I have the Yamaha RX-V659 receiver. Paid $300 new. 100RMS/ch 7 ch.
The reason I bought this receiver was because of the audioholics review that praised the amp section. I am happy with it, and cannot expect too much for $300.
When you mentioned the 'ghastly Far Eastern receivers', are you talking about the standard Asian style receivers? Denon/Yamaha etc?
What is it exactly that you don't like about them?
We agree on the tube amp philosophy. I had to laugh the other day when I found an article about a guy that was trying to isolate his tube amp from the bass vibrations so as to not disrupt the tubes. Tubes are better of usede as light bulbs.

Notice how buls are slowly being replaced with LEDs...
I am not doubting anything you say here, but have a few questions.
What are the benefits of passive crossovers over active crossovers?
Why can't/shouldn't the RCA level signal have a passive crossover?
I'm assuming the passive system creates issues with capacitance and impedance.
How about a reciever that is only low level with no internal amplification?
Why doesn't Yamaha have such a model? The Eclipse car deck I have is such an animal. No heat issues for the optical drive system etc. I will admit that it is more likely that external amps are used in the car audio scene.
I will be building an ddesigning my own speakers here shortly. I'd like to hear you points and info on why amp/w speaker is so effective. I may go that route. I wanted to assemble my own amps anyway...
Thanks Again,
I'm not as inept as I may have sounded. So building speakers/amp is not a pipe dream. I'm looking into the required audio test equipment etc and then do a box or 2 before hitting the amp side of things, if i do at all. I'll only do it if I can make something better than I can buy without having to sell my children...
Art
You have asked a lot of good questions!
First of all, at line level you are dealing with high impedances and low voltages. The loss of signal strength from passive components would be prohibitive and introduce noise. At line level you need an active filter. In the analog domain you do it with op amps and you can make a filter in the feedback loop of the chip. You cascade to get the order you want. Remember that phase and time shifts will be the same as for passive.
Now active filters before the amp avoid the interactions between the crossover and drivers and amps. There is no insertion loss and no waste of power from L-pads. Passive crossovers waste 15% of the power at a minimum and usually more.
The amps can be tailored for the frequency response required and the power in that band.
Crossover points below 350 Hz are possible. Crossovers below this point with passive components are highly unsatisfactory.
Also the speaker leads are very short. The length of usual speaker leads have a high enough resistance, when coupled with the inline resistance of passive crossover components to negate the damping factor of amps. This also has an impact on the total bass tuning of the system.
So there are very significant advantages from having the amps in the speakers with active crossovers. With the trend towards digital crossovers that have the potential to solve phase and time problems, then the benefits become even greater.
Now as far as receivers are concerned, I think it is a bad plan to place preamps, and above all signal precessing and decoding in the same case as the amps. Processor chips need a cool environment. There is therefore the need to increase amp efficiency and bias too heavily towards class B operation. These cheaper receiver amps tend to be highly current limited as evidenced by their poor ability to handle four ohm loads and double the power available into a four ohm load. The development of better class D amps will help in this regard, but still the power supply will have to be formidable.
Now there is a good reason that the higher end separates tend to shy away from these auto Eq sytstems.
It is a bad idea. Frst of all it is a myth that you can Eq away loudspeaker problems. You can't. Take a null at crossover for instance, if the auto Eq tries to correct it, the power increases to both drivers, but they are still out of phase and canceling. So the magnitude of the over drive can become immense. The auto Eq program has no idea where the resonant frequency of a tweeter is, for instance, and can inadvertently drive a tweeter to its area of resonance increasing distortion and harshness. There are so many things wrong with this I could be up all night listing them all.
The next issue is room Eq. Nothing says a room has to be Equalized to flat, and it will likely stress a speaker to do so. If you had a couple of instrumentalists in your room and they sound pleasing, then there is nothing intrinsically wrong with the room for the reproduction of music. Rooms intrude when a speaker's off axis response does not mirror the axis response, and the speakers have a poor waterfall plot.
With the long term and current state of microphone technology, the microphones have to be placed closer to the sound source than in the hall where a listener would obtain a pleasing balance. I'm speaking here about classical productions, pop productions are a different kettle of fish. Because of this phenomenon the recording will contain a balance favoring direct sound over reflected sound, where as the listener in the hall will hear a much higher proportion of reflected sound.
So random reflections in a listening room actually enhance a good recording if it is a good room. I and others have noted even a room with a definite echo does not preclude it it from being a good listening room. Auto Eq programs can, and often do spoil an otherwise pleasing room. I see no justification for them at all. Some systems and rooms can be in need of touch up, but I believe this is best done by ear with subtlety. Good loudspeakers will sound their best in the vast majority of rooms with NO equalization.
So in essence those are my objections. Really scarce resources are much better put towards improved basic engineering.
All these "peculiar notions" of the high end distract effort and resources away from solving real and important problems. There are for too many big problems out there to be wasting time on imagined problems.