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
Tube compressor shoot-out: Pendulum ES-8 vs QES Labs Variable-Gm Dynamics Processors (HW)
Old 27th November 2009
  #1
Gear Maniac
 
4blades's Avatar
 

Tube compressor shoot-out: Pendulum ES-8 vs QES Labs Variable-Gm

Here's your chance for a comparison between these two tube compressors:
Pendulum Audio ES-8 remote cutoff limiter, and
QES Labs Variable-Gm stereo compressor.

In the 2-bus, chain was:
aurora>compressor>aurora

Have fun.
Attached Files

Likyloo_flat.wav (5.94 MB, 6162 views)

Likyloo_es8.wav (5.94 MB, 6868 views)

Likyloo_Variable-Gm.wav (5.94 MB, 6584 views)

Old 27th November 2009
  #2
Gear Addict
 
krid's Avatar
 

It's the first time that I hear this brand : QES Labs.
The Variable-Gm is louder than the ES-8 but it sounds good, it's not too colored : I like it.
How much is their comp ?

Thanks for the test.
Old 27th November 2009
  #3
I like both compressors, but I like the ES-8 a tad more.
Old 27th November 2009
  #4
Lives for gear
+1 for ES-8
Old 27th November 2009
  #5
Gear Maniac
 
bewareofdogs's Avatar
 

My vote goes to the QES Labs Variable-Gm.
Absolutely.

Nice test indeed.

Old 27th November 2009
  #6
Lives for gear
Okay after a first try on my crappy laptopspeakers I listened to the files in the studio.
First of all the levels had to be matched, as the difference is around 2,5 db !!!
Second, the variable-GM is much more compressed than the ES-8, so no real comparison possible.

But in the studio I didn't really like the ES-8 as it darkens the music too much and sounds very squeezed.
Old 27th November 2009
  #7
Lives for gear
Here is my PL-2 for comparison. Please level adjust !
It is on the insert of my mixer, sorry I was too lazy to reconnect it all.
Attached Files

Likyloo PL2.wav (5.94 MB, 3448 views)

Old 27th November 2009
  #8
Lives for gear
 
Joram's Avatar
 

+1 for ES-8. Without doubt.
Old 27th November 2009
  #9
Gear Maniac
 
bewareofdogs's Avatar
 

Quote:
Originally Posted by soundgeezer View Post
Here is my PL-2 for comparison. Please level adjust !
It is on the insert of my mixer, sorry I was too lazy to reconnect it all.
Even if not a tube compressor like the others, I have to say that prefered the PL-2 over the ES-8.

Still my favourite was the Variable-Gm. It has a lot more 3D depth, sculptured low bottom, nice presence.
I confirm my preference.
Old 27th November 2009
  #10
Gear Maniac
 

+1 for the Qes labs variable-gm. Definitely.
Old 27th November 2009
  #11
Gear Addict
 
krid's Avatar
 

Does anyone can throw a Phoenix to the test ?
Old 28th November 2009
  #12
Lives for gear
I love the Phoenix. Take a look at vintageking.com and click "hear the gear" on the left ... they have some samples of the phoenix.
Old 28th November 2009
  #13
Shy
Lives for gear
 

PL-2 is great, maintains the punch perfectly and adds its little something while still being very transparent overall. Doesn't lose bass like the other two.
ES-8 adds a cool vibe to this. Clean and very tight.
The Variable-Gm to my taste is too distorted in this example and loses too much punch in the attacks, but I think that's mainly because the compression setting was too extreme. I'd like to hear it with settings closer to the ES-8 example.

Thanks for the samples.
Old 2nd December 2009
  #14
Gear Maniac
 

After a more careful listening through good monitors, I have to say that the comparison between the Variable-Gm and the ES-8 is embarassing.
The es-8 looks like a digital plug-in, too bright for my tastes, and squeezed.
The stereo openness the vari-gm imparts to the 2-bus is wonderful, nice also with the pl-2, didn't like that of the es-8.
Again, in the lower end, I liked the variable-gm more of the other two.
ES-8: It was more like I was hearing a little harder sound in the mix, it did not handle the mix the same way as the others.
PL-2: very transparent, across the whole spectrum. A bit cold.
Variable-Gm: my favourite one, offered a big, clear, sound with greater definition. The mix sounded larger and still very musical.

Thanks for the test.
Old 2nd December 2009
  #15
AvS
Lives for gear
What are the settings? The ES-8 sample sounds like it has a longer attack and release than the QES. Levels are pretty far off also.
Old 20th December 2009
  #16
Gear Maniac
 
Resis's Avatar
 

QES Labs Variable-Gm:
I was quite taken with it.
For me, the real test was the stuff.
It sounds terrific. It's great, and I love it.
It was a no-brainer for me. Out of the three units, the variable-gm was hands-down the best.

Quote:
Originally Posted by Shy
The Variable-Gm to my taste is too distorted in this example and loses too much punch in the attacks, but I think that's mainly because the compression setting was too extreme. I'd like to hear it with settings closer to the ES-8 example.
It seems that the variable-gm has a slower attack setting.
Anyway I liked it, loved to hear the original attack.

Probably a blind test would have been useful.
Anyway, the QES labs unit wins for now.
Old 11th March 2010
  #17
Gear Addict
 
fradoca's Avatar
the qes variable gm is clearly the winner.
I really really love the sound it produces!

thumbs up for the qes varible gm!
Old 11th March 2010
  #18
Lives for gear
 
everythinglouder's Avatar
 

I think the Pendulum sounds incredible.

Thanks for the test, it showed me how strong this unit can be. I'll have to get one in for demo'ing.
Old 13th March 2010
  #19
Gear Maniac
 
8inthemorning's Avatar
The Variable-Gm really wins hands down!
Old 14th March 2010
  #20
Quote:
Originally Posted by 8inthemorning View Post
The Variable-Gm really wins hands down!
I dig it too
Old 2nd September 2012
  #21
Registered User
Gearslutz much good

I was searching for audio samples of the Pendulum devices in the last half year or so for 10 to 15 hours or even more. Here they are! Thank you so much for that.

The track is nice. That's a style of rock music I still like to listen to. Who is the artist?

First I listened to the samples through headphones (Sony MDR-7506, a little Bass and Hi-Mids heavy, and AKG K-702, more even response). Both gave me the same impressions. I did the level-matching manually.

My favourite is the sample with the Pendulum ES-8. The stereo-mid has almost all of the attack of the original. The attack phases of Hihat, Snare, Kick, Bass are very little affected. And listen to that bass! The first notes come effortlessly deep and with a lot of authority. Gave me the shivers. The stereo-side is less emphasized than with the QES Labs Variable-Gm. But I like the bit of dark coloration and depth that the ES-8 introduced. So in the end the ES-8 sample had the most space in it. For my taste this supports the feel of the music without getting in the way of the music. The localization is also very good. A very stable stereo-image.

The Pendulum PL-2 sample impressed me by how close it stays to the original (with level matching). The Hihat seemed clearer than with the ES-8, the snare a little more squashed. But I would miss the sonic quality introduced by the ES-8.

In the sample of the QES Labs Variable-Gm the midrange seems to be emphasized which is good for the guitar. Also, even though the extreme stereo-positions seemed to be more emphasized (the delay on the guitar), the overall image is more compact or said another way less detailed than the ES-8. The attack phases of all instruments are more influenced - especially the snare, kick and the bass do sound less dynamic, to my ears. This compact sound seems to also make the stereo-image just a tiny little bit nervous.

Then I repeated the comparison through monitor loudspeakers. Again the QES Labs Variable-Gm produced more like a wall of sound image, while the ES-8 allowed for more separation between the instruments, e.g. the acoustic guitar behind the solo guitar or the snare still distuingishable as a separate instrument right next to the solo guitar. Also the ES-8 preserved a sense of a 3D natural space. I guess what some heard as being overly bright in the ES-8 sample are the still intact transients, and admittedly the sample of the Variable-Gm sounds rounder. But then I listened to the details of the articulation in the solo guitar or the pearly, sparkling yet still warm type of distortion the guitarist did choose - I do find these qualities to be presented best in the sample with the ES-8.

Probably that's only my preferences for hearing music. I like it, when every instrument has its own space but all instruments together make the big picture. I also like when music emotionally affects me. That's why details in the articulation of instruments or vocals are important to me. For what I think should be heard in the music I like to work with it probably has to be the ES-8 and half a year or so more before I saved enough $$$*#!*§ to get my mastering compressor/limiter. Why must good gear be so expensive?

Last edited by SoundKlang; 2nd September 2012 at 11:10 AM.. Reason: additionally heard on monitor loudspeakers
Old 5th September 2012
  #22
Gear Nut
 
Adkins's Avatar
I think both are top-notch units.

The track was a pleasure to listen.

The level and settings differences, as said, make the comparison difficult,
the QES Labs Variable-Gm has much more dBs of compression than the ES-8, that has slower attack settings.
Anyway...
I loved the Variable-Gm sample, for my tastes, it's action and sound fits perfectly in to that rock style,
and the sparkle it adds on the mid-high range is extremely interesting on the final mix.
The relationship between the Variable-Gm and the stereo mix is nothing short of love.
It gives the track that certain something that says, “This is a record”, while with the ES-8 sample, I felt there's still need for further processing.
The feeling I had is that the Variable-Gm sample is some step forward to the final version of the mix.
Even though the Variable-Gm has more compression, I noticed that the dynamic action was very well controlled.
The ES-8 was more on the linear side.
Again for my taste I prefer the sound of the Variable-Gm, with a really shimmery high-end and deep, solid bass.
The high-end on the ES-8 was a tad harsh-sounding, while with the Variable-Gm was particularly nice, as it sparkled but remained really smooth.

It would have been interesting to hear the Variable-Gm with less dB of compression and slower attack settings, I can’t say enough about how hi-fi and clear the compressor sounds when used in these more subtle ways on stereo material.
From the samples of the Variable-Gm I heard, the compressor always added a nice layer of harmonic interest to the sound. I can see why mastering engineers are eating these up.

So my pick would be the QES Labs Variable-Gm.
Old 5th September 2012
  #23
Registered User
Avalon Vt 747sp samples

Adkins, yes, these comparisons are probably not too meaningful, but I hope it is possible to get an impression of the character of a device. I would really like to hear the QES Variable-GM with a more subtle compression of the transients. Probably the ES-8 with all its possible different settings can sound as lively and ready for rock as the Variable-GM. And probably even so the QES Variable-GM can sound as stable as the ES-8, too.

I hope the Avalon 747 with its optical compressor section but a Tube Signal Path is not off topic in this thread. I tried to get as close to the ES-8 sample as possible with nearly same loudness. Again, this is not trying to let the 747 do what it could do (e.g. it could be about 1 to 2 dB louder) and who knows what else the ES-8 could have done.

I am surprised how close to the ES-8 result the 747 was able to get, because many people don't like its compressor too much. I did check for digital clipping but still, if I did miss something wrong with these samples, please let me know. It is true, the 747 did not easily catch attack phases of the snare. It seems, it is somehow inert even with the fastest attack setting when using small compression ratios. Maybe the 747 has a very very soft knee characteristic which even with extreme ratio settings behaves more like a type of compression than limiting. Anyway, the solution is to use a very high compression ratio (used here was 20:1) with the fastest attack and release and lower the threshold just enough to catch the peaks of the snare with the meter showing less than ca. 1 dB of reduction (while in fact achieving ca. 2dB loudness). Then raise the release time just a little bit to reduce more of the peak and decay but not the sustain. Then raise the input level to get as close to digital full scale as possible - if necessary readjust the threshold level.

The chain: PC > RME AIO Analog Out > Vovox cable > Avalon Vt 747sp > Mogami cable > Forssell MADC-2 > RME AIO AES/EBU In

The 747 settings

output gain: -6 dB (less output to be able to densify the signal)
input gain: used for same loudness (ca. -3dB/+7dB according to +10dB option)
Threshold: ca. +4,5dB
Ratio: 20:1
Attack: fastest setting
Release: almost fastest setting (see above)
Makup Gain: 0dB
EQ: off (intentionally not used the 747s built-in EQ in this comparison of compressors)

Variations in Tube Signal Path On/Off and +10dB Option On/Off lead to 4 different samples. What do you think? Can the 747 be used in mastering?

Added: Currently (7th Sep. 2012) the ProAudio part of the QESLabs Dotcom Website offers a section "dynamic processors" and there in the lefthand menu an item "audio samples" with dry/wet samples of the Variable-Gm with ca. 5 to 6 dB loudness increase. Also there is a youtube video (search engine "video variable gm") with not so good sound quality but still useful. I think these additional samples together with the above sample do tell something about the basic character of the QES Variable-Gm.
Attached Files

Likyloo_747TSP.wav (5.94 MB, 2962 views)

Likyloo_747TSP10dbOption.wav (5.94 MB, 2773 views)

Likyloo_747NoTSP10dbOption.wav (5.94 MB, 2716 views)

Likyloo_747NoTSP.wav (5.94 MB, 2671 views)


Last edited by SoundKlang; 7th September 2012 at 07:08 AM.. Reason: additional information
Old 6th September 2012
  #24
Lives for gear
 
666666's Avatar
Quote:
Originally Posted by soundgeezer View Post

...First of all the levels had to be matched, as the difference is around 2,5 db !!!

Second, the variable-GM is much more compressed than the ES-8...

...so no real comparison possible...

-
I agree!!!

The settings are way too different for any kind of fair comparison to be drawn. Most of the sonic attributes that people are picking out here are differences between less compression and more compression (differences in threshold, attack, release, etc), not necessarily differences in how the compressors might inherently sound.

And the PL-2, it's pretty much doing "brickwall limiting", not compressing, this is obvious. Again, this is more of a comparison of different TYPES of compression / limiting, not so much a comparison of the inherent characteristics of the units.

It may be reasonable to say that one example sounds "cold" and one sounds "more spacious" or one sounds "more natural" or one sounds "tighter" or whatever... but to RATE the individual UNITS based on these attributes makes no sense whatsoever.

But.... this IS a cool little piece of music, great for testing compressors, I'd encourage more folks to run this bit of audio through more compressors and post here, it's interesting to hear, regardless of differences in settings.

Old 7th September 2012
  #25
Gear Nut
 
Adkins's Avatar
Quote:
Originally Posted by SoundKlang View Post
Variations in Tube Signal Path On/Off and +10dB Option On/Off lead to 4 different samples. What do you think? Can the 747 be used in mastering?

I'm not a big fan of 747, I love units that bring a little something to the party, but I think it's down to personal tastes.

Not sure if this is the correct thread, since the comparison is between 2 tube vari-mu units, maybe you want to post the 747 samples in a dedicated thread.

In your samples, I felt the 747 is quite transparent, overall is a bit boxed and doesn't seem to produce any big, thick sounds, less focus, image and detail than the other units (ES-8 and Variable Gm).
Anyway if you're looking for this particular sound, I cannot see why can't use it for mastering.

Quote:
Originally Posted by SoundKlang
Added: Currently (7th Sep. 2012) the ProAudio part of the QESLabs Dotcom Website offers a section "dynamic processors" and there in the lefthand menu an item "audio samples" with dry/wet samples of the Variable-Gm with ca. 5 to 6 dB loudness increase. Also there is a youtube video (search engine "video variable gm") with not so good sound quality but still useful. I think these additional samples together with the above sample do tell something about the basic character of the QES Variable-Gm.
Nice, still not level adjusted, but reveal the round and solid tone this unit can produce.
The video is low quality, anyway I liked the drums working at 4'36", and the stereo mix at 7'58".
Did a search on YouTube but couldn't find anything on the ES-8.

Cheers.
Old 7th September 2012
  #26
Registered User
Quote:
Originally Posted by Adkins View Post
In your samples, I felt the 747 is quite transparent, overall is a bit boxed and doesn't seem to produce any big, thick sounds, less focus, image and detail than the other units (ES-8 and Variable Gm).
Thanks for the input. I agree with every point.
Old 8th September 2012
  #27
Lives for gear
 
TheBrightSide's Avatar
I thought that the 747 sounded pretty nice, though I would like to hear it with a lower ratio.
Also, a sample with the EQ included would be interesting.
(I just bought a 747 and I am waiting for it to arrive!)

And as already suggested, maybe start another thread with some more audio examples, perhaps a few people might be interested in that.
Old 8th September 2012
  #28
Registered User
I was looking into Adkins remarks, especially the boxiness. Interesting results:

1) Boxiness seems to be caused to a certain degree by high compression ratios (the PL-2 sample sounds more boxed than the ES-8 sample, too, if that is what Adkins means with the term 'boxed') or e.g. using the 747 sidechain EQ for compression or switching its passive graphical EQ in front of the compression instead of after.

2) Samplerates: What if the original was processed with 192kHz and then downsampled to 44.1 kHz - and the samples of the PL-2 and the Avalon 747 now had to use the flat file at 44.1 kHz for their processing? We would have a difference in sound quality which has nothing to do with the used devices nor their settings. Not to speak of the quality of the AD/DA conversion used (will try to get Forssell DA asap). I tried upsampling to 192 kHz - the Hepta SRC of Pyramix is one of the best I found - and, not only in this case, it seems low sample rates also introduce some kind of boxiness (44.1kHz more boxed, 192kHz less boxed). Anyway, all of my 747 samples were done entirely at 44.1kHz (avoiding two SRCs from upsampling to 192kHz and downsampling the result back to 44.1kHz).

3) The 747 compressor section is really almost not able to catch peaks. The loudness in the samples I posted previously must be the result of digital clipping (I did not see this in the Pyramix DAW mixer because my Forssell MADC-2 outputs max. -0.1dB, even when clipping). Sorry for that. Without digital clipping the previously posted samples would have about 1 to 1.5 dB less loudness, which means only very little loudness increase by the 747.

(With the following it is not really a comparison anymore, assuming no limiting was used in front of the ES-8 or the Variable-Gm) With manual peak limiting before the analog processing at -1.5 dBFS in the Pyramix DAW (using a volume automation curve, functionality called 'envelope punch'), it is possible to use lower compression ratios (here: 5 to 1) and increase the attack time a little bit with the 747. This time I also used the 747s graphical EQ with very little adjustments, the biggest being using the low-shelf to filter out some (a little too much) boom.

As Adkins wrote, not sure if this is the correct thread, but then TheBrightSide asked, this will be the last 747 sample I post in this thread. I think: less boxed, a little more big and thick, closer to but still less detail than the ES-8. (here the Tube Signal Path was used, because its warmth was more desirable to me than the slight increase in detail when not using the TSP). On the other hand: nice smooth tone with this bright (in my terms 'california sunshine') Avalon vibe, which was the reason why I wanted the 747.

Compared to the (level-corrected) original and focussing on the articulations in the lead guitar playing: The ES-8 sample seems to polish the signal, making it more 'noble', by taking away something while staying more true to the original, while the last 747 sample seems to add some 'shine' to it, imprinting something onto the signal. And so the value of the Variable-Gm becomes visible to me: the Variable-Gm enhances the articulation without obviously altering it. Still I do not like squashed drums, but the sample of the Varaible-Gm has to my ears 2.8db more loudness than the ES-8 sample - isn't that a lot and enough to get drums squashed?

@Adkins: Again thanks for the input - got me going and made me learn. Very much appreciated.

@TheBrightSide: It seems you are into synthesizers? I found the 747 was bringing true analog vibes to synthetic sounds, too. Crank up the input and make the tubes sing ...
Attached Files

Likyloo_ManualLimitOneAndAHalfdB_747_5to1_TSP_Eq.wav (5.94 MB, 3092 views)


Last edited by SoundKlang; 8th September 2012 at 12:29 PM.. Reason: addition
Old 8th September 2012
  #29
Lives for gear
 
TheBrightSide's Avatar
Hi SoundKlang, thanks for posting the new sample, it sounded much better than the first one. I am really looking forward to getting my 747 now!
A lot of people talk about the Avalon, but yours is the first audio example I have found here, so well done.
Yes I do record a mostly synths, but also a lot of vocals, guitar, bass and drums. I mix ITB, so I really wanted to get the 747 to add some tube warmth, and to help soften the "Bite" that digital can have. I hope that the avalon is good for this.

Thanks again for sharing
Old 8th September 2012
  #30
Registered User
I have to make one more remark about the quality of the 747 samples I posted in this thread. I did a comparison recording with the 747 in my chain and with the 747 unplugged from the same chain but still including all cables. Based on the result I may say that from all the loss in detail in my samples here in this thread at least 50% if not 75% are caused by the current state of my chain (probably the DA). It hurts a little to state this in public, but I have to be fair to Avalon and it is also another hint in the direction of how reliable example soundfiles are, if we do not know more about the chain.

Oh my, and away goes the comp ... next will be high quality DA ...

I followed your advice. Please see the thread Avalon Vt 747sp shootout for more example soundfiles. For the above reason unprocessed soundfiles will be rerecorded through the chain with the 747 unplugged from the chain.
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
theskids / Gearslutz Secondhand Gear Classifieds
4
derekbrin / Gearslutz Secondhand Gear Classifieds
0
bonusroommusic / Gearslutz Secondhand Gear Classifieds
2
Magic Genie / Gearslutz Secondhand Gear Classifieds
34

Forum Jump
Forum Jump