1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Will this be a good all round system, casual gamer mainly total war games

Discussion in 'Games' started by Dovah88, Apr 2, 2013.

  1. EarthDog

    EarthDog

    Joined:
    Dec 31, 2009
    Messages:
    3,316 (1.89/day)
    Thanks Received:
    683
    That benching system is jacked up as NEITHER option gives you a realistic result when playing the actual game...no?
     
  2. Frag Maniac

    Frag Maniac

    Joined:
    Nov 9, 2010
    Messages:
    2,645 (1.83/day)
    Thanks Received:
    551
    Keep in mind I earlier said, "not saying you're wrong". I just have a hard time assuming sites like AnandTech and Kitguru would run benches that are, from what you're saying, very misleading.
     
  3. crazyeyesreaper

    crazyeyesreaper Chief Broken Rig

    Joined:
    Mar 25, 2009
    Messages:
    8,162 (4.00/day)
    Thanks Received:
    2,805
    Location:
    04578
    they are misleading, very much so, the CPU bench is the more accurate bench.

    Average in game performance during large battles on a 4.6 i7 and a GTX 670 is closer to 30-40 fps during melee, time and time again over the last 300 hours,

    CPU bench is fairly accurate my minimum is around 30 the avg around 36 in the CPU bench. During battles thats what it tends to run at more often than not. In smaller battles CPu doesnt matter AS much, but once you pass about 3000 soldiers CPU starts to drag down and becomes the deciding factor.

    And since most fights in game look like this
    As you can see cheaper units have more men but here alone is roughly 15k soldiers
    [​IMG]

    Standing still zoomed out performance is 60-70 fps
    [​IMG]

    At this level of zoom CPU is heavily stressed due to animation sub system the more you zoom out the less 1 on 1 fights the CPU has to handle at this level of zoom in larger melees performance will drop to that seen in the CPU bench.
    [​IMG]

    Here is the CPU Frame Rate log from the CPU test

    Tested at 4.2 GHz GTX 670 FTW at 1006 core clock 8 gig 1866
    [​IMG]

    C:\Users\Ray\AppData\Roaming\The Creative Assembly\Shogun2\logs\frame_rate.log.txt:

    Start Logging
    58.000
    31.000
    49.000
    48.000
    52.000
    45.000
    50.000
    48.000
    47.000
    46.000
    42.000
    38.000
    38.000
    38.000
    37.000
    36.000
    37.000
    37.000
    37.000
    37.000
    36.000
    35.000
    34.000
    30.000
    28.000
    26.000
    28.000
    29.000
    27.000
    29.000
    28.000
    28.000
    27.000
    27.000
    27.000
    26.000
    27.000
    27.000
    27.000
    27.000
    27.000
    26.000
    27.000
    26.000
    26.000
    26.000
    26.000
    27.000
    26.000
    26.000
    26.000
    27.000
    25.000
    26.000
    26.000
    25.000
    27.000
    26.000
    27.000
    27.000
    26.000
    28.000
    27.000
    28.000
    28.000
    28.000
    27.000
    29.000
    28.000
    28.000
    29.000
    29.000
    29.000
    30.000
    29.000
    29.000
    30.000
    31.000
    31.000
    31.000
    33.000
    32.000
    32.000
    32.000
    32.000
    32.000
    33.000
    33.000
    34.000
    33.000
    34.000
    34.000
    33.000
    34.000
    35.000
    33.000
    34.000
    34.000
    35.000
    33.000
    34.000
    34.000
    34.000
    34.000
    33.000
    33.000
    35.000
    34.000
    35.000
    34.000
    35.000
    35.000
    35.000
    35.000
    35.000
    34.000
    35.000
    35.000
    35.000
    End Logging

    GPU Test

    Keep in mind this is 1080p High Preset not Maximum settings
    [​IMG]

    Average FPS: 72.9125

    benchmarks\BenchmarkDX11GraphicsHigh1080p_preferences.script.txt:

    C:\Users\Ray\AppData\Roaming\The Creative Assembly\Shogun2\logs\frame_rate.log.txt:

    Start Logging
    64.000
    68.000
    64.000
    66.000
    60.000
    80.000
    102.000
    102.000
    99.000
    99.000
    98.000
    100.000
    72.000
    69.000
    67.000
    69.000
    69.000
    68.000
    68.000
    67.000
    67.000
    64.000
    66.000
    63.000
    60.000
    64.000
    66.000
    65.000
    65.000
    64.000
    64.000
    65.000
    71.000
    81.000
    78.000
    84.000
    85.000
    87.000
    89.000
    94.000
    93.000
    92.000
    87.000
    67.000
    62.000
    68.000
    68.000
    68.000
    68.000
    67.000
    68.000
    67.000
    67.000
    65.000
    62.000
    64.000
    64.000
    62.000
    61.000
    64.000
    67.000
    68.000
    70.000
    80.000
    76.000
    82.000
    84.000
    87.000
    87.000
    86.000
    79.000
    66.000
    66.000
    64.000
    61.000
    67.000
    69.000
    70.000
    68.000
    64.000
    63.000
    64.000
    64.000
    63.000
    End Logging


    In Game tests

    Again this is 1080p High Preset you will notice many key things are disabled, Shadows / hit the CPU harder the Blood Pack DLC hits the CPU again, as such you can see Minimum is Lower the avg during melee scenes is around 29-32 once fighting stops CPU load drops in terms of needed calculations and performance begins to climb back up, So as you can see YES the benchmarks are in fact NOT telling the truth about performance CPU bench shows 30 avg thats what my avg was during Melee, graphics settings dont really matter that much in regards to the minimum.

    Intel 3770k 4.2 GHz
    CPU test- Minimum FPS = 26
    CPU test- Average FPS = 31

    FX 8350- In the CPU bench the FX chip runs at 4.2Ghz same as trinity A10 5800k the lack of cache will hit performance hard especially when using Blood Pack DLC which carries another 10-20% performance hit depending on the CPU
    [​IMG]

    GPU test- Minimum FPS = 60
    GPU test- Average FPS = 71

    In Game Minimum FPS = 26
    In Game Average FPS = 51

    Frames, Time (ms), Min, Max, Avg
    9456, 180000, 26, 111, 52.533

    Frames, Time (ms), Min, Max, Avg
    2209, 42775, 27, 78, 51.642

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]
     
    Last edited: Apr 4, 2013
    EarthDog says thanks.
  4. cadaveca

    cadaveca My name is Dave

    Joined:
    Apr 10, 2006
    Messages:
    14,050 (4.50/day)
    Thanks Received:
    7,208
    Location:
    Edmonton, Alberta
    Technically, if you place any faith in the "new testing" of multi-GPU configs, sites reporting things that don't relate reality are a fact. FPS given by benchmarks doesn't always = what's shown on the screen.

    And personally, I'll happily admit that. My memory reviews, for example, are made with benchmarks I hand-picked to show differences. If memory doesn't affect the results, the benchmark is not used. There's many instances where memory speed matters little, if any.


    Of course I do say that in my memory reviews. You need some sort of base to make a comparison, and any benchmark may use settings or such that relates to nothing useful...like 3DMark.

    I can post completely identical scores in any of those benchmarks using two different systems, and they'll have the same result score. Does that mean they perform equally? NOPE.

    3DMark is the classic example of it being widely known that to take the numbers as they are presented isn't the right approach.
     
  5. EarthDog

    EarthDog

    Joined:
    Dec 31, 2009
    Messages:
    3,316 (1.89/day)
    Thanks Received:
    683
    The GPU score in the modern 3DMarks are OK... however the overall, not so much as so many of them (read 06/Vantage/11) love CPU and CORES to come up with that overall score.
     
    cadaveca says thanks.
  6. crazyeyesreaper

    crazyeyesreaper Chief Broken Rig

    Joined:
    Mar 25, 2009
    Messages:
    8,162 (4.00/day)
    Thanks Received:
    2,805
    Location:
    04578
    Regardless those are my findings

    going from a 2500k to a 3770k minimum FPS at 4.4ghz i got a 19% increase on the 3770k.

    No matter how you slice it a 5ghz AMD FX or Trinity chip is still slower than a stock Intel i5 chip when it matters in a Total War game. So everyone here can take what ive present as they wish i dont really care. Im not here to hold hands what i presented isnt even MAX settings, which the avg drops even more yet minimum stays the same. and that 5k vs 5k is minor since i regularly experience double those troops in regular campaign gameplay.
     
  7. Dovah88 New Member

    Joined:
    Apr 2, 2013
    Messages:
    24 (0.04/day)
    Thanks Received:
    1
    well i might get this system instead

    Processor: Intel i5 3330

    Chipset: Intel B7M

    Memory: 8GB DDR3 RAM


    Internal Hardware:
    Storage: 1TB SATA
    Optical: DVD Burner
    Graphics: AMD 7770 OC 1GB

    LED Lights: Blue LED Strip

    Case Description:
    Case: ATX Tower
    Power Supply: Included

    Software:
    Operating System: Windows 7 Home Prem
    pls suggest or reply
     
    Last edited: Jun 30, 2013

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page