Jump to content

FM17 Performance Benchmarking Thread


Recommended Posts

In previous years first Yugo23 and then @Barside created new benchmark threads for releases of FM to do performance comparisons. As Barside has previously stated he won't have time to do it this year, and as I recently built a new PC to play FM17, I thought I'd initiate the benchmarking for FM17 this year.

As per previous years, I've created a save with all 116 leagues from all 51 nations loaded using a large database. Total player count is 167,000. Detail level settings are all at their default values.

Like Barside did for FM16, two separate benchmark files have been created:

Benchmark - A starts on 20th August

Benchmark - B starts on the 25th August and will capture the end of the transfer window.

To run the benchmark, download benchmark save games from the following mirror:

DropBox: FM17 Benchmark A

DropBox: FM17 Benchmark B

Once loaded, ensure that auto save interval is disabled, and also set processing to Fast (Less Responsive), found at Preferences, More Match Settings.

Optionally, restart your computer and close any unnecessary programs before running this benchmark.

After you load the appropriate benchmark save game, just go on holiday for a week, and measure the time needed for processing using a stopwatch on your wristwatch or phone.

Example:

Benchmark A: Holiday until 27/8/2016

Benchmark B: Holiday until 1/9/2016

If you're running on a laptop make sure it is plugged into the power socket.

Post your results in the following format:

Quote

 

Computer Type: Laptop

CPU: Intel Celeron G550

CPU Frequency: 2600 MHz

RAM: 8GB DDR3 1600 MHz CL9

OS: Win 7 64-bit

Storage: SSD

Test A

Time: 9 min 56 sec

Test B

Time: 11 min 12sec

 

Links to older threads for those who want to compare their scores:

FM 16 Benchmark

FM 15 Benchmark

FM 14 Benchmark

Additional Notes:

When I ran the FM17 benchmark against my new PC, it actually processed slower then last years FM16 benchmark file on the same PC, which was a surprise considering the new 64bit processing is enabled. To maintain consistency and ensure I was comparing like for like, I created new FM16 benchmark A & B files with the same settings as the new FM17 benchmark files. This was to make sure the files could be as the same as possible.

Savegame Data:

FM17 Benchmark database: 167440 players, 86421 staff, 57974 teams

New FM16 Benchmark database: 163785 players, 69892 staff, 55414 teams

Benchmark Results:

u2sTmr3.jpg

mFQjxyv.jpg

 

Link to post
Share on other sites

  • Replies 217
  • Created
  • Last Reply

My results:

CPU: Intel i7-6700K

CPU Frequency: 4.0GHz

RAM: 16.00GB Dual-Channel DDR4 @ 3200MHz

OS: Win 10 Pro 64-bit

Storage: M.2 SSD NVMe Samsung EVO 960 500GB

FM17 Benchmark A

Time: 2 min 38 sec

FM17 Benchmark B

Time: 5 min 11 sec

New FM16 Benchmark A

Time: 2 min 19 sec

New FM16 Benchmark B

Time: 4 min 10 sec

Link to post
Share on other sites

My results:

CPU: I5 6300HQ

CPU Frequency: 2.30Ghz  (Turbo 3.20GHz)

RAM: 8.00GB DDR3L 

OS: Win 10 Home 64-bit

Storage: SSD

Type: HP NOTEBOOK PAVILION "gaming" 15-AK020NB

FM17 Benchmark A

Time: 5 min 12 sec

FM17 Benchmark B

Time: 10 min 11 sec

I don't know if I should try the FM 2016 benchmark as I had performance issues on that version with windows 10. I could only play with maximum borderless mode after playing around with the task manager.

Link to post
Share on other sites

14 minutes ago, fabioke said:

My results:

****

I don't know if I should try the FM 2016 benchmark as I had performance issues on that version with windows 10. I could only play with maximum borderless mode after playing around with the task manager.

Thanks for the results.

Interesting you say you had performance issues with FM16 on Windows 10 before, I was about to submit up another post but with the opposite issue, FM17 is running slower for me on Windows 10. I've been running some other tests on what to me look near identical savegame files in FM16 and FM17 on the same PC and FM16 is running much faster despite being a 32bit process, compared to the FM17 64bit process.

I've been trying to decide if there something wrong with my PC build as everyone keeps saying how much quicker FM17 is but I am not experiencing it.

Link to post
Share on other sites

My results:

CPU: Intel Core i7 3820

CPU Frequency: 3.60Ghz

RAM: Corsair 16GB DDR3L

OS: Win 10 Home 64-bit

Storage: HDD (3TB)

FM17 Benchmark A:
Time: 4 min 35 sec

FM17 Benchmark B:
Time: 9 min 10 sec

Transfer deadline day is responsible for 5 mins 20 secs of the time in Benchmark B.
Apparently my computer doesn't handle those days very well.
Happy with Benchmark A though.

Link to post
Share on other sites

9 hours ago, roykela said:

Transfer deadline day is responsible for 5 mins 20 secs of the time in Benchmark B.
Apparently my computer doesn't handle those days very well.
Happy with Benchmark A though.

Yes the Transfer Deadline day certainly sucks up a lot of processing time when it runs.

Link to post
Share on other sites

Thanks for putting in the effort to set this up @Powermonger

CPU: Intel i7-6700K

CPU Frequency: 4.7GHz

RAM: 16GB Dual-Channel DDR4 @ 2400MHz Crucial Ballistic Sport

OS: Win 10 Pro 64-bit

Storage: Samsung EVO 250GB

FM17 Benchmark A

Time: 2 min 49 sec

FM17 Benchmark B

Time: 5 min 35 sec

Interesting that mine is clocked higher but I had worse results. I wonder if that's down to the RAM or something else. I did not do a fresh restart, but I'll try that next time.

Link to post
Share on other sites

1 hour ago, eple said:

Interesting that mine is clocked higher but I had worse results. I wonder if that's down to the RAM or something else. I did not do a fresh restart, but I'll try that next time.

Yes very interesting. My PC I am testing on is newly built and it is only running Windows and FM16/FM17, also all the latest drivers, that may contribute to it. I just ran the 'FM17 Benchmark A' test two more times and got timings of 2min 40 sec and 2min 36 sec. If you do the fresh restart and retry, let us know how you go.

Link to post
Share on other sites

51 minutes ago, Powermonger said:

Yes very interesting. My PC I am testing on is newly built and it is only running Windows and FM16/FM17, also all the latest drivers, that may contribute to it. I just ran the 'FM17 Benchmark A' test two more times and got timings of 2min 40 sec and 2min 36 sec. If you do the fresh restart and retry, let us know how you go.

I noticed that small things like testing on maximised borderless windowed costed me nearly 1 minute. 

Link to post
Share on other sites

1 hour ago, fabioke said:

I noticed that small things like testing on maximised borderless windowed costed me nearly 1 minute. 

I actually tested this earlier in the week, I ran the same benchmark 6 times: 3 times with full screen and 3 times with borderless windowed mode.

For my PC they performed the same, both getting with in a second of each other and sometimes slower. Maybe it depends on the graphic card but I would not be surprised this slowing down processing for others. 

Link to post
Share on other sites

CPU: 6700HQ 

Fre: 2.6/3.5

Ram: 16gb DDR4

windows 10 64bit

SSD

benchmark A : 4mins 7 seconds 

benchmark B: 8 mins 18 seconds 

just brought this today and I'm disappointed to be fair. I was hoping it would bench at around 3 mins for A. 

Any ideas how to speed it up? 

Graphics are pretty amazing though. 

Link to post
Share on other sites

6 hours ago, Baldyloxxx said:

just brought this today and I'm disappointed to be fair. I was hoping it would bench at around 3 mins for A. 

Any ideas how to speed it up? 

Only thing I can suggest is make sure you have all the latest Intel chipset drivers installed and turn off any unnecessary applications services.

Link to post
Share on other sites

On 7-1-2017 at 18:22, Baldyloxxx said:

CPU: 6700HQ 

Fre: 2.6/3.5

Ram: 16gb DDR4

windows 10 64bit

SSD

benchmark A : 4mins 7 seconds 

benchmark B: 8 mins 18 seconds 

just brought this today and I'm disappointed to be fair. I was hoping it would bench at around 3 mins for A. 

Any ideas how to speed it up? 

Graphics are pretty amazing though. 

Your result is perfectly normal if you compare it to my I5 6300HQ. Hyperthreading can only do so much. And with laptops you should also consider airflow.

Seeing your result makes me envious that I didn't buy the I7 6700HQ.

I'm hoping to see some Kaby, Zen (within a few months) and haswell/broadwell-e results.

Link to post
Share on other sites

7 hours ago, fabioke said:

Your result is perfectly normal if you compare it to my I5 6300HQ. Hyperthreading can only do so much. And with laptops you should also consider airflow.

Seeing your result makes me envious that I didn't buy the I7 6700HQ.

I'm hoping to see some Kaby, Zen (within a few months) and haswell/broadwell-e results.

Yeah I forget that the better times are on the desk tops PC'S. I'm actually pretty happy with the machine. 

Its got and gtx970 gpu too so the graphics are pretty flawless. I only play FM so it's actually probably overkill but hey ho. 

Link to post
Share on other sites

As soon as I saw this thread I wanted to see how my Kaby Lake i5 7600k stood up against the Skylake i7s. It faired quite well, but not what I was expecting.

CPU: I5 7600K, I couldn't be bothered waiting for zen anymore, I hope I don't regret it.

CPU Frequency: 4.8GHz, The CPU is a frequency monster. I can push it easily to 5GHz, but I get 90C after 5 minutes on 100% load with my Kraken x42 CPU cooler.

Ram: 16GB Corsair dominator DDR4 @ 2600MHz (it's quad channel ram in a dual channel board, I don't think it makes a difference)

OS: Windows 10 64bit

Storage: Cheap 500gb SanDisk SSD.

FM 17 Benchmark A:  2 minutes 49 seconds @4.6GHz (Funnily enough exactly the same as eple's i7 at a similar frequency)

2 minutes 41 seconds @4.8GHz still couldn't be the other i7 :( (might be because of the NVMe SSD and faster ram)

FM 17 Benchmark B: 5 minutes 37 seconds @ 4.8GHz (Blown away by both i7s, no idea why, considering it was on par in the first)

Hyper-threading definitely has a much greater influence in the second test, no doubt an NVMe SSD helps out a lot as well. I could repeat benchmark B, but the results would most likely stay the same. Overall I'm pleased.

 

 

 

 

Link to post
Share on other sites

16 hours ago, RubenL said:

As soon as I saw this thread I wanted to see how my Kaby Lake i5 7600k stood up against the Skylake i7s. It faired quite well, but not what I was expecting.

Not too bad of results for the i5, seems to match other benchmark results found on the net when comparing the 7600K and 6700K. Looking forward to see the results here when someone gets a 7700K.

Tonight I finally got around to doing some overclocking of my 6700K and some FM17 benchmarking.

Results:

Intel i7-6700K @ 4.7GHz OC

FM17 Benchmark A: 2 min 27 sec

FM17 Benchmark B: 4 min 51 sec

A 10 sec improvement for Test A, and a 20 sec improvement for Test B. I tried 4.8GHz but my system couldn't complete Test A without crashing.

Link to post
Share on other sites

I tried using a 5GHz, just because i wanted to see the absolute best my CPU could do. I would never run at this speed normally, the max temp got during the tests was 70C. The improvement isn't worth the temp increase, so once i finished i changed it back to 4.8GHz immediately.

CPU: i5 7600k @5.0GHz

FM17 Benchmark A: 2 minutes 39 Seconds

FM17 Benchmark B: 5 minutes 23 Seconds

I got a 2 second improvement from A, and somehow i got 14 seconds faster in B.

Link to post
Share on other sites

On 1/15/2017 at 03:09, RubenL said:

I tried using a 5GHz, just because i wanted to see the absolute best my CPU could do. I would never run at this speed normally, the max temp got during the tests was 70C. The improvement isn't worth the temp increase, so once i finished i changed it back to 4.8GHz immediately.

CPU: i5 7600k @5.0GHz

FM17 Benchmark A: 2 minutes 39 Seconds

FM17 Benchmark B: 5 minutes 23 Seconds

I got a 2 second improvement from A, and somehow i got 14 seconds faster in B.

Not too bad for the i5 OCed. It is going to be real interesting when someone upgrades to the i7-7700K and benchmarks it with FM17.

Link to post
Share on other sites

I tried a fresh restart and also changed voltage to manual from adaptive (1.375 IIRC). Dunno if it makes a difference.. not that computer savvy :) Anyways, results didn't differ enough from my 1st attempt to conclude anything. Dunno why yours is quicker @Powermonger but hope you enjoy it :) Probably multiple things that contribute to it, but I guess the lower temp and the quicker RAM is the prime suspects.

Link to post
Share on other sites

CPU: I7-4710mq

Frequency: 2.5/3.5

Ram: 16gb DDR3

Windows 10 64bit

SSD

Benchmark A: 4mins 20 seconds 

Benchmark B: 8 mins 26 seconds 

Pretty much what I expected, glad to see it only just off the pace of the 6700hq although benchmarking websites suggest as much.  All about power conservation these days rather than raw speed increases.

I'll benchmark my I5-4760k desktop and my fathers massively overclockable £50 Pentium G3258 as soon as I get a chance

Link to post
Share on other sites

50 minutes ago, bennytee said:

Pretty much what I expected, glad to see it only just off the pace of the 6700hq although benchmarking websites suggest as much.  All about power conservation these days rather than raw speed increases.

I'll benchmark my I5-4760k desktop and my fathers massively overclockable £50 Pentium G3258 as soon as I get a chance

Thanks for testing, I'll add your results to the table tonight.

 

On 1/17/2017 at 01:13, eple said:

I tried a fresh restart and also changed voltage to manual from adaptive (1.375 IIRC). Dunno if it makes a difference.. not that computer savvy :) Anyways, results didn't differ enough from my 1st attempt to conclude anything. Dunno why yours is quicker @Powermonger but hope you enjoy it :) Probably multiple things that contribute to it, but I guess the lower temp and the quicker RAM is the prime suspects.

Ok thanks for testing again @eple. Maybe the faster RAM and SSD is enough to add a few extra percentage differences between our two. What video card do you have?

Link to post
Share on other sites

1 hour ago, bennytee said:

As promised

CPU: I5-4670k

Frequency: 4.3GHz

Ram: 16gb DDR3

Windows 10 64bit

SSD

Benchmark A: 3 mins 29 seconds 

Benchmark B: 7 mins 0 seconds

Thanks I'll add the results to the table. I didn't get a chance to add your other results yet. My old PC is an i5-4560, similar to your specs, I should record the results from that too.

Link to post
Share on other sites

CPU: Intel i5 6600k 

Frequency: 3500 MHz (Overclocked to 4.4ghz)

RAM: 32 GB dualchannel DDR4 2400

OS: Win 10 64-bit

Storage: SSD

Test A

Time: 3 min 14 sec

Test B

Time: 6 min 40 sec

 

Overall "meh" feeling, rarely see i7 recommended for gaming but it makes a surprising amount of difference according to these benchmarks. Strangely the CPU utilization was sitting around 46% most of the tests, jumping to 100% occasionally.

Link to post
Share on other sites

To put things into perspective and satisfy my curiosity I decided to do this.....

CPU: Intel i3 2330M

Frequency: 2.2 GHz

RAM: 8 GB DDR3 1333MHz

OS: Win 10 64-bit

Storage: HDD

Test A

Time: 11 min 22 sec

Test B

Time: 22 min 19 sec

 

 

Link to post
Share on other sites

15 minutes ago, Powermonger said:

Thanks for that. You must be very patient to wait so long!

haha, no problem.

was wondering how it would compare - need to get myself a new laptop now I think!

Link to post
Share on other sites

On 17/01/2017 at 20:57, bennytee said:

CPU: I7-4710mq

Frequency: 2.5/3.5

Ram: 16gb DDR3

Windows 10 64bit

SSD

Benchmark A: 4mins 20 seconds 

Benchmark B: 8 mins 26 seconds 

Pretty much what I expected, glad to see it only just off the pace of the 6700hq although benchmarking websites suggest as much.  All about power conservation these days rather than raw speed increases.

I'll benchmark my I5-4760k desktop and my fathers massively overclockable £50 Pentium G3258 as soon as I get a chance

So DDR4 ram and the skylake make little difference. When the warranty is up on my laptop I may try upgrading my ram. I don't think it's very good. 

Link to post
Share on other sites

On 23/01/2017 at 18:57, bananahk said:

CPU: Xeon 1230v2

Frequency: 3.3 GHz

RAM: 16 GB DDR3 1333MHz

OS: Win7 Pro 64-bit

Storage: SSD

Test A

Time: 4 min 20 sec

Test B

Time: 9 min 0 sec

Nice to see a server class processor in the mix, not sure what to make of the results though.  Most of these results seem to follow quite closely with the passmark scores

Link to post
Share on other sites

12 hours ago, bananahk said:

is my performance limited by the Win 7?

I doubt it, extra things like browsers and processes may affect it though. I did a little research into your processor and your results seem about right tbh

Link to post
Share on other sites

On 2017/1/29 at 01:11, bennytee said:

I doubt it, extra things like browsers and processes may affect it though. I did a little research into your processor and your results seem about right tbh

O...I'm not mean I not satisfactory with the result, just question on OS effect CPU performance.

My CPU is equal to I7-3770 level

Link to post
Share on other sites

38 minutes ago, bananahk said:

O...I'm not mean I not satisfactory with the result, just question on OS effect CPU performance.

My CPU is equal to I7-3770 level

I don't think the OS would make too much of a difference, maybe 0.5% but your memory speed and type would contribute more. 

Link to post
Share on other sites

CPU: Phenom X4 955

CPU Frequency: 3600 MHz

RAM: 8GB DDR3 1333 MHz CL9

OS: Win 10 64-bit

Storage: SSD

Test A

Time: 7 min 25 sec

Test B

Time: 13 min 20sec

 

CPU: Intel i7 4700hq

CPU Frequency: 2400/3200 MHz

RAM: 8GB DDR3 1600 MHz single channel

OS: Win 1064-bit

Storage: HDD @ 5400rpm

Test A

Time: 5 min 5 sec

Test B

Time: 9 min 33sec

Link to post
Share on other sites

Just for a bit of fun I decided to do the A test again on various overclocks

i5-4670k

(my RAM is 1600mhz)

4.3ghz - 83°C - 3min 26sec
4.2ghz - 75°C - 3min 29sec
4.1ghz - 72°C - 3min 30sec
4.0ghz - 66°C - 3min 33sec
3.8ghz - 59°C - 3min 56sec (stock)

For those who are interested the difference in voltage from 4.0 to 4.3 is 1.200 - 1.365 (it's an awful chip)

Overall considering the increase in voltage and temps i'm quite happy at 4.0ghz considering the difference is only 7 seconds

I'd be interested to see the various timings from other peoples overclocks to see if they are having similar results and maybe see the value in doing it

Link to post
Share on other sites

23 minutes ago, bennytee said:

Just for a bit of fun I decided to do the A test again on various overclocks

i5-4670k

(my RAM is 1600mhz)

4.3ghz - 83°C - 3min 26sec
4.2ghz - 75°C - 3min 29sec
4.1ghz - 72°C - 3min 30sec
4.0ghz - 66°C - 3min 33sec
3.8ghz - 59°C - 3min 56sec (stock)

For those who are interested the difference in voltage from 4.0 to 4.3 is 1.200 - 1.365 (it's an awful chip)

Overall considering the increase in voltage and temps i'm quite happy at 4.0ghz considering the difference is only 7 seconds

I'd be interested to see the various timings from other peoples overclocks to see if they are having similar results and maybe see the value in doing it

Interesting progression of results, certainly doesn't look worth the performance gain and heat levels to jump from 4.0 GHz to 4.3 GHz.

Link to post
Share on other sites

  • 2 weeks later...

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...