Blackmagic Forum

Hi all,

I know this has been addressed to some extent in various posts but I have a specific question that I can't seem to find any info on.

My company recently bought a TVS pro hd and it works brilliantly for us. One of the types of content we broadcast is live presenters speaking with slides. Until this point I've been using the media player for a background, the USK with DVE to move the slides in position and one of the DSKs (cropped) for the camera on the presenter (the other DSK is used for lower third). This looks something like this:


It works fine, but makes it very difficult to frame the presenter, as I have to ask the cameraman to frame the subject like this (with very little room for error or they get cut off):


It also means while I can do DVE effects (fly in/out ect) on the slides, I can't do it for the presenter which for our content would be much better.

I would much prefer to use the USK for the camera so the cameraman can frame more normally and easily, and use the DSK for the slides. However this obviously would require moving the slides on screen which isn't possible with the DSK (only cropping).

I've seen some talk of using a multicam viewer to place a video in one of the 4 corners. While this could work, I'd love to know if there is any way (with hardware presumably) to move a source more specifically on a video stream as we'd prefer not to have the presenter completely flush with the corner for aesthetic reasons. Because the only way I can see at the moment is using an extra laptop with something like wirecast to move the vid and an i/o device to go back to the ATEM which would obviously be a bit overcomplicated...

Does anybody have any experience in this sort of thing? I love the device but the single USK is the biggest limiter for me at the moment given the requirement for great looking pic-in-pic in the field I work in (I'm really trying to squeeze every bit of functionality out of this thing ).

Many thanks for any advice or suggestions,

Mike

Sorry to bump this, but it got buried deep in the second page with only a few views by the time my account got sorted and the post was approved!!

If the slide show is PowerPoint you can specify a custom show size so it is not full screen. This results in a widow border around the presentation which you can then use a DSK mask to cut out. Presentation is on the DSK, Presenter is USK DVE and Program provides the full screen background image from Media Pool.
See https://ianmorrish.wordpress.com/2017/02/06/picture-in-picture-pnp-with-blackmagic-television-studio-and-powerpoint/ for PowerPoint configuration to achieve this.

Simply use a 2ME switcher instead. It has Super Source where you can composit up to 4 Sources in one picture with cropping, scaling and positioning. And you can even use a 5th source as Backdrop (f.e. the mentioned media player).

The TV Studio Pro is simply not powerful enough to do this.

Cheers

Robert

Thanks for the replies.

That's an interesting idea with Powerpoint, Ian. I wasn't aware of that feature so thanks for that. Only issue is that often I'm broadcasting a conference at which I'm not in control of the slide deck, I only have a video feed.

Robert, unfortunately we don't have the budget for that just yet so I'm trying to stretch the equipment we already have to its max.

All I really need is a device that allows me to resize and place a video feed in a certain area of the screen, then I can just mask out the black but short of another production switcher I can't seem to find any such device.

Thanks for the help anyway,

Mike

you might look into the Decimator DMON-QUAD. It's designed as a 4 input multiviewer, but supports any custom layout. I'm thinking you could use it as an external "super source" and bring it's output back into the ATEM. You could resize both your camera and powerpoint as you illustated, then switch between this, your camera full, and the powerpoint full. At $295 it is worth a look...

Thank you, I have been looking into this device as well and looks like you can greatly customise the multi view layouts to exactly what I need! Definitely going to pick one up.

Thanks for the help!

The Blackmagic web site shows an example of Pic in Pic.
I don't own this unit but may buy one if it will do this well.

PiP is definitely possible!


Skickat från min iPhone med Tapatalk

If you need this for livestreams, livestream studio can do this really easy. You need a computer with at least two blackmagic mini recorder (I use a macbook pro, it has 2 Thunderbolt).
If you also show your program local, there is some delay, so it´s not the best idea.

Then you could use one more TVS HD, if you don´t use the pro version it´s much cheaper.

Thanks for the input guys,

Wayne - a single source Picture in Picture as shown on the BM website is very easy to do, just choose the video you want overlaid as the upstream key source and resize it to where you want. You can save various positions for easy recall. However, with only one DVE on the device it makes using 2 sources on top of a background much more difficult - e.g. my original question with slides and a presenter in boxes on top of a branded background. You have to use one of the downstream keys, but these don't allow repositioning or resizing, only masking (cropping).

If you only want to use the device for PiP with a single source then the TVS will work perfectly for you. More than that and you need to think outside the box a little to get it to work as you want it. I think using a multiviewer is the best solution I've seen so far.

Markus - Thanks for the idea, I use Wirecast for streaming and had considered using a second Ultrastudio mini for the slides and placing them in Wirecast. However, I don't like the idea of having to add the slides back in post (I'm recording straight from the ATEM as the load on the laptop can be a little high when streaming and recording at the same time). As well as live switching, one of the main draws of getting the TVS was to effectively remove the need to edit from a project. It's a nice idea though and if I ever need more precise control over the positioning of layers this is probably the way I'll do it.

Thanks all,

Mike

Livestream Studio uses your GPU, so the load isn´t high anymore. My Macbook Pro from 2014 is working with a load of 35% at exactly that case.
But it costs about 800 Dollars of you don´t use livestream.com

To follow up on this, had a chance to experiment with a Decimator DMON-6S today, the results should be the same on the much cheaper DMON-QUAD.

I was able to take Mike's desired layout of a 2/3 size box and a 1/3 size box of 2 sources over a full screen 3rd source. The layouts are pretty easy to setup, and you can store several presets for recall as well.



The resulting output at 1080i59.94 was input as a source back into my ATEM, and shown here (poor cell phone shot, but you get the idea)



So it seems that for $300usd you can add a capable "Super Source" to the lower end ATEMs.

Good idea with the Decimator.

But have in mind that this will introduce more latency. You will have to compensate the amount of frames that the decimator needs to composit the multiview.

I think you have to add 2-3 frames to the delay of the switcher (1 frame) - so you will end up with 2-4 frames of latency. Way too much for imag. If you just output to a livestream this should be OK. But as soon as you switch back to a fullscreen source the audio will be out of sync.

Cheers

Robert

It might be better to only have the PowerPoint zoomed and panned into position using inline scaler (I'm using a TV One C2 scaler - normally expensive device but can be had 2nd hand on eBay real cheap if you are lucky) and displayed using DSK. Then camera can be displayed using USK with DVE.
This lets you move things around during the live presentation. Like this...

Using flying key for USK and conditional auto transition of DSK depending on size of camera PiP window (using PowerShell macro)

Robert Betzner wrote:But have in mind that this will introduce more latency. You will have to compensate the amount of frames that the decimator needs to composit the multiview.

I think you have to add 2-3 frames to the delay of the switcher (1 frame) - so you will end up with 2-4 frames of latency. Way too much for imag. If you just output to a livestream this should be OK. But as soon as you switch back to a fullscreen source the audio will be out of sync.

The Decimator has much lesser delay then you think.. True every processing introduces delay.. But if i may guess it will not be more then when you use the DVE on the mixer.. Which does also take 1 frame more delay when enabled..

I don't think that is true. Because the decimator does this similar to the ATEM. 1 frame for framesync and a second frame for compositing. Then you get a third frame of delay by re-entering the composited signal into the ATEM. You will end up with 3 frames delay if the decimator is really fast.

And the bad thing about this is that the camera that goes into the ATEM and the Decimator will have a difference of two frames to the composited decimator picture. So it will be really hard to keep audio in sync as all sources have different latency.


Cheers

Robert

wow, the things that are stressed about out there in the inter webs... here's a $300 tweak that's pretty cool to the low end crowd (well, all of us without a 2ME anyway) and all of this dissension about whether or not it is workable.

well...to set the record straight I offer this image



Using a timecode burn from an AJA KiPro as a source, it feeds into a DA. One DA output feeds the left monitor in the picture directly. Another DA out feeds input 1 of my ATEM. A 3rd DA output feeds the DMON-6S, and loops thru 3 inputs to the multi viewer. The DMON out is connected to input 2 of my ATEM. The right monitor in this picture is PGM out from the ATEM with a split screen wipe, the bottom half is input 1 (KiPro full screen), the upper half of the wipe is input 2 (DMON multi view feed).

As you can see, there is a net difference of.....ONE FRAME (yes, the time code in the little box reads 17:51:44:11)....between the kipro output directly and going through the DMON *AND* ATEM together.

Honestly, if you are IMAGing in a big room, the through the air delay of the PA speakers between the front row of seats and the back of the room will offer more lip sync delay issue than this solution...

Wow! some interesting solutions.

I would try reformatting the powerpoint presentation then add the presenter with pic in pic.
So Powerpoint looks like this.
Then presenter is added from Camera in the top right
Lip sync should come through without any delay.

MambaFiber.com wrote:As you can see, there is a net difference of.....ONE FRAME (yes, the time code in the little box reads 17:51:44:11)....between the kipro output directly and going through the DMON *AND* ATEM together.

Thank you Greg of proofing my point about the Decimator MV Still it is less then i even thoughed.. So yes that is pretty cool!

MambaFiber.com wrote:Honestly, if you are IMAGing in a big room, the through the air delay of the PA speakers between the front row of seats and the back of the room will offer more lip sync delay issue than this solution...

Very true!!

Love to see this all in action - Been wanting to pick the DMON QUAD up for a while and try this, but now that I know it works (with a few catches of course ) i'll have to get one!

Nice work

Wayne Rutherford wrote:Wow! some interesting solutions.

I would try reformatting the powerpoint presentation then add the presenter with pic in pic.
So Powerpoint looks like this.
Then presenter is added from Camera in the top right
Lip sync should come through without any delay.

This is the winner.

Though the Decimator box is cool and apparently fast it just creates more problems. The TVS doesn't have enough aux/PGM outputs to make it a viable solution, one way or another you're going to create feedback/re-entry issues. At the bare minimum with the decimator you'd need to use the aux out for one decimator input and another switch but not PGM for the other.

An expensive solution, but a very neat little box is the AJA ROI-SDI. I haven´t tested the delay yet, but it has an SDI in, SDI/HDMI out and an SDI-loopout. So for your needs (and mine! we do this type of presentation a lot) you can reframe a camera-input or the input of the presentation easily. No presets though, so you need to connect it to a computer via USB to configure it manually.

Our client have the same requirement that being discussed in this thread. My best alternative without 2ME switcher (yeah i only have budget for TVS HD at the moment) is having two sets of laptop with different powerpoint layout.

One is something like the one posted by Wayne Rutherford (few post above me)
And another one is the 'full version powerpoint' without reduction for safe area for PiP, so that in case of the slides need to be more clearly shown, i could just switch to full version.

The cons are it needs 2 laptop and if you are a one man show, can be quite demanding

the DMON 6S has loop outs, so you input the cameras in there, and loop them straight into the bmd mixer. then take the output of the DMON 6S as an input of the bmd mixer. problem solved

Koko Pudja wrote:Our client have the same requirement that being discussed in this thread. My best alternative without 2ME switcher (yeah i only have budget for TVS HD at the moment) is having two sets of laptop with different powerpoint layout.

One is something like the one posted by Wayne Rutherford (few post above me)
And another one is the 'full version powerpoint' without reduction for safe area for PiP, so that in case of the slides need to be more clearly shown, i could just switch to full version.

The cons are it needs 2 laptop and if you are a one man show, can be quite demanding

Why not just build the PP the way it needs to be shown instead of running 2 machines? Some graphics will be full screen others will be pipped but the pip will be in PP not in the TVS. This way you can build in any transitions you like without eating up any of the limited assets of the TVS. Alternately there are controllers on the market that slave the 2 machines so that they always follow, you can have 1 machine running full screen and another with the PP pip.

Hello Sir
PiP Atem Production Studio 4k Not Me
Plz Add PiP Function Basic All Brand mixer
Data Video, Vimix ,Live Stream,Wirecast,For A,Panasonic,Sony PIP Function Available
Plz requsted All Friends and Brothers Blackmagic Family
Users I waiting PIP Fuction Software Next Update 7.4 SDK

PIP is not possible in all mixers because of hardware limitations.

IF you really need PIP you might need to sell your Atem Production Studio and buy an Atem TVS HD or 1ME.