We are talking about the AL-811 and how ALC is good or bad. By comparing methods there is an aquired knowledge that can be used to make things work better.
There are amplifier manufacturers that do not include ALC such as Alpha.
An attenuators values are going to depend on the impedance they are working from and into. Usually the first resistor in the circuit takes the most power dissipation. There are programs available on the net to calculate these values for you. Here's one; http://www.microwaves101.com/encyclo...attenuator.cfm.
An attenuator will work on all modes including CW, FM and RTTY, though.
Some amps have one built-in. The older Alphas all did. Many solid state amps using FETs do also, as they only need a small amount of power to achieve full output.
I've homebrewed lots of 50 Ohm power attenuators over the years, but nowadays surplus commercial stuff from Bird et al. are so inexpensive at the Swap Meet, it would literally be cheaper to buy one! Last Bird unit I bought was 100W CCS, 50 Ohms, 10 dB, good to 1 GHz and cost $20. It has heatsink fins; I suspect blowing air across the fins would allow 200W operation. These seem to all have N connectors, so some may need to use adapters with them.
What if soy milk is just regular milk introducing itself in Spanish?
Now, back to 7QP, NEQP, and INQP.
How much attenuation do you think you need? Attenuators that would be suitable for this application are not continuously variable. You would need to design it for just enough attenuation to keep from overdriving the amplifier, with the exciter running at just below the 1 dB compression point. Not enough attenuation, and you can still overdrive the amplifier. Too much, and you can't drive it to its maximum output without undue distortion from the exciter.
That's why I say it's far easier to put an audio limiter between the microphone and transceiver. They're everywhere, and not all that expensive.