Are we close to getting plugin compressors that sound like real ones?

[munchkin]

I've never used high end or rare analogue equipment so I can't compare but having used digital compressors and listened to a lot of music that is probably almost exclusively digital I think the digital sound has become ubiquitous.

I don't use or know anyone who uses the old analogue synths. Perhaps some musicians still do but I assume most contemporary dance musicians don't - simply because of the inconvenience of integrating them into a digital environment.

The point I'm trying to make is that software emulations are like The White Stripes versions of old Led Zep/Grand Funk Railroad music. They're what people today think the old stuff sounded like - a rose tinted view of the past in plugin form.

I don't think there's anything wrong with that because I doubt that few of us would like to be in a smoke filled studio trying to sync tape machines while the drummer is overdosing in the corner. We like our past a bit tidier.

[Uncle E]

soorry,but its nothing close to what Shy wanted to hear

based on your previous posts here Im sure you have to hear the differences yourself.../...not so much compressed, frequency response is very different, snare sound is muffled.../

Yes, I didn't post this as an example of succeeding, I posted it because Himalaya asked me to, to show that even the crappy Maschine compressor wasn't falling apart at the settings I've got it at.

That said, have another listen to the two parts back-to-back and tell me if they're still as dramatically different as you describe above:

www.jrrshop.com/eric/eric_dahlberg-shymaschinetest.wav

[kvaca]

soorry,but its nothing close to what Shy wanted to hear

based on your previous posts here Im sure you have to hear the differences yourself.../...not so much compressed, frequency response is very different, snare sound is muffled.../

Yes, I didn't post this as an example of succeeding, I posted it because Himalaya asked me to, to show that even the crappy Maschine compressor wasn't falling apart at the settings I've got it at.

That said, have another listen to the two parts back-to-back and tell me if they're still as dramatically different as you describe above:

www.jrrshop.com/eric/eric_dahlberg-shymaschinetest.wav

yes,still different like night and day to me,sorry

I know youve improved attacks a bit but still they sound muffled compared to Shy Dragon,overall your example is still hearably quieter,less compressed and more dull...and Im starting to think that you cannot do much more with your compressor becouse if you increase volume it will distort peaks and if you increase compression it will fall apart...

And-theres no need to include Dragon demo /only with volume decreased/ in your file becouse I got it already,but I suppose you should balance the volume /peak and RMS/ of your demo according to original Dragon next time...same as Shy did before with Splitcomp /and same as I have done with my compdemo dry+/,becouse thats the way Im listening to it and also comparing it. Thank you for your effort.

[frankjcc2003]

Who knows exactly what it is about hardware that allows it to be pushed harder without clipping? surely we can get a comparable sound with software, but when I try my compressor emus, The sound, as in behavior, is very close but It will clip if I try to get the level matched by ear. I only imagine some subtle distortions are responsible for it's loudness over software. can someone enlighten me on this?

[Uncle E]

I know youve improved attacks a bit but still they sound muffled compared to Shy Dragon,overall your example is still hearably quieter,less compressed and more dull...and Im starting to think that you cannot do much more with your compressor becouse if you increase volume it will distort peaks and if you increase compression it will fall apart...

Yes, I agree. I tried every single Native Instruments compressor and couldn't make any of them sound like Shy's Dragon example. I even used the crossover in Guitar Rig to split the signal to two separate compressors (an 1176 on the bottom and an SSL Buss Compressor on the top got the closest), which technically makes it comparable to the Split Comp, yet even that couldn't get anywhere close. The more I tried, the more I appreciated the sound of the Dragon!

[Agreed]

Who knows exactly what it is about hardware that allows it to be pushed harder without clipping? surely we can get a comparable sound with software, but when I try my compressor emus, The sound, as in behavior, is very close but It will clip if I try to get the level matched by ear. I only imagine some subtle distortions are responsible for it's loudness over software. can someone enlighten me on this?

Hardware has effectively unlimited headroom insofar as it will continue to do things to the sound, with increasing levels of distortion, etc. (which can often sound pleasing), and the way we interface with hardware in terms of gain staging is very different than how we interface with a fully ITB setup. The AD/DA converters are not going to react like the input on a real Fairchild or Pultec or LA2A or whatever when you slam 'em, they're just gonna misbehave because you're exceeding their operational parameters.

It is still totally possible to get an awful sound with hardware, plenty of albums that suck were recorded at great studios to attest to that. But the way it's treated in the tracking process especially is very different from how, by necessity, we treat things in tracking digitally.

I imagine hardware would lose some of its "mojo" if you just stuck a straightforward AD/DA converter right before it with no volume knob (so you couldn't exceed the headroom of the converter).

Modeling the nonlinearities and distortion characteristics of hardware that's pushed isn't easy but it's mathematically knowable and therefore, at least in my view of how the universe works, can be simulated accurately. But there are some systemic limitations right now which need to be taken into account in order to get the best out of software.

It's not even a better or worse thing, just different. Yeah, maybe you can't track into a totally digital setup and just slam the input (minus the mics and preamps ), but that doesn't mean it's bad, it just means you have to treat the gain staging differently than if you were using hardware. Analog gear tends to be much more forgiving of excessive signal levels than digital emulations, but that doesn't mean toss out digital stuff as crap, it just means understand the limitations of your setup and adjust workflow accordingly. E.g. don't slam the crap out of your ADC.

Software can always get better. It has got a lot better just over the last few years. It will continue to do so. Hardware will, too. I don't see the reason for these arguments, honestly, it seems like an excuse for a fight where there doesn't need to be one.

[frankjcc2003]

Who knows exactly what it is about hardware that allows it to be pushed harder without clipping? surely we can get a comparable sound with software, but when I try my compressor emus, The sound, as in behavior, is very close but It will clip if I try to get the level matched by ear. I only imagine some subtle distortions are responsible for it's loudness over software. can someone enlighten me on this?

Hardware has effectively unlimited headroom insofar as it will continue to do things to the sound, with increasing levels of distortion, etc. (which can often sound pleasing), and the way we interface with hardware in terms of gain staging is very different than how we interface with a fully ITB setup. The AD/DA converters are not going to react like the input on a real Fairchild or Pultec or LA2A or whatever when you slam 'em, they're just gonna misbehave because you're exceeding their operational parameters.

It is still totally possible to get an awful sound with hardware, plenty of albums that suck were recorded at great studios to attest to that. But the way it's treated in the tracking process especially is very different from how, by necessity, we treat things in tracking digitally.

I imagine hardware would lose some of its "mojo" if you just stuck a straightforward AD/DA converter right before it with no volume knob (so you couldn't exceed the headroom of the converter).

Modeling the nonlinearities and distortion characteristics of hardware that's pushed isn't easy but it's mathematically knowable and therefore, at least in my view of how the universe works, can be simulated accurately. But there are some systemic limitations right now which need to be taken into account in order to get the best out of software.

It's not even a better or worse thing, just different. Yeah, maybe you can't track into a totally digital setup and just slam the input (minus the mics and preamps ), but that doesn't mean it's bad, it just means you have to treat the gain staging differently than if you were using hardware. Analog gear tends to be much more forgiving of excessive signal levels than digital emulations, but that doesn't mean toss out digital stuff as crap, it just means understand the limitations of your setup and adjust workflow accordingly. E.g. don't slam the crap out of your ADC.

Software can always get better. It has got a lot better just over the last few years. It will continue to do so. Hardware will, too. I don't see the reason for these arguments, honestly, it seems like an excuse for a fight where there doesn't need to be one.

Thank you for taking the time to explain that, and you did it very well. So as it appears, the people who advocate hardware has a right to do so, being that it can do things that software can't. Unless there is a breakthrough in technology that will allow the parameters of coding software to match or exceed the capabilities of hardware, This will always be a valid argument, for the sake of arguing.

However, as close as we are with software being able to replace our old hardware, In the end I don't think it will matter one way or the other, it's just what one may be used to using to get the results they want. Except for maybe 1db more rms in hardware.

[himalaya]

soorry,but its nothing close to what Shy wanted to hear

based on your previous posts here Im sure you have to hear the differences yourself.../...not so much compressed, frequency response is very different, snare sound is muffled.../

Yes, I didn't post this as an example of succeeding, I posted it because Himalaya asked me to, to show that even the crappy Maschine compressor wasn't falling apart at the settings I've got it at.

But the challange was to match the overall level/density/gain reduction in that Slate Dragon demo, not to provide how well behaved your compressor is at your chosen settings. I thought that was obvious.

[himalaya]

I know youve improved attacks a bit but still they sound muffled compared to Shy Dragon,overall your example is still hearably quieter,less compressed and more dull...and Im starting to think that you cannot do much more with your compressor becouse if you increase volume it will distort peaks and if you increase compression it will fall apart...

Yes, I agree. I tried every single Native Instruments compressor and couldn't make any of them sound like Shy's Dragon example. I even used the crossover in Guitar Rig to split the signal to two separate compressors (an 1176 on the bottom and an SSL Buss Compressor on the top got the closest), which technically makes it comparable to the Split Comp, yet even that couldn't get anywhere close. The more I tried, the more I appreciated the sound of the Dragon!

And yet, the plugin I've used has managed to get extremely close to that Dragon sound without any tricks whatsoever. That plugin also costs less than a hundred pounds, so in relation to the boutique price hardware unit it holds its own pretty well, I would've thought. Of course, the price may not matter to the high end studio, still it is good to know that a sub £100 plugin can get so close to a £1600 hardware unit.

It also flies in the face of Shy's alarmist attitude shown in his post on page three, that no plugin can do it, and only one (very expensive one) can, which now I'm convinced is an incorrect assessment.

[Aiynzahev]

Comon, lets stop the bickering and enjoy:

[himalaya]

Hey kvaca,
I've got one of your files here. Sorry for the delay.
This is your 10dbGR file only and my attempt at emulation. This time I don't think I got that close. I can hear the snare in your file being more alive and the kick being softer. Still, my file does not exhibit any of the stuttering artifacts, as far as I can hear. What do you think?

www.electric-himalaya.com/stuff/Compression_match_d.wav

By the way, I will be taking all of the WAV files I posted in this thread off my server soon as they clog up the storage space (I'm running low on space argh!).

Comon, lets stop the bickering

Pardon? Why this comment? Nobody is bickering.

[Fred_Abstract]

the closest to split comp should be ssl X-comp i think , i supect it a striped down version, exaclty like blue and X eq but they didn t advice it to not hurt the other product sell cause of the high price tag , could be true story.. duende dsp plugins have similar quality to the algorithmix one could be same peoples who made them

[kvaca]

Hey kvaca,
I've got one of your files here. Sorry for the delay.
This is your 10dbGR file only and my attempt at emulation. This time I don't think I got that close. I can hear the snare in your file being more alive and the kick being softer. Still, my file does not exhibit any of the stuttering artifacts, as far as I can hear. What do you think?

I think its close enought to rock'n'roll and if you want to be even closer use some combo of fast limiter-clipper after,this will keep your transient more alive and softer Try next one...with ratio 4:1

p.s.- I can hardly imagine some people with expensive hardware to obtain even more GR than 10db in this case...but if its possible I still dont think its anyhow desirable, becouse who will listen to such a brick than?

[himalaya]

Try next one...with ratio 4:1

I have. It's more alive, more dynamic, and at the same time less 'even', but still no 'break up' or 'stuttering'.

[kvaca]

the closest to split comp should be ssl X-comp i think , i supect it a striped down version, exaclty like blue and X eq but they didn t advice it to not hurt the other product sell cause of the high price tag , could be true story.. duende dsp plugins have similar quality to the algorithmix one could be same peoples who made them

very interesting- ty for the info!

so it looks like I need to test duende after all /I ve declined duende comp becouse of the apparently biggest aliasing from all SSL emus - but for drum compression theres no need to avoid aliasing at all costs/