Letter of Grievances by some FM Fans

[x42bn6]

No, automation is the term Civ themselves use for it when a human unit is assigned to AI control.

The AI's units follow the same automation procedures, except with leader traits (etc) modifying that behaviour.

It's a system, by a large order of magnitude, simpler than that used in FM.

I'm saying that that modification is not simple at all. If FreeCiv has 15,000 lines of code for its AI, then it is extremely unlikely that Civilization IV has a very simple process.

If the moves picked are always the same, then that just shows there is no randomness.

Take an example - if a worker has to make a move, and it has two resources in view - one that is nearby, but one that is miles away surrounded by a forest, which one is better? Perhaps the nearby one is a worthless resource, while the one further away is pretty good. The AI has to pick which resource is best (or indeed, whether both are worth it at all).

Then say that the nearby resource has an enemy unit in the vicinity, while the faraway resource is closer to one of your cities so you can easily shelter there if things go wrong. That changes the weighting of the resources and may make the faraway resource a better choice.

This is the difficult bit and the AI is not as simple as you think.

The leader profiles likely weight each scenario differently. Montezuma, for example, might weight enemy units higher than resources, so he might secure the risky resource and crush the enemy unit with reinforcements. Gandhi, on the other hand, might flee and perhaps bargain for peace. Where that weighting comes into place depends on the scenario, and there is little reason to suggest that it is a simple process.

If it "seems" simple and efficient, then that is down to the performance of the AI, able to pick good moves (*) the majority of the time, quickly. However, being able to pick moves (*) is a complex AI process and is the true AI grunt work - that is complex.

Rule-based AI as you are suggesting doesn't work out that well when there are a lot of variables like in Civilization, because some rules may be too broad and some may be too narrow. It works well for small simulations. For example, going to war with another civilization may be a good idea, unless that civilization has a lot of tough allies, or is located in a poor strategic location, or if your economy is shot to pieces, or if it will annoy your own allies, and so on. There are so many variables involved that simple rule-based AI simply won't cut it.

[shirajzl]

How about when you have no transfer budget and very small wage budget? Let's say your wage budget is 5k a week. You can go as far as 6k or so, but the board won't let you go over that. How can you exploit that? There isn't money to sign good players for your level.

[Guest]

Why do I have a feeling that Ackter was actually involved in the development of Civ 5, am I imagining this?

[Ackter]

No, you're imagining that

I've been involved in a lot, but that isn't one of them.

[Ackter]

I'm saying that that modification is not simple at all. If FreeCiv has 15,000 lines of code for its AI, then it is extremely unlikely that Civilization IV has a very simple process.

If the moves picked are always the same, then that just shows there is no randomness.

Take an example - if a worker has to make a move, and it has two resources in view - one that is nearby, but one that is miles away surrounded by a forest, which one is better? Perhaps the nearby one is a worthless resource, while the one further away is pretty good. The AI has to pick which resource is best (or indeed, whether both are worth it at all).

Then say that the nearby resource has an enemy unit in the vicinity, while the faraway resource is closer to one of your cities so you can easily shelter there if things go wrong. That changes the weighting of the resources and may make the faraway resource a better choice.

This is the difficult bit and the AI is not as simple as you think.

The leader profiles likely weight each scenario differently. Montezuma, for example, might weight enemy units higher than resources, so he might secure the risky resource and crush the enemy unit with reinforcements. Gandhi, on the other hand, might flee and perhaps bargain for peace. Where that weighting comes into place depends on the scenario, and there is little reason to suggest that it is a simple process.

If it "seems" simple and efficient, then that is down to the performance of the AI, able to pick good moves (*) the majority of the time, quickly. However, being able to pick moves (*) is a complex AI process and is the true AI grunt work - that is complex.

Rule-based AI as you are suggesting doesn't work out that well when there are a lot of variables like in Civilization, because some rules may be too broad and some may be too narrow. It works well for small simulations. For example, going to war with another civilization may be a good idea, unless that civilization has a lot of tough allies, or is located in a poor strategic location, or if your economy is shot to pieces, or if it will annoy your own allies, and so on. There are so many variables involved that simple rule-based AI simply won't cut it.

I will repeat a further, and probably final, time. Turn on all of the automatic options available to you as a human manager. Watch what happens. The AI is almost exactly the same, but with the leader traits modifying them.

It seems complex, but it's a lot simpler than it seems.

The movement in Civ's AI is incredibly predictable, which is why the horrible difficulty modifiers need to be used to actually present a challenge. Have you ever even got close to losing on Chieftain level?

[aderow]

I kinda agree with Eugene. The AI just isn't smart enough to compete with you once you really know the ins and outs of the game.

Its easy to grab the best youngsters around the world. Things become even easier when clubs offer you the world's best in their prime.

I don't think that you should restrict yourself in any regard while playing this game. At least I wouldn't. I scout the world for talent. Thats what every big club does. I get plenty of top quality youngsters without much hassle from barca or real or milan etc.

Shoot I usually graba youngster or 2 from barca/real every season. Clubs need to be better at ensuring their present and their future.

I also don't think that every wonderkid should want to head off to manchester at a moments notice.

With that said, every issue I've got with the FM has to do with the AI and the ME (we'll talk about the ME on another day).

I know that developing artificial intelligence is incredibly difficult and, like I said before, I like to think that SI are doing their best to improve it.

[x42bn6]

I will repeat a further, and probably final, time. Turn on all of the automatic options available to you as a human manager. Watch what happens. The AI is almost exactly the same, but with the leader traits modifying them.

It still has complex decision-making! The AI is able to identify potential city locations and obtain resources. The AI can identify threats on the map (i.e. it will try to fortify undefended cities). It can identify opportunities (i.e. if an enemy city is unoccupied, it will try to occupy it).

If the AI has a choice between copper and dye as resources for one worker to enhance, how does it pick which one to choose? AI.

How does the AI know that it is best to bombard a city, get free hits, before moving in for the kill? AI.

How does the AI identify missing technologies and attempts to trade for them? AI.

It seems complex, but it's a lot simpler than it seems.

The movement in Civ's AI is incredibly predictable, which is why the horrible difficulty modifiers need to be used to actually present a challenge. Have you ever even got close to losing on Chieftain level?

There is a difference between "AI easiness" and "AI complexity". Go, for example, has extremely complex AI, but is still too easy for competent human Go players (you need to give the best Go AI headstarts in order for them to match mid-ranked players).

I haven't played Civilization IV for several years, but I've poked around the custom AIs and FreeCiv and know that perhaps the AI is a little easy at certain levels, but it is certainly a very complex AI.

I'm talking about AI complexity, not AI easiness. The AI in Civilization IV has to be complex, full-stop, as FreeCiv's AI is inferior to Civilization IV's, and FreeCiv's AI is 15,000 lines of code.

Of course, it's still flawed, but that's down to the sheer number of options available in a single turn in Civilization. The fact that Civilization IV is a challenge for newcomers to the series shows that the AI is complex.

It would be very easy to write a Civilization IV AI that was "not complex".

You have to take a look at the mods, and FreeCiv's source code to realise the Civilization IV's AI module is actually probably really complex, but even though it is really complex, its "best" AI is still inferior to veterans of the series, and is forced to "cheat" to match them. That's not shameful in any way - that's just how difficult it is to write an AI for a game that is likely on par with Go when it comes to complexity.

[Yskel osminovski]

The game is a little too easy this year a quick look at the global stats on steam achievements will tell you that. The key one for me being:

Invincible: Your Team Played the Entire Season without losing a league game - 8.7% seeing as this has only been done twice in the history of english football and probably something similar around the world, I would suggest that the game is a little too easy.

[Bruce Drundrige]

Stop comparing it to the real world!

[Ackter]

It still has complex decision-making! The AI is able to identify potential city locations and obtain resources. The AI can identify threats on the map (i.e. it will try to fortify undefended cities). It can identify opportunities (i.e. if an enemy city is unoccupied, it will try to occupy it).

If the AI has a choice between copper and dye as resources for one worker to enhance, how does it pick which one to choose? AI.

How does the AI know that it is best to bombard a city, get free hits, before moving in for the kill? AI.

How does the AI identify missing technologies and attempts to trade for them? AI.

There is a difference between "AI easiness" and "AI complexity". Go, for example, has extremely complex AI, but is still too easy for competent human Go players (you need to give the best Go AI headstarts in order for them to match mid-ranked players).

I haven't played Civilization IV for several years, but I've poked around the custom AIs and FreeCiv and know that perhaps the AI is a little easy at certain levels, but it is certainly a very complex AI.

I'm talking about AI complexity, not AI easiness. The AI in Civilization IV has to be complex, full-stop, as FreeCiv's AI is inferior to Civilization IV's, and FreeCiv's AI is 15,000 lines of code.

Of course, it's still flawed, but that's down to the sheer number of options available in a single turn in Civilization. The fact that Civilization IV is a challenge for newcomers to the series shows that the AI is complex.

It would be very easy to write a Civilization IV AI that was "not complex".

You have to take a look at the mods, and FreeCiv's source code to realise the Civilization IV's AI module is actually probably really complex, but even though it is really complex, its "best" AI is still inferior to veterans of the series, and is forced to "cheat" to match them. That's not shameful in any way - that's just how difficult it is to write an AI for a game that is likely on par with Go when it comes to complexity.

Each tile has a "worth" value. A potential city location just looks for a minimum value of worth within a certain distance of your starting point/current cities.

The AI automatically defends cities and resources - that's hardly complex. It sees an empty city and chooses to occupy it - again really not complex (especially seeing as half the time they ignore it if they're not full on military civs). Most of the time it doesn't even defend it with the right units, it just fills it up with a number that relates to the type of leader again.

I thought you were a coder yourself? How can you not see that most of what you're talking about here is simple "IF x then y" style code?

I have no idea what Go is, so can't really discuss that.

Civ is a challenge for newcomers simply because of the learning curve of how to do things. As soon as you know the basics, even if you don't do them well, you can walk the first three difficulty levels. Once you get into city specialisation, something that the AI does incredibly poorly on every difficulty level, you have no chance of losing.

Civ is just not complex in the slightest compared to even small parts of FM. The match engine, for example, is the most daunting bit of coding I've ever seen.

[Ackter]

The game is a little too easy this year a quick look at the global stats on steam achievements will tell you that. The key one for me being:

Invincible: Your Team Played the Entire Season without losing a league game - 8.7% seeing as this has only been done twice in the history of english football and probably something similar around the world, I would suggest that the game is a little too easy.

Real world doesn't allow you to save and reload.

[aderow]

Real world doesn't allow you to save and reload.

......at least for now....

[Ackter]

Well that just wouldn't end well...

[aderow]

Well that just wouldn't end well...

Imagine the possibilities

[marty78]

Real world doesn't allow you to save and reload.

Thought exactly the same. I got this achievement on FM11 but it was the Scottish First Division, I had given myself a sugar daddy and had a team full of Premier League players(I went for the hero achievements). Of course some will be legit but my guess would be that many will have reloaded, edited, used scouting tools or exploited the game in some other way. Even the best FM'ers will have one of those games where they just can't score and the opposition gets a lucky break which in itself will make this achievement extremely rare to get without cheating.

[interesti88]

Wow another FM basher thread, I usually don't comment on these kind of threads but its seriously getting depressing now. If you don't like the game fine stop playing it, if you want the GAME to be more realistic become a real manager go manage Manchester United, throw your hat into the ring to become the next England manager, I know Inter is struggling right now and we could really do with an intelligent manager guiding us through this dark time.

If you want difficulty settings why not just go play FIFA? You can change the settings on whatever you want. I really wish I could experience your agitation with the video game, so that way I can see your point. Unfortunately I'm enjoying this game more than any other console game or for that matter more than Inter currently in the Serie A. To me it seems as if you and your friends are looking more for errors in the game rather than enjoying the game altogether.

If it seems like I'm defending FM well I guess you can YES! I am. But not for them, I'm doing it for me because I come to these forums looking for advice, giving advice or to see how others are getting on, but its more and more negative posts popping up ruining my fun coming here to the forums. I'm not a negative person but its seems that a lot of you guys are, why is that?

[Aytumious]

Personally I feel the problem lies in squad building and management. The AI buys the wrong players and uses them in the wrong way.

If I had to pick the major problems with the AI, these would be the two. I realize it could be extremely difficult to program an AI that could behave better in this regard.

I also agree that the PPA is too accurate. The career paths of players are way too predictable barring major injuries or a player somehow being stuck in an obscure place where he can't develop. You could also add very low professionalism and ambition as well, though I'm not sure how much ambition matters for a players' development.

[shirajzl]

Wow another FM basher thread, I usually don't comment on these kind of threads but its seriously getting depressing now. If you don't like the game fine stop playing it, if you want the GAME to be more realistic become a real manager go manage Manchester United, throw your hat into the ring to become the next England manager, I know Inter is struggling right now and we could really do with an intelligent manager guiding us through this dark time.

If you want difficulty settings why not just go play FIFA? You can change the settings on whatever you want. I really wish I could experience your agitation with the video game, so that way I can see your point. Unfortunately I'm enjoying this game more than any other console game or for that matter more than Inter currently in the Serie A. To me it seems as if you and your friends are looking more for errors in the game rather than enjoying the game altogether.

If it seems like I'm defending FM well I guess you can YES! I am. But not for them, I'm doing it for me because I come to these forums looking for advice, giving advice or to see how others are getting on, but its more and more negative posts popping up ruining my fun coming here to the forums. I'm not a negative person but its seems that a lot of you guys are, why is that?

We're not being negative, it's called constructive(more or less) criticism hoping to improve the game. If everyone shared your point of view, the game would never progress and we'd still have an exploitable ME from the old CM days, for example.

[Ackter]

If I had to pick the major problems with the AI, these would be the two. I realize it could be extremely difficult to program an AI that could behave better in this regard.

I also agree that the PPA is too accurate. The career paths of players are way too predictable barring major injuries or a player somehow being stuck in an obscure place where he can't develop. You could also add very low professionalism and ambition as well, though I'm not sure how much ambition matters for a players' development.

Actually I think it could be a "simple" fix - stop the AI basing purchases on reputation and base them on:

a) What position in their squad is the weakest (modified by the manager's attributes)

b) Which players perform the best in that position (modified by the manager's preferences towards youth/experience/style of play etc)

c) How large a percentage of their budget a manager/board is willing to spend on one player

Then the complex stuff like balancing how many players a team really needs, sacrificing one position to strengthen others etc comes in

[Ackter]

We're not being negative, it's called constructive(more or less) criticism hoping to improve the game. If everyone shared your point of view, the game would never progress and we'd still have an exploitable ME from the old CM days, for example.

Well, some of this is constructive criticism.

The rest of it falls into three categories:

a) Whining

b) Pretending nothing's wrong

c) x42 arguing for the sake of it

[milnerpoint]

I know nothing about programming but i do have a friend who works for Capcom now a days, he tells me the FM ME is one of the most complex things in any game out there, and he wouldn't touch it with a barge pole, no matter what they paid. Clearly he is not a super programmer like X42 who can make anything work just by thinking about it, but until we all get to that super level of intelligence i think we will all have to cut SI a bit of slack from time to time.

[Ackter]

And what's your capcom mate's name, seeing as that's where I work?

(Don't post it here though, obviously)

[milnerpoint]

you have been PM'd

[Ackter]

Ahah, he's three seats away from me

[Govnar1]

Ackter you hit the nail on the head with a), b) and c) but forgot:

d) milnerpoint

@x42bn6&Ackter please, please continue this FreeCiv/Civ IV malarkey elsewhere (Off Topic Discussion).

[milnerpoint]

Do i get my own rule? Wonderful!

Ackter, its a small world!!

[srinath]

the game has defenitely become easy .. loved one f the save games in FM07 where after 8 seasons i managed to qualify for champions league with sunderland.But in FM 12 it is super easy to finish in top4 .but it is difficult to win the league.

[wwfan]

I know nothing about programming but i do have a friend who works for Capcom now a days, he tells me the FM ME is one of the most complex things in any game out there, and he wouldn't touch it with a barge pole, no matter what they paid. Clearly he is not a super programmer like X42 who can make anything work just by thinking about it, but until we all get to that super level of intelligence i think we will all have to cut SI a bit of slack from time to time.

Is there an echo in here?

[milnerpoint]

hahaha did you say that before me? I didnt read much of page two, just had to chip in with my point you know Cant have people thinking the game is not perfect as it is.........

Edit - just read, sorry does read like an echo! lol

[Ackter]

Incidentally, your mate is actually one of the best programmers I've worked with. The guy can sense a memory leak before it's even happened.

[milnerpoint]

He is a genius at anything he puts his hand too, he did a 5/6 year degree in bio medicine before moving onto computers. He is easily the smartest person i know and for him to be daunted by something tells me it must be hard.

[Guest]

Careful, any more of this & the programmers at SI are going to be demanding a hefty pay rise.

[milnerpoint]

Neil Brock going on strike for a few weeks?

[Neil Brock]

More like a holiday I think

[x42bn6]

Each tile has a "worth" value. A potential city location just looks for a minimum value of worth within a certain distance of your starting point/current cities.

There's still a fair amount of AI here, no? For example, the AI settlers won't setup shop next to barbarians, and they will run if a barbarian is nearby.

The AI automatically defends cities and resources - that's hardly complex. It sees an empty city and chooses to occupy it - again really not complex (especially seeing as half the time they ignore it if they're not full on military civs). Most of the time it doesn't even defend it with the right units, it just fills it up with a number that relates to the type of leader again.

Ah, but that is complex. What if there are two cities - which one is the better one to pick? Maybe it's better to attack opposing units instead of occupying another city?

I thought you were a coder yourself? How can you not see that most of what you're talking about here is simple "IF x then y" style code?

No. In fact, if-then-else code is probably one of the worst things to base AI on, as it never covers corner cases and assumes independent conditions (while it is possible to implement if-then-else with dependent conditions, it is extremely inefficient). If-then doesn't scale well at all and is only worth considering for small scenarios (i.e. "there is a fork in the road, pick left or right?").

I think it is more likely that there is a weighting system where S, say, is a scoring function. So if the AI has a choice between defending city C1, city C2, or attacking an opposing unit U, it will pick s = max(S(defend(C1)), S(defend(C2)), S(attack(U))). This is a lot more flexible than lots of if-then-else cases. It allows the AI to consider cases where attacking a unit makes more sense than defending a city (i.e. this is the only enemy unit close by, and I have a 99% chance of winning this battle).

It's been a while since I played Civilization IV, but I do remember settlers running from wild animals, except where there is a fighting unit nearby who can take care of the thing (in which case the settler moves after the attacking unit). Assuming, of course, that the fighting unit kills the wild animal. That's 3-4 if-then-else conditions already, which does not scale well at all. In fact, there would be so many corner cases it would be unmaintainable code.

I have no idea what Go is, so can't really discuss that.

Go (basically) is a game where you place tiles on a board and when you surround enemy tiles, you remove those and score points.

The reason why it's so difficult is that Go is played on a 21x21 board, so the first move has 441 possible combinations, and the second has 440, and so on - so for the first players moving, that's 441 x 440 possible combinations to consider - which is why Go AI is not going to be any good any time soon. The poor AI only has a chance against intermediate opponents, on a smaller board, with a handicap.

It is estimated that the number of unique Go games exceeds the number of atoms in the Universe.

Civ is a challenge for newcomers simply because of the learning curve of how to do things. As soon as you know the basics, even if you don't do them well, you can walk the first three difficulty levels. Once you get into city specialisation, something that the AI does incredibly poorly on every difficulty level, you have no chance of losing.

The AI up to that point is still complex, as discussed above.

Civ is just not complex in the slightest compared to even small parts of FM. The match engine, for example, is the most daunting bit of coding I've ever seen.

Have you seen Civilization IV's source code?

FreeCiv's AI is a pretty nasty piece of work itself. With Firaxis's resources, it is likely that Civilization IV's AI is even more complex.

[Guest]

When did we change to the Sid Meier forum?

The horse is dead already so give it a rest. Why do you always have to ensure that the last word on subject is yours! Until you can confirm that you work in the field of game development I am inclined to accept that on this matter Ackter has greater knowledge than you do.

[x42bn6]

When did we change to the Sid Meier forum?

The horse is dead already so give it a rest. Why do you always have to ensure that the last word on subject is yours! Until you can confirm that you work in the field of game development I am inclined to accept that on this matter Ackter has greater knowledge than you do.

I'm a software developer, and have experience with AI and mathematical techniques around it (i.e. some information theory and modelling techniques that can be used in neural networks). I've also read FreeCiv's source code to some degree (FreeCiv is an open-source version of Civilization).

[FrazT]

Lets get back on topic guys please.

[darkfng]

Amen to that.

[glennuk]

I don't know if this has been mentioned, but i would really like to see a change in the whole fog of war/scout system.

The problem at the moment is i can go thorugh each of the U21/U19 squads for any national teams, look at the best players in the squad and scout them. A DAY (A bloody DAY) later i get a full report and get to see all/most of their stats, and have an easy star rating that says, he will be better than your current 4 star Leading Premiership Striker.

Firstly i don't think i should be able to go through each of the u21/u19 squads that simply, especially for lesser known nations. Also scouting should take longer. A report card should not tell me his exact stats, One Day is NOT enough. I mean how long do your scouts take reseacrhing players in lower leagues?

Secondly, even if i can get a scout report in the day, why give me a star system. At the moment i just don't look at any young player under a 4 star potential. Simple. Why not just, this kid has the potential to go far, this kid could improve, this kid looks like he is struggling at his current level etc. Then it is a gamble....how many players have been rejected from one club/nation only to come and haunt them and be a world class player in the future? If the club could see they had 5 star potential they would never have not signed them after a trial/report card.

Thirdly, The out of 20 rating is great, but it should defo be an around about figure, not a this guy has 20 finishing, this one has 20 pace etc. I like the +/- idea. For example Cristiano has a crossing rating of 18 +/- 5, if your scouts are world class. Cristiano has a crossing rating of 18 +/- 3 (Probably 3 would be my limit) after you have been training with cristiano and he has been in some games a more accurate picture is portrayed, crossing of 17. Or at least have this an option (Make it harder for those who like it to be difficult, or use normal settings who those who like a more easy game .

Just some suggestion but there has to be ways to make it more difficult which are simplier than changing the whole game. EG: Having options you can turn off/on

[glennuk]

Back on topic. PS: Please go talk about civ somewhere else. If you want to code, go code. I am using this thread to suggest ideas on how to improve, i still love FM, but would like improvements, which i have suggested above

PS: Happy valentines day to all those out there who didnt get any cards/presents (I did, but then again i am getting married, so less time on FM for me )

[markyosullivan]

I've found that since 12.1.1 the game has got a lot harder. I've yet to win the top domestic cup in a nation (maybe Northern Ireland), I haven't won the top league in a nation yet either.

[TomaZorakFM]

You have ruined the perfect game. FM2012 was one of the best FM games until the 12.1.1 patch. From then on, game is fu**d up real hard. Why putting this thing in which if you are a former sunday league player, the game is much harder? It should be a simulation, not an arcade. The only thing you have to do now in game is keeping your team's morale sky high. If you are a sunday league player it's impossible, and if you are a international footballer, than it is a piece of cake. FM2012 without patches was a very good game indeed, but now it sucks and I haven't played it for about month or two. I just hope that new patch will come out soon, and make the game playable like it was before this riddiculous 12.1.1 patch and difficulty changes...

[Neil Brock]

You have ruined the perfect game. FM2012 was one of the best FM games until the 12.1.1 patch. From then on, game is fu**d up real hard. Why putting this thing in which if you are a former sunday league player, the game is much harder? It should be a simulation, not an arcade. The only thing you have to do now in game is keeping your team's morale sky high. If you are a sunday league player it's impossible, and if you are a international footballer, than it is a piece of cake. FM2012 without patches was a very good game indeed, but now it sucks and I haven't played it for about month or two. I just hope that new patch will come out soon, and make the game playable like it was before this riddiculous 12.1.1 patch and difficulty changes...

The higher two 'past experiences' (international player and professional) haven't actually changed at all, it was just the Sunday League and Semi-Pro were changed as they were incorrect. We fixed a bug. Nothing at all has changed in regards to 'difficulty', just the starting point of the lower two experiences. Also we'd very much appreciate you don't swear on these forums.

[TomaZorakFM]

First of all, sorry for swearing, but only swear can express how I feel. It would not happen again

How it is not changed in regard of difficulty when if you start as sunday league player anything you say on team talk made players or angry or nothing specific happen, and if you start as International footballer, you only need to say that you have faith and everyone becomes happy or motivated.

[SiN8]

I don't know if this has been mentioned, but i would really like to see a change in the whole fog of war/scout system.

The problem at the moment is i can go thorugh each of the U21/U19 squads for any national teams, look at the best players in the squad and scout them. A DAY (A bloody DAY) later i get a full report and get to see all/most of their stats, and have an easy star rating that says, he will be better than your current 4 star Leading Premiership Striker.

Firstly i don't think i should be able to go through each of the u21/u19 squads that simply, especially for lesser known nations. Also scouting should take longer. A report card should not tell me his exact stats, One Day is NOT enough. I mean how long do your scouts take reseacrhing players in lower leagues?

Secondly, even if i can get a scout report in the day, why give me a star system. At the moment i just don't look at any young player under a 4 star potential. Simple. Why not just, this kid has the potential to go far, this kid could improve, this kid looks like he is struggling at his current level etc. Then it is a gamble....how many players have been rejected from one club/nation only to come and haunt them and be a world class player in the future? If the club could see they had 5 star potential they would never have not signed them after a trial/report card.

Thirdly, The out of 20 rating is great, but it should defo be an around about figure, not a this guy has 20 finishing, this one has 20 pace etc. I like the +/- idea. For example Cristiano has a crossing rating of 18 +/- 5, if your scouts are world class. Cristiano has a crossing rating of 18 +/- 3 (Probably 3 would be my limit) after you have been training with cristiano and he has been in some games a more accurate picture is portrayed, crossing of 17. Or at least have this an option (Make it harder for those who like it to be difficult, or use normal settings who those who like a more easy game .

Just some suggestion but there has to be ways to make it more difficult which are simplier than changing the whole game. EG: Having options you can turn off/on

While I agree that scouting should be less accurate, I want the human player interaction to be simplified. The last thing I want to do playing FM is having to repeatedly micromanage individual scouts. In fact, I'd like FM to look at Out of the Park (OOTP) baseball sim for their single scouting director model. At any given time, you are given some estimation (never the real attributes) of all players in the game. The accuracy of the estimation depends on the scouting budget and abilities of the director. Obviously, some 16 yr kid from Kazakstan will be wildly inaccurate. The scout probably saw nothing more than a youtube clip or a story from the local paper.

This way, you will have star ratings for all players at all times, but with a note on the perceived accuracy. In OOTP, you have the option to scout individual players as well for better accuracy.

[glennuk]

While I agree that scouting should be less accurate, I want the human player interaction to be simplified. The last thing I want to do playing FM is having to repeatedly micromanage individual scouts. In fact, I'd like FM to look at Out of the Park (OOTP) baseball sim for their single scouting director model. At any given time, you are given some estimation (never the real attributes) of all players in the game. The accuracy of the estimation depends on the scouting budget and abilities of the director. Obviously, some 16 yr kid from Kazakstan will be wildly inaccurate. The scout probably saw nothing more than a youtube clip or a story from the local paper.

This way, you will have star ratings for all players at all times, but with a note on the perceived accuracy. In OOTP, you have the option to scout individual players as well for better accuracy.

yeh i like that idea. I am not actually saying more micromanaging, i am saying that we shouldnt be able to automatically get a scout report after a day. Kind of unrealistic i think.

But some like the OOTP style or +/- style would make a huge difference to the game, making it harder and mroe realistic.

Imagine if i as a lower league club and found a newgen who looked like he had 15 finishing (Great for BSP, L2, L1) but it had a +/- 8 on it as my scouts are crap. Signing him would be a huge gamble, as he may turn out to have a horrible rating of 7, or it may be just reasonable like 10 or great like 15.

More risk, more reward, and a harder system.

[Smurf]

The higher two 'past experiences' (international player and professional) haven't actually changed at all, it was just the Sunday League and Semi-Pro were changed as they were incorrect. We fixed a bug. Nothing at all has changed in regards to 'difficulty', just the starting point of the lower two experiences. Also we'd very much appreciate you don't swear on these forums.

Past experience should have a greater affect on your performance and the shortness of the Board.

If I start as lowest level at Barcelona, dropping points should result in my dismissal. And, I shouldn't be able to attract high-class new signings.

I'm not saying to rubber-band the game, I'm saying that the Board reacts more severely to poor performances than they do a manager with more experience.

And if I'm an unknown manager, then players over a Certain Reputation wouldn't be inclined to sign for me. In essence it should be harder to sign players because I'm relatively unknown.

Or perhaps - limit the clubs the player can choose from, and if you pick that you are a Sunday League player and you attempt to take over at Barcelona, the Barcelona board could simply reject your "application".

C'mon you picked that status for a reason. You shouldn't be able to get that job at that level anyway.

Perhaps selecting Sunday League would limit the options to tier two leagues.

And the next level to tier one leagues.

Next level to top tier leagues.

While I agree that scouting should be less accurate, I want the human player interaction to be simplified.

This.

I said last year and the year before, I should not have to babysit my Scouts and send them out. Ok they sorta do it now, but they repeat the events they attend, the leagues they scout etc.

You should have Head Scout. And they continue to re-base the scouts to the Scouting areas you have little knowledge of, or to their best locations, or to the scouts weakest knowledge, strongest knowledge etc.

I see it like this

Head Scout (HS): You wanted to see me gaffer

Manager (M): Yes I want the scouts to scout their best regions and leagues

HS: Ok

HS: You wanted to chat to me

M: Yes please send the scouts to improve their knowledge in their weaker areas

HS: Hello gaffer

M: Send the Scouts with the best attributes to find Youth Players (and only those scouts with good Youth scouts these)

In team meetings - I don't think "you should scout this player" should come up.

The Scouts should just GO and DO IT.

To top it off. You shouldn't have to go to each individual scout to get a scout report.

You should simply Click on the Head Scout, and all the reports are accumulated in one.

yeh i like that idea. I am not actually saying more micromanaging, i am saying that we shouldnt be able to automatically get a scout report after a day. Kind of unrealistic i think.

Not if the scout is from that area. Or knew that league very well. Any scout, I feel with good knowledge of an area would be able to drum up a report on players in a day.

What is unrealistic is if I send my Scout that's never been to Africa to scout for players and he returns in a day... it really should take him longer than normal to drum up a report.