Ursa Mini Pro 4.6K G2 footage and first thoughts

The place for questions about shooting with Blackmagic Cameras.
  • Author
  • Message
Offline

Yiorgos Tryfonas

  • Posts: 60
  • Joined: Thu Jun 02, 2016 12:40 pm

Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 10, 2019 7:10 am

Hello guys, I've filmed a couple of things with the new G2 and currently preparing for a big project next month. I wanted to give some quick feedback here for the ones that are considering the update.

First the footage:

This is a video I created to test the slow motion and low-light of the camera. Slow motion is at 4.6K 120fps and 2K 300fps as well as Anamorphic 3K at 150fps. All shots use some level of the internal NDs except the night shots at the end.


This is a video I shot during a fashion shoot. The main shoot was on the Pocket4K but I had just received the G2 so I brought it along to capture some quick shots. In the beginning of the video I'm doing some 360 follows and quick pans while following the model in order to get a first taste of the motion of the camera. The rest are high dynamic range 120/300fps shots.

Both videos are filmed using Contax Zeiss MM lenses, more specifically the 35/1.4, 50/1.4 and 85/1.4. The anamorphic shots are with the ISCO Ultrastar RED cinemascope.


Here is a quick summary of what I think of highlights:
- I will mention it first because I believe it's the biggest and most significant point. Although it's hard to advertise something that can't be boasted with numbers (for example "8K resolution!") the motion is different from camera to camera. Some people call it motion "cadence", others simply the "feel" or "look" of the motion. It's one of those things that you can't really pixel-peep and you need moving objects and/or camera movement in a full clip to perceive. It is interconnected of course with the sensor readout speed and the shutter speed. The G2, with 120fps at full sensor readout, is one of the few cameras that can provide this fast readout. With 7.6ms it's only 1ms slower than the Alexa (but at higher resolution so it's faster at Alexa's equal resolution) and faster than the RED Gemini (which does 96fps at full sensor) it approaches to film level (estimated at around 5ms). For comparison the first UMP and Pocket4K have 14~16ms at full sensor - which is still highly and very well regarded. These are just numbers but how do they change things in practise? The easy thing to talk about it's the lack of rolling shutter artifacts, which is great but never had any problems with it on the first UMP anyways. It's good that it's much better now of course. What had a very big impact to my eyes though was the motion. It's something very hard to describe with words and can only be seen in the footage or better "felt" in the footage. It just look more "right", "sweeter" or more "professional". The equivalent mental feeling is like when you've been using a slow computer for years and you are used to that speed and waiting times then you one day upgrade and everything seems faster and more "right". That's how the motion in G2 feels. For my work, which is usually music videos that involve fast camera movements and choreography this is a game changer but it's something that positively affects every single thing you are filming from talking heads to commercials. It's in my opinion, the biggest thing that has happened to a BM camera in terms of image improvement since the 15 stops of dynamic range in the 4.6K sensor and Blackmagic RAW.

- There is something great going on with the color and "clarity" of the pixels coming out of the camera. It looks subtle at first but after reviewing the total of the footage I feel that compared to the first UMP there the sensor just performs better in this aspect. "Feel" is not exactly an accurate measurement but I'm pretty sure somebody with both cameras can give a test on this later on. JB mentioned the improved color filter array (bayer filter) on the camera and I believe it's what brings this improvement.

- While the USB-C recording might seem a little unprofessional or inconvenient to some compared to other options, it's a great addition that allows for cheap fast media options without any hustle. From velro-ing a T5 or Wise SSD somehere on the body (my favorite place is on top of XLR ports) to using a Pocket4K SSD holder on a full rig, it's a great solution. Coupled with a short and flexible USB-C cable and a 90 degrees USB-C angle its signature it's very small. The SSD recorder unit is great but it takes space, adds weight and consumes the rear SDI. The Cfast/SD cards are also great depending on what you are recording and if you are offloading. Personally, I love using the USB-C as I already have 5 drives from my Pocket4K and it's really nice to just record all day and go back home to offload without the need to bother about it during shooting.
Image

- After filming and inspecting my night shots, I can definitely say the image is a little cleaner than the first UMP. Perhaps in the same way the Micro has benefited compared to the Pocket, the G2 benefits from the improved electronics or the tweaked sensor. Or perhaps it's just the improved black shading and sensor calibration. Whatever it is, I don't feel the ISO1600 is that far from Pocket4K's ISO1600 - even when that falls into the "Dual ISO" spec - and I'm very confident filming anything with the G2.

- Do mind the G2 has pixel remapping included in the black shading feature. So you can stop worrying forever about getting some pixel stuck while travelling etc.

- The higher frame rates is probably the most advertised feature of the camera and doesn't need much explanation. It just does it. Mind of course, the incredible data rates of high frame rate recording and no matter the media you will have to settle down to a higher compression ratio for the highest frame rates.


In general, I think this is a great camera to have at this time and it will certainly last for a while. It's easy to get on the hype train with 8K and Large Format cameras but I will remind everybody that all those 8K large format cameras have slower readouts (closer to the first UMP) losing any of the motion benefits I talked above and of course retain those readout times on everything non-windowed. Which is another whole topic that many users struggle with 6K-8K cameras: the fact that when they just want lower resolution they have to crop-in. Nobody knows what/if/when BM is cooking but my conclusion is that the G2 is one of the most advanced cameras out there for this generation and I dare to include some brands "up there" into that statement. Perhaps the "next-gen" cameras will be at 8K, large format and who knows perhaps an analog gain circuit for every ISO jump (multi-native-ISO)? The sure thing is that in the process they will lose some of the benefits of this camera. If you want a current gen 4K cinema camera now, this is certainly the best value. If you already own the UMP for a while and thinking about switching, you won't be dissapointed with the G2. Besides, you can easily sell the G1 right now for $4000 or more (keeping all your accessories). Then with $2000 extra you got your upgrade.
Offline

AbdoulUK

  • Posts: 223
  • Joined: Mon Mar 18, 2019 4:53 pm
  • Real Name: Abdoul Mohammad

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 10, 2019 9:02 am

Thank you Yiorgos,
This is the first mini review I've seen on the camera. Everyone else seems really quiet on waging in on an opinion, but this helps a lot.

Sounds like a great camera.

I know the camera matches well with the pocket, does the jump in image quality justify the price over the pocket?
Offline

Yiorgos Tryfonas

  • Posts: 60
  • Joined: Thu Jun 02, 2016 12:40 pm

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 10, 2019 10:31 am

AbdoulUK wrote:I know the camera matches well with the pocket, does the jump in image quality justify the price over the pocket?


I could give you the cliche reply on that "totally different cameras for different purposes" but since you asked about the image quality, the G2 is definitely a step up from the P4K. This becomes really apparent when you film them side by side with the same lenses and while sure it looks similar (you are filming the same thing after all) and you could grade the one to match the other in general, the skintones and color fidelity in general that's coming out from the G2 are above the P4K - and of course the motion that I already talked about. It doesn't mean the P4K is bad but it's not an Alexa (and nobody expect it to be). The G2? Might sound exaggerated but I'm so impressed that I believe that if the same package was put in a cube and was called Arri Alexa 4.6K, not many would complain.
Offline

AbdoulUK

  • Posts: 223
  • Joined: Mon Mar 18, 2019 4:53 pm
  • Real Name: Abdoul Mohammad

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 11, 2019 8:04 am

This is great news, thank you.

I hope you don’t mind, but I have a few questions I’d love to hear answers too if you have the time.

1. Have you encountered any bugs with the camera or is it performing well?

2. Is the 300fps useable? Theres been a few reports its too soft/low quality for high end stuff

3. Is the colour much different than the original? I know it’s been improved, but is this easily visibly different?

4. I love the idea of the cfast hard drive options but scared the cable will get snagged mid recording and corrupt the whole drive. Has anything like this happened?

Thanks in advance
Offline

Yiorgos Tryfonas

  • Posts: 60
  • Joined: Thu Jun 02, 2016 12:40 pm

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 11, 2019 10:28 am

AbdoulUK wrote:This is great news, thank you.

I hope you don’t mind, but I have a few questions I’d love to hear answers too if you have the time.

1. Have you encountered any bugs with the camera or is it performing well?

2. Is the 300fps useable? Theres been a few reports its too soft/low quality for high end stuff

3. Is the colour much different than the original? I know it’s been improved, but is this easily visibly different?

4. I love the idea of the cfast hard drive options but scared the cable will get snagged mid recording and corrupt the whole drive. Has anything like this happened?

Thanks in advance


1) Hmm no bugs.
2) 300fps is at 1/4 of the full sensor reading. Like any other windowed mode it will be softer and exaggerate any sensor and optical defects. You are zooming in 4 times in the centre of the sensor, using only 25% of the glass (or even less if it's "full frame" glass). What's "usable" is up to interpretation to the individual. If you are asking for my personal opinion is that 300fps is such a specialised use (breaking things in a close up for example) that you can easily create an ideal environment when filming at it and avoid any issues. It's as usable as any other high frame rate camera in windowed mode (REDs for example). You still have perfectly great 120fps at full sensor and 150fps at UHD - those have no compromise. I find the less experienced a user is, the more sensitive they become in every little thing, from noise, to dynamic range, to aberrations, you name it. Experienced users understand why those defects exist, compromise/adapt and do their job. Expectations management and comparison with what's out there is very important.
3) In my opinion it's the clarity of the color that's improved - something not easily observable but in its totality makes a difference to the overal quality of the image. Perhaps there are further differences in the color science. As I don't have an UMP it's impossible to compare.
4) Removing the cable from the drive or the camera will simply result into stopping the recording. You will lose a few seconds from your last clip and that's all. There is no data/drive corruption - similar to what happens when the battery dies for example. That said, those cables are very sturdy and very unlikely to get damaged. Getting the cable off accidentally is extremely unlikely as it snaps hard both on the drive and the USB-C port. You can literally leave the drive hanging over and it will not come off unless you want it to (well don't leave it hanging anyways). I would be a little careful with disconnecting the cable from the USB-C port. Not because I had any problems (both on the Pocket4K and the G2) but because you never know. It snaps in quite hard and takes a bit of power to take off. A clumsy person on a bad day will definitely find a way to damage it.
Offline

AbdoulUK

  • Posts: 223
  • Joined: Mon Mar 18, 2019 4:53 pm
  • Real Name: Abdoul Mohammad

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 11, 2019 11:29 am

Such valuable information, thank you.

Im planning to shoot at Q0 so the cfast was really interesting to me, but I’m quite clumsy so will have to see if I can justify the price of cards.

Thank you again for your feedback, your footage is the cleanest example of the camera ive seen too and was a bug factor in me placing an order. If you ever release any of your first footage for download please let me know :)
Offline

Colourberry

  • Posts: 173
  • Joined: Thu May 31, 2018 7:12 am
  • Real Name: Simon Bailey

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSun May 12, 2019 11:16 pm

Yiorgos Tryfonas wrote:
AbdoulUK wrote:2) 300fps is at 1/4 of the full sensor reading. Like any other windowed mode it will be softer and exaggerate any sensor and optical defects. You are zooming in 4 times in the centre of the sensor, using only 25% of the glass (or even less if it's "full frame" glass). What's "usable" is up to interpretation to the individual. If you are asking for my personal opinion is that 300fps is such a specialised use (breaking things in a close up for example) that you can easily create an ideal environment when filming at it and avoid any issues. It's as usable as any other high frame rate camera in windowed mode (REDs for example). You still have perfectly great 120fps at full sensor and 150fps at UHD - those have no compromise. I find the less experienced a user is, the more sensitive they become in every little thing, from noise, to dynamic range, to aberrations, you name it. Experienced users understand why those defects exist, compromise/adapt and do their job. Expectations management and comparison with what's out there is very important.


I dont buy having to do a sensor crop as an excuse for any issues with the quality of its 300fps. My Sony F5 was released 7 years ago and for 240fps it cropped to HD also and that was a beautiful image.

There is nothing wrong with pixel peeping. A lot of us are either contemplating a UMP G2 or have one on order and want to know if there are issues in advance.

I find it concerning that BM hasnt provided any footage for download like they usually do.
Offline
User avatar

Robert Niessner

  • Posts: 4948
  • Joined: Thu Feb 21, 2013 9:51 am
  • Location: Graz, Austria

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostMon May 13, 2019 12:58 am

Yiorgos Tryfonas wrote:2) 300fps is at 1/4 of the full sensor reading. Like any other windowed mode it will be softer and exaggerate any sensor and optical defects. You are zooming in 4 times in the centre of the sensor, using only 25% of the glass (or even less if it's "full frame" glass).


I think you meant at 300 fps it is at 42% of the full sensor reading (4608 / 1920 = 2.4), and you are NOT zooming 4 times into the center of the sensor - because it is a 1:1 readout.
Saying "Thx for help!" is not a crime.
--------------------------------
Robert Niessner
LAUFBILDkommission
Graz / Austria
--------------------------------
Blackmagic Camera Blog (German):
http://laufbildkommission.wordpress.com

Read the blog in English via Google Translate:
http://tinyurl.com/pjf6a3m
Offline

Yiorgos Tryfonas

  • Posts: 60
  • Joined: Thu Jun 02, 2016 12:40 pm

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostMon May 13, 2019 4:27 am

Robert Niessner wrote:
Yiorgos Tryfonas wrote:2) 300fps is at 1/4 of the full sensor reading. Like any other windowed mode it will be softer and exaggerate any sensor and optical defects. You are zooming in 4 times in the centre of the sensor, using only 25% of the glass (or even less if it's "full frame" glass).


I think you meant at 300 fps it is at 42% of the full sensor reading (4608 / 1920 = 2.4), and you are NOT zooming 4 times into the center of the sensor - because it is a 1:1 readout.


Hello Robert. Not sure why you are dividing the horizontal resolutions. It's the total area that matters when we are talking about a window (two dimensions).
In 4.6K you have 4608 x 2592 = 11943936 pixels
In 2K you have 2048 x 1152 = 2359296 pixels (20% or 5 times less)
In HD you have 1920x1080 = 2073600 pixels (17% or 5.7 times less)
So I was actually kind when I was saying 1/4 of the full sensor - I generalised for 4K sensors. Since this is a 4.6K sensor it's actually 1/5.

Perhaps my wording was bad with the word "zooming". A better wording is that you are only using that little part of the sensor. Of course it's a 1:1 readout. It's in comparison with what you are getting to the full sensor readout. Nobody that knows what they are doing actually use windowed mode to shoot HD delivery content (unless it's some special S16 project). Instead, they scale down to HD from the full sensor readout. That windowed mode is for special cases, especially when you shoot in ultra slow motion. You are not going to display that windowed mode in a little square in the center in your final product. Instead your are going to "zoom in" to fill the full screen. It's actually the exact same thing as shooting the full 4.6K sensor readout, and zooming in the footage in post (only that in that way you still have the full sensor readout and no ultra high frame rates). In this sense, you are "zooming" into the (full readout) sensor.
Last edited by Yiorgos Tryfonas on Mon May 13, 2019 6:02 am, edited 2 times in total.
Offline

Yiorgos Tryfonas

  • Posts: 60
  • Joined: Thu Jun 02, 2016 12:40 pm

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostMon May 13, 2019 5:04 am

Colourberry wrote:
Yiorgos Tryfonas wrote:
AbdoulUK wrote:2) 300fps is at 1/4 of the full sensor reading. Like any other windowed mode it will be softer and exaggerate any sensor and optical defects. You are zooming in 4 times in the centre of the sensor, using only 25% of the glass (or even less if it's "full frame" glass). What's "usable" is up to interpretation to the individual. If you are asking for my personal opinion is that 300fps is such a specialised use (breaking things in a close up for example) that you can easily create an ideal environment when filming at it and avoid any issues. It's as usable as any other high frame rate camera in windowed mode (REDs for example). You still have perfectly great 120fps at full sensor and 150fps at UHD - those have no compromise. I find the less experienced a user is, the more sensitive they become in every little thing, from noise, to dynamic range, to aberrations, you name it. Experienced users understand why those defects exist, compromise/adapt and do their job. Expectations management and comparison with what's out there is very important.


I dont buy having to do a sensor crop as an excuse for any issues with the quality of its 300fps. My Sony F5 was released 7 years ago and for 240fps it cropped to HD also and that was a beautiful image.

There is nothing wrong with pixel peeping. A lot of us are either contemplating a UMP G2 or have one on order and want to know if there are issues in advance.

I find it concerning that BM hasnt provided any footage for download like they usually do.


Every BM camera since the first Ursa has a windowed mode. The Ursa Mini Pro had 120fps at 2K windowed mode. As it was mentioned plenty of times, the G2 carries an improved version 2 of that 4.6K sensor. So, in the G2, you get the same, if not better performance because of the sensor improvements. If you were happy with the 120fps of the Ursa Mini Pro, you would be happy with the 300fps of the G2. There is nothing to hide here as you imply. 120fps at 4.6K and 150fps at UHD is something to proudly shout about actually. The 300fps at windowed mode is pretty great for what it is as well.

You are causing a whole fuzz and bringing doubt to others about a great feature, simply based on a single shot with an aliasing-sensitive item, also embraced in lens CA, shot in unfavourable conditions. I saw that detail in the shot when I was editing but I didn't believe it would be something to picked up to this extend; I assumed the users will understand it's a condition-specific and easy to control issue, like aberrations in lenses for example. Your Pocket4K at 120fps has very similar performance compromises. I think the only way at this point, is to get a G2 in your hands and decide whether it's what you are looking for. Alternatively, you can wait for more and more footage as it comes out - but there is nothing like first hand experience. BM likes to put actual user footage on their site's library - they never pay to create commercials as they believe it's deceiving. The camera is just out so I'm pretty sure they will update their website with some nice user content.

I shall mention as a note that your F5 could do a 2K mode without cropping (I suspect by reading less pixels during a full sensor scan) along with a cropped 2K. There was a special OLPF filter for the 2K mode as well - without it aliasing would be bad. It also costed $35,000 or about 6xUMPG2. If you find the windowed mode on the G2 is not enough for your work you can always get an OLPF from Rawlite.
Offline

Colourberry

  • Posts: 173
  • Joined: Thu May 31, 2018 7:12 am
  • Real Name: Simon Bailey

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostMon May 13, 2019 6:23 am

If by unfavourable conditions do you mean daylight?? and aliasing sensitive item its some small straight lines and they arent even close together like a stripped shirt.

The 300fps for me is the main selling point of the camera but when you refer to it and say "for what it is" I cant tell if its just you opinion or if theres an issue.

I expect that HD crop to be the quality of HD. Not line skipping to achieve 300 fps. If it is then we should be told before we buy.

I'm so close to cancelling but I cant cancel based on your one shot. But it is the first shot with some items in the shot to show it up.

My F5 didnt have the OLPF and it looked great. It also didnt cost $35k your thinking of the F55.

I'm checking Youtube and Vimeo regularly and nobody is uploading anything. All we've seen of the 300fps is surfing shots which are too hard to judge and your shot.

Can someone go point their G2 at some cars please. Doesnt need to be pretty.
Offline

Tim Schumann

Blackmagic Design

  • Posts: 621
  • Joined: Tue Oct 23, 2012 6:21 am

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 2:12 am

Colourberry wrote:I expect that HD crop to be the quality of HD. Not line skipping to achieve 300 fps. If it is then we should be told before we buy.


We make no secret of the fact that to reach 300fps on the G2 you need to shoot windowed at 2K DCI or 1080p.

This is plainly stated in the paragraph titled 'High Frame Rate Recording' on the first page of the G2 web page and also on our tech specs page as well.

This is not line skipping in any way... it is a 2K DCI or 1080p window on a Bayer CMOS sensor.
Windowing a sensor down like this always has some trade-offs.

If you are not happy with the image quality at this crop level we also offer a myriad of other resolutions, frame rates and codecs you can shoot in. All of which you can do so internally in 12-bit non-linear RAW without an external recorder.
Offline

Stephen Fitzgerald

  • Posts: 226
  • Joined: Sat Sep 15, 2018 1:00 am
  • Real Name: Stephen Fitzgerald

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 2:54 am

Very straightforward answer this is the kind of responses I consistently appreciate.
Offline

Tim Schumann

Blackmagic Design

  • Posts: 621
  • Joined: Tue Oct 23, 2012 6:21 am

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 2:58 am

If you'd like to see some more examples of footage shot on G2 at high frame rates this is a nice one from Mark Wyatt. Search for 'Blackmagic Ursa Mini Pro G2 Footage - Westcoast of Canada' on Youtube
Offline

Colourberry

  • Posts: 173
  • Joined: Thu May 31, 2018 7:12 am
  • Real Name: Simon Bailey

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 10:53 am

Thanks for responding. I have no problem with a HD crop at all. Theres nothing wrong with HD. As long as its still the quality of HD.

If its not line skipping then I'm sure there wont be a aliasing problem.

Looking forward to seeing some footage somewhere some day thats not surfing :)

I really would love to see a rally car taking a corner in dirt close on the back wheel. Some footage of hard surfaces.
Offline

MishaEngel

  • Posts: 1432
  • Joined: Wed Aug 29, 2018 12:18 am
  • Real Name: Misha Engel

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 12:28 pm

Colourberry wrote:Thanks for responding. I have no problem with a HD crop at all. Theres nothing wrong with HD. As long as its still the quality of HD.

If its not line skipping then I'm sure there wont be a aliasing problem.

Looking forward to seeing some footage somewhere some day thats not surfing :)

I really would love to see a rally car taking a corner in dirt close on the back wheel. Some footage of hard surfaces.


The quality of HD is 720P. From a 1080 bayer sensor you should be able to derive a nice HD signal(1280x720), since the ratio is 1.5. Just remember that you need a lot of light to shoot 300 fps.
Offline

Colourberry

  • Posts: 173
  • Joined: Thu May 31, 2018 7:12 am
  • Real Name: Simon Bailey

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 1:48 pm

MishaEngel wrote:
Colourberry wrote:Thanks for responding. I have no problem with a HD crop at all. Theres nothing wrong with HD. As long as its still the quality of HD.

If its not line skipping then I'm sure there wont be a aliasing problem.

Looking forward to seeing some footage somewhere some day thats not surfing :)

I really would love to see a rally car taking a corner in dirt close on the back wheel. Some footage of hard surfaces.


The quality of HD is 720P. From a 1080 bayer sensor you should be able to derive a nice HD signal(1280x720), since the ratio is 1.5. Just remember that you need a lot of light to shoot 300 fps.



It clearly says on the UMP G2 webpage Windowed 1080p. Not 720p. Not sure where you got the impression its 720p.
Offline

MishaEngel

  • Posts: 1432
  • Joined: Wed Aug 29, 2018 12:18 am
  • Real Name: Misha Engel

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 2:34 pm

Colourberry wrote:
MishaEngel wrote:
Colourberry wrote:Thanks for responding. I have no problem with a HD crop at all. Theres nothing wrong with HD. As long as its still the quality of HD.

If its not line skipping then I'm sure there wont be a aliasing problem.

Looking forward to seeing some footage somewhere some day thats not surfing :)

I really would love to see a rally car taking a corner in dirt close on the back wheel. Some footage of hard surfaces.


The quality of HD is 720P. From a 1080 bayer sensor you should be able to derive a nice HD signal(1280x720), since the ratio is 1.5. Just remember that you need a lot of light to shoot 300 fps.



It clearly says on the UMP G2 webpage Windowed 1080p. Not 720p. Not sure where you got the impression its 720p.


1080p sensor is bayer pattern, your screen, display, beamer gives a RGB-output.
https://en.wikipedia.org/wiki/Bayer_filter
https://en.wikipedia.org/wiki/Demosaicing

With a normal demosaicing algorithm you need around twice the bayer pixels to get a decent RGB-output. ARRI uses a 1.5horizontal x 1.5vertical = 2.25 ratio with their ALEXA's
Sensor Active Image Area (photosites) : HD: 2880 x 1620 (from ARRI.com).
Offline

Colourberry

  • Posts: 173
  • Joined: Thu May 31, 2018 7:12 am
  • Real Name: Simon Bailey

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 2:53 pm

When it says 1080p I expect 1080p.

Im not sure what your point is. Are all 1080p images not subject to the same?

Hate to keep referring to it but my F5 had a beautifully detailed HD crop that looked nothing like 720p. I can say the same about the A7Rii.

If it's not line skipping then we can expect to see some beautiful pics from the camera at 300fps. I just wish there was more online.

And some BRAW for us to download.

Sent from my SM-G950F using Tapatalk
Offline

youlikeny

  • Posts: 173
  • Joined: Fri May 10, 2019 9:45 pm
  • Real Name: Alessandro Penazzi

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 2:55 pm

MishaEngel wrote:
1080p sensor is bayer pattern, your screen, display, beamer gives a RGB-output.
https://en.wikipedia.org/wiki/Bayer_filter
https://en.wikipedia.org/wiki/Demosaicing

With a normal demosaicing algorithm you need around twice the bayer pixels to get a decent RGB-output. ARRI uses a 1.5horizontal x 1.5vertical = 2.25 ratio with their ALEXA's
Sensor Active Image Area (photosites) : HD: 2880 x 1620 (from ARRI.com).


And how does this translates in the Mini Pro producing a 720p output?
Offline

Dan Shay

  • Posts: 45
  • Joined: Tue Apr 12, 2016 4:06 pm

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 3:38 pm

Does does the G2 compare to the 4.6k pro it terms of FPN at 800 iso and 1600 iso? I'd like to be able to shoot 1600 iso at daytime, to get the 7 stops of highlight protection.
Offline

Peter917

  • Posts: 3
  • Joined: Fri May 03, 2019 11:31 pm
  • Real Name: Peter Poulos

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 4:26 pm

I should be receiving my camera on Tuesday, so I can do some tests next week and upload a video. If there is anything people want to see let me know. I have a shoot June 1st so I want to test the hell out of it before then and pray I do not have the screen glitch issue or any others reported.
Offline

MishaEngel

  • Posts: 1432
  • Joined: Wed Aug 29, 2018 12:18 am
  • Real Name: Misha Engel

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 5:22 pm

youlikeny wrote:
MishaEngel wrote:
1080p sensor is bayer pattern, your screen, display, beamer gives a RGB-output.
https://en.wikipedia.org/wiki/Bayer_filter
https://en.wikipedia.org/wiki/Demosaicing

With a normal demosaicing algorithm you need around twice the bayer pixels to get a decent RGB-output. ARRI uses a 1.5horizontal x 1.5vertical = 2.25 ratio with their ALEXA's
Sensor Active Image Area (photosites) : HD: 2880 x 1620 (from ARRI.com).


And how does this translates in the Mini Pro producing a 720p output?


At 300 fps the sensor read out is 1920h x 1080v bayer pixels.
With the best currently available demosaicing algorithm this can give you a 1920v x 0.75 = 1440 horizontal and a 1080h x 0.75 = 810 pixel vertical resolution when you output it on a RGB screens.

The nearest standard output is 1280h x 720v which is 720p.
Offline
User avatar

Jamie LeJeune

  • Posts: 2012
  • Joined: Tue Feb 26, 2013 4:33 am
  • Location: San Francisco

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 7:27 pm

Dan Shay wrote:Does does the G2 compare to the 4.6k pro it terms of FPN at 800 iso and 1600 iso? I'd like to be able to shoot 1600 iso at daytime, to get the 7 stops of highlight protection.
I haven't done a side by side comparison with my G1, but in the two weeks I've had my G2 I haven't noticed FPN at any ISO. It's been delivering clean images with good color channel separation. I think the G2 will work just fine for you at 1600. For that use the G1 works just as well, in my experience, as long as you don't underexpose or try to raise the shadows in post.
www.cinedocs.com
http://www.imdb.com/name/nm4601572/
Offline

MishaEngel

  • Posts: 1432
  • Joined: Wed Aug 29, 2018 12:18 am
  • Real Name: Misha Engel

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostFri May 17, 2019 8:55 pm

Jamie LeJeune wrote:
Dan Shay wrote:Does does the G2 compare to the 4.6k pro it terms of FPN at 800 iso and 1600 iso? I'd like to be able to shoot 1600 iso at daytime, to get the 7 stops of highlight protection.
I haven't done a side by side comparison with my G1, but in the two weeks I've had my G2 I haven't noticed FPN at any ISO. It's been delivering clean images with good color channel separation. I think the G2 will work just fine for you at 1600. For that use the G1 works just as well, in my experience, as long as you don't underexpose or try to raise the shadows in post.


Good to hear and how is the rest of the camera? rolling shutter, DR, etc.. compared to the G1?
Offline

Colourberry

  • Posts: 173
  • Joined: Thu May 31, 2018 7:12 am
  • Real Name: Simon Bailey

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 12:21 am

MishaEngel wrote:
youlikeny wrote:
MishaEngel wrote:
1080p sensor is bayer pattern, your screen, display, beamer gives a RGB-output.
https://en.wikipedia.org/wiki/Bayer_filter
https://en.wikipedia.org/wiki/Demosaicing

With a normal demosaicing algorithm you need around twice the bayer pixels to get a decent RGB-output. ARRI uses a 1.5horizontal x 1.5vertical = 2.25 ratio with their ALEXA's
Sensor Active Image Area (photosites) : HD: 2880 x 1620 (from ARRI.com).


And how does this translates in the Mini Pro producing a 720p output?


At 300 fps the sensor read out is 1920h x 1080v bayer pixels.
With the best currently available demosaicing algorithm this can give you a 1920v x 0.75 = 1440 horizontal and a 1080h x 0.75 = 810 pixel vertical resolution when you output it on a RGB screens.

The nearest standard output is 1280h x 720v which is 720p.


But why make that comment here about the UMPG2 1080p crop?? Its like you were saying for some reason that the 1080p crop on the G2 is inferior to other 1080p HD cameras?

Your claim doesn't sound right anyhow but I dont want to spend time researching whats what. But if its right then it applies to all cameras doesnt it?

It says 1080p HD so lets go off that. Its easier language in a field with enough other complications and its an easy to understand.
Offline

MishaEngel

  • Posts: 1432
  • Joined: Wed Aug 29, 2018 12:18 am
  • Real Name: Misha Engel

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 12:25 am

Colourberry wrote:Your claim doesn't sound right anyhow but I dont want to spend time researching whats what. But if its right then it applies to all cameras doesnt it?


It applies to all cameras with a bayer patern sensor.
Offline

youlikeny

  • Posts: 173
  • Joined: Fri May 10, 2019 9:45 pm
  • Real Name: Alessandro Penazzi

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 1:32 am

Colourberry wrote:Your claim doesn't sound right anyhow but I dont want to spend time researching whats what. But if its right then it applies to all cameras doesnt it?


Exactly, all cameras out there today have bayer sensor... (there are some still cameras with a different technology, like Sigma Foveons, but not video or cinema ones)
Offline

MishaEngel

  • Posts: 1432
  • Joined: Wed Aug 29, 2018 12:18 am
  • Real Name: Misha Engel

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 12:16 pm

youlikeny wrote:
Colourberry wrote:Your claim doesn't sound right anyhow but I dont want to spend time researching whats what. But if its right then it applies to all cameras doesnt it?


Exactly, all cameras out there today have bayer sensor... (there are some still cameras with a different technology, like Sigma Foveons, but not video or cinema ones)


In broadcast 3 CMOS camera's are still a big deal
https://pro.sony/en_MY/products/4k-and-hd-camera-systems

With this one being the newest high-end broadcast camera
https://pro.sony/en_MY/products/4k-and-hd-camera-systems/uhc-8300

The biggest advantage of 3 CMOS over 1 bayer patern is that you can make your zooms a lot smaller with the same zoom range(3 small CMOS vs. 1 big bayer CMOS). Biggest drawback is price and less latitude.
Offline

youlikeny

  • Posts: 173
  • Joined: Fri May 10, 2019 9:45 pm
  • Real Name: Alessandro Penazzi

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 1:20 pm

MishaEngel wrote:
youlikeny wrote:
Colourberry wrote:Your claim doesn't sound right anyhow but I dont want to spend time researching whats what. But if its right then it applies to all cameras doesnt it?


Exactly, all cameras out there today have bayer sensor... (there are some still cameras with a different technology, like Sigma Foveons, but not video or cinema ones)


In broadcast 3 CMOS camera's are still a big deal
https://pro.sony/en_MY/products/4k-and-hd-camera-systems

With this one being the newest high-end broadcast camera
https://pro.sony/en_MY/products/4k-and-hd-camera-systems/uhc-8300

The biggest advantage of 3 CMOS over 1 bayer patern is that you can make your zooms a lot smaller with the same zoom range(3 small CMOS vs. 1 big bayer CMOS). Biggest drawback is price and less latitude.


Broadcast!?!?!?! You realize we are talking about cinema cameras with large sensors, right? Those cameras have 1" or 1.25" sensors!!!
Offline
User avatar

MScDre

  • Posts: 165
  • Joined: Mon Apr 30, 2018 8:31 pm
  • Real Name: Andrea Domenichini

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 1:23 pm

I wish the USB-C port was near the SDI outs and could be used without opening the display
Offline

Colourberry

  • Posts: 173
  • Joined: Thu May 31, 2018 7:12 am
  • Real Name: Simon Bailey

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 1:50 pm

MScDre wrote:I wish the USB-C port was near the SDI outs and could be used without opening the display


Absolutely. Sucks that is you just want to shoot with a viewfinder you have to have the door open. I assume they didnt move it because they didnt want to retool new molds etc and just use the existing design.

But surely at least the flipout monitor could have been modified just a little to have a cutout. While at it make it rotate all the way around.
Offline

MishaEngel

  • Posts: 1432
  • Joined: Wed Aug 29, 2018 12:18 am
  • Real Name: Misha Engel

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 2:25 pm

youlikeny wrote:Broadcast!?!?!?! You realize we are talking about cinema cameras with large sensors, right? Those cameras have 1" or 1.25" sensors!!!


Get Three Cameras in One
Digital film quality with broadcast camera features makes URSA Mini Pro perfect for film, broadcast and studio camera work!

https://www.blackmagicdesign.com/products/blackmagicursaminipro/workflow

One of the reasons we have them.
Offline

youlikeny

  • Posts: 173
  • Joined: Fri May 10, 2019 9:45 pm
  • Real Name: Alessandro Penazzi

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 3:12 pm

So you are saying that there are cameras out there with 3CMOS and tiny sensors, and that the Ursa Mini Pro, with a large sensor, can technically be used as a broadcast camera. Obviously it doesn’t work the other way around as you don’t use tiny sensors broadcast cameras for anything cinema/narrative/advertising/commercial, etc...
So I stand corrected that all cinema cameras have Bayer sensor, right?
So what’s your point?
Offline

MishaEngel

  • Posts: 1432
  • Joined: Wed Aug 29, 2018 12:18 am
  • Real Name: Misha Engel

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 3:26 pm

youlikeny wrote:So you are saying that there are cameras out there with 3CMOS and tiny sensors


No I'm not, I gave a link with a lot of Sony Camera's with 3 CMOS sensors.

youlikeny wrote:and that the Ursa Mini Pro, with a large sensor, can technically be used as a broadcast camera.


Yes.

youlikeny wrote:Obviously it doesn’t work the other way around as you don’t use tiny sensors broadcast cameras for anything cinema/narrative/advertising/commercial, etc...
So I stand corrected that all cinema cameras have Bayer sensor, right?


No.

youlikeny wrote:So what’s your point?


About what?
Offline

youlikeny

  • Posts: 173
  • Joined: Fri May 10, 2019 9:45 pm
  • Real Name: Alessandro Penazzi

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 3:31 pm

MishaEngel wrote:
youlikeny wrote:So you are saying that there are cameras out there with 3CMOS and tiny sensors


No I'm not, I gave a link with a lot of Sony Camera's with 3 CMOS sensors.

youlikeny wrote:and that the Ursa Mini Pro, with a large sensor, can technically be used as a broadcast camera.


Yes.

youlikeny wrote:Obviously it doesn’t work the other way around as you don’t use tiny sensors broadcast cameras for anything cinema/narrative/advertising/commercial, etc...
So I stand corrected that all cinema cameras have Bayer sensor, right?


No.

youlikeny wrote:So what’s your point?


About what?


I don’t understand, are you trolling me?
I literally just wrote: you are saying that there are cameras with 3CMOS and tiny sensor and you replied: no I have a link to cameras with 3CMOS and tiny sensors? You realize we said the exact same thing?

And then you are telling me that I’m wrong in saying that all cinema cameras have Bayer sensors, yet you can’t point to a single cinema camera without Bayer sensor???
Offline

Colourberry

  • Posts: 173
  • Joined: Thu May 31, 2018 7:12 am
  • Real Name: Simon Bailey

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSat May 18, 2019 3:36 pm

"I'm with you fellas"

To quote a moment in Oh brother where art thou that this reminds me of.
Offline

Wayne Steven

  • Posts: 3362
  • Joined: Thu Aug 01, 2013 3:58 am
  • Location: Earth

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSun May 19, 2019 10:28 am

What Latitude? Looks like it needs more latitude.

The colour reminds me of the gy-hd1, which was poppy in the vivid stuff. I wonder if this can grade like the original pocket, where you can make the scene pop in a pleasing way along with contrast. I wish the pocket 4k could do this.

It looks bland in detail and highlight cutoff, maybe this is the Braw? I wish it was still sub $3k.
aIf you are not truthfully progressive, maybe you shouldn't say anything
bTruthful side topics in-line with or related to, the discussion accepted
cOften people deceive themselves so much they do not understand, even when the truth is explained to them
Offline

Colourberry

  • Posts: 173
  • Joined: Thu May 31, 2018 7:12 am
  • Real Name: Simon Bailey

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSun May 19, 2019 10:36 am

Wayne Steven wrote:What Latitude? Looks like it needs more latitude.

The colour reminds me of the gy-hd1, which was poppy in the vivid stuff. I wonder if this can grade like the original pocket, where you can make the scene pop in a pleasing way along with contrast. I wish the pocket 4k could do this.

It looks bland in detail and highlight cutoff, maybe this is the Braw? I wish it was still sub $3k.
Have you posted that in the wrong place?


Sent from my SM-G950F using Tapatalk
Offline

Wayne Steven

  • Posts: 3362
  • Joined: Thu Aug 01, 2013 3:58 am
  • Location: Earth

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSun May 19, 2019 10:37 am

I think Panavision's original cinema camera may have had rgb filter. Sony had that tilted pixel top end camera which I think used some weird filter. Red dies monochrome ones. Has anybody imemented the random colour filter in a cinema camera? Some other weird filtering going around that supposed to be good, don't know if any got into top cameras. I think the hybrid complementary colour filter used in the hd1/10/PD1 etc, probably gave a good cinema like image when used correctly, though not a cinema camera. The fran8k and Zcam E2 both use a HDR adjusted Bayer array. The sharp, cough, cinema camera probably uses a quad Bayer, does the pocket 4k?
Last edited by Wayne Steven on Sun May 19, 2019 10:41 am, edited 1 time in total.
aIf you are not truthfully progressive, maybe you shouldn't say anything
bTruthful side topics in-line with or related to, the discussion accepted
cOften people deceive themselves so much they do not understand, even when the truth is explained to them
Offline

Wayne Steven

  • Posts: 3362
  • Joined: Thu Aug 01, 2013 3:58 am
  • Location: Earth

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSun May 19, 2019 10:40 am

Colourberry wrote:
Wayne Steven wrote:What Latitude? Looks like it needs more latitude.

The colour reminds me of the gy-hd1, which was poppy in the vivid stuff. I wonder if this can grade like the original pocket, where you can make the scene pop in a pleasing way along with contrast. I wish the pocket 4k could do this.

It looks bland in detail and highlight cutoff, maybe this is the Braw? I wish it was still sub $3k.
Have you posted that in the wrong place?


Sent from my SM-G950F using Tapatalk


Nope, I'm old enough to remember the hd1. Makes you appreciate what you had, despite it's flaws, took years before we saw a low end cinema camera do that sort of colour (except things like drake). Re-edit: Well, maybe not years then.
aIf you are not truthfully progressive, maybe you shouldn't say anything
bTruthful side topics in-line with or related to, the discussion accepted
cOften people deceive themselves so much they do not understand, even when the truth is explained to them
Offline

Colourberry

  • Posts: 173
  • Joined: Thu May 31, 2018 7:12 am
  • Real Name: Simon Bailey

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSun May 19, 2019 2:14 pm

Well I'm old enough to have edited 1 inch tape and what we had then makes me appreciate what we have now. I dont miss it.

Back to this camera.

Can anyone that has the G2 let us know if there is audio lag through the headphones??
Offline

Wayne Steven

  • Posts: 3362
  • Joined: Thu Aug 01, 2013 3:58 am
  • Location: Earth

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostSun May 19, 2019 9:40 pm

Colourberry wrote:Well I'm old enough to have edited 1 inch tape and what we had then makes me appreciate what we have now. I dont miss it.


I only remember editing 3/4 inch tape. But I'm talking about appreciating an "quality" aspect.
aIf you are not truthfully progressive, maybe you shouldn't say anything
bTruthful side topics in-line with or related to, the discussion accepted
cOften people deceive themselves so much they do not understand, even when the truth is explained to them
Offline

Denny Smith

  • Posts: 13131
  • Joined: Thu Aug 01, 2013 4:19 pm
  • Location: USA, Northern Calif.

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostTue May 21, 2019 4:50 pm

Well, the first Cinema digital cameras were RGB 3-CCD 2/3rds (B4), then 1-inch CCD sensors, no Bayer filter. While they re not used too much today, they did and do exist. Look at all the 2/3rds B4 Cine lenses (like the Zeiss DigiPrimes) that show up on sites like eBay.

The Panavision HD-900F is the first digital high definition camera, able to record using the standard motion picture frame rate of 24 frames per second. It is the result of a collaboration of Panavision and Sony In 1997. This was a 2/3rds RGB CCD sensor camera. It was fOllowed by several other models from other makers. This camera also gave birth to the Sony Cine Alta Cameras we have today, and the rest is history. ;)
Cheers
Denny Smith
SHA Productions
Offline

Wayne Steven

  • Posts: 3362
  • Joined: Thu Aug 01, 2013 3:58 am
  • Location: Earth

Re: Ursa Mini Pro 4.6K G2 footage and first thoughts

PostTue May 21, 2019 5:52 pm

Yep, I forgot those cameras. Thanks for pointing that out Denny.
aIf you are not truthfully progressive, maybe you shouldn't say anything
bTruthful side topics in-line with or related to, the discussion accepted
cOften people deceive themselves so much they do not understand, even when the truth is explained to them

Return to Cinematography

Who is online

Users browsing this forum: roger.magnusson, StrongOnline007 and 96 guests