The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Lets do it: The Ultimate Plugin Analysis Thread Dynamics Plugins
Old 25th May 2010
  #421
Gear Maniac
 

Quote:
Originally Posted by andy_cytomic View Post
Just note when looking at dynamic convolution plugins like liquid mix or nebula, typically they only model a fixed number of harmonics eg nebula does 9, so for a 910 Hz sin wave you will get 910 * 9 = 8190 Hz as the highest harmonic. This won't alias with a sample rate of 44100. What do get are problems when the engine tries to switch between which of it's convolution kernels it's using at any time, which happens quite a lot for compressors.

If the liquid mix and nebula have ssl buss compressor models, would anyone like to process the dry drum loop I posted earlier to compare their sound?
I am so glad you are on here. It is awesome to have a developer in here to set us straight and help keep the facts together.

There just isn't enough time to demo plugs and really shoot them out in their demo period. Especially with sessions in and out and deadlines. Can't afford to buy them all either. Luckily things are slow right now.

Here is the Liquid Mix's emulation of the SSL FX G384 Bus Compressor. 24-bit 44.1 using the same Dry Loop.

Same Attack/Release/Ratio settings. I have it smashing the hell out of it ~20db GR. Maxes out the meter on the LM Device, the plug has more resolution. Makeup gain matched by ear.

It would be interesting to do the Neve-esqe inspired ones too. Anyone have the hardware or DIY ones?
Attached Files

BFD-LoudKick110BPM_LiquidMix-SSLFXG384.wav (2.75 MB, 636 views)

Old 25th May 2010
  #422
Cytomic
 
andy-cytomic's Avatar
 

Quote:
Originally Posted by Robo View Post
What sine frequency should I use for 96khz?
Ok, I've worked these to be as close to 1000 hz as possible, be integer values frequencies, and space the aliased harmonics at close to 400 hz up and down from the main harmonic peaks for easy detection:

44100 -> 993 hz
48000 -> 992 hz
88200 -> 998 hz
96000 -> 996 hz
176400 -> 989 hz
192000 -> 998 hz
Old 25th May 2010
  #423
Gear Maniac
 
Stolle's Avatar
 

Quote:
Originally Posted by miro View Post
ok, i found the 4k preset for nebula. also attached TAL filter (free) which i think is a very good filter:
I think the FatFunker is actually a TL audio FATMAN.
Old 25th May 2010
  #424
Gear Addict
 
rob61's Avatar
 

Does this have to do with 24 bit math?

So is the aliasing peaks found due to integer math rounding errors? Would the results change if tested using 32 bit floating point instead of 24 bit integer? I know Nuendo does both. I usually use 24 bit, but I'm wondering if 32 bit floating would help this out and offer a reason to actually use 32 bit float in the DAW.
Old 25th May 2010
  #425
Lives for gear
 
miro's Avatar
 

Quote:
Originally Posted by Stolle View Post
I think the FatFunker is actually a TL audio FATMAN.



this is what happenes when they don't tell what is what!

sorry!

do they have an ssl comp though?


but i found another SSL type compressor plugin---->

The Conspiracy Buss Compressor...(which was quiet amazing)

2 screenshots + tried to match the settings to the Glue for audio example:
Attached Thumbnails
Lets do it: The Ultimate Plugin Analysis Thread-conpiracy1.jpg   Lets do it: The Ultimate Plugin Analysis Thread-conspiracy2.jpg  
Attached Files

LoudKICK_Conspiracy_Comp.wav (3.71 MB, 985 views)

Old 25th May 2010
  #426
Lives for gear
 
dotl's Avatar
 

Quote:
Originally Posted by aortizjr View Post
Here is the Liquid Mix's emulation of the SSL FX G384 Bus Compressor. 24-bit 44.1 using the same Dry Loop.
It behaves nice but the highs are slightly (well...more then slightly) filtered.
Old 25th May 2010
  #427
Gear Maniac
 
Stolle's Avatar
 

Quote:
Originally Posted by miro View Post

do they have an ssl comp though?
I don't think so. They haven't really worked much on compressors since the release. I think they're waiting to upgrade some part of the technology for compressors.
Old 25th May 2010
  #428
Lives for gear
 
Graal's Avatar
 

Old 25th May 2010
  #429
Gear Addict
Time for Eqs?

Ok, I'll try to explain some questions about digital Eqs.
I'm not gonna state if a Eq plugin is good or bad, or the way it sounds or if you must like it or not.

First you need to download the vst plugin analyzer
Christian’s Blog

Open the program and load the plugin you want to check.
Then measurement/frequency response
Now choose a band (is you are testing a parametric Eq) and pump up the gain a few db. You'll see the bell.
Now move to the right the bell.

Some Eq plugins have nice bells in the low field, but the problem comes when you approach the nyquist zone (over 20k).

If you see a deformation inthe bell as you approach the 20k area, then the plugin is not being "analog like".

Analog Eqs has the same bell EVER, no deformation at all in that high freq area.
If the plugin doesn't shows any distortion in this area, then is working similar to how the hardware do.

Let's see some graphics.
If you can see the deformation in the right zone the plugin is not working with analog style, although maybe you love the sound.
If you see no deformation, then the plugin is working similar to an analog Eq.

Same with LP Filters. HPF used to be fine in every digital Eq.

Same with the phase distortion. If the curve has deformation is not being "analog like". See attached 4 & 5.
If you can check the phase distortion in the vst plugin analyzer just press P in your keyboard.

Analog hardware Eqs have perfect coherence in bells, filters and phase. There is no deformation at all with hardware Eqs. Any distortion of the curves is a digital issue.

Now is time to check some plugins if you want to. I'll upload some plots soon. Get ready for big surprises.


More info:
Sonimus.com Artículo - Tutoriales - Eligiendo nuestro armamento de plugins III: La Fase
Attached Thumbnails
Lets do it: The Ultimate Plugin Analysis Thread-mal-lpf.jpg   Lets do it: The Ultimate Plugin Analysis Thread-bien-lpf.jpg   Lets do it: The Ultimate Plugin Analysis Thread-mal-band-freq.jpg   Lets do it: The Ultimate Plugin Analysis Thread-fase-bien.jpg   Lets do it: The Ultimate Plugin Analysis Thread-fase-mal.jpg  

Old 25th May 2010
  #430
Lives for gear
 
miro's Avatar
 

Quote:
Originally Posted by pacorro View Post
Time for Eqs?

Ok, I'll try to explain some questions about digital Eqs.
I'm not gonna state if a Eq plugin is good or bad, or the way it sounds or if you must like it or not.

First you need to download the vst plugin analyzer
Christian’s Blog

Open the program and load the plugin you want to check.
Then measurement/frequency response
Now choose a band (is you are testing a parametric Eq) and pump up the gain a few db. You'll see the bell.
Now move to the right the bell.

Some Eq plugins have nice bells in the low field, but the problem comes when you approach the nyquist zone (over 20k).

If you see a deformation of the bell as you approach the 20k area, then the plugin is not being "analog like".

Analog Eqs have the same bell EVER, no deformation at all in that high freq area.
If the plugin doesn't shows any distortion in this area, then is working similar to how the hardware do.

Let's see some graphics.
First one is DMG Equality in digital mode. You can see the deformation in the right zone.
Second one is DMG Equality in analog mode. No deformation. The plugin is working similar to an analog Eq.

Same with LP Filters. HPF used to be fine in every digital Eq.

Same with the phase distortion. If the curve has deformation is not being "analog like". See attached 3 & 4.
If you can check the phase distortion in the vst plugin analyzer just press P in your keyboard.


Now is time to check some plugins. I'll upload some plots soon. Get ready for big surprises.


More info:
Sonimus.com Artículo - Tutoriales - Eligiendo nuestro armamento de plugins III: La Fase

i wish there was an equal program for mac!
Old 25th May 2010
  #431
Lives for gear
 
dotl's Avatar
 

Quote:
Originally Posted by pacorro View Post
Time for Eqs?
Nah...that's boring
Old 25th May 2010
  #432
Lives for gear
 
Jeezo's Avatar
What a great thread , really ....

What's seems crazy is that aliasing and good sounding plugin are not an excact science as some like aliasing comp better than clean ones !

But i want to post an humble question : as we all know that all daw are not created equal (don't compare logic and sonar !!) so would a 64 double precision engine in sonar with its great audio engine and delay compensation have an impact in measurement for the same plugin in anotha daw ?

Even the platform , does it have an impact ?
Plugins are codes , codes are calculated , time is a function , latency and it's compensation one more value , so the end result might be diffrent no ?

By the way thks for you all to feed with great exemples and screen shot , will try this on my duende deeper ...

I also noticed that the oversampling in the FET softube comp is beautifull soundin , we loose high , but some time this is what i want (electro snare and all ..)and i had the feeling of a great reaction attack and realise speaking ....
Old 25th May 2010
  #433
Lives for gear
 

Quote:
Originally Posted by Graal View Post
Could someone please test this plug-in? It's FREE!

FerricTDS updates to 1.5

Ferric TDS.

Cheers!
Okay, I used Free-G to trim a few db so the input meter wouldn't be in the red. This is -6db reduction, the second one is with the saturation knob turned all the way up.
Attached Thumbnails
Lets do it: The Ultimate Plugin Analysis Thread-ferric-6db-reduction.jpg   Lets do it: The Ultimate Plugin Analysis Thread-ferric-6db-reduction-full-saturation.jpg  
Old 25th May 2010
  #434
Cytomic
 
andy-cytomic's Avatar
 

Quote:
Originally Posted by miro View Post


this is what happenes when they don't tell what is what!

sorry!

do they have an ssl comp though?
Miro, I think the 4k is meant to be the ssl bus comp, they just got the label wrong - there are no valves in it!
Old 26th May 2010
  #435
Lives for gear
 

Quote:
Originally Posted by Jeezo View Post
Even the platform , does it have an impact ?
Plugins are codes , codes are calculated , time is a function , latency and it's compensation one more value , so the end result might be diffrent no ?
From my experience doing null tests, plugins perform the same regardless of which daw it's in....but there are probably exceptions....but that is more likely due to a difference in the plugin coding, than the daw.
Old 26th May 2010
  #436
Lives for gear
 
Graal's Avatar
 

Quote:
Originally Posted by Tangible View Post
Okay, I used Free-G to trim a few db so the input meter wouldn't be in the red. This is -6db reduction, the second one is with the saturation knob turned all the way up.
WOW. COOL!
I hope you did it right, cause the results seem to be awesome and this is a free plug-in.

I used it on my mixes, and it sounded good, but I was skeptical as well.
I thought it probably had something I couldn't hear that was actually making my mix hurt in some way. Now I know there is nothing to fear.

Cheers.
Old 26th May 2010
  #437
Lives for gear
 

Quote:
Originally Posted by Graal View Post
WOW. COOL!
I hope you did it right, cause the results seem to be awesome and this is a free plug-in.

I used it on my mixes, and it sounded good, but I was skeptical as well.
I thought it probably had something I couldn't hear that was actually making my mix hurt in some way. Now I know there is nothing to fear.

Cheers.
The guy that made those plugs won awards for the best free plugins, so I wouldn't think you'd have to worry much about his coding skills. Never really used them much myself though. As plugins come down to programming and not physical parts, i.e. a cheap piece of hardware is usually made with inferior circuitry and materials, I don't think there's any direct correlation between the price of software and it's effect on audio fidelity.
Old 26th May 2010
  #438
Lives for gear
 
Surbitone's Avatar
Quote:
Originally Posted by Tangible View Post
I don't think there's any direct correlation between the price of software and it's effect on audio fidelity.

Totally. I think it's in the best interests of someone like Bootsy to really put his heart n' soul into it, both for pleasure & business - hence his awards
Old 26th May 2010
  #439
Lives for gear
 
SeniorityFedup's Avatar
 

Quote:
Originally Posted by andy_cytomic View Post
The cubase comp has a 1mS hold time, which is 1000 Hz, so if the test signal is 1000 Hz there will be no harmonics generated, which is what is shown in the plots.
Awww
Old 26th May 2010
  #440
Lives for gear
 

Here's an interesting one. Alex B's MBC console run using the regular Nebula and then Nebula Reverb. This is the MKI version of MBC not MKII PRO, I didn't install it yet. I also did the RN5088 Mixbuss pre, with regular Nebula, then Nebula reverb. I lowered the input gain inside of Nebula on all test, so the signal hit just below the red meters (stuff generally gets ugly when you overload any of the preamp programs).

I also tested many others, but I got a slow internet connection, so I'm not going to upload them....but moral of the story with Nebula, is to use the Reverb version of Nebula with the saturation programs, it doesn't seem to matter with the EQ's though (as far as aliasing is concerned.....however, on listening tests, the Reverb version is better on EQ's also).

*don't pay attention to the peak/rms meter in the back, the sine is being played on a muted bus, I'm also testing the plugs on some drums, which is what is showing on the peak/rms meter.
Attached Thumbnails
Lets do it: The Ultimate Plugin Analysis Thread-alexb-mbc-nebula.jpg   Lets do it: The Ultimate Plugin Analysis Thread-alexb-mbc-nebula-reverb.jpg   Lets do it: The Ultimate Plugin Analysis Thread-rnd5088-nebula.jpg   Lets do it: The Ultimate Plugin Analysis Thread-rnd5088-nebula-reverb.jpg  
Old 26th May 2010
  #441
Gear Addict
 
Robo's Avatar
 

Quote:
Originally Posted by andy_cytomic View Post
Ok, I've worked these to be as close to 1000 hz as possible, be integer values frequencies, and space the aliased harmonics at close to 400 hz up and down from the main harmonic peaks for easy detection:

44100 -> 993 hz
48000 -> 992 hz
88200 -> 998 hz
96000 -> 996 hz
176400 -> 989 hz
192000 -> 998 hz
Thanks Andy, that's great work.

Cheers
Robo
Old 26th May 2010
  #442
Lives for gear
 
Eloheim's Avatar
 

Quote:
Originally Posted by Tangible View Post
The guy that made those plugs won awards for the best free plugins, so I wouldn't think you'd have to worry much about his coding skills. Never really used them much myself though. As plugins come down to programming and not physical parts, i.e. a cheap piece of hardware is usually made with inferior circuitry and materials, I don't think there's any direct correlation between the price of software and it's effect on audio fidelity.
Yeah and if you read up on his site/blog it seems like his stuff is actually popular with a good amount of mastering engineers so that makes me think they have to be 'clean' at a base level.
Old 26th May 2010
  #443
Gear Maniac
 
gavriloP's Avatar
 

Hey Tangible!

I saw that you are still using older version of Ferric... there is an update on his site. Downloads
It has few good improvements, but word of warning, it's presets are not compatible with older one. So if you load a project with Ferric it resets all controls to default positions (if I remember right). Anyway, worth checking, just like his new Nasty VCS virtual console strip. It has very nice saturation FX and also phase tools, nice EQ and optocomp. Bootsie is a part of a royal family when it comes to PC freeware.
Old 26th May 2010
  #444
Gear Addict
 
Robo's Avatar
 

Cubase TestGenerator producing harmonics of it's own?

Just did a quick test using the Cubase 4 tone generator (might be a legacy plug actually) and on it's own it seems to produce strong harmonics... Pretty useless if it's true. I'll double check I've not done something wrong, but don't think I have.
Old 26th May 2010
  #445
Gear maniac
 

it does
Old 26th May 2010
  #446
Lives for gear
 

Quote:
Originally Posted by gavriloP View Post
I saw that you are still using older version of Ferric... there is an update on his site. Downloads
It has few good improvements, but word of warning, it's presets are not compatible with older one. So if you load a project with Ferric it resets all controls to default positions (if I remember right). Anyway, worth checking, just like his new Nasty VCS virtual console strip. It has very nice saturation FX and also phase tools, nice EQ and optocomp. Bootsie is a part of a royal family when it comes to PC freeware.
thanks, I've just recently moved over to doing my mixing mostly on my mac, but I'll definitely check for it if I go back to mixing on my PC.


It's funny that just a few minutes ago, I was reading through this article from Tape Op Magazine, and I came across a reason why the makers of Waveart's Tube Saturator plug-in asserts developers leave aliasing in their plugins:
"Aliasing is digital garbage; I've only ever heard aliasing sound good on Tom Yorke's The Eraser, where I am pretty sure he did it on purpose. Anti-aliasing filters remove the offending frequencies before the D/A. I asked, 'Why no anti-aliasing?' Simple, it increases CPU load even more, and they couldn't hear the difference. Plus, the claim is that no anti-aliasing is needed at 88.2kHz and above, so the manual suggests that if you can hear aliasing, up-sample before processing. The manual suggests up-sampling if you want to increase the resolution of the distortion modeling, and that it will be preserved when you down-sample. I did listening tests to confirm both of these assertions. "

His statement is intriguing and his "solution" also seems to match my earlier post about the conclusion I came to after running tests. For plugins that introduce aliasing, you will get cleaner results if you upsample to 88.2 and above and then down-sample back to 44.1kHz. On the other hand, plugins that don't alias, typically sound the same at 44.1k as they do when they were up sampled and processed, then downsampled.

I wonder what Andy has to say about the comment that no anti-aliasing is needed above 88.2 though.
Old 26th May 2010
  #447
Lives for gear
 

Quote:
Originally Posted by Tangible View Post
Here's an interesting one. Alex B's MBC console run using the regular Nebula and then Nebula Reverb. This is the MKI version of MBC not MKII PRO, I didn't install it yet. I also did the RN5088 Mixbuss pre, with regular Nebula, then Nebula reverb. I lowered the input gain inside of Nebula on all test, so the signal hit just below the red meters (stuff generally gets ugly when you overload any of the preamp programs).

I also tested many others, but I got a slow internet connection, so I'm not going to upload them....but moral of the story with Nebula, is to use the Reverb version of Nebula with the saturation programs, it doesn't seem to matter with the EQ's though (as far as aliasing is concerned.....however, on listening tests, the Reverb version is better on EQ's also).

*don't pay attention to the peak/rms meter in the back, the sine is being played on a muted bus, I'm also testing the plugs on some drums, which is what is showing on the peak/rms meter.

The Reverb version has higher quality settings so its no surprise it sounds better.
Old 26th May 2010
  #448
Lives for gear
 

Quote:
Originally Posted by Stolle View Post
I think the FatFunker is actually a TL audio FATMAN.

Your right it is a TL Audio FATMAN
Its a great sound Neb preset though

The 4k stands for 4 kernals
Old 26th May 2010
  #449
Lives for gear
 

Quote:
Originally Posted by redroom View Post
The Reverb version has higher quality settings so its no surprise it sounds better.
yeah, I knew it was better, by I was still surprised by how much cleaner it was. Night and day as far as aliasing is concerned.
Old 26th May 2010
  #450
Cytomic
 
andy-cytomic's Avatar
 

Quote:
Originally Posted by Tangible View Post
...I wonder what Andy has to say about the comment that no anti-aliasing is needed about 88.2 though.
This statement is false in general, but may well be true for the specific plugin they are talking about if the aliasing is already low at 44100.

The minimum sample rate required to contain the aliased harmonics is completely dependent on the process that introduces the aliasing.

For instance if you produce a hard clipped waveform from a sin wave by doing: if (sin > 0) 1 else -1, it has lots of aliasing as shown in the attached image, where the black dot indicates the only frequency that should be present. If you oversample to 5 MHz you can lower the aliasing to -80 dB, which is still not good - oversampling alone is not enough.
Attached Thumbnails
Lets do it: The Ultimate Plugin Analysis Thread-clippedsin88200.jpg  
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…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
n8tron / Music Computers
258
timbreman / Electronic Music Instruments and Electronic Music Production
22
Realziment / So much gear, so little time
6
djpolaco / High end
2
jchadstopherhuez / So much gear, so little time
11

Forum Jump
Forum Jump