FM20 Performance Benchmarking Thread

[Brother Ben]

Hello all and welcome to the FM20 Benchmarking Thread

Carrying on the sterling work of @SebastianRO and @Powermonger in previous years, I've decided to make it a little bit different this year and hopefully open up at least one of the benchmarks to more people who, like myself, may have less than stellar equipment

With that in mind I've made the first benchmark a lot less demanding but Benchmarks B & C will be familiar to those who have done these benchmarks in previous years

----------------------------------------------------------

The first benchmark, new this year, has 20 leagues in 10 nations loaded using a medium database and has a total player count of  51,330 players,

This seems like a popular type of setup for many and I've added it due to the feedback from this thread

Spoiler

Leagues loaded are

England - 2 divisions

Italy - 2 Divisions

Spain - 2 divisions

France - 2 Divisions

Holland - 2 Divisions

Germany - 2 Divisions

Portugal - 2 Divisions

Belgium - 2 Divisions

Brazil - 2 Divisions

Argentina - 2 Divisions

Just click on the hyperlink below to download the file, Save files to \Documents\Sports Interactive\Football Manager 2020\games

Benchmark A: Game starts on 20.08.19 and you need to go on holiday until 27.08.19 - Use a stopwatch to measure the time.

For the following 2 benchmarks, directly mirroring last year, I have loaded all 51 nations (117 leagues), using a large database giving a total player count of 162,590 which is higher than last years 141,670. 

Just click on the hyperlinks below to download the 2 Files.  Save files to \Documents\Sports Interactive\Football Manager 2020\games

Benchmark B: Game starts on 27.08.19 and you need to go on holiday until 07.09.19 - Use a stopwatch to measure the time.

Benchmark C: Game starts on 27.08.19 and you need to go on holiday until 02.09.19 - Use a stopwatch to measure the time. - Benchmark C will take longer because it has Full Detail activated for England, France, Germany, Italy and Spain.

---------------------------------------------------

In order to get the most accurate / consistent result possible, I advise you to restart your PC / Laptop and then close everything that is unnecessary, leaving only FM opened. If you run the game on a laptop, don't forget to plug it in the socket before you begin.

After you load the save, ensure that auto save interval is disabled, and also set processing to Fast (Less Responsive), both can be found by using the search bar in Preferences.

Below is the format in which I advise you to post your results, it's pretty much the same as previous years, don't worry if you aren't sure of the Base/Turbo frequencies of your processor as I can find this out for you as long as you know the CPU Model.

To find out the specs of your PC open the spoiler below

Spoiler

You can download this program HERE

once you've installed it the CPU tab will give you the type of CPU, it can be found under specification.  don't worry about CPU speed I can work that out from the model number

click on the "memory" tab and you will be presented with this

The type and size are top left

The speed is DRAM frequency.  You need to double this figure. (DDR = Double Data Rate)

So the above is 534 x 2 = 1068MHz 

-------------------------------------------------------

Below are my results for you to follow as a template:

Type: Laptop

Model: HP ENVY 17

CPU Model: i7 - 4710MQ

CPU Base Frequency: 2.5GHz

CPU Turbo Frequency: 3.50 GHz

RAM: 16GB

RAM Clockspeed: 1600Mhz

GPU: NVIDIA GeForce 840M - 2GB

Storage Type: SSD

 

Benchmark A: 1 min 42 Sec

Benchmark B: 12 min 42 Sec

Benchmark C: 14 min 02 Sec

-------------------------------------------------------------------

Any queries let me know, and if any of the more graphically inclined among you want to present the data in a more appealing visual nature feel free, for the moment I'll be keeping a spreadsheet (in a similar format that powermonger did for the FM18 thread) in the second post.

It'll be interesting to see how the results from FM19 compare to FM20 as the Benchmarks B and C are as identical as I could make them, even down to the club being managed

Here is the link for the 2019 Benchmarking thread by @SebastianRO where you can see last years times and he also gives some great tips in the first post for people wanting to get the best out of their computers / buying a new computer for FM

 

Happy Testing!

 

Edited June 10, 2020 by Brother Ben

[Brother Ben]

Results

Benchmarks at a glance

• Benchmark A - 20 leagues/10 nations/medium database/51,330 players - 7 day test
• Benchmark B - 117 leagues/51 nations/large database/162,590 players - 11 day test
• Benchmark C - 117 leagues/51 nations/large database/162,590 players/5 nations Full Detail - 6 day test 

 

Combined Benchmark A results

 

Combined Benchmark B results

 

Combined Benchmark C results

 

DOWNLOAD SPREADSHEET HERE

 

 

 

 

Edited November 19, 2020 by Brother Ben

[EdL]

My personal machine - if I have time at work I might try a few different machines we have

Type: Desktop

CPU Model:Ryzen 7 2700X

CPU Base Frequency: 3.7GHz

CPU Turbo Frequency: 4 GHz

RAM: 16GB

RAM Clockspeed: 2933Mhz

GPU: NVIDIA GeForce GTX 1060 6GB

Storage Type: SSD - Samsung 970 EVO

OS: Windows 10 1909 X64 (additional info but could be good to know as it can have an affect)

Benchmark A: 1 min 18 Sec

Benchmark B: 8 min 08 Sec

Benchmark C: 7 min 04 Sec

So compared to my FM19 times it looks like B where we have lost some performance, my C result is virtually identical which I'd take as a win given there are 20k more players in the save than last year. I will try and investigation the comparitive perfomance between 19 & 20 for B

[jonpt]

Type: Desktop

CPU Model:Ryzen 9 3900x

CPU Base Frequency: 3.8GHz

CPU Turbo Frequency: 4.6GHz

RAM: 32gb (4x8)

RAM Clockspeed: 3733MHz

GPU: NVIDIA GeForce RTX 2080s 8gb

Storage Type: NVME Corsair Force MP510  960gb

OS: Windows 10 1909 X64 

Benchmark A: 1 min 2s

Benchmark B: 6 min 34s

Benchmark C: 4 min 58s

 

[Brother Ben]

Type: Tablet

Model: Huawei Matebook E 2017

CPU Model: i5 - 7Y54

CPU Base Frequency: 1.20 GHz

CPU Turbo Frequency: 3.20 GHz

RAM: 4GB

RAM Clockspeed: 1600Mhz

GPU: Intel HD Graphics 615

Storage Type: SSD

Benchmark A: 2 min 22 Sec

Benchmark B: 20 min 13 Sec

Benchmark C: 39 min 52 Sec

Well that's an hour i'll never get back 

My poor old passively cooled tablet throttled down to base speed pretty quickly and despite that did remarkably well in Benchmark A all things considered.  The other 2 tests on the other hand created enough heat that I was able to turn off my central heating.  That said I do find my desktop PC a lot less convenient for my daily commute  

Edited March 6, 2020 by Brother Ben

[jwchriste]

Type: custom desktop

Model: n/a

CPU Model: i7 8700k

CPU Base Frequency: 3.7GHz

CPU Turbo Frequency: 5.2GHz (OC)

RAM: 32GB

RAM Clockspeed: 4000Mhz

GPU: NVIDIA GeForce GTX 1080 Ti

Storage Type: SSD - Samsung 850 EVO (1TB)

OS: Win 10 Pro 1903 x64

Benchmark A: 51 sec

Benchmark B: 5 min 46 sec

Benchmark C: 5 min 32 sec

Oh no!  I was slower in Benchmark C than @jonpt!  Must be time to build a new PC!

Kidding.  Looks like C is more thread dependent.  It would be interesting to see how that scales.  Anyone have a 32 or 64-core Threadripper?

[Brother Ben]

It’s nice to see more cores and threads being used when the detail is ramped up, the only issue is I think I’d rather use my processing power on more active leagues than compromise on less but with full detail 

what do others think?

[jonpt]

5 hours ago, jwchriste said:

Type: custom desktop

Model: n/a

CPU Model: i7 8700k

CPU Base Frequency: 3.7GHz

CPU Turbo Frequency: 5.2GHz (OC)

RAM: 32GB

RAM Clockspeed: 4000Mhz

GPU: NVIDIA GeForce GTX 1080 Ti

Storage Type: SSD - Samsung 850 EVO (1TB)

OS: Win 10 Pro 1903 x64

Benchmark A: 51 sec

Benchmark B: 5 min 46 sec

Benchmark C: 5 min 32 sec

Oh no!  I was slower in Benchmark C than @jonpt!  Must be time to build a new PC!

Kidding.  Looks like C is more thread dependent.  It would be interesting to see how that scales.  Anyone have a 32 or 64-core Threadripper?

Ha ha, I’m thinking the same about benchmark A and B

it would be interesting to see if the extra cores would add anything. Mine wasn’t getting close to using all of the cores. I have an AIO cooler and the fans were in zero rpm mode for quite a while so it certainly wasn’t taxing the CPU that much. 

Edited March 6, 2020 by jonpt

[jonpt]

47 minutes ago, Brother Ben said:

It’s nice to see more cores and threads being used when the detail is ramped up, the only issue is I think I’d rather use my processing power on more active leagues than compromise on less but with full detail 

what do others think?

I have found no noticeable speed difference by adding full detail to the 5 major European leagues and both European competitions on my save. I’m sure it is slower but it doesn’t feel it. Personally I like the added detail without loading the extra leagues. I currently have 20 nations and 30 leagues loaded. Hopefully one day FM will use more cores even more effectively so we don’t have to choose. 

[JamieTC13]

Type: Desktop (Custom Made)
  

CPU Model: i5 8600k

CPU Base Frequency : 3.6GHz

CPU Turbo Frequency: 5GHz Overclocked (all core)

RAM: 16 GB DDR4 Corsair vengeance 

RAM Clockspeed: 3200MHz 

GPU: RTX 2060 Zotac

Storage Type: PCIe M.2 SSD Samsung evo 970 500gb

OS: Windows 10 Pro 1909 x64
 

Benchmark A - 51 seconds

Benchmark B - 5  mins 51 seconds

Benchmark C - 6 mins 9 seconds

Edited March 14, 2020 by jckc221013jamie

[FMFutbol]

Type: Laptop

Model: Lenovo p51

CPU Model: i7 - 7700HQ

CPU Base Frequency: 2.8GHz

CPU Turbo Frequency: 3.8Ghz

RAM: 8GB

RAM Clockspeed: 2400Mhz

GPU: NVIDIA Quadro M1200 4GB

Storage Type: SSD

 

Benchmark A: 2 min 41 Sec

Benchmark B: 18 min 44Sec

Benchmark C: 17 min 29 Sec

[FMFutbol]

Thanks to the original poster for doing this.

Please change first post to update the year from 2018 to 2019 on the holiday dates. 

[EdL]

My new work desktop

Type: Desktop

CPU Model:Ryzen 9 3950X

CPU Base Frequency: 3.5GHz

CPU Turbo Frequency: 4.7 GHz

RAM: 32GB

RAM Clockspeed: 2933Mhz

GPU: Radeon 5600XT

Storage Type: SSD - Samsung 970 EVO Plus

OS: Windows 10 1909 X64

Benchmark A: 1 min 01 Sec

Benchmark B: 6 min 25 Sec

Benchmark C: 4 min 50 Sec

[jonpt]

12 minutes ago, EdL said:

My new work desktop

Type: Desktop

CPU Model:Ryzen 9 3950X

CPU Base Frequency: 3.5GHz

CPU Turbo Frequency: 4.7 GHz

RAM: 32GB

RAM Clockspeed: 2933Mhz

GPU: Radeon 5600XT

Storage Type: SSD - Samsung 970 EVO Plus

OS: Windows 10 1909 X64

Benchmark A: 1 min 01 Sec

Benchmark B: 6 min 25 Sec

Benchmark C: 4 min 50 Sec

That is interesting as in most games the Ryzen is really dependent on the RAM speed but not so much in FM it would seem as our times are really similar, even with an 800MHz ram difference and I’m guessing a 400MHz infinity fabric difference. Unless your 4 extra cores are being utilised 😀

[EdL]

yeah a bit surprising I expected you to beat me on B and me to win more on C

[Brother Ben]

3 hours ago, FMFutbol said:

Thanks to the original poster for doing this.

Please change first post to update the year from 2018 to 2019 on the holiday dates. 

Thanks, OP edited.

I'm surprised by your results, my laptop specs are a good comparison with yours and I would expect similar scores, indeed I would expect yours to be a touch quicker as its 3 generations newer.

I wonder is yours overheating at all?  I've very recently opened mine up and cleaned thoroughly around the fans (which were caked with dust) and did a fresh install of Windows 10.  The difference was like night and day

[Brother Ben]

Second post updated with timings and a link to a google sheets version added if people want to play around with the data

[FMFutbol]

39 minutes ago, Brother Ben said:

Thanks, OP edited.

I'm surprised by your results, my laptop specs are a good comparison with yours and I would expect similar scores, indeed I would expect yours to be a touch quicker as its 3 generations newer.

I wonder is yours overheating at all?  I've very recently opened mine up and cleaned thoroughly around the fans (which were caked with dust) and did a fresh install of Windows 10.  The difference was like night and day

it does heat up, thanks for the suggestion perhaps I will try to open it up and clean it out. Or perhaps the extra RAM made up the difference.

[EdL]

Type: Laptop

Model: Macbook Pro 16" 2019

CPU Model: Intel i9 - 9980HK

CPU Base Frequency: 2.4GHz

CPU Turbo Frequency: 5 GHz

RAM: 16GB

RAM Clockspeed: 2677Mhz

GPU: AMD Radeon Pro 5500M 8GB

Storage Type: SSD

 

Benchmark A: 1 min 04 Sec

Benchmark B: 7 min 31 Sec

Benchmark C: 6 min 22 Sec

[EdL]

9 hours ago, FMFutbol said:

it does heat up, thanks for the suggestion perhaps I will try to open it up and clean it out. Or perhaps the extra RAM made up the difference.

Yes this certainly seems like thermal throttling, it should be faster than that. Also check no anti-virus software is going crazy when you run FM - in task manager

[Brother Ben]

18 minutes ago, EdL said:

Type: Laptop

Model: Macbook Pro 16" 2019

CPU Model: Intel i9 - 9980HK

CPU Base Frequency: 2.4GHz

CPU Turbo Frequency: 5 GHz

RAM: 16GB

RAM Clockspeed: 2677Mhz

GPU: AMD Radeon Pro 5500M 8GB

Storage Type: SSD

 

Benchmark A: 1 min 04 Sec

Benchmark B: 7 min 31 Sec

Benchmark C: 6 min 22 Sec

Wow!

Now that's a laptop!

Hugely impressive results

[Brother Ben]

Type: Custom Desktop

Model: N/A

CPU Model: Intel i5 - 4670k

CPU Base Frequency: 3.4GHz

CPU Turbo Frequency: 3.8 GHz

RAM: 8GB

RAM Clockspeed: 1600Mhz

GPU: Nvidia GeForce GTX 760 - 2GB

Storage Type: SSD

 

Benchmark A: 1 min 27 Sec

Benchmark B: 10 min 50 Sec

Benchmark C: 14 min 20 Sec

My desktop struggles really badly this year after nearly dying a few months back and losing a stick of 8Gb RAM that broke, also lost all of my overclocked BIOS settings in the process so I may test again when I remember how to overclock a Haswell

[Faris Zulkifli]

Type: PC

CPU Model: Ryzen 36000x

CPU Base Frequency: 3.8 Ghz

CPU Turbo Frequency: 4.4  GHz

RAM: 2x8GB (16GB)

RAM Clockspeed: 3200Mhz

GPU: NVIDIA GTX950 2GB

Storage Type: SSD

Benchmark A: 1 min 5 Sec

Benchmark B: 7 min 25 Sec

Benchmark C: 6 min 42 Sec

Edited March 9, 2020 by Faris Zulkifli

[alcedo]

Type: Laptop

Model: Macbook Air 13" mid-2013

CPU Model: Intel i7 - 4650U

CPU Base Frequency: 1.7 GHz

CPU Turbo Frequency: 3.3 GHz

RAM: 8GB

RAM Clockspeed: 1600Mhz

GPU: Intel HD Graphics 5000 1536 MB

Storage Type: SSD

 

Benchmark A: 2 min 06 Sec

Benchmark B: 16 min 49 Sec

Benchmark C: 28 min 27 Sec

[roykela]

Type: PC

Model: Custom

CPU Model: Intel i7 - 9700K

CPU Base Frequency: 3.6GHz

CPU Turbo Frequency: 4.8 GHz

RAM: 16GB

RAM Clockspeed: 2600Mhz

GPU: NVIDIA Geforce GTX 980 Ti - 6GB

Storage Type: HDD

 

Benchmark A: 1 min 16 sec

Benchmark B: 6 min 31 sec

Benchmark C: 6 min 02 sec

Edited March 9, 2020 by roykela

[Bigpole]

this month I'm switching to PC after 5/6 years and I'll do a few tests with my low-mid pc made from popular parts in last year  -

Ryzen 5 1600AF (3.2->3.6 turbo)

Crucial Ballistix 16gb (2x8gb) 3000mhz cl15 

MSI Radeon Armor rx570 4gb

m2 SSD - samsung 970 evo 500gb 

And because I'll be overclocking this a little bit - I think I'll do a 2/3 tests with different ram frequencies (3000, 3200 and maybe 3400 if cpu will handle) and with cpu in box mode and after oc (I believe I could get 3.9/4.0 from this new revision ryzen). 

 

[FMFutbol]

On 06/03/2020 at 12:55, FMFutbol said:

Type: Laptop

Model: Lenovo p51

CPU Model: i7 - 7700HQ

CPU Base Frequency: 2.8GHz

CPU Turbo Frequency: 3.8Ghz

RAM: 8GB

RAM Clockspeed: 2400Mhz

GPU: NVIDIA Quadro M1200 4GB

Storage Type: SSD

 

Benchmark A: 2 min 41 Sec

Benchmark B: 18 min 44Sec

Benchmark C: 17 min 29 Sec

I ran all three tests today again, after dusting off the laptop and using a usb power air vaccuum on the exhausts. New results are much better!!!

 

Type: Laptop

Model: Lenovo p51

CPU Model: i7 - 7700HQ

CPU Base Frequency: 2.8GHz

CPU Turbo Frequency: 3.8Ghz

RAM: 8GB

RAM Clockspeed: 2400Mhz

GPU: NVIDIA Quadro M1200 4GB

Storage Type: SSD

 

Benchmark A: 1 min 39 Sec

Benchmark B: 11 min 47Sec

Benchmark C: 12 min 49 Sec

[Brother Ben]

1 hour ago, FMFutbol said:

I ran all three tests today again, after dusting off the laptop and using a usb power air vaccuum on the exhausts. New results are much better!!!

 

Type: Laptop

Model: Lenovo p51

CPU Model: i7 - 7700HQ

CPU Base Frequency: 2.8GHz

CPU Turbo Frequency: 3.8Ghz

RAM: 8GB

RAM Clockspeed: 2400Mhz

GPU: NVIDIA Quadro M1200 4GB

Storage Type: SSD

 

Benchmark A: 1 min 39 Sec

Benchmark B: 11 min 47Sec

Benchmark C: 12 min 49 Sec

Much better, glad you got it sorted, overheating laptops can be a nightmare

[JamieTC13]

On 06/03/2020 at 11:10, jckc221013jamie said:

Type: Desktop (Custom Made)
  

CPU Model: i5 8600k

CPU Base Frequency : 3.6 GHz

CPU Turbo Frequency: 4.8 GHz Overclocked

RAM: 16 GB DDR4 Corsair vengeance 

RAM Clockspeed: 3200MHz 

GPU: RTX 2060 Zotac

Storage Type: PCIe M.2 SSD Samsung evo 970 500gb

OS: Windows 10 Pro 1909 x64
 

Benchmark A - 51 seconds

Benchmark B - 5  mins 51 seconds

Benchmark C - 6 mins 9 seconds

My CPU overclock was actually 5GHz when i run the test sorry for any inconvenience so if you could please correct the chart to 5GHz overclock sorry again 

[Brother Ben]

15 hours ago, Bigpole said:

this month I'm switching to PC after 5/6 years and I'll do a few tests with my low-mid pc made from popular parts in last year  -

Ryzen 5 1600AF (3.2->3.6 turbo)

Crucial Ballistix 16gb (2x8gb) 3000mhz cl15 

MSI Radeon Armor rx570 4gb

m2 SSD - samsung 970 evo 500gb 

And because I'll be overclocking this a little bit - I think I'll do a 2/3 tests with different ram frequencies (3000, 3200 and maybe 3400 if cpu will handle) and with cpu in box mode and after oc (I believe I could get 3.9/4.0 from this new revision ryzen). 

 

Really interesting CPU this, great value/performance. Strange one to do a refresh on but looks like cracking value, I've added it to the spreadsheet.  Look forward to your results

Edited March 10, 2020 by Brother Ben

[Gee_Simpson]

Type: PC

Model: Custom Build

CPU Model: i9 9900k

CPU Base Frequency: 3.6GHz

CPU Turbo Frequency: 5GHz dual core/4.7GHz all core

RAM: 16GB

RAM Clockspeed: 3200Mhz

GPU: NVIDIA GeForce GTX 1660 - 6GB

Storage Type: SSD

 

Benchmark A: 56 sec

Benchmark B: 5 min 43 sec

Benchmark C: 5 min 06 sec

 

I'm really happy with my results this time around. That was on stock.

 

I will now post my 5ghz all core OC times:

 

Type: PC

Model: Custom Build

CPU Model: i9 9900k

CPU Base Frequency: 3.6GHz

CPU Turbo Frequency: 5GHz all core OC

RAM: 16GB

RAM Clockspeed: 3200Mhz

GPU: NVIDIA GeForce GTX 1660 - 6GB

Storage Type: SSD

 

Benchmark A: 53 sec

Benchmark B: 5 min 24 sec

Benchmark C: 4 min 43 sec

Edited March 10, 2020 by Gee_Simpson

[Gee_Simpson]

Based on those results, who would say the OC is worth it?

Edited March 10, 2020 by Gee_Simpson

[Brother Ben]

24 minutes ago, Gee_Simpson said:

Based on those results, who would say the OC is worth it?

Not really but marginal gains over a whole season/career can make a hell of a difference, assuming you haven't had to go crazy with the voltage and the heat is manageable

Edited March 10, 2020 by Brother Ben

[Gee_Simpson]

18 minutes ago, Brother Ben said:

Not really but marginal gains over a whole season/career can make a hell of a difference, assuming you haven't had to go crazy with the voltage and the heat is manageable

Max temp was 75 °C on one core, average temp was around mid 40's to 50's max. 1.2 volts default vs 1.3 volts 5GHz OC

Edited March 11, 2020 by Gee_Simpson

[Brother Ben]

28 minutes ago, Gee_Simpson said:

Max temp was 75 °C on one core, average temp was around mid 40's to 50's max. 1.2 volts default vs 1.3 volts 5GHz OC

Thats not bad at all really, I'm sorely tempted to build a new system in the near future myself, trouble is that I use the same save on my tablet on my commute so there's no real need to other than the fact I really want to, especially after seeing some of the times in this thread

[EdL]

No I wouldn't say it is worth it for a few seconds given its likely to shorten the life of the processor

[Gee_Simpson]

1 hour ago, EdL said:

No I wouldn't say it is worth it for a few seconds given its likely to shorten the life of the processor

Thanks. I will stick with stock settings then and also because I agree with this:

On 06/03/2020 at 08:54, Brother Ben said:

It’s nice to see more cores and threads being used when the detail is ramped up, the only issue is I think I’d rather use my processing power on more active leagues than compromise on less but with full detail 

what do others think?

I don't have much interest in running leagues at full detail, I would rather run lots more leagues at default detail level. This means the overclock would probably be unnecessary too.

Edited March 11, 2020 by Gee_Simpson

[Brother Ben]

8 hours ago, Gee_Simpson said:

Thanks. I will stick with stock settings

It's personal preference I suppose, If you are looking to run at default detail level then Benchmark B shows that it wouldn't really be worth the extra strain.  To be honest you're using high end equipment anyway. 

I'll look into Overclocking my i5-4670K, in my mind its more worth it at the lower end of the scale when you're looking to eek out a little more performance from older/lower end tech

Edited March 11, 2020 by Brother Ben

[Brother Ben]

Type: Mini PC

Model: Intel NUC

CPU Model: i7 - 7567U

CPU Base Frequency: 3.5GHz

CPU Turbo Frequency: 4.0GHz

RAM: 8GB

RAM Clockspeed: 2133Mhz

GPU: Intel Iris Plus Graphics 650

Storage Type: M.2 SSD

 

Benchmark A: 1 min 34 sec

Benchmark B: 12 min 36 sec

Benchmark C: 19 min 09 sec

 

My little NUC performs really well for what is essentially a media PC plugged into my main TV, the poor cooling (due to size) shows a fair bit especially on test 3 where it throttles down a fair bit.  For the occasional "big screen" game of FM though its more than adequate and even the 3D can be put on medium settings with a few settings put up to high

Edited March 11, 2020 by Brother Ben

[Gee_Simpson]

9 hours ago, Brother Ben said:

It's personal preference I suppose, If you are looking to run at default detail level then Benchmark B shows that it wouldn't really be worth the extra strain.  To be honest you're using high end equipment anyway. 

I'll look into Overclocking my i5-4670K, in my mind its more worth it at the lower end of the scale when you're looking to eek out a little more performance from older/lower end tech

I agree with this. Maybe I will use the overclocked settings when I've had it for 5 years or so. 

Edited March 11, 2020 by Gee_Simpson

[Brother Ben]

I know we've only got a small sample size so far but I wonder what, if any, conclusions can we draw from the results?

The most obvious question is what is the Football Manager processor of choice AMD or Intel?  

It would seem that Intel still has the edge in the kind of setup that 99% of players would run (A&B) but maybe for the full detail, full realism type player the latest AMD's are a clear winner, especially when it comes to price v performance and quite possibly in terms of future proofing as FM makes greater gains in multi core and threading support.

It would be nice if maybe we got to a point where we could recommend certain processors to people purely for FM, maybe -

Best budget Processor

Mid Range Processor

Best overall processor

Obviously we'll need a lot more samples for that but food for thought I guess

 

[Gee_Simpson]

Yeah my PC is a custom build, dunno if you inserted that question mark by mistake? 

[Brother Ben]

1 hour ago, Gee_Simpson said:

Yeah my PC is a custom build, dunno if you inserted that question mark by mistake? 

yeah not sure what happened there

[jonpt]

What really suprised me was when running benchmark C Hardware Monitor recorded 100% usage of all 24 threads on my 3900X for about 5 seconds whilst processing the matches on the Wednesday, Saturday and Sunday. 

Edited March 13, 2020 by jonpt

[Brother Ben]

2 hours ago, jonpt said:

What really suprised me was when running benchmark C Hardware Monitor recorded 100% usage of all 24 threads on my 3900X for about 5 seconds whilst processing the matches on the Wednesday, Saturday and Sunday. 

That's really good news, FM seems to be finally a lot more optimised for multi threads

[VP.]

Type: PC

CPU Model: i7 - 8700

CPU Base Frequency: 3.2GHz

CPU Turbo Frequency: 4.6 GHz

RAM: 16GB

RAM Clockspeed: 2400Mhz

GPU: NVIDIA GeForce 1050 TI - 4GB

Storage Type: 240GB SSD & 2TB HDD

 

Benchmark A: 1 min 5 Sec

Benchmark B: 7 min 17 Sec

Benchmark C: 7 min 01 Sec

[harrycarrie]

Type: Desktop

CPU Model:  i7 9700k

CPU Base Frequency: 3.6GHz

CPU Turbo Frequency: 4.6Ghz

RAM: 32GB

RAM Clockspeed: 2666MHz

GPU: NVIDIA RTX 2080Ti 11GB

Storage Type: M.2  SSD

 

 

Benchmark A - 1 min exactly

Benchmark B - 6 min 23 sec

Benchmark C -5 min 57 sec

 

[Bigpole]

Type: PC

CPU Model: AMD Ryzen 5 1600AF

CPU Base Frequency: 3.2GHz

CPU Turbo Frequency: 3.6 GHz (I've overclocked it so it runs on @3.7 without turbo)

RAM: 16GB DDR4 

RAM Clockspeed: 3000Mhz CL16

GPU: MSI Radeon RX 570 4GB

Storage Type: 500gb NVMe 

 

Benchmark A: 1 min 24 Sec

Benchmark B: 8 min 50 Sec

Benchmark C: 8 min 01 Sec

[FMFutbol]

On 12/03/2020 at 13:37, Brother Ben said:

I know we've only got a small sample size so far but I wonder what, if any, conclusions can we draw from the results?

The most obvious question is what is the Football Manager processor of choice AMD or Intel?  

It would seem that Intel still has the edge in the kind of setup that 99% of players would run (A&B) but maybe for the full detail, full realism type player the latest AMD's are a clear winner, especially when it comes to price v performance and quite possibly in terms of future proofing as FM makes greater gains in multi core and threading support.

It would be nice if maybe we got to a point where we could recommend certain processors to people purely for FM, maybe -

Best budget Processor

Mid Range Processor

Best overall processor

Obviously we'll need a lot more samples for that but food for thought I guess

 

Hopefully more people participate because like you said, the sample size is too small 

[Brother Ben]

18 hours ago, FMFutbol said:

Hopefully more people participate because like you said, the sample size is too small 

Hard to know how to get more people involved, may have to start blatantly plugging it around the forum!

edit - second post updated with latest results