I'm working on my own "value" formula for me that basically goes Price divided by Frequency divided by MS/s and that will get me the value indicator as far as raw functionality. (The lower, the better)
Then I will try to weigh that against screen size and res, and the other value added items (LAN, RS-232, VGA etc...
If you're wanting to know what I'll be using it for, strictly hobby level stuff. Microcontrollers, 555 timers, adding and subtracting capacitors, inductors, and maybe playing with sound. So I guess if that's it, really any scope would do, but I'm always up for a good value for the dollar.
i think you need to add more variables to your equation.... to be accurate.
guarantee, menu jump around factor, internal clock stability, stability/accuracy when measuring a 10MHz rubidium standard, FTT performance, dual trigger, probe 10x precision,
just to name a few...
also there are other brands like UNI-T,.
Bigger Screen without accuracy, speed or enough DSP is useless,
for starters,
think what´s the maximum SQR wave frequency you will be measuring, PWM from a Scanner? CNC Machine? Digital Audio? etc...
multiply that by x20 & thats your scope Miniumum requirement.
also more Real GSa/s the better.
memory is a controversial topic, More the better, but some times less is more.
for example:
s/pdif at 192khz goes at >6MHz = 120MHz scope.
FM Radio 110MHz, BUT FM Radio is NOT SQUARE waves, = for Sine a x2 should work, Nyquist minimum 220MHz scope, but more is better.
Those 555 can go really fast.
also there are other factors like Rise Time, if you really need to measure accurate Rise&Fall Times, you need a really fast Scope >500MHz has 0.7ns, 350MHz has 1ns, etc...
going cheap on BW is not a good idea.
if you need to detect digital glitchs, you need a fast wms/s, Agilent dsox3k has 1M, almost all others are under 100k waveforms per second, and cheap scopes are too slow.
...
is not all about value, it´s about what works.