Good idea for end users making new sample sets?

I guess we are listening to symphony concerts without noticing the lack of clarity and dynamics. The brain is marvellously capable of filling in what isn’t actually there and ignoring what is but shouldn’t be. For those without a large room and massive funds, what we have for Hauptwerk will have to do. I’d like more speakers but there’s little room and less money. There is also a very real consideration and that is the law of diminishing returns. We hear logarithmically and I believe the same idea would apply to the difference between let’s say 8 and 16 speakers. A fellow who used to be on this forum had a huge number (60?) of speakers. He spent much time fiddling with the vast number of ways to deploy them and was always on the hunt for that little bit better result. That’s assuming your aural memory was able to let you hear the improvement (minuscule at best) from yesterday’s result.
It’s a shame to always be thinking it could be better if only. That’s bad enough with regard to one’s playing technique let alone having to add an added disappointment.

There is definitely diminishing returns, no doubt about it; adding a second stereo pair is far more important than going from the 7'th to the 8'th pair.

My point though is that we all have limited funds, and choose to spend that money where we feel best. I think the entire premise of this thread (that end users shouldn't be making sample sets) is quite flawed, and was just pointing out that there are multiple ways to get better sound out of our organs.

Suggesting that one can't be happy with the result of merging various samples is just as flawed as suggesting you can't be happy with just 2 or 4 speakers. My point is... why focus on just that particular aspect (user defined samples)? The OP might find that user sample sets are poor quality, and I might find that his audio system needs some work, but yet we don't go around saying "users shouldn't be building their own audio system by grabbing random speakers/parts of a shelf and assembling it because "I" might find flaws in the sound".

A few years ago I spent quite some effort on designing CODM sample sets. Adri was one of my testers. At some point, he disapproved one of my designs. I did not publish it. No regrets: if a design is severely flawed, don't bother anyone with it.

Adri implicates that successful designs are possible, mentioning the Rouen model as the only example. If it is possible to successfully emulate a good organ, it should also be possible to design a good organ model from scratch...

So what design principles should aspirating builders of new virtual organs follow? What are the secrets of good organ builders and how can they be translated to our virtual environment?

To Josq:

Ultimately the question is: do we really need to emulate more sample sets by combining existing ones, when new ones of all kinds/types are coming out all the time?

The fake Rouen is from predominantly C-Coll samples; this is like restoring a real organ with pipes from another organ but by the same builder. With proper voicing, such stops can probably be adopted successfully. The same can be done in HW if you really know your stuff on all levels. The result has to remain utterly musical, and not just utilitarian.

By combining a few stops from here and from there, from different builders, different periods, old and new; I think you will indeed have great problems making it all sound cohesive, properly scaled, voiced, musical, satisfying, etc. Such a task is probably as hard and time consuming as recording a real organ from scratch and making a sample set.

When you look at a specification and the think "Oh I can play this and that on this organ" you are often quite mistaken, for a specification doesn't tell me or you anything about the sound. There are so many, many differences between organs, and that is kind of mind boggling.

Of all the sample sets you currently have, does any organ sound very much like another? Not in my vast collection. That summarizes my point exactly.

Dear Adri,
Just a few examples

* I occasionally play a small church organ, five or six stops, all divided in bass and descant. To my knowledge, nothing similar exists as a sample set
* Italian organs often have a stop for every single rank in a mixture, so you can build up a mixture while playing. Wouldn't that be an interesting possibility for other styles too?
* How do rare mutations such as 1 1/7 and 8/9 stops sound? I know of only one sample set that has such mutations in the stoplist - the PAB, but this is a rather dry sample set.

Of course, all kind of such things can be easily achieved using the CODM, but as you say, all depends on the quality of the result. It seems to me a valid question what can be done to obtain a musical result in these and other cases. Organ builders knowledge is hard to get!

As another example of why CODM can be useful, there isn't an ambient (non-dry) English sampleset from the early 1800s or earlier. The available ambient samplesets don't have large enough dispositions to reasonably play the 30 John Stanley voluntaries from the 1750s.

A CODM can arrange 3 manuals from the Little Waldingfield ranks with sufficient tonal resources to play that music comfortably, even though there are authenticity compromises involved. Better would be including Avery and Davis in a 3 manual, but their ranks are disallowed for CODM use.

Unless I'm mistaken, there are 293 sample sets mentioned on the PCOrgan website. And with some publishers not mentioned there is currently more than 300 sample sets that are available to Hauptwer users. There are all styles, sizes, prices (almost a hundred free), wet and dry. To not find the one that suits you without tinkering yourself you must make great efforts. Hauptwerk users are great spoiled children.

I guess it is all about hobbyism. Some people love to make recordings. Others love tinkering. Others are collectors. Hobbyists are no complainers, they just love their hobby. It does not make sense to suggest they should stop. It does make sense to challenge them to improve.

It is very important that hauptwerk offers this possibility of editing, however I agree with you that the results are not always satisfactory. Often, there is a combination of sample sets or simple extensions of an original sample set with questionable sound lines, some even bordering on the absurd. Among the composite sample sets I find the Zurek V5 very nice and well blended.

The ‘maker’ builds the parts mostly from scratch; makes his own pipes, makes his own designs, and employs craftsman and artisans, and builds according to historically proven and inspired approaches and styles, and pays enormous attention to pipe scales, metal alloys, acoustics, final voicing, wind pressures, tunings, etc. Such organs inspire, are beautiful to see, etc. Mostly they are mechanical action instruments, but sometimes with electric stop action.

I disagree with your observations. There are plenty of examples of poor (real) pipe organs whether through the makers lack of experience, or perhaps through experimentation that may have been less than successful. It is through the latter, whether real or virtual, that builders become artisans and find new possibilities. Exploration and creativity (in any context) should not be frowned upon and less than desirable results should be viewed as learnings rather than failure.

To jwillans: please give me some examples of what you are saying.
I am not sure what you wrote is actually in disagreement with my statements.
Thanks.

And this is why I don't spend much time reading this forum anymore.

I would like to not watch certain sub forums like this one but keep up e others. Any way to do that Stan?

Hauptwerk users are great spoiled children.

Yes, soured grandfather.

To not find the one that suits you without tinkering yourself you must make great efforts. Hauptwerk users are great spoiled children.

I would like to have an organ with a Glockenspiel (Carillon), as this is something I really miss (take a look at this organ: https://de.wikipedia.org/wiki/Orgel_der ... g_Kufstein). In HW1 this was not that much of a problem, especially since the .organ files were rather simple.

Now here is my task for you: I have an organ console with 4 manuals and I would like to use said console to play a virtual organ that uses all 4 manuals and has a Glockenspiel. Quite an effort, isn't it?