So you've seen the other threads. In some cases experience now means career success sooner, so why not just jump on a commodity scope (DS1054z?) and re-sell if it doesn't fit? The delay may cost even more.
What does that even mean? Reselling something takes time and usually a loss especially when you turn something new into something used.
It means that education is never free! Yes, you will sell for less than what you paid but the difference is 'education'. You learned a lot of stuff while using it (which has a value) and whatever residual value exists can then be applied to the next scope. The 'education' helps you select the next tool. You will know which features you care about.
My budget is about $400USD (since most things arnt in CAN on forums). The RF circuits are all under 1GHz for now.
Then you aren't within a decade or two of buying a 1 GHz scope. Look around! Keysight, Tektronix, etc. Not that it applies but here is a 1 GHz Keysight for about $16,000.
http://www.keysight.com/en/pdx-x202189-pn-MSOX3104T/mixed-signal-oscilloscope-1-ghz-4-analog-plus-16-digital-channels?nid=-32541.1150366&cc=US&lc=engFor $400, you are stuck with entry level scopes and one of the more often recommended is the Rigol DS1054Z. After the first firmware revision (hopefully), the Siglent SDS1202X-E will be a nice 2 channel 200 MHz scope. At the moment, the Rigol OWNS the low end market. There's a reason for that!
A signal generator would be good for testing stages in an RF circuit but can you use the same scope that's making the signal to measure the signal?
The scope internal signal generators are not very sophisticated. An RF signal generator isn't much of an expense ($179)
http://www.newark.com/tenma/72-585/rf-signal-generator-frequency/dp/66F3578It also isn't as sophisticated as some of the high frequency arbitrary waveform generators
https://www.rigolna.com/products/rf-signal-generators/ I'm not really advanced enough to look for packets in a digital signal so I guess that doesn't matter. I also want to learn more with transistor circuits and I'm always curious to see what the signal is doing between components. I learn best by trial and error or building things then changing out random pieces or values. For instance in my motorbike turn signal circuit I totally removed what I thought was an absolutely necessary transistor's (collector?) resistor in one stage and the circuit still works. Everything I know says that it either shouldn't work or at least stop at that stage. I think a scope would tell me why this is. But don't et too hung up on the fact I'm experimenting with RF stuff. I might get hung up on something as I learn more and decide that I would rather build other circuits. That's what makes this difficult, people ask very specific question as to what you are doing at the moment instead of thinking about the range of things you would do with it. Just because I'm building a circuit now that might need a very specific function of a scope doesn't mean I want to invest and extra 100$ just to build that one circuit and have a piece of equipment I'll never use again.
There is no perfect scope when budget is finite. Even if you work on RF, you are seldom interested in the carrier, you want to see the modulation. Or you want to see the spectrum and a scope is the wrong tool.
https://www.rigolna.com/products/spectrum-analyzers/I don't work on RF (although my logic circuits run at 50-100 MHz) but I do have one of the less expensive RF signal generators (above) to help align an AM-FM radio kit I bought for my grandson. I also have a nice Siglent arbitrary waveform generator but it only goes to 80 MHz and that is more than I will ever need.
http://www.siglent.com/ENs/prodcut-detailxx.aspx?id=1134&tid=16&T=2Two different tools, two different areas of application.
For your budget, you are pretty much constrained to the Rigol DS1054Z or that Siglent SDS1202X-E. I might hold off until the first firmware update on the Siglent but the Rigol is pretty well understood at this point. It isn't perfect but it is inexpensive.
There are $300k scopes and I'll bet they work pretty well. I'll never see one or even drive by a building that has one!