Zacuto Camera Shootout 2018 : Panasonic EVA1 vs Canon C200

Mitch Gross said:
I'll point out a couple of things.

I feel that EVA1 has greater information in the shadows. There is simply more detail before the image collapses. You could say that's 10-bit but I find it true compared to the Canon RAW-lite as well, so there's that.

EVA1 colors are more accurate. In the interior location the walls were a neutral gray tone, perhaps leaning a touch to the blue in daylight. This is rendered quite accurately by the EVA1, but I did not find that at all with the C200. Those colors were just wrong.

EVA1 has greater accuracy in the reds. Look at the 400x images in the interior. Ours looks right. The C200 looked ... otherwise.

EVA1 has far greater color information in highlights. On the day exterior the sunlit buildings nearly went white on the C200 while the EVA1 correctly showed their colors. Likewise, in the night exteriors the colored highlights on buildings had far greater (and much more accurate) saturation than those on the C200. This was made clear in the magnified views as well.

These are only some of the examples and differences, but I think they are telling ones. You can get a pleasing image out of either camera, but if you wish to combine that with accuracy and with not having to struggle in post to create colors or power window like crazy, then I liked what I saw n the EVA1.
Click to expand...

It is a shame that simple testing did not occur to substantiate your "feelings"

Grey - shoot a grey card (!)
Colours- shoot a chart
Shadow noise - shoot a chart

For shadow my personal preference is to shoot a side lit cylinder (large) a test setup that examiners tonality in the shadow area

Of course "better in the shadows" needs careful consideration as one must accurately pin the exposure to the clip point and with a non linear file the top end "knee" may make absolute clip a complex variable as close to clip might be poor quality data
 
Last edited:
morgan_moore said:
Of course "better in the shadows" needs careful consideration as one must accurately pin the exposure to the clip point and with a non linear file the top end "knee" may make absolute clip a complex variable as close to clip might be poor quality data
Click to expand...

Yo man, unless you be Science-man, yo might want to spress yoself diffantly.
 
morgan_moore said:
It is a shame that simple testing did not occur to substantiate your "feelings"

Grey - shoot a grey card (!)
Colours- shoot a chart
Shadow noise - shoot a chart

For shadow my personal preference is to shoot a side lit cylinder (large) a test setup that examiners tonality in the shadow area

Of course "better in the shadows" needs careful consideration as one must accurately pin the exposure to the clip point and with a non linear file the top end "knee" may make absolute clip a complex variable as close to clip might be poor quality data
Click to expand...

Bruce was adamant that he did not want to shoot charts. He wanted to shoot real world scenes and judge based on real world issues.
 
filmguy123 said:
It is. But it's not so bad once you understand the nature of the beast. 5.7k full sensor just isn't possible, that's a ton of data. So, you've got to pull less data hence using the 4K resolution (which on most S35 cameras would mean no crop). But given the 5.7k, you have to crop in to the 4k which equates a 4/3 crop. The 2K is pulled off of the same 4K using a MIX mode so it does not incur any further crop. 2K RAW @ 240fps with a 4/3 FOV is pretty solid.

The 5.7k, for all the super high crisp detail it renders, does come with some drawbacks and that's one of them. Shooting raw at wide angle on EVA1 will be rough, especially if you want to shoot wide AND fast. But for anything more standard focal length it's a great option.
Click to expand...

This doesn't appear entirely clear as written, so let me express what is going on with the RAW output.

5.7K up to 30p. This is the data limit of what can be passed through a 6G-SDI connection, which is how EVA1 outputs RAW. 5.7K is the native resolution of the Super-35 fame.

4K up to 60p. 5.7K is almost twice as many photosites as 4K, so the EVA1 can output twice as many frames per second in 4K at the same data rate. So the camera will output native 4K off the sensor, which is a crop to approximately the 4/3" image area.

2K up to 240p. Once again, 2K is 1/4 the data of 4K, so the EVA1 can output four times the frame rate. Rather than have the camera crop to a tiny part of the sensor, the engineers came up with a clever solution. A standard cluster of photosites on a Bayer-pattern sensor is top line RED / GREEN, bottom line GREEN / BLUE. Take four of these clusters adjacent to one another (two top, two bottom), take the values of each corresponding photosite and average them to create one "virtual cluster." By doing this, a 4K RAW image becomes a 2K RAW image while retaining the same field of view. So the 2K RAW out of the EVA1 will have the same 4/3" image area as the 4K RAW.
 
Mitch Gross said:
Bruce was adamant that he did not want to shoot charts. He wanted to shoot real world scenes and judge based on real world issues.
Click to expand...

Mitch Im on you side here - like I said your 'feelings' need to be clearly demonstrated.

I dont respect Bruce's standpoint however long his IMDB arm is. It is the classic position of an ex chemical filmer who has not really 'got' digital.

I would certainly want to shoot some locked down scenes at least without clouds bouncing in an out and an undefined random approach to highlights shadows and the rest.

For example the bridge daylight had no consistent approach.

There was a skyline that looked to me like it should have been blue, but I have not idea.
Fair play dont do charts but at least make a consistent approach..
'we are exposing for the sky and seeing how well the characters in shadow hold'
'we are placing the characters a stop under and then will examine how the sky holds'
'Im trying to make a beautiful image and theses are the compromises I hit at these Tstops'
we then need to hear the colourist commentat while he tries to deliver the DPs intent

To me shooting some cards and a DR wedge is like a 10min part of the testing day, same with a locked down DR test - one of the interns could have done it before the royalty showed up from the cafe.

You do that basic ABC123 testing then you go play and try and discover some mojo. But to not do the basics make it feel like as I said, an old schooler back sratching session leaving me with no desire to buy either camera.

As it happens I would be most interested in upgrading my FS7 becuuse I think the codec is a little thin and dont like its raw mess (im therefore your target audience?) but cannot work out if these cameras bring anything much new to that sub $10k party.
 
Last edited:
yoclay said:
Yo man, unless you be Science-man, yo might want to spress yoself diffantly.
Click to expand...

I did study physics and engineering before moving to images but to me cinematography is a mix of science and art, testing a camera is the science bit, making emotionally valuable material is the art bit.

As it happens I beleive well conceived testing (usually of blowing highlights and digging in to shadows) can reveal 'mojo'

Edit stumbled on this
Yeldin
https://m.youtube.com/watch?v=A4F29I27oTM
 
Last edited:
Mitch Gross said:
This doesn't appear entirely clear as written, so let me express what is going on with the RAW output.

5.7K up to 30p. This is the data limit of what can be passed through a 6G-SDI connection, which is how EVA1 outputs RAW. 5.7K is the native resolution of the Super-35 fame.
Click to expand...

Hi Mitch,
I own fs700 + odyssey 7q..
I have a doubt .. how is it possible to Sony pass 12 bits of 4k 50p data over a 3G HDSDi cable to Odyssey with no crop and not to Panasonic after 3 years later?

Panasonic have a 6G .. why output at 50/60p 4k in cropped mode? how many bit?

This thing is very important for me because the fs700 + odyssey is a great combo also in 2018 considering the other brand.. i'm quite confusing..

can you help me to understand?

thank you very much!
 
imax said:
I have a doubt .. how is it possible to Sony pass 12 bits of 4k 50p data over a 3G HDSDi cable to Odyssey with no crop and not to Panasonic after 3 years later?
Click to expand...
Because the Sony's output is compressed. Panasonic's is uncompressed.

Panasonic have a 6G .. why output at 50/60p 4k in cropped mode? how many bit?
Click to expand...
It would need 12G to output 60P with no crop. Or, it would have to be compressed.
 
morgan_moore said:
Mitch Im on you side here - like I said your 'feelings' need to be clearly demonstrated.

I dont respect Bruce's standpoint however long his IMDB arm is. It is the classic position of an ex chemical filmer who has not really 'got' digital.

I would certainly want to shoot some locked down scenes at least without clouds bouncing in an out and an undefined random approach to highlights shadows and the rest.

For example the bridge daylight had no consistent approach.

There was a skyline that looked to me like it should have been blue, but I have not idea.
Fair play dont do charts but at least make a consistent approach..
'we are exposing for the sky and seeing how well the characters in shadow hold'
'we are placing the characters a stop under and then will examine how the sky holds'
'Im trying to make a beautiful image and theses are the compromises I hit at these Tstops'
we then need to hear the colourist commentat while he tries to deliver the DPs intent

To me shooting some cards and a DR wedge is like a 10min part of the testing day, same with a locked down DR test - one of the interns could have done it before the royalty showed up from the cafe.

You do that basic ABC123 testing then you go play and try and discover some mojo. But to not do the basics make it feel like as I said, an old schooler back sratching session leaving me with no desire to buy either camera.

As it happens I would be most interested in upgrading my FS7 becuuse I think the codec is a little thin and dont like its raw mess (im therefore your target audience?) but cannot work out if these cameras bring anything much new to that sub $10k party.
Click to expand...

I understand your points and I would say that this isn't the test for you. Fortunately there are a lot of other tests out there. Plus you could always arrange to go to a sales house to test one out yourself. I'm sure they exist and it's likely worth the trip for such an important purchase. No test performed by someone else will ever show you as much as the one you would do for yourself.
 
imax said:
Hi Mitch,
I own fs700 + odyssey 7q..
I have a doubt .. how is it possible to Sony pass 12 bits of 4k 50p data over a 3G HDSDi cable to Odyssey with no crop and not to Panasonic after 3 years later?

Panasonic have a 6G .. why output at 50/60p 4k in cropped mode? how many bit?

This thing is very important for me because the fs700 + odyssey is a great combo also in 2018 considering the other brand.. i'm quite confusing..

can you help me to understand?

thank you very much!
Click to expand...

The FS700 RAW signal is compressed, I believe between a 4:1 or 5:1.
 
Mitch Gross said:
The FS700 RAW signal is compressed, I believe between a 4:1 or 5:1.
Click to expand...

ok thank you! i din't know it! anyway is a great signal and amazyng job from CD.
I'll hope that Panasonic will open to CD or other vendor to maximize the performance of this great little camera.

thank you! ;)
 
Mitch Gross said:
Oh, you mean the camera that required a huge amount of adjustment to the lighting and tons of power windows in post? The one that Francis Coppola liked but a lot of others didn't? The one that didn't have the depth of field characteristics of Super-35 (neither did the iPhone!!)? Of course there was another camera on that same test where the operator walked into the base lighting setup and said, "yeah, our camera can handle that." That camera shot the scene in about 1/10th the time and the grade was super fast as well. So even though it was the most expensive camera on the stage that day it was easily the most affordable given the minimal lighting requirements, the minimal lighting time (zero!) and the short time spent in color correction. But yeah, that little GH2 looked nice when you worked it well.

Tools matter. If you want to cut off your nose in spite of your face then that's fine, but don't fool yourself into believing you're doing yourself any favors in the process.
Click to expand...

THIS! All the way.
 
yoclay said:
Extremely well put Mitch. My sentiments exactly. Could not have stated it better.

As someone who does a lot of grading for others, you have no idea how many times I cringe when I hear "It's a mini-Alexa at a tenth the price".
Click to expand...

This is probably the biggest BS statement going around for the last few years. I think it started with the first blackmagic camera. Wishful thinking and delusion at its strongest! There's only one mini-Alexa and that is the Alexa Mini.
 
Run&Gun said:
You kids, get off my lawn!

I cut my teeth on Betacam. Shot it from the beginning of my internship in 1996(bought my first one in '97) until probably at least 2012- nine years after I had started shooting HD and over six after I had bought my first VariCam.

Yep, B&W CRT VF's. You really had to know what your camera would do under different lighting conditions. And we rarely had incorrect white balances and exposure issues. I actually didn't get my first color VF until after I bought my first P2 Vari. Which is why it baffles me today as to why I see so much poorly exposed and white balanced video since almost every camera has a color VF or monitor attached.
Click to expand...

The reason is most people can afford cameras today. Even those who can't shoot. :laugh:
 
morgan_moore said:
Its a start, but no concept of DR? (or Mojo)
Click to expand...

No mojo tests there, that's for sure. The Zacuto shootout might help with that.

But for DR, you have over-under exposure tests, which is exactly the kind of experiment I'd use to measure dynamic range:
- All cameras look good at ETTR.
- All cameras look good in the bright side of the face at ETTR-4, but on the shadow side only the Helium, BM4.6k and Terra6k are really OK, the FS7 has little noise but is soft because of the amount of NR already applied in-camera, Varicam, EVA1 and C200 look kind-of-OK but with some noise (with maybe the C200 slightly behind the others), and the GH5 is already pretty noisy.
- And ETTR-6 on the bright side of the face confirms those previous results and helps differenciate at the top a bit more: only the Helium sensor is keeping up, BM4.6k and Terra6k don't show much noise but have a bad color shift, FS7 is also clean but only because it seems to have heavy NR already applied in-camera, the Varicam, EVA1 and C200 seem to be just good enough that some NR would save them, and the GH5 has tons of noise.

So, in terms of DR, it tells me that Helium > BM4.6k = Terra6k > Varicam = EVA1 >= C200 = FS7 > GH5. I'm guessing there that if you apply NR to the C200 you'll end up with something relatively similar to the FS7 image. If I had the full range from ETTR-3 to ETTR-6 I'd be able to say exactly how many stops each difference is.

To me, the winners there are first the Helium sensor, of course, and then the BM4.6k, looking great at a much more affordable price, and with a good balance of DR, colors, and detail (to me, it shows just the right amount of skin texture). And the loser *for me* would be the FS7: I'm usually a Sony fanboy, but I had never noticed how much noise reduction is happening in the shadows.
 
Last edited:
BTW, viewing the results of this shootout on different monitors ( Samsung S8 phone, Dell 27 inch 4K HDR Monitor, Samsung 55 inch 4K HDR TV ) gave very different impressions. I found the EVA1 color looked the best to me when viewed on my 55 inch TV, but only with the correct exposure. ( over or under exposed footage looked quite bad )
 
You must log in or register to reply here.
Back
Top