Thanks for the explanation about switching to DCI-P3/REC2020.
I don’t understand though how a display could score so well in HDR color volume while doing only so so in SDR. Can you elaborate?
Hi chairman_mao,
Great question!
Aside from differences between the ITP and CIELAB color spaces, much of the discrepancy comes down to how we score HDR vs. SDR Color Volume.
In the case of HDR Color Volume, 100% of the score is based on a TV’s color performance when receiving a 10 000 cd/m² BT.2020 signal. As you may expect, covering 100% of BT.2020 @ 10 000 cd/m² isn’t something we’re likely to see anytime soon in any consumer display. Most TVs have a rather limited coverage of 10 000 cd/m² BT.2020; in fact, the best we’ve tested so far is the Samsung S95D QD-OLED and it got 59.4%. The scoring spline used for HDR Color Volume is thus very lenient.
On the other hand, for SDR Color Volume, we test at a reference white point of 100 cd/m², which any TV can achieve. The scoring splines used for CIELAB DCI-P3 and BT.2020 are thus significantly harsher than the one we use for HDR Color Volume, since we should expect most TVs to at least perform acceptably in this test.
So in short, the apparent scoring imbalance often reflects how we tune our expectations: our SDR Color Volume testing assumes mature performance, while our HDR Color Volume testing still accommodates significant room for growth. As display tech keeps improving, we plan to revisit the HDR Color Volume scoring to gradually make it more demanding.
Hope this clears it up! Feel free to reach out if you have more questions or suggestions.
Rtings,
I like the Gamut Ring measurement idea and commend the update. Despite the explanation, I don’t quite understand why you are measuring DCI-P3 and Rec2020 for SDR color volume. It seems like BT601 and BT709 color space would make more sense for SDR, and that most readers might confuse the Gamut Ring as representing the HDR color space.
Admittedly, it’s possible one could be watching 4K SDR content with wide color, but that seems less frequent than SD or HD content. I don’t have a sense of how many SDR video games would use wide color – can you provide some context and examples?
It also makes sense to cap the brightness at 100nits (though it does seem unlikely that a user will do the same) as some HDR TVs do not maximize the SDR brightness at the same level as HDR, but in a TV that offers local dimming, is that 100 nits measured on a small window, which represents a peak brightness, or on a full field white?
Thanks,
Hi Amazon_Fan,
Thanks for taking the time to reach out and share your thoughts on our new SDR Color Volume test—we really appreciate your detailed feedback and questions!
You’re absolutely right that BT.601 and BT.709 are the standard color gamuts for SDR content. However, one of the key reasons we opted to use DCI-P3 and BT.2020 for our new SDR Color Volume test is that modern TVs already cover BT.709 so completely—often 100%—that measuring within that gamut offers little to no differentiation between models, even lower-end ones. In contrast, using larger color gamuts like P3 and BT.2020 allows us to highlight differences in color rendering capabilities between displays in SDR picture modes.
It’s true that traditional SDR content rarely exceeds BT.709. But there are still relevant scenarios where a TV might display colors outside of BT.709 in SDR picture modes. For instance, some users choose to set their TV to a wider gamut manually by personal preference, a notable one is when gaming in SDR to get more saturated colors (e.g., setting the TV to P3 color when using a Nintendo Switch which outputs BT.709). In regular video content, it’s possible that some pixels on screen may fall beyond the limits of the BT.709 color space, depending on how it was encoded. This is something we also often see in HDR, where a lot of content mastered in P3 will have pixels falling into BT.2020 territory. We’ve updated the test’s tooltip to make this nuance clearer—thanks for bringing it to our attention!
The brightness cap at 100 cd/m² is intentional since it’s the reference diffuse white point for SDR content. More importantly however is the fact that it’s a luminance output any TV can easily reach, which is crucial to ensure the comparability of our gamut rings results. At different diffuse white luminance points, results lose their comparability as the measured coverage of the color spaces will differ. We measure it with a test window with constant APL to ensure brightness stability across TVs, especially OLED models.
At a broader level, the motivation behind adding SDR Color Volume isn’t about emphasizing the presence of wide color in SDR content itself, but rather addressing a gap in how we assess color volume. Our current HDR Color Volume test uses the ICtCp (ITP) color space, which is excellent for HDR but not applicable to SDR gamma transfer functions. This means our HDR Color Volume results only reflect performance in HDR picture modes. Gamut Rings, however, are specifically suited for SDR gamma, making them ideal for evaluating SDR modes regardless of the color gamut of the content.
We’re also closely following industry efforts to expand Gamut Ring methodologies for HDR applications. There’s promising work underway, but no finalized standard yet. Until then, combining SDR and HDR Color Volume results provides the most complete picture of a TV’s overall color output capability, across all picture modes and content types.
I hope this gives you clearer insight into why we designed the test the way we did, and we’re grateful for your thoughtful engagement with it.
I like the Gamut Ring measurement idea and commend the update. Despite the explanation, I don’t quite understand why you are measuring DCI-P3 and Rec2020 for SDR color volume. It seems like BT601 and BT709 color space would make more sense for SDR, and that most readers might confuse the Gamut Ring as representing the HDR color space.
Admittedly, it’s possible one could be watching 4K SDR content with wide color, but that seems less frequent than SD or HD content. I don’t have a sense of how many SDR video games would use wide color – can you provide some context and examples?
It also makes sense to cap the brightness at 100nits (though it does seem unlikely that a user will do the same) as some HDR TVs do not maximize the SDR brightness at the same level as HDR, but in a TV that offers local dimming, is that 100 nits measured on a small window, which represents a peak brightness, or on a full field white?
Hi chairman_mao,
Great question!
Aside from differences between the ITP and CIELAB color spaces, much of the discrepancy comes down to how we score HDR vs. SDR Color Volume.
In the case of HDR Color Volume, 100% of the score is based on a TV’s color performance when receiving a 10 000 cd/m² BT.2020 signal. As you may expect, covering 100% of BT.2020 @ 10 000 cd/m² isn’t something we’re likely to see anytime soon in any consumer display. Most TVs have a rather limited coverage of 10 000 cd/m² BT.2020; in fact, the best we’ve tested so far is the Samsung S95D QD-OLED and it got 59.4%. The scoring spline used for HDR Color Volume is thus very lenient.
On the other hand, for SDR Color Volume, we test at a reference white point of 100 cd/m², which any TV can achieve. The scoring splines used for CIELAB DCI-P3 and BT.2020 are thus significantly harsher than the one we use for HDR Color Volume, since we should expect most TVs to at least perform acceptably in this test.
So in short, the apparent scoring imbalance often reflects how we tune our expectations: our SDR Color Volume testing assumes mature performance, while our HDR Color Volume testing still accommodates significant room for growth. As display tech keeps improving, we plan to revisit the HDR Color Volume scoring to gradually make it more demanding.
Hope this clears it up! Feel free to reach out if you have more questions or suggestions.
Thanks for the explanation about switching to DCI-P3/REC2020.
I don’t understand though how a display could score so well in HDR color volume while doing only so so in SDR. Can you elaborate?
Hi Amazon_Fan,
Thanks for taking the time to reach out and share your thoughts on our new SDR Color Volume test—we really appreciate your detailed feedback and questions!
You’re absolutely right that BT.601 and BT.709 are the standard color gamuts for SDR content. However, one of the key reasons we opted to use DCI-P3 and BT.2020 for our new SDR Color Volume test is that modern TVs already cover BT.709 so completely—often 100%—that measuring within that gamut offers little to no differentiation between models, even lower-end ones. In contrast, using larger color gamuts like P3 and BT.2020 allows us to highlight differences in color rendering capabilities between displays in SDR picture modes.
It’s true that traditional SDR content rarely exceeds BT.709. But there are still relevant scenarios where a TV might display colors outside of BT.709 in SDR picture modes. For instance, some users choose to set their TV to a wider gamut manually by personal preference, a notable one is when gaming in SDR to get more saturated colors (e.g., setting the TV to P3 color when using a Nintendo Switch which outputs BT.709). In regular video content, it’s possible that some pixels on screen may fall beyond the limits of the BT.709 color space, depending on how it was encoded. This is something we also often see in HDR, where a lot of content mastered in P3 will have pixels falling into BT.2020 territory. We’ve updated the test’s tooltip to make this nuance clearer—thanks for bringing it to our attention!
The brightness cap at 100 cd/m² is intentional since it’s the reference diffuse white point for SDR content. More importantly however is the fact that it’s a luminance output any TV can easily reach, which is crucial to ensure the comparability of our gamut rings results. At different diffuse white luminance points, results lose their comparability as the measured coverage of the color spaces will differ. We measure it with a test window with constant APL to ensure brightness stability across TVs, especially OLED models.
At a broader level, the motivation behind adding SDR Color Volume isn’t about emphasizing the presence of wide color in SDR content itself, but rather addressing a gap in how we assess color volume. Our current HDR Color Volume test uses the ICtCp (ITP) color space, which is excellent for HDR but not applicable to SDR gamma transfer functions. This means our HDR Color Volume results only reflect performance in HDR picture modes. Gamut Rings, however, are specifically suited for SDR gamma, making them ideal for evaluating SDR modes regardless of the color gamut of the content.
We’re also closely following industry efforts to expand Gamut Ring methodologies for HDR applications. There’s promising work underway, but no finalized standard yet. Until then, combining SDR and HDR Color Volume results provides the most complete picture of a TV’s overall color output capability, across all picture modes and content types.
I hope this gives you clearer insight into why we designed the test the way we did, and we’re grateful for your thoughtful engagement with it.
Cheers
Rtings,
I like the Gamut Ring measurement idea and commend the update. Despite the explanation, I don’t quite understand why you are measuring DCI-P3 and Rec2020 for SDR color volume. It seems like BT601 and BT709 color space would make more sense for SDR, and that most readers might confuse the Gamut Ring as representing the HDR color space.
Admittedly, it’s possible one could be watching 4K SDR content with wide color, but that seems less frequent than SD or HD content. I don’t have a sense of how many SDR video games would use wide color – can you provide some context and examples?
It also makes sense to cap the brightness at 100nits (though it does seem unlikely that a user will do the same) as some HDR TVs do not maximize the SDR brightness at the same level as HDR, but in a TV that offers local dimming, is that 100 nits measured on a small window, which represents a peak brightness, or on a full field white?
Thanks,