[Yves De muyter]

You might want to wait a few days. BMD have acknowledged the problem, a fix has been promised. 🙂

-Yves

[Yves De muyter]

https://www.virtualvtr.com/

-Yves

[Yves De muyter]

https://www.blackmagic-design.com/support/detail.asp?techID=76

It’s always a good idea to talk to an integrator, even some apple dealers supply macs using decklink and all the apple goodies. If anything like this happens, it’s their fault and have to supply another mac.

-Yves

[Yves De muyter]

[Russell Lasson] “Here is our situation:

So my questions are:

1. What’s the difference between how the Decklink Extreme cards transcode video from an SDI source and how Compressor 2 transcodes video, given that the input codec and output codec are the same?
If the source is DV50, there will not be a big difference IMHO, better yet, going from one codec to another is often a bad idea. SDI is component-video uncompressed and decklink will compress it a little bit again.

2. Typically the Blackmagic Codecs use 720×486 frame size, but my DV50 material is 720×480. Is there anything wrong or dangerous about using the Blackmagic 10bit codec at 720×480?

I would’nt do that. I’d try to convert them to 486 height. I don’t know in this situation since we’re using PAL here. I guess your NLE should detect this and force
a render on the clips, maybe even resizing it, and that is known to be slow. I’d try to add 4 lines black while transcoding. Depends on your transcoding tool if it can do that.

This is really going to be a big issue when people really start using P2 for aquisition because there is no tape to go back to for the online.”

Yes. But often the idea is to put those clips on some central terrabyte storage as rushes and keep them there until the project has finished. Diskspace is cheap these days.

-Yves

[Yves De muyter]

There is no DeckLink board/device that has s-video input. Only models with composite input/outputs. Also, make sure your composite signal is stable enough, the decklink is picky about its composite input. Professional stations use time base correctors and frame synchronizers to bring in consumer material.

-Yves

[Yves De muyter]

[Mike_S] I don’t have smooth playback of DV files via Decklink – i.e. with fields in correct order for non-DV PAL – without adjusting settings for each file individually..

You can adjust the settings for the file. And what if you apply an effect, like “blinds” or anything that is heavily fields-dependant ? Here the result will be lowerfields again, and don’t come out of the decklink correctly.

– Yves

[Yves De muyter]

I have and I think they added the issue as a bugreport (since they asked a sample clip + system specs and never replied back). I don’t know when they’ll have a fix though, and they say they don’t see the issue on their machines. Not very promising imho…
BTW, i tested it on 3 machines now (HP8000, HP8200, SuperMicro X5DA8), all of them suffer from this problem.

I think the PAL users are the only ones who have this problem.

-Yves

[Yves De muyter]

You could try to run an older version of the drivers?
I can’t imagine BMD would not provide a fix for this…

-Yves

[Yves De muyter]

Paul,

This is to adjust the timing of the decklink output to compensate when there is a small difference. Not used in standalone or small installations, but when you have a large installation with long cable lengths, your reference can sometimes be slightly off, moving your video out of its window (or other devices that are sligtly off and don’t want to take the decklink as a source etc…).

Most systems do it via the registry, BMD has controls for it. 🙂

-Yves

[Yves De muyter]

Dear BMD,

> Also, you can play out the PAL DV sequence to tape with DeckLink, and then
> recapture to uncompressed, and it will be OK. We set the fields correctly
> when playing a PAL DV clip out to the SDI output.

This doesn’t seem to apply on our customer’s system. Any playout of DV in a DV-project seems to be upperfield, and thus, has fields swapped.
Playout includes export-to-tape and just previewing the playlist.

I have 2 solutions I proposed to the customer:
1. is to create a new sequence, import the main sequence into that sequence and move the video 1 pixel up and render it out.
2. is to create an upperfield DV project and reverse field dominance of all captured clips. This is far the best solution, the only problem with this is that exported DV tends to be upperfield and that might become a nightmare exchanging media with other systems.

-Yves