Hahaha! Funny how two people can read the same sentence and have completely different understandings...
I guess that's one of the limitations of this form of communication. Also, English isn't my first language, so I'll probably make a fair bit of mistakes.
Apart from one offer I saw a while ago, I am still to find another really competitive offer from Keysight's eBay store for hobbyist/enthusiast level oscilloscopes. But that may be just me...
I doubt that, and it can sometimes be difficult to find the good one amongst the masses of overpriced crap on ebay. As to the Keysight store, not everything there is priced attractively (especially older gear can be overly expensive), but they seem to not list everything they have, so it might be worth just asking them if they can offer you the item you want.
No contest there. Despite I still think Rigol's struggle with firmware is understandable from a newcomer perspective, I don't think it should have been released with so many issues from the get go.
I wouldn't call them a newcomer, in fact, Rigol makes scopes for much longer than say Rohde & Schwarz (the first Rigol scope came out around 2002 if I remember right). They have also produced entry-level scopes for Agilent. You'd think that being in the business for over a decade that by now they have figured out that software is as important as hardware. And they really got the hardware bit right (the hardware quality is pretty good), they (unlike Siglent) even managed to come up with some decent UIs for their kit.
A while ago I talked with some colleagues who worked in China, and they said that the reason why software in Chinese products regularly sucks is pretty much cultural, as software isn't considered valuable over there (one reason why piracy is so high there). Which isn't that surprising, considering that the same thinking as (and still is) prevaillent in some Western companies, but by now it should be pretty clear that to get somewhere in T&M the whole package (software and hardware) must be right. Good support is a big thing as well. That Rigol still doesn't understand that is worrying, even more so when they now have a US subsidiary which you'd think would help get the software quality issues sorted out, but it doesn't look like that's the case.
It is hard to dissociate brand from overall quality perception, but I would not probably be swayed either by price alone if I had my rear end on the line (nobody is fired for buying IBM). However, one point that A-brands should be concerned is that, as people use Rigol's gear as hobbyists, they start to get comfortable with its quirks and interface. If Rigol survives long enough in this marketplace, there is a strong chance they will eat the market from the ground up.
I doubt that. Any hobbyist who decides to pursue the career of an EE will very likely get into contact with big brand kit on college, and even if not there then very likely when starting to work as EE. And the enthusiasm for Rigol will hardly last long when confronted with the superior capabilities of a A-brand mid-range scope, plus the fact that the big brand scope is unlikely to suffer from similar firmware issues as the Rigol gear. Most EE graduates we get know Rigol (usually because they own some Rigol kit) but the majority seems to be well aware that there is quite a difference between Rigol and big brand, especially outside the entry-level class. And as big brands are essentially giving gear away for education this is unlikely to change anytime soon.
We also have had a few people who really didn't know much more than their Rigol scope (or their old analog banger). We usually sit them on a bench with a modern DSO and some low complexity circuit and ask them to perform some simple tasks (i.e. measure signal jitter). If they then start to fiddle with cursors then they get a quick introduction into the jitter analysis package, at which point you can hear the penny drop
Seriously, unless Rigol comes up with something groundbreaking they will remain stuck to the entry level/hobbyist/startup market, and even if they do they'd still have to dramatically improve support to even make a dent.
In hindsight, I think my comments were swayed by the price I got for my DS4104 (US$1.6k with new 500MHz passive probes), which was pretty difficult to match with eBay's offers of newer generation oscilloscopes from the A-brands (yes, I did a large search as I mentioned here). Closer to US$3k the search can be widely expanded.
Well, $1600 isn't a bad price for a hackable 4ch scope, and in the end it only comes down on what you want the scope to do and if you can live with its bugs and Rigols support.
Back to the topic, for what the OP wants to do and the amount he could spend I'm sure there would have been a few big brand options available for the money he was willing to spend and considering that he's obviously fine with only 3months warranty. Especially if the requirement is ETS (although for that price you could also find a scope with proper jitter analysis software).