The No.1 Website for Pro Audio
SSL plugins shoot out! (UAD / SSL / Waves)
Old 13th July 2013
  #31
Lives for gear
 
Michael E's Avatar
3, but love the colour of number 2.
Old 13th July 2013
  #32
Gear Maniac
 
Fuseburn's Avatar
 

2 all the way !
Old 13th July 2013
  #33
Gear Head
If i was stuck in a room with that on loop forever i'd choose 3.
but I thought 2 was also pretty cool.
Old 13th July 2013
  #34
Lives for gear
 
dotl's Avatar
 

No.3 I like the best but it's the loudest (less compressed?) and the high end is boosted a bit (or i just perceive like that for other reasons). I like the result but something's not right.

It's followed by No.2
Old 13th July 2013
  #35
M2E
Lives for gear
 
M2E's Avatar
 

I like 2 and 4. 4 seems a little louder. 2 has a thing to it. 4 has pumping going on a lot.
Both sound nice.

Thanks,

Marc
Old 13th July 2013
  #36
Lives for gear
 
jalcide's Avatar
Wow, what a difference in these compressors [EDIT: I retract this statement, considering the test was done with VBC pre-bug-fix and also with the auto, ratio, make-up gain and GR being set so vastly different between files. I'll rephrase it: "wow, what a difference in these results." For fairness sake, a retest would be needed for the VBC bug-fix, alone.]

This is one of those times, that often happens in electronic dance music production, where adding compression can, paradoxically, have the desired effect of adding dynamics rather than reducing them.

#2 shows this the most. It really controls the low end (in a good way). A side effect is that it's also less low-end heavy. This allows the overall RMS of low-mids/mids to increase without raising the overall RMS loudless level (which is highly bass sensitive). Less muddy, punchy, clear, warm might describe a mix like this. For these reasons, it was my favorite.

#1 was similar, but slightly louder and slightly less dynamic. I liked it second best.

#3 was loudest and most compressed (yet still having more compression-induced dynamics than the original). I didn't find anything objectionable about it. To be fair, I'd need to hear it matched (mainly, dynamics-wise) to be closer to #1 and #2 before judging it.

#4 was, dynamics-wise, all over the place. Yikes. It didn't control the bass well and even went so far as fluctuating, in lurching fits and starts, when parts with slightly different dynamics came in! Not solid sounding, and by quite a bit -- it went from being the least loud to the most loud (in the lineup). For EDM, this instability is not at all desirable. I won't speak for other genres as I'm not an expert in them. This was definitely my least favorite for this reason alone. I mean, it almost sounded glitchy/buggy. Like something that needs to be fixed in the product.

I'm not going to guess which is which. Mostly, because I'm terrified that #4 has to be a new product that hasn't been vetted yet. I can't tell you how much I hope this isn't VBC (which I own and have been loving the harmonics it adds).

Erol, thanks for doing this test. Please do PM me the results. I almost don't want to know.

Also, I have a few questions, Erol...

Quote:
Originally Posted by erol View Post
Treshold : between -6 and -12db, depending on the plugin.
Attack : 30ms
Release : 1 or auto, depending on the plugin.
Ratio : 10:1 or 20:1, depending on the plugin.
When you say "1" I assume you mean ".1"? But, they all have an auto, why not use auto on all of them for apples-to-apples? This could potentially explain what I was hearing on #4 (as auto tends to be smoother when it has a groove to work with).

Also, "threshold between -6 and -12db": threshold is labeled differently on some of the plugins' GUIs. Was there a target GR range you were shooting for? I think that might be more telling.

Finally, was the VBC's HPF in its default, disabled position? If not, that could explain some things.

Cheers.
Old 14th July 2013
  #37
Lives for gear
 
jalcide's Avatar
Okay, I've got a working theory and will guess:

#4 might be SSL, because with its 20:1 ratio, that would explain what I'm hearing. More GR, therefore more makeup gain needed. Then, during dynamic changes, you get a more pronounced jump.

#3 I'll say VBC cuz it seems to have some harmonics / thickness going on.

#2 is probably UAD due to some harmonics mojo they'd bring to the table.

#1 is probably Waves, maybe the cleanest sounding. They probably had to play it safe with SSL looking over their shoulder and pulled back on the harmonics stuff.
Old 14th July 2013
  #38
VBC has a subtle more bottom end because of this emulated Neve transformer in the signal path. Because most people like 3 the most, I expect this is the VBC "Grey"
Old 14th July 2013
  #39
Lives for gear
 
stinkyfingers's Avatar
 

VBC is #4
Old 14th July 2013
  #40
Here for the gear
 

I guess 2 is VBC. It sounds best to me. It has an awesome groove.

I think 4 is Waves cause it sounds more smeared and dead.

Can you PM me the results?
Cheers!
Old 14th July 2013
  #41
Gear Addict
 
ABBA's Avatar
 

I prefer nr 2.
Results please (PM).

Thanks for the test.
/Mike
Old 14th July 2013
  #42
Lives for gear
 

2-3 sound best
Old 15th July 2013
  #43
Lives for gear
 
Progmatic-Studios's Avatar
PM's send!

So far:

1: I
2: IIIII II
3: IIIII IIIII IIIII I
4: II
Old 15th July 2013
  #44
It's interesting... the results are not the things I expect... maybe a good time to rethink future investments :P
Old 15th July 2013
  #45
Lives for gear
 
Progmatic-Studios's Avatar
Quote:
Originally Posted by 4damind View Post
It's interesting... the results are not the things I expect... maybe a good time to rethink future investments :P
I also compared the SSL Channelstrips.. in that case, no3 did not sound the best...
Old 15th July 2013
  #46
the funny thing (if I remember this correctly) the UAD SSL G-Bus is on the cleaner side (no modeled transformers) and there was always the comparison between UAD <>Waves and people never found a clear winner.
Some of the other SSL emulations are well known for the additional harmonics (modeled transformers) but... don't having this in mind and only listening... big surprise
Old 15th July 2013
  #47
Quote:
Originally Posted by erol View Post
I also compared the SSL Channelstrips.. in that case, no3 did not sound the best...
Afaik the SSL native channel is from the digital C200 while most of the other emulations are from the analog 4000?
Old 15th July 2013
  #48
Lives for gear
 
Lupez's Avatar
I didn't receive my results...PM please!
Old 15th July 2013
  #49
Lives for gear
 
davidwilson's Avatar
 

Well listening on my earbuds.
I like 1 followed by 3 they have a very cool vibe. 2 seemed a bit too clinical to me. 4 seemed a little week.

results Pleeeese
Old 16th July 2013
  #50
Gear Maniac
 

I have no idea which is which but my order would be

3 - 2 - 1 - 4

curious about the results !!!
Old 16th July 2013
  #51
OJS
Gear Maniac
 
OJS's Avatar
At the risk of not being very original in my findings:

3 (I think this is the Slate VBC)
2 (UAD?)
1 (Waves?)
4 (SSL)

Please send me the results!!
Old 16th July 2013
  #52
Gear Addict
 
Fabrice Gabriel's Avatar
 

Someone brought my attention to this thread,
I found it interesting so I'd like to point out a few things.


The files seems to be dated of the 10th of July, so the VBC version used in these tests was released before the FG-Grey algo update that we did a few days ago, which fixed a bug.

https://www.gearslutz.com/board/new-p...ml#post9223348



Also I did a few measurements on the .wav files,
and here is what we have :



Loudness of the drum parts only (start and ending of the song cut, so measured within about 12s to 1min35s of the dry file) :

Average Loudness (compressed drum part only)
1 : -14,2 dB
2 : -14,4 dB
3 : -13,7 dB
4 : -14,4 dB
Dry : -13,9 dB

So the #3 is louder by 0.7 dB compared to the #2 and #4, which can be related to a very interesting article from Sound on Sound :

Is Loud Music Better?



But more importantly, while the loudness of the drum part remains within the same range, even compared with the dry audio, I measured the loudness of the very end of the song, which is supposed to be not compressed. (between 1min45s55 and 1m46s4 of the dry file)

Loudness of the last second of audio
1 : -47 dB
2 : -49.9 dB
3 : -54.8 dB
4 : -50.5 dB
Dry : -58.3 dB



So we can calculate which make up gain has been used compared to the dry file for each processed audio file.

Unless the compression timing is so different (one very long release time for instance) that the last second of audio is still affected by gain reduction on some processed audio files.. Which can happen with the auto release on, which is changing the compression sound and behavior a lot, especially with an SSL type compressor.

So, if the auto release or timing discrepancies are not influencing the last second of audio gain reduction (meaning the gain reduction came back to zero), we have :

Make up gain
1 : 11,3 dB
2 : 8,4 dB
3 : 3,5 dB
4 : 7,8 dB


Otherwise, the timing (auto release or release) being way different, this becomes tricky to make any comparison.


If we assume that the last second of audio is not affected by any gain reduction, and if we also get the processed loudness difference into the calculation, we can get the average gain reduction of the compression for each processed audio example, during the drum part :

Average gain reduction
1 : 11,6 dB
2 : 8,9 dB
3 : 3,3 dB
4 : 8,3 dB


Hope this helps !




Fabrice
Old 16th July 2013
  #53
Lives for gear
 
Progmatic-Studios's Avatar
@Fabrice,

Thank you for replying! I'm not a scientist like you guys and I only tried to match the sound, not the settings. If you want, you can try the latest VBC on the dry WAV file to match the no.3 sound a lot of people here prefer?! I did use the first VBC release without the fix btw.
Old 16th July 2013
  #54
Gear Maniac
 

Interesting. I picked #2 based on perceived dynamics and given Fabrice's input, I would have thought I would pick #3 for dynamics given the lower gain reduction of that file. The fact that 2 sounded better to me with so much more gain reduction was unexpected.

Last edited by 1rechtien; 16th July 2013 at 05:23 PM.. Reason: typo
Old 16th July 2013
  #55
Gear Addict
 
Fabrice Gabriel's Avatar
 

Quote:
Originally Posted by erol View Post
@Fabrice,

Thank you for replying! I'm not a scientist like you guys and I only tried to match the sound, not the settings. If you want, you can try the latest VBC on the dry WAV file to match the no.3 sound a lot of people here prefer?! I did use the first VBC release without the fix btw.
Well I don't know if it's my role to make plugin comparisons.. What I can tell is that the best way to do it is to have exactly the same make up gain and average gain reduction in order to compare the things.
Also indeed, the new FG-Grey version may have a different sound that the one you used for processing the VBC audio example on your system.


Quote:
Originally Posted by 1rechtien View Post
Interesting. I picked #2 based on perceived dynamics and given Fabrice's input, I would have thought I would pick #3 for dynamics given the lower gain reduction of that file. The fact that 2 sounded better to me with so much more gain reduction was unexpected.
Don't forget that this compressor design has an auto release, which, once engaged, does not make the average gain reduction necessarily match the inherent dynamics of the gain reduction, because of the auto release behavior.

Which means that a comparison should be made with the same timing and auto release adjustements.



Fabrice
Old 16th July 2013
  #56
Lives for gear
 
jalcide's Avatar
Quote:
Originally Posted by 1rechtien View Post
...I would have thought I would pick #3 for dynamics given the lower gain reduction of that file. The fact that 2 sounded better to me with so much more gain reduction was unexpected.
with the original being most compressed (or at least simulating that from its low-end distribution), and with the slow 30ms attack, less GR will often equate to less compressor-induced dynamics. #2 having most GR, with that slow attack, will be shaping those nice punchy sounding transients.

i chose #2 for the same reason: more compressor-induced dynamics (which i'm a fan of for this style of music).

often, during mastering, i'll find increasing the GR, with a slow attack, will restore dynamics and pull it back into "green" territory on the DR meters -- DR value gets larger (less compressed). it's a paradoxical effect.
Old 16th July 2013
  #57
Lives for gear
 
jalcide's Avatar
Quote:
Originally Posted by Fabrice Gabriel View Post
...comparison should be made with the same timing and auto release adjustements.
If anyone ever does a retest, this (above), along with post-bug-fix VBC, exact make-up gain and matched GR, it'd also be interesting to hear the 4:1 ratio being used (which all units have in common).

The overall threshold/GR could still be dialed up to pronounce the effect.

I'm more a fan of 2:1/4:1 for buss compression, anyway. 10:1 / 20:1 is really more limiting than compression.

I never have really liked compressors being used as limiters. I think specialized peak-limiters, with multiple internal stages in them, like Pro-L, etc. do better at those ratios.

The one exception to this (to my ears / style of music) might be the Brainworx VSC2, which seems to limit in an interesting way (sometimes).
Old 16th July 2013
  #58
Gear Maniac
 

Quote:
Originally Posted by jalcide View Post
with the original being most compressed (or at least simulating that from its low-end distribution), and with the slow 30ms attack, less GR will often equate to less compressor-induced dynamics. #2 having most GR, with that slow attack, will be shaping those nice punchy sounding transients.

i chose #2 for the same reason: more compressor-induced dynamics (which i'm a fan of for this style of music).

often, during mastering, i'll find increasing the GR, with a slow attack, will restore dynamics and pull it back into "green" territory on the DR meters -- DR value gets larger (less compressed). it's a paradoxical effect.
Thanks for this feedback. Your explanation is a factor that is simple but also turned on a light that I wasn't able to reach on my own.
Old 16th July 2013
  #59
Lives for gear
 
Progmatic-Studios's Avatar
Quote:
Originally Posted by jalcide View Post
If anyone ever does a retest, this (above), along with post-bug-fix VBC, exact make-up gain and matched GR, it'd also be interesting to hear the 4:1 ratio being used (which all units have in common).

The overall threshold/GR could still be dialed up to pronounce the effect.

I'm more a fan of 2:1/4:1 for buss compression, anyway. 10:1 / 20:1 is really more limiting than compression.

I never have really liked compressors being used as limiters. I think specialized peak-limiters, with multiple internal stages in them, like Pro-L, etc. do better at those ratios.

The one exception to this (to my ears / style of music) might be the Brainworx VSC2, which seems to limit in an interesting way (sometimes).
I will try to do this tomorrow!
All 4 plugins with exact make-up gain, matched GR and ratio 4:1.
Old 17th July 2013
  #60
Lives for gear
 
jalcide's Avatar
Quote:
Originally Posted by erol View Post
I will try to do this tomorrow!
All 4 plugins with exact make-up gain, matched GR and ratio 4:1.
nice! you're a god among audio engineers.

this should be very interesting.

hmmm, maybe we should test the glue and fxpansion's dcam ssl buss compressor, too (i've got both). seize this rare opportunity to get a comparison against the UAD plugins.

Quote:
Originally Posted by 1rechtien View Post
...simple but also turned on a light...
yeah, that happens to me too, on here. glad to have helped.
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
Forum Jump
Forum Jump