Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video

Started 3 months ago | Discussions
Markr041 Veteran Member • Posts: 8,110
Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video
2

A "true" HDR video is a video with an extended color gamut and more color gradations - more colors - and greater dynamic range compared with standard definition (REC709, 8bit), which characterizes almost all monitors and TV's and video shot by most cameras. 12bit DNG RAW clips have the extended color gamut (REC202), and more color gradations (12bit), and the full dynamic range and resolution of the sensor (12.5 stops versus the standard of about 6).

And making an HDR video, in simple terms, just means using the original extended color gamut of the clips and at least 10bit and taking advantage of the increased dynamic range in rendering the video as HDR (there are categories in DaVinci Resolve). The only difficulty is you cannot see the video in HDR while editing (without special HDR viewing equipment), so you have to rely on scopes. I usually edit the video first in SDR, the make an HDR version by extending the dynamic range in the scopes and rendering appropriately.

In terms of viewing HDR:

a.  Most new cell phones will automatically show YouTube HDR videos in HDR.

b. Most new 4K TV's will also automatically play HDR YouTube videos in HDR.

But what if you cannot view the HDR video in HDR? Well, YouTube converts the HDR video for viewing in SDR. And it still looks pretty good. You see that converted version if you cannot see HDR.

If you just click on these HDR videos here, you will see them in SDR for sure. If instead  you see them in the YouTube app or at the YouTube web site (by clicking in the frame on the video title), you will see them in HDR, if you have an HDR viewing device (eg, cell phone). The difference is like night and day, almost literally - as the highlights will beam. It is not like comparing 4K to HD, which is actually subtle if the HD is good.  It is really different. And HDR is well-accepted in professional circles (unlike 60 fps!) Most new Netflix series and movies are HDR, as is much Amazon PrimeVideo material.

So, here are two true Sigma fp HDR videos, made from 12bit DNG RAW sources:

 Markr041's gear list:Markr041's gear list
Panasonic Lumix DMC-FZ200 Panasonic LX100 Sony RX100 IV Panasonic ZS100 Olympus TG-5 +12 more
xpatUSA
xpatUSA Forum Pro • Posts: 17,680
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video

Markr041 wrote:

A "true" HDR video is a video with an extended color gamut and more color gradations - more colors - and greater dynamic range compared with standard definition (REC709, 8bit), which characterizes almost all monitors and TV's and video shot by most cameras. 12bit DNG RAW clips have the extended color gamut (REC202), and more color gradations (12bit) ...

It's "Rec. 2020".

But what percentage of people here own output devices that have the Rec. 2020 primary colors of 630 nm for the red, 532 nm for the green, and 467 nm for the blue?

https://en.wikipedia.org/wiki/Rec._2020

...and the full dynamic range and resolution of the sensor (12.5 stops versus the standard of about 6).

My cheapish NEC sRGB monitor is capable of 1000:1 real contrast, just under a DR of 10EV, my SD9 had a measure DR of 9EV (Bill Claff), so where does "the standard" of 6EV come from? Please explain ...

<>

Are you familiar with "just noticeable color difference"?

https://en.wikipedia.org/wiki/Color_difference

Discerning a 1-bit difference in 8-bit sRGB is virtually impossible, so how would anyone appreciate the "more color gradations" in 12-bit that you are promoting above?

-- hide signature --

There are more ways than one to skin a cat ...
Seba Smith - The Money Diggers, 1840
Ted

 xpatUSA's gear list:xpatUSA's gear list
Panasonic Lumix DMC-LX1 Panasonic Lumix DMC-G1 Sigma SD9 Sigma SD15 Panasonic Lumix DC-G9 +13 more
OP Markr041 Veteran Member • Posts: 8,110
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video

xpatUSA wrote:

Markr041 wrote:

A "true" HDR video is a video with an extended color gamut and more color gradations - more colors - and greater dynamic range compared with standard definition (REC709, 8bit), which characterizes almost all monitors and TV's and video shot by most cameras. 12bit DNG RAW clips have the extended color gamut (REC202), and more color gradations (12bit) ...

It's "Rec. 2020".

But what percentage of people here own output devices that have the Rec. 2020 primary colors of 630 nm for the red, 532 nm for the green, and 467 nm for the blue?

https://en.wikipedia.org/wiki/Rec._2020

...and the full dynamic range and resolution of the sensor (12.5 stops versus the standard of about 6).

My cheapish NEC sRGB monitor is capable of 1000:1 real contrast, just under a DR of 10EV, my SD9 had a measure DR of 9EV (Bill Claff), so where does "the standard" of 6EV come from? Please explain ...

<>

Are you familiar with "just noticeable color difference"?

https://en.wikipedia.org/wiki/Color_difference

Discerning a 1-bit difference in 8-bit sRGB is virtually impossible, so how would anyone appreciate the "more color gradations" in 12-bit that you are promoting above?

This poster, besides correcting a typo - thanks!, questions where the 5-6 stops of Rec. 709 (SDR) video I claim come from, and he also claims that having more than 8bit color is not noticeable, nor is having a very wide color gamut. I guess he is questioning any claim that HDR video is visibly superior to Rec. 709 or SDR video. This is based on ignorance of the theory, ignorance of video standards, and has to also come from never having actually viewed an HDR video in HDR. I know of no one who cannot notice the difference.

Where does the REC. 709 standard come from and is it relevant? Rec. 709 is a true, adhered-to standard. It was set by the ITU-R in 1990 *for all HDTVs*. You can look up what the ITU-R is and where it is headquartered (hint: Geneva). This standard is the default in almost all video cameras. It is the standard for blu-rays. The standard limits the both dynamic range and the colors you can see - and you can see the difference when those limits are exceeded.

The REC. 709 standard specifies that there are 5 stops of dynamic range. It is slightly more than that with superwhites. It does not matter what one's display is capable of, video shot in ITU709 or Rec. 709 has only 5-6 stops of dynamic range. That is a specification. HDR video allows many more stops than that. Almost all the video you see has 5-6 stops of dynamic range even if your camera's sensor has 15 stops of DR or your display can show 12 stops.

See

http://www.xdcam-user.com/2012/04/rec-709-a-true-rec-709-camera-should-only-have-5-stops-of-dynamic-range/#:~:text=In%20theory%20if%20two%20cameras%20are%20both%20set,stops%20of%20dynamic%20range%20from%200%20to%20100%25.

As to the number of color bits. Look up "banding." The sky, for example, is not one uniform shade of blue, it has many gradations, but 8bit = 256 shades per pixel, so a very pure blue sky can only have 256 shades in a Rec. 709 video, but in reality it has a lot more. And in 8bit color all those gradations are put into visible bands instead of finely differentiated shades.

See: https://www.rtings.com/tv/tests/picture-quality/gradient

It is true that very few displays can show all those Rec. 2020 (ok?) colors (now we are talking about number of different colors, not shades of a single color). But many displays now certainly can show a wider color gamut than that specified by Rec. 709 (DCI P3 is another standard color gamut that is larger than that for Rec. 709 that many displays can provide).

So, with HDR you can actually make use of those extra stops of your camera or your display and you can see more colors (not just more shades but actually more colors, up to the limits of the human eye (which are actually less)).

Note: To be even more thorough: one will not see a difference between an HDR and an SDR version if the actual scene has no more than 5-6 stops of dynamic range. That should be obvious, but I say this just in case someone tries to come up with a video in which you cannot see any difference. The videos I rendered in HDR and chose to post are specifically ones where the actual dynamic range of the scenes is very high.

 Markr041's gear list:Markr041's gear list
Panasonic Lumix DMC-FZ200 Panasonic LX100 Sony RX100 IV Panasonic ZS100 Olympus TG-5 +12 more
xpatUSA
xpatUSA Forum Pro • Posts: 17,680
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video

Markr041 wrote:

xpatUSA wrote:

Markr041 wrote:

A "true" HDR video is a video with an extended color gamut and more color gradations - more colors - and greater dynamic range compared with standard definition (REC709, 8bit), which characterizes almost all monitors and TV's and video shot by most cameras. 12bit DNG RAW clips have the extended color gamut (REC202), and more color gradations (12bit) ...

It's "Rec. 2020".

But what percentage of people here own output devices that have the Rec. 2020 primary colors of 630 nm for the red, 532 nm for the green, and 467 nm for the blue?

https://en.wikipedia.org/wiki/Rec._2020

...and the full dynamic range and resolution of the sensor (12.5 stops versus the standard of about 6).

My cheapish NEC sRGB monitor is capable of 1000:1 real contrast, just under a DR of 10EV, my SD9 had a measure DR of 9EV (Bill Claff), so where does "the standard" of 6EV come from? Please explain ...

<>

Are you familiar with "just noticeable color difference"?

https://en.wikipedia.org/wiki/Color_difference

Discerning a 1-bit difference in 8-bit sRGB is virtually impossible, so how would anyone appreciate the "more color gradations" in 12-bit that you are promoting above?

This poster, besides correcting a typo - thanks!, questions where the 5-6 stops of Rec. 709 (SDR) video I claim come from, and he also claims that having more than 8bit color is not noticeable, nor is having a very wide color gamut. I guess he is questioning any claim that HDR video is visibly superior to Rec. 709 or SDR video. This is based on ignorance of the theory, ignorance of video standards, and has to also come from never having actually viewed an HDR video in HDR. I know of no one who cannot notice the difference.

Where does the REC. 709 standard come from and is it relevant? Rec. 709 is a true, adhered-to standard. It was set by the ITU-R in 1990 *for all HDTVs*. You can look up what the ITU-R is and where it is headquartered (hint: Geneva). This standard is the default in almost all video cameras. It is the standard for blu-rays. The standard limits the both dynamic range and the colors you can see - and you can see the difference when those limits are exceeded.

The REC. 709 standard specifies that there are 5 stops of dynamic range. It is slightly more than that with superwhites. It does not matter what one's display is capable of, video shot in ITU709 or Rec. 709 has only 5-6 stops of dynamic range. That is a specification. HDR video allows many more stops than that. Almost all the video you see has 5-6 stops of dynamic range even if your camera's sensor has 15 stops of DR or your display can show 12 stops.

See

http://www.xdcam-user.com/2012/04/rec-709-a-true-rec-709-camera-should-only-have-5-stops-of-dynamic-range/#:~:text=In%20theory%20if%20two%20cameras%20are%20both%20set,stops%20of%20dynamic%20range%20from%200%20to%20100%25.

As to the number of color bits. Look up "banding." The sky, for example, is not one uniform shade of blue, it has many gradations, but 8bit = 256 shades per pixel, so a very pure blue sky can only have 256 shades in a Rec. 709 video, but in reality it has a lot more. And in 8bit color all those gradations are put into visible bands instead of finely differentiated shades.

See: https://www.rtings.com/tv/tests/picture-quality/gradient

It is true that very few displays can show all those Rec. 2020 (ok?) colors (now we are talking about number of different colors, not shades of a single color). But many displays now certainly can show a wider color gamut than that specified by Rec. 709 (DCI P3 is another standard color gamut that is larger than that for Rec. 709 that many displays can provide).

So, with HDR you can actually make use of those extra stops of your camera or your display and you can see more colors (not just more shades but actually more colors, up to the limits of the human eye (which are actually less)).

Note: To be even more thorough: one will not see a difference between an HDR and an SDR version if the actual scene has no more than 5-6 stops of dynamic range. That should be obvious, but I say this just in case someone tries to come up with a video in which you cannot see any difference. The videos I rendered in HDR and chose to post are specifically ones where the actual dynamic range of the scenes is very high.

A lot of words and no specific references as regards Rec. 709.

From the horses mouth, not one mention of "dynamic range":

https://www.itu.int/dms_pubrec/itu-r/rec/bt/R-REC-BT.709-6-201506-I!!PDF-E.pdf

From perhaps the most respected in the field:

"The dynamic range associated with code 1 is close to a million to one, not just 1/255."

http://poynton.ca/notes/PU-PR-IS/Poynton-PU-PR-IS.pdf

It would appear that video lives in a completely different world than that of normal digital imaging, so I'll bow out from this discussion before I get drowned in a sea of verbiage.

-- hide signature --

There are more ways than one to skin a cat ...
Seba Smith - The Money Diggers, 1840
Ted

 xpatUSA's gear list:xpatUSA's gear list
Panasonic Lumix DMC-LX1 Panasonic Lumix DMC-G1 Sigma SD9 Sigma SD15 Panasonic Lumix DC-G9 +13 more
OP Markr041 Veteran Member • Posts: 8,110
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video
2

xpatUSA wrote:

Markr041 wrote:

xpatUSA wrote:

Markr041 wrote:

A "true" HDR video is a video with an extended color gamut and more color gradations - more colors - and greater dynamic range compared with standard definition (REC709, 8bit), which characterizes almost all monitors and TV's and video shot by most cameras. 12bit DNG RAW clips have the extended color gamut (REC202), and more color gradations (12bit) ...

It's "Rec. 2020".

But what percentage of people here own output devices that have the Rec. 2020 primary colors of 630 nm for the red, 532 nm for the green, and 467 nm for the blue?

https://en.wikipedia.org/wiki/Rec._2020

...and the full dynamic range and resolution of the sensor (12.5 stops versus the standard of about 6).

My cheapish NEC sRGB monitor is capable of 1000:1 real contrast, just under a DR of 10EV, my SD9 had a measure DR of 9EV (Bill Claff), so where does "the standard" of 6EV come from? Please explain ...

<>

Are you familiar with "just noticeable color difference"?

https://en.wikipedia.org/wiki/Color_difference

Discerning a 1-bit difference in 8-bit sRGB is virtually impossible, so how would anyone appreciate the "more color gradations" in 12-bit that you are promoting above?

This poster, besides correcting a typo - thanks!, questions where the 5-6 stops of Rec. 709 (SDR) video I claim come from, and he also claims that having more than 8bit color is not noticeable, nor is having a very wide color gamut. I guess he is questioning any claim that HDR video is visibly superior to Rec. 709 or SDR video. This is based on ignorance of the theory, ignorance of video standards, and has to also come from never having actually viewed an HDR video in HDR. I know of no one who cannot notice the difference.

Where does the REC. 709 standard come from and is it relevant? Rec. 709 is a true, adhered-to standard. It was set by the ITU-R in 1990 *for all HDTVs*. You can look up what the ITU-R is and where it is headquartered (hint: Geneva). This standard is the default in almost all video cameras. It is the standard for blu-rays. The standard limits the both dynamic range and the colors you can see - and you can see the difference when those limits are exceeded.

The REC. 709 standard specifies that there are 5 stops of dynamic range. It is slightly more than that with superwhites. It does not matter what one's display is capable of, video shot in ITU709 or Rec. 709 has only 5-6 stops of dynamic range. That is a specification. HDR video allows many more stops than that. Almost all the video you see has 5-6 stops of dynamic range even if your camera's sensor has 15 stops of DR or your display can show 12 stops.

See

http://www.xdcam-user.com/2012/04/rec-709-a-true-rec-709-camera-should-only-have-5-stops-of-dynamic-range/#:~:text=In%20theory%20if%20two%20cameras%20are%20both%20set,stops%20of%20dynamic%20range%20from%200%20to%20100%25.

As to the number of color bits. Look up "banding." The sky, for example, is not one uniform shade of blue, it has many gradations, but 8bit = 256 shades per pixel, so a very pure blue sky can only have 256 shades in a Rec. 709 video, but in reality it has a lot more. And in 8bit color all those gradations are put into visible bands instead of finely differentiated shades.

See: https://www.rtings.com/tv/tests/picture-quality/gradient

It is true that very few displays can show all those Rec. 2020 (ok?) colors (now we are talking about number of different colors, not shades of a single color). But many displays now certainly can show a wider color gamut than that specified by Rec. 709 (DCI P3 is another standard color gamut that is larger than that for Rec. 709 that many displays can provide).

So, with HDR you can actually make use of those extra stops of your camera or your display and you can see more colors (not just more shades but actually more colors, up to the limits of the human eye (which are actually less)).

Note: To be even more thorough: one will not see a difference between an HDR and an SDR version if the actual scene has no more than 5-6 stops of dynamic range. That should be obvious, but I say this just in case someone tries to come up with a video in which you cannot see any difference. The videos I rendered in HDR and chose to post are specifically ones where the actual dynamic range of the scenes is very high.

A lot of words and no specific references as regards Rec. 709.

From the horses mouth, not one mention of "dynamic range":

https://www.itu.int/dms_pubrec/itu-r/rec/bt/R-REC-BT.709-6-201506-I!!PDF-E.pdf

From perhaps the most respected in the field:

"The dynamic range associated with code 1 is close to a million to one, not just 1/255."

http://poynton.ca/notes/PU-PR-IS/Poynton-PU-PR-IS.pdf

It would appear that video lives in a completely different world than that of normal digital imaging, so I'll bow out from this discussion before I get drowned in a sea of verbiage.

You are correct. You know nothing about video. You dont know what you are talking about. Your links are bizarre

You don't shoot video, you can't view HDR video, and you dont want to learn. I gave you links about the dynamic range limits of Rec. 709. About the limits of 8bit. You do not need to thank me for trying to educate you about video. Please stick around and do not be afraid of being ignorant so you can learn.

 Markr041's gear list:Markr041's gear list
Panasonic Lumix DMC-FZ200 Panasonic LX100 Sony RX100 IV Panasonic ZS100 Olympus TG-5 +12 more
xpatUSA
xpatUSA Forum Pro • Posts: 17,680
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video

Markr041 wrote:

xpatUSA wrote:

It would appear that video lives in a completely different world than that of normal digital imaging, so I'll bow out from this discussion before I get drowned in a sea of verbiage.

You are correct. You know nothing about video. You dont know what you are talking about. Your links are bizarre

You don't shoot video, you can't view HDR video, and you don't want to learn. I gave you links about the dynamic range limits of Rec. 709. About the limits of 8bit. You do not need to thank me for trying to educate you about video. Please stick around and do not be afraid of being ignorant so you can learn.

That's pretty rude. "Flaming" complaint filed. Goodbye, ventral orifice.

-- hide signature --

There are more ways than one to skin a cat ...
Seba Smith - The Money Diggers, 1840
Ted

 xpatUSA's gear list:xpatUSA's gear list
Panasonic Lumix DMC-LX1 Panasonic Lumix DMC-G1 Sigma SD9 Sigma SD15 Panasonic Lumix DC-G9 +13 more
OP Markr041 Veteran Member • Posts: 8,110
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video
2

xpatUSA wrote:

Markr041 wrote:

xpatUSA wrote:

It would appear that video lives in a completely different world than that of normal digital imaging, so I'll bow out from this discussion before I get drowned in a sea of verbiage.

You are correct. You know nothing about video. You dont know what you are talking about. Your links are bizarre

You don't shoot video, you can't view HDR video, and you don't want to learn. I gave you links about the dynamic range limits of Rec. 709. About the limits of 8bit. You do not need to thank me for trying to educate you about video. Please stick around and do not be afraid of being ignorant so you can learn.

That's pretty rude. "Flaming" complaint filed. Goodbye, ventral orifice.

Actually I agree that this was too rude, and I was editing it completely, taking my time to be careful, when I got stopped from saving the new reply because you replied! I couldn't change it. Sorry about that. You did lose some sympathy by your juvenile name calling, though (I do understand the reaction).

I would be perfectly happy if that post was removed. And my apology is sincere; no matter how rude and aggressive and purposely obscuring you were, you did not deserve that response.Along with taking out the rude bits - my overeaction to a video flat earther - I was obtaining more links about the dynamic range of Rec. 709.

More links:

https://www.cinema5d.com/the-link-between-hdr-sdr-and-hlg-explained/

https://www.google.com/url?sa=t&source=web&rct=j&url=https://pro.sony/s3/cms-static-content/file/15/1237494951415.pdf&ved=2ahUKEwj54KiKhODqAhWomHIEHYHXAW8QFjACegQIDRAI&usg=AOvVaw0FyJHUTH688eg0whIU7IOD

https://en.m.wikipedia.org/wiki/Standard-dynamic-range_video

 Markr041's gear list:Markr041's gear list
Panasonic Lumix DMC-FZ200 Panasonic LX100 Sony RX100 IV Panasonic ZS100 Olympus TG-5 +12 more
SigmaChrome Forum Pro • Posts: 12,715
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video
2

xpatUSA wrote:

Markr041 wrote:

xpatUSA wrote:

It would appear that video lives in a completely different world than that of normal digital imaging, so I'll bow out from this discussion before I get drowned in a sea of verbiage.

You are correct. You know nothing about video. You dont know what you are talking about. Your links are bizarre

You don't shoot video, you can't view HDR video, and you don't want to learn. I gave you links about the dynamic range limits of Rec. 709. About the limits of 8bit. You do not need to thank me for trying to educate you about video. Please stick around and do not be afraid of being ignorant so you can learn.

That's pretty rude. "Flaming" complaint filed. Goodbye, ventral orifice.

And you're  filing a complaint, Ted? Wow!

-- hide signature --

Regards,
Vitée
Capture all the light and colour!
http://www.pbase.com/vitee/galleries

 SigmaChrome's gear list:SigmaChrome's gear list
Sigma DP1 Sigma DP2 Merrill Sony RX100 Sigma DP3 Merrill Sigma dp0 Quattro +35 more
xpatUSA
xpatUSA Forum Pro • Posts: 17,680
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video

SigmaChrome wrote:

xpatUSA wrote:

Markr041 wrote:

xpatUSA wrote:

It would appear that video lives in a completely different world than that of normal digital imaging, so I'll bow out from this discussion before I get drowned in a sea of verbiage.

You are correct. You know nothing about video. You dont know what you are talking about. Your links are bizarre

You don't shoot video, you can't view HDR video, and you don't want to learn. I gave you links about the dynamic range limits of Rec. 709. About the limits of 8bit. You do not need to thank me for trying to educate you about video. Please stick around and do not be afraid of being ignorant so you can learn.

That's pretty rude. "Flaming" complaint filed. Goodbye, ventral orifice.

And you're filing a complaint, Ted? Wow!

always ready to jump in, eh?

-- hide signature --

There are more ways than one to skin a cat ...
Seba Smith - The Money Diggers, 1840
Ted

 xpatUSA's gear list:xpatUSA's gear list
Panasonic Lumix DMC-LX1 Panasonic Lumix DMC-G1 Sigma SD9 Sigma SD15 Panasonic Lumix DC-G9 +13 more
Jozef M Senior Member • Posts: 1,710
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video
2

SigmaChrome wrote:

xpatUSA wrote:

Markr041 wrote:

xpatUSA wrote:

It would appear that video lives in a completely different world than that of normal digital imaging, so I'll bow out from this discussion before I get drowned in a sea of verbiage.

You are correct. You know nothing about video. You dont know what you are talking about. Your links are bizarre

You don't shoot video, you can't view HDR video, and you don't want to learn. I gave you links about the dynamic range limits of Rec. 709. About the limits of 8bit. You do not need to thank me for trying to educate you about video. Please stick around and do not be afraid of being ignorant so you can learn.

That's pretty rude. "Flaming" complaint filed. Goodbye, ventral orifice.

And you're filing a complaint, Ted? Wow!

Yes, I agree with this opinion too ... The OP has a good point.

Jozef.

 Jozef M's gear list:Jozef M's gear list
Sigma DP3 Merrill Canon EOS 30D Panasonic Lumix DC-G9
tagscuderia
tagscuderia Senior Member • Posts: 1,695
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video
1

Hi Mark, your response to Ted was utterly uncalled for and comparable to a previous thread’s response to myself. From what you've written thus far, it appears that you yourself are learning, and excited about the SIGMA fp, which is great but your replies are confrontational − if people bother to take the time to reply, I'm sure that it's to help you or to educate themselves.

One small technical remark: Cinema DNG does not have a colour space per se i.e. Rec. 2020, and the gamut of the fp may or may not exceed that of Rec. 2020. It is our job as photographers/graders to interpret the data to fit within the confines of our output medium, just as you're doing with LDR and HDR.

Will you be upgrading your display for editing HDR? When I last upgraded, most displays advertised as HDR... were not e.g. DisplayHDR 400

P.S. I preferred the Black Point in your LDR video but it's a compelling comparison! I have calibrated my TV, separately for both LDR and HDR.

 tagscuderia's gear list:tagscuderia's gear list
Sigma DP3 Merrill Sigma SD1 Merrill Sigma 18-35mm F1.8 DC HSM Art Sigma 50-100mm F1.8 DC HSM Art
xpatUSA
xpatUSA Forum Pro • Posts: 17,680
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video
1

Markr041 wrote:

You know nothing about video. You [don't] know what you are talking about. Your links are bizarre.

You don't shoot video, you can't view HDR video, and you [don't] want to learn. I gave you links about the dynamic range limits of Rec. 709.

You gave me links to opinions that refer to Rec. 709.

Read this again and tell me where it SPECIFIES a dynamic range limit or even mentions dynamic range.

https://www.itu.int/dms_pubrec/itu-r/rec/bt/R-REC-BT.709-6-201506-I!!PDF-E.pdf

Here's a link to Rec. 2020:

https://www.itu.int/dms_pubrec/itu-r/rec/bt/R-REC-BT.2020-2-201510-I!!PDF-E.pdf

It does not SPECIFY any dynamic range either. Dynamic range is only mentioned in passing with reference to future output devices:

"ultra-high definition television (UHDTV) is expected to become available in the near future with, inter alia, larger screens, higher spatial/temporal resolution, wider colour gamut, wider dynamic range, etc. taking into account developments of display technology"

So please refrain from misleading this forum by stating or even implying that the recommendations specify a dynamic range limit.

About the limits of 8bit. You do not need to thank me for trying to educate you about video. Please stick around and do not be afraid of being ignorant so you can learn.

Note to Members:

In spite of the OP's dire warnings about blue sky 8-bit color and consequent banding, here is an 8-bit sRGB blue gradient. (Rec. 709 blue primary):

Pure linear blue, running from about 4% HSV value up to 100% at top, 1080px from top to bottom. Anyone see dire horizontal banding? Apart from the OP, that is ...

-- hide signature --

There are more ways than one to skin a cat ...
Seba Smith - The Money Diggers, 1840
Ted

 xpatUSA's gear list:xpatUSA's gear list
Panasonic Lumix DMC-LX1 Panasonic Lumix DMC-G1 Sigma SD9 Sigma SD15 Panasonic Lumix DC-G9 +13 more
bizi clop
bizi clop Regular Member • Posts: 428
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video

xpatUSA wrote:

Note to Members:

In spite of the OP's dire warnings about blue sky 8-bit color and consequent banding, here is an 8-bit sRGB blue gradient. (Rec. 709 blue primary):

Pure linear blue, running from about 4% HSV value up to 100% at top, 1080px from top to bottom. Anyone see dire horizontal banding? Apart from the OP, that is ...

Well, there is a good reason why Stellarium includes a dithering option, up to 10bits.

https://github.com/Stellarium/stellarium/issues/131
"Color banding artifacts make atmosphere ugly in low light #131"

Here I made an example from two otherwise identical screenshots, one half is dithered, the other is not. (To be viewed only at 100%.)

Here I stretched the contrast with Levels to make it more visible, and also put the picture of the relevant settings:

So I think 8 bits are not always enough.

(BTW, it's Bayer dithering applied here, invented by the same Bryce Bayer responsible for the Bayer mosaic.)

 bizi clop's gear list:bizi clop's gear list
Canon PowerShot G2 Sigma DP2 Merrill Canon PowerShot SX50 HS
xpatUSA
xpatUSA Forum Pro • Posts: 17,680
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video

bizi clop wrote:

xpatUSA wrote:

Note to Members:

In spite of the OP's dire warnings about blue sky 8-bit color and consequent banding, here is an 8-bit sRGB blue gradient. (Rec. 709 blue primary):

Pure linear blue, running from about 4% HSV value up to 100% at top, 1080px from top to bottom. Anyone see dire horizontal banding? Apart from the OP, that is ...

Well, there is a good reason why Stellarium includes a dithering option, up to 10bits.

https://github.com/Stellarium/stellarium/issues/131
"Color banding artifacts make atmosphere ugly in low light #131"

Here I made an example from two otherwise identical screenshots, one half is dithered, the other is not. (To be viewed only at 100%.)

Here I stretched the contrast with Levels to make it more visible, and also put the picture of the relevant settings:

So I think 8 bits are not always enough.

Nice example of banding!

Quite right, it doesn't take much post-processing for banding to show up in the 8-bit world - but I do not recall any mention of such post-processing per se by the OP.

(BTW, it's Bayer dithering applied here, invented by the same Bryce Bayer responsible for the Bayer mosaic.)

When I made my gradient, I did switch off dithering in the GIMP and of course I set the image precision to 8-bit integer so as to preclude any obfuscation.

In ImageJ a vertical profile plot of my blue gradient does reveal banding of more than one pixel - as does the blue channel histogram. However that banding is invisible in the posted view on my monitor and would be even more so had I produced a perfect gradient - which belies the OP's claim quite handsomely.

-- hide signature --

There are more ways than one to skin a cat ...
Seba Smith - The Money Diggers, 1840
Ted

 xpatUSA's gear list:xpatUSA's gear list
Panasonic Lumix DMC-LX1 Panasonic Lumix DMC-G1 Sigma SD9 Sigma SD15 Panasonic Lumix DC-G9 +13 more
tagscuderia
tagscuderia Senior Member • Posts: 1,695
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video
1

bizi clop wrote:

... So I think 8 bits are not always enough.

But then you have to take gamma into consideration! Lifting shadows on an image that has gamma correction hard encoded is a bad idea.

Also 8-bit sRGB in general is perfectly good (8-bit ProPhoto, not so much), but it's an output file (both JPG and TIFF) not designed for manipulation.

 tagscuderia's gear list:tagscuderia's gear list
Sigma DP3 Merrill Sigma SD1 Merrill Sigma 18-35mm F1.8 DC HSM Art Sigma 50-100mm F1.8 DC HSM Art
OP Markr041 Veteran Member • Posts: 8,110
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video

tagscuderia wrote:

Hi Mark, your response to Ted was utterly uncalled for and comparable to a previous thread’s response to myself. From what you've written thus far, it appears that you yourself are learning, and excited about the SIGMA fp, which is great but your replies are confrontational − if people bother to take the time to reply, I'm sure that it's to help you or to educate themselves.

One small technical remark: Cinema DNG does not have a colour space per se i.e. Rec. 2020, and the gamut of the fp may or may not exceed that of Rec. 2020. It is our job as photographers/graders to interpret the data to fit within the confines of our output medium, just as you're doing with LDR and HDR.

Will you be upgrading your display for editing HDR? When I last upgraded, most displays advertised as HDR... were not e.g. DisplayHDR 400

P.S. I preferred the Black Point in your LDR video but it's a compelling comparison! I have calibrated my TV, separately for both LDR and HDR.

I am not sure what you mean that DNG RAW does not have a color space "per se." The sensor has some color gamut. We need to inform Resolve  what the gamut is so it can make the correct transform to, say, rec. 709. There are only pre-sets. But I am happy to learn more.

It is good now that displays have the standardized HDR rating so one can distinguish which are truly HDR-capable.

I have already apologized for the response to the person who was questioning basic facts that I could not remove. However, the person I was responding to was not in any way trying to be helpful and at the end resorted to name-calling, revealing his true character. And I was not defending the fp, but rather truth and my integrity. But let's not rehash that.

 Markr041's gear list:Markr041's gear list
Panasonic Lumix DMC-FZ200 Panasonic LX100 Sony RX100 IV Panasonic ZS100 Olympus TG-5 +12 more
xpatUSA
xpatUSA Forum Pro • Posts: 17,680
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video
1

Markr041 wrote:

tagscuderia wrote:

One small technical remark: Cinema DNG does not have a colour space per se i.e. Rec. 2020, and the gamut of the fp may or may not exceed that of Rec. 2020.

I am not sure what you mean that DNG RAW does not have a color space "per se." The sensor has some color gamut. I am happy to learn more.

No, the sensor does not have a color gamut. Don't take my word for it, please read this:

https://www.provideocoalition.com/camera-not-color-gamut/

You may find it helpful ** in avoiding statements like the above.

I have already apologized for the response to the person who was questioning basic facts that I could not remove. However, the person I was responding to was not in any way trying to be helpful and at the end resorted to name-calling, revealing his true character. And I was not defending the fp, but rather truth and my integrity. But let's not rehash that.

You just did!

** hope springs eternal in the human breast ...
Ted

 xpatUSA's gear list:xpatUSA's gear list
Panasonic Lumix DMC-LX1 Panasonic Lumix DMC-G1 Sigma SD9 Sigma SD15 Panasonic Lumix DC-G9 +13 more
OP Markr041 Veteran Member • Posts: 8,110
Re: Sigma fp: What's the Point of 12bit DNG RAW? One is True HDR Video

xpatUSA wrote:

Markr041 wrote:

tagscuderia wrote:

One small technical remark: Cinema DNG does not have a colour space per se i.e. Rec. 2020, and the gamut of the fp may or may not exceed that of Rec. 2020.

I am not sure what you mean that DNG RAW does not have a color space "per se." The sensor has some color gamut. I am happy to learn more.

No, the sensor does not have a color gamut. Don't take my word for it, please read this:

https://www.provideocoalition.com/camera-not-color-gamut/

You may find it helpful ** in avoiding statements like the above.

I have already apologized for the response to the person who was questioning basic facts that I could not remove. However, the person I was responding to was not in any way trying to be helpful and at the end resorted to name-calling, revealing his true character. And I was not defending the fp, but rather truth and my integrity. But let's not rehash that.

You just did!

** hope springs eternal in the human breast ...
Ted

Thanks, that was helpful (the first part).

 Markr041's gear list:Markr041's gear list
Panasonic Lumix DMC-FZ200 Panasonic LX100 Sony RX100 IV Panasonic ZS100 Olympus TG-5 +12 more
Keyboard shortcuts:
FForum MMy threads