Bug? Capture component, some fixture models have no orientation markers

Welcome to the Avolites TitanOne Support Forum

Moderator: Moderators

revenant
Posts: 51
Joined: 25 Oct 2020, 14:36

Bug? Capture component, some fixture models have no orientation markers

Postby revenant » 02 Nov 2020, 11:08

Titan Go V14
Fixture Library (V14) 2020-11-02 05:54:06

Most of the fixture models within the Capture component have a small circular element on the fixture to denote the power tail, I use this to determine the fixture's orientation. There are some fixture models that do not have this element or any other element that could be used as an orientation marker. Unless you are very familiar with the physical fixture the lack of any sort of orientation marker makes it difficult to impossible to determine the fixture orientation.

The fixture models missing this detail that I have noticed today are all from Clay Paky, the specific fixtures are -

Sharpy Standad, 16 DMX
Shapry Wash 330 Standard (CMY)
Scenius Spot Standard, 32 DMX
HY B-EYE K15 Standard, 21 DMX.
User avatar
niclights
The eManual
Posts: 4442
Joined: 24 Sep 2004, 01:06
Location: UK

Re: Bug? Capture component, some fixture models have no orientation markers

Postby niclights » 02 Nov 2020, 11:30

Thank you for highlighting this. It appears to be related to where the fixture has a detailed 3D model as opposed to a generic. I will discuss this with Capture.
revenant
Posts: 51
Joined: 25 Oct 2020, 14:36

Re: Bug? Capture component, some fixture models have no orientation markers

Postby revenant » 02 Nov 2020, 13:43

Thank you niclights,

There is another related issue that you may wish to raise with them at the same time, all of the fixtures I listed are placed in Capture with a non-standard default orientation.

The majority of fixtures I have used in Capture have a default orientation of power tails pointing stage left when initially placed, although it does not appear to be documented anywhere, as it seems to be the case with the majority I have taken this as the undocumented system default orientation?

Of the fixtures I listed, the Sharpy and Sharpy Wash are placed with a default orientation of power tails pointing downstage, the Scenius Spot and HY B-EYE K15 default to power tails pointing upstage, I have also just placed a Clay Paky MINI B in Capture and it's default orientation is power tails pointing stage right.

Apart from the inconsistency, a side effect of this is when you perform a fixture exchange any offsets applied to the 3D models are also applied to the models that are exchanged. For example, if I take the HY B-EYE K15 and apply an offset of 180° on the Y axis to orientate it power tails pointing stage left, then I exchange with a fixture that defaults to power tails stage left a 180° offset will be applied to the Y axis and it will end up being orientated power tails pointing stage right. A minor annoyance at best but if it's another fixture without orientation markers then you have idea it's happening. If there is no actual default orientation I do think there should be, I feel that consistency in this area would be beneficial.
User avatar
niclights
The eManual
Posts: 4442
Joined: 24 Sep 2004, 01:06
Location: UK

Re: Bug? Capture component, some fixture models have no orientation markers

Postby niclights » 02 Nov 2020, 13:45

Further to this Capture have confirmed this is currently true for some fixtures.

The default orientation in Titan (ie. X = 0°, Y = 0°, Z = 0°) is such that when pan is set to locate (almost always 50%/127 DMX), increasing the tilt value will tilt the head towards the camera. Many fixtures have an arrow indicating this where the default orientation in Titan will be equivalent to the arrow pointing towards the camera. This helps in ambiguous situations such as where there is both a power in and power out connector located on opposite sides of a base or where the display might be located on the head rather than the base.

I appreciate this isn't ideal and I have suggested it would be very helpful to include some standardised orientation details for all fixtures in the future if possible. In the meantime hopefully this will be enough to ensure your simulation matches the real world.

edit: Regarding default orientations if you are referring to the stand-alone Capture then to quote Capture's reply directly:
An unrotated fixture will have it's display along the positive Z-axis in the direction from the stage towards the audience. You can also swith the view to plot mode and see the location of the pigtail/power whip which is indicated by a power symbol / bolt.
revenant
Posts: 51
Joined: 25 Oct 2020, 14:36

Re: Bug? Capture component, some fixture models have no orientation markers

Postby revenant » 02 Nov 2020, 14:18

Thank you again niclights,

This additional information is appreciated and I now understand why these fixtures are being placed with such orientation. I would think this is beneficial to those who want to quickly mock up a stage to try out some programming but seems less useful if you want to accurately model a physical stage.

Can you clarify, should I be seeing an arrow of some sort on the Clay Paky fixture models I listed previously? I'm not seeing any arrows and I don't recall ever seeing any arrows on any fixture models within Capture desktop or component.
User avatar
niclights
The eManual
Posts: 4442
Joined: 24 Sep 2004, 01:06
Location: UK

Re: Bug? Capture component, some fixture models have no orientation markers

Postby niclights » 02 Nov 2020, 14:38

There are currently no arrows shown in the models. This is essentially what I was suggesting could be improved in the future.
revenant
Posts: 51
Joined: 25 Oct 2020, 14:36

Re: Bug? Capture component, some fixture models have no orientation markers

Postby revenant » 02 Nov 2020, 15:31

Ah, I see, understood, thank you.
revenant
Posts: 51
Joined: 25 Oct 2020, 14:36

Re: Bug? Capture component, some fixture models have no orientation markers

Postby revenant » 02 Nov 2020, 21:07

niclights wrote:The default orientation in Titan (ie. X = 0°, Y = 0°, Z = 0°) is such that when pan is set to locate (almost always 50%/127 DMX), increasing the tilt value will tilt the head towards the camera.

I am not sure if you checked this against the fixtures I listed previously, I have put it to the test and the results I'm seeing are a little inconsistent. Using the same V14 2020-11-02 05:54:06 fixture library version as previous, I patched the following fixtures -

1x Clay Paky Sharpy Standad, 16 DMX
1x Clay Paky Shapry Wash 330 Standard (CMY)
1x Clay Paky Scenius Spot Standard, 32 DMX
1x Clay Paky HY B-EYE K15 Standard, 21 DMX
1x Clay Paky MINI B Basic RGBW, 16 DMX
1x Philips SL Beam 300 FX 16-Bit, 1 Cell, 25 DMX

If I open the Capture component window, select the fixtures, locate them and then increase the value for tilt, I see that the HY B-EYE K15 and MINI B do tilt towards the camera, the others tilt away from the camera. All the Clay Paky fixtures pan in the same direction when increasing pan, the SL Beam 300 pans in the opposite direction, although I am assuming the orientation rules are only relevant for tilt. From your description it seems that the Sharpy, Shapry Wash 330, Scenius Spot and SL Beam 300 FX have a default orientation some 180° out from expected.

The results of placing the SL Beam 300 FX fixture with the orientation rules you describe is that the orientation of the top box is very much on the wonk which is what I expected and is correct, it just looks a little odd. The SL Beam 300 fixture was really just for my own curiosity as it is a little bit of an oddball, contrary to the manufacturers documentation the fixture has a physical total pan range of 600°. Also, if you subscribe to the notion that the default home position for pan/tilt will have the yoke set up either parallel or perpendicular to the power tails then this fixture also has an asymmetric pan range. I consider this fixture to have a pan range of -240~360°, applying the typical 50%/127 DMX to locate pan on this fixture would normally result in pan being on the wonk relative to everything else in the rig.

This is really just an FYI, I rarely use the internal Capture component, I've only been using it it at the moment to have a quick poke at V14.
User avatar
niclights
The eManual
Posts: 4442
Joined: 24 Sep 2004, 01:06
Location: UK

Re: Bug? Capture component, some fixture models have no orientation markers

Postby niclights » 03 Nov 2020, 10:29

My understanding is that we use a property in the associated Capture fixture definition to determine the orientation when patched in Titan. If these do not result in the beam moving towards the camera when increasing tilt this suggests the Capture fixture is wrong.

Ultimately, just as with the Titan personality files, the accuracy is only as good as the information available and unfortunately in my experience very few manufacturers provide sufficient details and sadly, as with your poignant example, the details are often wrong. I find this very disappointing. However, we are not restricted to manufacturer's documentation and user feedback is very valuable, particularly for things like calibration.

The intention is that the simulation is as close to real-world as possible. If (ignoring default orientations for now), you find when using the latest library that the simulation does not match the real-word (for example default pan/tilt directions, range and pan position relative to the base), it would be very helpful if had the time to provide the correct details.

In terms of pan/tilt calibration Capture have a simple guide for this. If you are a stand-alone Capture customer then the most efficient process would be to contact them directly at library <at> capture <dot> se explaining the issue. If it is pan/tilt they will probably send you the guide and ask that you provide answers to the three questions. We use the same library so any changes to the Capture library will automatically propagate to Titan - these are incorporated on nightly builds so will normally be available in the Titan Fixture Library installers from approx. 09:00 GMT the day after a fixture update confirmation notification from Capture. To avoid confusion I would recommend sending a separate email for each fixture.

Finally, regarding pan direction, the important thing is that it matches the real world. If one type of fixture turns clockwise and another turns anti-clockwise when increasing the pan value then the solution to make them move in the same direction would be to invert the start and end DMX ranges for the control in the Titan personality. This is something I have considered however it is not something we currently do, mainly because we rarely have the calibration information to determine this. We would also need to agree on a standard although this would be purely arbitrary.
revenant
Posts: 51
Joined: 25 Oct 2020, 14:36

Re: Bug? Capture component, some fixture models have no orientation markers

Postby revenant » 03 Nov 2020, 22:37

I can say with certainty that in July of last year Capture's fixtures library file and model for the SL Beam 300 FX fixture was accurate for range and direction of movement on pan and tilt, the only way it would be incorrect now was if it has since been updated for the worse. I have also just patched in 1 x Varilite VL2600 Profile 16bit, 42 DMX and 1 x High End SolaFrame 1000 48 DMX to Titan's Capture component, both of these fixtures also tilt away from the camera when increasing the tilt value, but, Capture's fixture library files for these two fixtures were correct for pan and tilt earlier this year. So, I am not so sure that these fixtures tilting away from the camera when increasing tilt values is down to incorrect base fixture library files from Capture.

I have in the past feed details back to Capture, I'll spare you the details, suffice to say that the level of involvement and time required at my end was just way too much. Now, if using a visualiser I either adjust my library and personality files to match the visualiser then update the files when I get with the physical fixtures, or I do all my programming with fixture types I know have accurate visualiser models then fixture exchange when I get to the physical rig.

The personality files I use do have pan and tilt defined such that when you increase/decrease pan/tilt values the fixtures all move the same way. As an aside, the remote focusing feature from Capture desktop via Titan's ConsoleLink depends on ascending DMX values, if you invert the DMX range within the personality file remote focusing from Capture desktop breaks (fixtures move in the opposite direction within Capture from expected) or that was certainly the case recently when I looked at it.
revenant
Posts: 51
Joined: 25 Oct 2020, 14:36

Re: Bug? Capture component, some fixture models have no orientation markers

Postby revenant » 06 Nov 2020, 11:26

I'm not sure if this helps or not, however, I have a HY B-EYE K15 with me today and can confirm that Capture's fixture model this fixture is correct with regards the range and direction of movement for pan and tilt, this was one of the fixtures that when placed in Titan's Capture component did tilt towards the camera for increasing tilt values from default orientation. No idea what that means, but if the orientation is deemed worthy of investigation it may be helpful to note.

Straying a little off topic but, I noticed that the current Titan system personality file for HY B-EYE K15 does not contain any RDM details other than the manufacturer ID, as that is the area I'm in at the moment here are some details from what I'm seeing with this fixture today -

Code: Select all

Fixture Software Version 2.24.001

RDM DEVICE MODEL - 0x02e2

List of Personalities (12 of)

1. Standard (21 SLOTS)
2. Shapes (35 SLOTS)
3. Standard Freq. (22 SLOTS)
4. Shapes Freq. (36 SLOTS)
5. Standard RGB (21 SLOTS)
6. Shapes RGB (35 SLOTS)
7. Standard Freq. RGB (22 SLOTS)
8. Shapes Freq. RGB (36 SLOTS)
9. Standard RGBW (21 SLOTS)
10. (NOT RESPONDING) (--- SLOTS)
11. Standard Freq. RGBW (22 SLOTS)
12. (NOT RESPONDING) (--- SLOTS)

Guessing 10 is "Shapes RGBW (35 SLOTS)" and 12 would be "Shapes Freq. RGB (36 SLOTS)"

When setting a personality via RDM the controller reports "RESPONSE TIMED OUT" but selected personality is set on fixture. Controller also reports too long to respond from the fixture when scanning rig but it does show up and it is responsive (for most operations).

The "RGB" and "RGBW" personalities are the equivalent of the "Extended RGB" and "Extended RGBW" personalities (SLOT count seems to include basic engine and frequency option only) of the A.LEDA B-EYE K10 that used to be documented within the earlier versions of the "DMX Channels" document for that fixture.

When selecting any of the "RGB" or "RGBW" modes, the underlying "Pixels Engine" mode is also set but "Source", "Universe" and "DMX Address" remain unchanged. Universe can be set via manufacturer specifc PID 0x8618 "Pixel Engine Universe (RGB RGBW)" (min value 0x0 max value 0x2) but there does not appear to be any way to set pixel engine source or address via RDM.

The "Freq." personalities adds 1 additional channel onto the end of the basic engine (standard or shape) and provides control of PWM refresh rates via DMX.

There is a whole host of other things going on with this fixture but I feel way too deep for this thread.

For what it's worth, there was an issue with Capture's model in that the dimmer control of the basic engine was not being seen as a global dimmer so the pixel engine was stuck @ full with no control, this I did feedback to Capture and has been rectified. I have also provided them with the details to correct their cell/pixel layout as currently they are 180° out, I have received a mail to say the layout mappings have been revised and an updated model should be available alter today, it'll be nice not to have to continually fudge my cell mappings to account for their offset. I did inform them that the cell maps are incorrect for all Clay Paky B-EYE models but unclear if they revised all models or just the K15 at this time.
User avatar
niclights
The eManual
Posts: 4442
Joined: 24 Sep 2004, 01:06
Location: UK

Re: Bug? Capture component, some fixture models have no orientation markers

Postby niclights » 06 Nov 2020, 12:53

Apologies. I have not had a chance to reply until now.

Regarding the orientation originally I also tested using some fixtures I was confident in the pan/tilt calibration and found they moved towards the camera when increasing tilt but now I can see that was just coincidence and some move away.

As things stand my understanding remains that the default orientation in Titan is such that the arrow on the fixture would be pointing towards the camera (assuming sufficient and accurate information is available of course).

I had assumed that, while the pan/tilt implementation varies by fixture, the arrow has some sort of standard meaning where with the pan control set to 127 the yoke would be perpendicular to the direction of the arrow and, with pan in this position, increasing tilt would either move the head toward or away from the arrow (one or the other but the same for all fixtures). But perhaps this is not the case and there is no particular standard. I have asked whether this is correct (in which case something might be wrong with the default orientation or the way it is determined in Titan) or whether it only defines the pan. I am waiting a response.

Regarding the Clay Paky HY B-EYE K15 I have added the RDM device ID, the 'Freq.' modes and associated RDM personality ID's (corresponding to numbers 1-4 in the list or personalities detailed). Thank you for providing these - they were not available before. In my experience it is extremely rare for manufacturers to (publically) document the required RDM information. I have also added the 'Freq.' modes to the Clay Paky HY B-EYE K25 and forwarded to Capture to add these too. All being well the revised Titan personalities should be available in the Titan Fixture Library installers from approx. 09:00 GMT 7/11/20. If you find any further problems with these or other personalities please submit a separate bug report or new personality request for each fixture via the personality website so that we can keep track and avoid confusion.
revenant
Posts: 51
Joined: 25 Oct 2020, 14:36

Re: Bug? Capture component, some fixture models have no orientation markers

Postby revenant » 07 Nov 2020, 16:50

I was at the polar opposite end as all the fixtures I had confidence in were coincidentally all tilting away from the camera.

I have not seen or noticed any fixture models with arrows on them so I could not comment on what the arrow may be denoting. There are many physical fixtures that have arrows on them, this usually points to what the manufacturer considers to be the front of the fixture, it seems unlikely but I guess it could be possible that Capture is using this or something similar on their models?

For me, all plot paperwork I have seen includes the power tail of the fixture and it is this that is used to determine orientation, from the plot view of Capture desktop the small lightning bolt icon associated with the fixture is the only indicator displayed that could be used to determine fixture orientation. Of the few fixture manufacturers that document the direction of pan/tilt movement, the images they use always include the location of the power tail, some also include the location the display but that is an additional and not very helpful when the display is on the yoke arm. I consider the reference to the power tail in orientation to be an unwritten industry standard, it does seem to be the best way to exchange orientation details (cue all the people with battery operated fixtures).

If I were to give you standard plot paperwork, I would think you would really struggle to place and accurately orientate the fixtures within Titan's Capture component when the 3D models have no element to denote the power tail, the Capture component does not include a plot view. I may take this up with Capture as although you can determine the orientation from the plot view of Capture desktop, most of the 3D models have included an element to denote the power tail so you could orientate the fixture just fine from within any of the 3D views. The need to now switch to plot view seems like a step backwards to me.

Regarding the HY B-EYE K15, I checked out the updated personality file and it works all good, the fixture appears in the RDM view of the patch workspace and it appears to match the personalty file's modes with the RDM modes. Note, the fixture does not support RDM over Art-Net, I was using an ethernet->DMX gateway that handles the RDM<->RDM over Art-Net conversion. There are some issues with the RDM view and features, I may post some feedback on that but it's not really a feature I make use of.

I see the file revision history of the HY B-EYE K15 file notes that there is nothing for the pixel modes, I'm not sure where best to post the info on that as it's not really a bug it's more an FYI, Although the pixel modes are documented in such a way as they may be considered separate modes, they are not modes in the standard sense. They are just the protocols for the two modes of the pixel engine part, but, the pixel engine itself cannot run standalone. Although the pixel engine can be independently addressed, the idea is that you can patch the basic engine (the fixture part) to a console, and then patch the pixel engine directly to a media server. The dimmer control of the basic engine is also the master/global dimmer for the pixel engine and the pixel engine is only active when the function channel of the basic engine is set to DMX 103~105 (Pixel map enabled), so the pixel engine depends on the basic engine in order to function. There are some "+ pixel engine" modes, for example, RDM personality 9, is essentially "Standard mode + RGBW Pixel engine enabled". Another important piece of detail that isn't documented anywhere is that when the function channel is set to "Pixel map enabled" then the RGBW channels (1 thru 8) of the basic engine are disabled, their value have zero affect on output, presumably the CTO and macro colour channels are also disabled, I have never actually confirmed this. Also note that the Pixel map enabled option is stateless. The latest revision of the protocol does add some more options to the function channel.

As all that seems as clear as mud, let me know if you have any questions or need any of the above clarified.

I have considered seeing if Capture can model the pixel engine enable/disable functionality as at present they don't and I think it could be helpful if it were more accurate in this regard.
User avatar
niclights
The eManual
Posts: 4442
Joined: 24 Sep 2004, 01:06
Location: UK

Re: Bug? Capture component, some fixture models have no orientation markers

Postby niclights » 08 Nov 2020, 11:31

Regarding orientation I have to say my experience of standards and plots is the same as yourself. I had always considered tails (or possibly display) as the standard way of communicating orientation and is how I have always explained when designing and, similarly, is my expectation when receiving plots. When discussing this following your first post in this thread I was told about the arrow and that this is what is used. This was news to me. I had seen arrows on real-world fixtures but never considered using them as a reference and I have never seen any plots that do. But apparently others do/have...

To clarify there have never been any arrows on Capture models representing this. 'Generic' models in Capture have a display and a power representation (either guessed or calibrated depending on information available). The problem that you have highlighted here is with detailed 3D models. My understanding (although I may be incorrect) is these are provided by the manufacturer and used 'as-is'. So if the manufacturer's model includes the display then it's there, otherwise not (as with the Clay Paky Sharpy).

Capture fixtures have a property that (quoting Capture's response to my query) "means that with the fixture in front of you and receiving 50% pan, it can rotate equal amounts clockwise and anti-clockwise and it tilts either away from you or towards you. Which one of these it is depends on the fixture manufacturer's choice of tilt direction. The same goes for in which direction the fixture pans when you increase the pan value". It is this property that is used to determine the default orientation in Titan and this agrees with what we see. I am trying to get confirmation as to whether we are certain this can be considered equivalent to the arrow on the fixture pointing towards you.

I can see the logic in not relying on display and perhaps there could be confusion using power if a fixture has both power in and out and they exist in different locations (or if they are only battery powered). However, I think the vast majority still do have a power in and display on the base. And I suspect more than have an arrow!

I completely agree that it is currently very difficult to orientate fixtures in Titan where the 3D models have no reference points. Even if we are confident the default (0,0,0) orientation = arrow pointing towards the camera it is not necessarily straight-forward how you would translate this to a display or power reference point if you only have this. And the situation is not much better in stand-alone Capture, although at least here you have a guarantee that 0,0,0 = display facing camera and the plot view if necessary.

I have suggested including a representation of the arrow in the (wireframe) model such that you could identify it from any orientation (ie. not just from above/below but also from the front/back/side). Personally I think there should also be a representation of the display and/or power too. This would provide all the tools required and avoid any confusion. I have also proposed that Titan uses the same orientation values as stand-alone so that, even if we have a different default orientation, setting to 0,0,0 would = display facing camera, maybe with a complementary 'locate orientation' to return to default. I think this would be helpful in fixtures with unusual calibration such as the Philips SL Beam 300 FX where in the real-world I think it is extremely unlikely you would ever rig it in the way it appears in Titan. Rather it will almost certainly be with a side of the base parallel to the rigging structure - if we have the same orientation values as stand-alone then this would just be a multiple of 90°.

One way or another I think this needs to be improved. Accurate orientation of fixtures is fundamental to visualisation and pre-programming.


Regarding the Clay Paky HY B-EYE K15 I understand the reason why the basic and pixel engines are separated. As they can be addressed separately they have to be provided as separate modes.

The intention is to provide mode options that match those in the fixture menu. I can't find any supporting documentation for this specific fixture but my understanding from other Clay Paky fixtures with similar implementations is that the menu effectively has a 'Basic' and 'Pixel' section, each with address and mode which is how we are currently supporting. If, in the fixture menu there is a specific option to select a combined mode where the channels are contiguous (such as RDM personality 9) then we should add equivalent modes. If these combinations only exist in RDM then this doesn't make much sense and I don't know how Clay Paky expect them to be supported. It would be helpful if Clay Paky could provide more comprehensive documentation detailing all of the modes.
revenant
Posts: 51
Joined: 25 Oct 2020, 14:36

Re: Bug? Capture component, some fixture models have no orientation markers

Postby revenant » 09 Nov 2020, 10:49

I wonder if the arrow is pointing to the side of the fixture that Capture has determined has the display? I think my memory is failing or things have changed recently, I don't create or edit Capture project files very often and it's been a while, I was sure that the default fixture orientation in stand-alone Capture was power tails pointing stage left, but looking at it today, I think it might be display facing downstage out to house. For 3D models that lack the power tail element, stand-alone Capture does provide the possibility of viewing the power tail location from within it's plot view, it's still an inconvenience to switch into plot view just to determine orientation but at least the possibility exists.

For fixtures that have both power in and out, you just need to detail the in I don't see any benefit to including the out as it just introduces a potential level of confusion as you note. As all the generic 3D models seem to have the power tail and display detail I have to conclude at some point in time someone at Capture must have considered this an important detail.

With regards the HY B-EYE K15, I think my referencing the old extended mode of previous fixtures was a poor choice and poorly worded also resulting in misleading detail. You are correct, the menu system/options are the same as previous Clay Paky fixtures. The "RGB" and "RGBW" modes just set/enable the modes for both basic and pixel engines but do not change the start address or source values for either engine. Option 9 from RDM personality list is really just "Standard mode with RGBW Pixel Engine" but does not assign the start address to be a contiguous block. I'm still not sure I'm explaining that very well, but I think you are correct in that personalities 1 thru 4 are the only ones that can really be supported directly at this time.

Who is online

Users browsing this forum: No registered users and 29 guests