Raw Sample Sony A7 III UHD SDR vs HDR HLG BT2020 25p 700 nits YV12 8bit скачать в хорошем качестве

Raw Sample Sony A7 III UHD SDR vs HDR HLG BT2020 25p 700 nits YV12 8bit

Comparison between an HLG HDR BT2020 700 nits (cd/m^2) and a Linear BT709 100 nits SDR. The source file is an H.264 100 Mbit/s --ref 2 YV12 (4:2:0 planar) 8bit exactly as it comes out from the Sony A7 III. Despite the fact that the HLG curve is far too big to be encoded in 8bit, the added benefits of an high dynamic range are immediately noticeable as skies are not clipped out, whites are well defined and the picture looks just good in HDR (left) while SDR (right) is showing all its limits. The sensor of the A7 III is capable of recording many stops, but the Linear BT709 SDR limits it to 6 stops which is a shame. To record as many stops as the sensor can record, HLG comes in handy and the Sony can produce files with up to 700 nits. As to the considerations about HLG vs Slog, we gotta remember that HLG is an hybrid curve meant to be backwards compatible with SDR BT2020nc monitors, which means that such a content can be displayed on a TV which only interprets the color matrix (BT2020nc) but not the color curve (HLG). To achieve this, HLG (Hybrid Log Gamma) starts as a Linear BT2020 at the low frequencies (blacks) and slowly embraces its logarithmic part as it goes up (towards the whites). This means that recording in HLG will NOT add any benefit to the blacks (compared to Slog) but only to the whites in which you're gonna have a dynamic range. If you record in Slog, instead, you're gonna be able to have a dynamic range on the black values as well, however, given that the A7 III only records in 8bit and there are only 0-255 values available, it may or may not be worth recording in Slog and then using a matrix of linear transformation to bring it to PQ and you might just wanna record in HLG and have a file that doesn't need any linear transformation.