C200: Alexa Mini VS C200

The reason people would like more DR is so their final grades (which are usually a bit crushed/contrasted) can retain more information. And better control of highlights, of course. Because if you're starting with 12, you'll likely have a final product with 10, 10.5.

So people want 15-16, so you can end with 13-14.

It's been this way for like 15 years...you always want more to work with.

And you always tried to squeeze out as much as possible out of the camera.
 
Mark Smith said:
Dynamic range is over rated in my estimation as there are so few viewing situations where you can capitalize on 13,14,15 stops of DR. Granted, that. The trend is towards greater DR these days but stillI think its a fad. In some recent years 3D was the fad- there was this huge stampede over to the side of the film space where everything was going to be shot 3D, and then poof no more 3D. DR seems like the same thing to me. No way, no how more DR fills more theater seats, its just the latest thing that people in the film industry have adopted as a means of differentiation which is for the most part meaningless.
I am not advocating going back to rec601 standard def with 6-8 stops of DR, but the quibbling over 13,14 or 15 stops of DR just strikes me as kind of silly since a huge % of displays in use these days can’t replicate those numbers.
Click to expand...

Any screen can show 20 stops of DR !

Ansel Adams could get 15+ stops onto a bit of paper with no back illumination.

Consider an INT scene with a window - if you use a secondary in post to 'pull down' the scene outside the window then you are 'showing' 20 stops of DR on the screen.

But if when you shot all outside the window was clipped then you cannot pull it back in post.*

Wide DR cameras are therefore entirely useful even for delivery onto low DR screens.

--

This is why I keep saying all the 'tests' are dumb - none of the tests are pulling secondaries in the grade and seeing where the cams do or dont perforfom.

All of the DPs the test lack rigour or method.. because they would need to expose for the brightest outside when shooting an INT to do proper testing.

All we are really seeing on these 'tests' is that the middle few stops can be matched pretty well.. which is indeed a good reflection on the C200. But then a decent colourist can match basically anything that is not clipped.

S

*clearly I would not propose exposing for the out on an INT scene leaving the int 10 stops under and lifting it in post to be a method that one should actually use on set.. only during the extreme end of a camera stress test.
 
Last edited:
Matt Turner said:
Who shoots the C200 by itself! I thought everyone would dual shoot with an Alexa so they could use it as a reference to get the C200 closer in post.

Jokes aside, most people would slap on a LUT (like from this guy or Arri Log C or @jcs?), in which case they would really have no idea how close their shot was to an actual Alexa, just an assumption it is close.
Click to expand...

As you know from working with a variety of cameras over the years, you start to see patterns/styles in the color science and tone curves of various cameras. As a fan of the ARRI look (anyone not a fan lol?), which IMO is simply the most accurate 'as the eye sees it' compared to the competition. There's still room for much improvement, especially mixed light and bounced environment light, though simply being ~100% accurate would be great!

After dropping $16K for a C300 II at launch, I thought I finally acquired the mythical perfect camera haha. Canon 5D3 color with really excellent autofocus, built in NDs, and pro-audio XLR with phantom power. For a long time I just shot with the 'Alexa' compatible settings, Canon Log 2, Cinema Gamut and Production matrix, where I could simply slap on an Alexa LogC to Req709 LUT and be good to go. While that looked good most of the time, it didn't look as 'Alexa' as I'd hoped.

I had purchased a 1DX II to get 4K60 and to upgrade my 5D3s for stills and was using it more and more for on-the-go, run & gun, and travel shooting. The Canon default settings looked really bad to me, especially compared to the 5D3 which could look decent by tweaking the default settings (Neutral, Faithful, etc.). For example, the blacks are crushed to oblivion and if you turn the contrast down too much, skin tones get really ugly (general case). So I experimented with creating a log emulation (including testing CineStyle) and found that also messed up skin tones (general case) and ended up doing something outside the box which got the 1DX II looking better IMO than the C300 II with 'Alexa" settings + LUT. So I spent way too much time tweaking the C300 II trying to get as good as the 1DX II with the tweaked custom picture style. I got pretty close, but the 1DX II IMO still looks better- e.g. for live streaming or when shooting for in-camera looks. I found Canon Log 3 was the best overall log, and coupled with better actual DR and NDs etc., the C300 II is still a much better video camera for higher-end kinds of shoots. However for what we shoot- totally controlled lighting green screen and mostly simple talking heads, walking outside using wireless audio (no CMIT5U for the 1DX II :( ), I realized the 1DX II was all we needed and the rapidly depreciating C300 II got sold.

Like Mark, the fact that the C300 II doesn't really provide 15-stops doesn't bother me at all. From my own use I could see it was better than the 1DX II (which when shooting for in-camera looks is fine for us so log wasn't really needed). What bothers me (and I think everyone) in general is that camera manufacturers lie and probably even more annoying is the crippling. C300 II can't do 4K 60- great I'll just get the 1DX II. So it worked on me haha. However, what I found out was that the C300 II didn't even do real 4K (.67 oversampling is needed, e.g. 6K to get real 4K), has terrible aliasing (couldn't believe it first time I saw it after the 5D2 => 5D3 phase), and after comparing C300 II tweaked color to 1DX II tweaked color, didn't look as good. Even with all the warts, I didn't really want to sell the C300 II because it was still very convenient to use and provided adequate image quality. Rumors of the A7S III and other recent high-spec cameras got me thinking about depreciation and felt the camera deserved a new home for someone who could really use everything it could offer (sold to a production company- camera was mint). There was one other reason I kept the C300 II a bit longer than otherwise- to finalize and sell improved quality Custom Picture settings. That didn't turn out to be a big market so the camera found a new home (the settings should work on the C200 if anyone wants to be a guinea pig ;) ).

Back to ARRI- they're the one camera company who typically (until the Alexa LF "Large Format look" claims lol) doesn't lie, and is conservative in their marketing. Red also exaggerates and has burned a lot of people with reliability issues (same with Black Magic). However, if it weren't for reliability issues, there'd be a lot more folks using Red and BM cameras- lately the footage coming from these cameras is really, really good, right up with ARRI (but again not as fault tolerant to exposure etc.).

A lot of working pros get annoyed with my frequent comments about how cameras are getting smaller. The Alexa Mini should be proof enough that is where the market is heading (stripped Sony VENICE is pretty small; also the removable sensor block). If one has a 'large' video camera and it's not actively making one money, probably best to sell it now as the depreciation is going to accelerate- perhaps why Sony's never actually going to release the mythical A7S III haha.

The C200's AF makes it usable in cases where the Alexa Mini isn't. The Mini's slomo and latitude and overall better color science make it more useful than the C200 when top image quality is required. In terms of 4K, neither camera shoots real/true 4K, however I'm sure the average person can't tell the difference! (especially if accutance was matched in post).

In the final analysis Sony with their AI AF with VENICE like color in a compact pro camera wins the game. The tech of which they could put into the A7S III, today! haha. Hopefully ARRI innovates and stays relevant- competition is good! Lots of new camera tech coming in the future (next phase (computational+AI cameras), which is a few years out, will makes expensive lenses massively drop in value).
 
jcs said:
However, what I found out was that the C300 II didn't even do real 4K (.67 oversampling is needed, e.g. 6K to get real 4K), has terrible aliasing...
Click to expand...

To quote the Canon whitepaper:

"The central design goal was to offer a substantial extension of the 2K / HD options beyond the HD-only capability of the C300 while also adding a modest 4K / UHD capability"

Which is what you are describing. I've got no way of delivering 4K to viewers without putting it on YouTube, so luckily none of that makes much difference!

I've been meaning to try out your LUT one day hopefully I'll get a chance to give it a go soon.
 
If you capture 20 stops and you think you monitor can display 20 stops, shouldn’t one be able to see 20 stops on the monitor without using a secondary window to pull down the exterior highlights?
By using a secondary window you are taking some highlight info that is recoverable and bringing it down into a range that the screen can display.
 
jcs said:
...Lots of new camera tech coming in the future (next phase (computational+AI cameras), which is a few years out, will makes expensive lenses massively drop in value).
Click to expand...

This has already happened because the sheeple have moved on to FF. Shooting S35 cameras and lenses is so "2018" Sarcasm emoji. All that S35 killer glass has seriously dropped in value in less than a year because the majority has decided that it's all crap and digital cinema is only worth shooting in FF, so ridiculous. A complete triumph of marketing combined with brainless, lock step thinking by people who supposedly know what they're doing and should know better than to fall for a "new" technology push that is a LOT more expensive, more hassle to shoot and provides negligible improvements over the status quo. The only term that comes to mind is "suckers". For those who have done the homework and decided FF provides exactly what they need, want and desire, bravo. To the rest of the brainless followers who buy into this "innovation", "a fool and his money are soon parted". Go elsewhere on discussion boards and you'd think that you can't actually make a film without FF technology.

Seriously though, it's alarming at how most people in our business don't have a brain of their own and are too lazy to devise their own gear preferences based upon testing, research and comparitive analysis done for real, in person, by spending a few hundred bucks to rent or borrow candidates. That's one thing you do that I admire, you really test your gear and tweak it, just like DPs used to do in the film era. I still do it, although not to the extent you do, I rented the FS-7 for a few shoots, had shot numerous time in with the C300 MKII and even tried the BMD UMP before settling on owning the C200. I shot test charts, tweaked with footage, not to the extent that you do, but enough to satisfy my curiosity before pulling the trigger. I also debated on the EVA 1 over the C200 but the EVA 1 wasn't available yet and I had a deadline on beginning filming one of our documentaries. But at least I did my homework and didn't buy a technology just because it was the flavor of the week.
 
Mark Smith said:
If you capture 20 stops and you think you monitor can display 20 stops, shouldn’t one be able to see 20 stops on the monitor without using a secondary window to pull down the exterior highlights?
By using a secondary window you are taking some highlight info that is recoverable and bringing it down into a range that the screen can display.
Click to expand...

Again any screen could display 20stops without secondaries - but the image would be horrifically flat - not something any viewer would enjoy - like un kitted log - but way worse.

A normal display cannot show 20 stops at proper contrast ratio.

“Proper contrast ratio” is something people are very woolly on.
 
The same folks just did another C200 comparison with the RED Helium. Haven't gotten a chance to actually watch this one, but I'd expect there to be more of a difference here because of the higher DR and double the Ks.

 
Since we're comparing the C200... curious about everyone's thoughts about this video comparing the C100 vs C200 that I came across.
In most of the shots, the C100 looks better to my eye. Is the similar detail/resolution a lens limitation here? Makes me hesitant to ever upgrade to the C200 from my C100 if I'm never going to shoot raw...

 
Last edited:
As an owner of two C200’s I would definitely not recommend upgrading from a C100mk2 to a C200 unless you are going to utilize raw on a decent percent of your projects OR if you’re going to utilize doing 4K punch-ins during interviews. Having owned and used C100’s and C100mk2’s, a C200 in .mp4 is a 4K C100mk2.
 
*or if your clients that don’t know any better want 4K because someone told them it should be shot in 4K. (For me, this alone would justify the upgrade)
 
BillP said:
Since we're comparing the C200... curious about everyone's thoughts about this video comparing the C100 vs C200 that I came across.
In most of the shots, the C100 looks better to my eye. Is the similar detail/resolution a lens limitation here? Makes me hesitant to ever upgrade to the C200 from my C100 if I'm never going to shoot raw...
Click to expand...

The C100 uses a 4K sensor to get 1080p. The C200 (and C300 II) use a 4K sensor to get 4K. In order to get True 4K, you need 1.5x Bayer sensor resolution, or 6K downsampled to 4K. The C100 was effectively 2x Bayer sensor resolution to get HD, which explains why the C100 HD looks so good, and why the C200 (and C300 II) 4K doesn't look that great, they're really only doing 4K/1.5 = 2.67K which isn't a big leap over HD vs. True 4K. In the examples the C200 did have less sharpening artifacts in some clips (not clear what the settings were on each camera). Full discussion and analysis here: http://www.dvxuser.com/V6/showthread.php?356356-Actual-Camera-Resolution-Before-Aliasing

When I found the C300 II 4K YUV 422 10-bit didn't key as well as HD RGB 444 12-bit (at almost 1/2 the bitrate), I stopped using the C300 II in 4K and used only HD RGB 444 12-bit and upscaled to 4K and sharpened in post: overall a better result vs. the fake in-camera 4K (which also of course aliases! (related to the 1.5x / .67 oversampling factor and Nyquist sampling theory (see link above)). Canon finally provided True 4K in the C700 FF (5.9K sensor). Now that the cat's out of the bag, no one's going to buy a fake 4K camera anymore. (except perhaps the Alexa Mini :)). Most C300 II owners realized this intuitively and found HD/2K RGB 444 12-bit to be the best way to use the C300 II.

As for the C200/C300 II sensors not looking as good as the prior generation in terms of color- I noticed that the 5D Mark III looked better than the C300 II as well as the 1DX II, for video (color for skin tones). After tuning the color on the 1DX II (custom picture style), it finally looked as good as the 5D3 (close enough), and then the 1DX II was clearly better than the C300 II. Even after tuning C300 II color, the 1DX II produced better skin tones (and detail for that matter). The C300 II has more dynamic range and lower rolling shutter; not clear if there were design tradeoffs that affected color, or if Canon did this on purpose for business reasons (related to the next upgrade cycle). On that note, it appears the new EOS R and RP have better out-of-the-box color for video vs. the 80D and 1DX II (based on comparing online videos), possibly even better than the 5D3. Maybe I'll test the EOS R against the 1DX II and 80D in the near future.
 
morgan_moore said:
The C200 is a raw camera with a crappy proxy embedded.

If you are not going to shoot raw you are not buying the right camera.
Click to expand...

Agreed, 100%. My $1,399.00 Fuji X-T3 blows away the Canon C200 8-bit frankly, but the Cinema RAW Light from the C200 is excellent. And the 8-bit isn't bad, it's just there are far cheaper cameras that look better and shoot 10-bit. But the X-T3 is a mirrorless and mirrorless sucks for a lot of video shooting situations that the C200 is great in. Glad I own both, choices are good.
 
puredrifting said:
Agreed, 100%. My $1,399.00 Fuji X-T3 blows away the Canon C200 8-bit frankly, but the Cinema RAW Light from the C200 is excellent. And the 8-bit isn't bad, it's just there are far cheaper cameras that look better and shoot 10-bit. But the X-T3 is a mirrorless and mirrorless sucks for a lot of video shooting situations that the C200 is great in. Glad I own both, choices are good.
Click to expand...

Our lens choice was different, but this is the same loadout I roll with as well. It's the most versatile setup I've found that fits my workflow. We're spoilt for choice at this point. The C200/X-T3 combo costs $9,000US for the bodies. For that money, you can get a pretty amazing amount of gear; from multiple entry level cinema or mirrorless cameras all the way up to a used RED Dragon (with factory warranty).

PS - The X-T3 is an amazing camera with no qualifications. That's especially true if you shoot both stills and video. I'm finding myself reach for it more than my 1DC.
 
Fahnon said:
Our lens choice was different, but this is the same loadout I roll with as well. It's the most versatile setup I've found that fits my workflow. We're spoilt for choice at this point. The C200/X-T3 combo costs $9,000US for the bodies. For that money, you can get a pretty amazing amount of gear; from multiple entry level cinema or mirrorless cameras all the way up to a used RED Dragon (with factory warranty).

PS - The X-T3 is an amazing camera with no qualifications. That's especially true if you shoot both stills and video. I'm finding myself reach for it more than my 1DC.
Click to expand...

I agree, I continue to be impressed at how plain great the images look, although I am yet to be convinced on skin tones. I shot something the other day, available light in a subject's kitchen before sunrise on a documentary, they were lit solely by an overhead CFL and the color cast and skin tones are not great. But then again, it didn't look very good to my eye either. It is lifestyle footage for a documentary so I couldn't actually light it. I think my C200 might have had a better skin tone but I am still messing with CC and it's probably just my lack of color correction skils too.
 
puredrifting said:
I agree, I continue to be impressed at how plain great the images look, although I am yet to be convinced on skin tones. I shot something the other day, available light in a subject's kitchen before sunrise on a documentary, they were lit solely by an overhead CFL and the color cast and skin tones are not great. But then again, it didn't look very good to my eye either. It is lifestyle footage for a documentary so I couldn't actually light it. I think my C200 might have had a better skin tone but I am still messing with CC and it's probably just my lack of color correction skils too.
Click to expand...

Not to derail the thread, but I find the skin to be good on X-T3 when I shoot F-Log and use curves+sat in resolve. I don’t shoot Eterna or apply a LUT in post and it looks great to my eye. I wonder if we have different workflows?

Not a colorist, but I actually find that with most cameras shooting log curves+sat looks better to me than transform LUTs.
 
You must log in or register to reply here.
Back
Top