Jump to content

V3ntricity

Members+
  • Posts

    73
  • Joined

  • Last visited

Posts posted by V3ntricity

  1. No, that is not the issue! I should have been more clear: Martinelli and Ugarte went away to a u23 youth championship in january. They are first team starters for me. I can understand players going away for the africa championship, as that is national teams on the highest level. But youth tournaments should not take away players in the first team without the option to refuse.

  2. I'm in the middle of my second season at Arsenal and in january, several players are unavailable for me due to them going away on some youth tournament. We are talking established first team players like Martinelli and my new multimillion signing, Ugarte. This is in the middle of a packed match program. In fact, i'm having difficulites mustering a lineup not consisting of reserve players.

    Is this normal in the real world? I can't recall this happened in january. In game i'm powerless to stop it.

  3. why has Van Dijk (*spelling) better technical attributes than world class midfielders? He even has high dribbling stats, but he hardly ever dribbles. And Salah just isn't that fast anymore (even though he is still good).

    In response to those complaining that defence don't pick up on runs, I think this is realistic if you play a high line, high pressing game. It should then be difficult to track runners

  4. 38 minutes ago, themadsheep2001 said:

    In terms of giving useful advice in regards to FM, it's crap. The advice its been giving as examples was both woolly and incorrect, and that's what people were talking about in those examples before you came in halfway through a thread, and fired off without reading what people were responding too. Again please read the context of what people are saying, otherwise its a bit of waste of time. 

    it's not crap because you are using it wrong. That is what my post was about. This whole thread is about how it can contribute to FM.  Also, if your understanding and use of ChatGPT is through the old prompt, you don't really know its capabilities.  Asking it randomly to give you advice on tactics is obviously not very useful

  5. 2 hours ago, enigmatic said:

    Maps and Wikipedia excerpts could be one for the feature request section, nothing really to do with AI (although ChatGPT can make up a "sporting culture" for a lower division club that doesn't really have a well documented one and sanitise some of the famous ones by not mentioning the ultras...)

    It is just an example of the use of AI tools, and thus have alot to do with the discussion at hand. Meaning even static information like plain info can be made interesting and tailor made with the help of AI

  6. On 20/01/2023 at 14:01, themadsheep2001 said:

    They aren't arguing against AI, but this particular AI, which is currently crap, especially as a help tool. That's been the core of the discussion. AI in general is different discussion. If you're going to have a pop, at least have the courtesy to read the detail and context they have posted

    Crap? This is objetively wrong. It depends on what you use it for and how (also, ChatGPT comes in more than one form). The limited understanding of these tools at display, combined with the negative attitude to explore possibilities, is quite disappointing (including your snide and pointless remark).  

  7. 22 hours ago, Platinum said:

    Mostly agree which is why I dont think its worth the effort (I dont think buggy is a given). I think simple things like news reports and can be rewritten using LLMs though to get rid of the repetitiveness. Would have to do some testing to see how often the rewriting loses the original meaning though. You could do the same for player interactions  and press conferences however I think we actually want the interactions to be repetitive in this case and instead want our range of responses to be adequate - which is completely separate to any type of ChatGPYT technology.

    Ultimately though I would guess that the cost likely wouldnt make sense. If it uses an API from OpenAI for example  it would be expensive and require the internet or require the user to run it locally on their machine which may or may not alter the system requirements in a significant way.

    Regardless I hope they are putting A LOT of resources in to improving interactions and immersion.

    It's unambitious to think like that for a big game like FM. The costs could be too high, but that will be remedied by implementing their own AI services over time, which would be both simpler and more specific to FM.

    AI could be used plenty of places in FM that not only would make the game less repetative, but also considerably more informative. How cool would it be to use a simple text service through AI to show tidbits of information on the city your team is playing against? I would love to have a little map and a little text box when playing against some lower league team or international team, where it shows the city on a map and some information about it. 

    That is not in any way complicated to implement, it certainly shouldn't be. The benefit would be huge for a game like FM. It's the main thing I miss in game, and it's more of a closeness/discoverability to the environment and culture.  A tool like ChatGpt would be perfect for this. If the costs would be to high, then there are ways to go about it.

    1) set up a server that is basically a proxy between FM and ChatGPT. It would be like a Redis server, but for information to be used in FM.  So in game FM would enqure for info on Torquay (for example) to the proxy. If it is already saved in the db, it would return it, else it reaches out to ChatGPT, inserts this info in the db, and returns. It's cheap and effective. From time to time,  this database could be updated for more fresh info if desired.

    For information that needs to be more dynamic, then you could directly use ChatGPT, or preferrably FMs own AI tools (it would be worth building in the long run). But doing it this way, alot of the operations would be stuf that don't need to be immediately dynamic, so the costs will be substantially reduced.

    2) Not use AI services, and instead just download wikipedia to this server. You could even add a feature where FM users could update information themselves. It's amazing how people like to contribute to things they care about. it gives them a sense of ownership and pride. This is a huge opportunity for FM, like many other things I've listed. 

    It's not about these specific examples, but these features are just something I would want. There are many use cases (generating pictures and backgrounds for example). 

    I sometimes believe the developers/leadership/owners of FM to lack vision and ambition. It's a really great game, but they really neeed to inject some vitality into their game or they could quickly grow stale.  A FM that includes women's football is not really that innovative (it's more of the same). Using the opportunities of internet services would make the game considerably more dynamic and adaptable. 

    It really isn't a complex thing to do for most developers (I suspect it's more about having a culture that is afraid of change, which most companies grow into). An indicator of this is people who obsess over details that are wrong and then come to the conclusion that ChatGPT is terrible or not useable. You simple don't see the forest for the trees.  

  8. I can't believe there are people in here that argues against this. Either you don't understand AI or you are nitpicking AI answers for faults. 

    But it's not about using ChatGPT as it is, that would be terrible. It's about using it (or tools like it) with an added context that is FM, training it and developing models for use in game. This could greatly enhance immersion and solve one of the major problems of FM, which is the tedious repetition of text and processes.

    This is not complicated to do anymore, especially on a smaller scale. Just making the same game reviews and conversations in game a little bit more interesting would help alot.

    Look, I get that it is not necessarily easy to hook up a service like this in decades old c++ code just to inject some text.  Game developers tend not to be the best at writing maintainable code. But in theory this is easy, and in reality this is a challenge and opportunity you probably should take. 

  9. 22 hours ago, Welshace said:

    Son has been incredibly underrated? in game there are only a handful of players with a higher CA than him...  

    Harry kane has something like 194 goals in less than 300 matches in the premier league, I can't see how anyone could call him overrated in any sense.. slow and lumbering? is Lewandowski overrated because he's slow and lumbering?

    I think the obvious issue here isn't that English players are overrated in the game... it's that your bias against them is clear.

     

    I'm not saying Son is underrated in FM23. I'm saying he has been incredibly underrated for years in FM untill now, while players like Kane has been world class for years in game.

    I'm not saying Kane and Bellingham and many other english players aren't good. I question the realism in that so many english players are world class (in my opinion Lewandowski has much quicker reaction time than Kane, which is why he is actually world class).

     

  10. 1 hour ago, Welshace said:

    Lets look at this objectively... lets take Bellingham... can you honestly name many better performing young midfielders in the world right now? or Kane.. how many strikers are better than him in the world at the moment and howare they reflected in game?

    I have seen plenty of both Bellingham and Kane. I just don't see it with Bellingham, where are his insane numbers? Kane is slow and lumbering. He is technically gifted and tall, which makes him a perfect premier league player. He scores consistently because he is a focal point for a good football team, where he takes all the penalties. I equate him with other reliable goal-machines for even lesser teams in the past like Kevin Philips, Vardy and Defoe.

    They are alll very good players, I just think it's unrealistic that there are so many world class english players in the game (when in reality England isn't that good).

    Is it a fame issue? Meaning, english players gets more press within the english media, so people naturally think they must be the only ones out there. You dare me to name better young midfielders than Bellingham, but I suspect you actually don't know of any because all you read is english media. Gavi won recently some best young player award, to the surprise of Arsenal fans who meant Saka should have placed higher than 7th. 7th is good.

    Take Son, who just recently has been upgraded to reflect his ability compared to Kane. In reality his numbers are just as good, and that is without the penalties. For the last 4-5 seasons Son has been incredibly underrated comparatively to the likes of Kane.  I mean, Dele Ali was world class for many seasons in FM, and still is kinda good.  

  11. Alot of english players are always overrated in FM. I can understand the bias, being an english game, and I think it is difficult for them to see the outside perspective. Reece James, Rice, Sterling, Arnold (most Liverpool players too), Saka, Kane, Bellingham, Mount, Sancho, etc - all these are some of the very best players in the game. You would think England are by far the best national team in the world.   

  12. What would be the ideal winger that will run down the line and put crosses in? What would be the ideal stats and traits for this? Any examples of players in game, because I have trouble finding them?

    Good crossing stat is a given. But what else in descending order? I'm thinking vision, technique, teamwork(so he isn't selfish), anticipation (i never really know what this stat does)? dribbling, flair and agility if they want to get past

  13. I see many choose to spend big right away. I tried to keep it a bit simple. I lucked myself to the last europe league spot in the very last match of the season.

    My best buys were Jonas Wind (12M) and Bruno Petkovic (7,25M), who both shared the DLF(attack) role alongside Wilson with a Advanced Forward. They worked so well in this role, scoring alot of headers and providing assists. 

    Gianluca Mancini (19M) was another good piece of business. Livakovic everyone knows about and is a must. Then I needed a playmaker and bought Aouar for too much money, thus ruining my modest strategy (but he was good and injury free all season).

    Start of second season I've bought Sesko (17M) and Tomas Soucek (34M on transfer list) who i'm very exited about to take us to the next level. Waldemar Anton (4,9M) seems like a steal too.

  14. my post is more a theory based on what other have written and my own understanding of the game engine (i am in fact a professional developer in C#, C++ like the game is written in). I could have stated it differently. I know attributes go from 1.0 o 20. But in terms of actual appliance, there's hardly any difference from an a player with dribbling 13 or 14. But there is proportionally more between 14 and 15 (the cutoff can be different than this). Something like that, more or less, but i'm pretty sur there's a range. 

    Furthermore there's alot of strance choices, like an ability cap, so that means low rated teams can't have any players that are as good as the top players in a higher rated club (but they can have high potential players). So the game operates with alot of set caps, which is unfortunate if you want to have a truly dynamic game.

  15. i don´t think spending that kind of money on one player is wise when you are a relative small club. Felixs wages are also not worth it. Many players improve after 22, but it depends how much the´ve been developed/played and if the have a professional personality (positive). Players who lack professionalism don´t develop well

  16. No one really knows how attributes work. But I know a secret alot of people do not know about (i´m pretty sure an official person said this); the stats are basically 1-10. So it´s something like 19-20 = 10, 17-18 = 9, etc.  So the difference between 16 and 17 is bigger than 15 and 16. 

×
×
  • Create New...