chroma_offset for 4:4:4 looks suspicious

[kornelski]

Q_MODEL_MUL and Q_MODEL_ADD is used only for 8-bit depth. This seems odd, because perceptual effects of chroma's spatial resolution are not dependent on the bit depth.

The adjustment is not used for intra frames either. I'm not sure why. It would be nice to have appropriate chroma quality for still AVIF images.

rav1e/src/rate.rs

Lines 534 to 538 in c247d53

	if !is_intra && bit_depth == 8 {
	log_q_y = log_target_q
	+ (log_target_q >> 32) * Q_MODEL_MUL[chroma_sampling as usize]
	+ Q_MODEL_ADD[chroma_sampling as usize];
	}

I suspect that the slopes and bases for 4:4:4 may be overfitting something, or lack clamping before the calculation:

rav1e/src/rate.rs

Lines 510 to 523 in c247d53

	fn chroma_offset(
	log_target_q: i64, chroma_sampling: ChromaSampling,
	) -> (i64, i64) {
	let x = log_target_q.max(0);
	// Gradient optimized for CIEDE2000+PSNR on subset3
	let y = match chroma_sampling {
	ChromaSampling::Cs400 => 0,
	ChromaSampling::Cs420 => (x >> 2) + (x >> 6), // 0.266
	ChromaSampling::Cs422 => (x >> 3) + (x >> 4) - (x >> 7), // 0.180
	ChromaSampling::Cs444 => (x >> 4) + (x >> 5) + (x >> 8), // 0.098
	};
	// blog64(7) - blog64(4); blog64(5) - blog64(4)
	(0x19D_5D9F_D501_0B37 - y, 0xA4_D3C2_5E68_DC58 - y)
	}

That's because in dc_qi and ac_qi, u can end up having higher qi than y:

 QuantizerParameters {
    log_base_q: 596939234966375734,
    log_target_q: 540310075640714472,

x	10 bit 444	8 bit 444	10 bit 420	8 bit 420
dc_qi	[110, 131, 108]	[110, 132, 108]	[110, 101, 75]	[110, 101, 73]
ac_qi	[99, 118, 97]	[98, 118, 97]	[99, 90, 65]	[98, 88, 61]
lambda	20.83157578250051	21.039473974368523	20.83157578250051	21.039473974368523
dist_scale	[1.0, 0.54229736328125, 1.0629119873046875]	[1.0, 0.5428314208984375, 1.0639495849609375]	[1.0, 1.2977294921875, 2.5435638427734375]	[1.0, 1.3011627197265625, 2.5502777099609375]

[kornelski]

@barrbrain Is this behavior correct? Is there a reason for ac_qi for luma to be so low compared to chroma for 444?