I was really excited when Blackmagic Design announced the new Ursa Broadcast camera. Grant Petty showed us a camera that can be used with a standard B4 lens and has a sensor/converter combination that can handle the image circle that a B4 lens projects. I first thought that Blackmagic Design would have done it the right and straight way by creating or buying a 2/3inch sensor and putting it into an existing body. But that is not the case.
As far as I know it, the camera uses the same sensor that the Mini-Studio-Camera uses. It has exactly the same dimensions and resolution, but not the same dynamic range. This could be due to the fact that the Ursa body has a better option to cool the sensor and more space to use more electronics for better processing. The problem with this sensor is, that it is too big to be used with B4 lenses. A B4 lens projects an image circle that is too small to fill the sensor of the camera. The result would be dark edges and a huge lack of resolution in the border area. Blackmagic Design compensates for that with an optical adapter that is built into the B4 mount that comes with the camera. Unlike a speed booster that reduces the image circle, it expands it. That means you will lose light and suffer from some other problems like resolution loss and other optical artifacts. I think that Blackmagic Design does that because it is much cheaper to build a camera around an existing and well-known sensor. Blackmagic has had a lot of issues with all of their camera sensors in the beginning like Fixed Pattern Noise, Banding, Black Sun Effect, Purple Cast and so on. It is not a trivial task to build a camera from scratch so they relied on existing technology that they already use in other cameras. That is good for Blackmagic Design but not as good for the customers because of the disadvantages that are introduced by the optical converter. Insofar it is really interesting how the Ursa Broadcast performs against a well-known camcorder from a vendor that is widely accepted in any broadcast environment: Panasonic. We compared the Ursa Broadcast to the HPX3100 from Panasonic.
It would be a really nice if you give us a like on our facebook page https://www.facebook.com/mindandvision.tv/ if you like this test. It would be even better if you rate us with five stars on facebook. To be clear: This is not necessary but it would be a nice touch if you honor the effort that we put into this test. Thank you for that!
Effective Sensor Size
2/3 inch sensor size when using 4K B4 mount (Actual sensor size 13.056mm x 7.344mm)
B4 2/3 inch
Electronic control via 12-pin broadcast connector. Control also possible via EF mount pins when using optional URSA Mini Pro EF Mount.
Project frame rates of 23.98, 24, 25, 29.97, 30, 50, 59.94 and 60 fps supported. Off-speed frame rates up to 60p.
Built in ND Filters
Four position ND filter wheel with clear, 2-stop, 4-stop and 6-stop ND filters.
Focus button turns on peaking, auto focus available using compatible lenses with focus servo.
Total Video Inputs
1 x SDI
Total Video Outputs
1 x SDI, 1 x SDI Monitoring.
1.5G, 3G, 6G, 12G.
Analog Audio Inputs
2 x XLR switchable between mic, line and AES digital audio. Phantom power support.
Analog Audio Outputs
1 x 3.5mm headphone jack, supports iPhone microphone for talkback.
2 x 2.5mm LANC input for Rec Start/Stop, plus Iris and Focus control using compatible lenses.
USB Type-C for software updates.
Power Supply DC12V (11V to 17V)
Power Consumption 34 W (main unit only)
Operating Temperature 0°C to 40°C (32 °F to 104 °F)
Operating Humidity 10 % to 85 % (relative humidity)
Keeping Temperature −20 °C to 60 °C (−4 °F ro 140 °F)
Operating Time Approx. 150 min., when using DIONIC90 battery
Weight Approx. 3.9 kg/8.6 lbs (main unit only)
Dimensions (W x H x D) 14.0 cm x 27.05 cm x 33.58 cm excluding handle and option cove
Camera Section CCD Elements CCD x 3 (2/3-type interline transfer type, RGB, 2,200,000 pixel)
Picture Elements Total: 2010 (H) x 1120 (V)
Lens Mount Optical image stabilizer lens, 22x motorized zoom, F1.6 – 3.2 (f=0.39 cm – 8.6 cm), 3.5 cm conversion: 2.8 cm – 61.6 cm (16:9)
Filter Diameter 2/3-bayonet type
Optical Color Separation Prism system
CCl Filters A: 3200K, B: 4300K, C: 5600K, D: 6300K
ND Filters 1: CLEAR, 2: 1/4ND, 3: 1/16ND, 4: 1/64ND
Quantizing 14 bits
Horizontal Drive Frequency 59.94 Hz: 74.1758 MHz, 50 Hz: 74.25 MHz
Sampling Frequency 59.94 Hz: 74.1758 MHz, 50 Hz: 74.25 MHz
Digital Signal Process 59.94 Hz: 74.1758 MHz, 50 Hz: 74.25 MHz
Programmable Gain −6 dB, −3 dB, 0 dB, 3 dB, 6 dB, 9 dB, 12 dB, 15 dB, 18 dB, 21 dB, 24 dB, 27 dB, 30 dB selectable
Digital Super Gain 6 dB, 10 dB, 12 dB, 15 dB, 20 dB, 24 dB, 28 dB, 34 dB selectable
Super Gain 30 dB, 36 dB, 42 dB selectable
Shutter Speed 1/60 (50 Hz) sec., 1/100 (59.94 Hz) sec.,1/120 sec., 1/250 sec.,1/500 sec., 1/1000 sec., 1/2000 sec., HALF 180.0 deg, 172.8 deg, 144.0 deg, 120.0 deg, 90.0 deg, 45.0 deg
Shutter Speed (Syncro Scan) 1/61.7 sec. to 1/7200 sec. (1080/59.94i, 480/59.94i) 1/30.9 sec. to 1/3600 sec. (1080/29.97p,480/29.97p) 1/24.7 sec. to 1/2880 sec. (1080/23.98p,480/23.98p) 1/51.4 sec. to 1/6000 sec. (1080/50i,576/50i) 1/25.7 sec. to 1/3000 sec. (1080/25p,576/25p)
Sensitivity 1080/59.94i: F11, 1080/50i: F12 (2000 lx, 89.9 % reflect)
Minimum Luminance 0.005 lx (F1.4, S.GAIN 42 dB + DS.GAIN 34 dB)
Video S/N DNR ON: 59 dB, DNR OFF: 54 dB (standard)
Registration Less than 0.03 % (whole zone, without lens distortion)
Horizontal Resolution 1,000 TV lines (at center standard)
Although the Ursa Broadcast is capable of recording and delivering a resolution of 4k, we only tested the camera in HD resolution. We did this because all major broadcasters around the world still rely on HD and 4k is still a niche market. We would use the camera in 99% of all cases in HD resolution, so this is the important task for us. The Ursa Broadcasts allows recording in different codecs and formats. Apples ProRes, Avids DNxHD, and RAW are the codec options – 4k (3840x2160) and HD (1920x1080) are the resolution options. Besides that, there are a lot of framerate options from time-lapse up to 60 frames per second. There is also an interlace option but only in HD resolution and with the DNxHD codec.
In theory, if you do a broadcast comparison of two cameras, you would use 1080i50 (for Europe) as a resolution, framerate, and scan mode. That is the worldwide standard of most of all HDTV-Stations. In the USA it would be 1080i60 or 1080i59.94. If you use this format you will have the maximum of compatibility with all broadcast gear and markets. We always use this production format and use format converters to get a 1080p25 stream for internet use. If you want to record interlaced with the Ursa Broadcast, you have to use the DNxHD codec and HD resolution. In 4k there are only progressive framerates available. So we switched the camera to this format to match the format that the Panasonic HPX3100 records.
I like the viewfinder of the Ursa very much. We own an Ursa Mini Pro with a viewfinder and had a lot of experience with that in advance. The viewfinder is razor-sharp, has a great image quality and a lot of possible adjustments. So i was very puzzled as I saw a lot of moiré and artifacts on diagonal lines when I switched on the Ursa Broadcast in 1080i50 mode for the first time. It looked very bad compared to the image of the HPX3100. I first thought that these problems were due to the fact that the Ursa Broadcast doesn‘t do a sensor crop when switched to HD resolution but uses the full 4k sensor and scales this signal down to HD. During this process (if it isn‘t done right and with enough processing power) you can introduce a lot of moiré issues. But finally I switched the camera to 50p and all the obvious and big moiré issues went away. So I decided to record 1080p50 with the Ursa Broadcast and convert that in our NLE back to 50i. Even in live applications, this would work well, because you can still switch the SDI output of the Ursa Broadcast to 50i even if you internally record 50p.
Edit (9th of April 2018):
Blackmagic released a new firmware version for the camera that adresses some issues in 1080i modes. I couldn't test it because this firmware simply was not available during the test of the camera. Maybe this issue is already solved.
Let‘s compare the form factor of both cameras first. The Panasonic comes in a shape that is used for many years and by all vendors like Sony, Ikegami, Hitachi, GV and so on. Every button is easily accessible – all functions like white balance, record switches, and gain options are in a place where a single camera operator can access them in an easy way, even while the camera is rolling. The connectors for microphones or mobile mixers are on the back. You have a little LCD monitor that can be flipped out and rotated by 180 degrees. You can even flip the monitor back in when you rotate it by 180 degrees. Timecode options are instantly accessible on the side of the camera. An LCD display shows the timecode, audio levels, and some other camera options. The camera has a uni-slot that you can use to install a wireless microphone receiver. It has a very solid top handle and two installation points on that to install headlights or monitors. You have 3 record switches: One on the lens handgrip, one on the camera under the lens an finally one on the top handle. You can access the two P2 slots for the P2 recording media on the left side. You could even change a full card while recording on the other during shooting.
Blackmagic Design adopts a lot of these ergonomic design. The Ursa Broadcast has a big LCD display that shows audio levels and Timecode too. Some switches on the left side allow changing gain, shutter and white-balance settings. Unfortunately, they didn‘t adopt everything. For example: They put a record button on the left side above the mentioned switches. That is one of the worst places where you would place a record button. In any non-shoulder situation, you would search that button and are unable to find it because of the missing raised position. Most of all broadcast cameras have a record switch under the lens. And this switch is clearly tactile for the user. Another problem arises if you flip out the LCD monitor. If the monitor is flipped out, you will hardly be able to use the switches.
All broadcast cameras have their XLR-audio inputs on the back below the battery mount. The Ursa Broadcast has them on top of the camera. That is a rather uncomfortable position because everytime you plug some cables into them you end up having the cables in the way. The Ursa doesn't have a uni-slot option for a wireless receiver. If you want to attach a receiver you would have to use one of the mounting options that the Ursa provides. That is the good part. There are a lot of them, 4 on the top handle, 2 on both sides of the shoulder mount and one on the right side of the camera if you don‘t use the handgrip.
Another disadvantage is, that the Ursa Broadcast has big air vents on top of the camera. In all ENG situations, it is possible to have changing weather conditions. If it starts to rain during a shot, I don‘t think that the Ursa Broadcast will like it that much. Other ENG cameras can handle a slight amount of rain very well. Of course, you would put a rain-suit onto the camera if you would shoot in rain all day long, but we‘ve had lots of situations where we shot in short rain situations without that suit. You simply should not do this with the Ursa Broadcast.
Regarding the card slots on the Ursa Broadcast – they are in the right position but it is hard to change the cards while the camera is running because the LCD monitor is so stiff. It is nearly impossible to open the LCD monitor without shaking the camera. Be prepared for that.
A really good thing is the HFR-button on the left side. This button allows you to define another framerate setting than you current project-framerate. If you want to record a special slow-motion shot, you only have to press that button to change the camera settings to high framerate. After the shot, you would press the button again and the camera will go back to your project-framerate. But this function makes not really sense on the Ursa Broadcast because the camera is only capable of doing a max framerate of 60fps. As you will mostly have a project framerate of 50fps, 10fps more is no real advantage for a slow-motion shot. For the Ursa Mini Pro, this makes perfect sense as the camera can record up to 120fps.
Regarding audio: The Ursa Broadcast records only two channels of audio which is a big disadvantage to other ENG-camcorders. Most other cameras allow four channels of audio recording. That makes perfect sense – there are many situations in ENG where you would record a field mixer on channel 1&2 and a stereo ambiance on 3&4. Although it is planned by Blackmagic to activate channel 3&4 on the Ursa Broadcast (there is already a switch to change the monitored channels from 1&2 to 3&4) it is not possible right now to record more than 2 channels.
The menu settings on the Ursa Broadcast are great. All features are easily accessible. The menu structure is really good to understand. You can even save a setup and recall it later – but be aware that all you saved setups are lost if you update your firmware. So please backup the setups before updating. A little drawback is the fact that you cannot shade the camera with the onboard menus. You can only set the white balance and that is it. Other cameras (like the Panasonic) have huge shading and painting menus to set a specific look to your production. The Ursa Broadcast has nothing like this. But you can still do some image tweaking if you use Resolve for that and save a LUT. The camera can apply this LUT to the recorded image so that is possible to set some special painting and shading settings to your recorded clips. But this is somewhat complicated compared to other cameras and you always have to use some external computer equipment to do this.
Speaking of timecode: You can set a preset for the timecode on the Ursa Broadcast. And you can switch the camera to advance timecode only while recording or continuous timecode. Be aware that it is not possible to apply a preset to the continuous timecode. The camera will always use the time of the day for that. So you would have to change the time of the day inside the menu-settings in order to have a different continuous timecode than the time of the day. The camera has a timecode input BNC connector on the right side to sync the camera to other timecode sources. But the camera doesn't have a timecode output! That is a real showstopper to me. We often record interview situations with 3 cameras, where one camera would be the master and the two others are timecode slaves. That is not possible with the Ursa Broadcast because the camera has no timecode output. You would have to bring a timecode generator like from Tentacle Sync or Lockit and some kind of distribution to distribute this timecode signal to all cameras. With other ENG-cameras, it would be a simple loop through. That is a major flaw on the Ursa Broadcast! But the continuous timecode syncs to a provided timecode on the timecode in at least. I didn‘t do any tests on how stable the internal clock of the Ursa Broadcast is. But I read on some forums, that it is not that stable. One user reported a drift of up to one frame per 20 minutes. But this is only something that I've heard. Maybe Blackmagic Design already adressed this. Please drop me an email if you did some tests.
A question that arose during use of the Ursa Broadcast was: What if at some point my Ursa Broadcast will suffer from dead pixels? Regular ENG cameras can mask these dead pixels by performing a black balance. The Ursa doesn‘t have such a function. The Ursa Mini Pro has a function called „calibrate sensor" but this option is not available on the Ursa Broadcast. That is a question that only Blackmagic Design can answer, but I doubt that they will ever do that.
How well do the cameras sit on your shoulder? Well, I think both cameras do it well. A slight advantage goes to the Panasonic. The weight is a little more evenly distributed. But that may be a personal perception. A huge problem is the shoulder mount of the Ursa. It simply doesn‘t fit to all Sony VCU14 plates. We have about 10 different tripods here – all with their own shoulder mount plate. I can slide the Panasonic onto every single plate without any problem. I cannot do this with the Ursa. On some plates, it is not possible to lock the camera. You can‘t push it forward enough to get the plate locked. A problem that is somewhat annoying! Blackmagic should have tested these plates better! We have a similar problem with our Ursa Mini Pro by the way.
Regarding weight: a slight advantage for the Ursa. Although you don't think that this heavy body can weight less than a Panasonic shoulder camera – it can. With lens and 90WH battery it weights less than 6,7kg. The Panasonic is around 7,4kg fully equipped.
Bootup time is good on both cameras. The Panasonic is slightly faster with providing a picture to the viewfinder but takes some more time to figure out if P2 cards are loaded and have space for recording. The Ursa takes a little more time to provide a picture to the viewfinder but is instantly ready to record at this time.
Formatting the cards to get ready for recording is clearly easier and faster on the Ursa. The menu is intuitive and easy to operate. On the Panasonic, you have to go to a different mode to format the P2 cards. Then step through several menus to start formatting. But the advantage of the Panasonic is, that all of your recordings have thumbnails and you can simply select a single recorded shot for viewing. The Ursa has none of these features. You can only watch the recorded clips one after another. No thumbnails – no selecting of a particular clip. Even worst you can only playback clips that were recorded with the same codec and resolution that your Ursa is currently set to. Not very handy. On the Panasonic, it is possible to delete single clips to free up recording space. On the Ursa, this is not possible. What is great on the Ursa is the fact that you can use simple SD-cards for recording. Depending on the speed of your card you can record up to 4k in some of the ProRes formats. Although the higher end formats are to data-throughput heavy for SD cards. If you want to record these HQ-formats or even raw, you have to use C-Fast cards that are way more expensive. And have in mind that you need a lot of them because of the huge data rates that raw recording produces. For broadcast needs, you can still work with SD-cards as you will record HD resolution most of the time. The camera will also warn you if the card is not fast enough during recording and you can configure it to stop if there are dropped frames.
Regarding the power consumption, the Panasonic has a clear advantage. It draws only 34 watts of power while recording (with viewfinder). The Ursa needs 42 watts while recording with a viewfinder. So you will need more batteries. I had a 150WH battery on it while shooting the test footage and it was good for about 2-3 hours of shooting while switching the camera off and on sometimes. I‘ve some experience with the Panasonic and I would say that you can easily get 3-4 hours recording time on that with the same battery. If you consider using an additional headlight on the Ursa, make sure to get some really heavy batteries because your batteries will drain very fast.
We had the Ursa Broadcast for just one weekend to test. So I thought it was good to take it out to some outside conditions in German winter. The temperature was around -2°C and it was quite cloudy. Most of the shots were done using ND2 on the Ursa. I switched off all detail to get an idea of the native sharpness of the sensor. I used the native ISO of ISO400 (0dB) on the camera. I performed a manual white balance before I started shooting. Although I wanted to record in 1080i50 I decided to record in 1080p50 due to the problems I had with moiré. In 1080p50 all the moiré problems went away. You can easily convert 1080p50 to 1080i50 in the post. I hadn‘t enough time to investigate the 1080i50 moiré problem more. Maybe it is normal because of the different scanning of the picture (first lines 1,3,5… than the lines 2,4,6…). But I haven‘t noticed any of these issues on the footage of the Panasonic.
I shot some scenes in our dog obedience school. We have a Rottweiler called „Gandhi" that we train with 2-3 times a week. So I thought that would be some good test footage. The whole shot took about 2 hours and none of the shots was done with a tripod. There were too many different locations to carry the tripod around by myself while simultaneously operating the camera. So please excuse the shaky footage. I did some low light tests on a tripod later that day but I‘ll get to that later.
I used the provided lens from Teltec which was the Fujinon XA20sx8.5BERM-K3.
The overall performance of the camera during the shot was great. The viewfinder is exceptionally good. I would confess that this is one of the best viewfinders I ever worked with regarding sharpness and picture quality. In that special case, you can really experience an advantage to the Panasonic viewfinder which is good (black and white btw.) but not as sharp an crisp as the Blackmagic Design viewfinder. The rubber eyepiece is somewhat loose so that you have to rearrange it on every shot - really annoying. It is even so loose that you can lose it if you carry the camera around and have it touching your legs or anything else in the right way.
What is a little bit confusing is the option to superimpose camera data and some focus and exposure assistance. There are two ways to do that:
#1 you can tell the camera to superimpose this data to the SDI out on the front of the camera. The neat thing about this is that you can use any viewfinder or monitor that you want to display the data.
#2 You can tell the Blackmagic viewfinder to superimpose this data over the image. The advantage using this method is that you can use the buttons on the viewfinder to modify these superimposed data or switch it off. The bad thing is, that it is possible to switch on both – camera superimposing and viewfinder superimposing. So you can end up with some garbage letters and scales on your viewfinder that overlap each other. So make sure you only use one of the described methods.
One thing about the source of this data: The origin of this superimposed viewfinder data is the camera that embeds some of the information in the data section of the SDI stream. The viewfinder decodes that and superimposes this data onto the image. You should have that in mind because you won‘t find any setting for zebra in the viewfinder settings. This setting is done in the camera menu which is confusing as you can set the peaking level and color inside the viewfinder menu. So make sure you get used to this camera and know where all the settings are before going to a shoot. The policy of Blackmagic regarding a clear and homogeneous workflow is not as good as the one from other well-known brands like Sony or Panasonic in my opinion.
A handy feature of the viewfinder is the ability to assign a button to the zoom-in feature. With this function enabled the viewfinder zooms into the picture to get a better indication of focus.
But this function is really dangerous too. There is absolutely no indication that the viewfinder is in zoomed-in state. And what makes it even worst is that the viewfinder doesn‘t switch back to default view (not zoomed in), when the camera is switched off and on again. Even if you power cycle the camera the viewfinder remains in the zoomed-in state with no indication that this is the case. That leads to the problem, that a lot of my footage wasn‘t usable due to wrong framing. I simply saw another picture in the viewfinder (zoomed-in) than the camera recorded. This must be addressed by Blackmagic Design. The has to be a big blinking warning when the viewfinder is zoomed in. So be warned if you use this feature!
Another little glitch was a bug in the viewfinder software. Although the camera was set to record 1080p50, the superimposed data in the viewfinder showed me a framerate of 24p. The LCD screen of the Ursa still showed the right framerate of 50p nevertheless. Even after power-cycling the camera, this glitch remained persistent. Finally, after switching the framerate of the camera to something different and then back to 50p, the viewfinder showed the right framerate of 50p again.
Besides that, it was really fun to shoot with this camera. Everything worked well the whole day through. Even in the cold environment, there wasn‘t any problem with it. It was a little difficult for me to start the camera when not having it on my shoulder – for example on bottom view shots, where I held the camera between my legs. The Panasonic has a nice little record button on the top handle. The Ursa Broadcast doesn‘t. The record button on the side should have better haptics that you can identify this button at once without having a look at it. But that may be something that you get used to if you use the camera a lot.
A big drawback is the fact that the Ursa Broadcast only has only one memory for the white balance. The Panasonic has two plus a preset setting if everything has to be very fast and there is no time for a manual white balance. This is really a problem in ENG situations. Especially for reality documentaries where you would eventually walk from an outside scene into a tungsten lit scene with a running camera. You cannot do this with the Ursa because you can‘t switch to a second white balance memory – there simply is none. You can indeed switch to a preset but for that you have to open the side LCD and tap on the „tint" or „white-balance" info on the LCD screen. There you can change the white balance to some established presets like tungsten, fluorescence lights, sunlight, cloudy and some more. You can even dial in a manual white balance by adjusting the temperature and the tint. But in real-world situations, this is a problem. You simply don‘t have the time in ENG situations to flip out the screen, go into the menu and change the settings. On all other ENG camcorders, the presets are instantly accessible by a button or switch on the side.
There is indeed a semi workaround for that. There are two (mostly) freely configurable function switches on the side of the Ursa Broadcast. You can assign a preconfigured white-balance setting to them. So you could put a tungsten preset on f1 and a daylight preset on f2. If you have enough time and you have to do a shot that goes from outside daylight to inside tungsten in one shot, you could do a manual white-balance in both situations and write down the temperature and tint settings. Then you would dial in these settings into the presets of f1 and f2. When you did this you can perform a white-balance change while recording. But be aware that this is no smooth transition but a hard cut from one preset to another. You could also try to change the temperature by using the white-balance switch. But during a shot, it is simply impractical to do the framing, keep the focus and have one hand on a switch to change the temperature by holding the switch down until the temperature increased or decreased to the desired level. And the other bad thing about this method is that you can only change the color temperature, not the tint. So if you plan to do a transition between tungsten and neon-light this workflow is not an option. Maybe Blackmagic Design can change those shortcomings with a future firmware update.
The iris control is another issue where Blackmagic is not quite ready. On the Panasonic, you can switch the iris to automatic mode or push a momentary button to adjust the iris. On the Ursa Broadcast, you have nothing like that. There is simply no automatic iris mode. Even if you switch the lens to the auto iris, nothing will happen. In this mode, you can control the iris manually by twisting the knob on the side of the camera which makes no sense. Because if you would want to control the iris manually, you would just turn the iris ring on the lens. Although I must admit that you wouldn‘t use auto-iris in most of all ENG situations anyway. After the test I discovered a method on our Ursa Mini Pro to get a momentary auto iris: You set the iris to "auto" and push the iris button on the left side of the camera that is located under the LCD display. Then the camera will perform a momentary auto-iris. But it is not very accurate and you can't reach the button if the camera is on your shoulder.
Audio is really not the best part of the Ursa Broadcast. There are only two XLR sockets on top of the camera – so you have only two audio inputs. The Panasonic has four. There is a build in microphone for ambience sound recording. The mic is not very good and has a lot of issues if there is some wind, you can use it in some cases but just for ambience recording. If you want a better option you can attach a microphone holder to the camera and use your own mic. But have in mind that you will lose one of the two XLR inputs for that and the camera will be somewhat bulky with the external microphone attached and cables running to the upper side of the camera. Another shortcoming is, that the Ursa Broadcast has no option for automatic audio levels. This option is very useful for the ambience mic recording. In my shooting, I had more than one situation where one of the dogs barked so loud that the recorded ambience was totally distorted. Obviously, the camera doesn‘t react well on to hot audio. There is no limiter too, so you have to be very conservative on audio levels.
The sound quality is not really equal to the Panasonic. An attached Sennheiser MKH60 shotgun microphone sounds not as good on the Ursa Broadcast than on the Panasonic. It seems that there is a lowcut filter applied. All the low frequencies are missing. You can fix a lot of this issue in the post – but who wants that and who has the time for this in the ENG-business. It‘s not that bad but you can clearly hear the difference. Although I didn‘t try to switch the XLR inputs of the camera to line and use a field mixer. That is the common workflow in ENG. Maybe the line inputs sound better than the microphone inputs. I didn‘t have the time to test this. As a third input option, you have a digital AES/EBU input. If you have a field mixer that has an AES/EBU digital output, this should be the best option to maintain sound quality. Right now it is only possible to record two audio tracks. Maybe Blackmagic Design adds four-track recording in a future firmware update. There are already some switches for monitoring track 3 and 4 on the camera. But the limitation of 2 physical XLR inputs is still there – even if you have the option to record 4 tracks of audio.
I must admit that the overall picture quality of the Ursa Broadcast is great. The details are fascinating even if there is no additional detail applied in the processing of the camera. The picture has a very natural feel and is very pleasing for a camera with this price factor. The colors are rich and beautiful. Skin tones are very natural and simply look great. Even if you overexpose the skin tones they don‘t look as nasty as an overexposed skin on the Panasonic. The highlight roll-off could be a little better although that isn‘t really an issue in ENG situations. There is no knee function on the Ursa Broadcast but you can set the camera to extended dynamic range in the menu and have a similar effect. The dynamic range turns out to be really good. Blackmagic claims that the camera has a dynamic range of 12 stops. If you set the camera to extended dynamic range (which I did during the whole shot) you can really save some shots that would be otherwise over- or underexposed in some areas of the picture. Clearly an advantage of the Ursa Broadcast. The sensitivity seems to be slightly lower than the Panasonic on the same gain setting of 0dB. I think it is only half a stop. But the good news is you won‘t lose as much detail and add noise if you apply gain than on the Panasonic. More to the sensitivity in the low light test section.
Regarding rolling shutter – I couldn't produce any real-life situations that cause visible rolling shutter artifacts. It seems that the readout time of the sensor is really fast. I did, in fact, some extreme panning to test this. At 30% speed, you can see that there are some yello-issues, but they are not that bad compared to other cameras. The Panasonic doesn‘t suffer from rolling shutter at all because it has CCDs thus a global shutter. But the newer Panasonics and even the Sony ENG camcorders all have CMOS sensors and rolling shutter issues.
Just for the record – The Ursa can record in DNxHD or Prores (I used Prores 422 for my tests). The Panasonic can record DVCPro and AVC-Intra. I used AVCIntra 100 for my tests.
I performed a low-light test with both cameras in my living room. The room was dimly lit by some halogen tungsten lights above the table. I recorded clips in all gain modes of both cameras. I only used a maximum of +12dB on the Panasonic although the HPX3100 is capable of much higher gain settings. I did this because the Ursa Broadcast maxes ot at +12dB gain. I just recorded some seconds in each gain setting and you can hear me on the audio track telling the current gain setting of the camera.
To be honest – the Ursa Broadcast is no low-light monster. The Ursa broadcast has a native ISO of 400. If you compare the recordings you will find that the Panasonic performs slightly better. I would say that the Ursa Broadcast has about half a stop lower sensitivity. This could be due to the fact that the B4 lens mount of the Ursa Broadcast has a lens in it that spreads the image circle of the B4 lens in a way that it covers the 1inch sensor of the camera. This means you will lose some light during this process.
Both cameras are really close to each other regarding the picture quality. But If you take a closer look at the images you will see that the Ursa Broadcast performs much better. This is a real surprise! If you zoom onto the test chart on the table you can see that the Panasonic produces a lot of noise and compression artifacts even at 0dB gain. The Ursa Broadcast is much cleaner and has a lot more detail while having visually no compression artifacts. The noise of the Ursa Broadcast has such a fine structure, that is doesn‘t disturb you at all. The Panasonic, on the contrary, has a very noticeable noise with big blobs and resolution loss. It isn‘t really noticeable if you don‘t zoom in but the overall picture looks a little bit more uneasy on a professionally calibrated HD-monitor. The Panasonic suffers a little bit from detail sharpening in the camera. I didn‘t turn it off because I simply forgot it. That makes the picture look not as natural as the picture of the Ursa Broadcast. The Ursa has a lot of natural detail (I turned off all the detail sharpening in the menu) and a very pleasing noise print. If you switch the Ursa to +6dB gain the noise is absolutely acceptable and now it has more sensitivity than the Panasonic. Even at +6dB, I would say that the noise is still not as bad as the Panasonic at 0dB gain. If you switch the Panasonic to +6dB gain, you can clearly see the pronounced noise – even if you look at the whole picture, not zoomed in. With +6dB on the Panasonic, you reach the absolute edge of what I would use for any production. If you switch the Panasonic to +12dB the picture breaks apart totally. A lot of visible noise all over the picture and the resolution drops to a point, where I don‘t find it acceptable anymore. +12dB on the Ursa Broadcast is still quite good. The noise is a little bit more visible but by far not as bad as on the Panasonic. And even the resolution is good and absolutely usable. Allthough the Panasonic looks way overexposed on the +12dB test and the Ursa does not, this may be because the the Ursa was set to "extended dynamic range".
To sum it up: the Ursa has about half a stop of lower sensitivity but performs much better with any of the available gain settings. The Ursa, to my surprise, is the absolute winner regarding the picture quality.
I extended the test by adding another camera. You can now compare the Ursa Broadcast to the Ursa Mini Pro. Some facts to this camera: It has a bigger S35mm sensor. If you use a B4 lens with the optional B4 mount (what I did), you can only record a resolution of 1920x1080 with the sensor set to windowed. In this mode only a fraction of the sensor is used. That means that the effective sensor size gets smaller. But you also lose resolution (from 4k to HD). The Ursa Mini Pro has a native ISO of 800. If you compare the images you can see that the picture of the Ursa Mini Pro looks brighter at the same ISO setting than the picture of the Ursa Broadcast. The reason for this should be the B4 lens mount and the different sensor. On the Ursa Broadcast the B4 lens mount has a glass inside that spreads the image circle in a way that it matches the sensor size of 1 inch. That introduces a loss of light. I would say that the difference between both cameras is about half a stop/1 stop. Please note that I had not time to mount a proper HD lens to the Ursa Mini Pro. I only had a SD lens available. I think the images of the Ursa Mini Pro would have looked sharper if I‘d have used a HD lens. I switched off all additional detail in the Ursa Mini Pro – exactly as on the Ursa Broadcast. If you want to download the original files, you can find them in the download section.
Panasonic has an excellent repair policy. We had HPX3000 failing last year and we contacted Panasonic support to get it repaired. The camera was 4 years out of warranty but Panasonic told us that they will repair the camera for free. All we had to pay was the shipping. The whole time from shipping to repair and get it back was about one week. Great service and really fast!
In contrary Blackmagic Design did not perform as good. We once had a defective Blackmagic Cinema Camera where the sensor suddenly produced a garbage picture in all rainbow colors in the middle of a production. The camera was still under warranty so they exchanged it. But this process took more than 3 weeks. Way too long for any professional production situation. And it gets even worse: If the camera wouldn‘t have been under warranty, Blackmagic maybe would have refused the repair of the camera. I've read lots of forum-posts, where Blackmagic gear fails, is out of warranty and Blackmagic Design refuses to repair it. This is simply a matter of costs. A good repair policy and infrastructure is expensive for any company. The Blackmagic products are really cheap. This price point has one of its origins in that fact. So better be prepared to buy a new camera when you're out of warranty camera fails. I must admit that this would hurt not as much as with a Panasonic and 20k Euro and more.
To sum it up:
The Ursa Broadcast is a really great camera for ENG situations. You can work with it in most of all cases when you are willing to change the proved and learned workflow with other ENG cams. There is much light, but also a lot of shadow that comes with it. For the price point of about 3k Euro, it is an absolute bargain. But have in mind that there may be no real support from Blackmagic Design if your camera fails and is out of warranty. The same probably applies to problems with dead pixels. Time will tell. Especially if you use the camera for live-multicamera situations, the Ursa Broadcast is a fantastic investment because you get everything in one box: camera, tally, intercom, and of course a software CCU with all shading options via the ATEM software. And I must mention the price point again. There is simply no other vendor that offers a camera with this amount of features and picture quality for this price.
If you want to have a camera that is absolutely reliable in all situations, offers perfect ergonomics and won‘t disappoint you at any time you should invest in a Sony or Panasonic ENG camera. In particular, if you do only ENG stuff and no live-multicamera productions. These cameras are completely ready. There are no hiccups or missing features. You get great service from the established companys if anything fails on your camera. They are pure workhorses but cost 6 to 8 times more than the Ursa. You have to spend a lot more time working, to get this camera paid off.
-low noise in any gain setting
-sd card recording
-bad microphone audio quality and only 2 audio tracks
-missing timecode out
-tripod mount doesn‘t fit all plates
-only one white-balance memory
-no auto-iris function
-Blackmagic Design has bad out of warrenty repair reputation
-bad file management (review of clips only one after another)
-perfect form factor
-absolutely ergonomic in any way
-all inputs and outputs that you ever need
-professional sound quality on up to 4 tracks
-reliable for years and years
-Panasonic has a fantastic repair policy
-good clip management
-low power consumption
-bad picture quality compared to the Ursa
-Much more noise / not acceptable at +12dB gain
-codec adds a lot of artifacts compared to the Ursa
Here are some download links to the original footage that I shot to compare it for yourself. I even recorded a short raw file with the Ursa Broadcast in the Low Light Test situation. Copy these links into a new browser window or rightclick the link and selsct "open in new frame/window" to get the file downloaded.
HPX3100 Original P2 Footage of Low Light Test:
Ursa Mini Pro Low Light Test Files:
Ursa Mini Pro Testchart ISO run
Ursa Broadcast Low Light Test Files:
Ursa Broadcast Rolling Shutter Test:
Ursa Broadcast RAW Low Light Test shot:
Ursa Broadcast Real-Life Footage: