john why are you doing that to me brother you ain't tell me nothing about the limiter message i sent you.

Recommendations are on the first page: Digital Limiter Quality Test

what would be the most who's the most true to the source and damage the sound the lesser, you just gave me a lot of recommendation here.

This is version 1.9. Oversampling did not occur until 2.0. I Installed that but reverted it for some reason. Did not work well with cubase. However i have been many updates on both cubase and this since then.

Oversampling turned on should make a real difference, a bit unfair to show an older plugin that people won't realise now has oversampling but understand why.

Also to not forget if you plan on crancking up the lindell 80 series preamp on 1073 section a a lot it can start to show some alias a little bit, just saying for people who's gonna crank it on a guitar better put it in metaplugin if that's you're plan but if used normally i don't see any problems also the elysia alpha compressor can aliases if extreme compression is dialed but if used normally or subtlety there will be no problems, just saying for parallel compression fan boys and people who uses their tools to the extreme, anyways i love using them inside metaplugin anyway i can still get them to show on my screen when minimizing my daw but these plugins are top tier i have you know, also weiss is a good digital compressor for mastering it has that compression curve that grab and make everything cohesive with a open wide top.

Consider putting plugins that you know generates a lot of harmonics saturation into metaplugin. plugin alliance saturation plugins are a good example of this as they support very high oversampling.

less than 10dbs THD in the lindell 80 is enought for vocal or guitar saturation but as i told you for the hard rockers out there who's gonna crank it to 30 to 70 it will start to aliases as the neve preamp are a lot complex.

It does. Here is what it look like with OS.

But with this test I also realised why I did not use 2.x. It is all the bloatware.
All preset handlings assumes that you have all their products. This garbage plugin will be reverted, so back to 1.9!

Thanks johann

Garbage? I've done some really good productive work using them.

After having done some tests myself, I am relieved to see (and hear) that my regulars from DMG, Tone Projects, PSP and Valhalla are clean (enough). However, putting the Eventides into Meta and oversampling 2x (project at 44.1) improved them a lot. DDMF has a new customer.

j,

It is a bloatware. The designer of this plugins have not read the VST-SDK so the work flow it totally ****ed up. The installation procedure a disaster and install things that I have not bought or asked for. And their support is quite rude. I have pointed one problem that make this not work as VST should and make it broken on cubase with remote controllers. There are two common consumer controllers for from Steinberg. Steinberg is the creator of VST. And when pointing out that Steinberg controller does not work with their plugins the response is that they can not test all remote controllers. Sure that is fair, but Steinberg controller is the reference for Steinberg software. And VST is steinberg software. Plugins does not only need to sound good. They need to fit in to a workflow as well.

Even when driven hard, beautiful, simply beautiful:

it does aliases if cranked but pretty good at handling aliasing.

Everybody… big news!!!!!!

Look at what Christian at DDMF has put together for us in Metaplugin:





Separate realtime and render processing, and expanded oversampling multiples for each!!!!!!!

Available right now in version 3.8 of DDMF Metaplugin!!!!

https://ddmf.eu/metaplugin-chainer-v...s-aax-wrapper/

Sorry to write this as an ad, but these changes were brought to us from a request I made to DDMF because of this thread, and I am very appreciative for it!

Please support this dev if you haven’t yet!

I will need to get confirmation, but I believe the oversampling multiples are still based on our project sample rate, so that can require some care when using realtime oversampling at higher project sampling rates.

Just keep in mind, if we have a crash when oversampling, it’s because we are probably pushing the plugin we are trying to oversample too hard for its design.

For example, if the plugin already has some type pf internal oversampling that we don’t already know about, it will likely croak or crash when pushed maybe even as low as 4x.

So with this great expansion of this tool, we also bear some responsibility to not blame this dev for these types of crashes, which are essentially the laws of computed physics biting us if we push them too hard.

Let’s go!!!

I have had it confirmed that the oversampling remains a true multiple of our current project sample rate.

Can you take a plugin that aliases (Waves TG12345 comes to mind) and show with SPAN what this does? I tested doubling the project sample rate at some point but it wasn't reducing the amount of aliasing...

Unfortunately when I did a fresh install of Big Sur last week I decided not to install anything which intentionally aliases, and cut out anything with online installers.

So I decided to install Klanghelm IVGI 2 right now for this test because as a saturator without user controllable OS, I know I can make it alias, and it is a free plugin but also without copy protection, like all of Klanghelm’s plugins.

I have a 44.1kHz project sample rate for this test.

No oversampling (44.1kHz)


Realtime 2x (88.2kHz)


Realtime 4x (176.4kHz)


Realtime 8x (352.8kHz)


Realtime 16x (705.6kHz)


Offline 32x (1.411MHz)


I had a 64x offline render but messed up taking the picture, but it looked very similar to the 32x.

This plugin definitely behaves differently than most of the Waves I’ve tested which usually have less complex nonlinearities as usually just odd or even harmonic series and should clean up easier.

Obviously I’m pushing the plugin internal controls hard on this test to make the aliasing more visible on the spectrum.

Let me know if this was useful

Actually no. If you've done great music with them then just move on and don't drag yourself into this "anti-alias" abyss.

This Thread refuses to die, and I feel as if I'm in a time warp with the Plugin Analysis Thread that's been circulating for years. I will happily unsubscribe, as I have the tools and techniques with which to analyze in place. I suppose there's a time and place for everything, though neither fits for me. Carry on, as I'm sure you will.

incredible!!!!

so can i oversample plugins which does not support sample rate more than 192khz since it doesn't upsample the sample for oversampling.

Most work but if you get in that situation just follow your ears, as it will depend on the plugin. If it sounds better, then run with it. Since it is a simple click of the OS button, it should be an easy decision.

Basically the advantage to listen for is that we don’t load our mixes with inharmonic high frequency content which then in turn gives us plenty of clean headroom, just like analog mixing would from lacking aliasing.

Ok good to know will try the Brainworx plugin and see how it goes.

The best thread ever on GS. Thank you OP for starting this.

i think the offline oversampling doesn't work.

It was, thank you.

Here is an example with a harmonic profile more similar to Waves plugins.

I am ignoring Voxengo’s Overtone GEQ oversampling features for this example and using Metaplugin’s instead.

I’m also using a much lower frequency tone to underscore that aliasing does not only occur from higher frequency content.

No oversampling (44.1kHz)


Realtime 2x (88.2kHz)


Realtime 4x (176.4kHz)


Realtime 8x (352.8kHz)


Realtime 16x (705.6kHz)

When we get up there weird things can happen.

It’s doing something for sure but with some plugins at certain volumes will hit the law of diminishing returns.

64x is more useful for higher volume situations, such as the mastering chain.

AFAIK, not all oversampling is equal and to some extent depends on the quality of the filters in the plugin. I could be wrong though.

Is that really an issue?

Is Metaplugin’s oversampling “high quality?”

Is Metaplugin light enough to be used all over the place, with say, 50+ instances?

It would obviously suck to end up with worse results after oversampling IF that is a thing.

In before “just use your ears bro.”