[Player valuations still make no sense at all, and it's still a problem]

It probably has a lot to do with the expected national transfer/wages values baked in the game. Happened to me in Slovenia eons ago: players who would be asking for millions per week in England were happy at a tenth of that (which is still very solid money mind you) in Slovenia because 1) I found them when they were nothing and cost nothing and 2) expected wages value sin Slovenia are very low. Same for transfer values when selling a player... to a point: when my club became a 5 star reputation club (and soon after the most reputable club in the world), which happens when you win the CL since its reputation is the maximum 200 out of 200, I started having "normal" transfer bids from other big European clubs for my star players even though the Slovenian first division wasn't top 10 in reputation rankings. But until then, it was terrible offers after terrible offers that don't really respect the quality of the player, the financial situation of the club (as you don't need the money at all), the money you've spent bringing him in, nothing. Basically the offers were indexed on the wages and in-editor values the AI should abide by. That being said, obviously if you decide to, say, quadruple his wages, incoming offers will at least reflect that. 20k per week isn't much for a world class player; I'm repeating myself, but that's also partly why you get terrible offers: the players don't cost much to you either.

A lot of aspects in FM are pretty stiff and rigid and don't contort well to fit very unexpected situations, like a club from absolutely nowhere becoming a worldwide superpower. Regarding winning the CL, FM also doesn't value well having excellent consistent runs as much as getting the trophy. There's a huge gap between the reputation earned by winning the CL once (which will skyrocket you to the top of the world) and doing nothing of note for years, and say, getting 3-5 semi-final runs; even if the latter is more reflective of how sound your club actually is. Likewise in Slovenia and until I won the trophy, for like five years in a row the end of season recap for the CL noted our club as the most surprising outfit... even though I was really consistently getting to the latter parts of the competition, which any real life analyst would have noticed. Then I won it, and the game went "oh, they're obviously world class now"... which we already were since it wasn't exactly our first rodeo deep into the competition. Consistency and doing well without quite winning the trophy isn't well rewarded in FM, and there's a lot of inertia regarding reputation.

[*Official* Football Manager 2023 Feedback Thread]

4 hours ago, forameuss said:

Scheduling is definitely an issue, particularly as you go deep into a season being "unrealistically" successful with some teams.  But I don't think it's necessarily something you can just go in, spot a glaring bug and resolve.  I imagine it's like a microcosm version of the match engine where you've got to balance this equation perfectly with a million variables.  You'll likely get it for 95% of those variables, but it's the 5% that throws up the weird cases where the algorithms come out with something that doesn't fit.

I might be repeating myself a bit, but I'll try adding something new. As far as the computer go and when left to its own devices, the fixture scheduler tends to stack matches early and late in the season, which leaves some stretches of mid-season perfectly empty around February to April for teams that got knocked out of continental football (or cannot be involved in continental football at all like in most second divisions). You can help the game by tweaking the database, adding alternate dates, adding different season starting and ending dates to accommodate for big competitions, allow matches to be rescheduled, make pitch heating mandatory for top division clubs in big countries to avoid waterlogged pitches in winter. If you're too heavy handed and not careful enough by mandating something like having all matches played on the same day, then you end up in a hot mess.

Basically it's not just the fixture scheduler, but the overall care given to databases outside of big countries that leaves to be desired. It's not easy since SI themselves aren't specialists and leave the community to fill the gaps, but the Editor is not exactly a friendly piece of software (far from it), and the nerds dedicated enough to even bother with it are few and far between. Not even SI staff often visits the Editors' Forum to give pointers on how the damn thing works; at the very least I haven't seen them often at all. I can tell you that even if a well done database is very satisfying, I don't exactly miss the time I've spent battling with it. I'm grateful to the people from that par of the forum who've made it possible for me, but yeah, it's rough.

I hope that this is something that can be improved moving forward in FM25. I haven't delved in Unity modding, but I'm aware that the Unity engine allows for mods and has a pretty enthusiastic community. I have no idea on how it would or even could translate to an hypothetical version of Football Manager that's not even out yet.

[*Official* Football Manager 2023 Feedback Thread]

20 minutes ago, andu1 said:

Thanks for the reply. I digged a bit in the editor but i would have to start a new save if i do any changes to the league format.

I am invested in this save since i always like to do in a FM a save where i take a league from ground bottom to the top of the European Coefficient table and this save is going well enough for now.

That's why i even saved a copy of the save where i am now which i will continue in FM24 and compare my progress with FM23.

Then force matches on Wednesdays and Saturdays, that will help. If you carry your save over, I expect that you will play with the rules you started your save with so be mindful that if you don't like it now, I doubt you're going to like it next year as well. It's the same when you have an incremental update: saves created on 23.1 use 23.1 rules even on 23.4. I'd be surprised if it were any different to be honest.

As far as  the bolded part: you're going to have an issue with Reputation Manager™. The other clubs in your league will spend absurd amounts of money to get fairly mediocre players, or not spend money at all because they only can attract mediocre players because of their very low reputation and complete inability to scout quality players all over the world like the human manager would to get out of this deadlock. To "help" them help you on the European front, you should hire as many decent youngsters or cheap quality players and loan them out to the clubs of your league while paying for their wages. Yes, you're going to subsidize the clubs in your league. If you can sell them for no money at all that's fine, but they'll ask for absurd wages so be mindful of that.

Subsidizing clubs in your league shouldn't be a problem since once you win the CL a couple of times, you can end up winning every match every year in every competition (even the CL) really quickly as the AI isn't too good at countering you when you get your reputation going and your scouring on point (it happened to me and it was ridiculous). You should earn incredible amounts of money that you have no use for since you can't exactly pay English wages either by collecting huge paychecks from either selling kids you've scouted around the world, selling older players at your club to make way for youngsters, and going far in Europe; rinse and repeat.

[*Official* Football Manager 2023 Feedback Thread]

2 hours ago, andu1 said:

Luckily i managed to get through.
But, as i specified and as you can see, there was a free date to play the game on Saturday, 3 days before the Champions League game so i stand by my reasoning that the scheduling must be fixed.

 

The fixture scheduling being questionable at best isn't quite news. If you don't manually schedule matches very tightly in FM, you can end up with fairly absurd results. You need to create the fixture list, schedule holiday breaks correctly, the hierarchy and importance of fixtures and schedule alternative dates in case of conflicts, all that while the game will generally refuse to extend or contract the season beyond the dates that are fixed in the database. You also need alternative dates or schedules once every two years to account for international tournaments. And sometimes you still get absurd dates because some dates are fixed and the author of the schedule didn't account for every possibility while IRL the powers that'd be can be more flexible about it... except when they don't want to be. Was it Liverpool that was expected to play both in the League Cup and the Club World Cup in two different countries a few years ago?

Few years ago I had that in Slovenia where matches were scheduled on Saturdays for some clubs with few if no alternative dates, which meant that teams playing the Europa league on Thursdays would get mauled in league games on Saturday. A more lenient scheduling would see them playing on Sundays or even the next Monday if necessary. The general rule is that the further you manage from England, the less care has been put into getting really tight database rules; to the point that some of the amateur projects have tighter scheduling than some official databases especially from more obscure countries. And sometimes the Editor will still fail to validate database rules and find conflicts that don't exist.

I hope that with the rewrite to another game engine that SI can crack down that kind of annoying relics from the past. But that has existed for years, it's probably not the first time me or others have talked about this, and I doubt SI's going to fix this next year with the direction FM's going regarding its future. If you're curious, you could load the pre-game Editor and check the rules for the Romanian league. You could try your hand at editing them if you desire, but that's a deep rabbit hole. There's another alternative that allows you to keep your current save: there's a setting that forces matches on Wednesdays and Sundays and almost completely eliminates the problem at the cost of less "realism" in terms of fixture dates. It also applies only from the next season onwards, so if you end up not liking it you'll have to play a whole season like that before changing it.

[*Official* Football Manager 2023 Feedback Thread]

On 28/06/2023 at 19:36, Tyburn said:

If we’re allowed feedback on the future here … 

https://www.footballmanager.com/news/future-football-manager

 

To be honest, I wouldn't have locked the thread since it's genuinely one thing I'd want people to talk about. It would be hell to navigate and moderate, but that's par for the course. That, unless the powers that'd be have a very good reason not to allow further discussion about it within that thread at least.

[How long in the same club?]

I prefer doing my whole career at a single club. Unfortunately much like the real life market, loyalty is poorly rewarded in FM. Your wages (and therefore job security) as well as the board's compliance to your requests are worse when you stay at your club for a long time, even if everyone loves you. If you want better conditions for yourself, you're better off leaving, even if you plan to rejoin later. During that time, you'll have the absolute "pleasure" of watching an AI manager tearing down your work to pieces.

[Defending with a 3-3-3-1 formation]

On 13/12/2022 at 19:54, Skywalk3r83 said:

Yeah maybe for a new experiment later on in the save👍🏻

I've done the 3-4-3 diamond in the past thanks to the thread below.

Unless the ME has changed a lot, the answer is WB(A). CWB(A) seems not to be as smart and effective in its choices for some reason even though the role description is what you'd want out of him. WB(A) is a lot more straightforward: they will occupy and offer width, they'll dive into the box, they're actually a bit more free in their movement than CWB(A) or IWB(A), and they're not completely selfish either like Wingers on Attack duty. CWB(A) stays wide a lot, IWB(A) always tries to cut inside, WB(A) can do either depending of your team's width. You don't even need someone who's particularly good at defending there, you can very well stick a AML/R or a ML/R in the slot; you'll just have to tolerate your AssMan moaning about Player X being played out of position. The only issue is that you'll defend with a back 5 when you're out of possession with a narrow diamond in the middle. Not only it isn't really how the 3-4-3 was played defensively, but you'll have to be somewhat careful on the wings. Try to retain possession as much as possible to prevent WB(A)s from running too much to cover both their offensive and defensive prerogatives.

Ah yeah, and IME the CAR role doesn't do anything interesting. A player can perform with that role, bu the role itself just doesn't seem to do much even though you'd think it would be welcome in such a scenario. It's puzzling how it's tough to figure out what they actually do differently from other roles... if they do anything at all. This video is from FM21, but I'm not convinced the role has changed with FM23; I'm open to being mistaken though.

A customized CM(S) or BBM are more interesting, alongside partnerships like DLP(S)/RPM(S) or double MEZ. CM(S) offers recycling options while also offering a threat from range, BBMs run forward and especially backwards a lot as they have for hidden trait the fact that they track back deeper when the team's defending, MEZ is more offensively inclined while also working in the channels/half-spaces a bit more. For those midfield roles, the players are going to matter a lot more than the role itself: they need to be able to find space for themselves and make versatile choices.

[Netherlands Lower Leagues]

On 15/12/2022 at 16:09, Zek said:

The main reason is there's still no mandatory promotion/relegation between Keuken Kampioen Divisie (Tier 2) and Tweede Divisie (Tier 3). It's supposed to be introduced next season, but I wouldn't be surprised if it was postponed for a few more seasons once again. In the past decade it has been postponed again and again. Not many clubs from the top three leagues have much interest in promotion either.

The Dutch FA also owns the licence rights for the teams in Tier 3 and below, whereas the licence rights for the top two tiers are owned by two different partnerships that aren't owned by the Dutch FA. The Dutch FA is really touchy about the licences, hence the Dutch NT coach being named Jan Jansen since FM13 or so and the FIFA and PES games in the 2000's either not having the Dutch NT in the game at all or having fake names. So SI may consider it's not worth getting those licence rights, even if they wanted to.

 

Those who never had their attack spear-headed by von Mistelroum have greatly missed out.

[Netherlands Lower Leagues]

8 hours ago, phnompenhandy said:

I get the impression that not only has the pregame editor improved over the years, but so have amateur editors who create these database add-ons. I see very few complaints about them crashing the game these days, and I haven't experienced any problems, as someone who plays the game exclusively with lower leagues added.

More so the latter than the former unfortunately; or at least that was my experience of it. The Editor is a very finicky piece of software. I'm still very thankful to the people who helped me, have tried the options out and overall paved the way, but I cannot say that the Editor is something that was a pleasure to deal with; actually I've just cut a good rant about it from this very post you're reading!

[Low CPU use in the game.]

Short answer: no.

Medium-length answer: FM is very, heavily single-threaded.  That doesn't mean there isn't a couple of cores that are working hard; just that FM doesn't use everything. Unless you have tons of leagues enabled, the number of cores of your CPU has little impact on performance, and you always will see low overall usage.

Longer answer: There's a good post about it below that sums it up.

 

 

[Netherlands Lower Leagues]

5 hours ago, phnompenhandy said:

Okay. Maybe you'd like to volunteer to be a researcher for the Dutch lower leagues.

It's not even a research problem; having spoken about it with the Dutch researchers, the lower leagues are well researched as far as structure and clubs go. Most clubs have their correct stadium, city, kits and league; only the ruleset and a lot of players are missing. The latter can be managed by using the "Add players to playable teams" option and by waiting it out since there's quite a few free agents in the country. The former requires either passionate users... or for SI to actually want to buy the rights. Nevermind that the Dutch FA cannot seem to figure out their own structure IRL, with the situation between the Tweede an Eerste divisie still not really solved AFAIK.

SI just doesn't have the rights to the lower leagues and that's about it, like many other leagues. You could try the Feature Request forum section... though I doubt there's a lot of interest in it. Having built such a db in the past, I know there are passionate people there; just not that many of them on the grand scale of things.
https://community.sigames.com/forums/forum/353-football-manager-feature-requests-pcmac/

 

[[FM23] Netherlands (D7)]

On 20/11/2022 at 21:50, Wolf_pd said:

It may be merged, but would still have the situation that Saturday teams would strictly play the Saturday (at least the religious ones). Besides there is still the lower levels below that that are split.

FM's match scheduler is notoriously picky at times. IIRC in the older DBs that have been made (tho' I can't check mine), it's a league rule that matches occur on a designated day (Saturday or Sunday). FM will gladly move matches to the next or previous day if it detects a conflict like a stadium being used for another fixture, cup matches, TV rescheduling of higher division matches affecting lower division clubs, TV rescheduling of cup matches, reserves/youth team matches, winter break, waterlogged pitches (happens often in winter!) or just for the lulz because FM can and therefore will. This happens even if you set your days very accurately since the lower divisions have very low fixture priority. That being said, considering how FM isn't particularly sensible with its fixture scheduling, it can at times be a good thing to move matches around a bit... and seriously infuriating at other times when you cannot easily figure out why a match was moved around.

In the older dbs it would often happen because there weren't stadiums for every teams early in the save, especially at lower levels. The very complex and long playoff between the Eerste Klasse and Hoofdklasse would get extended for weeks on end due to the lack of stadium availability, as a given stadium can only be used once per day. As the Dutch research team have added stadiums to the database throughout the years, it's been less prevalent. The disappearance of the Amateur Super Cup, which happened at the end of the season between the winners of the Districtsbekers also helped. But it still happens, and there's not much that can be done about it.

On 23/11/2022 at 09:39, DurLennart said:

I believe the only 'fantasy' element that is currently in play is the 2 relegation places and relegation/promotion playoffs between the Keuken Kampioen Divisie and the Tweede Divisie. I believe it has been decided that from 2024 there will be 1 relegation and promotion spot in the KKD (no playoffs). But I think the implemented promotion system in the database suits the pyramid better as it is the same in all the other Dutch leagues. 

It was indeed an explicit choice I've made for purpose of homogeneity (all leagues having similar rules regarding promotion/relegation and Periodes) and fluidity between the leagues. Moreover, due to the transfer/wages values and how in vanilla FM there are only two divisions, the money and reputation drastically increase once you reach the Eerste Divisie and therefore are forced to turn pro. It also helps to explain why there's not much incentive for IRL amateur clubs to turn pro due to the financial commitment it requires. I'm not a specialist of Dutch football, but the issue of that Tweede to Eerste Divisie promotion's been going on for years at this point with no definitive answer.

[[FM21.4] Netherlands (D7)]

On 10/11/2022 at 11:44, wfm18 said:

Would you know someone who can update this file? I'm good at changing kit colors, but bad at competition rules haha. I always get errors and I don't understand what to do.

 

On 14/11/2022 at 00:41, Tomijssel said:

@wfm18 Advanced rules settings can indeed throw the most funky error's withour giving any guidance. I'm far from having the skills and knowledge of the original creators of this database, but I managed to port this database over to FM23, check it out if you like!

 

Thanks to the work of the Dutch reasearchers, most of the leagues should be in the right order. However not only the Districtsbeker do not exist as a database item in the vanilla db (you have to create them, although importing a db from an older version of FM may work), I also had split teams in their respective Districtsbeker groups manually in order to have a fairly even number of teams per Districtsbeker and achieve geographical consistency, so from one version to the next you may get mismatches or errors because the number of teams in each group may not be right for the ruleset. But since there's a flexible amount of teams in the all the Districtsbekers and KNVB Beker, it may work with a dirty port. I cannot speak to the quality of the work done for FM22 or FM23 because I had no hand in it, I can only mention that I (and many others that I thank yet again for their work and help) left most if not all the references and discussions and processes I went through in the thread below which you can find in the OP.

TL; DR : Aside from the Districtsbeker that may need extensive manual tweaking, most of the rules can be imported from one version of FM to the next without a huge amount of effort (assuming the IRL rules haven't changed too much). Ah, and I forogt: be careful of the COVID rules since I suppose they do not apply in more recent versions of FM. You may have to edit or delete them, as well as the rules related to the Qatar WC.

[[FM21.4] Netherlands (D7)]

4 hours ago, wfm18 said:

@Tomijssel @Xavier Lukhas Will this file be continued for FM23?

Unless there are major changes in how FM is and works, I don't see myself playing the game anymore.

[Evidence Based FM's testing seems to show that Match Prep sessions increase the likelihood of losing the next match compared to no training at all. Is this a bug or something else?]

This time around, we do have the schedule.

 

[Will you be buying FM23 ?]

On 10/09/2022 at 17:49, kertiek said:

see we are referring to different things, the graphical representation of plays are not my main point, they could use only one lame animation for every dribble/feint over and over and i wouldnt have a problem with it, in fact if the solution to get more dribbles in game is watch the game in 2d then i would play it in 2d but is not the solution because those plays simply are very rare to happen ingame matches, the graphical representation of those plays is not the problem for me.

the problem is that those plays are so rare,  plays from skilled players literally leaving players behind them due to a skill/dribble used, currently in fm22 most of the "skilled" players get a defender glued to them specially when u play with wide attackers, most of IFs and IWs have the instruction "cut inside with the ball" but that rarely happens, what is the most common play they do? run wide with a defender glued to them and they either cross or pass it back to recycle the ball when they have the ball, when they dont have the ball the run into space and end up scoring one taps or assists to other players but thats not what the "cut inside" instruction is, the ingame description of it is "ask wide players to look to come into central areas when running WITH the ball, driving inside their opponent and heading towards the goal" and that rarely happens.

i would rather see mbappe messi and neymar attempting dribbles and losing  the bals 10+ times per game than what i currently get is just a running wide with the ball to cross or recycle ball back to the midfielders, if dribbling/feints/skills from players arent going to be a common occurrence even when instructed to it, then what is the difference between top tier matches and a match from serie C in italy from example?.

in the last 2 or 3 iterations of the game dribbles have been nerfed greatly, and by this i mean actual plays not looking at the match stats (because that is another whole problem in fm22 por example running wide with the ball is counted as a dribble but the defender is glued to the attacker the whole run) play a game in fm15 or fm16 and watch for actual dribbles, defenders left behind walking because they cant keep up with the attacking player, those plays rarely happen in fm22, its not normal or immersive to see those plays 3 times per season at most.
 

what i mean for dribbles/feints/skills ?  attacker players literally leaving defenders behind due to feints/dribbles and, not having the defender glued to them the whole play, players like mbappe, neymar, antony, luis diaz, dybala, Ousmane Dembele should  be doing plays like those way more than watch they do ingame.

 

 

It's fun to dig old FM videos at times and see where we at compared to where we were.

It's so crudely over-the-top, yet so hilarious to watch. A lot more fun than when people were comparing old regen graphics vs new ones... or old pitches against new ones.

[Most of the players I have sold flop in their new clubs]

3 hours ago, TokyoWanderer said:

The AI picking a player based on a single attribute of the opposing team wouldn't be a good idea and it isn't realistic either. A team may adapt its style depending on the opposition, and if the tactics/style change, it may affect who starts.

Unrealistic I don't know; I would've though it could've given the edge to some players over another by being exceptional in a given department compared to the mass of other players, or make a team being able to single out someone in a mass of data (after all IRL players don't have attributes either, that's not realistic). Baseball enthusiasts tend to crunch number to that kind of degree, and certain attributes do work on their own without supporting attributes, so I don't see it as unthinkable. I can see where you come from though, I'm simply thinking that if the human player can select players with such minutia, you could consider that the AI also could. Now whether the AI should is a different topic...

Anyway, we're drifting further from the topic, so I'm happy to leave it there.

[Most of the players I have sold flop in their new clubs]

6 minutes ago, TokyoWanderer said:

No, you wouldn't. There even isn't such as thing as "training program for a player". Teams work in groups. Look for a video of a team training in real life and you will realize that the concept of min-maxing makes zero sense.

Also, how is a forward gaining 1-2 points out of 200 in marking something notable? Do you guys know that forwards play defense too, right? Players can get marginally better at something even if they are not doing specific training on it outside of matches.

I don't know, I've read reports of players given specific tasks, like Beckenbauer's OM staff specifically training Basile Boli to do 100m sprints followed by a cross over and over again. And to some extent, players do have personal programs in FM, even if they do practise together overall. That's this kind of min-maxing I was thinking about, the kind you see in the gym when you want to target a specific muscle group and design/choose exercises around that.

6 minutes ago, TokyoWanderer said:

On your other point about CA and the AI picking players, different ratings have different weights when it comes to CA, so for example high pace already gives a striker a high base for his CA. You can get pacey strikers to score a lot of goals because the ME is not perfectly balanced, but you can't design an AI to buy players who exploit the match engine. If they are buying the players with the highest CA possible for positions of need, the AI is doing a good job. The human player will always have an advantage because it can exploit the ME in ways the AI can't.

I was thinking about it not in the sense of exploiting the ME, but that the AI knows and tells you in their reports that they value pace. They also have report on how the attributes of your team stack up vs. the league or another team. So I wouldn't have thought that it was far-fetched to have an AI say that this player is faster than the average defender from this team, therefore let's play him instead of another option. It's not exactly unknown of IRL, or so I summarize. By extension, it could be part of the scouting setup to search for a given player profile to fit a given role. Or at least, that's how I was thinking about it.

[Most of the players I have sold flop in their new clubs]

13 hours ago, kentonizking said:

I love how his marking has gone up for......reasons?!

If anything, the thread below shows how opaque at best training is... and how unintuitive it is at worst to be polite. It's kinda jarring when you don't know what you'll get out of training. I wager that IRL you would know what you expect out of a training program for a player and how to min-max your gains, but maybe I really don't know anything about it. There can be surprises, but I'd have thought that trainers, coaches or sports scientists would have a pretty good understanding of how to get results that are somewhat in line with what they expect.

 

[Most of the players I have sold flop in their new clubs]

The AI also works under the assumption that having good depth is having at least three first-team level players per position (roughly 3 stars relative to their team's strength). That's how you see clubs with more money than sense stacking 3 or 4 strikers on their bench coupled with at leat another couple other players who can play the position, even though they play a one striker formation. A  few editions ago, Mourinho's Man Utd was notorious for buying basically every striker under the sun and end up with a main squad with 5 or 6 players who can play the position. A human player with good rotating skills (which is anything better than the AI, the bar is that low) would only really require two strikers and perhaps one or two players whose main roles are AML/R but can do the job at the ST position in case of injuries.

in other words, IRL big clubs also have tight squads that they rotate a lot, while smaller clubs have big squads with little rotation. In FM, big clubs have gigantic squads with no rotation.

[Players leaving on free for significantly less salary]

On 28/08/2022 at 17:17, RoryWatt said:

I don't know if this is a bug or not, but it seems very unrealistic:

- Eriksen on 75k p/w, 6 months left, 32 years old.
- Wants 190k p/w for 2 years.
- Rejects 130k p/w for 2 years
- Signs with Ajax on an end of contract of 44k p/w for 2 years.

This can't be right, right? I know he loves Ajax... but still.

Wage demands in the Eredivisie are comically lower than they are in the EPL. You can check the values for each country in the pre-game editor. I'll try to summarize, but it's not easy: let's say a player has a reputation of 140 out of 200. A player in the Netherlands with such a reputation will have wages expectations of 50k per week, while a player in England will have wages expectations of 250k per week. Now if the Dutch club has better reputation and/or plays in the Champions' League (which turns every players' heads, almost no exceptions), he could agree to 50k in the Eredivisie while rejecting a 190k offer in the Premier League. The wage and reputation values are made up, but that's the gist of it. It also works in reverse: if you manage a big club in a small country, you can retain a world class player by having them sign a new 75k per week contract even if PSG comes knocking at the door with literally ten times what you offer him. Because for a 180 reputation player in Boldhagivskan, 75k is amazing for a 180 reputation player while in France, 750k is almost an insult for a player of such calibre and way too little.

It's indeed ass backwards, but I guess it was designed to prevent players from asking way too much money relative to the country they're in, or importing anything and everything to countries where it wouldn't be likely to do so. In the Netherlands the wages and transfer values are **** low despite being a well developed footballing nation, so a pro or semi-pro outfit actually struggles to attract foreign players while players already in the country will ask for breadcrumbs to play. It can indeed lead to very nonsensical moves, although very high-calibre transfer moves tend to be immune to that phenomenon: even if a small country with horrible transfer values, you can still ask and get paid 200M for the reincarnation of Maradona... to some degree. Those expected wage/transfer values are also part of the reason why a player's value can wildly inflate when they transfer to a new country, even if the transfer fee wasn't anything special. The AI does take it in account in general... maybe too much in some cases.

Basically it makes some kind of sense in FM, but it's not very realistic as the living standards aren't so far apart between England and the Netherlands that you'd rather live with 44k in Amsterdam than 190k in London. And sometimes you have some transfer moves that just don't make much sense, even all things considered. Like players signing for a team in a lower division of your own country, for lower wages when it's not just no wages, in a club with less Reputation and with worse playing expectations.

[Most of the players I have sold flop in their new clubs]

5 hours ago, DMVian said:

Could player reputation be the reason? [...] But your post and other examples in my game makes me think player reputation is a really underrated aspect in value.

I'd bet a couple of drinks that it's the case. Back in FM17, I had the most reputable club in the world in Slovenia by some margin (it was ridiculous, it was difficult to actually even draw a game even in the UCL). When I bought some kids that I stored in the Reserves because I had to make room in the main squad, I was at times welcomed by messages of their NT manager calling them. And I'm not talking about U19s or U21 nonsense, I mean playing for the big boys despite having less than 10 first team apps, and most likely 0 (as in the number "zero") first team starts. And I'm not talking about some small time NT from Central Asia, I'm talking about really good European countries like France. The kids weren't even that good yet; they would eventually be, but they were not top European NT material at all, and considering how NT worked in reasonably older FMs, there was no such things as news articles about "battle for citizenship" or whatnot. The bigger country had dibs and that was the end of it, so there should be no pressure to "forcefully" cap the player in case he decides to defect.

Yet, just the fact that they've signed for the reserves of the biggest club in the world meant that they could do a job for the NT. Nah, that's not how it works. But m8, there's this nobody who always get into NT despite being invisible with his club, so that's realistic!

On 05/09/2022 at 14:24, Draakon said:

Is AI scouting really that bad?

The AI is also completely unable to create a tactic that exploits the strengths of players properly, nevermind properly scout a player profile for a given tactical style (although they should be able to as all managers have preferences in players and or/tactics, I just don't see it). It's also not able to use players that have a distinctive advantage over the competition despite having low CA. I recall that in the Netherlands, I had a striker with just 100-110 CA. Even though the average quality of Eredivisie players isn't great, that's still a pretty weak CA amount for a top division player, and even my coaching staff agreed that he wasn't good enough. Except that he also was one of the top scorers in Eredivisie. Why? 15 Pace and Acceleration. Even with otherwise garbage attributes, being that fast is enough to make a mockery of the average Eredivisie defender, and even give some Ajax players a good workout. But if you leave it to the AI to decide what the worth of that player is, he probably would struggle to even find a professional contract, nevermind play any sort of professional football at all.

The AI cannot really perceive attribute distribution separately from overall ability; or at least they cannot infer meaningful tactical and recruiting conclusions from it, nevermind build a tactic or a team around the quality of their players. The AI generally has pretty poor tactical awareness; it works well enough to make the game playable, but it's not so good that you don't notice some obvious cracks. At least it somewhat prevents the AI from going all in on players who can only work in a given system, so there's a balance to be found... even if it results in obvious poor recruitment choices at times that are more reputation driven than attribute driven.

Oh wait, actually I found the guy on my other computer. Only 95 CA, third best goalscorer in Eredivisie (11 vs 13 and 15), second highest average rating at 7.45. I even have his hidden attributes on display. He's simply faster than the average defensive Eredivisie player, so he'll keep scoring over and over again even if he's unable to do anything else correctly. I do think you wouldn't go out of your way to recruit such a player, but there most likely are some D2 clubs who could use a player that can score plenty at D1 level.

And Finishing is overrated, of course.

[Evidence Based FM's testing seems to show that Match Prep sessions increase the likelihood of losing the next match compared to no training at all. Is this a bug or something else?]

@werogatda has uploaded a new video to his channel. Spoiler alert: there's no proposed training schedule yet as he wants to run further tests.
 

 

[Will you be buying FM23 ?]

57 minutes ago, fc.cadoni said:

@harrycarrie & @Xavier Lukhas

Happy to see there is an interesting in this topic. This year with Inside Forward situation was an interesting topic as an example; reminds me the situation with Bad Company where your movement was left - right - forward - backward, but not diagonal left-forward. Prefer to see that 2-core CPU (supported) being a thing of past in FM. It's like image & audio in my world. Recording 4K will look better in 720P display from 720P recording in 720P display, but recording 720P video and being displayed in 4K will look ugly. That's what I want to see from Sports Interactive, not if they fix or add new stats in Data Hub as an example.

The way I see it, it wouldn't make dual cores unsupported: it would just offer the ability for basically every other CPU to stretch its legs. The only negative is that it's potentially very hard since FM hasn't fundamentally changed in almost two decades in the way it processes gameplay. A lot of other things have changed, but FM22 runs about the same as FM12. Technically FM22 can stretch its legs a bit on higher core count CPUs if you wall it with leagues (as evidenced in the benchmarking thread), but it's not a whole lot of distinction for not a whole lot of difference.

EDIT: Deep inside, I do think that FM should be rewritten from scratch because of many reasons that are not worth opening the can of worm for. I don't believe it's going to happen anytime soon though; I'd be happy if it could at least run decently well on modern hardware though.

[Will you be buying FM23 ?]

@fc.cadoni

If SI cared enough about my opinion to at least convince me to take a look at FM23, making FM run well enough so that a R5 3600 isn't outpaced by a i3 10100F would go a long way. I actually could somewhat overlook the stagnant nature of FM's development and progress over the years if it didn't take ages to run leagues; nevermind checking rules and conflicts in the Editor for anything somewhat complex, which takes absurd amounts of time to load, and even longer to verify. That's part of the reason that FM can feel like a second job (to me at least): it actually runs bad enough that it takes as much time as a second job if you have the gall to load 10 or 15 leagues, which barely covers the top leagues in the world even if you only play in Europe. But someone else has said it a bit more eloquently in another thread, so I'll just quote them instead.

On 13/11/2021 at 21:15, PineappleBlender said:

I've run another set of benchmarks, this time disabling 12 out of 16 cores in the BIOS i.e throwing away basically 75% of the CPU.

Benchmark results, Ryzen 5950X 4/16 cores enabled, otherwise as above:

Benchmark A - 1:02
Benchmark B - 4:47
Benchmark C - 14:14

Notes

This was achieved by naively disabling the cores in the BIOS, with no attempt to select optimum cores. With Benchmarks A & B we're within margin of error. For most practical purposes under normal usage, effectively zero parts of the game use multiple cores. In theory this should allow PBO to boost to higher core speeds because there's extra thermal overhead. For scientific accuracy more detailed testing would be needed, but it shows the general idea.

As a reminder, results using the full 16 cores i.e. ~4x more processing power were:

Benchmark A - 0:54
Benchmark B - 4:43
Benchmark C - 6:59

Conclusions

If you look at Benchmarks A & B which is how almost everyone will play the game, the result is frankly embarrassing. I remember years back asking Miles to look into this on Twitter, and he was refusing to accept the game doesn't scale with multiple cores, insisting that telemetry from AMD and Intel showed how good his team's programming was. Send him these results. Or get someone to test it themselves, because this is blatant denial.

Some people might look at this and say "it doesn't matter about optimising for people with 16 cores". And you're right, it doesn't. There's probably only a handful of FM players with this many cores. But it's badly optimised for anything modern. That basic laptop you got from PC World? I'm sure you'd like FM to be 3-4x faster if it was optimised properly. There is zero reason to think the game scales great on 4 cores, but badly past that, because the same technique that would allow it to scale for 'normal' processors would work on high end stuff.

The point is not that the game isn't designed efficiently for supercomputers, it's that it's not designed to run well on anything past 2005.

Every single aspect of this game is able to be parallelised. QME could simulate different leagues per core. Transfer targets could be done per core. Training improvements could be done per core. Even for stuff like the full match engine which is complex, you don't need to parallelise that, because you're doing that already by simulating 20 games as part of the league.

FM should be scaling near linearly with core count, like video encoding. It is probably one of the best games in the world for scaling potential, but poor coding practices and a refusal to modernise or even accept this on the part of SI means the game is incredibly suboptimal.

It isn't easy to rewrite the game to use multiple cores. No one says it is. But at least Miles and the like should honestly admit their software does not work well on modern hardware, instead of dismissing criticism even with evidence.

  

It would require a lot of work, it's not easy. But we're not talking about only getting a 5% speedup for 10% of users. We're talking about massive speedups for every single player.

 

On 14/11/2021 at 03:31, PineappleBlender said:

I can't prove what Miles said on Twitter several years ago to me on an account I've long since deleted, so I'll understand not taking some random person on the Internet at their word making such an allegation, though it was completely dismissed.

It's not easy, I completely agree. I'm sure it would be a lot of work and you'd have to rewrite probably almost the entire code base to convert it into a format suitable for multi core. It would however be a lot of work that would pay massive dividends. Every single player having a faster game, twice as fast, four times as fast, ten times faster. Players being able to run large amounts of leagues with acceptable loading times.

It's also frankly crazy that this hasn't been a priority since c. 2010 when multi core started becoming a major thing. Especially these days as single core performance is much harder to improve, and there's a trend towards more and more cores.

One thing Miles has done recently is say women's football is being implemented, not for economic factors, but because it's the right thing to do. Yet apparently no one seems to have seriously argued about the massive benefits over the past decade of using 21st century programming design to achieve massive game speedups that other companies could only dream of! If Miles thinks "our game is 10x faster" isn't a good selling point, then I'm not sure what to say. If that isn't a good reason, how about "it's the right thing to do".