sample set integrity

[jordan312]

When you acquire a sample set for HW or GO or one of the various VPO systems, do you prefer it to be an exacting representation of the original pipe organ's specification or are you welcoming of "editorial enhancements" ?

[jbird604]

I've only dabbled in the VPO world, so my opinion isn't based on much experience. I'd opine that it would depend greatly on my intended usage of the sample set. For example, I'd want a sample set of the Notre Dame de Paris to be a genuine replica of the original, seeing as how it's such a significant organ, one of a kind, and lately had a close call that we all found quite disturbing. I'd want it to be a faithful replica because the purpose of buying it would be to "experience" the playing of that organ without actually traveling to Paris!

(That said, I certainly wouldn't mind if the publisher had tuned up any sour notes that happened to show up in his recordings. No need to preserve for eternal posterity the problems with the organ on a given date. I just would prefer that the stop list itself, the disposition, the layout would be authentic, and that the acoustics would be authentically "wet" so as to give the full effect of actually playing that organ in that space.)

OTOH, if I were seeking out a sample set to buy for use as an everyday home practice organ, I might prefer that it's disposition be "normalized" so that the stop list is more typical of the majority of organs we actually play on, rather than representing just one particular organ, which might have a rather quirky layout. After all, many organs in the real world are the product of someone's often biased or quirky concept, and might have some stops in the "wrong" division by accepted standards, or have other anomalies that are "interesting" to a student of organ design perhaps, but possibly frustrating and confounding for a person just wanting an organ to practice standard literature on.

Just my thoughts. As I said, not based on much experience, just opinions.

[Admin]

For the most part, it makes no difference to me. I'm more interested in how the organ sounds and its versatility in handling repertoire of its period, rather than its historical correctness. On the other hand, I'd prefer not to see enhancements for the sake of enhancement. Enhancements that overcome a serious defect or omission in the original design are welcome, while an enhancements that represent an uninformed idea of what would make the organ better are not.

Regardless, if the sample set is being sold as a representation of a particular organ, the enhancements should be well documented and capable of being disabled so that the organ can be played with its original specifications.

[voet]

Excellent post. I completely agree. I think the problem often arises because the uninformed often think they are informed.

[Nutball]

What would an example of an enhancement from an uninformed idea be if you can think of one?

[jordan312]

Now two more questions:
1. So who is qualified to make those editorial decisions? Who ARE the informed? Whose bias is the correct bias?
2. So if an organ is not sold as a representation of a particular organ, but rather as an expression of a particular form of organ, would that be of any interest? or is historicity the governing factor?

I know asking such questions is a power keg, but I ask questions, it's what I do.

[samibe]

I'll jump in. As far as editing and editorial decisions are concerned, I think most of the editing (done on VPO samples) is mainly removing obvious issues (a couple of individual pipes that are wildly out of tune, overblown, clogged, or have some other mechanical error affecting playing/recording and that will likely get fixed the next time a tech works on it). When playing a pipe organ, the occasional sour note or mechanical issue is forgivable and we do our best to work around them. There are just too many moving parts to keep all of them functioning perfectly year round. That is not the case for virtual organs. We expect our virtual instruments to work and sound good always.
Editing also involves removing noise from and prepping the samples to cleanly loop and transition from attack to sustain to decay.
None of what I have listed involves tonal finishing or voicing. That is an extensive area of study and practice with lots of opinions and discussions. And I don't know enough about it to offer a useful opinion.

Addressing your second question, I think there is plenty of interest. I think most digital organs specifically went that route. Most digital organs are a conglomeration of ranks sampled from various pipe organs that fit a specific form (classical, baroque, symphonic, theatre, etc.). Many newer digital organs come with several "suites" of samples that can easily be switched through. The suites may be typical forms for the era, but the stoplist will be adjusted to reasonably fit the labels on the organ's physical stops. I'm hoping to do something similar with my organ (once I get around to implementing midi on the stops of my organ). My current plan is to build a VPO that has 4 to 8 different voices per stop (typical theatre, symphonic, classical, and baroque stops along with a couple alternates and/or some fitting instrumental patches).

There is some value for a reasonable representation of particular organs though. I've thought about sampling each of the organs I play regularly (most of them digital) and building a VO of each of them so that I can work out registrations and tonal experiments at home. IIRC, there are a few contributors to jOrgan that specifically sampled and built dispositions of pipe organs to preserve a record of them before the organs were sold or destroyed.

[Admin]

1. So who is qualified to make those editorial decisions? Who ARE the informed? Whose bias is the correct bias?

First, it might be useful to define what I mean discussing "enchancements" in this context. I'm talking about specification augmentation, not cleaning up out of tune notes, misspeaking pipes, or other remediable defects that were not the intent of the builder.

As to the question of who is qualified to tinker with the specification of a historic organ, my answer is nobody and anybody.

Nobody is in the position, or is rarely so, to second guess the intent of the builder. The exception of course would be when the complete specification was not implemented or is otherwise missing. In that case the "editor" is acting as the agent of the builder in completing or restoring the specification. The catch here is in creating a rank of samples that is in keeping with tonal scheme and period of the existing organ.

Anybody, who is not fearful of being arrested by the organ police, can make any change or "enhancement" they see fit, but such changes run the risk of alienating a segment of their market, especially if those changes are not in keeping with the historical context of the organ, are not documented, or interfere with playing the organ as it existed in the physical world.

2. So if an organ is not sold as a representation of a particular organ, but rather as an expression of a particular form of organ, would that be of any interest? or is historicity the governing factor?

I don't have a problem with composite organs, provided that they are not marketed as being a historical representation of a particular instrument. I own and play several composite sets. One is the 80 rank "Father Willis" series from Silver Octopus. These samples are all from Willis organs, but not from the same one.

[jordan312]

Okay, then how about instead of a composite organ, what if someone created a blended organ? - bringing specific ranks from various builders and instruments (all complementary to the instrument as a whole.) I'm not talking about putting Wurlitzer ranks in a Holtkamp with a few Kimball ranks just for fun, but rather an appropriately blended arrangement of "like minded" voices. ((A sophisticated hot-rod, if you will.))

[tbeck]

Okay, then how about instead of a composite organ, what if someone created a blended organ? - bringing specific ranks from various builders and instruments (all complementary to the instrument as a whole.) I'm not talking about putting Wurlitzer ranks in a Holtkamp with a few Kimball ranks just for fun, but rather an appropriately blended arrangement of "like minded" voices. ((A sophisticated hot-rod, if you will.))

See this post for a .wav file demonstrating a blended sample set.
https://organforum.com/forums/forum/...ite-sample-set

[Admin]

Okay, then how about instead of a composite organ, what if someone created a blended organ?

I'm not sure I understand the difference between a "composite" and "blended" organ, but I have no philosophical problem with a disposition created from a different organs whether they be the same or different builders provided that the blending is seamless.

That said, I think such dispositions are challenging. Not only do the ranks have to be compatible in scale, but they also must be in the similar acoustic spaces when sampled. To me, that pretty much means either using bone dry samples or masking the different spatial profiles by adding a bunch of artificial reverb.

[tbeck]

I'm not sure I understand the difference between a "composite" and "blended" organ, but I have no philosophical problem with a disposition created from a different organs whether they be the same or different builders provided that the blending is seamless.

That said, I think such dispositions are challenging. Not only do the ranks have to be compatible in scale, but they also must be in the similar acoustic spaces when sampled. To me, that pretty much means either using bone dry samples or masking the different spatial profiles by adding a bunch of artificial reverb.

This is definitely a challenge. I have tried to make a composite sample set from samples from different organs with different acoustics. I tried to shorten the reverb tails to the same length, and then add reverb, but it really didn't sound convincing. I think if you shorten the reverb tails too much you also lose the natural release of the pipes.

[Admin]

The problem with shortening tails is that the recording is still in that reverberant space so you're still aware, on some level, of the room and your distance from the source. Reverberant samples with truncated tails always sound a bit hollow to me.

[j reimer]

I am in strong agreement with jbird's remarks in the second post of this thread, but I send this reply with some hesitancy, as I am not directly addressing the subject of jordan312's initiating post, which was to do with whether or not people prefer a sample set to be “an exacting representation of the original pipe organ's specification”.

What I look for in a sample set is, first and foremost, a satisfying representation of the original pipe organ's SOUND. The operative words there are “satisfying”, and “sound”. In the subject of Virtual Pipe Organs, both words are capable of a very wide range of meaning. For what satisfies one person, won't necessarily satisfy someone else. And of course there are various ways of listening to the sound of a pipe organ, varying from up close to quite distant, plus everything in between!

I like to vary my playing of VPO's and my listening to them, across that whole range. My first real experience of playing (i.e. learning to play) was on a small tracker pipe organ in my local church, and over the 60-plus years since then, most of my playing has been on such instruments. I regard playing such a pipe organ, in close proximity to the pipes, to be an almost unique musical experience. There is a fascination in taking note of the physical placement of the pipes, and the way each pipe almost has a personality of its own, while one is playing.

So I am putting in a plea for more sample sets to be produced which use DRY recordings. This is the only way to capture and reproduce the close-up sound of the pipes. I would far prefer to use a small VPO which gives me that sound, than a large one that doesn't, but I readily admit that not all users will have that preference. I might add that I have the choice with my VPO set-up, to add excellent reverberation which I can adjust to my liking, so I think that this way I have the best of both worlds.

There is another very important issue that I feel bound to raise. At the end of his post, jbird speaks of using a VPO at home for “practice”. I'm sure that John didn't mean to imply this, but in many pipe organ circles there is the attitude that an organ substitute is sometimes necessary in order to have something to practice our pieces on, but we musn't expect to enjoy it! I am of the opinion that if we make available sufficiently excellent VPO's, for use in the home especially, affordable VPO's which true musicians will enjoy enormously, then we may do something significant in maintaining interest in the real thing.

John Reimer

[jordan312]

Excellent comment and well said.
FYI- all the sample sets that I have created are recorded "dry" with the stereo microphone positioned a few feet from and above each chest. This would allow me to mix ranks from different instruments and create a more complex or varied specification from the original instrument, that is - when the character of the ranks would complement one another.