PDA

View Full Version : Speed of Game - Myth or Speculation



O'Leary's Trainee
08-04-2010, 20:20
Ok. This is going to be a weird thread, but the question's been going round and round in my head for a long time now. This has got to do with the speed at which the game processes... and no, it's not anything to do with calculations vs. computer ram/processor/memory or anything like that. No this is a question about what actually affects game speed when setting up a new game.

Let's start by eliminating a few of the common suggestions that I know of, and has nothing to do with the actual discussion I want to have. I am not interested in knowing I can set fixtures on low or high detail - I am not interested to know that leagues can be run as view only... and so forth.

So here is the thing that interests me.

I frequently hear the theory that it's the number of players loaded into the database that is the main culprit in slowing down game speed... and to me that makes sense to a certain degree. The number of players and clubs that needs having stats entered and events calculated certainly would add to the time it takes for the computer to calculate a day.

Last year, I ran a game with every single Premier level team loaded into the game, and with all but the Asian leagues running as Active. This gave me a huge number of players and staff, and produced a slow game which I still was able to enjoy.

The thing is that, in theory, I had pretty much every player in the database loaded into the game, which means (according to the above statement) that should I decide to start a game with Every Single League running this year I would not notice much of a difference from last years.

The reason I skipped Asia last year was because it just doesn't feel right over there without J-League active. This year, I can have J-League thanks to the editor and it opens up so many new possibilities for me... being able to manage in Asia where there are many nations I am yet to try my luck in (and I love exploring new nations to play in).

My further problem is... I've always loved to manage in Africa. Anyone who frequents the Editor's Hideout knows I've been working on a couple of African leagues... and I want to load up a few of those as well for the full African experience.

Now, that means I am probably adding about 10-15 more countries to the list of playables in this years game when I get it started. This would suggest a totally unplayable game in theory...

...and that is what leads to this discussion. Since I retained almost every player in the database in last years game - even with numerous added leagues, I would only add 10-20.000 players/staff to the game (and that's a generous calculation). And adding that number to 250.000+ isn't really all that terrifying.

It is a given that game day calculation would be slowed down even with minimum setting on all leagues - but I can live with that. It is the "dead time" after matches have been calculated and the game (probably) generates news items, events, contracts, transfers and updates all stats that I don't really like to wait for for too long. During match calculation I can watch some scores scroll down... I can alt+tab out of the game and doodle with something else on the computer as well... so I don't mind a slow game, but there is a breaking point somewhere.

Basically - I'd like to know (from someone in the know, preferably) just what to expect when setting up my game. Is it the number of players/staff/clubs that are loaded from the database - or does the number of active leagues really matter?

Say I retain every single club and staff, but only have the English league active.

Compare this to a game with every single league Active.

Apart from calculating fixture results... would there be any significant difference in game speed?

x42bn6
08-04-2010, 22:51
There should be. Having lots of data (i.e. players) in your database isn't a huge deal compared with simulating every single league, even "virtual" or "grey" leagues which are quick but slower than unsimulated.

sporadicsmiles
08-04-2010, 23:12
Well, having lots of leagues active means their games will be processed, news items to be generated, bans to be appealed against and everything else that goes with having a league running. So I guess that loading more leagues would have a greater effect on the overall speed of the game than adding more players.

However, the more players, the longer it will take to dish out random injuries in training, process unhappyness, and everything that goes with being a player that will have to happen regardless of whether a league is active or inactive.

In other words, I would say both contribute to varying degrees. Simulating games and leagues would be my bet for the main proccessing time. An easy test of that actually, does the game run faster pre-season (assuming most of the leagues active run over the same time period), when there are fewer games/league related things to process?

O'Leary's Trainee
10-04-2010, 09:59
...

In other words, I would say both contribute to varying degrees. Simulating games and leagues would be my bet for the main proccessing time. An easy test of that actually, does the game run faster pre-season (assuming most of the leagues active run over the same time period), when there are fewer games/league related things to process?

What I found in last years game was that pre-season was actually the slowest time of year - that and during the shift of year. I contribute this to the vast majority of leagues having active transfer windows at this time, which again suggests that the number of players added cause most of the slowdown to a game.

Let's not discuss the fact that fixtures need to be generated, because it is (of course) a natural thing that this area of the game will take longer the more leagues you have active. That part is not what I'm interested in - but more about the "dead-time" processing... after fixtures have been simulated and the game needs to create the News, Player Events, Team Events and Updating of Staff/Players/Histories.

To me... it kinda makes sense that the number of players in the db would be the main thing that slows down the processing here. As the vast majority of things that needs updating involve the players and staff.

Of course - more leagues IS going to be slower... but is it going to be significantly so, compared to less leagues but with the same number of staff loaded?