We were doing Işın Karaca at the open-air venue in Harbiye; the monitor desk was a PM5D-RH. At soundcheck, since the orchestra was used to me, I'd build the monitor mix first and then move to FOH. Right then, the percussionist, Sonat, said there was too much hiss in his in-ears. Back then — and still today — everyone claims Yamaha desks have a lot of hiss. It's not entirely wrong; they just don't know the reason. I listened to the in-ear mix; Sonat was right. He played a four-piece set of tumbas (Dora, Conga, Quinto, Requinto) and liked a bright sound for attack. Pile high frequencies on top of an already hissy signal and the hiss becomes unbearable — but that couldn't be the real cause. Listening to the channel in PFL, I opened the gain slowly. On the PM5D, gain starts at +6 dB; as I raised it, I noticed something surprising: between −10 dB and −13 dB there was incredible hiss, but the moment I passed −14 dB the hiss vanished completely. I checked every channel, moved all the gains in that range up to −14 dB — and the problem was gone. Sonat started to smile. Later I spoke with Yamaha and learned the reason. It turns out that from +6 dB to −13 dB it works as a pad — attenuation. Listen closely and you can even hear the relay switch as you cross from −13 to −14 (on the PM5D and M7CL). Yamaha used two different preamp technologies across those two ranges. Here's the logic: an op-amp can't turn down a hot incoming signal; but you can pad the signal first with a front-end circuit, then raise it with the op-amp. The best example is EQ bands: even though a band reads ±15 dB, all the frequencies are first padded by −30 dB, then raised or cut per band with separate op-amps. That's why experienced, technically-minded engineers switch their EQ out when they're not using it — otherwise your sound is split apart, cut by 30 dB, eating terrible phase, and pushed back up by a 15 dB amplifier just to sit at zero.
The practical rule: on every digital desk built on the Yamaha PM1 platform (LS9, M7CL, PM5D-RH), don't work between −10 dB and −13 dB. (The PM5D doesn't have this problem, because it has analog gain.) Is it a flaw? Absolutely. But that very gap is what led me to the cornerstone of gain management. Note: on the CL and QL the balance is different — gain no longer starts in the positive; it starts at −6 dB and climbs into plus values, and the "−14 dB" of the M7CL corresponds to +18 dB there.
Do you think it's the engineer's job to know a manufacturer's quirk like this — or not?