*** NOOB Warning *** Attempting CODM - First Post

This past week I've been experimenting with CODM and find the documentation quite useful and reasonably easy to follow. As well, the step-by-step examples are excellent. It is well-done!

I'm at the requirements for the GUI interface at the moment and have created some pretty wicked-looking hypothetical organs in a week - doing so by cutting and pasting into an XML program. Since I'm going to be driving quite a few real physical ranks - a hybrid - I have to get good at this so that's where I'm concentrating at the moment.

With respect to stop jambs, all of the example organs contain no 3-D images of the stop activated (or not); for example, an activated stop simply lights up and an unactivated stop goes dark in comparison. The HauptwerkStandardImages - as far as I can tell - contain no 3-D images of a traditional stop in activated/unactivated position. So, I'm assuming that if I want that - and I do - I have to do my own bmp images of these states. This is not a trivial undertaking even though I'm pretty good with Photoshop.

Am I correct in this assumption?

I cannot seem to find the 3-D images of the standard St. AnneMosley's stops to use as an example; perhaps they're encoded in a way that obfuscates that. Anyways, I'm learning (but not nearly fast enough), and am excited at the enormous potential of this marvelous program!

This past week I've been experimenting with CODM and find the documentation quite useful and reasonably easy to follow. As well, the step-by-step examples are excellent. It is well-done!

I'm at the requirements for the GUI interface at the moment and have created some pretty wicked-looking hypothetical organs in a week - doing so by cutting and pasting into an XML program. Since I'm going to be driving quite a few real physical ranks - a hybrid - I have to get good at this so that's where I'm concentrating at the moment.

With respect to stop jambs, all of the example organs contain no 3-D images of the stop activated (or not); for example, an activated stop simply lights up and an unactivated stop goes dark in comparison. The HauptwerkStandardImages - as far as I can tell - contain no 3-D images of a traditional stop in activated/unactivated position. So, I'm assuming that if I want that - and I do - I have to do my own bmp images of these states. This is not a trivial undertaking even though I'm pretty good with Photoshop.

Am I correct in this assumption?

I cannot seem to find the 3-D images of the standard St. AnneMosley's stops to use as an example; perhaps they're encoded in a way that obfuscates that. Anyways, I'm learning (but not nearly fast enough), and am excited at the enormous potential of this marvelous program!

Hauptwerk will display images in a series of 'states', usually 2 for stops, being on or off. For stops that appear to move, you would normally use an off (or in) image or an on (or out) image. With tabs, images would be up or down. With expression pedals etc, you would normally have up to 16 different images, which represent each stage of the pedal. These can either be photographs, suitably sized and shaded, or drawings.

There is an art to it and it's very time consuming, especially if you want to make the engraving (text) appear to move correctly with the stop image. I have found that the most realistic method is to 'construct' the stop image using Xara and superimpose suitably sized text on the image, exporting the final result. The stop must change size as well as the text, to give movement.

As it is so time-consuming and often time is money, I can understand Hauptwerk not producing free '3D' images to give away.

The St Anne's images are stored as .bmp files within the sample set's internal folders and are viewable. Take a look, then I suggest you get out the camera and see what you can do.

You don't need to use .bmp files with masks, you can use .jpg files or probably preferable, the newer .png images, with the alpha channel set.

The St Anne's images are stored as .bmp files within the sample set's internal folders and are viewable. Take a look, then I suggest you get out the camera and see what you can do.

Found them - THANKS! - I guess I wasn't looking hard enough in the right places. It's easy to 'capture' these images using a snipping tool (no camera required) and then alter them; of course, you could use those bmp files if you wanted that 'look' and those stops.

I do notice the intentional glare on the stop to give a more lifelike impression (easily accomplished using Photoshop). Still a lot of work - especially angles of the text upon the knob to make a more realistic impression. There is also a lot of shading to go along with this and the text seems low resolution (blurry) - probably intentionally. I'm going to try making a couple of my own this afternoon to see how it goes.

I've never used XARA (just looked it up now - emmm ... looking interesting) ... but I do know Photoshop and know that I think I can do what I want there.

There is an art to it and it's very time consuming, especially if you want to make the engraving (text) appear to move correctly with the stop image. I have found that the most realistic method is to 'construct' the stop image using Xara and superimpose suitably sized text on the image, exporting the final result. The stop must change size as well as the text, to give movement.

Yes, I knew that that would be so.

As it is so time-consuming and often time is money, I can understand Hauptwerk not producing free '3D' images to give away.

So can I; all I was expecting was to find ONE drawknob (two images), but they have, in essence done that with the St.Anne'sMosley stop images.

You don't need to use .bmp files with masks, you can use .jpg files or probably preferable, the newer .png images, with the alpha channel set.

I'll use the .png images for the reasons you suggest.

THANK YOU for your reply to my first query on the FORUM. It's so gratifying to see those who have gone before helping the new entrants who are still wet behind the ears!

Respectfully,
Richard

Found them - THANKS! - I guess I wasn't looking hard enough in the right places. It's easy to 'capture' these images using a snipping tool (no camera required) and then alter them; of course, you could use those bmp files if you wanted that 'look' and those stops.

If you mean capture the St Anne's images and snip them, that would be most unwise, as it would be a breach of the Hauptwerk license conditions. The St Anne's images are proprietary to Hauptwerk/MDA. Believe me, by an error on my part, I nearly found out the hard way!

THANK YOU for your reply to my first query on the FORUM. It's so gratifying to see those who have gone before helping the new entrants who are still wet behind the ears!

You are most welcome.

Good luck.

If you mean capture the St Anne's images and snip them, that would be most unwise, as it would be a breach of the Hauptwerk license conditions. The St Anne's images are proprietary to Hauptwerk/MDA. Believe me, by an error on my part, I nearly found out the hard way!

Nope ... wouldn't do that for the reasons you specify and also that I don't particularly like them - especially the shading and blurriness. I want to design my own 'look' and started by taking a digital photo of one of the drawknobs on our Casavant organ - surely (the ubiquitous) they can't complain! I'm importing those two images into Photoshop 5 and working (using PNG) to accomplish the task that way.

This alone is not an insignificant task - ie. circles aren't circles, but slight ellipses, when viewed in a 3-D context. Anyways, we'll see how it goes. Also perspective means that EVERY stop should be slightly different (don't think that's going to happen, though). And Left and Right jambs ... I will probably just reflect the left jamps in the y-axis.

Again, thanks for your ongoing hints, traps, and otherwise invaluable data..

[mdyde]

Found them - THANKS! - I guess I wasn't looking hard enough in the right places. It's easy to 'capture' these images using a snipping tool (no camera required) and then alter them; of course, you could use those bmp files if you wanted that 'look' and those stops.

We don't mind you editing copies of the St. Anne's images files, as long as it's strictly for your own use in Hauptwerk and you don't redistribute them. But if you wanted to redistribute them in any way then you'd need to make your own images from scratch.

There is also a lot of shading to go along with this and the text seems low resolution (blurry) - probably intentionally.

The text is actually taken from photographs of the draw-knobs on the real organ. If you want a draw-knobs's text to move with the draw-knob, then you'll need to include the text in the image itself.

This past week I've been experimenting with CODM and find the documentation quite useful and reasonably easy to follow. As well, the step-by-step examples are excellent. It is well-done!

Thanks - glad you like it!

Nope ... wouldn't do that for the reasons you specify and also that I don't particularly like them - especially the shading and blurriness. I want to design my own 'look' and started by taking a digital photo of one of the drawknobs on our Casavant organ - surely (the ubiquitous) they can't complain! I'm importing those two images into Photoshop 5 and working (using PNG) to accomplish the task that way.

This alone is not an insignificant task - ie. circles aren't circles, but slight ellipses, when viewed in a 3-D context. Anyways, we'll see how it goes. Also perspective means that EVERY stop should be slightly different (don't think that's going to happen, though). And Left and Right jambs ... I will probably just reflect the left jamps in the y-axis.

Again, thanks for your ongoing hints, traps, and otherwise invaluable data..

It's my opinion that you can go too far with realism, especially on a flat screen and at the expense of getting the best sound out of the instrument, which is really the main criteria.

The St Anne's concept is a great flagship and as Martin says, actual photographs of real stops, complete with text, are probably as good as it gets with current technology.

Nothing wrong with Photoshop, I just prefer Xara.

It's my opinion that you can go too far with realism, especially on a flat screen and at the expense of getting the best sound out of the instrument, which is really the main criteria.

The St Anne's concept is a great flagship and as Martin says, actual photographs of real stops, complete with text, are probably as good as it gets with current technology.

Nothing wrong with Photoshop, I just prefer Xara.

Thanks for your continuing comments.

I took high-resolution pictures of a draw-stop on our Casavant organ - at different angles with the stop both 'ON' and 'OFF'. The results were very unsatisfying - mainly because the stem of the stop is black and it's quite hard to get it looking realistically not only because of the blackness, but also because of the distortion of the text showing the engraving on the stop. You're right - that was a step too far in realism - and, I think, it failed the test.

So, I finally just took a (hi-res) picture front on and used Photoshop to give the two states. Here are two examples (PNG format) of what it looks like: (400 x 390 pix)
with the corresponding


I wanted something that was easily read and yet clearly shows the state of the stop. I jettisoned the 3-D image tryouts altogether although it looked sort-of cute ... it was harder to read. This is the result so far.

In investigating the mess of varnish-coated wiring in the organ itself (!! ) to which I'm having to connect MIDI driver boards, I found that the stop action energizes the rank of pipes to go from unconnected to ground (via a relay). To play the note, then, you have to nominally apply +15VDC to the magnet. Is it possible in Hauptwerk to have that occur (via MIDI) - ie. choosing the desired stop will then ground the rank of pipes while the midi chooses the notes (normally) via Hauptwerk? An Aside: I think satan wired our organ - it's so messy - too many hot fixes by different organ technicians! I'd give a picture, but I'm embarrassed.

Now, when I (shortly) buy a sample set - say the Vol I and II Salisbury samples sets (I know the latter is not yet released), I'll obviously have to use CODM with images like the above for a hybrid result. Will Hauptwerk allow this? You don't have to tell me it's a lot of work - it takes me 10-min to make a pair of stop PNG images in Photoshop alone.

While the whole daunting process is grinding along - one step at a time - I'm trying to read both Hauptwerk user manuals whether ranks of pipes can start on any of 96 (my MIDI driver boards have 96 physical notes each) position on that board. For example, I have an 8' (real pipes) Oboe that has 61-notes. If that was on, say, channel 5 MIDI out, that leaves 35-notes on that board - on channel 5. Am I able to start the next stop at position 62 to 96 and carry the rest over to another MIDI driver board starting at another user-defined location?; otherwise, I'll run out of MIDI channels and have many unsubscribed drivers per board not used.

Apologies for both the volume and in-depth analysis in all of this. Yes, I'm reading both of the manuals but there's so much there and so much to learn.

[B. Milan]

Hello,

Now, when I (shortly) buy a sample set - say the Vol I and II Salisbury samples sets (I know the latter is not yet released), I'll obviously have to use CODM with images like the above for a hybrid result. Will Hauptwerk allow this?

Most sample sets will allow you to create a CODM format of it, however there are some available that the licensing does not allow this.

However our sets do, so this would allow you to create your own CODM of the Salisbury set if you wish.

Am I able to start the next stop at position 62 to 96 and carry the rest over to another MIDI driver board starting at another user-defined location?; otherwise, I'll run out of MIDI channels and have many unsubscribed drivers per board not used.

Hauptwerk allows using any channel any any event number for stops/couplers etc. and they do not need to be located on the same board, so that should work for your circumstance so long as they do not conflict with the keyboard MIDI note on/off events used to play the keyboards.

[mdyde]

Hello RAWAbacus,

While the whole daunting process is grinding along - one step at a time - I'm trying to read both Hauptwerk user manuals whether ranks of pipes can start on any of 96 (my MIDI driver boards have 96 physical notes each) position on that board. For example, I have an 8' (real pipes) Oboe that has 61-notes. If that was on, say, channel 5 MIDI out, that leaves 35-notes on that board - on channel 5. Am I able to start the next stop at position 62 to 96 and carry the rest over to another MIDI driver board starting at another user-defined location?; otherwise, I'll run out of MIDI channels and have many unsubscribed drivers per board not used.

MIDI output to any given rank/windchest/keyboard does require that each of the pipes/pallets/keys within that rank/windchest/keyboard have consecutive MIDI numbers. Hence you can't split a rank of pipes across two MIDI controllers, for example. However, as Brett mentioned, you can use spare encoder pins for moving/illuminated draw-knobs/tabs or lighted pistons, for example.

otherwise, I'll run out of MIDI channels

You can get multi-port MIDI interfaces, such as the MOTU Microlite:

http://www.motu.com/products/midi

In investigating the mess of varnish-coated wiring in the organ itself (!! ) to which I'm having to connect MIDI driver boards, I found that the stop action energizes the rank of pipes to go from unconnected to ground (via a relay). To play the note, then, you have to nominally apply +15VDC to the magnet. Is it possible in Hauptwerk to have that occur (via MIDI) - ie. choosing the desired stop will then ground the rank of pipes while the midi chooses the notes (normally) via Hauptwerk?

You can configure Hauptwerk to send MIDI output from its virtual draw-knobs.

Hauptwerk allows using any channel and any event number for stops/couplers etc. and they do not need to be located on the same board, so that should work for your circumstance so long as they do not conflict with the keyboard MIDI note on/off events used to play the keyboards

I didn't know that it was possible for MIDI inputs (keyboards, pistons, etc.) to conflict with MIDI outputs. I'm trying to keep everything straight.

MIDI output to any given rank/windchest/keyboard does require that each of the pipes/pallets/keys within that rank/windchest/keyboard have consecutive MIDI numbers. Hence you can't split a rank of pipes across two MIDI controllers, for example. However, as Brett mentioned, you can use spare encoder pins for moving/illuminated draw-knobs/tabs or lighted pistons, for example.

I think I understand (and at the 11th hour as my order for MIDI driver boards was about to be dispatched.) So, I'm assuming that each rank of (real) pipes has to have its own MIDI driver board or, perhaps, if two (full) ranks could fit on ONE physical MIDI driver board, that this would be OK. For example, many pedal stops are 32 physical notes, and one of those could co-exist with a 61-note rank ... in my case, 32 + 61 = 93 contiguous notes in two ranks - leaving 3 spare outputs on a 96-note MIDI driver board. It would seem then that each physically different rank of pipes - excluding pedal ranks - would have to have its own MIDI driver board and that would mean that I would need separate multi-port MIDI interfaces because of too many channels occupied. I hope I'm right in that.

THANKS ever-so-much for your ongoing help as I struggle to get up-to-speed. Your help is invaluable.

[mdyde]

Hello RWAbacus,

I didn't know that it was possible for MIDI inputs (keyboards, pistons, etc.) to conflict with MIDI outputs. I'm trying to keep everything straight.

No - it isn't possible for MIDI input to conflict with MIDI output. I think Brett was referring entirely to MIDI output (or entirely MIDI input).

I think I understand (and at the 11th hour as my order for MIDI driver boards was about to be dispatched.) So, I'm assuming that each rank of (real) pipes has to have its own MIDI driver board or

Whether you need MIDI output at the rank level or only at the wind-chest (=division) level depends on how your wind-chests and their relays are wired.

Basically if a wind-chest needs separate on/off signals for each pipe individually (as opposed to needing only one per note, with that note-on/off controlling all pipes of the same pitch on that wind-chest simultaneously, for ranks whose stops are on - like a slider wind-chest) then you need rank-level output (and thus potentially a MIDI decoder per rank).

Hence it's essential that you determine that fully first.

The relevant two screens in Hauptwerk, which you would use to configure either of those two approaches respectively, are:

- 'Organ settings | Advanced MIDI applications | Direct MIDI input/output for ranks/pipes' (Output tab):



- 'Organ settings | Advanced MIDI applications | Direct MIDI output from divisions':

or, perhaps, if two (full) ranks could fit on ONE physical MIDI driver board, that this would be OK. For example, many pedal stops are 32 physical notes, and one of those could co-exist with a 61-note rank ... in my case, 32 + 61 = 93 contiguous notes in two ranks

As long as you only ever use your MIDI pipework with organs you've created yourself using the CODM, then yes - that would be fine, since within your CODM organ definition files (ODFs) you could specify any key-shifts (between -128 and +128) for any divisions or ranks.

However, there is a potential benefit to using only the conventional MIDI note numbering for each decoder (e.g. MIDI note numbers 36-96 for a 5-octave non-unified rank or division): you would then have the option to use off-the-shelf (non-CODM) organ definitions with your MIDI pipework by setting some of the virtual ranks or divisions to send MIDI output to your hardware ranks or divisions and disabling the corresponding virtual ranks in Hauptwerk.

For example, with the St. Anne's sample set you might disable the virtual Swell Oboe rank's sound (using the 'Organ | Load organ, adjusting rank audio/memory options/routing' screen) but set that virtual rank to control a rank of your MIDI pipework via MIDI output from it. You MIDI rank would then sound in place of the virtual Swell Oboe rank.

That wouldn't be possible if your MIDI rank needed a different numbering (key-shift), since Hauptwerk doesn't allow key-shifts to be specified for rank or division MIDI output via user settings (only within the CODM).

would have to have its own MIDI driver board and that would mean that I would need separate multi-port MIDI interfaces because of too many channels occupied. I hope I'm right in that.

If you have more than 16 decoder boards (=MIDI channels), then yes - you would need a multi-port MIDI interface with at least one MIDI OUT port per 16 MIDI channels (decoder boards). The MOTU Microlite is well-proven (I use one myself and can recommend them).

Whether you need MIDI output at the rank level or only at the wind-chest (=division) level depends on how your wind-chests and their relays are wired.

Basically if a wind-chest needs separate on/off signals for each pipe individually (as opposed to needing only one per note, with that note-on/off controlling all pipes of the same pitch on that wind-chest simultaneously, for ranks whose stops are on - like a slider wind-chest) then you need rank-level output (and thus potentially a MIDI decoder per rank).

Hence it's essential that you determine that fully first

I'm strongly with Martin on this. It's important to work out exactly how and what needs to be driven/energised in the organ before investing (probably heavily) in MIDI hardware that you might or might not need and which might or might not be suitable.

As an example, It sounds as though you are intending to drive pallet magnets directly from the MIDI decoders - are you sure that they are capable of this?

Nick

I'm strongly with Martin on this. It's important to work out exactly how and what needs to be driven/energised in the organ before investing (probably heavily) in MIDI hardware that you might or might not need and which might or might not be suitable.

As an example, It sounds as though you are intending to drive pallet magnets directly from the MIDI decoders - are you sure that they are capable of this?

This organ was originally a tracker organ but when these mechanisms failed some time ago and before my time, it was subsequently electrified, with electric valve magnets under each pipe. I've determined that those magnets are activated in two ways.

Principally, each rank of pipes is connected together on one side of the direct pallet magnet and that common-wired line was left floating - neither grounded or at any electrical state. When a stop is activated - say an 8' oboe, the stop drawknow causes that common side to be grounded using one relay. Then +15VDC is applied to the other side of the magnet via the keys; this would be the place where MIDI negative common pipe driver boards would be connected - supplying +15VDC to individual pipes.

I've tried & verified this by manually applying these voltages to the required places and the pipes do sound as predicted.

While most of the stops work this way, a few other stops work differently. The pipes are wired to ganging switches - some with up to 73 contacts each and these long switches are grouped with common silver crossbars. Applying +15VDC to these common crossbars in these long switches causes the pipe to sound. I'm finding that these switches seemed to be used for mutation stops (some of them have offset starts).

So, yes, I've verified that MIDI pipe driver boards would work.

Basically if a wind-chest needs separate on/off signals for each pipe individually (as opposed to needing only one per note, with that note-on/off controlling all pipes of the same pitch on that wind-chest simultaneously, for ranks whose stops are on - like a slider wind-chest) then you need rank-level output (and thus potentially a MIDI decoder per rank).

So, I think the above description rules out slider wind-chest.

In the dialog box 'Organ settings | Advanced MIDI applications | Direct MIDI input/output for ranks/pipes' (Output tab): which I have studied before, I see designation of the MIDI OUT port (using micro lite: Port 1), and the MIDI channel (1 - 16), (and a test button for middle "C"), but I don't see where you can set an offset within that channel - again using my example of physical output 1 --> 61 for Oboe 8' and then using the rest of the MIDI driver board for, say, Bourdon 16' using the same channel but offset +61 to get to unused MIDI outputs (for 32 notes). That's the only thing I don't see how to do at the moment (at this design stage).

The picture is certainly clearing


Again, thanks to all for commenting.

[mdyde]

Hello RWAbacus,

In the dialog box 'Organ settings | Advanced MIDI applications | Direct MIDI input/output for ranks/pipes' (Output tab): which I have studied before, I see designation of the MIDI OUT port (using micro lite: Port 1), and the MIDI channel (1 - 16), (and a test button for middle "C"), but I don't see where you can set an offset within that channel - again using my example of physical output 1 --> 61 for Oboe 8' and then using the rest of the MIDI driver board for, say, Bourdon 16' using the same channel but offset +61 to get to unused MIDI outputs (for 32 notes).

My previous reply covered that.

Summary: there is no user setting for rank output key-shift, so if you want to use more than one rank per decoder then you would be restricted to using only your own custom CODM organ definitions, since only within the CODM can specify such a key-shift. Hence my recommendation would probably be only to have one rank per decoder.