When does engine strength/quality actually matter for you?

Here's an attempt to get some opinions on this.

I find it strange that players who have trouble crawling past 1200 USCF get all excited about the latest Rybka version and how they can't wait to receive it in the mail. 

I've been happy using free (downloadable or online like Chess.com) engines in post-mortem review sessions to go over tactics in my played games for the past 400-600 rating points and will continue to do so until I get strong enough to warrant needing a "better" engine. 

So the question really is : At what federation rating (FIDE/USCF) does a specific engine outshine the generic, enough to actually "help" you at the level you are?

Of course, I'd like to "state" that the majority of engine use is for quickly going over missed tactics but for anything non-tactical that *just* doesn't make sense to me,  I go bug stronger players/coaches.  It's not like an engine is actually going to "explain" the position that well, anyhow.

Thanks in advance!

I have wondered the same thing.

I have Fritz 11 and use it for tactics/blunder checking along with the huge database and large openings book.  I can't see myself needing anything better for a long time.  I've wanted something like aquarium for checking database statistics, since I don't think I can do that with just Fritz.  I'd love to see my games broken down by opening type/lines with results, or the same for any other database.  And if I were looking for novelties in opening systems, I bet a stronger program/computer would be helpful, but I'm not a titled player so that strikes me as irrelevant. 

I think it's like the person that has to have a turbo charged car that does zero to sixty in two seconds, but they do  99% of their driving in heavy city traffic, and when at a traffic light that  turns green, it takes them 5 seconds to step on the gas....

 Very much so.

I was about to draw an almost identical analogy but yours is better.

Well, about the chess.com analysis you're talking about, I think it's very poor and even when I was like a 1400 there were too many holes in the analysis for me to be happy with. Something like Fritz 10 that I have right now is fine, and probably will be forever. It doesn't always evaluate a position correctly, but it usually does, and it seems to see every little tactical idea possible in a game. But chess.com engines are simply too weak to be reliable when using to analyze many of your games.

Unfortunately I have no coaches to go to, but I simply try to use my prior understanding of chess to figure out why the engine thinks a certain position is better (keep in mind there are some situations computers may misjudge, that's why they aren't 100% relied on for generating opening theory . Usually with enough work and playing out some variations, you can figure it out, and it makes you a much better player doing things that way, in addition to seeing the flaws in your thinking, and the weaknesses in your game.

So, I would prefer strong engines, but they certainly don't have to be the strongest.

I disagree just a bit (well, for my level anyways). Although it won't make a HUGE difference, I don't want a chess engine to be "only" say 2500 strength. I would prefer it to see as much as it can, and indeed I think this would help me more because I could see more variations that I missed and be able to trust it completely. So I probably could use an engine like that worst case scenario, but I think it's worth getting at least a decent one.

There are quite a few free chess engines that are extremly strong, no reason not to get one of those instead of some rubbish free one.

Good answers!  Thanks!

Would it then be fair to say, that if you are not a Federation rated expert (2000+ USCF, for example) or above, the free 2500+ engine you have at your disposal is more than sufficient?

When I say free, let's assume the discerning downloader will be wise enough to  get the best of the free? (Rybka 2.0 or Fritz 5.32 for example?)

For lower-rated players it's a matter of prestige and fashion ("I use the same engine as the top guys"). For pros it's about finding new ideas and flaws in opponents' analysis or coming up with new ideas. The stronger the engine, the easier it is.

If the free 2500+ one is not enough you can always go for the free 3000+ one.

I don't think you have to have the newest and best engine, but I don't want something crappy either.  I used to use the computer a lot to analyze opening lines against some of the other club players - granted neither of us are near the engine's strength, but it's nice to have good analysis.  I think the most important thing is finding an engine whose analysis you trust:

I love playing the white side of the Najdorf, which is really sharp but you often have to sacrifice material for long-term positional compensation (if Black can navigate through the tactics) and it's at that point that I want an engine who I trust.  Rybka is pretty good with it's positional evaluations - I find that Fritz 9 (which is the last version of Fritz I used) frequently gave much larger advantages to the side with more material than Rybka did - up to 2 or so pawns advantage in its evaluation, but what it boils down to is in the long run the positions were still playable.  

Bottom line is that at the club level, any engine will help you practice tactics.  To use an engine to get a better feel for positions, you need to find an engine that you're comfortable using.  It takes time to get a feeling for an engine and it's strengths and weaknesses in evaluation, but if you're willing to put in the effort to learn with the engine than I think any reasonably strong engine will do just fine.  Only the GMs need the best engines, but even that isn't a complete necessity (look at Topalov - his early copy of Rybka 4 didn't help him win the match!).  

@NM ajgreen: great reply ... really liked the part about getting a feel for a certain engine's evaluation scores.

Engine strength does matter, not because of the tactics (which are good enough in weaker engines too) but because of the positional evaluation.

Rybka may think that a position is quite nice, and be able to show you why (as you go through the variations) where another engine that is weaker would think its equal or maybe even slightly worse.

I understand that if I 'misunderstood' things like a weak engine I would still be a rather strong human player, but I just find it more appealing to have a really good coach that is trustworthy in difficult closed positions and in tactical scenario's.

After all, if you can't afford a real coach, the engine must show you why positions are good. I would think that, taking away the tactical prowess they all have, the positional 'understanding' of engines varies quite widely and this is maybe more important to humans (as you can really learn from it) than their tactical strenght. (which you can't really take with you to an OTB game)

I therefore think owning the strongest engine isn't entirely like owning a car so fast that you'll never really use it. It is more like having a really strong coach. You may never become as strong, but your learning path may well be more efficient.

Just out of curiousity: how good are the chess com engines now, Abril 2024?

Stockfish 11 Lite (HCE only) to be specific, and using it for analysis of chess positions (not for game review in other words).