Relation in volume and voicing between ranks?

Hi all,

I may have asked a question similar to this in the past but now that I have more experience with Hauptwerk I'm now perhaps better experienced to ask it in a better way. When I sit down and play I sometimes wonder if certain ranks are sounding too loudly or perhaps not loud enough, too brightly, too dull and so on, which brings me to the next question: Is the organ sounding as it should or are certain ranks because of the way I either have them adjusted or have neglected to adjust them, causing the organ to sound considerably different than it should? I also wonder as the sample set is being recoded, if the mics are getting moved around to get the best sample of a certain rank, if this doesn't kind of skew things as well? Or is the standard practice to leave the mics in the same position throughout the recording process in order to better give a more accurate rendering of how the entire organ and it's individual ranks sound from one predetermined position?

I'm not looking to get into the recording business here but instead am trying to better understand the voicing process. I've watched videos of technicians voicing an organ but have no clue how they make their determinations. They start with a bunch of pipes mounted in a windchest and start hitting notes, listening, adjusting, tweaking and so on. How do they determine how loud or quiet one rank should be up against another? Is it just preference and subject to their choice, some kind of rule of thumb, exact science, only comes with experience or what? I have to believe over the hundreds of years of organs being around there must be some kind of general guide, rules or certain practices or guidelines that come into play. If anyone here can share either some knowledge or provide some info I can read up on the suject I would greatly appreciate it!

Thanks

Marc

[mdyde]

I also wonder as the sample set is being recoded, if the mics are getting moved around to get the best sample of a certain rank, if this doesn't kind of skew things as well? Or is the standard practice to leave the mics in the same position throughout the recording process in order to better give a more accurate rendering of how the entire organ and it's individual ranks sound from one predetermined position?

Hello Marc,

The standard practice for wet sample sets is to keep the microphones in fixed positions, since that ensures that all pipes are recorded with the same room acoustics and thus sound natural together. Also (with wet sample sets) the balance between the levels of the ranks and pipes is usually kept approximately as it was recorded, again to keep the recording true to the original instrument. (Some tweaks to individual pipe levels/voicing might be necessary if the original organ hasn't recently been maintained. Also it might occasionally be necessary to move the microphones for a few pipes in exceptional circumstances, such as for very quiet Pedal pipes that are placed elsewhere in the building and for which sufficient signal-to-noise ratios couldn't be achieved by any other means.) The primary aim when creating a wet sample set is usually that, if heard through headphones (or a pair of near-field monitors, so as to minimise the influence of the listening room), the sampled organ should sound as closely as possible (including an accurate representation of the acoustic for every pipe, which imparts realism, clarity and three-dimensionality to the recording) to what you would have heard if you had listened to the real organ from exactly the same (fixed) position at which the stereo microphone pair was placed.

With dry sample sets there are no rules, since, in order to record the organ with minimal influence from the acoustics of its building, you would normally have to move the microphones frequently (to get them close to each rank/pipe). However, in that case it wouldn't matter since you wouldn't then be aiming to preserve the acoustics anyway (being the reason not to move microphones). Hence with dry sample sets the levels and voicing are likely to need lots of adjustments later, in order to approximate the perceived original sound of the organ. That might be done by the sample set producer, or left up to the user. Even if done approximately by the sample set producer, the end user would probably still need to voice them for his or her speakers and listening room.

Hello Martin,

Thanks for the explanation, that helps. After I posted this I realized I probably should have made a distinction in my question between dry and wet sets, but your explanation none the less does confirm some things I wasn't sure about, in particular on the mic positioning part when recording to produce a wet set. So thanks for that and assures me if the set in question is a wet set, that the volume between ranks in the final product is likely to be very accurate as if we are there and listening in person.

One of the main reasons I ask this is with a few wet sets I have, I seem to either struggle to get a nice balance between certain ranks, or have a tendency to doubt as they are added to the others that their volume is where it should be. Is it too high, too low and so on and is why I ask if there is some kind of guideline to go by? Parts of a sample set that can really vex me from time to time are the mixtures, 2' and some 4' in particular. When I work my way up the keyboard, they seem to jump out more and I wonder if it is supposed to be this way. When I play a fairly full registration on some sets with these ranks, in some cases they seem to just scream and in others they get buried. Tinkering with either their amplitude or their brightness is touchy business and can make a noticeable difference, but again I ask if I'm kind of messing them up and not giving them their due justice.

Something I pondered, if I were to start out at a set volume for the entire organ, would it make any sense to use a db meter set in a fixed position in my space to measure each rank as a person voices a set, that is providing I knew where the approximate / correct db of a given rank should be up against the others, or would this be a complete exercise in frustration?

Thanks,

Marc

[mdyde]

Hello Marc,

If you're listening through speakers in a non-anechoic listening space and/or especially if you're using multi-channel audio, then your listening room will affect frequencies differently, as well as each speaker's output differently. Your speakers themselves inevitably will also colour the sound to some extent (and even amps, to a lesser extent). Hence what you hear from your playing position in your listening room will be significantly different from what's recorded in the samples (and from what would be heard through head-phones). That may make some ranks/pipes sound louder and/or brighter than others.

You could try comparing the result with a headphones mix.

Hello Martin,

Yes, sir, all good points. Due to the fact as you point out that speakers can either emphasize or de-emphasize things, what do you think of the db meter idea to get the rank levels adjusted or pipes within the ranks even sounding when voicing a live (speakers / multi-channel) set-up? I just did a quick search online and there's a bunch of them for real reasonable money, some for under $20. I do trust my ears pretty well but it would be interesting to see how well a db meter might do. Do you think using one while voicing would be a worthwhile tool?

Thanks,

Marc

I attempted to post the following just after Marc posted his original questions. In effect, some of my questions have been answered, at least to some extent. Here is what I had intended to post (my ISP had a tech issue).

Along the same line as Marc's questions above, I've wondered how much actual volume difference can a voicer of real pipes make? I have an idea of some of the mechanics of real pipes to change the characteristics as to how they sound, but volume? My impression was that a lot has to do with the air pressure. Also, maybe to a lesser extent, the physical position of the rank itself in the chamber.

For the most part, I use the volume sliders to "compensate" for room deficiencies such as a pipe sounding too loud with respect to other pipes in that rank because of room resonances.

I've only been to a couple real pipe installations in a home environment. In both cases, I've perceived that the pipes were very much too "in your face" because the pipes themselves were constructed for a much larger environment. So how does (real) pipe scale enter into the formula? BTW, the best installation at least to my ears was where the (real) pipes were installed in the home's basement and there were controlled "vents" from which you could hear the pipes (upstairs).

And of course, there is the variation of volume (and "equalization") of the ranks that are enclosed when swell shades are opened and closed..

Rgds,
Ed

Hi Ed,

My questions are indeed much along the same lines as yours. I've watched videos of the very beginnings of an organ being constructed, from the pipes being made in the shop, to the very end of the project, very interesting by the way but boy talk about having patience! They start out in the shop with a completed pipe they just made by hand no less, cutting and trimming and tweaking and bending, throw it on a wind chest, hit the key and listen to it and then kind of go from there and make a few adjustments, looks very seat of the pants. How in the world they know once that pipe lands in the actual wind chest in a much larger space surrounded by all kind of other pipes and what not if it will be too loud or not loud enough or what is beyond me? Then as I ask, how in the world do they know if the rank of pipes they just completed will even be in the ballpark amplitude-wise in comparison to the other ranks? Some very fine science in craftmanship here obviously, or they have certain formulas worked out to give them the desired outcome. In this day of modern technology do they use db meters to measure the amplitude of a pipe? It's all quite amazing to me.

I watched the video here when they voiced the newly refurbished Skinner / Quimby at the Cathedral of St. Paul, MN and they had a guy up in the middle of the pipes, two guys at the console would play a note and listen, then tell the guy by the pipes "louder" or "quieter" or it's too sharp or flat or whatever, he would be banging or tapping away at things, lifting this or bending that, they would then play the note again before finally saying "OK, that one sounds good and move on to the next one. This obviously all takes a very trained ear with experience to know what to do and what to expect for an outcome when various stops are drawn on top of each other. Martin covered very well my questions on the recording and playback part, the voicing is the other part of the equation I'm still trying to get a grip on.

Here's a short video of the Quimby gang in action in St. Paul, MN explaining and demonstrating some of the procedure, they do have a way of making them louder or quieter but he really doesn't go into much detail.
https://www.youtube.com/watch?v=D-pDu5n ... Yi2zOF9a9w

Thanks,

Marc

Here's a short video of the Quimby gang in action in St. Paul, MN explaining and demonstrating some of the procedure, they do have a way of making them louder or quieter but he really doesn't go into much detail.
https://www.youtube.com/watch?v=D-pDu5n ... Yi2zOF9a9w

A personal "trade" secret maybe?

You mentioned using a meter. Never heard that Father Willis or any of the other organ building greats of old had even conceived such a thing were possible. They, no doubt, had calibrated their ears over time. There are uses for meters, but I don't think voicing an organ is the proper place. Like the saying goes, "If we all sang the same note on perfect pitch, there would be no harmony".

Rgds,
Ed

[mdyde]

Due to the fact as you point out that speakers can either emphasize or de-emphasize things, what do you think of the db meter idea to get the rank levels adjusted or pipes within the ranks even sounding when voicing a live (speakers / multi-channel) set-up? I just did a quick search online and there's a bunch of them for real reasonable money, some for under $20. I do trust my ears pretty well but it would be interesting to see how well a db meter might do. Do you think using one while voicing would be a worthwhile tool?

Hello Marc,

I'll leave people here with more experience of real-world voicing than I have to comment on that. Except for when testing multi-channel audio, I almost always use headphones.

I have severe hearing issues, so I used a VU meter to help set things up. Now of course a VU meter isn't the best way to go about things, but it seemed to do what I wanted it to do. I purchased one will excellent accuracy. You do have to take your readings in the "sweet spot". I basically recorded using midi, then played it back and noted the various readings for each pitch, for each stop.
Glen

I did some research this afternoon and found volumes of information concerning how a pipe speaks, theory and so on, but when it comes to voicing it's either a highly guarded secret or more so a trade of one opinion vs. another. Within a rank, matching timbre and volume are the two key factors but I couldn't find a thing concerning volume level from one rank / family of similar sounding pipes to completely opposite ones other than if you're off the outcome can be very disappointing. Well no kidding, who would have guessed?

Top secret stuff here from the looks of things.

Marc

You need to adapt the sample set to your sound system, acoustics of the room, and finally, to your personal taste. At first I felt very sheepish about retouching an organ made by Schnitger or other great legend, but am very happy I did so. Once you start hearing the differences in pipe volume and tone within each rank you will really appreciate evening out the inconsistencies. However, some of the variability in tone from one note to the next adds charm and life to the sound, so it's good to leave in some "imperfection".

I've now spent a lot of time voicing my sample sets, first individual notes (especially pedal notes which can be really too loud or too soft depending on your sound equipment), then whole ranks, to arrive at a sound which I want to hear. For example, on the Zwolle organ I have boosted the entire rank of 8' Quintadena to make a useful solo stop that balances with an accompaniment on another manual. Ditto the 8' regal. On the French organ of St.Michel en Thiérache I have considerably softened both the tone and the volume of the reeds because they were very harsh. I made the treble louder, especially at the top because the pipes were biased towards the bottom. I made the larigot softer especially in the top. I revisited the volume of all the stops to even up the differences between notes. These changes made a great improvement and a much more satisfying experience. In the final analysis, you the player and listener are the one you are trying to please, so feel free to tinker. Your modifications can always be undone.

Disclosure : I have been a professional harpsichord maker and technician, and piano tuner so I have a lot of voicing experience.

Thanks robsig,

Yes, you're right. The nice thing is things can always be undone quite easily with HW vs. a real organ and it seems evening out things is the key. I just picked up the St. Max and is the first set in awhile I've taken an interest in voicing. The 16' and 8' on the pedal were waaaay too loud, some of the reeds, in particular the trumpets are also too loud as far as I'm concerned. I was always of the thinking that one or more trumpets have a softer or more quiet tone than some of the others, but so far all are very close to each other in volume, so now to decide on which manual I will chose to quiet one or two down. I've also noticed a bit more un-eveness within the same rank on some of the stops, the tierce and larigot needed work, some notes too quiet, some just screamed. I will slowly work my way through the entire set, but so far what I've done has really made a nice improvement.

Marc

Marc,

Just a thought, but are you voicing your St Max with the BBE kit in place?
Will a Sonic Maximizer affect absolute loudness levels? In other words, is the Maximizer making your trumpets sound the same?

I ask because I understood them to use a multi-band compression technology which I gather would "even up" the sound level in the various frequency bands.

Regards,

Marc,

Just a thought, but are you voicing your St Max with the BBE kit in place?
Will a Sonic Maximizer affect absolute loudness levels? In other words, is the Maximizer making your trumpets sound the same?

I ask because I understood them to use a multi-band compression technology which I gather would "even up" the sound level in the various frequency bands.

Regards,

Hi Richard,

No, that's not what the BBE's do at all, I would liken them more to an adjustable cross-over of sorts but the main effect is they really clean up the sound. Of course if a person goes crazy and has them way out of normal adjustment they will make the lows too boomy and highs exceedingly bright, but this range of adjustment is there in order to work with a wide variety of different responding speakers, so in other words the BBE's should have a positive effect on any sound system.

What I've run into with the St. Max is pretty typical from past experiences when first trying out new sets. With some, the entire set is really spot on or needs very little work, on others it seems the pedals, 16' and 8' flutes, or 16' bombard in particular can be a bit much at first and need some toning down. Another thing I've discovered with the St. Max is some of the pedal notes speak rather slowly, so one needs to hold the note long enough to find out what it really does before making any decisions on adjustments. Other than that I've encountered a minor amount of unevenness between notes within a rank and although all the trumpets are quite loud to me, they do have different timbre. Pretty much typical stuff so far but I can say it is by far one of the most enjoyable sets I've come across, a complete joy to play and I absolutely love the sound!

Marc