forameuss

But this has always been a problem historically, and an extension of it in software in general. It's the unrealistic input/unrealistic output problem. There's a reason why every one of these custom tactics looks to be an absolute mess, and it's because the ME is - presumably - developed from a sensible standpoint to make sure that the "normal" stuff is as balanced as possible. With millions upon millions of combinations being prohibitive to test extensively, someone who makes it their mission to find out those gaps is probably going to be able to cobble together some monstrosity that the engine simply doesn't know what to do with. And not saying it should be this way, the really obvious stuff that shouldn't work genuinely shouldn't work, but making an orthodox 4-2-3-1 behave correctly is probably going to always take priority over making sure someone can't play a 2-1-2-5 and get weird results.

Except the match engine has no idea whether it's simulating AI vs AI or AI vs User, and as such can't tell who is who. So no, the game isn't broken. It's a you problem.

I've no idea how it was done, but I expect them having to find a solution fairly quickly would've had something to do with it being hard-coded. That's always going to be the easier option. Always a chance they could just carry it over and not waste time implementing something more flexible when the current way just works, but given more time, if it's something that would make the game more stable, they might have done it "properly" this time. The situations aren't really comparable at all. The Brexit thing was done because they simply had no idea what way the vote would go (nor did people who had a financial stake in it even once polls had shut). They offered random options to not pick a side either way. The pandemic approach I disagreed with, but their decision at least had some basis and was explained. Seemed largely popular too. With this, they are modelling what the current state of the world is based on what punishment FIFA have handed out. There is absolutely nothing that would give them a basis to put an end date on it, and it would be crazy to try. Would likely also open them up to certain questions they shouldn't have to be answering if they're seen to be saying to "the baddies" in this war that they're good to have all sanctions lifted.

I don't know for sure, but I wouldn't be surprised if it's true. If someone who actually has access to the code is telling you that though, they at least know 50% of the picture, significantly more than most of the people who like to throw out wild conjecture about something they know little about. So on that conjecture, I wouldn't be surprised. It's a fairly unique game, as even ones you could perhaps compare - other genre management etc - probably aren't having to model something as complex as 22 individual players trying to make up a football match. There's a reason even EA and their deep, filthy pockets abandoned their management title, and why no other developer has decided to jump in (quite a few reasons, but that's not the topic). And outside the match engine there's still the complexity of tens of thousands of individual agents having to interact with each other and support a gameworld that's supposed to feel alive and "real". They're sorely lacking on that latter part, but that's not to say that it isn't already complex, in fact the complexity is likely what makes further advancement hard. Long story short, from experience on software that's nowhere near as complex, I expect the real picture with regards to bugs is probably far worse than any individual user will ever see. But that's just software. Could SI do better? Undoubtedly, and I expect many - if not all - within SI would agree. But throwing out stuff like some do (not directed at you) questioning whether they can be arsed or not must be absolutely maddening to people working really hard to deliver what is ultimately not looking like much progression.

You're surprised that a YouTuber who can financially benefit from more eyes on their content might care more about attracting said eyes to the content? They probably do care about the game, but only because if the interest in it dries up, so does a source of income. XaW is absolutely correct. It doesn't matter if someone knows the game inside and out, or not at all, their opinion will still not matter anywhere near as much as your own.

It's not that it doesn't take any of that into account, it's that it's incredibly difficult to model the intricacies and vague feelings that turn the sort of nation that is completely set up to be that successful, but just...doesn't. England have all the money, they have the facilities, they have the history in the game and the means of securing talent that should make them that good. But when you ask 1000 people why England lose, and you get near enough 1000 answers back, it's clear that it's going to be pretty difficult to model realistically. Because there's no way some Nations League League B diddy should be that successful right?

He's right though. Instant result is taking away the control you have over the match itself and putting it into the hands of the AI. The AI, by your own admission, gets better results. It's not exactly a hard parallel to draw.

Your biggest problem might end up being European nations to be honest. With the way the schedule is, if you don't draw them in competitive matches, they're probably not going to be regularly available for pre-tournament friendlies either. I think there's still a few nations we haven't met in ages, and there might even be some we've never met.

I've never seen the point in this really. 50 years, 100 years, 1000 years...the only reason FM's simulation remains interesting is with you inserted into it and having a stake. I've always struggled to see the interest in how, say, Yeovil Town have won 100 Premier Leagues in a row in 2672. Anyway, the tech stuff will be the bottleneck. If you're wanting to do something that simulates a lot of leagues, it's going to perform pretty badly, and it will take you an incredible amount of time to actually get far into the future.

I like to think of it as a million very small and narrow boxes, each one with an interaction or situation in it. Ideally, you're not having all these individual boxes, you just have one box, with every decision inside it taking into account other decisions to be able to better choose for themselves. But as it is, you enter the box and you try to "solve" the problem it poses. And furthermore, said problem is completely abstract and often nonsensical. And once you do make a decision either way and you leave the box, nothing else really comes with you.

This is a decent point actually. I would take issue with that particular wording. It's not falling behind, as it's probably still the best specific example in the genre (or the only one, they killed the rest). But being the prettiest turd only gets you so far. What we currently have is massively lacking - plenty of bits are going to take real steps forward to fix and are very difficult to do, but there's a few places where it likely wouldn't take a huge leap to design things a bit better to disguise the really difficult stuff. That specifically, yeah, but the "easy" fix to that isn't really a true fix, it's just adding another fixed rule for your agent to follow. Now obviously to us that still looks better, but it's not true AI, it's not true context. And that's where the big strides need to be made. I don't want a player just to be told that you need to wait for x date to evaluate, rather than y. I want the player to be able to make an educated decision on y date, with the ability to know what the state was going to be on x. If that makes sense.

Really? You really believe there isn't reason to complain about an AI that lacks any sense of context, or even the ability to make sense a lot of the time? About literally the biggest, most glaring flaw the game has, and has had for a number of years?

And it'll likely still sell a very similar number to always. A 40% review drop likely wouldn't change that, let alone a 4%.

If the need to save 20% overrides the need to see whether you're entirely wasting your money or not, then that's a you problem and you kind of forfeit the right to complain. I see where you're coming from, but the tools are there to get a good idea of whether the product is worth your money. In that case, you're not part of what I'm saying. It's those that don't use it to supplement, but as the entire basis. I get the subtleties part, but at the same time it can often lead the thinking, much like visiting here will. There will be people who don't watch YouTubers, don't visit here, and who happily play the game blissfully unaware of any problems in the game. It's like the slightly squint bannister that you never noticed, but now can't not notice. Would it have been better just to trust your own eyes and enjoy it, or hear someone else's words and ruin it?

Well if you keep buying a product that apparently hasn't changed in years, probably as long as you do that. Given there's a completely free demo coming in a couple of weeks, why not wait for the most insider knowledge of all - your own? You're hardly alone in that thinking, but it seems more and more the case now that people are completely abandoning forming their own opinions and taking other people's. You might try and demo and think it's an absolute turd not worth your time, or you might find it scratches the itch perfectly. I've never understood why what someone else thinks - regardless of if they play FM for a job or are just a stranger on a forum - has such a bearing on what an individual thinks.

The issue I could perhaps see with that is that the more options you have for ticking and unticking things, the more different possibilities you have. The more possibilities you have, the more complex and difficult testing could be. Theoretically, of course, but logically it follows.

Here's a crazy idea...maybe don't?

I'd be more critical of the graphics if the game behind it was a lot closer to perfect. There's so many more areas severely lacking to go after something which is just...fine. It represents the match engine well enough. But outside of matches the game still has pretty considerable problems that need a lot of time and care. The match visuals just don't. They need work, absolutely, but buying a more sparkly glitter probably isn't going to make much difference if you're pouring it over a puddle of cold sick. At least mould that cold sick into something better before you do.

Well at least this stuff is completely objective and no-one's allowed to have any kind of alternate opinion I suppose. Saves a lot of bother.

The word "basically" doing a whole lot of work there. Have there been enough changes to perhaps warrant a new version? Not always. Have there been enough to make a noticeable different to how the game feels? Yes. Am I really going to have to add that to a 4th post after this one?

Yes, because like I said, all the minor features add up to a different experience to anyone that looks beyond major feature announcements and actually waits until the product is actually here. You're also assuming that this hypothetical product that arrives in 3 years time won't just be roughly the same product you would have gotten with the current model. 3 years of features vs 1 year of features is going to probably need 3 times the testing effort, so the difference would likely be negligible. It's not like they sit down in November with a blank sheet of paper and say "right, what about the next version?" We already know they've got women's football arriving in a future version, and they've said in the past that they're planning multiple versions in advance. The only difference this change would likely make would be a superficial one to stop people making the same complaints every Autumn, and move to them making them every third Autumn.

How it's delivered doesn't really make a difference to me, whether it's a boxed product, DLC, whatever. I'll likely buy FM23, but I wouldn't purchase any product that was solely labelled as a database update. And obviously that's just personally me, but I'd be surprised if I'd be alone in that, and if that's true, then that would be a problem for SI.

I wouldn't pay for something sold as a pure database update. Not much changes from edition to edition, but there's still a noticeable difference as all the little features add together into the overall experience. I'd have no interest in paying for the same game as the previous year, with just the addition of something I could get for free - albeit in much lower quality - elsewhere. And a subscription model can get itself in the sea, as I can't even muster the effort to put it there myself.

Yeah, the hype really, really didn't help. I was one that got it at launch and it was...fine. A puddle, rather than an ocean. Endlessly wide but incredibly shallow. A result of letting a developer - albeit the senior one - become the marketing guy when he really, REALLY wasn't suited to be that. Looking back, it's just a guy who was absolutely passionate about what he was doing getting carried away, but he never should have been in that position in the first place. Particularly when people are fickle, and when they get an idea in their mind that didn't end up coming to fruition, they went a bit scorched earth. But everything from then, they've been great. They got their heads down, they worked themselves ragged, and they tried to make good on what they initially talked about. I just don't think using them as a yardstick works given the completely unique case. Not many developers could get away with following their example

It always amuses me when Rockstar are held up as this shining example in the industry, despite probably being as predatory with microtransactions (if not more so) than the likes of standard whipping boys EA. If we're talking about "need" then no-one at Rockstar ever needs to do anything ever again given how much they make through shark cards. While I see where you're coming from, I'm not sure NMS is a good comparison to make to anything else really. They're a freak case in the way they've dealt with the product, and even though they've arguably had one of the most amazing resurrections in recent times, it still bears repeating that the only reason they were in the position to resurrect was because they screwed up so incredibly badly in the first place. Having said that, it is interesting to wonder how Hello Games are staying afloat with their strategy. Can only assume they made an outrageous profit initially (and I think they had a very lucrative deal with Sony from the outset). Ultimately, this is true. For all the complaints that come up this time of year, every single year, how many times does that translate to that person actually not buying the product? If it did, in large enough numbers, then we'd probably see a change in approach. But...well, what do people expect them to do if sales stay high? What better validation for their plans is getting money in their pocket? And also, worth noting that just because the natives are restless here, or Reddit, or YouTube or whatever, doesn't mean the majority aren't still happy. I imagine there's plenty of people who will pick up the game on day 1, play it to death, end up with thousands of hours, and then do it all again next year. All the while being delighted doing so. Ultimately they're no more or less wrong than someone who dislikes the game.