I've read this post a few times as I appreciate the data provided. One question and one thought;\n\nI can't tell if Amir says ECO mode degrades DAC performance only in the channels using onboard amps or also those of LR in pre-amp mode? It appears using an X3700 in pre-amp mode with a Monolith 5X (even though not tested on ASR, bench results on here were very good) and two Emotiva A-150s (I already own one) for the Atmos channels would be an excellent and relatively cost-effective solution.\n\nThis would save speculation on if the extra processing in the X4700 & X6700 would actually cause higher levels of noise and therefore not test as well.\n\n\nI don't remember ASR measured the DAC performance with ECO on. From what I gather, it would degrade in the sense that ECO reduces the rail voltage so the onboard amps will reach their clipping point sooner, resulting in the pre-out (I don't prefer the DAC term that Amir used all the time) distortions to increase sooner, that is, at a lower output voltage.\n\nI don't prefer the term "DAC" but it is probably just me, because very few people actually use the direct\/pure direct modes using analog inputs, when the ADC\/DAC path is not used.\n\nThe X4700\/6700H should measure better because their rail voltages are a little higher and the power supply a little larger, so even if the THD+N is the same as the X3700H, it will at least be maintained up to a higher voltage.\n\nI suspect the X4700 and X6700H can maintain the same SINAD of about 100 up to 1.5 V, but that's just my educated guess. There should not be extra processing in those units but even if there is, it wouldn't result in more noise in those kind of measurements.\n\nEdit: "not" was missing for some reason.