Do hardware VA's alias as badly as software?

[Leslie Sanford]

Aliasing is what you get when the stagecoach rides into town and the wheels start rolling backwards.

[IncarnateX]

Tell that to my jp8000 that wouldn't give a fart which way the wheels are going and yet the aliasing bitch sounds great.

[cron]

Aliasing is what you get when the stagecoach rides into town and the wheels start rolling backwards.

It's pretty much exactly the same thing! The sampling rate of the device (number of frames per second in the case of film) isn't high enough to capture what's going on leading to an incorrect representation of the signal.

[@midnight]

Diva doesn't have an option for oversampling. It already oversamples internally many times per sample (because of the zero-feedback-delay filter prediction). You only get "Accuracy" parameter, which boils down to the amount of iterations of predictions for each sample...

Never used Diva but what are the draft and divine settings for if not oversampling?

[@midnight]

Aliasing is what you get when the stagecoach rides into town and the wheels start rolling backwards.

I'm at work and this has me lolling all over the office.

[xh3rv]

Diva doesn't have an option for oversampling. It already oversamples internally many times per sample (because of the zero-feedback-delay filter prediction). You only get "Accuracy" parameter, which boils down to the amount of iterations of predictions for each sample...

Never used Diva but what are the draft and divine settings for if not oversampling?

It is iterations of predictions, instead of calculating more samples it's more calculations per sample. It's iterative in the sense that the algorithm can evaluate an input value to improve the precision of that value and then recycle this output value as an input again. Behind that there's an idea of how to compute a problem which can be functionally defined but not computationally finite. But there's a geometric progression of precision which I think Urs or Clemens has mentioned relates to bit-depth in Diva's filters. (Fun stuff IMO ) Also draft is not 0df, worth noting since there's a big jump in sound between draft and fast but less from fast to divine.

[e] And (if I'm understanding things correctly) it's worth noting, this isn't about oversampling to avoid aliasing. Oversampling can also have some effects on models of time-variant types of stuff that are fundamentally tricky in DSP, this type of solution is more powerful - it's expensive for CPU, but orders of magnitude less CPU than the oversampling needed to get the same results.

[Lotuzia]

Xils-Lab Synthesizers : Aliasing free

[@midnight]

Xils-Lab Synthesizers : Aliasing free

A lot of them say that. Fabfilter has "industry leading aliasing free oscillators" yet their synths alias horribly in the upper register due to their (not so fab anymore) filter code.

This is slightly off topic but fabfilter really needs to revisit their filter code, which I believe is nearing a decade old. The newer plugs are great, timeless and volcano and any other plugs using the fabfilter one code are aliasing city @ 44.1khz

[Lotuzia]

Xils-Lab Synthesizers : Aliasing free

A lot of them say that. Fabfilter has "industry leading aliasing free oscillators" yet their synths alias horribly in the upper register due to their (not so fab anymore) filter code.

This is slightly off topic but fabfilter really needs to revisit their filter code, which I believe is nearing a decade old. The newer plugs are great, timeless and volcano and any other plugs using the fabfilter one code are aliasing city @ 44.1khz

Well you can try the demo versions and verify by yourself .... Risk free, aliasing free, what can be better ? Try Oxium, pure crispy sound in the higher octaves

[mkdr]

Great that you finally found few audio examples. But still no one has any idea what the aliasing on synths like Nord Lead, Virus, V-Synth is.. And if it even is actually aliasing or something else. None who claim that those synths alias has one. Wtf..

That's all i was saying.

Anyone can quote theory all day long. Without audio proof it's just hearsay.
I'm not saying aliasing doesn't exist, just that you are baboons who repeat other baboons nonsense.



The sync and fm on Nord Lead was a nice suggestion. Points for that. Got to try those out to see what they do. Under normal use (making music) this hasn't crept up.. and i do use fm and sync too. Maybe i just don't do music for bats

[Sendy]

Aliasing is to modern VAs what drifting was to analog vintage synth. When they were popular everybody hated it. When analogs became vintage, everyone tries to emulate it. One of the problems with making a clone of Jp8000 is actually to emulate the way it aliases, which is a part of it's special sound character. Ironically, Synapse audio has just released the Antidote RE synth. One of its waveforms is called "digital saw" and is a special aliasing waveform, which is meant to emulate the string and pad sounds of older digital VAs. There you go, aliasing is becoming a vintage feature.

Cheers

I completely agree with this. Not only is aliasing an essential part of the sound character of some synths, but I predict that when we reach the point where every digital synth is free from aliasing and that becomes the norm, more and more synths will include optional alising algos as musicians find the output "too clean".

Same thing seems to happen to all technology. In the late 90's all gaming was 3D and aimed at being realistic and older style games were almost unheard of and considered very primitive. Once 3D graphics became ubiquitous and reasonably advanced in their realism, it suddenly became "ok" (not unfashionable) for developers to make 2D sidescrolling games, platformers, low-bit styled games etc.

About aliasing, the simplest test you can do is to play a high note with a bright waveform and use pitchbend. If you hear a part of the sound moving the opposite direction to your desired sound, that could well be aliasing. If you pass that test, try the same thing with a patch that features sync, AM, FM and audio rate modulation. If it passes all these tests, it's fairly safe to assume your oscillators are alias-free for most intents and purposes.

Oscillators aren't the only thing that can alias, as has already been mentioned, filters can do this as well, as can distortion stages. I think filter aliasing sounds different from oscillator aliasing but I could be wrong on that... But the truth is there are lots of ways that aliasing can be introduced at every turn in the digital signal path.

[realmarco]

nothing stopping Harware VA makers to filter out potential aliasing frequencies of the Oscillator waveforms.

and then claim its aliasing free. just sayin'

is a Virus in 44.1khz or 96k hz ?

[pdxindy]

Aliasing is to modern VAs what drifting was to analog vintage synth. When they were popular everybody hated it. When analogs became vintage, everyone tries to emulate it. One of the problems with making a clone of Jp8000 is actually to emulate the way it aliases, which is a part of it's special sound character. Ironically, Synapse audio has just released the Antidote RE synth. One of its waveforms is called "digital saw" and is a special aliasing waveform, which is meant to emulate the string and pad sounds of older digital VAs. There you go, aliasing is becoming a vintage feature.

Cheers

I completely agree with this. Not only is aliasing an essential part of the sound character of some synths, but I predict that when we reach the point where every digital synth is free from aliasing and that becomes the norm, more and more synths will include optional alising algos as musicians find the output "too clean".

Do people accuse analog of being too clean?

[EvilDragon]

But still no one has any idea what the aliasing on synths like Nord Lead, Virus, V-Synth is.. And if it even is actually aliasing or something else.

Now this is insulting. I am studying DSP and I know what aliasing is, and I've had enough of experience with the mentioned boards (I don't need to OWN THEM, I tried them out in the shop with headphones and pushed them far enough to alias) to confirm they indeed are aliasing at one point or another.

[IncarnateX]

But still no one has any idea what the aliasing on synths like Nord Lead, Virus, V-Synth is.. And if it even is actually aliasing or something else.

Now this is insulting. I am studying DSP and I know what aliasing is, and I've had enough of experience with the mentioned boards (I don't need to OWN THEM, I tried them out in the shop with headphones and pushed them far enough to alias) to confirm they indeed are aliasing at one point or another.

I once made a thread where I asked which of a line of artifacts I percieved that was aliasing. I was given a good example from another thread as best answer (scroll down)

in this thread i posted a comparison of an anti-aliased sawtooth and a non-anti-aliased one:

http://www.kvraudio.com/forum/viewtopic.php?p=3298573

along with pictures of the waveforms and spectra. somwhere else, someone described the non-anti-aliased version as 'the one with the lawnmower in the background', which i found pretty fitting

Here the difference between the aliasing and non aliasing sawtooth is pretty obvious if anyone should have doubt about what aliasing is from a perceptual point of view.

Cheers