I agree. I break out the Ninja for projects where ProRes is desired by the client for editing purposes (easy to ingest in FCP7) and then I tend to just set it to LT, which gives sizes slightly larger than XF. Otherwise, I shoot a lot of AVCHD for clients I know won't do much in post, and mostly use the Ninja for projects with C-log.
C300: C100 better blown out whites than C300?
- Thread starter Alvise Tedesco
- Start date
I agree. I break out the Ninja for projects where ProRes is desired by the client for editing purposes (easy to ingest in FCP7) and then I tend to just set it to LT, which gives sizes slightly larger than XF. Otherwise, I shoot a lot of AVCHD for clients I know won't do much in post, and mostly use the Ninja for projects with C-log.
Gary, what kind of stuff are you shooting that requires C-log and for what clients do you deliver standard AVCHD files (and using what gamma? wide dr?)? I just sold my external recorder because I find it's not really necessary when shooting wide dr and when using the C100 as an a-cam, but I'm likely going to rent a Samurai for an upcoming gig shooting b-cam for an Alexa (which will be shooting in 2k, so I have to upscale and grade to match). Does this sound sane? I feel like the C100 (and to some extent the C300) is more a big boy dSLR than a baby Alexa (like the F3 or F5, for instance), and I don't see myself booking gigs that F5/F55 guys are bidding on.
Finnegan
Well-known member
If you're talking about the "tests" I did with Rudy, they're not scientific at all. I don't have time to pixel peep -- generally I test my cameras by throwing them into a real world production and then I judge the footage in post. In these situations I've found that to my eye the C300 (particularly in C-Log) retains more detail in the shadows at low ISOs, has a slower and more balanced roll off into the highlights, and has a wider dynamic range (although this last bit might be perceived dynamic range because I make the judgment based on how much I can push/pull the footage in post). Lastly, because of the bit-rate and color space advantage, I find it's way easier to push the footage from the C300 around than the footage from the C100.
I was shooting in a warehouse with lots of high contrast scenes and the client wanted to do some serious color correction on the footage, so I shot C-log with that to the Ninja-2 in 10-bit 4:2:2 ProRes HQ. If I was shooting a personal project that I wanted to do some serious work with, that's what I'd use as well.
I have a lot of clients where I know the video won't be touched at all with a single drop of color correction, so I tend to shoot AVCHD with them or ProRes LT on the Ninja.
Makes sense for extreme circumstances (such as a warehouse might be), but since the camera has such poor tonality relative to the 10-bit guys, doesn't it make sense to maintain as much of it as possible in the midtones (using wide dr gamma) rather than spreading it out all over with the log curve? With the Alexa or F3 to prores you have the information to manipulate, but with the C100/C300 you start to lose tonality, and with wide DR you don't lose dynamic range relative to c log so why shoot anything but?
Just deciding how to approach this next project. I guess to intercut with the Alexa I'll go c log to 422 HQ, but wide dr seems so much more useful for everything else.
Just deciding how to approach this next project. I guess to intercut with the Alexa I'll go c log to 422 HQ, but wide dr seems so much more useful for everything else.
NeedCreative
Well-known member
I own and operate both a C300 and C100 (with latest firmwares). I almost exclusively use the C100 with a Ninja 2 recorder - keep that in mind.
I don't see any difference between the two cameras at all. None - nothing. I shoot both on the same profile (almost always C-Log, using Brian Weed's matched profiles). The picture is exactly the same thing. If I shoot the same subject with both cameras, it's exactly the same thing. Both to my eye, and via scopes. They intercut identically, and I can't tell the difference between them at all - dynamic range wise.
The Ninja 2 image is cleaner off both cameras - especially the C100 compared to it's AVCHD. I see a tiny bit more detail on the C300 than the internal MXF, mostly in the presentation of noise. That's the only difference between both cameras when I cut between them, and that's the recorder not the camera.
Bottom line, I'm not sure why you're seeing what you are seeing, but on my C300 and C100 with the Ninja? Using the latest firmware, using matched profiles, the image is exactly the same.
Shoot only AVCHD on the C100 and the 422 Canon XF on the C300 of course and this comparison goes out the window. The AVCHD is clearly worse at retaining detail, especially in expanses of non-focused color. See: http://filmmakermagazine.com/64918-canon-c100-does-422-with-the-ninja-video-recorder/ (examples from me)
dustylense
Veteran
@needcreative.
THe only problem I have with your blog test is that I have to wipe the nose grease off my monitor now. The very fact that you are taking the dreaded AVCHD and comparing it to a codec that is shooting 220mbits/sec and seeing at most a 5-10% improvement speaks volumes. Especially when you have to blow up the images to 400% to see the difference.
I've seen many comparisons with the C100 avchd vas ninja. First and this is for everybody, BACK OFF YOUR MONITOR when viewing material. Didn't your mothers teach you not to sit so close to the TV?
Now I get it. If you are going to add lightsabre's and spaceships to your work, you might want the higher bit depth and we ALL understand why. But man, the pixel peeping needs to really be curbed these days. You guys are going to go blind and have hairy palms.
The only reason I would add a ninja or such to my C100 is workflow. Other than that, I would guarantee that if one shot 220mbit Prores and I transcoded with 5d2rgb from the C100, we could press out faces to the monitor all we want and would barely see a difference.
But you are right. Add a ninja and you basically have the C300 with the C100.
THe only problem I have with your blog test is that I have to wipe the nose grease off my monitor now. The very fact that you are taking the dreaded AVCHD and comparing it to a codec that is shooting 220mbits/sec and seeing at most a 5-10% improvement speaks volumes. Especially when you have to blow up the images to 400% to see the difference.
I've seen many comparisons with the C100 avchd vas ninja. First and this is for everybody, BACK OFF YOUR MONITOR when viewing material. Didn't your mothers teach you not to sit so close to the TV?
Now I get it. If you are going to add lightsabre's and spaceships to your work, you might want the higher bit depth and we ALL understand why. But man, the pixel peeping needs to really be curbed these days. You guys are going to go blind and have hairy palms.
The only reason I would add a ninja or such to my C100 is workflow. Other than that, I would guarantee that if one shot 220mbit Prores and I transcoded with 5d2rgb from the C100, we could press out faces to the monitor all we want and would barely see a difference.
But you are right. Add a ninja and you basically have the C300 with the C100.
The Ninja-2 footage helps with isolating colors in the picture. Want to use Cosmo on your interview to help the talent feel better about how they look on-camera? The ProRes footage will let you isolate the skin tone better without the Cosmo filter applying to background objects in the 4:2:0 colorspace. Want to shoot C-log? The Ninja provides a higher bitrate so that pushing and pulling on those midtones doesn't introduce noise into the picture. Same for pulling keys.
There are definite reasons why the Ninja footage can be of benefit. However, for most people they don't go very in-depth with these post tools and thus the benefits aren't something they'd be likely to notice.
There are definite reasons why the Ninja footage can be of benefit. However, for most people they don't go very in-depth with these post tools and thus the benefits aren't something they'd be likely to notice.
digitalinnovations
Well-known member
I believe the links to the profiles are stickies at the top of this Cinema EOS forum titled 'Cowpunk's Custom Profiles". Mr. Weed goes by Cowpunk on the forums here.
Thanks, what are the advantages of these profiles, btw? I want to match the colors of canon log as closely as I can to the colors of the Alexa. I know the gamma curve will vary, but I want to get everything as close as possible otherwise. Should I create my own look with a color checker chart or are these a good starting point?
Also, HDMI out preserves all the highlight detail found in AVCHD, correct? I've read otherwise, but it would make sense that it would record superwhites the same.
Also, HDMI out preserves all the highlight detail found in AVCHD, correct? I've read otherwise, but it would make sense that it would record superwhites the same.
digitalinnovations
Well-known member
If you're looking for a custom profile to attempt to match the colors of an Alexa, Art Adams created a profile specifically for that. You can find his daylight and tungsten formulas here: http://provideocoalition.com/aadams/story/color-matching-a-canon-c300-to-an-arri-alexa/P3
As for your question; there will probably be others with views on highlight retention. Both "for" and "against", more than likely
I don't tend to get tied up in pixel peeping or the like. I find a tool that is useful to me, provides the features I need, makes nice imagery, and helps me make a living doing what I love. Beyond that...
As for your question; there will probably be others with views on highlight retention. Both "for" and "against", more than likely
I don't tend to get tied up in pixel peeping or the like. I find a tool that is useful to me, provides the features I need, makes nice imagery, and helps me make a living doing what I love. Beyond that...
NeedCreative
Well-known member
Maybe those aren't the best examples.
But trust me - wide expanses of color like walls, skies, etc. can fall apart under under the AVCHD when pushing the footage. It's not a mess worth risking on a paid shoot. It's not about pixel peeping, it's about professionalism and quality in the finished product. I demand the best I can get out of a tool, period, and if the Ninja helps workflow too it's a no brainer.
NeedCreative
Well-known member
It depends on what you're doing with it yes. If people are shooting "video by the pound", the target viewer for that is happy with compressed YouTube never mind the difference between AVCHD and ProRes capture. Other clients aren't as forgiving...
anatusa
Veteran
what is the gamma used in arri alexa cp ?
It doesn't matter, you can make it whatever you want.
digitalinnovations
Well-known member
I believe it was determined that Art prefers the Cine gammas (Cine1 if memory serves), but you could use whichever you prefer.
Do you have a video of this?