Jay, this is one reason why I say use the color knob and set the mix to 100%.

DG

Here is a direct quote from Peter Bakker of AudioEase from their forum:


Re: Direct Vs. Dry Sound...
Reply #2 - 26. Mar 2007 at 09:19 Quote The 'flat direct' sound is the dry signal, but delayed and gained to take the place of the direct sound in the IR. This was done so you can use uncolored sound for the speaker placing options.

Both colored and flat direct clash with untreated dry sound: they are so close together that they will cause comb filter effects.


Re: Direct Vs. Dry Sound...
Reply #9 - 27. Mar 2007 at 09:27
The issue is not that different from the ones described above: the direct sound in the IR can cause phasing problems with the dry sound, and that will be worse when it's color is turned to flat.

The direct sound in the IR is later than the dry sound, giving a few samples delay, causing filtering and phasing effects.

@JBacal said:

Here is a direct quote from Peter Bakker of AudioEase from their forum:


Re: Direct Vs. Dry Sound...
Reply #2 - 26. Mar 2007 at 09:19 Quote The 'flat direct' sound is the dry signal, but delayed and gained to take the place of the direct sound in the IR. This was done so you can use uncolored sound for the speaker placing options.

Both colored and flat direct clash with untreated dry sound: they are so close together that they will cause comb filter effects.

Jay, this is exactly my point. If you think about it, in the real world adding close mikes doesn't cause phasing, so why should the dry sound when it is added to a mix?

DG

WHAT???!!!!


Why is there a wet-dry knob if it should not be used?

(trying to remain calm... trying to remain calm... trying to remain calm...)

The issue has to do with the way Altiverb creates their IR's. There is a direct signal component to their IR's. Some companies remove the direct signal from their IR's. A synthetic hardware unit like the Lexicon has no direct signal (I might be making up this fact and could be completely wrong about this!!!). That is why adding a dry signal to a Lexicon wet signal does not create any phasing. But the Altiverb IR's have a direct signal plus early reflections plus reverb tail. This is why adding a dry signal MIGHT cause some phasing. If the direct signal and the dry are too similar you may get an unwanted phasing effect. Therefore, if you want to add dry (and I like dry in my mixes) then it might be necessary to set the color knob to IR and turn down (or even mute) the direct signal. Of course, if you don't hear any phasing (or if the slight phasing adds a desirable warmth) then you can just leave the direct signal alone.

Best,
Jay

@William said:

WHAT???!!!!


Why is there a wet-dry knob if it should not be used?

(trying to remain calm... trying to remain calm... trying to remain calm...)

You can use it, just don't (probably, possibly, maybe) add dry sound as well when it's not set to 100% wet. I think that it is of most use when not using stage positions.

DG

Hi guys, William asked me to chime in. A few tips I figured out along the way if this will help:

1. Don't widen past "default" of a certain IR: Correct - don't move the altiverb speakers further apart. There is normally plenty of width in the VSL files they don't need widening -do this over the mix via placement of instruments and groups in the field, not by widening them individually, narrow only if you wish.

2. Don't narrow to mono - but narrow a fair amount if you wish.

3. The speakers are not really panning at all, they are placement, which is different. It is hard to achieve a studio style panning effect with these speakers, and I figured out, they are not really for that. My suggestion is to use Waves S1 or another stereo panning utility before altiverb, do you narrowing of the field and positioning there if / how you want to, and then add to that effect by placing the altiverb speakers as well. In other words, do a bit of panning with S1 then finish off the job with altiverb - stereo speakers in altiverb should remain at default for this scenario, no need to narrow and narrow again or narrow then widen! all you are doing in altiberb is adding to the panning effect. [edit I forgot to mention the reason I think S1 is just fantastic is the assymtry parameter - it's awesome I don't know of another placement plug in that has this feature it is not only left / right and width the assymetry gives a whole new dimention of placing and shaping the sound and is absolutely invaluable, I can't rate this highly enough. If you can I suggest you demo it because it's really good. Instead of just placing the sound on the stage where it is like the instrument is playing 90 degrees to the stage you can turn it, and that kind of changes the sound source without filtering the sound in anyway, and you can achieve both MUCH more realistic panning techniques and also must more, where you wish, precise technqiues. Finally another small pointer, never "monoise" a stereo source with either S1, and never use the built in panning of your host. Instead, just take the stereo file, and mute one channel - you now have a mono recording.]

4. Very important. The direct signal should only be ON at ONE stage in the process NEVER more. This is where the phasing effects come in. In fact, you can leave this off altogether throughout, and just use the mix buttong but this is less effective for the placement effect.

5. Like they say, think of altiverb as a real room, what you are doing is placing speakers in a room, not the instrument you are playing through them - the result IS different - so they are there to emulate and to assist but they are not the real thing - that's why you need to use S1 or your choice of panner to refine the signal before going into altiverb.

6. William: The wet/dry knob is for use without "direct" sound. Direct sound *is* your dry signal *through* the IR being used. - adjust short reflections and tail to adjust the mix. When turning off direct signal, use the wet/dry as your wet dry [[:)]]

7. If you are using the altiverb "preset" style placement technique, you should turn down the short reflections on the buses - the placement altiverbs - don't leave them at -0db - these are adding short reflections. OR, you should turn OFF short reflections at the master - that might work, and only mix in the tail, but just be careful not to double up on SF in the chain.

8. The colour knob is important, I leave it set to IR most of the time. The IR used is also important.

9. The "master tail" should only have signals coming into it from the placement reverbs - no other dry signals - and those signals from the placement altiverbs should naturally be 100% wet from those altiverbs, so when you mix with direct signal off, and use just the wet/dry to mix your tail in you shouldn't get phasing effects since the dry signal we are talking about is actually the dry "non" tail version of the processed signal not the original dry sound from the instruments if that makes sense.

I hope that helps a bit [[:)]]

For my 2c worth, I think of all convolution reverbs available AV is the best one right now. It can be frustrating but once you work it out, and realise that what is frustrating is simply the many possible options, then you just have to tailor your options down and focus them into something that works.

Best,
Miklos.

A futher thing based on what William said in an earlier post: When mixing at 100% with the dry signal, treat the short reflections and tail as your wet/dry mix. If you turn them off, you are getting only direct signal - the dry sound of the room (but not a dry signal). Then just mix them back in to the degree you want - this might be why it sounded terrible before.

Miklos.

O.K., thanks Miklos I will try some of those options...

Though to add to my previous post, I will need to take the Krell Brain Boost that Walter Pidgeon in Forbidden Planet survived, but Warren Stevens did NOT, in order to do all that Miklos says here. It is far too complex for human brains. Of course, Miklos is actually an alien, one of the few Krell who survived the destruction of the Id force that was let loose on the planet. Dr. Morbius actually did not know that a few Krell survived, and emigrated to Earth. Miklos is one of them. I'll bet you people did not know that? Well, now you do. And he is benefitting us all with his advanced knowledge. Though many humans are incapable of enacting such knowledge.

Ha ha, I just read this - v funny. Thanks William, although I would say that a truly brainy person would have been able to explain the principles of AV much more simply than I have.

I thought about how to simplify what I said eaerlier, and really it comes down to this:

There are 2 modes in altiverb.

1. You USE the direct sound going through the IR, then you turn the wet / dry knob to 100wet always.

2. You MUTE the direct sound going through the IR, then you can use the wet / dry know to control the mix of the IR.

It's really basically as simple as that.

When using method 1 - you control the mix between the direct "dry sound" of the IR (which is different to a completely dry signal) by using the volume of the short reflections and tail and direct sound to mix the sound as you like.

When using method 2 you can also use the principles behind method 1 - mixing up and down the short reflections and tail but you should always leave the direct sound OFF.

The problem is that the direct sound going through the IR which is basically a dry sound of the ROOM (as in specifically different from a dry sound that has not been touched by AV) is so close to the actual dry sound in most cases, that you will get phasing and other ugly effects, this is why it is to be avoided - you need to set the wet / dry knob to 100% to make sure that 100% of you dry sound is coming from altiverbs direct sound (which is not dry, but processed through the room).

Make sense?

Miklos

I'm curious how you guys are actually implementing these options - I use Digital Performer (just always have) as my sequencer, driving 4 Gigastudios, each using Gigapulse reverbs. If I was going to use AltiVerb, wouldn't it need to be a plug-in for my Pro Tools rig, and if so, I guess I'd have to implement it on each section recorded separately. Which is fine, except then I can't hear it properly during composing, etc.

In fact, that brings up another issue, which is that I like to record my sections into Pro Tools via optical digital, which requires me to manually repatch a cable, and do 4-5 passes per cue. The sound is vastly superior to just taking a stereo sum off of my mixing board into Pro Tools, but it is also vastly more time-consuming. What are you guys doing?

_Mike

the proper way is to have multiple inputs into your rig - you can still record simultaneously - you will also need to run 2-4 instances of altiverb to get proper distance. And having it on while you write is pretty much a necessity. You really don't need to do stereo spacing just depth spacing. So 4 inputs is actually fine - but more is nice as well. I don't think you'd need to do more than 12 stereo inputs in any case.

Hi guys! I've been reading this thread with interest as I've been using VSL samples for a year or so now and I just got Altiverb.

I'm running VSL Horizon samples through Kontakt 2 on Cubase 4.

My idea is to use the built-in stereo modeller in K2 to correctly narrow and pan the various instruments and then use Altiverb for depth.

So, I'm creating 4 groups for the separate sections (winds, strings, brass, percussion) and each of them has an Altiverb insert for ERs.

Each of these groups is then being routed through another group for Ambience which has an Altiverb insert for reverb tail.

The problem I'm getting is that the Altiverb ERs appear to be reducing my specifically placed stereo sound to mono and then spreading them to stereo again (or at least a similar effect). My placement appears to be lost and the ERs just spread the sound across the stereo field.

At the moment I'm simply reducing the wet/dry mix to compensate for this effect but obviously this is a workaround (and a poor one at that). Is there any way to avoid this?

Having just finished my first major project ever where there was no live element, I thought that I would chime in and give my version of workflow. All samples were VSL VI and the only reverb(s) used was Altiverb.

1) All instruments were panned, either by narrowing the VST outputs or by using a panning plug (StereoPan).

2) All MIDI programming converted to audio.

3) Pre-mix done using a studio reverb with 5 instances of Altiverb set at 100% wet and using Stage Placement for depth. Speakers moved as wide as I like...!

4) All files mixed down separately (again using MEAP) with reverb printed, and imported into a new session template.

5) My engineer then treated the audio files much as he would a live recording with various reverb sends, EQs etc.

Time wise this was a very inefficient way of working due to the time it took in the double audio conversions. However, as my engineer usually works in ProTools with live orchestral mixes it was the best way to give him what he is used to. Also all audio conversions were done by my assistant in Studio 2, so it didn't slow me down, which was just as well, as I was averaging around 20 working hours a day for six weeks!

DG

I'd like to hear the results

@magates said:

I'd like to hear the results

It's not really up to the standard of the demos in this board, but I'll try to find a small section that doesn't sound too bad. It's a real bummer having to do something quickly and not even have time to use the best articulations for the job. However, having been to the premiere of the show last night, the sound system was so bad that I could have used a SoundBlaster and GM sounds for all the difference it would have made. [:(]

DG

Thanks for the info DG.

It seems you are saying that by keeping the speakers at maximum width, using Altiverb for depth placement didn't effect the stereo panning/narrowing.

Is this correct? If so, did you have to change an option to do this because I'm sure all my panned instruments are moving to the middle after I add Altiverb.

That is not true in my experience, widening the speakers in altiverb does generally speaking negatively affect the stereo imaging it is good for some types of stereo material, not for VSL particularly, but that is a general rule not a fixed one. I find the VSL samples are generally recorded with an almost archivalist mic technique that provides the fullest amount of stereo image as possible in a stereo technique and they did this so that they would have the maximum amount of material to work with when it comes time to implement MIR - as opposed to recording things with another technique that would give a more controlled image for lack of a better word, but then being limited later on in terms of shaping the sound flexibly in the future. They did the right thing. This technique narrows well and at "center" provides great imaging, but does not widen well with stereo width tools and that is what Altiverbs speakers essentially are as well. I have found you get a better result from leaving them at stereo 0 and using Waves S1 to place the source more precisely, since you can skew the assymetry of the source not just the width and panning - believe me once you play around with it you realise how indispensible it is and you can't live without it - it's far superior than just using width and pan together, you can really place the sound's centre better instead of just hoping as you sometimes have to do, and many times you can avoid using the width paramter altogether, which I find is generally good.

Also a great tip from Angelo Clematide on the forum here - all the samples are 2 channel stereo. If you want a mono recording of any instrument or section in the library load a seperate two channel mix channel in your DAW (NOT a stereo linked single fader as is commonly used) and mute one channel - and you have a mono recording. Far cleaner and more true to life than monoising using a stereo tool which is pretty crude way of doing it. By the way personally I have never found it to be preferable from a purely aesthetic standpoint to use a mono recording of any VSL sample - I find they always sound better in stereo and I've never needed to use this to place a sound but, sometimes that sound may be desirable.

Miklos.

Miklos.

@bluejay said:

Thanks for the info DG.

It seems you are saying that by keeping the speakers at maximum width, using Altiverb for depth placement didn't effect the stereo panning/narrowing.

Is this correct? If so, did you have to change an option to do this because I'm sure all my panned instruments are moving to the middle after I add Altiverb.

You don't have to take my word for any of this. It is easy to test, without even using your ears (the best test of all).

1) Narrow the stereo width of your sample, either at the VST outputs (not possible in some programs) or by using a proper panning plugin. Notice I'm not talking about silly panpots.

2) Look at the output on your Master Fader. You should be able to see exactly how much sound is coming up each channel.

3) Now introduce an Altiverb with stage positioning. Leaving the speakers at default distance will drag your sound more towards the centre. Also, as in the real world, you never hear the sound only from the direction where it is played, so this will also have an effect on where the perceived direction is.

4) Now do the same with the speakers at maximum width. The effect of your original panning is more pronounced, although there will still be some "room displacement" that affects your panning.

With my method of using a very short, relatively dry reverb for the depth placement the room displacement effect is much less apparent than with a long reverb.

Finally, the best test is to use your ears. I hear no degradation of the sound by increasing the width of the Altiverb speakers, and to be fair, this is all fake anyway, so who is to say what is right and wrong....!

DG