This is pretty old now and almost not worth posting in again, but I wanted to give a final update in case anyone was still interested or curious.
As I've added to the top of the first post, I've fallen victim to spec creep. I'm blaming it on the Rigol DS2072 (2-channel, 70MHz). This scope was so clearly superior to the DS1102E (2-channel, 100MHz) in every aspect except bandwidth that it seemed well worth the ~840 USD cost (vs the 1102E at ~400 USD). I was sorely tempted to buy the DS2072, despite being so close to finally deciding on the DS1102E.
It made me think quite a bit more about bandwidth, though. 100MHz versus 50MHz for ~$70 was an easy choice (DS1052E at ~330 USD). $300 for the DS2000 series 70MHz > 100MHz upgrade was much harder to swallow. I wondered if 70MHz would really be enough. In the end, I found even 100MHz was a little "risky" as I researched more on the importance of bandwidth.
Looking at rise times, frequency components/Fourier transforms, "rules of thumb" (BW 5x the highest clock frequency), and noting that any signal or component even approaching the 3dB BW would be significantly attenuated it seemed that even a 100MHz BW could really be limiting. That was especially true after increasing my digital area of interest to the 8MHz to 80MHz range from my original 20MHz cap.
The Rigol DS2202 (2-channel, 200MHz) was just too darn expensive though at ~1600 USD. I noticed another scope in this price range and my spec creep was complete! The Rigol DS1204B is a 4-channel, 200MHz scope selling for ~1400 USD. Though the DS2000 series is quite amazing, I was going to have a hard time paying more for a 2-channel scope than a 4-channel scope of the same BW.
All of the sudden I was seriously considering 4-channel, 200MHz scopes at 3x the cost of the original 2-channel, 100MHz DS1102E front-runner. Sure the DS1204B has a much worse display, waveform update rate, memory, and several other features, but it had the key features that mattered most: channels and BW (with sampling rate to match).
I wrote Rigol concerning the crazy $800 cost for the 70MHz > 200MHz, DS2072 > DS2202 upgrade (at least, compared to the $400 cost in the DS1000 and DS4000 series that flank the DS2000 series in the Rigol line up). As expected they didn't have much to say about the pricing. One representative echoed the importance of bandwidth, saying generally you can make due with less channels and features (with some frustration) but not less bandwidth. They thought I'd probably be happiest purchasing the DS1204B, with the confidence of an easy return process within 30 days if the small screen, memory, waveform update rate, etc, etc were really too bothersome. They were really quite helpful.
I then discovered used Agilent equipment. Not old analog scopes, their new DSOX2000 and DSOX3000 series scopes! With warranty and calibration, even. I'm currently within maybe a week or so of buying an MSOX2024A (4-channel, 200MHz, 8 digital) for ~$2000.
The Agilent X2000 series is inferior to the Rigol DS2000 series in several ways, perhaps most notably in lack of memory depth and segmented memory. However, for a $400 premium over the Rigol DS2202 I'd be getting a 4-channel, mixed-signal scope. On top of the key features of two more channels and the digital inputs, I'd be getting a true made-by-Agilent scope, and the brand does carry some weight with me. I think the $2k could be a good buy and worth it to me.
I've written Rigol again to see if they offer certified used scopes with warranty and calibration but haven't heard back yet. If they do, and offer their used equipment at a similar discount, perhaps I can spring for a Rigol DS4024 4-channel, 200MHz scope for less than the Agilent. It wouldn't have the digital inputs, but that isn't so important to me and the other features like memory, update rate, etc would make up for it.
So there it is. I still have a little voice in the back of my head saying "Buy the DS1102E. Save your money and buy a 4-channel Rigol DS2000 series if/when it comes out. Or save it to buy whatever comes out 5 years from now. Or save your money and put it towards something responsible like the mortgage. Or save your money and do anything else with it - just don't blow $1600 more than you need to."
I don't know. I think the MSOX2024A is worth 5x the DS1102E. Twice the bandwidth, 3x the channels (counting 8 digital as worth 2 analog), 100x the update rate, 5x the pixels, the Agilent name, etc. But does that mean I should buy it?
Anyhow, there's my last update for this thread. People are still welcome to post here if they find something relevant to them, especially about the less expensive scopes mentioned in older posts, but please use the new thread I'll make shortly to make any comments about the new scopes I'm looking at.
Thank you all again for all your input. It certainly hasn't gone to waste, and as I wrote in the first post I'm positive I'll be much happier with the purchase I make as a result of the discussion and your ideas.
Even writing things out and having to come up with words to explain what I'm thinking has been a helpful exercise. Thanks for humoring me as I've worked my way through it.
Jacob