Re: Home practice: model, attenuate, or baby tube amp?
Instead of an attenuator, I'd suggest a dummy load - if you want to use your Mark five quietly. That's a way of separating cooking the power tubes from making a lot of pretty noise. The drawbacks are that you'll probably miss the speaker cab's coloration and there is no easy way around it. You can use impulse responses but they lack the nonlinearities of the real thing and introduce phase aliasing, which is why I think amp modelling sucks in general. However I still use this method, because a grossly underpowered guitar cab sucks too, although in a different way and for a different reason. If you don't mind a digital cabinet simulation, I suggest finding out about the Torpedo loadbox with integrated IR convolver. I think I saw Keith Merrow use one.
Then, there are isolation cabinets, which are a 1x12 cab and mic in a soundproofed box. These will help you practice close-miking as well, but you won't hear the raw speaker tone unless you leave it open and find out it's either too loud for your fellows or too quiet to sound good. Guess what - isocabs suck too, because they take up space, and need some auxiliary equipment before they can be properly used (mic, preamp, monitoring solution, cables). They leak a bit too.
Then there is baby tube amps. My dear, these are so cute. But the problem is everybody seems to be stuck in a "watts per dollar" mentality, as if guitar amps were clothes irons. Yes, you could build a low watt mark five but it would wind up costing almost as much as the big brother. Iron and tubes are cheaper but designing and building a small amp takes the same amount of work as a big one. So most commonly, baby amps are made simple and cheap because fitting a zillion knobs into a small package is difficult and because few people would actually buy a $1500 practice amp.
