When did you start being able to see several moves ahead?

Yesterday.

Eh, you think you''re asking about 1 thing, but you're talking about 3 or 4 different things.

1) When could I visualize a 3-4 move (6-8 ply) sequence and clearly see the end position? After a few years. When could I visualize one small area after a 3 to 4 move sequence (like checking if I have enough defenders in a capture-recapture sequence) probably after a few months.

2) When could I analyze any 3-4 move sequence correctly? Depends on the position. Some are practically impossible even for GMs. (Analysis is not only visualizing, but finding the best moves and rendering the correct evaluation of the end position).

3) When could I instantly be aware of a tactic? After I knew the tactic. For example one of the first patterns I learned was back rank mates. As a very new player (less than 1 year) I was spotting those opportunities practically instantly. Another one is the smothered mate which is often a mate in 5. I can see that opportunity instantly without having to calculate the individual moves to make sure it works (although it still may not work see #4).

4) When could I instantly know that a tactical pattern in the position will definitely work out to my favor? Never. You always have to calculate (at least a little) to be sure. This is another one that's basically impossible no matter how good you get.

 Slightly off topic but you might understand why.  Can I please pick the collective IT brains? How does one set up a database using FEN and be able to toggle from one position to the next? Can it be done using Arena for example?

Fair point. When I do puzzles I look for checks, captures, loose pieces, overloaded pieces, and exploitable direct or indirect pins with little to no thought. Then I start to check for familiar tactics or scenarios. If I still don't see it I start to use a sort of process of elimination way of thinking. All of this I do without calculating. Then comes a stalemate puzzle with 4 knights on the board and not much else and I'm dumbfounded. Probably why I've gotten just as many puzzles wrong as I have right.  So I agree that any chess player seeking improvement shouldn't rely on that alone. 

Slow-pawn, combinations which require boxing in a piece (which is what stalemate and even checkmate is) requires a different approach to calculation. Most combinations end on the win or loss of one square (or rather what material is on it). The box type of combinations requires domination of all possible moves by a piece and is often not that simple to calculate because we get very little practice in the game itself as it tends to be rare (except for those involving the king, of course).

Well said, thank you

With me it came after I lost with every conceivable tactic being sprung on me, I remembered them better after I lost to them. You gotta suffer to sing the blues

When you've solved thousands upon thousands of tactics on the board and on tactics trainer .  If you are having difficulty with mate in 4 or 5 try mate in two on chess tempo.  Or get that book with over 5000 chess tactics.  It has a lot of mate in 2's.