Ebrahim Saadawi
Carbonite Member
Removed for copyright matters.
Last edited:
What makes a high quality image? The technical aspect of Image quality in Cameras
- Thread starter Ebrahim Saadawi
- Start date
FrameFarmMedia
Veteran
Sony has many flavors of XAVC.
XAVC-S is the consumer version but should be pointed out that not all Sony XAVC is created equal.
XAVC-I that is Intra, 10bit & 422 colour. Its on their higher end cams (FS7 and up). Sure its not Pro-res but it does a great job of giving you 10bit 422 in decent file sizes.
Lots of great information there. Must have taken a while to write up. well done.
XAVC-S is the consumer version but should be pointed out that not all Sony XAVC is created equal.
XAVC-I that is Intra, 10bit & 422 colour. Its on their higher end cams (FS7 and up). Sure its not Pro-res but it does a great job of giving you 10bit 422 in decent file sizes.
Lots of great information there. Must have taken a while to write up. well done.
mainstreetprod
Veteran
Very informative article, and not just for beginners.
Ryan Patrick O'Hara
Veteran
You are 100% incorrect in regards to bokeh.
You are 100% incorrect in regards to bokeh. Bokeh shape is completely subjective. 100% a matter of opinion. While many people hold the opinion that circular bokeh is favorable, the fact that one is more 'cinematic' or 'better' than another is complete and utter nonsense. I partly blame this misconception to inventive advertisers have many-a-time advertised round apertures as 'cinematic' when in reality nothing is cinematic. In fact, if we were to be forced to define what 'cinematic' is, I would argue 'cinematic' is simply the qualities we have recognized in the history of cinema. Things that we attribute to looking 'cinematic'. Thus I would further argue that the vast majority of cinema lenses in our 100+ years of making motion pictures have been varying degrees of polygonal bokeh.
I think the shape of bokeh has gained more attention in recent years due to the current trend of super shallow DoF. I would furthermore argue that if one looks at the following image and thinks it's the bokeh shape, circular or polygonal, which is the main concern for distracting a viewer from story, their perception is a little off.
Also,
Lowlight performance is a result of many different factors. Larger sensors have greater potential for better low light performance, but don't necessarily guarantee it. To simplify something that is far from simple, larger photo-sites provide better low light performance. It's really about the size of the photo-sites.
Bokeh: it's the quality of the blurry out-of-focus areas of the image, a good lens will have round/circle blur with a creamy smooth look while a bad lens will have a harsh non-creamy look. This is best demonestrated here:
The right image the bokeh isn't circular, it's more distracting, while left one looks dreamy, smooth.
Perfect creamy/circular bokeh
You might wonder why a lens produces non-circular bokeh and some produce circular bokeh, it's determined by the shape of the lens "opnening", which is controlled by the number of iris blades. To demonstrate here: a lens with a low number of blades makes non-circular bokeh while a lens with a high number of blades makes circular more pleasing bokeh
In any lens specifications the number of iris blades will be mentioned, so now you understand why a higher number is better to have in a lens.
-An ideal lens would not have any of those defects, and the higher the quality the lenses are, the less defects they show, and the more expensive they get! So in low-budget photography/videography you'll have to make a few sacrifices and live with a few lens defects, so it's better to know what they are and try to hide/avoid them.
You are 100% incorrect in regards to bokeh. Bokeh shape is completely subjective. 100% a matter of opinion. While many people hold the opinion that circular bokeh is favorable, the fact that one is more 'cinematic' or 'better' than another is complete and utter nonsense. I partly blame this misconception to inventive advertisers have many-a-time advertised round apertures as 'cinematic' when in reality nothing is cinematic. In fact, if we were to be forced to define what 'cinematic' is, I would argue 'cinematic' is simply the qualities we have recognized in the history of cinema. Things that we attribute to looking 'cinematic'. Thus I would further argue that the vast majority of cinema lenses in our 100+ years of making motion pictures have been varying degrees of polygonal bokeh.
I think the shape of bokeh has gained more attention in recent years due to the current trend of super shallow DoF. I would furthermore argue that if one looks at the following image and thinks it's the bokeh shape, circular or polygonal, which is the main concern for distracting a viewer from story, their perception is a little off.
Also,
Lowlight performance is a result of many different factors. Larger sensors have greater potential for better low light performance, but don't necessarily guarantee it. To simplify something that is far from simple, larger photo-sites provide better low light performance. It's really about the size of the photo-sites.
Last edited:
Patryk_Rebisz
Veteran
Cinematography is NOT about making images beautiful.
That's like saying that baking is about mixing water and flour. Yes, it's one of many (sometimes necessary) steps but that in itself doesn't make you a baker.
Moreover, you can watch short youtube demonstrations simplifying the basics of one's philosophy or one's art... but unless you go the the source and struggle to comprehend it yourself you will NEVER get the craft's spiritual essence. In other words, it IS magic.
That's like saying that baking is about mixing water and flour. Yes, it's one of many (sometimes necessary) steps but that in itself doesn't make you a baker.
Moreover, you can watch short youtube demonstrations simplifying the basics of one's philosophy or one's art... but unless you go the the source and struggle to comprehend it yourself you will NEVER get the craft's spiritual essence. In other words, it IS magic.
Samuel H
Major Contributor
I'm not resolution-obsessed, I'm DR-obsessed regarding cameras, and bokeh-obsessed regarding lenses.
But I don't worry a lot about the shape of bokeh either. Quality has a lot more to it than shape. If those out-of-focus circles have a strong halo around them and a lighter center, bokeh is bad, no matter how circular it is (because out-of-focus stuff will look distracting).
From my bokeh tests:
This is a perfect-bokeh lens, the Leitz Summilux-R 50mm f/1.4, both at f/1.4 and f/4:
(uniform and low-contrast, out of focus stuff just *disappears*)
And this is the Helios 44-2 58mm f/2.0, at f/2.0 and f/4.0
(much more circular when closed down, but also much more distracting in every case)
Also, for an extreme example: this is a bad copy of the Carl Zeiss Jena 180mm f/2.8 (a lens that usually has great bokeh, but not when you get a bad copy, as I unfortunately did) (f/2.8 and f/4):
(color aberrations that make out of focus stuff jump out, no matter how circular the highlights look)
edit: BTW, apart from this bokeh thing, great write-up
But I don't worry a lot about the shape of bokeh either. Quality has a lot more to it than shape. If those out-of-focus circles have a strong halo around them and a lighter center, bokeh is bad, no matter how circular it is (because out-of-focus stuff will look distracting).
From my bokeh tests:
This is a perfect-bokeh lens, the Leitz Summilux-R 50mm f/1.4, both at f/1.4 and f/4:
(uniform and low-contrast, out of focus stuff just *disappears*)
And this is the Helios 44-2 58mm f/2.0, at f/2.0 and f/4.0
(much more circular when closed down, but also much more distracting in every case)
Also, for an extreme example: this is a bad copy of the Carl Zeiss Jena 180mm f/2.8 (a lens that usually has great bokeh, but not when you get a bad copy, as I unfortunately did) (f/2.8 and f/4):
(color aberrations that make out of focus stuff jump out, no matter how circular the highlights look)
edit: BTW, apart from this bokeh thing, great write-up
Ebrahim Saadawi
Carbonite Member
Corrected the Bokeh area based on your information. Is it better now? (further corrections welcomed/wanted, I still haven't printed or sent it)
I want this to be as correct and easy to read as possible for a group of complete beginners.
Can anyway explain and give a few examples of what "Colour science of different manufacturers" mean or about "Motion Cadence" I always hear of but never understood. Those would be great additions!
I want this to be as correct and easy to read as possible for a group of complete beginners.
Can anyway explain and give a few examples of what "Colour science of different manufacturers" mean or about "Motion Cadence" I always hear of but never understood. Those would be great additions!
Ebrahim Saadawi
Carbonite Member
Yes Liam I completely forget mentioning anamorphic bokeh, I'll add that too. Thank you.
That's exactly what I want, corrections and additions.
morgan_moore
Major Contributor
Lenses..
The 'ugly' view is from the camera being close - not the lens angle of view.
If you photograph and crop/with two (say a 50 and a 24) lenses from the same distance the image is identical (apart from things like loss of res from cropping the wide image)
Yes wides tend to have barrel/moustache - but that is a separate issue from the critical determinant of 'ugly' with is subject to camera distance!
S
Personally I think people can be 'ugly' if photographed from too far away too as the compression of perspective is also unpleasant to some eyes
The 'ugly' view is from the camera being close - not the lens angle of view.
If you photograph and crop/with two (say a 50 and a 24) lenses from the same distance the image is identical (apart from things like loss of res from cropping the wide image)
Yes wides tend to have barrel/moustache - but that is a separate issue from the critical determinant of 'ugly' with is subject to camera distance!
S
Personally I think people can be 'ugly' if photographed from too far away too as the compression of perspective is also unpleasant to some eyes
Ebrahim Saadawi
Carbonite Member
Interesting perspective. But in order to get the same headshot for example, with a wider lens you need to get closer, and with a longer lens you need to go farther, therefore having distortion changes. Plus the fact that wide lenses have barrel distortion while longer lenses have pincusion distortion.
I do get you point: it's similar to saying that a larger sensor does not produce shallower DOF, but effectively it does, as with a larger sensor you need to get closer to get the same image, while with a smaller sensor you stand farther, so you focus more towards infinity, making you get deeper dof. So if you stand in the same position, use same lens, same everything, a larger sensor does not produce shallower dof. But in real usage it does, as it changes one of the elements that directly control DOF (distance)
I do get you point: it's similar to saying that a larger sensor does not produce shallower DOF, but effectively it does, as with a larger sensor you need to get closer to get the same image, while with a smaller sensor you stand farther, so you focus more towards infinity, making you get deeper dof. So if you stand in the same position, use same lens, same everything, a larger sensor does not produce shallower dof. But in real usage it does, as it changes one of the elements that directly control DOF (distance)
morgan_moore
Major Contributor
Yep indeed - but it is IMO important to separate the difference between bad lenses and perspective/FOV (although many wide lenses are indeed bad/distorted)
They, FOV/Perspective and lens distortions, are two separate considerations in cinematography.
For example I have nikkor 50 1.4 and 55 2.8 - similar FOV but the first is way more barrel distortion.
I also have a nikkor 18 classic and my sony 18-200 plastic - the former presents a much squarer image than the latter.
---
On a side note - I like to pick my distance and shoot from there to sustain constant relationship (or not) with the character - a reason I feel zooms are more cinematic than primes (no one will agree for the first 20 years of their career)
I also like to get close eg a ECU with a 50 not an 80 because it may be 'ugly' but it is also 'intimate'
S
They, FOV/Perspective and lens distortions, are two separate considerations in cinematography.
For example I have nikkor 50 1.4 and 55 2.8 - similar FOV but the first is way more barrel distortion.
I also have a nikkor 18 classic and my sony 18-200 plastic - the former presents a much squarer image than the latter.
---
On a side note - I like to pick my distance and shoot from there to sustain constant relationship (or not) with the character - a reason I feel zooms are more cinematic than primes (no one will agree for the first 20 years of their career)
I also like to get close eg a ECU with a 50 not an 80 because it may be 'ugly' but it is also 'intimate'
S
Motion quality, beyond just common rolling shutter artifacts, is a factor in high quality images too. H.264 4:2:0 AVC is a web streaming format and a lousy camera original format in most of its iterations. Poor motion quality is one of the reasons. It falls apart on motion in highly detailed scenes. There also is very little data depth for post production, especially for color and banding artifacts. There is a huge quality difference between 8 bit 4:2:2 broadcast i-frame codecs at 100mbps or higher and 4:2:0 H.264 long GOP codecs.
For me compression in any form is the enemy. The higher the compression, the lower the quality of an original even though it may not be casually visible on single stream playback. If you only judge video by what you see streamed via the web, you can't see the differences.
For me compression in any form is the enemy. The higher the compression, the lower the quality of an original even though it may not be casually visible on single stream playback. If you only judge video by what you see streamed via the web, you can't see the differences.
Hi Ebrahim, thanks for the write-up. Small correction (bit picky).
The 'p' in 1080p 720p 2160p stands for 'progressive scan', not pixels. The reason resolution is indicated by line count is that digital video wasn't taking photo after photo. each 'frame' would be cut into interlaced fields, I'm sure you're familiar with the artifacts this can cause in motion. 1080i means that every frame is an interlaced set of fields. p means each frame is an independent photo.
tv was in 'lines' like that. that carried over to digital cinema it seems. It's 1080, not 'under 2k'. Cinema is 4K measuring the horizontal length of the scan.
It's weird that both specs coming from different backgrounds are part of the same 'talk'. It does 1080, and 4k. I would think it would've been either 'it does 1080 and 2160p' or it does 4k, and a little under 2k.
it's the same as 24p. 24 progressive frames per second.
Nice picture for Banding example
Tarek
The 'p' in 1080p 720p 2160p stands for 'progressive scan', not pixels. The reason resolution is indicated by line count is that digital video wasn't taking photo after photo. each 'frame' would be cut into interlaced fields, I'm sure you're familiar with the artifacts this can cause in motion. 1080i means that every frame is an interlaced set of fields. p means each frame is an independent photo.
tv was in 'lines' like that. that carried over to digital cinema it seems. It's 1080, not 'under 2k'. Cinema is 4K measuring the horizontal length of the scan.
It's weird that both specs coming from different backgrounds are part of the same 'talk'. It does 1080, and 4k. I would think it would've been either 'it does 1080 and 2160p' or it does 4k, and a little under 2k.
it's the same as 24p. 24 progressive frames per second.
Nice picture for Banding example
Tarek
In addition to what greymog said (which wasn't picky at all; the "P" has nothing at all to do with "pixels"):
The number refers to the lines of resolution from top to bottom, not pixels. It's the vertical resolution of the image. A video image is made up of individual horizontal lines drawn from left to right on the screen. Sure, each line is a pixel high, but it's the lines which are referred to. Thinking of the number as just "vertical pixels" confuses the issue.
The number refers to the lines of resolution from top to bottom, not pixels. It's the vertical resolution of the image. A video image is made up of individual horizontal lines drawn from left to right on the screen. Sure, each line is a pixel high, but it's the lines which are referred to. Thinking of the number as just "vertical pixels" confuses the issue.
Ebrahim Saadawi
Carbonite Member
Yes I didn't say that the p stands for Pixel, it's progressive, but this article was meant to be as simplified as possible, just saying the number before the P you see in youtube (720p/1080p) stands forthe vertical resolution. I didn't want to get into what progressive and interlaced mean because that's separate article, or perhaps I should add it. Does anybody still shoot Interlaced and does it have many effects on IQ? It would be a worthy addition to the article if so.
David: so a video image is not made of individual pixels like a photograph, but complete lines of pixels?
David: so a video image is not made of individual pixels like a photograph, but complete lines of pixels?