It is currently Fri Mar 29, 2024 9:12 pm


bad sounding sample sets because too perfect

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

Theorbe

Member

  • Posts: 97
  • Joined: Sat Jan 11, 2020 7:10 am

Re: bad sounding sample sets because too perfect

PostSun May 30, 2021 11:47 am

Deleted by author.
Last edited by Theorbe on Sat Jan 13, 2024 6:53 pm, edited 1 time in total.
Offline

TubaFan

Member

  • Posts: 54
  • Joined: Fri May 15, 2020 8:57 am

Re: bad sounding sample sets because too perfect

PostMon May 31, 2021 9:01 am

Theorbe wrote:
TubaFan wrote:This is detuning entire ranks, correct? What is the maximum detuning allowed?


Hi TubaFan

For each virtual rank which remains listed after any manual removals, the app randomly detunes every note separately of each rank. So for the avoidance of doubt, it's not simply moving an entire rank up or down by a random amount.

The maximum Standard Deviation is 10 but than can easily result in several random values exceeding Hauptwerk's maximum of 50 or -50 (and so will be limited to that). So more than enough to make it sound awful.

Regards

Andy


Thanks for your reply. I'm going to try it out.
Offline

1961TC4ME

Member

  • Posts: 3144
  • Joined: Mon Feb 07, 2011 1:45 pm
  • Location: Lake Minnetonka, Minnesota

Re: bad sounding sample sets because too perfect

PostTue Jun 01, 2021 11:19 am

Hello all,

A couple observations and a couple questions as well. First, I've stayed with the 2.5 detuning default for starters and tried this utility on 2 instruments so far. Other than some amplitude voicing here and there to either tame down a few overly pronounced pedal stops / notes and a raising of amplitude here and there for a few stops that were a bit weak, both instruments otherwise have been left as delivered, so no previous de-tuning on my part.

On the St. Max I can't really say it helped anything, maybe more so it just 'changed' things is the best description and perhaps it's meantone tuning in the first place is the main contributor. The reeds do have a bit more of an interesting chorus sound to them, but overall the instrument sounds too out of tune to my ears using the 2.5 default, so perhaps I need to try a different value for the St. Max.

On the other hand, for the Armley Schulze I'd say the outcome was much more positive. Playing a Trumpet Tune in particular really made the trumpet part stand out more. A few of the trumpet notes really stand out and you can hear them, but overall not in a really annoying way, I'd say more so in a pleasing way. Overall the rest of the instrument sounds pretty good, however on both instruments the 2.5 default (to me anyways) makes the instruments sound 'too' out of tune.

Questions: How do I re-set the values back zero to try a new detuning value, or in other words set the instrument back to as it was delivered with no additional retuning? Or, if I simply enter a new value, say 2.2 as an example and then click 'start detuning' will the utility then re-tune to the new value? I'm assuming that's the case.

Not knowing how dramatic a difference between values makes, is going from say 2.5 to 2.4 a pretty big jump, or is it more like going from 2.0 to 2.5 that is a rather big jump?

And finally, is anyone else thinking 2.5 is too much and have they come up with a better value to try?

I'd say overall well done and it certainly helps add realism. More experimenting to come.

Marc
Offline

josq

Member

  • Posts: 912
  • Joined: Tue Jan 24, 2012 12:11 pm

Re: bad sounding sample sets because too perfect

PostTue Jun 01, 2021 11:47 am

Yes, in my experience 2.5 is too much. I choose 1 or 1.5 in most cases.
Offline

CWEB

Member

  • Posts: 43
  • Joined: Thu Jun 22, 2017 8:11 am

Re: bad sounding sample sets because too perfect

PostTue Jun 01, 2021 12:06 pm

Here's a thought.

Real organs get tuned pipe by pipe, harpsichordists traditionally do there own tuning. Why don't we do the same with hauptwerk? At least we know there will be no drift or instability once we are done! :twisted:
Offline
User avatar

Jan Loosman

Member

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

Re: bad sounding sample sets because too perfect

PostTue Jun 01, 2021 4:10 pm

Hello Marc

You mentioned the st Max that the detuning didn’t do much.
The strenght of this app is that it will synchronise all channels. So you will only notice a positive effect on multichannel sets. But as far is i know the sint Max is just stereo and in that case you can use the original Hauptwerk detuning it will perform the same as the app.
I use 1.5 detuning it sounds good so far.

Correction this set is multichannel i have the stereo version and thought being an old set it was stereo only.

Jan
Last edited by Jan Loosman on Tue Jun 01, 2021 4:57 pm, edited 2 times in total.
Offline
User avatar

CarsonCooman

Member

  • Posts: 173
  • Joined: Sat Dec 03, 2011 10:01 am
  • Location: Cambridge, MA

Re: bad sounding sample sets because too perfect

PostTue Jun 01, 2021 4:32 pm

Jan Loosman wrote:But as far is i know the sint Max is just stereo...


No, the currently available version is surround (front/rear). It's possible an earlier version was stereo only, but the surround version has existed for quite some time now.
Offline
User avatar

Jan Loosman

Member

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

Re: bad sounding sample sets because too perfect

PostTue Jun 01, 2021 4:51 pm

CarsonCooman wrote:
Jan Loosman wrote:But as far is i know the sint Max is just stereo...


No, the currently available version is surround (front/rear). It's possible an earlier version was stereo only, but the surround version has existed for quite some time now.


You are right. I own the stereo version and did a quick check on the SP site and still didn’t notice.

So my comments to Marc are not correct!

Thanks Carson!!
Offline

1961TC4ME

Member

  • Posts: 3144
  • Joined: Mon Feb 07, 2011 1:45 pm
  • Location: Lake Minnetonka, Minnesota

Re: bad sounding sample sets because too perfect

PostWed Jun 02, 2021 7:47 am

Hello Carson and Jan,

Yes I indeed have the surround St. Max. It's not that it didn't do much, it's more so the utility set at 2.5 didn't do the St. Max any favors. But, more experimenting as we speak.

EDIT: 1.2 seems about right for St. Max, maybe 1.0. As for Armley Schulze closer to 2.0 works good although the default of 2.5 is interesting, but is perhaps still a bit too much. In either of the 2 cases the detuning does make for a more lively and pleasing sound.

Marc
Offline

jean.lemoine

Member

  • Posts: 10
  • Joined: Mon Jan 27, 2020 5:12 am

Re: bad sounding sample sets because too perfect

PostThu Jun 03, 2021 2:47 am

Theorbe wrote:For those who may be interested, my utility for automated random detuning is available from the following OneDrive link:

https://1drv.ms/u/s!ArppdH7JT_uasCD2SWaxzH9gIsOr?e=X1FVxe


Many, many thanks for this, Andy!
I have fiddled with it while waiting for being able to install Caen 2.66 and I fell in love with it!
I actually prefer Caen with your utility than with v2.66's pipe detuning+coupling features...

I wish all great pleasure using it and can't help but looking forward to seeing what nitty features you might add in the future (like, applying specific patterns to reeds or distinct divisions...).

Many thanks again!

Jean

PS: I'd love to be refreshed on detuning factors and their precise influence, the different patterns, etc. But this might rather have its place in another thread.
Offline

jean.lemoine

Member

  • Posts: 10
  • Joined: Mon Jan 27, 2020 5:12 am

Re: bad sounding sample sets because too perfect

PostThu Jun 03, 2021 4:02 am

1961TC4ME wrote:Hello all,

Questions: How do I re-set the values back zero to try a new detuning value, or in other words set the instrument back to as it was delivered with no additional retuning? Or, if I simply enter a new value, say 2.2 as an example and then click 'start detuning' will the utility then re-tune to the new value? I'm assuming that's the case.



Open the voicing window (should be like "Organ settings > Ranks routing ... panning" or something).
Select ALL the stops at once, then select the tuning entry in the drop down menu (I guess it's the first entry).
Now, at the lower-left end of the window pane that display the sliders for each note, you shall see a "Reset" button.
Click it and voilà :-)

1961TC4ME wrote:And finally, is anyone else thinking 2.5 is too much and have they come up with a better value to try?


As for St Maximin, I don't play it very often but I was so far satisfied with the original (untuned) temperament. I just need from times to times to check that I have not inadvertantly reverted to the default equal temperament.
The original temperament being the actual tuning at recording time, that may explain why you need to set a very low value.

I've not very much played with the utility yet and it was mostly with Caen while waiting for 2.66.
I've first tried with a 2.5 deviation. I was not very impressed.
Then I've tried with 4. I was then like "woah that organ deserves a good tuning!" It was so realistic!
Then I've tried with 3.5. I found it amazing with a few notes deserving a bit of tuning :-)
I'm now experimenting with the regions of 3.2~3.4.

I'm currently performing tests with Zwolle. This organ is so dead flat... I haven't found the sweet spot yet, perhaps because this organ has many multi-ranks stops that are so sterile by themselves. But it will probably be in the 3~3.5 region too.

I'm not in a hurry to test with other instruments as I think Oloron, St Michel are already FANTASTIC and my daily work horse of Dudelange (I have my actual lessons there) is already realistic enough (I use the untuned flavor).

I wish you lots of renewed joy playing with this app and many many thanks again to Andy for producing this life-changing piece of software!
Offline

Theorbe

Member

  • Posts: 97
  • Joined: Sat Jan 11, 2020 7:10 am

bad sounding sample sets because too perfect

PostThu Jun 03, 2021 12:34 pm

Deleted by author.
Last edited by Theorbe on Sat Jan 13, 2024 6:54 pm, edited 1 time in total.
Offline

kasterling

Member

  • Posts: 19
  • Joined: Thu Feb 20, 2020 7:04 pm

Re: bad sounding sample sets because too perfect

PostThu Jun 03, 2021 7:17 pm

Andy, all these reports coming in about the success of your utility with various sample sets is really exciting. I was particularly thrilled to hear the benefits to the Armley Schulze set, as it is one that I own and think very highly of. I can only hope that your efforts will eventually turn toward making your utility compatible with HW4.2. As I am unable to switch to the later versions, I am waiting with eager anticipation for the release of your utility in a version compatible with what I have. I know many others reading this feel the same thing. Three cheers to you for stirring such excitement!

Kevin
Offline
User avatar

Doug S.

Member

  • Posts: 498
  • Joined: Tue May 20, 2008 8:17 pm
  • Location: Massachusetts USA

Re: bad sounding sample sets because too perfect

PostFri Jun 04, 2021 12:06 pm

In CODM, is there a default value for random detuning? What value would one need to achieve the desired effect being discussed?
Thanks,
Doug
Doug
Offline
User avatar

mdyde

Moderator

  • Posts: 15446
  • Joined: Fri Mar 14, 2003 1:19 pm
  • Location: UK

Re: bad sounding sample sets because too perfect

PostFri Jun 04, 2021 1:04 pm

Doug S. wrote:In CODM, is there a default value for random detuning? What value would one need to achieve the desired effect being discussed?


Hello Doug,

The defaults for attributes are listed via the 'Design tools | View Custom Organ Definition ... format definition' document. (As listed there) there's only attribute in the CODM for controlling random detuning (so to keep the CODM format simple), which is Rank.Pitch_MaxRandomTuningErrorHz, the default for which is 0.25:

"Each time that an organ is loaded or activated, Hauptwerk applies a randomized detuning to each pipe, to model slight tuning errors in the organ. Thus setting determines the maximum detuning in Hertz that is allowed for the rank. The probability function is weighted towards perfect tuning, so large tuning errors are fairly unlikely. Often, tuning errors are larger for reed ranks than diapasons."

(I can't help you with how its probability distribution compares to that of Andy/Theorbe's utility, I'm afraid.)
Best regards, Martin.
Hauptwerk software designer/developer, Milan Digital Audio.
PreviousNext

Return to Hauptwerk instruments

Who is online

Users browsing this forum: No registered users and 12 guests

cron