I have both of them now and like them for different reasons.
The main reason I picked up a 590 was that it had lower current draw on rx, about 1.25a measured. My late model 2000 pulls about 2.1a, and 600ma of that is due to the backlight! Unfortunately, turning the backlight off does not really result in a usable display from external light sources, so I always had to run it with at least the lowest setting. With backlight off, it pulls about 1.5a, but any form of backlighting pushes that to 2a quickly. I run solely from batteries as a hobby in itself, so current draw is extremely important to me with these bigger rigs.
I suppose my 590 is a pretty expensive led-mod to the 2000.
This may not be an issue for most people running from a power supply obviously.
Despite that, all things considered, I still like my 2000 very much. I mostly did HF and airband monitoring with it on vhf, and the ability to mix audio with HF without having to run two speakers if you want to go that route is a plus. Although the slowness of the scan is definitely not scanner status.
And for me, nothing beats Kenwood audio. Either radio shines in that department. The problem with many comparison videos, is that you have no way of knowing if the op has the rx-eq set to any of the following:
High Boost 1
High Boost 2
Bass Boost 1
Bass Boost 2
I normally use either OFF or Flat, but unfortunately you have no way of knowing in most videos if these have been set properly, or long forgotten about after getting tweaked early on.
I think that the 2000 gets a bad rap from those that try to put it into a contest-radio category, or attach it to a 6 over 6 array and wonder why it jumps off the desk.
Turning off the preamp when you don't really need it does wonders. Then again, I was using a very late model.
So far, I'm very happy with my limited use of the 590. But quite frankly, had the 2000 used led's instead of incandescents and didn't draw so much current, I don't think I would have sprung for the 590 with the way I operate.