It is currently Thu Mar 28, 2024 9:21 am


OPEN LETTER TO SAMPLE SET PRODUCERS

Existing and forthcoming Hauptwerk instruments, recommendations, ...
  • Author
  • Message
Offline

adri

Member

  • Posts: 1545
  • Joined: Tue Aug 31, 2004 6:33 am
  • Location: Belgrade, Serbia

OPEN LETTER TO SAMPLE SET PRODUCERS

PostSat Sep 16, 2017 3:52 pm

An open short letter to sample set makers:

I believe the following to be quite obvious:

A singing and speaking human being is also really a wind instrument. Breath is very human with all its fluctuations. A monotone way of speaking and singing would sound very robotic and un-human.

Likewise, an organ is a wind instrument, and naturally it breathes. Its sounds have fluctuations of differing degrees.

Why are harmoniums/reed organs not sampled? Lack of interest? May be (and regretfully so, if the case), but more importantly, how in the world are sample set makers going to incorporate the effect of pumping such reed instruments, whether of the suction or pressure type? Emulate the difference between normal pumping and engaging the “Expression” on pressure wind harmoniums? You would need to have a physical MIDI pumping system device on the floor whereby a sample set would respond fully to its input. Sounds very complicated. No one has attempted this yet, as far as I know.

Modern organs with extremely steady wind systems, in my opinion, do not sound very human. They simply don’t seem to breathe. They sound robotic and stilted. Modern organs sound often cold because their wind is too steady and they have lost their humanity.

And this is exactly the problem with sample sets: their wind behavior needs to be also “sampled” or at least emulated more perfectly, before we can truly say that such sampled organs sound like the original.

Sample set makers, and I am not saying this to offend anyone, are “wasting” precious RAM on 4 or 6 microphone setups, instead on more fantastic and realistic wind modeling. I would rather have my RAM emulate the humane wind behavior in an organ than whether I can have a slider that can bring my ear unrealistically close to the pipes, or too far away from the organ.

Every space has an ideal listening position, and I think the 2 stereo mikes should be at that location, about 7 inches apart to mimic the human ear. I want an organ sample set to sound like the way I hear it at the ideal distance from the instrument.

I know many will disagree with me, but do we ever hear an organ as one person what two people would hear in two locations? No, we are individuals. Being at two locations at once is obviously going beyond reality. Should virtual reality go beyond reality? Have we gone too far?

Right now I cannot play because my organ is in storage until our new (old) apartment’s remodeling is finished. Mid November may be. So, I listen a lot, an awful lot to HW users on CBCH, with my excellent new headphones. I listen really a lot. And I listen also to the new sample sets, and am making the following main observation:

Sample sets in general do not sound as good as CD recordings of the same instrument. Why is that?

Simple: apart from the miking (and this is also still super important) it is because the CD recording catches every nuance of the breathing of that organ, while sample sets do not even come close. Not even close. Subtle nuances in wind behavior cannot be heard in sample sets, but in CD recordings, yes, we can hear them.

We have come a long way since HW1 and non-multiple samples/releases, etc. Great, but….

And this may very well be the final frontier in organ sampling: A vastly improved wind modeling, especially affecting fuller registrations, but even also the more modest registrations. On smaller organs even the effect of a bass pipe on a higher pipe. It’s not there in samples.

Piano sample set producers had to struggle how to emulate the sympathetic string behavior of a real piano. The old trick of holding down a key without playing it and making it com alive by hitting the overtone series, or even by just singing or speaking into the piano, is well known. This natural behavior is being more and more perfectly emulated in piano samples. Way to go!

Wind is everything in an organ. This cannot be stressed enough. When we cannot mimic the wind behavior of an instrument, but are concentrating too much on multiple miking (which requires crazy amounts of RAM), we are still making robot-like sample sets, and probably focusing our attention too much in one direction only. It has become a new rage. A new fad, and fashion.

Let’s put some more realistic virtual wind in these sample sets!
I would prefer that over 6 mikes. In this I am not alone.

This would be energy and time well spent!

Thank you.
Offline

erikds

Member

  • Posts: 220
  • Joined: Fri Dec 22, 2006 6:19 am
  • Location: Belgium, Zandhoven.

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSat Sep 16, 2017 5:11 pm

Let’s put some more realistic virtual wind in these sample sets!
I would prefer that over 6 mikes. In this I am not alone.


Indeed, you are not alone.
I fully support your recommendation.

All the best.

Erik.
Offline
User avatar

Csaba Huszty

Member

  • Posts: 366
  • Joined: Tue Nov 28, 2006 3:18 pm
  • Location: Budapest, Hungary

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 2:27 am

Hi Adri,

As a sample set producer, I can also confirm that you are not alone. We are experimenting on this problem for considerable time now, and also working together with organ builders to work out something useful. I cannot tell anything further at this point, just confirm that I also think you have a point.

However, to make the 'wind effect' work, more than wind modeling is needed. There seems to be some more phenomena that makes the virtual organ experience appealing and inspirational. Transient effects of and during the pipe speech, touch responsiveness on some organs, the vast dynamic range a larger organ can produce, the sound reproduction, head tracking, the quality of reverberation tails, the balance of wet and dry, signal processing complexity, and so on. Whether you are listening an organ through headphones or loudspeakers also affect how the recording should be made to match the reproduction enviroment (e.g., ambisonics, WFS or BoSC, object based systems, simple binaural, traditional microphone techniques, etc.). This topic has its own art and science and makes a great difference, as you well pointed out.
Yours truly,
Dr. Csaba Huszty
______________________
More at: inspiredacoustics.com
Mátyás II | PAB | MTH | Subscribe to the IA Newsletter for discounts and more
Offline

organsRgreat

Member

  • Posts: 561
  • Joined: Fri Nov 16, 2012 5:30 am
  • Location: England

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 3:59 am

It would be interesting to know how many organists using Hauptwerk at home have anything more than about 24 speakers. Even 24 runs into severe compromises once one uses more than a handful of stops.
The most realistic and musical organs I've heard on Contrebombarde were those recorded by the late Leo Christopherson. I believe he used around eighty speakers, which may well be a sensible minimum, rather than a maximum to aspire to.

Using this many speakers brings two problems:

(1) Cost: not just the cost of speakers, but of soundcards to drive them. And the effect on one's electricity bill!

(2) Space - always limited in a domestic set-up in comparison with even a small church or hall.

In my opinion too few speaker channels (and wrong types of speakers – another big topic) have more to do with lack of realism than wind modeling.
Offline

Mixtuur4st

Member

  • Posts: 97
  • Joined: Fri Mar 01, 2013 5:13 pm
  • Location: Breda, the Netherlands

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 6:39 am

organsRgreat wrote:
In my opinion too few speaker channels (and wrong types of speakers – another big topic) have more to do with lack of realism than wind modeling.


Absolutely true, in my opinion.
I am experimenting with a 100-plus channel audiosystem and the first results are overwhelming.
It is the mixing of the individual pipes (samples) in the air instead of in an electronic device that drastically improves the experienced reality of a sampled organ.

To Adri: listening by headphones, no matter how good they are, will never let you hear that mixing in the air.
Unless, of course, you listen to a CD that recorded a real organ.

Have fun,
Jack
Offline

münsterorganist

Member

  • Posts: 135
  • Joined: Sat Aug 27, 2016 2:24 am

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 7:55 am

Dear adri,
at last the right discussion. For me it is important to change the static sound of the samples.
Multi Channel at home is not so important for living sounds. You should need the gigantic Ram for improvement
the static sound.
I be looking forward to the notes von Sonus Paradisi, the sample set Kreszow is for me going
In a good way.

Best regards
Offline

adri

Member

  • Posts: 1545
  • Joined: Tue Aug 31, 2004 6:33 am
  • Location: Belgrade, Serbia

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 8:23 am

My post has nothing to do with sound reproduction via speakers, headphones, etc, although I believe that good sound systems will produce better virtual reality.

This is why I spoke about CD recordings of real organs vs. Hauptwerk's reproduction of the same instrument.
If I can hear absolutely the differences between them and hear the wind of an organ on a CD, and not from a HW recording, you know that there is work to be done.

I keep repeating: the best virtual reality must come from vastly improved wind behavior emulation. Not multiple mike setups.

Listen for example to the short samples here: https://www.muziekweb.nl/Link/CLX4995/Ibach-orgel-Grote-of-St-Gertrudiskerk-Bergen-op-Zoom and then go and listen to the same organ on ContreBombarde. In my humble opinion, the microphone placement for this CD recording was very good. Probably ideal in this large space.

Listen e.g. to the Leyding pieces on this site: https://www.muziekweb.nl/Link/CLX0407/Orgelmusik-aus-Ottobeuren; you hear the wind responses very well. It's fantastic!

And here to the last Toccata by Callearts: https://www.muziekweb.nl/Link/CLX1532/Bespeelt-het-B%C3%A4tz-orgel-van-de-Domkerk-te-Utrecht

During the Orgel Reform Movement (Orgel Bewegung), the focus was on how to make new organs sound as good as the old ones, which they didn't. Now, today, they can (e.g. listen to the Flentrop organ of the Catharinenkirche in Hamburg)(https://www.muziekweb.nl/Link/CLX4990/Regina-Renata-Die-Orgel-in-St-Katharinen-Hamburg.

Now, somewhat likewise, our focus in the virtual world has to be on making the virtual organs sound as good as the real ones; it's a challenge that can be tackled with the high tech equipment and computers we have these days.
Offline

ludu

Member

  • Posts: 997
  • Joined: Wed Nov 02, 2011 8:52 am
  • Location: Tournai (Belgium)

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 10:08 am

Jack wrote:I am experimenting with a 100-plus channel audiosystem and the first results are overwhelming.
It is the mixing of the individual pipes (samples) in the air instead of in an electronic device that drastically improves the experienced reality of a sampled organ.

In the case of 100 speakers: 50 stereo outputs or 100 mono?
Luc
Offline
User avatar

Jan Loosman

Member

  • Posts: 380
  • Joined: Sat Dec 20, 2008 4:33 pm
  • Location: The Hague, Netherlands

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 10:41 am

Hello Adri

The latest sample sets on CB often use the 6 channel recording technics to use with perspective sliders for variable organ positions. Recordings on CB are often recorded with some degree of mixing these perspectives.
Personally i always experienced some loss in detail and flattening of the sound when mixing the front perspectives, because mixing 2 perspectives is not the same as a genuine recording in between. You will always introduce phase poblems.
In an other thread about perspective sets Martin gave this Comment, quote!!

"Mixing the recorded samples from two separate microphone positions (e.g. 'wet' and 'dry') does actually give a significantly different result acoustically from a recording of a real microphone having been placed between them physically. I.e. mixing microphones is an acoustical approximation, and does inherently lead to a certain amount of perceived loss of clarity (due to contradicting spatial cues in the mixed result). However, it's a very common method used in classical recordings, and many people prefer it anyway. (Plus of course it allows people to adjust the sound, or, if you want a result that is strictly 'correct' from an acoustics point of view, choose to load either one or the other perspective, but not both together.)"

Could this also be a reason, or one of the reasons you are having a worse listening experience of organ music from CB compared to a cd recording. Mp3s on CB also have a compression of the sample rate!

When you listen to older stereo or 4 channel sets on CB do you experience the same sound problems or do they sound better??

Regards Jan
Offline

OrganoPleno

Member

  • Posts: 652
  • Joined: Thu Dec 03, 2009 4:08 pm

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 12:57 pm

adri wrote:I keep repeating: the best virtual reality must come from vastly improved wind behavior emulation.
Listen for example to the short samples here...


Those CD excerpts sound pretty good, at least on my cheap Desktop PC speakers. What we'd have to know is, WHAT did the SOUND ENGINEERS do to modify and distort the recorded sound, in order to produce this commercially-desirable result?

One thing we DO know is that commercial CD's are NOT interested in the accurate, faithful reproduction of some original sound. Rather, they aim to give a conventionally pleasing effect for consumer ears when played on ordinary grade home audio equipment.

To me, none of that is particularly relevant.

In many cases, I consider that recorded performances on Hauptwerk sound substantially BETTER than comparable recordings made directly on the live Organ. Particulary, improved clarity of sound enabling every detail of every note to be clearly heard and appreciated.

I've no objection to there being an improved Wind Model. This is of course an ongoing challenge in programming. The existing Wind Model in Hauptwerk was a TREMENDOUS step forward, for which I shall be forever grateful.
Offline

Mixtuur4st

Member

  • Posts: 97
  • Joined: Fri Mar 01, 2013 5:13 pm
  • Location: Breda, the Netherlands

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 1:07 pm

ludu wrote:
In the case of 100 speakers: 50 stereo outputs or 100 mono?


In case of 100 channels: 50 stereo ouputs.
(biggest problem is administrative/ logistic: which audio output goes to what amplifier, connected to what speaker and where is that placed).
Offline

adrianw

Member

  • Posts: 160
  • Joined: Sun Dec 18, 2011 5:10 am
  • Location: Lincolnshire, UK

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 2:08 pm

Sample sets in general do not sound as good as CD recordings of the same instrument. Why is that?


In your opinion.

On the contrary, I believe that HW sample sets often sound much "better" than CD recordings of the same instrument. Less "realistic", perhaps. But often better than real. A sort of Platonic hyper-reality.

This view seems to be shared by many experienced CD reviewers when they (unknowingly) encounter HW recordings and believe then to have been recorded live. For example, from reviews of Todd Fickley's Bach project, using SP's Zwolle:

“This is a valuable disc solely as a fantastic recording of some of Bach’s finest music, but as a harbinger of future releases in The Bach Project it is priceless... The sound of the instrument used for this recording is arresting from the first chord of the opening selection, and special mention must be made of the fidelity with which the organ is recorded. This is one of Northern Europe’s greatest organs, and recording it from a distance of 32 feet as supervised by engineer Jiri Zurek​ for this project both enables preservation of the clarity of the instrument’s registers and appreciation of the acoustical ambiance of the space and the five-second reverb...Its robust construction and unique timbre make it an ideal instrument for playing Bach’s organ music, and Mr. Fickley’s performances on this disc are enhanced by the grandeur of the instrument at his disposal... Hearing a great organ in a space worthy of its sound is an experience like no other, and this first installment in Todd Fickley’s and MSR Classics Bach Project recreates that experience on compact disc as vibrantly as any recording has ever managed to do.

Joey Newsome, Voix des arts [April 2015]


(My italics).

I do not believe that the difference between HW and CD is wind modelling. (Indeed, I routinely turn it off for modern organs. Except when recreating a baroque model, wind instability is rarely desired by organ builders.
Many modern organs actually have rock-steady wind, while HW organ models often seem keen to exaggerate the effect.)

Rather, CD recordings, which one listens to for no more than an hour or two, can be set up to capture the ambience and generous acoustics of big spaces in an indulgent way that would simply make an HW sample set unplayable.

The Bergen op Zoom recording link you give is a fine example: it sounds very splendid indeed but I would never want to play a sample set that sounded like this. (Indeed, SP has received flak just for releasing the distant version of this sample set.)

At its best, HW mixed mike sampling "enables preservation of the clarity of the instrument’s registers and appreciation of the acoustical ambiance of the space" (as the reviewer above expressed it) in a way impossible in reality.

- Adrian.
Offline

adrianw

Member

  • Posts: 160
  • Joined: Sun Dec 18, 2011 5:10 am
  • Location: Lincolnshire, UK

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 4:29 pm

Perhaps it is worth quoting another CD critic, apparently blissfully unaware of Hauptwerk but enthusing over an HW recording. This time a review of "Litany", Erik Simmons playing the music of Carson Cooman on the SP Marcussen organ of Laurenskerk, Rotterdam.

Classical Music Sentinel
“Great music, performed on a magnificent pipe organ, by an organist at the top of his game, all captured in vividly realistic sound. In other words, an outstanding recording. I never give out stars or number ratings but if I did, this CD would get 6 out of 5 stars, or a rating of 11 out of 10. Cooman’s music is highly original and, on this recording anyway, all the pieces are as varied from one another as they are captivating. A state of the art recording so realistic you would swear you are sitting right there at the organ console. I believe Divine Art have released a benchmark recording. Recommended for everyone and definitely not to be missed by pipe organ fans. ” —Jean-Yves Duperron


My italics again.

- Adrian.
Last edited by adrianw on Sun Sep 17, 2017 4:45 pm, edited 1 time in total.
Offline

ludu

Member

  • Posts: 997
  • Joined: Wed Nov 02, 2011 8:52 am
  • Location: Tournai (Belgium)

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostSun Sep 17, 2017 4:35 pm

I agree totally with Adrian about the wind modelling. I turn it off, not only for modern organs but for romantic organs too. Let’s take the example of Cavaillé-Coll. One of his greatest invention was the perfect domination of the wind. Over more, his art has inspired composers as Widor, Vierne and Dupré, developing the absolute legato, a technique that only the organ is able to restitute. The human voice and other wind instruments need to cut melodies regularly in order to breathe. For me the comparison with the organ is inappropriate, except for baroque music.
Luc
Offline

adri

Member

  • Posts: 1545
  • Joined: Tue Aug 31, 2004 6:33 am
  • Location: Belgrade, Serbia

Re: OPEN LETTER TO SAMPLE SET PRODUCERS

PostMon Sep 18, 2017 1:29 am

When I say that organs on CDs sound better than in HW, I was referring only to the behavior of the wind, nothing else.

There are many important non-wind related points brought up in this discussion, but I only want to talk about wind. The other topics are actually truly worthy of their own separate discussion threads.

It is true that Cavaille-Coll and other romantic organ builders tried to produce a steady wind, and yet, even in those organs there is wind behavior. And earlier Cavaille-Coll organs didn't yet have an electrical wind system.

But in baroque organs, who had smaller wind supplies, the wind behavior can range from actually quite unstable to more stable. Even JS Bach didn't like jittery wind supplies and would tests organs about this. But should one man's preference be taken as the way it should be done?

But there is something very subtle and extremely complex about wind behavior. Multiple sampling is still random and doesn't quite imitate the real behavior under naturally occurring circumstances of a live baroque instrument. I'm talking not only about obvious wind behavior when playing tutti chords whereby the bellows really have to keep up with the player, but also about very subtle differences in smaller registrations.

For example, let's say I play a Holpijp/Gedackt 8' only, and let's say I'm playing a chord or even just one note with the right hand, and then in the left hand I'm playing up and down passages. Most certainly in baroque and earlier organs with wedge shaped bellows, I am going to hear the differences in the right hand notes; they will be ever so slightly affected. Now that's natural wind behavior. This we don't have yet in HW. Just like the very subtle and infinitely variable sympathetic string vibrations in a live piano, which has been a real challenge for piano samplers to emulate, we have a parallel like challenge to introduce the subtle effects of wind under all kinds of playing circumstances/registration loads.

Also, there is a great difference between a hand/foot pumped organ, and one driven by electrically generated wind. When Taylor & Boody built the organ for Holy Cross in Worchester, Massachusetts, they installed motors on the wind system that would pump the bellows. Needless to say, this produced a very different effect on the music than just using the electric wind motor. The Hinsch organ in Zandeweer has a computer driven electrical system that actually pups the bellows; I think there is no override on it.

I know of organs whereby, during a concert, the bellows are pumped by human power to get an authentic experience. Listen here for example: https://www.youtube.com/watch?v=tHoXn_dQQGY

and here: https://www.youtube.com/watch?v=K0Fom3BEZkI&list=PL914B9899CA346D53
and: https://www.youtube.com/watch?v=Q-xtMd20b14&list=PL8Svq8PmThX48Snu4fnWaBZDcLlYuYooP

and: https://www.youtube.com/watch?v=3gKGJ_A5paw Note here that the man pumping has to increase his speed when the organ gets louder.

new organ, reconstruction of 15th century organ of Utrecht (now Middelburg): https://www.youtube.com/watch?v=EKF34fl9LeQ

And a magazine bellows system: https://www.youtube.com/watch?v=b03Pes5JAm8

Before the age of electricity, organs were pumped by humans, whether of the old wedge shaped type or the magazine bellows type. We seem to forget this.

For example, take the sample set of Kreszow with its 32' and plethora of 16' stops in the pedal; could we in real life in the 18th century draw all these pedal stops at once? And all the stops to create a tutti in the manuals? No, the bellows would not be able to handle that at all.

Are we ever going to have the virtual equivalent of human powered winding? An intelligent system that would respond differently according to how many stops were drawn? And increase the virtual peed of pumping when more stops are drawn? And would produce audible windsag when the organist draws too many stops?

We must not ever forget what change of effect an electrical system had on the sound of organs that were previously driven by human power.
Next

Return to Hauptwerk instruments

Who is online

Users browsing this forum: No registered users and 2 guests