The No.1 Website for Pro Audio
SPL IRON mastering compressor by Brainworx
Old 7th September 2020 | Show parent
  #961
Lives for gear
 

Quote:
Originally Posted by Bob Olhsson View Post
Aliasing typically masks the midrange.
Indeed. And not just masking, it somehow "congests" the mids and high-mids in a really nasty way that I can only describe as "fatiguing".

It's a bit similar to truncation noise in the way it takes away "space" and clarity. What it replaces with that is not just a typical noise profile either. If you add similar amounts of white or pink noise to what you are getting aliasing and truncation, you'd still have no issues at all enjoying the mix.. but truncation noise and aliasing is harsh.

I've often wondered if it's some kind of disconnect that our brain detects. Sort of similar to the studies that Genelec did on positioning and how our brain gets slightly stressed when signals aren't arriving exactly the way our brain anticipates (phase offsets, group delays etc). Maybe we have the possibility of following harmonics somehow and when the bounce happens and the harmonic gets re-translated into the wrong register, we get stressed. I don't know.. pure speculation here of course. I just know it's unpleasant.


The interesting thing though is that I do not find heavy use of bit crushing nor sample rate reduction unpleasant at all (as long as it's properly filtered). Perhaps when the effect is very obvious, the brain switches to another mode and simply goes "Aha! This is supposed to be so over the top!"?
Old 7th September 2020
  #962
Lives for gear
 

Audio example posted of aliasing, here:

Lets do it: The Ultimate Plugin Analysis Thread

That's how much better SPL Iron could sound if PA bothered to implement proper anti-aliasing filters and not truncate the output from the plugin.

EDIT: Mind you, this is only one of the several PA plugins with some serious digital artifacts. Just looked at the bx_2098 EQ and it seems to have pretty much similar issues to the Iron. Dare I look at the channel strips?

EDIT2: .. and as there are some obvious PA fanboys here who feel very hurt by my experiment, lets just put it straight out there. ALIASING IS NOT JUST A PA PROBLEM. It is universal for all digital non-linear processes. Thus minimizing those artifacts, no matter what technique is used, is of real importance. Not just for the pretty pictures.

EDIT3: .. and here are the SoundToys plugins aliasing put through the grinder. Yes, SoundToys plugins alias heavily too. Booohooo.

Lets do it: The Ultimate Plugin Analysis Thread

Last edited by bmanic; 7th September 2020 at 11:56 PM..
Old 7th September 2020 | Show parent
  #963
plx
Gear Addict
 

Quote:
Originally Posted by bmanic View Post
Audio example posted of aliasing, here:

Lets do it: The Ultimate Plugin Analysis Thread

That's how much better SPL Iron could sound if PA bothered to implement proper anti-aliasing filters and not truncate the output from the plugin.
they probably will - rename it "Class A", charge 399$ for it than drop it to 29$ in few months.
MEGA
Old 8th September 2020
  #964
Gear Addict
 
Durk Diggler's Avatar
 

Old 8th September 2020 | Show parent
  #965
Lives for gear
 

@ Durk Diggler Should have photoshopped the graph to show aliasing rather than a notch in hearing.

You didn't want to go the whole mile. Oh well 5/10 for effort.
Old 8th September 2020 | Show parent
  #966
Gear Addict
 
Durk Diggler's Avatar
 

Quote:
Originally Posted by Neptune45 View Post
You didn't want to go the whole mile. Oh well 5/10 for effort.
I tried. The secret to success is to raise oneself up after falling down and keep going. I will either make it one day or just die.
Old 8th September 2020 | Show parent
  #967
Lives for gear
Quote:
Originally Posted by Neptune45 View Post
@ Durk Diggler Should have photoshopped the graph to show aliasing rather than a notch in hearing.

You didn't want to go the whole mile. Oh well 5/10 for effort.
Was it supposed to be a hi-cut?
Old 8th September 2020 | Show parent
  #968
Lives for gear
 
BM Grabber's Avatar
 

Quote:
Originally Posted by bmanic View Post
Indeed. And not just masking, it somehow "congests" the mids and high-mids in a really nasty way that I can only describe as "fatiguing".
How many plugins, with how much aliasing "problems" do you have to populate your mix with for the masking to be noticeable?

How "hard" do you need to "drive" the plugins to let the aliasing congest your mids? Or is it just other forms of distortion that "fills the frequeency specter"?

I have certainly heard some great mixes the last 20 years produced within the digital world, so how big is this problem really?

PS. I am just curious... No bet in the "fight"
Old 8th September 2020 | Show parent
  #969
Lives for gear
 

Quote:
Originally Posted by bace View Post
Was it supposed to be a hi-cut?
It's probably meant to be a notch around 4khz if it is to do with hearing loss as it's well known that the older you get this is the frequency range which tends to get notched out. The old wear and tear around where the ears are most sensitive. Also can be exacerbated by overly loud noise exposure for prolonged periods, so don't play your mixes too loud for too long lol.
Old 8th September 2020
  #970
Gear Guru
Didn’t Tchad Blake love Decapitator? Black Keys sound pretty good. Aliasing aside..,,
Old 8th September 2020 | Show parent
  #971
Lives for gear
 

Quote:
Originally Posted by ardis View Post
Didn’t Tchad Blake love Decapitator? Black Keys sound pretty good. Aliasing aside..,,
I think he should browse Gearslutz.
Old 8th September 2020
  #972
Motown legend
 
Bob Olhsson's Avatar
 

This is why working at 96k or above is a rather good idea. Developers/design engineers must always trade between practical efficiency and audio quality.
Old 8th September 2020 | Show parent
  #973
Lives for gear
Quote:
Originally Posted by Neptune45 View Post
It's probably meant to be a notch around 4khz if it is to do with hearing loss as it's well known that the older you get this is the frequency range which tends to get notched out. The old wear and tear around where the ears are most sensitive. Also can be exacerbated by overly loud noise exposure for prolonged periods, so don't play your mixes too loud for too long lol.
The thing is that anti aliasing filter is hi-cut filter. So that is what's missing.
Old 8th September 2020
  #974
SPL Iron is definitely a colorful plugin, and I’ve not used the hardware as a comparison.

However, if you use one of the Germaniums or LED and set the tubes to high bias, the aliasing drops way down to levels that are not really problematic. So, there is a bit of learning how to get a cleaner sound out of it if that is what you want.

I bought it on a whim sale and never got along with it; I’m getting more controlled color other places, but it’s not unusable. Just another shade of color in an endless box of colors.

If you love the compressor action of Iron, run it in Metaplugin oversampling if you’re worried about aliasing and get on with life.
Old 8th September 2020 | Show parent
  #975
Lives for gear
 
dirtROBOT's Avatar
Quote:
Originally Posted by ardis View Post
Didn’t Tchad Blake love Decapitator? Black Keys sound pretty good. Aliasing aside..,,
Trent Reznor likes it too. He was also fine with plugging in zoom 505 pedals into the board so maybe aliasing isn't getting in the way of making works of staggering genius.
Old 8th September 2020 | Show parent
  #976
Lives for gear
 

Quote:
Originally Posted by BM Grabber View Post
How many plugins, with how much aliasing "problems" do you have to populate your mix with for the masking to be noticeable?
Lets put it this way:

As soon as you have any multiple series of plugins (a "chain") that deal with dynamics in any way (this means saturation, distortion, gate/expander, compressor, limiter, dynamic EQ, de-essing etc) you may want to experiment with rendering your final track at higher sample rates. However, if your above mentioned plugins allow internal oversampling, use it!.

Rendering at higher sample rates will lead to other issues, like harmonics going into ultrasound territory that then can cause additional distortion on playback equipment if not dealt with. This is why I use TDR Ultrasonic Filter after any non-linear plugins if I render or operate at higher sampling rates. So this gets messy and quite hard to manage. That is the down side.

The more of these chains that you have in your mix, the more noticeable the differences will be. However, if your chains happen to be on audio sources that only contain low frequency content, then you should be fine. But if you have anything that contains mainly 5kHz and above, then you definitely should consider either higher sample rate rendering or making sure that the plugins you are using allow for oversampling.

Alternatively you could get a plugin "host" chainer thing, for instance DDMF Metaplugin, that allows any plugin to be oversampled. Yet here I also recommend using TDR Ultrasonic Filter after each instance within the Metaplugin.

The plus side of doing all this work is that the final track, when sent to mastering, will be a lot easier to deal with. Further analogue processing (or digital) will not amplify the nasty aliasing artifacts as easily and thus the mastering engineer can do more drastic moves, if needed.


Quote:
Originally Posted by BM Grabber View Post
How "hard" do you need to "drive" the plugins to let the aliasing congest your mids? Or is it just other forms of distortion that "fills the frequeency specter"?
Aliasing is just harmonics that bounce back and thus become in-harmonic. Think about a chord where you have dissonance and compare it to a pure fifth. Now think how much more open and "free" the pure fifth is compared to a dissonant chord. You can apply tons of pure chords all over the place and the end result wont feel all that filled-in.. whereas with dissonant chords the arrangement quickly feels crowded.

As for drive, it completely depends on the plugin. But any properly analogue modeled plugin will simply produce more harmonics and louder, the harder you drive it. Thus more of these will bounce back as aliasing and also louder.

So lets say you drive a basic tube EQ with very low levels. This might give you some 3rd harmonic distortion down at -70dBFS and perhaps 5th harmonic at -110dBFS.. and thus 7th and 9th harmonic will be way below the noise floor and thus not even calculated nor created.

If you now decide to really hammer that tube EQ hard, you suddenly have 3rd at maybe -30dBFS, 5th at maybe -50dBFS (or sometimes it even goes past the 3rd!), 7th is now audible at maybe -70dBFS.. also get 9th, 11th etc.

Now calculate from there. 3rd harmonic is x3 up from 3kHz, so at 9kHz. As you can see.. the 5th, 7th and 9th harmonic very quickly reach nyquist and beyond if you send anything higher than a few kHz.

This is why "driving" a plugin causes much more aliasing. You simply get much more harmonics and at a much higher level. In some cases they bounce all the way down into the low-mids and lows!

Quote:
Originally Posted by BM Grabber View Post
I have certainly heard some great mixes the last 20 years produced within the digital world, so how big is this problem really?
There are of course tons of mixes that have been produced with horribly aliasing digital equipment. But as you may recall, a very common trend has always been the saying "you can't push digital as hard as you can analogue". Especially in the early 90's and 00's, most professional people with "sane" ears, used digital for what it did best.. and that was NOT distortion/compression.

Aliasing isn't really a problem if you avoid it.. with sensible settings and basic usage only of non-analogue modeled stuff. Yes, basic digital compressors still produce a lot of aliasing if not dealt with, but it is directly proportional to the attack, release and ratio settings. It's unlikely that you'd hammer every single compressor in your mix.

Today things are a bit strange with so many "analogue modeled" plugins. Every single one of these, if actually modeling properly, will cause aliasing. Thus it is not at all unlikely that a person with 100 tracks ends up with 300+ plugins that all alias. Lets say somebody inserts 64 instances of Plugin Alliance or Waves SSL channel to their project. That's immediately 64 plugins that alias in multiple parts of each plugin! You get aliasing at the input stage (if modeled), then you get aliasing at the compressor and the gate.. then in the EQ and finally at the output stage.. then take this times 64! Ouch.

We are living in a new analogue modeled plugin revival period.. but that also means we are dealing with much more aliasing than ever before in the history of digitally produced stuff. We get all the cool new toys but also the not so cool uncle Alias.

Thus, dealing with aliasing should be taken seriously by any company dealing with Digital Signal Processing.

Cheers!
bM
Old 8th September 2020 | Show parent
  #977
Lives for gear
 

Quote:
Originally Posted by dirtROBOT View Post
Trent Reznor likes it too. He was also fine with plugging in zoom 505 pedals into the board so maybe aliasing isn't getting in the way of making works of staggering genius.
Bad sounding or intentionally bad sounding sounds have never been a decider between good or bad music.

Music and the fidelity of music are completely separate topics.


I know, it's a fun straw man argument to always fall back on when anybody says anything critical about the actual fidelity of any process. None the less, it's still just a straw man argument.

You can get some absolutely terrific tasting street meat in various places in the world.. just don't ask what it was made from or how. Still tastes good. Thus two separate things.

See my point?
Old 9th September 2020 | Show parent
  #978
Lives for gear
 
ionian's Avatar
Quote:
Originally Posted by bmanic View Post
You can get some absolutely terrific tasting street meat in various places in the world.. just don't ask what it was made from or how. Still tastes good. Thus two separate things.

See my point?
I just cover my plugins in a lot of tzatziki and try not to think about it.
Old 9th September 2020 | Show parent
  #979
Gear Addict
 
Durk Diggler's Avatar
 

Quote:
Originally Posted by bmanic View Post
Rendering at higher sample rates will lead to other issues, like harmonics going into ultrasound territory that then can cause additional distortion on playback equipment if not dealt with. This is why I use TDR Ultrasonic Filter after any non-linear plugins if I render or operate at higher sampling rates. So this gets messy and quite hard to manage. That is the down side.
In situations where you insert TDR UF after a non-linear plug, you do not put an instance before it as another user suggested?
Old 9th September 2020 | Show parent
  #980
Lives for gear
 

Quote:
Originally Posted by Durk Diggler View Post
In situations where you insert TDR UF after a non-linear plug, you do not put an instance before it as another user suggested?
I put the TDR UF anywhere I think there might be stuff going past my final target sample rates nyquist frequency.

As my example was a simple 44.1kHz drumloop, I didn't have to put TDR UF before the first non-linear plugin. The upsampling of the audio material didn't cause any spikes or such that went past 22050Hz.

So yeah, TDR UF after every process that creates stuff past nyquist. Like this:

Audio source -> upsampling -> saturation plugin -> TDR UF -> 2nd saturation plugin -> TDR UF -> 3rd plugin EQ -> 4th saturation -> TDR UF -> etc. etc.

If I understood Fabian's description of the UF plugin, the idea is to not let high harmonics get to the next instance of saturation as they can eventually bounce back, even at very high sample rates and thus bounce back down. There were some other benefits to it also but I've forgotten what they were.. whatever it does, it does something good. Definitely sounds better using TDR UF than not. Settings wise I'm not entirely sure how to use it most effectively but so far the default settings seem like a good starting point.

This may be placebo but I do feel like I get slightly better results forcing each TDR UF plugin instance to have a slightly different cutoff frequency. Pretty sure it's placebo. I know for a fact I can't hear **** past 20kHz. Probably not even 18kHz at this point in my life.

.. which reminds me, I gotta get my ears checked. Been years since I last did it.

Last edited by bmanic; 9th September 2020 at 06:52 AM..
Old 9th September 2020 | Show parent
  #981
plx
Gear Addict
 

Quote:
Originally Posted by dirtROBOT View Post
Trent Reznor likes it too. He was also fine with plugging in zoom 505 pedals into the board so maybe aliasing isn't getting in the way of making works of staggering genius.
you can make a hit track with a usb mic and a god damn ukulele, so why bother with anything at all?
And you know, 250years ago, mozart made iconic music only with a piano and his music has held for two hunder and fifty years!!

If i paraphrase your witty remark, it reads: Why improve anything, ever?

That's a philosophical question, not one about aliasing or SPL IRON
Old 9th September 2020 | Show parent
  #982
Lives for gear
Quote:
Originally Posted by plx View Post
you can make a hit track with a usb mic and a god damn ukulele, so why bother with anything at all?
And you know, 250years ago, mozart made iconic music only with a piano and his music has held for two hunder and fifty years!!

If i paraphrase your witty remark, it reads: Why improve anything, ever?

That's a philosophical question, not one about aliasing or SPL IRON
What was the point of doing that piano in the first place? Germans could make flutes 43000 years ago.
Old 9th September 2020 | Show parent
  #983
plx
Gear Addict
 

Quote:
Originally Posted by bace View Post
What was the point of doing that piano in the first place? Germans could make flutes 43000 years ago.
germans?
pft.
https://en.wikipedia.org/wiki/Divje_Babe_Flute
Old 9th September 2020
  #984
Lives for gear
 
M Albazy's Avatar
I don't use Iron anymore due to the shallow response even with a slight compression. Unisum way better in Vari-mu mode.
Old 9th September 2020 | Show parent
  #985
Gear Guru
Quote:
Originally Posted by bmanic View Post
Lets put it this way:

As soon as you have any multiple series of plugins (a "chain") that deal with dynamics in any way (this means saturation, distortion, gate/expander, compressor, limiter, dynamic EQ, de-essing etc) you may want to experiment with rendering your final track at higher sample rates. However, if your above mentioned plugins allow internal oversampling, use it!.

Rendering at higher sample rates will lead to other issues, like harmonics going into ultrasound territory that then can cause additional distortion on playback equipment if not dealt with. This is why I use TDR Ultrasonic Filter after any non-linear plugins if I render or operate at higher sampling rates. So this gets messy and quite hard to manage. That is the down side
The more of these chains that you have in your mix, the more noticeable the differences will be. However, if your chains happen to be on audio sources that only contain low frequency content, then you should be fine. But if you have anything that contains mainly 5kHz and above, then you definitely should consider either higher sample rate rendering or making sure that the plugins you are using allow for oversampling.

Alternatively you could get a plugin "host" chainer thing, for instance DDMF Metaplugin, that allows any plugin to be oversampled. Yet here I also recommend using TDR Ultrasonic Filter after each instance within the Metaplugin.

The plus side of doing all this work is that the final track, when sent to mastering, will be a lot easier to deal with. Further analogue processing (or digital) will not amplify the nasty aliasing artifacts as easily and thus the mastering engineer can do more drastic moves, if needed.




Aliasing is just harmonics that bounce back and thus become in-harmonic. Think about a chord where you have dissonance and compare it to a pure fifth. Now think how much more open and "free" the pure fifth is compared to a dissonant chord. You can apply tons of pure chords all over the place and the end result wont feel all that filled-in.. whereas with dissonant chords the arrangement quickly feels crowded.

As for drive, it completely depends on the plugin. But any properly analogue modeled plugin will simply produce more harmonics and louder, the harder you drive it. Thus more of these will bounce back as aliasing and also louder.

So lets say you drive a basic tube EQ with very low levels. This might give you some 3rd harmonic distortion down at -70dBFS and perhaps 5th harmonic at -110dBFS.. and thus 7th and 9th harmonic will be way below the noise floor and thus not even calculated nor created.

If you now decide to really hammer that tube EQ hard, you suddenly have 3rd at maybe -30dBFS, 5th at maybe -50dBFS (or sometimes it even goes past the 3rd!), 7th is now audible at maybe -70dBFS.. also get 9th, 11th etc.

Now calculate from there. 3rd harmonic is x3 up from 3kHz, so at 9kHz. As you can see.. the 5th, 7th and 9th harmonic very quickly reach nyquist and beyond if you send anything higher than a few kHz.

This is why "driving" a plugin causes much more aliasing. You simply get much more harmonics and at a much higher level. In some cases they bounce all the way down into the low-mids and lows!



There are of course tons of mixes that have been produced with horribly aliasing digital equipment. But as you may recall, a very common trend has always been the saying "you can't push digital as hard as you can analogue". Especially in the early 90's and 00's, most professional people with "sane" ears, used digital for what it did best.. and that was NOT distortion/compression.

Aliasing isn't really a problem if you avoid it.. with sensible settings and basic usage only of non-analogue modeled stuff. Yes, basic digital compressors still produce a lot of aliasing if not dealt with, but it is directly proportional to the attack, release and ratio settings. It's unlikely that you'd hammer every single compressor in your mix.

Today things are a bit strange with so many "analogue modeled" plugins. Every single one of these, if actually modeling properly, will cause aliasing. Thus it is not at all unlikely that a person with 100 tracks ends up with 300+ plugins that all alias. Lets say somebody inserts 64 instances of Plugin Alliance or Waves SSL channel to their project. That's immediately 64 plugins that alias in multiple parts of each plugin! You get aliasing at the input stage (if modeled), then you get aliasing at the compressor and the gate.. then in the EQ and finally at the output stage.. then take this times 64! Ouch.

We are living in a new analogue modeled plugin revival period.. but that also means we are dealing with much more aliasing than ever before in the history of digitally produced stuff. We get all the cool new toys but also the not so cool uncle Alias.

Thus, dealing with aliasing should be taken seriously by any company dealing with Digital Signal Processing.

Cheers!
bM
Tchad is that you lol! Thx for actually answering my question. Obviously a top guy can work around limitations and that’s refreshing. I learned something from this....
Old 10th September 2020 | Show parent
  #986
Lives for gear
 
Calagan's Avatar
 

I did my own test with a drum loop compressed through a chain of 3 SPL Iron in serie, running at 44khz with and without 8X oversampling inside the demo version of Metaplugin, then rendered in 44kHz / 16 bit with a triangular dither.

After each instance of Iron in Metaplugin, I used an Ultrasonic filter from TDR with the default preset.

For the sake of easy reproductibility, I used the same preset in Iron for each instance (MR Drum buss), but I did some tweaking : 100% wet / Max threshold / +2db output / high tube bias / fastest attack
The GR needle goes from 1 to 3 db of gain reduction, so it's not a lot in theory, but as you can hear it's already quite smashed.

You'll find attached the original dry loop, then A & B version.

I can definitely hear a difference in clarity : it's true about the congested mids, there's something congested in the non os version. But it's not so crazily bad IMHO.
When listening both versions without overthinking, I'm more listening two different versions than a "bad" one and a "good" one... Actually, there is too a tiny difference in the dynamics (I don't know why, but I'm sure it's not a difference in the settings).

What do you think ?
Which version is the oversampled ?
Attached Files

drums loop - DRY.wav (3.03 MB, 587 views)

drums loop - A.wav (3.03 MB, 618 views)

drums loop - B.wav (3.03 MB, 628 views)


Last edited by Calagan; 10th September 2020 at 05:07 AM.. Reason: improving clarity of my explanations
Old 10th September 2020
  #987
Lives for gear
 
Macaroni's Avatar
 

I say B is oversampled.

I've been using IRON on several tracks since getting it and I'm not hearing any unpleasant congestion when I bypass it in various chains. And that's just using one instance.

I can't imagine SPL and PA won't provide OS options soon. Most of their other plugins have it standard.
Old 10th September 2020 | Show parent
  #988
Lives for gear
 
Calagan's Avatar
 

Quote:
Originally Posted by Macaroni View Post
I say B is oversampled.

I've been using IRON on several tracks since getting it and I'm not hearing any unpleasant congestion when I bypass it in various chains. And that's just using one instance.

I can't imagine SPL and PA won't provide OS options soon. Most of their other plugins have it standard.
Thanks for your participation. I'll wait for some more answers before giving the results.
I'm agree with you, I never noticed any nasty stuff while using Iron "normally" (not compressing a sweep for exemple) and I like very much this plugin, even if I must recognize it sounds a bit better in my exemple when it is oversampled.
But it's an X8 oversampling, something you can only achieve offline (at least on my 2012 laptop). I guess most dynamic plugins sound better at 352 kHz .

Regarding your hope, the problem is PA/Dirk Ulrich officially says that Iron is oversampled just like any other PA plugin. Which seems to be false. Actually, all their compressors have some latency (that seems to indicate some OS somewhere). Iron is the only one to have zero latency - and that's pretty strange for a "mastering compressor"...
Old 10th September 2020 | Show parent
  #989
Gear Nut
 

Quote:
Originally Posted by Calagan View Post
Thanks for your participation. I'll wait for some more answers before giving the results.
I'm agree with you, I never noticed any nasty stuff while using Iron "normally" (not compressing a sweep for exemple) and I like very much this plugin, even if I must recognize it sounds a bit better in my exemple when it is oversampled.
But it's an X8 oversampling, something you can only achieve offline (at least on my 2012 laptop). I guess most dynamic plugins sound better at 352 kHz .

Regarding your hope, the problem is PA/Dirk Ulrich officially says that Iron is oversampled just like any other PA plugin. Which seems to be false. Actually, all their compressors have some latency (that seems to indicate some OS somewhere). Iron is the only one to have zero latency - and that's pretty strange for a "mastering compressor"...
I have multiple plugins that are oversampled and don't have any latency.

Tal DAC, and Audiothing Outerspace come to mind, zero latency doesn't necessarily mean not oversampled.
Old 10th September 2020 | Show parent
  #990
plx
Gear Addict
 

Quote:
Originally Posted by smellypants View Post
I have multiple plugins that are oversampled and don't have any latency.

Tal DAC, and Audiothing Outerspace come to mind, zero latency doesn't necessarily mean not oversampled.
Yes that's true. Voxengo even allows you to switch between minimal phase and linear phase filter for oversampling - the latter having latency.

However it's visible on the phase response.

Majority of PA's oversampled plugins have 32 samples of latency tho.

Quote:
Originally Posted by Calagan View Post
I did my own test with a drum loop compressed through a chain of 3 SPL Iron in serie, running at 44khz with and without 8X oversampling inside the demo version of Metaplugin, then rendered in 44kHz / 16 bit with a triangular dither.

After each instance of Iron in Metaplugin, I used an Ultrasonic filter from TDR with the default preset.

For the sake of easy reproductibility, I used the same preset in Iron for each instance (MR Drum buss), but I did some tweaking : 100% wet / Max threshold / +2db output / high tube bias / fastest attack
The GR needle goes from 1 to 3 db of gain reduction, so it's not a lot in theory, but as you can hear it's already quite smashed.

You'll find attached the original dry loop, then A & B version.

I can definitely hear a difference in clarity : it's true about the congested mids, there's something congested in the non os version. But it's not so crazily bad IMHO.
When listening both versions without overthinking, I'm more listening two different versions than a "bad" one and a "good" one... Actually, there is too a tiny difference in the dynamics (I don't know why, but I'm sure it's not a difference in the settings).

What do you think ?
Which version is the oversampled ?
ooh i'll check this out later today. thanks for the effort.

Quote:
Originally Posted by Macaroni View Post

I can't imagine SPL and PA won't provide OS options soon. Most of their other plugins have it standard.
idk. he called me a gearslutz conspiracy theorist, and two supports tickets to the matter were closed inexplicably.
He then posted a c/p about oversampling pretty much from the Alpha Comp manual, which has a section dedicated to OS/AA (absent in SPL IRON).

fwiw, SMHC Class-A ("the mastering edition") has the typical 32samples of latency like other bx/pa oversampled plugs, SMHC has no latency as well.

I reckon if they ever do SPL IRON with oversampling, it will be a separate product
📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 11798 views: 1149120
Avatar for wjmwpg
wjmwpg 27 minutes ago
replies: 75 views: 20196
Avatar for MegaGiga
MegaGiga 27th October 2015
replies: 12544 views: 964785
Avatar for monkeyxx
monkeyxx 4 hours ago
replies: 30 views: 7587
Avatar for jakeg70888
jakeg70888 22nd April 2020
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
🖨️ Show Printable Version
✉️ Email this Page
🔍 Search thread
🎙️ View mentioned gear
Forum Jump
Forum Jump