hyalinejim

Exactly the same but stretched.

On Windows 10, if you go in to Display Settings you can change a percentage slider which will scale the size of the Premiere interface. Not sure about Resolve. So yes, you can buy a bigger screen with a higher resolution and have the Premiere interface take up the same area relative to your current monitor. I have a 26" monitor and find that 1440p is plenty resolution for me. But at 32" you might want to go full 4K in terms of resolution. However, trying to find a 4K monitor that can be calibrated is probably very difficult / expensive.

Here is an XC10 and 5D3 Magic Lantern RAW matched: This was the starting point for each: I achieved this by carefully matching the tonality of each camera. The colours were matched in Resolve using white balanced and tonally matched shots of an XRite Colorchecker. Then I applied the same look to each clip and tweaked levels slightly.

This should get you started. It's only 720p, and is compressed for vimeo, and white balance seems off, but.... This might be a better resource, Log-C to Rec709 JPEGs: http://www.drewmoe.com/digitalnoise.shtml

That's quite a provocative question, but I like the way you think! Theoretically, it should be possible to get the GH4 to more closely match the Alexa. This is what I would do. 1. Shoot in 10bit V-Log to an external recorder 2. Use LutCalc to generate a VLog to Alexa LUT, to get the Panasonic gamma and gamut into the Alexa range 3. Shoot a colour chart with a GH4 and Alexa (this is the hard part) with the same exposure. Then bring both into Resolve. Apply the VLog to Alexa LUT to the GH4 shot and adjust the hue and saturation curves so that each square on the aligns with those in the Alexa shot, as seen on the vectorscope. Using this kind of method I've got my XC10 to match my 5D3 MLRAW almost perfectly, both shooting a colour calibrated version of C-Log. That's not entirely dissimilar from what you're talking about. OK, they're both Canons but actually the colours coming from each of them were actually quite different. Maybe you can find a shot of a chart shot on an Alexa somewhere and exactly replicate the exposure and lighting on your GH4. Good luck and don't forget to post your results when you succeed!

I could be wrong here, but I think that noise problem with 10bit was due to a wrong black level or something and now it's fixed.

I got an Xrite Passport recently and have been pissing about with colour in Resolve. Forget about its Colormatch function - it's a shot in the dark as to whether it works. But it's only 5 minute's work to selectively adjust the hue and sat of the colours on the chart so that they align with their correct targets on the vectorscope. Where are the correct targets on the vectorscope for the Xrite? Download one of these http://www.babelcolor.com/colorchecker-2.htm#CCP2_images and mask out the relevant squares using Photoshop or something and stick it as an overlay above your chart shot. Then copy that grade to all other relevant clips, or save it out as a lut for use elsewhere

Well, visually it looks great! Hope you had a couple of Zorglons to celebrate.

Both of these are an Iscorama pre-36 on a 5D2. Still going strong these days with the Isco for documentaries on ML5D3. It requires a measured approach to shooting. Not good for when you need to change lenses quickly in order to get the shot.

To be honest, if Magic Lantern team can ever get a usable preview for 10 bit recording then I'd be very happy to stick with continuous 24p 2.8k 10bit on the 5D3 for a number of years yet. The downsample for HD or 2K DCP delivery would just tighten up the resolution nicely while still allowing for a little wiggle room with stabilisation or reframing. At the moment the preview freezes when recording in the crop mode that allows resolutions above 1080.

Hey, awesome comparison @DaveAltizer I would be forever in your debt if you could upload a 16 bit TIFF of each of the ungraded Canon Log shots. This would be a godsend for trying to match my 5D3ML with other cameras.

I know what you mean - the XC10 feels so right in the hand and it's almost weightless. In fact, even with stabilisation off you can get a very stable shot because it's nicely balanced. It's a pity the lens is slow. That would help a hell of a lot as f2.8 seems to be ok for general interiors. Does that setup of your C100 work for handheld as well, or does it need to go on a tripod?

Exactly! It's like two cameras in one: a really great image in bright light and a really shitty image in low light. The point at which it stops being great and starts being shitty is down to individual taste and requirements. BTW, I'm getting a significant improvement in colour rendition by using an XRite colour checker and manually adjusting hue and saturation in Resolve, getting the vectorscope spikes on the chart shot to match an overlay I got from here. In general, when a shot is white balanced accurately, the hue and saturation adjustments tame the reds and warm the yellows quite noticeably. It also helps with matching the XC10 with other cameras.

The real life corollary is when you need to shoot something that moves significantly from frame to frame in low light. In practice ghosting is a problem from 1250 ISO up. But aggressive noise reduction artifacts aren't limited to motion. Check out the image softening that occurs with static scenes going from 500 to 1000.

When they got to their vows... I nearly hurled

Whack the ISO up to max and wave the camera around.

Absolutely agree with this. Even the 10bit footage was no cake walk.

Interesting! I didn't know that other cameras showed the same problem. As many people have mentioned, the average viewer wouldn't spot it, but it seriously gets on my nerves every time I see it. For anyone just popping into this thread who doesn't feel like trawling through pages of ghosting tests the bottom line is this: To prevent ghosting, shoot in EOS Standard and don't go over 1250 ISO. Explanation: the different camera profiles demonstrate differing amounts of ghosting. C-Log is the worst. EOS Standard is the best. But what about dynamic range? EOS Standard captures the same dynamic range as C-Log if you expose 1.66 ISO stops down (5 clicks). EOS Standard 160 = C-Log 500. If you then want to get back to a C-Log gamma for whatever reason, just bring down your superwhites to below 100 IRE and use the attached lut. I'm kind of amazed that this works as I thought C-Log was some special voodoo. It's actually an evil voodoo as it's applying a hell of a lot of noise reduction which reduces image detail and increases temporal ghosting. Every review I read said "Don't use EOS Standard - it's too contrasty!". I've learned a lesson about trusting the value of my own experience, based on my own tests. XC10 EOS Standard to C-Log v02.cube

@kidzrevil that video above sums up for me how I feel about the XC10. In good light, like the outdoors shots, the camera gives a lovely image which can be pushed quite hard in grading. But when light is low, the image is not good - it's mushy, falls apart in the grade, colour rendition suffers and ghosting is a PITA. For what kidzrevil needs to do - fashion, music vids, etc. - image quality is paramount and the XC10 doesn't cut the mustard. For more casual stuff like travel vids and documentation it's absolutely fine. I promised to post a few frame grabs to show off what the camera can do when conditions are right. The weather was great the other day so we went for a walk. I brought along the XC10 as I wanted to check out how an X-Rite Colorchecker might improve colour separation using MBR Color Corrector 2 plugin for AE (it did help). No look LUTs here. Everything is EOS Standard, corrected for colour using the X-Rite, and a simple white balance, contrast and saturation adjustment.

Yes, in good light the image from the XC10 can be beautiful. I'll upload a few more stills in a while. I've been playing around with an X-Rite Colorchecker in an effort to get more pleasing colour.

Also, compression artifacts will affect our perception of motion.

Saw this reposted on dvxuser the other day and found it absolutely fascinating: Film v Alexa: http://www.yedlin.net/DisplayPrepDemo/ Background: https://storify.com/tvaziri/steve-yedlin Philosophical implications: http://www.yedlin.net/160105_edit.html

The short version is: it's not the XC10, and the XC10 will not give you the great low light performance, 4k slow motion and shallow DOF that the video promises - I wish it did!

In the comments on that same vid: (Although commenters don't seem to believe it's that camera either) And when "Michelle" isn't exhaustively testing cameras, she's making gems such as this:

It can't be the XC10: 1. XC10 has 8 aperture blades. Video shows 10 sided bokeh 2. Video is supposed to be shot in 4k, yet it's in slow motion. There's no 4k slow motion on XC10. 3. In the mirror shot, the lcd screen is flipped horizontally. XC10 screen only tilts vertically. That's a bullshit account - here is the same vid on someone else's account: Description: Trying Out My New Canon EOS 1D 4K PROFESSIONAL CAMCORDER camera With 5 - 100 mm Lens