squirrel_plays_fm

I find doing this works well and can unlock all sorts of value by moving players to positions they fit better. Takes about half a season of playing time for them to adapt, with timing varying somewhat based on their hidden adaptability rating. Most frequent position switch I find myself doing is midfielder to wing-back and vice versa

Ha thanks mate, good to hear

That's a really good point I had forgotten - to focus on the clubs that are local to you. Particularly for u21 releases from clubs far higher up the pyramid

Get expensive (= high wage) players off your wage bill as soon as you can. Keep your squad small so you can be on the lookout for needle-mover signings all the time in-season as per the point below Free transfers and loans only. Free transfers: Have a look on 1 July after all the contracts end on 30 June...but recognise the real needle-moving players won't be willing to sign with you until their other options are exhausted so you will likely sign them at various points during the season. Transfer window doesn't apply at that level. Find players by going to World > Nations > England > Transfers > Released players. Move all of those into a shortlist (you can do 500 at a time) and offer trials / look at the attributes yourself. Loans: Try to do on deadline day as more likely to be free. Find players by putting all the players (senior + youth) from each of the clubs in the two divisions above your division into a shortlist and using the 'slightly interested' filter in the player search screen Simple tactics *Big* emphasis on physicals (esp pace and accel) - see this from Ian Graham when he went on the Freakonomics podcast - I think FM tried to simulate this:

Agree entirely. Was thinking more about the potential rating i.e. when you're looking at a 17 year old and trying to figure out what they will be like at age 24

Yes on one level it's a bit solvable and therefore 'easy' but I find taking on the harder challenges in the game is still plenty of fun. Taking on a club with massive debts is a favourite of mine. This year I've done a pentagon challenge, and to do that 'properly' and get around the five continental challenges in the shortest amount of time is great fun and hasn't been easy for me

Hey guys Just wondering what experienced FM pros make of a player like this guy - Viligiardi (picture hopefully pasted below). For a player to be this good at age 20 you normally see them go on to be a champions league level player i.e. seriously good. In my save I'm a champions league winning club so 3 stars is a solid prem-level player, but I'm surprised to see the potential star rating this low. Generally I'm spending more time lately tracking star ratings and find that a lot of 5-star 15 year olds are 3.5 stars at age 18 for example...with not much in the player's development to help explain why. Generally how much do you guys trust the star ratings, especially around PA? Cheers

Thanks man, appreciate that. Extending it to all roles is a big job - I did post a thread on twitter FWIW on how one might do a simplified version of it if you want to adapt it for your own save

Hey man...very cool to see you using the scoring system tool on here

I wonder if that's because the thread has been so successful this year in figuring out what the best hardware is for FM at the moment. I can't find that written out clearly in one place anywhere but FWIW I read the benchmarking data as follows: - if you want to play FM on a desktop the latest Intel and AMD CPUs are both very fast and about as fast as each other. 12th series and recent (using Intel's naming system) on desktop are faster than Apple silicon - if you want to play FM on a laptop Apple silicon is best by a clear distance, with M2 better than M1 and the various Max and Pro variants perhaps making a small difference over the base model CPUs but not a huge one. PC laptop CPUs are a long way behind at present, particularly when run on battery i.e. not plugged into the wall I'll hazard a third thought - but I'm much less convinced this is correct than the first two - because Apple silicon is as good as it is at present, the gap between laptop and desktop peformance for FM is smaller than it has been in the past i.e. the benefits of the full-on desktop rig are smaller than they were before FM was optimised to run on M1 / M2 chips. I think this is probably true but can't say that for certain. For me personally the need to build a desktop rig to replace my 2019 PC has gone and FM is a laptop game for me now, but YMMV on that one Is all this in line with the consensus view? I think this thread does a very good job at setting out what's going on and that's how it looked to me when I looked at it anyway. (I should add in case this post annoys anyone - I'm not a hardware fanboy in any direction. Just want to take a data-driven approach to ensuring I have the best computer to do the things I want to do, hence why I'm here)

And now after the new computer is here. Base spec M2 Pro Macbook Pro, 10 core 16 / 512 Type: Laptop Model: Macbook Pro 14, on battery CPU Model: M2 Pro (10 core) CPU Base Frequency: 2.4 GHz CPU Turbo Frequency: 3.6 GHz RAM: 16GB RAM Clockspeed: 6400Mhz GPU: M2 16 core GPU Apple Graphics Level in 3D: High Benchmark 1: 37 Sec Benchmark 2: 01 min 41 Sec Benchmark 3: 05 min 33 Sec Benchmark 4: 18 min 19 Sec

Yep, can confirm FM play on (1) a 9900k 32gb desktop - pretty much the top of the range self-build PC possible in 2019 is noticeably slower than (2) a M1 mac mini. A pretty wild result. Here's a 'real world' FM23 example...I holidayed through the January transfer window (28 Dec to 25 January) on my save which is in the Greek first division about 12 seasons in with about 10 leagues loaded and around 100,000 players in database. Detail turned down where possible. Same starting point on each computer. M1 Mac Mini took 8 mins 40. 9900k desktop took 11 mins 22. T480s laptop took 19 mins 11 plugged in. I didn't bother doing it unplugged on the T480s as I expect based on other tests I had done it would have been >25 minutes.

Hi guys, do you think a M2 Macbook Air or an M1 Macbook Pro would be a better machine for playing FM? Posting as they now cost about the same following the M2 Pro/Max Macbook Pros being released...and the benchmarks indicate both would be very good FM machines. For those interested I dug out the detail on the benchmarks etc and posted it to the benchmarks thread here:

In case of interest to others, I have been looking at the correlations between the tests that have run here and generic CPU benchmarks. Correlations between Geekbench and the test scores here for all posted benchmarks )from the M1 apple silicon and faster) were as follows: Test 1 - 0.851 to Geekbench single-thread, 0.602 to Geekbench multi-thread Test 2 - 0.792 to Geekbench single-thread, 0.888 to Geekbench multi-thread Test 3 - 0.798 to Geekbench single-thread, 0.886 to Geekbench multi-thread Test 4 - 0.930 to Geekbench single-thread, 0.936 to Geekbench multi-thread I looked at Passmark also but Geekbench has higher correlations. At first glance this runs against the idea that FM puts heavy single-thread loads on the CPU without much parallelisation. Having said that I ran some real-world benchmarks across my three computers using my current save (10 leagues loaded, detail turned down) and test 1 was a better predictor of real-world performance than tests 2, 3 or 4. Flagging this here because I can't figure out whether I think a M1 Pro Macbook Pro or a M2 Macbook Air would be faster for playing FM in the real world. They cost about the same now that the new M2 Pro Macbook Pros have come out (January 2023) - and I think for FM performance they might be about the same - although before I had looked at any numbers I would have guessed the M1 Pro was faster. Now I have looked at the numbers I think the 10 core model might be, I'm less sure about the 8 core model. In case of interest, Geekbench benchmarks for macbooks and various other computers are as follows (can you tell I can't figure out which one I want to buy yet! LOL) M1 Macbook Air - 1,707 single, 7,427 multi M2 Macbook Air - 1,893 single, 8,739 multi M1 Pro (8 core) Macbook Pro 14 - 1,731 single, 9,543 multi M1 Pro (10 core) Macbook Pro 14 - 1,737 single, 12,039 multi M2 Pro (10 core) Macbook Pro 14 - 1,958 single, 12,077 multi M2 Pro (12 core) Macbook Pro 14 - 1,964 single, 15,116 multi i9-13900K (fastest CPU available today) - 2,237 single, 25,374 multi i9-9900k (CPU on my desktop built in 2019, fastest available then, how times change) - 1,304 single, 8,472 multi

Three benchmarks from me...I'm having a look at all my computers and deciding whether now is the time to upgrade. Thanks for this thread, it's great. It looks like a M1 or M2 laptop is in my future. -- Type: Desktop Model: Mac Mini CPU Model: M1 CPU Base Frequency: 3.20 GHz CPU Turbo Frequency: 3.20 GHz RAM: 16GB RAM Clockspeed: 4266Mhz GPU: M1 7 Core Apple Graphics Level in 3D: High Benchmark 1: 43 Sec Benchmark 2: 02 min 22 Sec Benchmark 3: 06 min 28 Sec Benchmark 4: 25 min 47 Sec -- Type: Desktop Model: Self-build CPU Model: i9-9900k CPU Base Frequency: 3.60 GHz CPU Turbo Frequency: 5.00 GHz RAM: 32GB RAM Clockspeed: 3200Mhz GPU: 1080 Graphics Level in 3D: High Benchmark 1: 59 Sec Benchmark 2: 02 min 17 Sec Benchmark 3: xx min xx Sec Benchmark 4: xx min xx Sec -- Type: Laptop Model: Lenovo Thinkpad T480s CPU Model: i7-8550u CPU Base Frequency: 1.80 GHz CPU Turbo Frequency: 4.00 GHz RAM: 16GB RAM Clockspeed: 2400Mhz GPU: none Graphics Level in 3D: [not sure] Benchmark 1: 1 min 57 plugged in, 2 min 28 on battery Benchmark 2: 07 min 44 Sec plugged in, 07 min 48 on battery ...I'm afraid this laptop was pretty shot-out thermally by the time I got round to running benchmark 2...looks like it's time to upgrade away from it for FM and simply use it for surfing the web / doing Excel from now on