Welcome

Please register for Total War Access to use the forums. If you're an existing user, your forum details will be merged with Total War Access if you register with the same email or username. For more information please read our FAQ’s here.

Categories

Mortal Empires Performance test results

makar55makar55 Posts: 1,613Registered Users
edited July 2018 in General Discussion
Mortal Empires 7350K, 7700K, 8350K, 8600K, 8700K and Pentium turn time massive test
Test rig:
CPU: Intel Core i7 8700K
GPU: Nvidia GTX 1070
RAM: HyperX Predator 3200MHz 16(4x4)GB
SSD: Plextor M6Pro 256GB
Motherboard: Gigabyte Z370 HD3P

Game settings:
All ultra(shadows are extreme) + FXAA, Resolution: 2560x1080
Mid-end game, turn 133-134, Chaos warriors.
With 8700K you can get another Intel CPU(architecture is the same), just disabling cores, on/off multithreading, changing clock and voltage.

1) Intel Core i3 7350K, 2 cores, 4 threads, 5GHz
Turn time: 48 sec
http://tinypic.com/view.php?pic=zvakig&s=9#.W1X-fdL7QbU

2) Intel Core i7 7700K, 4 cores, 8 threads, 5GHz
Turn time: 47 sec
http://tinypic.com/view.php?pic=1zejkpj&s=9#.W1X-edL7QbU

3) Intel Core i3 8350K, 4 cores, 4 threds, 5GHz
Turn time: 47 sec
http://tinypic.com/view.php?pic=xnysyr&s=9#.W1X-2NL7QbU

4) Intel Core i5 8600K, 6 cores, 6 threads, 5GHz
Turn time: 46 sec
http://tinypic.com/view.php?pic=24gvfj5&s=9#.W1X-1NL7QbU

5) Intel Core i7 8700K, 6 cores, 12 threads, 5GHz
Turn time: 45 sec
http://tinypic.com/view.php?pic=6f72pg&s=9#.W1X_C9L7QbU

6) Intel Pentium G5600, 2 cores, 4 threads, 3.9GHz
Turn time: 60 sec
http://tinypic.com/view.php?pic=14116bp&s=9#.W1X_0NL7QbU

There is no difference in campaign and battle with these processors(exclude Pentium). You can play with 150$ processor in 5GHz and you get almost the best result in this game.

Total War: Warhammer II utilizes only one core for game factions computing, but in battle and campaign map you need more cores, because your videocard demands results from processor for in-game graphic.

This game is fully videocard-dependent, neither NASA computer nor Intel Core i9 processor will not help you to reduce turn-time.

Original thread>
https://steamcommunity.com/app/594570/discussions/0/1711816348629696833/
Post edited by makar55 on
Tagged:

Comments

  • NeverBackDownNeverBackDown Senior Member Posts: 1,815Registered Users
    edited July 2018
    Not that great of a test.

    Just disabling and turning of HT isn't 100% accurate.

    You still have cache to worry about. For this game it won't matter much but if all you did was turn cores off it's not entirely accurate.

    On top of that most of those CPUs won't hit 5ghz and the ones that do will need some serious cooling. A more realistic test would have been stock clocks. To better show the difference. I understand the purpose is to show how each CPU handles the game but 5ghz isn't realistic for most CPUs and most people won't bother to overclock that high (if at all).

    By leaving everything at the same clock speed you take out half the equation for speed. That's why there is little difference. You didn't specify which API you ran it in and you also didn't test multiple sets of turns. A picture of the current state of the campaign would also help.

    Needs more work done to make this a valid test. As of now it's not entirely done.


    Now with that said. This game isn't very well multi threaded so the results won't change much. But still needs to be more accurate before you can conclude anything.
  • makar55makar55 Posts: 1,613Registered Users

    Not that great of a test.

    Just disabling and turning of HT isn't 100% accurate.

    You still have cache to worry about. For this game it won't matter much but if all you did was turn cores off it's not entirely accurate.

    On top of that most of those CPUs won't hit 5ghz and the ones that do will need some serious cooling. A more realistic test would have been stock clocks. To better show the difference. I understand the purpose is to show how each CPU handles the game but 5ghz isn't realistic for most CPUs and most people won't bother to overclock that high (if at all).

    By leaving everything at the same clock speed you take out half the equation for speed. That's why there is little difference. You didn't specify which API you ran it in and you also didn't test multiple sets of turns. A picture of the current state of the campaign would also help.

    Needs more work done to make this a valid test. As of now it's not entirely done.


    Now with that said. This game isn't very well multi threaded so the results won't change much. But still needs to be more accurate before you can conclude anything.

    Thanks for info, I will tell the guy who tested the game everything you've suggested.
  • NeverBackDownNeverBackDown Senior Member Posts: 1,815Registered Users
    makar55 said:

    Not that great of a test.

    Just disabling and turning of HT isn't 100% accurate.

    You still have cache to worry about. For this game it won't matter much but if all you did was turn cores off it's not entirely accurate.

    On top of that most of those CPUs won't hit 5ghz and the ones that do will need some serious cooling. A more realistic test would have been stock clocks. To better show the difference. I understand the purpose is to show how each CPU handles the game but 5ghz isn't realistic for most CPUs and most people won't bother to overclock that high (if at all).

    By leaving everything at the same clock speed you take out half the equation for speed. That's why there is little difference. You didn't specify which API you ran it in and you also didn't test multiple sets of turns. A picture of the current state of the campaign would also help.

    Needs more work done to make this a valid test. As of now it's not entirely done.


    Now with that said. This game isn't very well multi threaded so the results won't change much. But still needs to be more accurate before you can conclude anything.

    Thanks for info, I will tell the guy who tested the game everything you've suggested.
    It's a good baseline for 5ghz. So it's not the worst benchmark i've ever seen but it's not entirely representative of the mass population who play this game. That's all I meant:)
  • LestaTLestaT Senior Member Posts: 3,269Registered Users
    Just upload that save game and anyone can benchmark their own base on their own rig and post the results.
Sign In or Register to comment.