AJA Cion Redshark Review

At last reviews are coming through. That sounded positive, and the stills on the review look very nice. If they can get 120 fps 4k going - that would be a real biggie.
Steve
 
Andrius Simutis said:
The Cinema5D guys tested the DR and found it to only be 8.2 stops.
Click to expand...
Well, the Cinema5D guys have their own method of testing DR. So we don't know whether it is right or not. Let them also use the same methodology to test the Alexa to tell us what its DR is. And also they should tell us the DR numbers of other cameras, so that we can have some context. Else, it is just a number and doesn't make sense. May be Cinema5D has done a few tests and the dynamic range of various cameras are there in other posts. I am not aware. If some one knows, then please share.
 
indiawilds said:
Well, the Cinema5D guys have their own method of testing DR. So we don't know whether it is right or not. Let them also use the same methodology to test the Alexa to tell us what its DR is. And also they should tell us the DR numbers of other cameras, so that we can have some context. Else, it is just a number and doesn't make sense. May be Cinema5D has done a few tests and the dynamic range of various cameras are there in other posts. I am not aware. If some one knows, then please share.
Click to expand...

http://www.4kshooters.net/2014/07/2...canon-5d-mark-iii-c300-1dc-and-panasonic-gh4/

I don't understand how their 5D3 tested higher than the C300 for DR.
 
Cinema5D is giving what they call the "usable" dynamic range, which is lower than what is normally told. There is no such thing as "absolute" dynamic range. While you can clip the highlights and say, "that's exactly as far as it goes," the shadows descend into ever-increasing noise, and who's to say how much noise is okay? And noise is always present, even in the highlights, so you can't say, "I'd like the number to be when there is zero noise."

Even so, there is an international standard for measuring dynamic range, ISO 15739. Traditionally the makers of sensors and cameras seem to be using this standard. It says the lower end is the point where the noise and the signal are equal. This may be a higher figure than any of us are willing to use. But at least it is a standard.

In fact I call into question the usefulness of usable dynamic range. Most of us don't go out measuring the dynamic range of scenes and then look for a camera. Instead most of us, I would say, have looked at sample footage of various cameras, then looked up the dynamic range of the ones we liked. We don't go out with a high-end reflected light meter: "Well, I'm going to shoot Dave here in a black shirt under this tree, and Sally in a white shirt in sunlight, so lets see . . . that's about 11 stops. I need a camera with 11 stops!" Instead, it's more like, "Whoa, the Alexa looks awesome! What's it rated at? 14 stops. Okay, what has 14 stops that's in my price range. . . Nothing. Okay, well, the Canon 5D looks pretty good, and it has 11.7 stops. So, maybe I can get by with a camera that has 11 or so stops."

In fact, if any of us did measure the dynamic range of scenes we intended to shoot, we would see that we almost never need 14 stops. The aforementioned example shot, with someone in the shade in a black shirt and someone in the sun with a white shirt, is 8-10 stops difference, depending on how thick the shade is (from my crude tests. Please chime in if you come up with something different). And that's an extreme example. On an overcast day, there is just about a stop difference between sun and shade, and there is always just a 5- or 6-stop difference between black and white in the same light. So, 6-7 stops, for overcast days (if my tests weren't flawed).

Anyway, I would prefer we just stick to the ISO standard, instead of trying to be "helpful" by interjecting "usable" dynamic range, which is subjective, unregulated, and unstandardized. Now we have two numbers (at least) for every camera, and we will have to make it clear which number we're talking about. It just causes confusion.
 
Last edited:
I find the Cion price point interesting. It's hard to compare it to other cameras since it falls in between the Black Magic Production cam which is around 6,000 USD and the Canon C500 which is around 16,000 USD.
 
indiawilds said:
Well, the Cinema5D guys have their own method of testing DR. So we don't know whether it is right or not. Let them also use the same methodology to test the Alexa to tell us what its DR is. And also they should tell us the DR numbers of other cameras, so that we can have some context. Else, it is just a number and doesn't make sense. May be Cinema5D has done a few tests and the dynamic range of various cameras are there in other posts. I am not aware. If some one knows, then please share.
Click to expand...
Yeah, the C5D method using Imatest software isn't the same as the manufacturer's method, but I think it's a much better representation of what a camera is actually going to give you in the real world vs a brochure. The Cion was the lowest of any camera they've tested at 8.2 while the Arri Amira was 13.1, FS7 12.4, and "most Canon cameras come in at around 10-11 stops". That's a big difference. It's also not to say that the camera overall isn't any good, just that the DR portion of it isn't as good as we'd hoped. The sample shot in the Redshark review seems to show that clipping with the blown out highlights in the window.
 
It's the low ISO that is the big issue with this camera.......it takes great pictures, a great, massive signal recorded and yes, lots of room to play with in post....but our clients have all said..."320 to 800 max ISO!......the fact is, you must look at this camera as a "Film Camera" in terms of lighting... a film camera using 400 film means you need the appropriate amount of lights.

Jim Martin
 
Last edited:
Andrius Simutis said:
Yeah, the C5D method using Imatest software isn't the same as the manufacturer's method, but I think it's a much better representation of what a camera is actually going to give you in the real world vs a brochure. The Cion was the lowest of any camera they've tested at 8.2 while the Arri Amira was 13.1, FS7 12.4, and "most Canon cameras come in at around 10-11 stops".
Click to expand...

Do you have links to their other cameras dynamic range tests?
 
Played with one today. Over 40sec boot up time and was told this won't improve. Looked more like 9 stops at best. 320asa. 1000 is possible but it automatically changes gamma to hide noise (seemed very strange). No Log mode but did have an "expanded mode". In hard sun I felt like it was hard to not get video/DSLR looking images. No built in ND's seems strange. Like I guessed, front heavy with a small prime; the pad needs to be forward 3 inches or more. 40+ watt draw so big batteries with small run times. Seemed built well and was light but not sure I understand where this fits in.
 
I saw that. The facts are that - of course - it's not a paid advertisement. (It's Dave, the editor of RedShark here.) When we do carry "Advertorial" we clearly label it with something like "Sponsored Article", and even then, the have to comply with our normal editorial guidelines.

The seemingly awkward fact is that Peter Haas, who looked at the camera for us, liked it. We will have it tested by other writers as well, as we normally do with important cameras, to get other viewpoints. The only reason we haven't done this yet is that review models are hard to get hold of.

Personally, I like it as well. It's clear that its output is designed to go straight to ProRes and will therefore look very contrasty. It's not a low-light camera by any means, but sometimes you have to spend a bit of time with lighting to make a great movie: there's nothing new about that.

There's nothing wrong with Cinema 5D's testing I assume (I know those guys and they are very good at what they do) but if there is an issue with rigid testing it's that it can paint a unidimensional picture of a camera that - despite the test results - is actually rather good. I would also say that you should NEVER judge a camera from how it looks on YouTube or Vimeo.

With cameras in this price bracket, you really need to test them yourselves to see if they're right for you. We can only tell you what the people that we ask to try the camera think about it. And that's what Peter thought.
 
I signed up for the TryCion program, but I haven't heard any news about it, so I don't know whether or not I'm going to get one to test out. I'm hoping so, because it would be interesting see what it can do. Other than dynamic range, it seems like AJA nailed just about everything else, in that it's actually quite easy to use. The menu system feels a bit odd at first, but even just playing with it at NAB, it didn't take long to get used to.
 
I've also signed up for the TryCion program - Hoping to get a week with it to try it in everything from corporate, to a wedding (not that its meant for that, but hell, why not see how it does!!)

AJA was in Toronto last week with a Cion demo. Looks very nice! They seem to want to push the idea of 'get a good image straight out of camera'. It does feel very balanced on your shoulder, and does feel nice.

Pros
Tons of outputs (4+ monitors @ 1080 and an EVF all at the same time, with p-tap on top as well)
Quick offload time (thunderbolt I believe -> 5 gigs in a 5-7 seconds)
Can control it with Ethernet and a laptop with any browser.
Menu system is very straight forward. 4 Options I believe.. Their guy highlighted you'd never see an "Other" under the menu for "stuff they couldn't decide where to put"
If using a laptop you can group several cameras together and control them all at the same time (start/stop record, feed all the same pic profile info, etc etc).
2x XLR and a proper volume knob for headphone monitoring!


Quirks (possible cons depending how you look at it)
Comes with PL Mount, you can get the Canon EF mount, but it is a passive mount -> no electronic communication, so CN-E, CP2, Rokinon, Other Cine lenses etc are good to go. Out of luck for any Canon L glass / non-manual lenses.
No Weather sealing (not significant, but good to know)
Definitely need an EVF or monitor (side screen or 'confidence monitor' doesn't provide much confidence, hah).
8 stops seems to be what their tech was saying for most profiles (unless going to expanded, which can max out at 12... but if you go to the max ASA of 1000, expanded gamma is locked out, so you can't get 12 anyway).
3 white balances and AWB -> Possibly adding in custom WB with firmware in the future, but they couldn't say for sure. Maybe I'm too used to the C100, but I'd want control myself.


They did a quick demo recovering an underexposed shot (recordered 4444 if I remember correctly), extremely good results. Heres hoping to try one out for a week!
 
You must log in or register to reply here.
Back
Top