Canon 5D Mark III - Resolution Review

[psolberg]

Stephen- hard to say until we hear from Canon and Adobe.

psolberg- check out the bear picture and Andrew's comments here: http://www.eoshd.com/content/7608/cinestyle-on-the-5d-mark-iii-and-fixing-softness-in-post#d2

He too was surprised, but it works. A clean, anti-aliased image can be sharpened quite a bit by removing the anti-aliasing with a convolution sharpen filter. The trick is to remove enough to increase sharpness while not introducing excessive aliasing.

Sharpening 700 lines of resolution helps but it ain't the 1000 lines I hoped for or want out if the box. It is off course the only choice short of a c300 or Sony fs.

 

[jcs]

In my informal resolution test with a printed test chart comparing still to video, it looks closer to 800-900 lines. The GH2 was rated around 800 lines; Andrew @ EOSHD matched the 5D3 to GH2 footage, so that sounds about right.

Look forward to seeing a proper resolution test with a real test chart. In any case, the price is fair for the combination of low aliasing, high resolution, and full frame capability.

 

[psolberg]

http://www.eoshd.com/content/7620/3000-nikon-d800-thrashes-flagship-6000-nikon-d4-for-video#d2
EOSHD still seems to think the 5Dmk3 is a 720p faux 1080p camera (his own words).

If you are afraid of 720p levels of resolution in a faux 1080p mode get the Nikon D800 because it is far more detailed.

Sharpening or not, 720p level is very dissapointing. I decided to skip the 5D3. Maybe that 4k body won't cost 16k , although I'm sure it will be C line meaning it will make the 1DX cheap in comparison.

 

[psolberg]

In my informal resolution test with a printed test chart comparing still to video, it looks closer to 800-900 lines. The GH2 was rated around 800 lines; Andrew @ EOSHD matched the 5D3 to GH2 footage, so that sounds about right.

Look forward to seeing a proper resolution test with a real test chart. In any case, the price is fair for the combination of low aliasing, high resolution, and full frame capability.

new post from EOSHD
http://www.eoshd.com/content/7631/panasonic-gh2-vs-5d-mark-iii

quote from the article:

The 5D Mark III’s image cannot be sharpened in post to match the GH2. That camera has a sharper image yet it is organic and fine not digitally over sharpened. Yes the 5D Mark III does benefit from a little bit of sharpening with certain shots over the master footage direct off the card but it adds to your workload in post – you have to be very careful about which shots you sharpen and which you don’t. Unlike moire and aliasing, resolution affects every frame so this is quite an important area and my single biggest frustration with the mushy 5D Mark III.

As I suspected, sharpening can only go so far when your base resolution is low to begin with. we know this from still images and given the low resolutions of HD video on top of the soft 5dMKIII output, sharpening is just going to help contrast (preceived sharpness) but won't bring detail out that was smudged by the codec.

 

[peederj]

OK now I've done some more pixel peeper tests and I've come around to agreeing with the Crooked Path recommended picture style: Faithful 2, -4, -2, 0. Faithful is truer to skin tones, which you usually will want, though without humans in the picture Neutral may sometimes be better. But the important thing is two clicks of sharpening is the right setting. Sharpening at 0 is the right setting for stills I think but that doesn't translate to video. The picture looks a bit blurred at 0 sharpening...they may actually be intentionally blurring it as an anti-aliasing or anti-moire measure (just speculation...I have no idea...maybe they did it that way for the 5d2 and it's carried over). So it's possible the "native" sharpening setting is 2, but regardless, I am getting a sharper picture without haloing on the setting of 2.

Crooked path also has his own picture styles built off faithful. Though personally if I'm going to use a flat picture style it will be Technicolor Cinestyle which Crooked Path explains on his own blog: rather than a hacked-up faithful, Cinestyle is a deeper interaction with H.264. And I do think Sharpening 2 on Cinestyle is probably a better idea too than sharpening 0.

Just sharing my thoughts as I update them. Try it out and see what you think.

 

[peederj]

EOSHD has a new video up comparing the 5D3 to the GH2. Bizarrely, he is using Neutral 2, 2, 3, 4. Which gives an awful green tint! I have to think someone told him to use that setting, but they were counting clicks to the right on the scale, but he interpreted it as Canon's actual numbers! I think they intended for him to shoot with Neutral 2, -2, -1, 0 which is a reasonable setting. Again my suggested setting for post grading is Faithful 2, -4, -2, 0.

http://www.eoshd.com/content/7631/panasonic-gh2-vs-5d-mark-iii

My resolution tests were a bit depressing though. It's true the 5D3 video is fairly low-res, and the GH2 with the Ptools hack may well beat it handily. My shots didn't all get up to the bitrates that are nominal for this system...it's VBR and I was only getting half the nominal rate on locked down shots. I guess that's the GOP compression working well, but I would hope for more resolution available with more bitrate available.

However, varying resolution might look worse...suddenly things would appear to focus and defocus. So maybe what Canon is doing with resolution is drawing a compromise: fast-changing scenes will have comparable resolution to unchanging scenes, so that there is no visible degradation or blurring in specific places and times, but instead an overall softness that ensures the video will look consistent straight from camera.

Possible?

 

[JasonATL]

I'm growing a little weary of EOSHD's soapbox. I get it. The 5D3 isn't what he hoped it would be in terms of resolution.

JCS has shown here that it appears best to set sharpness off in order to get the best out of this camera. Why would someone doing comparison tests set sharpness up rather than down? Why not set it the way JCS has and then sharpen in post? Why not put 5D3, 5D2, and GH2 through actual resolution tests with resolution charts? Why continue to use ALL I-frame if IPB doesn't have the problems?

I'd love to answer these questions myself, at least as far as the 5D3 is concerned, but I'm still waiting on mine. Perhaps this weekend... In the meantime, I'm learning from others' and appreciate the reasonable approach that I've seen from JCS, as opposed to the approach some have taken who seem more emotional and less interested to get the best possible out of the 5D3.

 

[jcs]

peederj- I too have found the bitrates to be a bit low: detail is being lost by overcompression.

I did a detailed test of I-only vs. IPB. I-only is noisy, and appears to have a macroblock issue in PPro (couldn't replicate in FCPX on the Mac, though their sharpen function is Unsharp Mask looking and can't really bring out fine detail since there is no 'radius' option). I could not see any resolution/detail increase by using I-only; still recommend IPB until the issue(s) are worked out.

JasonATL- I have found 'Faithful' works the best so far. You can also sharpen fine detail with an Unsharp Mask (USM) filter with radius set to < 1.0. Local Contrast Enhancement via USM can be achieved with a 30-100 radius.

 

[LetTheRightLensIn]

peederj- I too have found the bitrates to be a bit low: detail is being lost by overcompression.

I did a detailed test of I-only vs. IPB. I-only is noisy, and appears to have a macroblock issue in PPro (couldn't replicate in FCPX on the Mac, though their sharpen function is Unsharp Mask looking and can't really bring out fine detail since there is no 'radius' option). I could not see any resolution/detail increase by using I-only; still recommend IPB until the issue(s) are worked out.

JasonATL- I have found 'Faithful' works the best so far. You can also sharpen fine detail with an Unsharp Mask (USM) filter with radius set to < 1.0. Local Contrast Enhancement via USM can be achieved with a 30-100 radius.

The compression loves to compress away any detail in shadows or even fine details that are not of extreme contrast and it tends to go to a very video-type noise look instead of a nice tight 'grained' sort of noise. I think Canon is wayyy too paranoid about anything looking, horrors, grainy, at the non-C300 plus level that they think it better to NR and compress away nicer looking quality in favor of wax or something.

 

[LetTheRightLensIn]

Sharpening up a fair amount some original file 5D3 files of twigs and branches that someone posted, there is still something weird. Looking closely I can occasionally see some frame where very fine twigs have parts of the branch missing, almost like something line skipped (but later smeared to avoid AA). Weird. Not quite sure what they are doing to produce the video on the5D3 yet.

 

[psolberg]

peederj- I too have found the bitrates to be a bit low: detail is being lost by overcompression.

I did a detailed test of I-only vs. IPB. I-only is noisy, and appears to have a macroblock issue in PPro (couldn't replicate in FCPX on the Mac, though their sharpen function is Unsharp Mask looking and can't really bring out fine detail since there is no 'radius' option). I could not see any resolution/detail increase by using I-only; still recommend IPB until the issue(s) are worked out.

JasonATL- I have found 'Faithful' works the best so far. You can also sharpen fine detail with an Unsharp Mask (USM) filter with radius set to < 1.0. Local Contrast Enhancement via USM can be achieved with a 30-100 radius.

I don't understand how the bitrates can be too low. it is TWICE the bitrate of the 5DII and nearly 3X that of the IPB codec which IIRC is 28mbps. I've seen more detailed footage from nikon's internal IBP codec than from the all-I canon codec. I think the problem is sensor sampling. I suspect the way canon was able to reduce the amount of moire was by trading fine detail. so no matter how much bits per second you throw a it, a soft image is being created early enough before it is even compressed.

 

[JasonATL]

I've made a resolution and sharpness test of the 5D3 using a resolution test chart. I'm no pro, so I don't make more of this than it is. Still, I thought it performed quite well. My Canon 600D shows about 700 lines of resultion and this appears to show 800 or more lines. My Sony PMW-EX1 shows the full 1000 lines on the same test, which makes me think I've got it set up close to correctly - but I'm open to pointers and insights.

The video should be live on Vimeo in a few minutes.

https://vimeo.com/39517721

 

[peederj]

The bitrates they are advertising are maximum bitrates. In practice I'm getting something close to 60% of the advertised figures on steady-state shots. High-motion shots may get up to those bitrates. It's VBR, and a very stingy one at that. I want a superfine IPB mode with double what we're currently getting, and given ALL-I is rated as streaming 90mbps there's no way they can claim they can't give it to me if they are willing.

crippleware...

 

[psolberg]

interesting post at EOSHD

http://www.eoshd.com/content/7727/james-miller-removes-optical-low-pass-filter-from-5d-mark-iii-for-resolution-increase

apparently, the soft video output of hte 5DmkIII may be due (at least in part) to the sensor OLP filter. This means anybody hoping that either a bitrate increase or a firmware update would result in more resolutions needs to keep their expecations in check. The problem may very well be hardware and if the data being sample is already smudged, then the image quality is as good as it is going to get.

 

[peederj]

interesting post at EOSHD

http://www.eoshd.com/content/7727/james-miller-removes-optical-low-pass-filter-from-5d-mark-iii-for-resolution-increase

apparently, the soft video output of hte 5DmkIII may be due (at least in part) to the sensor OLP filter. This means anybody hoping that either a bitrate increase or a firmware update would result in more resolutions needs to keep their expecations in check. The problem may very well be hardware and if the data being sample is already smudged, then the image quality is as good as it is going to get.

Easily disproven: take a look at the resolution 10x live view ("focus assist") gives you vs. what streams out of the cripple codec. The hardware is more than fine, and can generate amazing resolution.

 

[psolberg]

interesting post at EOSHD

http://www.eoshd.com/content/7727/james-miller-removes-optical-low-pass-filter-from-5d-mark-iii-for-resolution-increase

apparently, the soft video output of hte 5DmkIII may be due (at least in part) to the sensor OLP filter. This means anybody hoping that either a bitrate increase or a firmware update would result in more resolutions needs to keep their expecations in check. The problem may very well be hardware and if the data being sample is already smudged, then the image quality is as good as it is going to get.

Easily disproven: take a look at the resolution 10x live view ("focus assist") gives you vs. what streams out of the cripple codec. The hardware is more than fine, and can generate amazing resolution.

I don't know that I agree with your assesment but removing the OLP filter seems to help. I'm sure more testing will be done to prove or disprove that modification as a way to fully realize the potential of the camera. I guess stay tuned.

 

[JasonATL]

I'm with peederj on this on. This is simply illogical. If the optical low pass filter is, indeed, optical, then it is in hardware. In live view, zoom in as peederj suggests. The resolution is there on the sensor AFTER the optics of the camera. So, any softening would logically come from the downconversion AFTER the sensor and the optics.

Having said that, if you provide an even sharper (dare I say even aliased) image to the same downscaling algorithm, it seems plausible to me that the result might appear sharper. But, that won't necessarily make a better image overall.

I won't be running out to mod my 5D3 any time soon, assuming this is even serious.

 

[psolberg]

I'm with peederj on this on. This is simply illogical. If the optical low pass filter is, indeed, optical, then it is in hardware. In live view, zoom in as peederj suggests. The resolution is there on the sensor AFTER the optics of the camera. So, any softening would logically come from the downconversion AFTER the sensor and the optics.

Having said that, if you provide an even sharper (dare I say even aliased) image to the same downscaling algorithm, it seems plausible to me that the result might appear sharper. But, that won't necessarily make a better image overall.

I won't be running out to mod my 5D3 any time soon, assuming this is even serious.

just curious, did you watch the video? phillip bloom had his own take
http://philipbloom.net/2012/04/01/a-drastic-solution-to-increasing-sharpness-with-the-5dmkiii/

To my eyes, it looks better, but as I said, I'm sure we'll find out more in time.

 

[jcs]

I compared P.Bloom's sharpened material to J.Miller's OLPF hack on a computer monitor and on an HDTV (D8000). Though it's not the same subject matter, I don't see any significant advantage to no-OLPF vs. post sharpening*. I did see aliased edges with the no-OLPF footage, however.

*There is at least one disadvantage to post sharpening- it effects edges of bokeh. Fortunately, sharpening can be controlled, per shot. When noise is added as 'film grain' (not straight Gaussian noise- processed a bit), it helps fix issues caused by sharpening in post: https://vimeo.com/39523633 (link to instructions on dvxuser).

 

[psolberg]

I compared P.Bloom's sharpened material to J.Miller's OLPF hack on a computer monitor and on an HDTV (D8000). Though it's not the same subject matter, I don't see any significant advantage to no-OLPF vs. post sharpening*. I did see aliased edges with the no-OLPF footage, however.

*There is at least one disadvantage to post sharpening- it effects edges of bokeh. Fortunately, sharpening can be controlled, per shot. When noise is added as 'film grain' (not straight Gaussian noise- processed a bit), it helps fix issues caused by sharpening in post: https://vimeo.com/39523633 (link to instructions on dvxuser).

yes. sharpening just to achive what IMO should be the output we should get unedited blows big time. Not only do you have to be selective which ads work in post but you have to watch out the amount or introduce artifacts. I know there are ways to improve the footage we've seen so far but it is just so annoying to have to add it each and every time...... :-\