Voxengo Marquis Compressor VST 1.4 released

[scam_artist]

Has anyone realised just from reading this thread that we have two absolutely brilliant minds here at work? Imaging what they could accomplish if they join forces. That is my take on it.

You're fcuking joking ttoz! You literally said, word for word, what I was gonna say... but then I saw you had already said it!

I totally agree. I remember when I bought the "Warm colors bundle" of Crunchessor and GPP, and was impressed by the professional nature of it. And now you're both arguing, what's happened? You both develop fantastic products and are leaders in the comp/eq/limiting area IMO, so if you ever did join forces you would fidn great success I would think.

Devine Machine (Link and Steve) work on other sides of the world from each other, and you guys aren't that far apart, so there's no reason why it couldn't work - just keep the goodwill there. Like Urs Heckmann does, he's a champ

[ObsoleteAcc99]

[DELETED]

[egbert]

I don't think combining would be so great from the consumer's point of view at this stage of the game. The greatest advances happen when you have more competition and more creative people given the chance to put their work out there. As industries mature, and the vast majority of players are eliminated, you end up with absolute stasis - need I mention Internet Explorer?

The bigger the corporate entity, the more conservative they become. An extreme case would be Hollywood studios - they won't let anyone do anything they haven't already done successfully. Those guys would rather release Blair Witch 4 than let someone put a genuinely new idea out there.

Audio DSP software is a field where a single coder can enter the market and use internet distribution to make a living from their spare room. My impression is that this is more or less what Torben and Aleksey have done so why fix something that ain't broke: Let a hundred coders bloom!

Perhaps in time some of the current small players will form together into teams to tackle bigger projects like fully featured hosts. When that happens the sort of direct communication they have with customers will likely become a thing of the past and product updates will become annual events

[papawillow]

, war does fuel the wheels of progress.

[lavoll]

kjaerengo mlephant

[ObsoleteAcc99]

[DELETED]

[Cabinfever]

Fair and friendly competition is good for everyone. It's the fundamental strength of captialism.

[bmanic]

Here is a proposal for such test using r8brain PRO (demo version will be enough), this is applicable to all brickwall limiters:

1. Take an already loud sound clip.
2. Process it first with MPL (+6 dB), and then with Elephant (+6 dB of gain as well). Elephant should be in 4x oversampling mode. Save both results in different files.
3. Now launch r8brain PRO and select both files for conversion. Specify target sample rate to 4x of the original sample rate (176400 in case of 44.1kHz original sample rate). Then check the 'prevent clipping' checkbox. LinPhase mode should be used.
4. Process the files and open both in a sound editor. You will see that Elephant is dead brickwall while MPL has some intersample peaks appeared (r8brain PRO's clipping prevention dropped overall gain by 0.4 dB in my test).

From this test I conclude that MPL's peak detection - though sophisticated, does not work as expected - it does not suppress intersample overshoots.

BTW, iZotope Ozone passes this test as well. So, maybe the latest Ozone ( 3.08 ) already uses full oversampling - or maybe I just do not know what they know.

Although I think technical jargon is good and all it still doesn't sell a product for me personally. I always listen to the sound of the plugin, no matter what it has as technology (at least I try!! ). I at least am not in the least bit worried about samples that go past 0dBFS. They are a moot point in modern pop/rock/distorted music. Smooth jazz and classical don't need limiters anyways so it's a moot point there as well I think.

What I'm trying to say: The perfect limiter on paper might not be the best when it comes to sound. This is at least very true in the analogue domain where a lot of funky old hissy tube gear that have sucky specs kick serious ass.

Cheers!
bManic

[bmanic]

Oh btw, as an interesting experiment I suggest you guys rip one of the current top 20 US singles from the CD with some quality CD ripper software and then open up the file in WaveLab or similar and check for overshoots. As incredible as it sounds the most I've seen is +3dBFS. My little CD player (which is rather old) and my boombox plays these still without any noticeable "crack!" or "pop!" sounds.

Cheers!
bManic

[christianmusicmaker]

Has anyone realised just from reading this thread that we have two absolutely brilliant minds here at work? Imagin what they could accomplish if they join forces. That is my take on it.

My thoughts exactly.

I suppose thier discussion here is a type of collaboration as they discuss the best way to do things technically.

Also well done to Torben and Aleksey for keeping things freindly and professional. The information shared between them on this thread is great to read. As many have said both of their product sets are superb. I own plugs from both of them.

I think a joint effort, Limiter, 5 band compressor , Eq or some sort of "KV" series of plugs would be possibly be more lucrative for both developers. If the following of Aleksey and Torben on KVR is anything to go by a mini series of plugs developed by both of them would be a guaranteed overnight success.

Of course it's not necessary as their plugs are just so good already but...what if?

[egbert]

A lot of these things do not work well with a committee system - even if it is just a committee of two. All those predictable ends to formulaic Hollywood movies? - for those we can thank the committee system.

So we're building a limiter - will we oversample all the way through or just the peak lookahead (just to borrow a binary choice from this discussion)? These are two ways to go and, probably, you would have a discussion/argument between the partners and your team would go down one of these roads to produce a single product that has to make enough money for both of you. One person may have lost the argument but still feel his way was best - and they might be right too! Better, I think, for one guy to go with his gut feeling, his energy and belief in himself and his ideas and do the best he can with his preferred approach and the other guy to try his way and let the results speak for themselves.

Bob Moog decided what his filters and oscillators should look and sound like - ARP synths got different ones. Is anyone sorry they didn't merge earlier and form the Giant Synthesiser Company and just make one model?

Remember also that people have lives - they are not just coding machines. They have a language, friends, family etc - they might be happy to have found a way to do what they want where they are and teamwork between two people in different countries may necessitate somebody moving.

So vwot eggzactly iss going on here? I think it is time we put KVR onto ze couch.

As some may know, there was this psychoanalyst geezer called W.R. Bion. During WW II, he was placed in charge of a military hospital in London where he observed a lot of strange things going on with the morale of the patients and workers etc. He got some groups together and made a study of their psychology.

His observation was that many groups behaved as if the members had all made what he termed a basic assumption. He talked, for example, about basic assumption dependency groups where the members acted as if they had collectively decided that all the resources in the group resided in the leader - eg The Reverend Jim Jones and the Koolaid Krew.

Then there was the fight-flight group where the group acts as if it's purpose was to defend itself from threat or attack. Stuff like this seems to be at play on music forums when platform wars and host wars rear their ugly heads.

Another kind of group he called a pairing group - in these groups two people became prominent and the others became quiet - they sat and watched and it seemed to Bion that they shared the fantasy that the two active players would form a pair - and perhaps give birth to a saviour

I suspect something like this is going on here - people want Aleksey and Torben to unite and give birth to the one true plugin which will ... what exactly ... save our mixes?


"Insanity in individuals is something rare—but in groups, parties, nations and epochs, it is the rule." Friedrich Nietzsche

[Torben]

Oh btw, as an interesting experiment I suggest you guys rip one of the current top 20 US singles from the CD with some quality CD ripper software and then open up the file in WaveLab or similar and check for overshoots. As incredible as it sounds the most I've seen is +3dBFS. My little CD player (which is rather old) and my boombox plays these still without any noticeable "crack!" or "pop!" sounds.

Cheers!
bManic

The only thing that can give +3dB FS is a pure square wave (or DC) - nothing else. So it only depends on your meter how long time there should have been a square wave before it shows +3dB FS. This is the absolute maximum for an AES17 calibrated meter. It doesn't surprise me though but I really hate to hear music that has been treated that way. I sometimes thinks its hilarious that HIFI nerds spends thousands of dollars on loudspeakers and amplifiers that makes almost no distortion while the labels see to that the music is distorted 50%

Torben

[defjamm]

^^it's the same with the whole recording, mixing and mastering:
people use 5000$-mics into 3000$-preamps into 2000$-converters, record in expensive rooms, mix with expensive gear, master in expensive rooms with expensive gear by an expensive engineer just to distort it at the end.

some distort it even in mixing, so it's not always the mastering engineer(+ he doesn't like it but his clients want it to be distorted)

[Aleksey Vaneev]

Torben, but your MPL *does not* suppress intersample overs. That's what I'm saying. If Ozone has implemented it right (CPU meter shows they are not doing full blown oversampled processing), then it means MPL does not implement it correctly.

I will be adopting a similar approach in the future - but I'm sure I will implement it right (I've already analyzed how to implement it right, with the same results as what Elephant shows at 4x oversampling).

Dandruff, of course I'm talking about Ozone's limiter with that 'prevent intersample-peaks' switch enabled. I'm just trying to show that Ozone works correctly, while MPL does not. While it seems that both are using the same approach.

[Torben]

Torben, but your MPL *does not* suppress intersample overs.

According to our tests it does! Maybe your up and downsampler are not excact enough for this test.

Torben