VST Oversampler 1.4 (with SSE/3DNow! support)

[nollock]

*oops*

[hesnotthemessiah]

Wow Chris!!!! This is amazing!!! I don't pretend to fully understand the mathematics involved but this really enhances the plugins I have just tried it out on. I use a combination of Cubase VST5 with EnergyXT running as a VSTi and VSTfx plus another 2 PCs running EnergyXT purely as a plugin host. 2 Pentium4s and 1 AMD Sempron. Only had about half an hour trying it out but it seems to work great on all the PCs.

Tried Oversampler with Filterbank3 synth, Z3ta+ and Vanguard. The Filterbank3 synth preset SYN FM is great to show the benefits of using Oversampler - the lows and highs are much more clearly defined. Most sounds are now much fuller and defined - especially sounds with alot going on in the lower or upper end of the audio spectrum. Reverbs are also less muddy. It works great with EnergyXT. I just renamed the Oversampler file to that of my EnergyXT files (as per your instructions) and then I can load up the EnergyXT file and insert what ever plugins I want to upsample into it. Can't wait for your updates. I will happily pay a donation if you could point me in the right direction.


Thankyou!!!

[metamorphosis]

Yes, higher sample rates are better, sure, I know that. But are you saying that you'll get better results from rendering a synth at 192khz resampled back down to 96khz, than you would if you just rendered 'straight' at 96khz?
Cause I'd have to say, if you can hear a difference between the two, I'm really surprised-

Thats exactly what ive just said isnt it?

Ok heres a 1Khz naive saw...

http://www.flak.clara.co.uk/NaiveSaw.wav

Heres the FFT analysis of it...



Everything between the main peeks is alising, notice how densely its packed. Thats because the aliasing has wrapped round many times, up and down..

Heres the same sawtooth 4x oversampled..

http://www.flak.clara.co.uk/4xOSSaw.wav

Heres the FFT analysis of it...



Notice how the aliasing is a bit lower in respect to the main peaks and a lot less densely packed.

Yes, that's wonderful. Read my question again. Can you 'hear' it? Not 'see' it. Not write it down on paper. Can you actually physically hear the -actual difference- between a synth offline-rendering at 96khz, and a synth offline-rendering at 192khz and then being downsampled, to 96khz? Not between 44khz and 88khz, or 48khz and 96khz, but between 96khz and 192khz. I doubt that, in the extreme-
Cheers,
m@

[metamorphosis]

If only 1 or 2 generators/effects you are using will prosper from it, why waste the CPU on everything else? For instance if you have 4 non-aliasing synths on a project that you love the sound of, and a 5th that you really like but it aliases a lot, why waste the CPU on the other 4 as well?

Because you're rendering, and rendertime does not matter, and because the entire signal chain benefits from using higher sample-rates- generally-speaking. Most plugins, EQ's, effects, benefit from increased resolution-
cpu time is only an issue when mixing, not when offline rendering.

Also, not all VST's work outside of 44.1Khz. Say for instance you are working with one while also working with one you want to use at 88.2Khz, for instance.

[Edit: I get what you mean here, after reading more carefully - I think a 44k-limited plugin is a bit of a limitation in your signal path, in itself - I'd be inclined to dispose of it]

[nollock]

Yes, that's wonderful. Read my question again. Can you 'hear' it? Not 'see' it. Not write it down on paper. Can you actually physically hear the -actual difference- between a synth offline-rendering at 96khz, and a synth offline-rendering at 192khz and then being downsampled, to 96khz? Not between 44khz and 88khz, or 48khz and 96khz, but between 96khz and 192khz. I doubt that, in the extreme-
Cheers,
m@

Well i guess I rather naively thought that given a good explanation you would be able to figure it out for yourself. I did specificaly show that the aliasing can wrap round many times, it aint just a bit of hf noise around 20-40khz..

Anyway one 'writen down' on paper example before the audio clips...

A saw drops off at 6dbs an octave, which means a 1Khz saw has harmonics at 96khz that are roughly 39dbs down from the fundamental. So when a naive saw is rendered at 96khz any signal energy at around 96khz will be aliased to around 0hz. So even at 96khz that wil result in 39db alising noise. Do you think that would be audible? It's probably around 10 dbs better of than rendering at 48hhz though.

Anyway as examples seem to be what you want, or need.. here goes..

Heres a naive Saw rendered directly at 96khz

www.flak.clara.co.uk/96kSaw.wav

Heres one rendered at 192Khz, downsampled to 96Khz.

www.flak.clara.co.uk/2x96kSaw.wav

Both saws, 96k first, 192k second....

www.flak.clara.co.uk/BothSaws.wav

Does that answer your question??

Anyway i give up now, i took me 4 bloody edits to get this right, so i think i'd bettr get back to nursing my hangover..

chris

[punkrockdude]

Just want to let you know that I am REALLY looking forward to an update where the plugin is less unstable and the high end attenuation is gone. I can't wake because what it made everything else than the really bright stuff sound was so good. Regards

[triple p]

hey check out the wusik oversampler demo, it goes up to 4x and sounds amazing.
Completely stable to, unless the plug your using does not support higher sample rates.
give it a try, works amazing.

[Sam@Megablastic]

The bottom line is that any dsp fx sound better at higher sampling rates, why do you think that the uad-1 top plugins run at 192khz.

It's not about hearing higher than 22khz, it's about all the aliasing that is past back down through the 20hz to 20khz range.

And yes I can hear it Higher sample rates really help the bass funnily enough, not just the higher frequencies. I'll be putting this plugin through it's paces this week. Top job on this and from William K for his plugin as well.

For those that refuse to see the merits of higher sampling rates and would rather theorize than listen to the results..... well, I'd rather use my ears and go with what sounds best

[Chris Walton]

I AM AN IDIOT

Ok, after spending about 16 hours simply not looking at the code, I found the problem I've been having all this time

I will release an update shortly, and I suspect much better compatibility with most of the hosts out there. As in, MUCH better.

Oh dear god I am an idiot

Sorry for all those that were inconvenienced by my stupidity

[sc_a]

Please will Work Window ME?
Not work now - PLease help!

[Chris Walton]

Please will Work Window ME?
Not work now - PLease help!

I cannot guarantee it, because I have no Windows 9x or ME computer or installation to test it on. I don't even have my original Windows 98 disk anymore (I think it's across the atlantic ocean, if my dad hasn't thrown it away yet).

[Chris Walton]

Alright, released, see the original post!

I'm guessing now 99.999% of hosts out there should work with the oversampler now. It was a really stupid bug of mine that I didn't catch, and after a day of working non-stop and not seeing any of the code, the source of the bug finally stood out clear as day.

I also implemented host synchronization, although there is still a bug present (everything runs at double rate until the host tempo is changed once).

[tony tony chopper]

The bottom line is that any dsp fx sound better at higher sampling rates, why do you think that the uad-1 top plugins run at 192khz.

For those that refuse to see the merits of higher sampling rates and would rather theorize than listen to the results..... well, I'd rather use my ears and go with what sounds best

some would also use their brain instead, and understand what this is all about, to avoid oversampling things like echo effects or what's already antialiased by other means.

You will only discredit the real advantage of oversampling with idiotic claims. In this thread you have someone who knows what he's doing, in another thread someone who claims his plugin does magic, sadly you're just food for marketing.

[triple p]

in the other thread you speak of, if it is the one i think, I don't see anywhere he claims it to be magic. Just that it works well

[Chris Walton]

in the other thread you speak of, if it is the one i think, I don't see anywhere he claims it to be magic. Just that it works well

I promise that once I've got the kinks worked out of mine, it'll work better than his (completely different approach).