matkeane

It would be nice if we could rename media cues in the timeline. On a recent project, each scene ended up with a bunch of copies of 'black square blur 50px.jpg'. It would be great if each instance of a media cue could be renamed with something more informative like 'Mask top', 'Mask left', etc. I know we can rename layers, but there always seems to come a time when I have to quickly move cues to a different layer to make room for a new clip, and then the naming scheme gets messed up. Likewise, I guess I could add and rename a 'Play' cue, but that means moving 2 cues around or else the label will get separated from its media cue. And since hovering over a cue reveals the media name, and the 'locate media' button takes me to the source, it seems like renaming media cues wouldn't break any workflows.

Hi, I'm in the process of setting up a new production laptop, and I'm encountering errors when trying to import & playback HAP video files in Watchout. When I tried to add HAP media to the project, I got an error message: 'Error: Not recognized as a media file'. When I opened an older project, which already has HAP media imported, playback results in a 'Quicktime error -2095'. I've tried a 'repair' installation, as well as a complete removal and re-installation of Watchout 6.6.4. I also tried installing Quicktime, but that just results in a 'Quicktime Error -8961' error (as mentioned here: https://forum.dataton.com/topic/6365-hap-q-from-avf-batch-converter-quicktime-error-8961/) when attempting playback, so I uninstalled Quicktime and am back to a barebones Watchout installation. This is a new installation of Watchout 6.6.4 at the root of the C drive. Other video files with h264 and Mpeg2 codecs are importing and playing back fine in Watchout. I started to have doubts about the Hap encoding, so I encoded some test files in HAP, HAP-Q, and HAP-Alpha flavours, using Adobe Media Encoder and AVF batch converter. All of the HAP files play with no errors in VLC, Disguise and Notch, but I still get an error when I attempt to import them into Watchout. I wondered whether there is a possible conflict with the other software (Disguise, Notch) but they are cohabiting fine with Watchout on another Win10 machine. That machine also has several versions of Watchout installed (via the zipped installation folder method) and all seem to be working OK. As an aside, on the other laptop, I notice that HAP files encoded with Adobe Media Encoder show generic file icons in the Media window (but playback is fine), while those encoded with AVF batch converter show as thumbnails. The laptop is a new MSI machine running Windows 10 Pro (20H2 Build 19042.789) with a Nvidia RTX2070 Super card (driver DCH 461.40). I've deactivated many of the Windows services but, since this is a production machine, I haven't applied everything in the display machine tweaking list. I'm pretty sure there are no dodgy codec packs installed. So it looks like it's something related to this new machine - if anybody has any ideas or pointers to the problem, I'm all ears...

Perhaps I'm missing something but, on the new site, there doesn't appear to be a way to view other users' suggestions so, unlike the epic 'feature request' thread on this forum, there's no way to chime in with a '+1 great idea', or 'no, please don't change things because...' comment. And, presumably, no feedback from Dataton on whether an idea is in the works, or a 'won't fix'. In this forum, suggestions, problems and their solutions can be shared - Personally I think it would be a shame if all feature & support requests disappeared from view.

I agree that constantly expanding/shrinking layers to access tween properties is a pain. I've long thought that it would be nice if the timeline panel could toggle between cue layout and a curve editor (similar to After Effects), so that we can say a long-overdue goodbye to 'smooth' keyframes. Perhaps being able to toggle the panel between multiple uses - cue layout, curve editor, and a property inspector - would make life easier still (especially if keyboard shortcuts were available). I'm not entirely sure I like the Disguise approach of opening infinite menus until you run out of screen space, but it's better than concertina-ing the timeline all day.

Hi Quentin, I had been looking at using the RTTrPM protocol as a way of getting data into Watchout at relatively high frequency - more for general interactive purposes than object tracking - and so I was thinking about building an app to translate data from various sources to RTTrPM. When I read about the FreeD support in the Panasonic I started thinking about how that might be used in Watchout - although camera tracking data might not be as immediately useful as object tracking, I guess. So the aim was to build a tool that could translate, or simulate, data from various sources/formats to RTTrPM. And then lockdown ended, things got busier again and I haven't had time to make any more progress on it! But if you have any tips on integrating RTTrPM & FreeD, I would be grateful for any advice. By the way, were you present at a Panasonic Road show / BlackTrax demo event in Saint Denis a couple of years ago? Thanks, Matthew

I think Jhannah is referring to this page: https://www.liminalet.com/zoomosc which shows 'Zoom OSC' software sending/receiving OSC commands to/from media software like Qlab and Isadora. The main problem there is that Watchout doesn't (currently) work with OSC so you would need a little bridge app to deal with communication if you need synchronised switching of sources and graphics, for example. However you could add your Zoom meeting as an NDI source to Watchout, add PIPs and graphic overlays as required in Watchout and then - as others in this thread have suggested - capture the output from Watchout as an NDI source in OBS for streaming. Native NDI output from Watchout has already been posted on the feature request thread but would make this kind of setup much easier - and built-in NDI would (hopefully) mean support for an alpha channel, so that one could send graphic overlays to external software without dealing with key & fill or chroma keying. Using Watchout for streaming projects might seem like overkill but I think it offers some advantages: Final output may just be a single video stream, but a virtual studio might have several surfaces on which content can be displayed: animated backdrops, screen inserts, graphics overlays. Being able to manage content for each destination with separate Tasks & Displays in Watchout seems like a good way to manage that. For a simple project, cueing a video file from PlaybackPro or Qlab might be enough, but being able to use the compositing features of Watchout to overlay different backgrounds, PIP frames and other video content can make life much easier once things get more complex. Apart from anything else, being able to import and cue content in a different system from the streaming software (where the operator is already managing live camera switching and remote connections) seems like a good idea!

My guess is that Jhannah was asking whether it's possible to use Watchout as a playback system and inject the output into OBS (or other video systems) for streaming. My other guess is that a lot of media-server operators are being asked the same question at the moment and, unfortunately, the answer is that it's not as easy as it could be. With no support for NDI output from Watchout, your options are either piping Watchout output into a capture card on your streaming server (and eating the 2-3 frame latency you will no doubt get) or using a device like a Birddog to convert a HDMI output to NDI (and maybe only dealing with 1-2 frames of latency). Which, incidentally, is one reason why I see lots of Watchout servers gathering dust in warehouses at the moment while vMix and OBS systems are flying off the shelves.

Hi Michael, Not yet, but when I saw that the Panasonic 150s support the FreeD protocol, I did start wondering about whether it would be possible to build an application that could convert FreeD data to Blacktrax RTTrPM data which could then be piped into Watchout. The RTTrPM format is well documented here: https://github.com/RTTrP/RTTrP-v1.8.7.0 . The FreeD protocol is explained here: http://docs.vizrt.com/tracking-hub-guide/1.0/description_of_the_freed_protocol.html . But I haven't yet had the chance to get my hands on any FreeD compatible equipment to try capturing a data stream. An alternative might be to use something like TouchDesigner (somebody has something working here: https://forum.derivative.ca/t/udp-free-d-protocol/8655/ or there is an experimental CHOP here: https://docs.derivative.ca/Experimental:FreeD_CHOP) to convert the FreeD data to raw numbers and adjust Generic Inputs with the setInput command but, although performance was improved in v6, I don't think setInput was designed with high-frequency updates in mind. A Blacktrax input would probably be more robust.

Another way to manage the load on a laptop producer, while still using video previews, is to use Stage Tiers. If you assign parts of your show to separate Tiers, you can choose which to preview on the production machine. Alternatively you can encode low-res/low data-rate preview files and place them on a 'preview' Tier, only rendered on the production machine, so that you have a visible reference (even a BITC in case of corrections requiring a re-render) for timing, but without overloading the machine.

Salut Benoit, Ah, you're right! For some reason I thought the Datapath accepted both analogue and digital signals, but I guess I got mixed up between their older VisionAV cards and the VisionSC, which does appear to be digital only. That's a shame, but not a big problem really... Merci et à+ Matthew

Apologies for what is not strictly a Watchout question, but I wondered if anybody has managed to successfully capture a composite PAL source with a Datapath VisionSC-SDI4 card. The Datapath card is installed in a Watchout player (purchased from a Dataton distributor and otherwise working perfectly) currently running v6.2.2. However, I can't get the Datapath to recognise a PAL source on any of the 4 inputs in the Datapath Vision software - so I'm not even getting as far as configuring the video inputs in Watchout. I did wonder whether the Datapath still supports SD formats, but the data sheet for the VisionSC-SDI4 seems to show support for both PAL & NTSC signals (576i25, 480i29.97) https://www.datapath.co.uk/datapath-products/video-capture-cards/visionsc-range/visionsc-sdi4 The camera is a HikVision DS-2CC12D9T which outputs either PAL or NTSC on the composite output. https://www.hikvision.com/uploadfile/image/9900_E资料模拟相机NewtemplateDatasheetD5T(18)DS2CC12D9T(A).pdf The reason I don't use a HD signal is that the camera is actually used with a legacy motion capture system which is configured for a 4:3 PAL source. Since the camera hangs alongside the main video projector for the show and needs to frame the stage, it would be handy to be able to see what I'm doing while adjusting the camera. Currently I have to use a laptop to VNC into the machine in FOH that's connected to the camera, but I was hoping I could make life a little easier by using the Datapath card to display the live input through the video projector. When the camera is connected - with the same video cable - to the machine running the motion capture software, which uses a Blackmagic capture card, the signal is recognised as a PAL source. Usually I find Datapath cards are very good at auto-detecting the input format, while Blackmagic cards are more picky. Unfortunately I don't have access to the machine right now, and it will be a while before I can try this again, but I was just wondering if I'm missing something obvious... Thanks, and I hope everybody is staying well out there.

Hi, If you search for 'countdown' on this forum you will find various ways of doing this, as well as a few feature requests for an automatic way of doing this. One easy way is to add a video with a countdown timer alongside your video media. The downside is that this adds some extra load to your machines and, if you switch your production machine to 'Video as thumbnails' you won't see the timer. Another way might be to use Watchnet to show the status of your timelines. I haven't actually done this, but I believe you can get the status of your timelines and show the remaining time until the next cue. The downside is that it requires another machine and Watchout licence. The way I usually do this is to create a composition with a simple clock animation made from image layers. That way there are no extra video files to be played back and it's still visible when displaying 'Video as thumbnails'. I drag it into my tasks and slide it along until it snaps to the end of my video layer, and then place it somewhere on the Stage where I can see it during playback. I often add it to a preview monitor so that colleagues can see it without me having to call it out. The downside is that it has to be placed manually and I haven't got around to making a version longer than 1 minute, but I figure if you have more than a minute to go, there's nothing to worry about yet... There's an example Watchout project file with the media for the clock on my github here: https://github.com/matkeane/watchout-snippets/tree/master/countdown-clock

I never got around to testing it, but I downloaded this spout-nidi utility a while back (Doesn't look like the site got their certificate sorted for https access, but the links still work if you edit to use http). http://spout.zeal.co/download-spout-to-ndi/

I haven't tried nested free-running/looping. My impression, from comments on this forum by Mike Fahl and others, was that free-running and looping should only be done at one level, otherwise things get unpredictable. If I want to loop a video in a composition, I don't set free-running and looping on the cue within the composition, but only on the composition itself and that has worked reliably so far (haven't tested in v6.6 yet).

Hi Tim, Nice - thanks for posting this. It's working for me on MacOS 10.13.6 using a non-commercial licence for TouchDesigner099.2019.20140. The icon images in the transport and cue-list windows seem to be missing, but other than that, it's working as described.

I usually setup a static backdrop with logos or something so that it's hopefully not as noticeable when I switch to Standby. However, on a recent job using Watchout 6.5.0 and a 6 output rental machine (W9100?), anytime I did an update (CTRL-D) - whether in Online, Live or Standby mode - each output would sequentially switch to black and then back to picture output... Needless to say that didn't go un-noticed. Standby content was in a composition on the main timeline, but the outputs were fed via Virtual Displays. There was too much else going on that day to try and identify a workaround, so I don't know if this was related to the Watchout version, the project setup, or the hardware.

In the past, whenever I've had to deal with very large format Powerpoints, it's been split across multiple PCs. You just have to make sure you have _exactly_ the same number of steps on each slide (going forwards and backwards) and then use something like a PerfectCue to control both machines with the remote control (or a multi-cue, if you need 3 or 4 machines in sync). Managing changes across 2 presentations can be a pain, but it seems more likely that you will manage to get a UHD output from a powerpoint PC working than an 8K signal. Bear in mind that you probably won't get perfect sync between machines though - even on slide transitions with supposedly identical machines - so you probably don't want to try playing video. Although if you need 8K video playback, then you have the Watchout so...

The question is not entirely clear, but I wonder whether the OP was asking about anamorphic lenses - like the concave mirror adapters used to shoot 360 degree video - or motorised mirror heads which allow a circular image to be projected in different directions... I've never seen the concave lenses used for projection, but there was a discussion on another forum about dome projections using a single projector and a convex hemispheric mirror - although apparently it's both fiddly to align and expensive. An example of motorised mirror heads can be found here: https://www.dynamicprojection.com/mirror-head-en/ - and I believe I saw a demo of something similar from Panasonic.

Recent forum posts about slow network transfers when dealing with large media files reminded me of this... The ability to constrain which media gets pushed to which displays in Live Update mode would be useful in some cases. On the last couple of projects, I had two distinct sets of displays, with media created specifically for each area - content for the front screen will never be displayed on the back, and vice-versa. When I activate Live Update, so that I can tweak some effects in real-time, all media is pushed to all servers - resulting in a long wait while several hundred Gb of files is transferred to players which will never use that content. The only workaround seems to be to temporarily deactivate displays, but then the client only gets to see half the content update live, so it’s not really a solution in most cases. Other media server software seems to manage this problem with named servers and string matching with media filenames so that, for example, a media file named ‘my_file_front.mov’ won’t be transferred to a server named ‘back’. It seems a bit of a clunky solution, but it does work. I wonder whether, in Watchout, it would be possible to exclude certain folders in the media window from selected stage Tiers… That way I could put all media files destined for the front screens in one folder, right click to bring up options for that folder, and select only the ‘front’ Tier to ensure media was never pushed to displays on other stage tiers.

The 'Format' shows as Mpeg-4, which is a bit surprising to me. I've only ever used HAP encoded Quicktime movies with a .mov extension. Is this Hap video in an MPEG-4 container, or did it accidentally get renamed as an .mp4? Could that be what's confusing Watchout?

Are you looping by jumping back to the start of the aux timeline, or by placing the audio tracks in a composition and setting free-run/looping on that in an aux timeline? I have found that the composition method usually works better, but I haven't actually tried with a seamless audio loop.

Hi Mike, The problem I have encountered with compositions is that (sometimes) when media cues within a composition use a blend mode other than normal, the cues 'pop' on and off instead of fading smoothly, once I apply an opacity tween to the whole composition in a task. I initially ran into this with a fairly complex wait loop containing free-running loops and various blend modes. Nesting the whole thing in a composition gave hiccups with opacity tweens and looping, but placed in a task everything ran smoothly, but then it was difficult to fade everything out mid-sequence if necessary.

I would find it really useful to have a master opacity and volume control for all cues within an auxiliary timeline, so that I can say something like 'Task 11 - fade out nicely over 2 seconds and then stop!'. Currently I'm doing this by creating generic inputs for my_task_opacity and my_task_volume, but then I need to add them to every media cue in each task timeline, which can be a slow process. For some simpler tasks, like live inputs, I nest a composition in a task with a fade-in/pause/fade-out which achieves the same thing but, having to create compositions for every task is also time consuming. If I could do something like 'setInput my_task_01.opacity 0 1000', I'd find that really useful.

To piggyback on Cowboyclint's suggestion, it would be nice if it were possible to change the timecode display format for a timeline (or for all timelines based on the Project frame rate) and let Watchout deal with the conversion to milliseconds. Content creators (in my part of the world at least) are more familiar with working at 25fps, and doing timecode calculations in my head while also multiplying by 40ms per frame seems like something a computer would probably do faster and more accurately! Also, the ability to enter relative timecodes would be great - e.g. hit ctrl+J and then type +12.20 to jump forward 12 seconds and 20 frames from the current timeline position.

I'm currently working on a Watchout installation at the Paris airshow. In addition to the production and player machines, I have a Watchnet 1.4 server running and a SurfacePro that the client can use to cue various clips on demand. So far the Watchnet setup is working well, but I have a few questions... Occasionally, before or after the show opens, I need to make changes in the control booth and so I put the show in Standby (triggered by Production, not Watchnet). Is there a way to show the standby status in a Watchnet panel, so that the user outside on the floor can understand why the buttons are unresponsive? I wondered about creating a 'standby' panel and forcing the Watchnet UI to navigate to a holding page, but that would require somehow triggering a 'navigate' command on the remote UI. I can just switch the screen off, of course, but a blank screen tends to worry the client, plus it means a lot of walking back and forth! Is it possible to trigger Watchnet commands from Production? Now I have all my scripts set up in the Watchnet server, It would sometimes be handy to be able to trigger them when I need to launch a specific task from production, without recreating the same events in my Tasks. Is there a way to temporarily disable buttons via a script? Once certain tasks are running, I'd like to disable the other buttons until the current task is finished to avoid lots of videos being launched at the same time. And finally, just a detail, is there a way to add a newline to button text? I was trying to add the clip duration under the title, but my attempts at adding '\n' and '<br/>' didn't get me anywhere, and extra spaces seem to get stripped out. Thanks for any help and suggestions!