The URSA broadcast seems still suffer from noise

Questions about ATEM Switchers, Camera Converter and everything live!
  • Author
  • Message
Offline

Wilson Luniz

  • Posts: 71
  • Joined: Thu May 22, 2014 4:19 pm

The URSA broadcast seems still suffer from noise

PostTue Feb 06, 2018 6:03 am

I just take a look from the official live video on Youtube:

As Grant said this video was shooted by the new URSA broadcast. And the video looks noisy. I don't know the lighting setup in the room and I know Youtube compression reduce the quality. But I never seen URSA mini 4.6k perform as bad as the video above. Take a look of the screenshot below.
Attachments
ursa_broadcast3.png
1:1 crop with 400%(2x width, 2x height) enlarge
ursa_broadcast3.png (721.1 KiB) Viewed 2136 times
ursa_broadcast2.png
1:1 crop with 400%(2x width, 2x height) enlarge
ursa_broadcast2.png (175.92 KiB) Viewed 2136 times
ursa_broadcast1.png
1:1 crop with 400%(2x width, 2x height) enlarge
ursa_broadcast1.png (932.32 KiB) Viewed 2136 times
Wilson Luniz
Generator Stage Production, Macau
Offline

Wilson Luniz

  • Posts: 71
  • Joined: Thu May 22, 2014 4:19 pm

Re: The URSA broadcast seems still suffer from noise

PostTue Feb 06, 2018 6:04 am

One more screenshot...
Attachments
ursa_broadcast4.png
1:1 crop with 400%(2x width, 2x height) enlarge
ursa_broadcast4.png (613.33 KiB) Viewed 2091 times
Wilson Luniz
Generator Stage Production, Macau
Offline

Asgeir Hustad

  • Posts: 105
  • Joined: Tue Sep 26, 2017 10:13 am

Re: The URSA broadcast seems still suffer from noise

PostTue Feb 06, 2018 8:18 am

This could equally well just be compression artifacts from the live webcast, in my eyes. Guess we'll need to get someones hands on it to post a proper review with example recordings.
Offline
User avatar

Tom_Bassford

  • Posts: 1606
  • Joined: Wed Aug 22, 2012 8:12 am
  • Location: Europe / UK

Re: The URSA broadcast seems still suffer from noise

PostTue Feb 06, 2018 3:20 pm

yeah looking at anything post web compression is pointless.

i'd like to see some uncompressed footage from the cameras SDI output. I'm interested in seeing how low light, high light and CA artifacting from b4 lenses is managed.
http://www.atemuser.com
if it was easy it wouldn't be called engineering
Offline

Denny Smith

  • Posts: 9106
  • Joined: Thu Aug 01, 2013 4:19 pm
  • Location: USA, Northern Calif.

Re: The URSA broadcast seems still suffer from noise

PostTue Feb 06, 2018 5:07 pm

Also, the Broadcast should have a fairly clean output, depending on exposure and lighting. But a UM 4.6 it will not be.
Cheers
Denny Smith
SHA Productions
Offline
User avatar

Tarek Saneh

  • Posts: 146
  • Joined: Fri Jun 27, 2014 8:28 pm
  • Location: DUBAI

Re: The URSA broadcast seems still suffer from noise

PostWed Feb 07, 2018 6:56 am

Dont give great hopes on low light capabilities of this camera
Tarek Saneh
Creative & Technical Director
http://www.wonderweb.ae
Dubai - UAE
Offline

Wilson Luniz

  • Posts: 71
  • Joined: Thu May 22, 2014 4:19 pm

Re: The URSA broadcast seems still suffer from noise

PostWed Feb 07, 2018 7:06 am

Asgeir Hustad wrote:This could equally well just be compression artifacts from the live webcast, in my eyes. Guess we'll need to get someones hands on it to post a proper review with example recordings.


Tom_Bassford wrote:yeah looking at anything post web compression is pointless.

i'd like to see some uncompressed footage from the cameras SDI output. I'm interested in seeing how low light, high light and CA artifacting from b4 lenses is managed.

Yeah, but look at the video in 1:1(2560x1440) instead of screenshoot, I think you will start to worry.
Wilson Luniz
Generator Stage Production, Macau
Offline

Wilson Luniz

  • Posts: 71
  • Joined: Thu May 22, 2014 4:19 pm

Re: The URSA broadcast seems still suffer from noise

PostWed Feb 07, 2018 7:12 am

Denny Smith wrote:Also, the Broadcast should have a fairly clean output, depending on exposure and lighting. But a UM 4.6 it will not be.
Cheers

That's the point! Broadcast may need some heavy post-processing in DSP. It's different market orientation compare to URSA mini. As live may not be able to de-noise.
Wilson Luniz
Generator Stage Production, Macau
Offline
User avatar

Tarek Saneh

  • Posts: 146
  • Joined: Fri Jun 27, 2014 8:28 pm
  • Location: DUBAI

Re: The URSA broadcast seems still suffer from noise

PostWed Feb 07, 2018 1:50 pm

It might be the same sensor as the studio 4k
Tarek Saneh
Creative & Technical Director
http://www.wonderweb.ae
Dubai - UAE
Offline

Wilson Luniz

  • Posts: 71
  • Joined: Thu May 22, 2014 4:19 pm

Re: The URSA broadcast seems still suffer from noise

PostMon Feb 12, 2018 5:35 am

No, they have different dynamic range.
Wilson Luniz
Generator Stage Production, Macau
Offline

Pavel Lavrov

  • Posts: 642
  • Joined: Wed Dec 19, 2012 5:24 am

Re: The URSA broadcast seems still suffer from noise

PostMon Feb 12, 2018 5:47 am

Tarek Saneh wrote:It might be the same sensor as the studio 4k


There’s good chance they use the same noise reduction capabilities as Micro Studio 4k...


Sent from my iPhone using Tapatalk Pro
Pavel L
Offline

Robert Betzner

  • Posts: 316
  • Joined: Fri Sep 28, 2012 7:54 am
  • Location: Cologne, Germany

Re: The URSA broadcast seems still suffer from noise

PostTue Feb 13, 2018 5:04 pm

Wilson Luniz wrote:No, they have different dynamic range.


Even then it seems to be the same sensor. It is possible to get a different dynamic range due to different read out modes of the sensor or better cooling.

My guess: It is the same sensor.

Maybe BMD could clarify this?


Cheers

Robert
Offline

Denny Smith

  • Posts: 9106
  • Joined: Thu Aug 01, 2013 4:19 pm
  • Location: USA, Northern Calif.

Re: The URSA broadcast seems still suffer from noise

PostTue Feb 13, 2018 7:02 pm

They have, it may the same sensor, but is being pushed to higher limits with a bigger video processor, and better cooling, enabling faster readout times, from a post BM made last week, but I can’t find it now. So I could have misread it? Will keep looking for the reference and post it here.
Cheers
Last edited by Denny Smith on Wed Feb 14, 2018 12:12 am, edited 2 times in total.
Denny Smith
SHA Productions
Offline

Howard Roll

  • Posts: 304
  • Joined: Fri Jun 03, 2016 7:50 am

Re: The URSA broadcast seems still suffer from noise

PostTue Feb 13, 2018 7:27 pm

Denny Smith wrote:They have, it is the same sensor, but is being pushed to higher limits with a bigger video processor, and better cooling, enabling faster readout times.
Cheers


Where did they confirm it's the same sensor?

The improved DR could simply be a matter of the new gammas. The current 4K studios are Rec709 only. I don't even know how you'd get 11 stops of DR into a 709 pipe without an aggressive knee or black stretch which they don't have.
Offline
User avatar

Tom_Bassford

  • Posts: 1606
  • Joined: Wed Aug 22, 2012 8:12 am
  • Location: Europe / UK

Re: The URSA broadcast seems still suffer from noise

PostTue Feb 13, 2018 9:17 pm

Why do people speculate about what sensor is or isn't in cameras they have never personally seen?

Seriously who cares what sensor it is, without the rest of the chain in place it's as meaningless as discussing the gearbox in a car.
http://www.atemuser.com
if it was easy it wouldn't be called engineering
Offline

Denny Smith

  • Posts: 9106
  • Joined: Thu Aug 01, 2013 4:19 pm
  • Location: USA, Northern Calif.

Re: The URSA broadcast seems still suffer from noise

PostTue Feb 13, 2018 11:59 pm

Howard, I think I read this for a Post Tim (BM) made. Edited my original post.
OK, found it, my bad, I was reading between the lines :roll:, and Howard you asked Tim if it was the same sensor, no reply. Here is the reference:
Tim Schumann wrote:On the URSA Broadcast -6dB is 200ISO, 0dB is 400ISO, 6dB is 800ISO and 12dB is 1600ISO.

That is the current range. We are driving the sensor differently to how it has been done previously, more similarly to a traditional broadcast camera, and are happy with the results we have been getting out of it at each of these settings.
You will need to update your ATEM switcher and software to 7.3 and the camera to Camera 5.0 Update so that these line up correctly between camera and switcher.


So his comment about driving the sensor differenty, implied they are using the same or similar sensor previously used, is where I got the idea it was the same BM Studio 4K sensor, again I am reading between the lines here. Sorry for the confusion.
Cheers
Denny Smith
SHA Productions
Offline

Howard Roll

  • Posts: 304
  • Joined: Fri Jun 03, 2016 7:50 am

Re: The URSA broadcast seems still suffer from noise

PostWed Feb 14, 2018 12:30 am

Tom_Bassford wrote:Why do people speculate about what sensor is or isn't in cameras they have never personally seen?


Didn't you kind of answer your own question, if I had used it I wouldn't be speculating? I don't know what the camera can or can't do, indeed. But I have been using the Micro 4K for well over a year and I know what it can or can't do. If the sensor is the same I can draw a lot of conclusions without ever having to touch it. The same way I don't need to use my fingers (or toes) to count to 10, or know that 8 bit is going to mostly suck.
Offline
User avatar

Tom_Bassford

  • Posts: 1606
  • Joined: Wed Aug 22, 2012 8:12 am
  • Location: Europe / UK

Re: The URSA broadcast seems still suffer from noise

PostWed Feb 14, 2018 11:30 am

No you absolutely don't understand. The sensor itself has a minimal impact on the overall image output. It is the sensor, combined with the electronics and software driving the camera which makes the image quality. The sensor in the Micro 4k may or may not be the same or similar to the sensor in the Broadcast URSA. However we absolutely know that it is driven totally differently, cooled differently and has different software. Thus we know that the image quality will bear little or no resemblance to the micro cameras.

Until there are in the field results from real world tests of the complete system there is no point talking about the quality and usability of this system.
http://www.atemuser.com
if it was easy it wouldn't be called engineering
Offline

Howard Roll

  • Posts: 304
  • Joined: Fri Jun 03, 2016 7:50 am

Re: The URSA broadcast seems still suffer from noise

PostWed Feb 14, 2018 6:54 pm

So you're speculating that my speculation based on extensive camera tests of a similar system are somehow less valid than your speculation based on marketing?

Then you are correct, I absolutely don't understand.

Be assured as soon as one of these becomes available for rental I'll get one and put all this speculation to rest.
Offline

Denny Smith

  • Posts: 9106
  • Joined: Thu Aug 01, 2013 4:19 pm
  • Location: USA, Northern Calif.

Re: The URSA broadcast seems still suffer from noise

PostWed Feb 14, 2018 7:02 pm

Good Howard, let me know when you do this, and I will help you, this could be fun!
Cheers
Denny Smith
SHA Productions
Offline

Howard Roll

  • Posts: 304
  • Joined: Fri Jun 03, 2016 7:50 am

Re: The URSA broadcast seems still suffer from noise

PostWed Feb 14, 2018 9:32 pm

If anyone is interested you can use a CST to look at the differences between the different gammas offered by the Ursa Broadcast because they're already in the latest version of Resolve. Broadcast video looks to be much the same as Rec709 but with a knee at the top (thank you). Extended video is like an HLG and Film is a log curve as would be expected.
Offline
User avatar

Xtreemtec

  • Posts: 2682
  • Joined: Wed Jan 02, 2013 11:48 am
  • Location: The Netherlands

Re: The URSA broadcast seems still suffer from noise

PostThu Feb 15, 2018 3:21 pm

I did a test last week at the ISE tradefair.. They had a 4K lens in front of it. UHDP50 Shutter set to 50.
I put the lens to F1.8 and pointed the cam to a point about 30/35 meters away zoomed in to 2 people talking in a not well lit corridor. While there was a little bit of noise.. I had to look for it in the viewfinder.. My JVC HM700 has much more noise in the same conditions... So if we would propper lit the persons it would be more then oke.

What you seeing in the images posted above i would describe as encoding quality.. And btw.. who says that this was filmed by the new ursa broadcast cams? This studio could be very well filmed by the old ursas or studio cams for that matter... ;)
Daniel Wittenaar .:: Xtreemtec Media Productions ::. -= www.xtreemtec.nl =-
4K OBV, ATEM TVS HD, 1M/E 4K, 2M/E Broadcast Studio 4K, Hyperdeck Studio Mini, SONY FS7 4K, Ursa Broadcast 4K
Offline

Wilson Luniz

  • Posts: 71
  • Joined: Thu May 22, 2014 4:19 pm

Re: The URSA broadcast seems still suffer from noise

PostFri Feb 16, 2018 9:10 am

Xtreemtec wrote:I did a test last week at the ISE tradefair.. They had a 4K lens in front of it. UHDP50 Shutter set to 50.
I put the lens to F1.8 and pointed the cam to a point about 30/35 meters away zoomed in to 2 people talking in a not well lit corridor. While there was a little bit of noise.. I had to look for it in the viewfinder.. My JVC HM700 has much more noise in the same conditions... So if we would propper lit the persons it would be more then oke.

What you seeing in the images posted above i would describe as encoding quality.. And btw.. who says that this was filmed by the new ursa broadcast cams? This studio could be very well filmed by the old ursas or studio cams for that matter... ;)

We're both not accurate. Nethier watching on camera's screen or a compressed youtube stream. I can't wait to see what Howard will done. And Grant said that at the begining of when he was introducing URSA broadcast.
Wilson Luniz
Generator Stage Production, Macau
Offline

Denny Smith

  • Posts: 9106
  • Joined: Thu Aug 01, 2013 4:19 pm
  • Location: USA, Northern Calif.

Re: The URSA broadcast seems still suffer from noise

PostFri Feb 16, 2018 6:28 pm

I also think, Grant said, they were using the new Ursa Broadcast cameras to shoot the program.
But as previously pointed out, trying to judge the webcast program which has compression artifacts is kind of pointless. BM might have feed the final output to their Web Presenter to stream it, which also drops the resolution to 720p.

We need original footage, not streamed to make any real judgments. Meanwhile it is educated guess work (Howard) or speculation (the rest of us). :roll:
Cheers
Denny Smith
SHA Productions
Offline

Wilson Luniz

  • Posts: 71
  • Joined: Thu May 22, 2014 4:19 pm

Re: The URSA broadcast seems still suffer from noise

PostSat Feb 17, 2018 9:23 am

Denny Smith wrote:I also think, Grant said, they were using the new Ursa Broadcast cameras to shoot the program.
But as previously pointed out, trying to judge the webcast program which has compression artifacts is kind of pointless. BM might have feed the final output to their Web Presenter to stream it, which also drops the resolution to 720p.

We need original footage, not streamed to make any real judgments. Meanwhile it is educated guess work (Howard) or speculation (the rest of us). :roll:
Cheers

No, it's 1440p(2560x1440) as Youtube offer more than HD live stream. And it's sharp enough when you watch it at 1:1.
Wilson Luniz
Generator Stage Production, Macau

Return to Live Production

Who is online

Users browsing this forum: Andy Coulthurst, Google [Bot] and 6 guests