushy** » Fri Jun 28, 2019 7:56 am

Sadowick seems a lot calmer on YouTube than on forums

I was using your 100Hz to 5kHz trick with great success last night

Songwriter101 · Post by **Songwriter101** » Fri Jun 28, 2019 7:57 am

Waiting for X-Mas sales price reduction.

EvilDragon · Post by **EvilDragon** » Fri Jun 28, 2019 8:00 am

You mean X-Massive?

Native Instruments Massive X Synth - Sequel to Massive (Out Now!)