Page 1 of 2

wet vs dry samples

PostPosted: Fri Feb 12, 2010 9:22 am
by sonar11
Some of the organs I've looked at come in both wet and dry flavours... If my audio card (e-mu 0404) has built-in reverb, should I buy the dry sample and use the card for getting the "cathedral" effect, or should I turn off all effects on my card, buy the wet flavour, and let hauptwerk do it's thing. (The organ is in a house, so there is no natural reverb).

Any thoughts?

Re: wet vs dry samples

PostPosted: Fri Feb 12, 2010 10:34 am
by mdyde
If you search the forum for 'wet' and 'dry' you'll find lots of topics discussing the differences and strengths/weaknesses of each approach. E.g. here are a couple of relevant discussions:

http://forum.hauptwerk.com/viewtopic.php?f=6&t=2905

http://forum.hauptwerk.com/viewtopic.php?t=2293

People generally prefer either one approach or the other, according to what they want to do with a sample set, so you'll probably find lots of different opinions.

Re: wet vs dry samples

PostPosted: Fri Feb 12, 2010 11:06 am
by sonar11
Thanks for your quick reply! I read through those posts, and I must say, some of you really know your stuff :)

My question though is slightly different... I'm after the wet sound since I have a small room with no reverb itself... but my question is, should I let the sound card handle the reverb effect (the e-mu 0404 comes with built-in reverb) and so purchase a dry organ, or let hauptwerk do the reverb and purchase a wet organ? What will give me the better effect, and will there be a performance difference between the two options? (I would assume that letting the sound card handle the reverb would free up the cpu to handle other parts of Hauptwerk)

Thanks.

Re: wet vs dry samples

PostPosted: Fri Feb 12, 2010 11:22 am
by mdyde
Using hardware reverb (e.g. that in an EMU card) with dry sample sets is usually significantly less demanding in terms of computer resources, since the release samples are much shorter, thus requiring less RAM to load them and also less polyphony (CPU power) to play them.

However (at least for classical organs) wet sample sets will typically give you the most realistic spatial impression of the original organ's room, i.e. the sound will be closest to the sound of the real organ in its original acoustic. To achieve a similar degree of realism in that respect with a modelled acoustic (dry + reverb) you would effectively need a separate convolution instance (or reverb unit) per pipe, and it will be a few years before computers or hardware are sufficiently powerful to manage that.

I.e. because the position of each real pipe is different, you would need slightly different reverb/acoustic for each pipe, and thus a separate reverb unit/instance for every pipe, to keep the same spatial impression that a wet recording can capture. By using just a single convolver/reverb you are effectively mixing the dry sounds of all of the pipes together and then applying one single overall reverb to the mixed result, which doesn't give the same thing as recording each pipe with its own (slightly different) natural reverb/acoustic.

However, there are other many other reasons why some people prefer dry samples and/or artificial reverb/convolution (ability to adjust the reverb to taste, suitability for use in reverberant or different types of listening environments, etc.), as discussed in those topics I mentioned above.

Re: wet vs dry samples

PostPosted: Fri Feb 12, 2010 12:27 pm
by sonar11
I guess that answers my question, thanks!

Re: wet vs dry samples

PostPosted: Fri Feb 12, 2010 12:48 pm
by micdev
Sonar,

Having investigated reverb/convolution since the beginning of my Hauptwerk experience, I agree with Martin and while convolution is a lot better than most hardware devices and can create a very realistic environment, a well recorded wet sample set will always be impressive. A wet sample set provides a great "out-of-the-box" experience on the first load; with convolution, you will need to tweak a lot and experiment...

I do use some wet sample sets with convolution (with the release tail truncated). Sometimes because I don't like the orginal acoustics or simply because it allows me to customise the reverb; with the same sample set, I can recreate the acoustic of a small church, then by changing the impulse response feel like being in a large cathedral. With experience, you will achieve great results, but you will need to "play a lot"... for me it is part of the fun, for other, it is way to much trouble.

There are now a lot more choices when comes the time to select a convolution engine and plug-ins. While it is not for everyone, it is a great thing to have the possibility to use dry set within this environment.

You can have a look at my tutorials about adding convolution with "Sonar", and my latest one (publish yesterday) using Bidule here

Best regards
François

Re: wet vs dry samples

PostPosted: Fri Feb 12, 2010 1:12 pm
by sonar11
I'll probably just keep things simple for now, and get the wet samples... Interesting though, I can only wonder what another 5 years in computing power will bring :)

Re: wet vs dry samples

PostPosted: Fri Feb 12, 2010 6:29 pm
by kaspencer
You have raised a very interesting question, Sonar.

I am also a guitarist, and like most guitarists, when I play a certain style of electric (and even acoustic/electro-acoustic) I like to add certain effects, including reverberation and delay.

But something really interesting happens when I compare my attitude to guitar+reverb to that of organ+reverb. It is simply this: I have yet to hear an "artificial" reverb, no matter what technology has been used, which sounds as effective as a real large space, when used with an organ.

I know that Francois is an expert on this and I have read his accounts of his setting up of reverberation methods and technology.

Although I have some understanding of the physics & acoustics, I am in this case referring simply to my perception of the effect in my ears and brain. So I always prefer the reverberation recorded with the samples to that obtained artificially.

Martin has, I think, given us the answer: namely that a large number of pipes occupying different positions in the acoustic space, produce a much richer interaction with the reverberation characteristics than a small instrument such as a guitar, where the sound emanates from a tightly defined locus.

So, I'll cut the waffle and say: "go for the wet samples" unless you are to play the organ in a space which itself has a very interesting acoustic.

All the best

Ken.

Re: wet vs dry samples

PostPosted: Fri Feb 12, 2010 7:56 pm
by Grant_Youngman
micdev wrote:
Having investigated reverb/convolution since the beginning of my Hauptwerk experience, I agree with Martin and while convolution is a lot better than most hardware devices and can create a very realistic environment, a well recorded wet sample set will always be impressive. A wet sample set provides a great "out-of-the-box" experience on the first load; with convolution, you will need to tweak a lot and experiment...


I have yet to hear a sample set with added reverb that sounds as convincingly good as natural ambience. They're getting better, theres no doubt. All the tweaking is getting somewhere. But still something is not quite on the mark.

IMO, theater sets with added reverb seem to sound better than classical dry (or truncated) sets with reverb added. Not sure why that is. Maybe there's something wrong with my ears. Given how long I've been using (and probably abusing) my ears, that wouldn't be surprising :-)

Or it could just be something equivalent to the placebo effect -- if I think it's added reverb, it sounds like added reverb :-)

Re: wet vs dry samples

PostPosted: Fri Feb 12, 2010 9:20 pm
by Mike 353
If you compare the St George's Casavant sample set to other "dry sets, you will notice just a slight hint of reverberation. It is just enough for me to be satisfied with it, as compared to sets that are absolutely dry.

While I have not tried any of the newer "wet" sets, I have some of the older ones, such as St. Annes, Marcussen, and First Baptist of Riverside, and I always come back to the St. Georges set, because I like the sound of the sample, and the realistic feel of the "ambience". I also have two sample sets from Silver Octopus which are absolutely dry, and, as much as I admire the sound of the Trio samples, the dryness just turns me off.

It may be because I am used to playing in smaller churches that do not have a lot of reverberation, or it may be that it just sounds more realistic in my fairly small (22' X 12' X 8') music room.

Michael Moore

Re: wet vs dry samples

PostPosted: Sat Feb 13, 2010 9:47 am
by telemanr
While you might think there is a "realistic" ambiance playing the St. George's dry, the organ is actually in, and voiced for, a large cathedral-like stone church with a quite reverberant acoustic. I've had the pleasure of playing the actual organ and I would say it is much more realistic with added reverb. Having said that, it will never sound exactly like it does in real life, but it will be closer. Mind you, if you like the sound as it is found then who's to argue.

Re: wet vs dry samples

PostPosted: Sat Feb 13, 2010 10:17 am
by Mike 353
While you might think there is a "realistic" ambiance playing the St. George's dry, the organ is actually in, and voiced for, a large cathedral-like stone church with a quite reverberant acoustic. I've had the pleasure of playing the actual organ and I would say it is much more realistic with added reverb. Having said that, it will never sound exactly like it does in real life, but it will be closer. Mind you, if you like the sound as it is found then who's to argue.


Your post didn't surprise me; I had thought as much. I guess what I'm getting at is that I don't expect, or even want, my home organ to sound as if it is in a cathedral. Just a little ambience is enough to satisfy me.The ambience of the St. Georges samples is about the same as I experience at the console of the pipe organ I play on Sunday mornings (only the ambience, of course; not the pipe sounds themselves!).

I have occasionally played in churches which had quite a bit of reverberation, and have been able to compare the sound at the console with that heard out in the auditorium. While appreciating the fact that the organ may sound more glorious farther out in the listening space, to me it does not seem appropriate to have that sound at home, but rather to hear the sound more "close-up", as you would hear it at the console. I have heard the Washingtom National Cathedral organ both well out into the nave and in the choir area, and, if that organ was ever sampled, I would prefer the sound as heard in the choir, whereas in recordings I expect to hear the sound as if I was listening out in the nave.

On the other hand, as I pointed out, the Silver Octopus samples beg for a little reverb to be added to them. I love the Trio organ samples, and would like to add some of them to the St. Georges samples for a custom organ, but it is a lot of effort to add that little bit of reverb necessary to make them work with the St. Georges samples. I did do it with the Celestes, and I am preparing to do it with the Mixture from the Willis set. but it is not fun doing this sort of thing.

Re: wet vs dry samples

PostPosted: Sat Feb 13, 2010 10:35 am
by telemanr
I understand. Sometimes though, I do like to be "in a cathedral" in the comfort of my own home. For one thing, it's usually warmer. :)

Re: wet vs dry samples

PostPosted: Sat Feb 13, 2010 1:26 pm
by sonar11
I've never had the pleasure of playing a massive organ in a cathedral (the one church in Holland that I went to wouldn't let me play their large organ (they had a smaller one too) because I didn't have any "papers"), but I've listened a few times in a concert setting... I'm sure the reverb is different for the organist playing the organ (on a mechanical organ he is a few feet away from the pipes), then it is for the audience, but there still has to be some reverb where he is, right? If you're playing a dry sample at home, wouldn't that still be far different then the experience an organist has, even if he / she is playing "in the loft"?

The sound doesn't instantly decay for an organist when he finishes the piece... the audience will hear the final chord swirling around them, but the organist would still hear that. While he is playing, he may hear more actual pipe noise vs reverb, whereas the audience will hear a lot more reverb in that ratio, but again, when he stops playing, there are still those 2 to 7 seconds of sound.... I'm sure he must hear most of that...

Re: wet vs dry samples

PostPosted: Sat Feb 13, 2010 2:55 pm
by telemanr
If you listen to some of the recordings on Youtube of Daniel Roth playing the St. Sulpice organ you definitely hear lots of reverb. And these were recorded from right beside him as he plays. You also hear the Barker levers which are quite noisy when everything is coupled down.
As far as the St. Georges goes, you still hear reverb at the console since it sits across from the pipes and your left ear is pointing out into the length of the church. The Swell box has two sides with shutters each of which can be independantly coupled to the swell shoe. One side faces into and down the length of the church while the other faces the console.