Have a look at some of Morphy’s games @ chessgames.com
He was playing enjoyable tactics over 150 years ago. 
His games are fairly short and not too difficult to understand.
The early levels mentioned above are just roughly based on your rating.
From FIDE Master onwards the main way to the title is to achieve a set % result in an International Tournament.
(The actual score required will depend on the strength of your opposition.)
You can read more here:
That is an amazing accomplishment. I know this because the reason I gave up serious chess study is that I wanted to live a full high school teenaged life, and all the hours of study and play would have been too much of a sacrifice even for a nuclear powered 17-year old. How you were able to handle a mentally exhausting job and still reach that level boggles my mind.
This is exactly why we need to nurture and fund talented players. If you were able to reach 2390 under those circumstances, what might you have achieved if you were able to devote all your time and energy to it? I can only imagine how many others there are out there in the same boat.
You’re very kind! 
Probably the main reason that I was able to work full-time and still do well at chess was that I have Asperger’s Syndrome.
Because of this autism I found chess immensely satisfying:
- there were strict rules
- there was almost no social contact (just handshakes and post-game analysis)
- similar patterns kept coming up
I considered turning professional, but chess is not like golf or tennis. 
A golf or tennis player rated 250 - 500 is very well-off. There’s far more sponsorship as well.
Not so in chess.
I’m happy I made the right decision - by having a career outside chess I have a house, a pension and savings.
Anyway I never had the combination of chess talent and determination to reach the very top at chess. (Of course I admire those who have!)
I still don’t really get it but thanks. 
Not when you’re playing a computer engine that takes 5-10 seconds per move, which as I stated is how people typically play against a computer engine. Because who wants to sit there forever waiting for the computer to think? At such a short think-time, the engine doesn’t reach a deep-enough depth to approach what humans might label strategy. (Although, to be honest, comparing human chess-playing heuristics to computer heuristics is a little misleading to begin with, since the heuristics are rather different).
By “calculation” in this context, I mean chess calculation, i.e. “If I move here, my opponent will likely move here, and then I will move here… on the other hand, if my opponent makes this other move, then I can respond by …, etc…” Apologies for the lack of clarity.
I agree that computers are not very good to learn against. If you have no other option, it’s better than nothing, but otherwise you should definitely be playing against humans.
Fortunately this is just as easy as playing against the computer in our modern times.
RickJay, at your current level games are largely decided by all the material (pieces and pawns) that are left hanging, i.e. free for the taking. In an 850 rating vs. 850 rating game, there will be stuff hanging all over the place throughout the game. This is where computers and humans are very different. A human misses these things for very complicated, human reasons: tunnel vision to a different attack; overlooking that a piece is pinned; missing that a bishop is hiding in the corner all the way over there; overlooking that a move looks great but unprotects something else; seeing only the obvious opponent response and not something they can just do instead that is stronger; and on and on. These are dominant mistakes in your game, and the mistakes you can try to punish in your opponent’s game. At your current level they are the deciding factor.
Computers, though, don’t approach the game like a human. They will punish your mistakes inhumanly brutally – missing nothing and also freely using tactics that no 850-rated player would ever find – and then make compensating weird computer-like mistakes that don’t even look like they have a purpose. The computer might even just fail to make a basic recapture when you take something – something a human would never overlook (“You took my piece, so I will take yours back and we’re even.”) – just because the engine decides it’s time to play a “bad move”. Their bad moves have no soul and make no sense and feel nothing like playing a human, while their very good moves are too good.
So:
(1) Play humans in the longest time-controls you can find and tolerate. On chess.com, 15|10 (15 minutes plus 10 second increment) is the typical “long” game.
(2) Analyze your games after using the computer. When you get past regular 1- and 2-move tactical blunders, keeping the engine off when you start your analysis is better. But for now, just switch on the computer evaluation and walk through the game. When the evaluation changes drastically, stop and understand why? Did someone just leave a piece hanging? Did you notice it in the game? If not, why? If you don’t understand why the computer says a move is bad/good, look at the lines given by the computer to understand the ideal sequence and what happens if you deviate from it in a human-like way.
(3) Study tactics both by learning the basic tactical themes (they have names because they come up again and again) and by working through tons of tactics problems such as those available on chess.com. Take your time on them – there’s no (explicit) clock! When you get one wrong, open it in the analysis board to understand why.
Do those three things, especially (2) and lots of (3), and you will improve.
If you want to post here or DM me your chess.com username, I would be happy to look at a few of your recent games and give you some feedback here.
Any chess player will play better if you let it think longer. But speed is a much less significant limitation for a computer than for a human. If you want to make a computer play its moves in half the time, all you need to do is make the computer twice as fast. That’s easy. If you have a slow chess program, just wait a few years, and with no extra effort on your part, you now have a fast chess program.
A question. I just played a game on chess.com (I lost.) It occurred to me at the end of the game that my opponent and I had probably played a game of chess that was unique in all of human history.
At what point - how many moves in a game - is it likely that the game you have played has never occurred before? It’s obviously not after one, two or even three moves.
I don’t see how to give you an answer.
I hope you won’t mind if I say it depends on how well you and your opponent played! 
I’m confident that over a billion games have been played over the Internet (if that helps.)
There is a huge amount of opening analysis and I know of Grandmaster games that have started exactly the same way for 25 moves (by each side i.e. 25 pairs of moves.)
I’ve had the same 15 moves (by each side) once. (I won both games.) 
Here’s an opening I played in a Dutch International Tournament that I doubt has ever been repeated:
The amusing thing is that after 10 moves White has only moves his King and pawns. 
“Likely” is the tricky word here.
There are extensive databases of games accessible for free online. On chess.com you can find 2.8M masters’ games in the opening explorer accessible through the analysis board. Actually, it might be that you need a premium membership to use the opening explorer extensively, in which case you can get a fully free database at lichess.org instead. In the lichess.org opening explorer, you can also switch it to use not the masters database (which is based on real-life over-the-board games played over the past 100 years or so) but to instead use all the games played on lichess.org itself by players of whatever ratings you choose. It you pick the whole database (which does have a lower rating cutoff) including all time controls, you get about a quarter billion games.
Using that large database, you can make some silly moves and find unique games after the third half-move (“ply”), but since this full database includes people clearly goofing around with each other, you need some really silly moves. For instance, the completely bonkers sequence 1. f3 e6 2. Kf2 Ke7 3. e3 Kd6 4. d4 has happened three times in the past three years with completely different players.
I found a unique third ply after 1. Na3 a5 2. Rb1. It would be interesting to see if there are any two-ply games (of which there are only 400 different possibilities) that aren’t in that database. I would guess not.
So what about games where the players are trying to make sensible moves? At very beginner levels, games are probably unique after five moves or so (full moves now, not plies). At beginner-to-intermediate level, maybe six to eight moves. Comparing your own openings to the masters database is a good way to see what early moves you make that masters never play. You can then ask “why don’t they?”
At a decent club level, “likely unique” probably happens around the eighth to twelfth move, but this is just a guess at the average. Any one game can obvious be unique much earlier or much later. If both opponents are playing a line that they have good “prep” for, even at the club level it’s not uncommon to get to the teens of moves and still be “in book” (i.e., still be playing moves that masters have considered worth playing). Looking at my last online game, I see that my opponent’s (white’s) 15th move was the first move not in the masters database nor the larger lichess database. If I play the same opponent multiple times in a row, we’ll regularly get deep into the same lines as we each try to find subtle improvements.
At even higher levels, as mentioned by glee, 20+ moves of “theory” is not uncommon at all. Just clicking blindly through the most common reply to the most common reply to the most common reply … until the database stops letting me click 25 moves in, there are still 105 games at that point after those 25 moves, and that’s passing up soooo many other branches along the way.
Using the chess.com database instead and clicking less blindly to select lines I think might generate longer sequences, on my first try I got to two matching 32-move games from 1977 and 1980. Those games ended at that move, so that line couldn’t have kept going. There are surely even longer repeated games, though.
A related question that I’ve sometimes pondered: If you somehow magically had a database of every valid chess game ever played, right down to the five-year-old who has just learned how every piece moves but is otherwise playing randomly, all the way up to those practice games Alpha played against itself… And if you then selected games from that database uniformly at random, what average level would those games be at? On the one hand, there are a lot more novices than masters, but on the other hand, each master plays many times more games than each novice.
Honestly, I think the best advice I can give you is have fun with the game.
Yes, there are super good players and AIs that can destroy them too. Rating systems are all over the place–someone’s 800 is someone else’s 1200–the population can vary wildly. Playing someone wildly different from you in skill can be educational, but it’s not much fun on its own.
I find chess to be fun because of the tactics. I generally don’t play ‘gotcha’ sorts of tricks, but being able to win a game because you’ve outwitted your opponent is a blast; meanwhile I don’t feel a lot of guilt or shame for losing a match either.
I’m not a huge fan of memorizing openings or doing a lot of deep analysis, and I’m not a professional player, but getting a decent game is mostly practice. Play chess puzzles, they really help your tactical skill–I think that’s my major edge. At my level of play (1400ish on Chess) I generally am outmatched on well practiced opening players–until they fall out of their opening book.
Block poor sports and people you don’t enjoy playing against, and play more chess to improve.