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. cadaveca

    cadaveca My name is Dave

    Joined:
    Apr 10, 2006
    Messages:
    14,049 (4.51/day)
    Thanks Received:
    7,205
    Location:
    Edmonton, Alberta
    Trinity is great. I was praising it before the launch. even.


    But at the same time, I know Crazy's listed builds for tonnes of people on the total war forums, and everyone has been very happy with those builds. He's also spent considerable time playing those titles on both AMD and iNtel systems. FOr that engine, he's my go-to expert.
     
  2. EarthDog

    EarthDog

    Joined:
    Dec 31, 2009
    Messages:
    3,316 (1.89/day)
    Thanks Received:
    683
    Makes sense assuming in your reviews (DrD) you tested that game specifically. :)
     
  3. CJCerny

    CJCerny

    Joined:
    Nov 25, 2008
    Messages:
    855 (0.40/day)
    Thanks Received:
    249
    Location:
    Akron, OH
    So why didn't anybody bother to look at the Shogun 2 numbers from Cadaveca's review of Trinity on this web site?
     
  4. crazyeyesreaper

    crazyeyesreaper Chief Broken Rig

    Joined:
    Mar 25, 2009
    Messages:
    8,162 (4.00/day)
    Thanks Received:
    2,805
    Location:
    04578
    I posted them on the last page

    Trinity APU gets 16 fps in the CPU bench which has 0 GPU input and represents the AVG minimum frame rate during game play

    Ivy Bridge quadcores aka the 3350p and above offer nearly double the avg frame rate during melee battles.

    The GPU bench just tests GPUs
    CPU bench tests just CPU

    however with 40 vs 40 units or hell even 20 vs 20 slow down is still possible on stock Ivy Bridge systems, the game is extremely CPU limited where even a 3770k + 2x 7950s only musters around 29 FPS in cadaveca's testing. CPU is the limiting factor period.

    Sadly the CPU bench is also limited in game performance is entirely different compared to both benchmarks and rests somewhere in between.

    But having tested,
    FX 4100
    Phenom II 965
    Phenom II 1090T
    Sandy Bridge 2500k
    Sandy Bridge 3960X
    Ivy Bridge 3770k

    with higher end GPUs the AMD chips consistently deliver sub par performance in battles.
     
    Last edited: Apr 3, 2013
    Dovah88 says thanks.
  5. Aquinus

    Aquinus Resident Wat-man

    Joined:
    Jan 28, 2012
    Messages:
    6,460 (6.47/day)
    Thanks Received:
    2,188
    Location:
    Concord, NH
    You mean this?

    [​IMG]
     
    Dovah88 says thanks.
  6. crazyeyesreaper

    crazyeyesreaper Chief Broken Rig

    Joined:
    Mar 25, 2009
    Messages:
    8,162 (4.00/day)
    Thanks Received:
    2,805
    Location:
    04578
    Also most info on the net is now old

    the flash pack update changed the performance numbers

    those numbers above are pre flash update in game

    after update performance improved dramatically on Intel systems
     
    Dovah88 says thanks.
  7. cadaveca

    cadaveca My name is Dave

    Joined:
    Apr 10, 2006
    Messages:
    14,049 (4.51/day)
    Thanks Received:
    7,205
    Location:
    Edmonton, Alberta
  8. CJCerny

    CJCerny

    Joined:
    Nov 25, 2008
    Messages:
    855 (0.40/day)
    Thanks Received:
    249
    Location:
    Akron, OH
    That's the one. Took me about 12 seconds to find it. Granted, there isn't a graph that lays it all out for every question every poster has on this site, but there sure was in this case. Let's not get lazy out there.
     
  9. Aquinus

    Aquinus Resident Wat-man

    Joined:
    Jan 28, 2012
    Messages:
    6,460 (6.47/day)
    Thanks Received:
    2,188
    Location:
    Concord, NH
    In the review list there is only one other review that has Shogun 2, and that was the 8350 review and it doesn't include Trinity on it so I don't know how Trinity changes with the patch. Am I looking in the wrong place? You made the reviews so you might know where a better example would be.
     
  10. crazyeyesreaper

    crazyeyesreaper Chief Broken Rig

    Joined:
    Mar 25, 2009
    Messages:
    8,162 (4.00/day)
    Thanks Received:
    2,805
    Location:
    04578
    Trinity performs within 2-3% of the 8350 in Shogun 2 specifically.

    At least from what posters on the TWC forums have told me, along with a couple of buddies owning FX 4100 based systems

    Trinity performs a bit better at the same clocks speeds when compared to Bulldozer aka FX 4100 / 8150 etc etc

    compared to the Vishera update Trinity is a bit slower but there so close that its still a wash. Most likely its cache related.
     
    Last edited: Apr 3, 2013
    Aquinus and Dovah88 say thanks.
  11. Frag Maniac

    Frag Maniac

    Joined:
    Nov 9, 2010
    Messages:
    2,643 (1.83/day)
    Thanks Received:
    549
    Here's some TW:S2 benches with the Trinity APU mentioned. Unfortunately no direct comparison with the Intel CPU OP mentioned, but a rough idea of what Trinity can do:

    http://www.anandtech.com/show/6332/amd-trinity-a10-5800k-a8-5600k-review-part-1/3
    http://www.kitguru.net/components/g...2a85x-up4-amd-a10-5800k-review-w-discrete/23/

    My personal opinion though, I feel it's silly to build an entire gaming system around one game or game series. You never know when your gaming tastes will change or broaden.
     
    Dovah88 says thanks.
  12. Dent1

    Joined:
    May 18, 2010
    Messages:
    3,158 (1.95/day)
    Thanks Received:
    915
    I agree. Like a few years back when people dropped thousands in Crysis builds. In the end the game still ran like crap. They couldn't have spent half the money on a rig capable of playing everything else BUT Crysis.

    It's best to build a rig capable of playing a varied set of games, with longevity.

    In regards to the OP, if he can afford an i5-3330. He can definitely afford a Piledriver FX-6300 Six Core, then use the saving towards a better GPU.
     
    Last edited: Apr 3, 2013
    Dovah88 says thanks.
  13. crazyeyesreaper

    crazyeyesreaper Chief Broken Rig

    Joined:
    Mar 25, 2009
    Messages:
    8,162 (4.00/day)
    Thanks Received:
    2,805
    Location:
    04578
    Both those reviews are the graphics performance not CPU side, in those benchmarks the troops dont move they pretty much just stand still while a camera pans around so next to 0 CPU input.

    example

    SHogun 2 i get over 100 FPS in the regular graphics Benchmark
    CPU Benchmark i get 30 min 36 avg

    In game large battles i actually have a minimum of around 25 and an average of 30-33 during extremely large melee fights.
     
    Last edited: Apr 3, 2013
    Dovah88 says thanks.
  14. Dovah88 New Member

    Joined:
    Apr 2, 2013
    Messages:
    24 (0.04/day)
    Thanks Received:
    1
    I mainly play the xbox but i do like to play comp every now and then and the total war games i like but i go back to playing some of the old games as well like AOE and stuff i would like a build that would be able to resonably play total war games doesn't have to be super maxed out just upgrading from old tower as its useless.
    So it would be cheaper to build the comp myself but i have never built one before so not sure how complicated it is. Thankyou for all the ideas
     
  15. Dent1

    Joined:
    May 18, 2010
    Messages:
    3,158 (1.95/day)
    Thanks Received:
    915
    If you're just playing old games like Age of Empires III from pre-2006, you don't need a fancy rig, anything can run it. Integrated video from the Trinity APU is enough. In all honesty if you build around an I5 or FX 6 core with an 1GB 7770 or better you'll be wanting to play modern games otherwise you'll be wasting the build as it'll be capable of playing todays games reasonably well.

    I think you should challenge yourself and attempt to build your own. There is plenty of tutorials on the web, youtube videos, and help on this forum.
     
    Dovah88 says thanks.
  16. Dovah88 New Member

    Joined:
    Apr 2, 2013
    Messages:
    24 (0.04/day)
    Thanks Received:
    1
    Ok thankyou, so it would be better if i bought all the components my self and put it all in the case and evrything?
    comp is about 6 years old now and the psu that came with the case is still working so they ain't that bad.
    thats y i am updating
     
    Last edited: Apr 4, 2013
  17. LDNL

    LDNL

    Joined:
    Jan 25, 2011
    Messages:
    465 (0.34/day)
    Thanks Received:
    106
    Location:
    Finland
    What kind of PSU is it (Company and model name)?
    E: Nvm just googled it. It has all necessary cables and is quite capable of running even a higher end configuration.
    And yes it would be better that you build the pc yourself. You would save money and learn a thing or two. It will always be hard and confusing for the first time but as you take the old computer apart, you'll see where all the cables go. The front I/O (power button, reset, usb and audio jacks) might be abit tricky but thats what the motherboards manual is for. Some times they are even labeled on the actual motherboard. But when its complite and you boot it up for the first time successfully, it's that feeling of accomplishment that will make you want to build another one as soon as possible.
     
    Last edited: Apr 4, 2013
    Dovah88 says thanks.
  18. Dovah88 New Member

    Joined:
    Apr 2, 2013
    Messages:
    24 (0.04/day)
    Thanks Received:
    1
    Ok thankyou i will get the parts soon or anytime and figure it out just as long as it all fits in well and will be much more of an upgrade than the one i got now:toast:
     
    Dent1 says thanks.
  19. Dent1

    Joined:
    May 18, 2010
    Messages:
    3,158 (1.95/day)
    Thanks Received:
    915
    One last thing I want to add.

    Before you buy any components make sure somebody in the forum checks over it. Post the website and links too so we can see if you're getting a good deal.
     
  20. Frag Maniac

    Frag Maniac

    Joined:
    Nov 9, 2010
    Messages:
    2,643 (1.83/day)
    Thanks Received:
    549
    If you're referring to the Anandtech and Kitguru benches I posted, I didn't see them mention at all what part of the game was being tested. Are you guessing?
     
  21. crazyeyesreaper

    crazyeyesreaper Chief Broken Rig

    Joined:
    Mar 25, 2009
    Messages:
    8,162 (4.00/day)
    Thanks Received:
    2,805
    Location:
    04578
    No im not guessing

    CPU bench = 10-40 fps

    GPU bench is 50-150 depending on card or setup its not rocket science. (fixed it :roll)

    CPU bench


    Graphics bench
     
    Last edited by a moderator: Nov 20, 2013
  22. Frag Maniac

    Frag Maniac

    Joined:
    Nov 9, 2010
    Messages:
    2,643 (1.83/day)
    Thanks Received:
    549
    In other words based on what you've seen of in-game frame rates with spec you or others have tested on without really knowing what part of the game was tested, pretty much like I implied.

    Not saying you're wrong, just sounds like too much blind speculation to be sure. That game series has too broad a result FPS wise and obviously hardware used can vary quite a bit too.

    And btw, you made a Freudian slip on the "rocket science" part.
     
  23. crazyeyesreaper

    crazyeyesreaper Chief Broken Rig

    Joined:
    Mar 25, 2009
    Messages:
    8,162 (4.00/day)
    Thanks Received:
    2,805
    Location:
    04578
    uh no, everyone runs the benchmark if your to stupid to understand the difference thats your problem

    the game has a built in CPU bench and Built in DX11 1080p and 720p benchmark it also has custom benchmark settings by using In game settings,

    DX11 1080p is the normal ran in tests, Andantech posting say ULTRA settings is just everything turned on in game graphics menu and then use the Bench game settings option

    [​IMG]

    The results are not all over the place,

    The CPU has to handle all Animation / Physics / 1 on 1 battle sync on 1 CPU core AMD fails in this regard.

    In my own testing using large scale battles aka 2 vs 2 thats 20 units + 20 units vs 20 units + 20 units in battle which does happen in game is on average anywhere from 3000-5000 vs 3000-5000 men, when join in melee the CPU must handle the 1 on 1 fighting between all individual soldiers while also calculating overall unit statistics

    So unlike most games if you have a 7850 2gb or better your CPU will be the limiting factor.

    One thing I have noticed with more recent patches is the game will drop CPU handled animations frames while keeping graphics running, this means the game can run at 60 fps according to say the benchmark but the animations are running at 15-16 fps causing

    Originally the game would DROP to 15-16 fps solid no matter the GPU in huge melee battles unless on Intel ivy or sandy bridge chip. now FPS will stay higher but it drops animation frames regardless its extremely distracting depending on the system and hardware.

    Easiest way to avoid it period is to run Intel i5 set shadows to medium, because on Ultra graphics settings all soldiers bodies remain on screen, those dead soliders are still technically animated, aka it has to keep the dead body animation in sync with the rest of the engine, setting Unit detail to High causes some bodies to dissappear.

    Regardless watch the two videos you will see the difference I am talking about, large melee still lags on 90% of systems. Its an inherent flaw in the game engine, and it wont change going forward to Rome II.

    The Warscape engine used in Empire / Napoleon / Shogun 2 is ment for GUN POWDER era aka guys standing still shooting muskets, so no heavy animation load on the CPU trying to sync all these fighters. Shogun 2 is heavily melee Rome II will be as well.

    Older games like Medieval 2 and Rome can also handle extreme unit sizes and battles but its also single core with less optimization aka text files etc used for making things work. In this regard Intel also comes out ahead.

    No matter the GPU I have used playing these games CPU has always been the limitation 8800 GTX 640 in Rome / 4870x2 5850 xfire 6970s 6950s 7970 gtx 670 my frame rate is more often than not limited by the CPU during battles.

    Basically Minimum frame rate is entirely dependent on the CPU used and the worse the CPU more often the game stutters and hitches even with todays best GPUs
     
    Last edited: Apr 4, 2013
  24. Frag Maniac

    Frag Maniac

    Joined:
    Nov 9, 2010
    Messages:
    2,643 (1.83/day)
    Thanks Received:
    549
    If you can't engage in a discussion without childish flames, maybe I should ask the staff what they think. This is the kind of skepticism that results when you make blanket statements that are assumptions based on assumed testing methods and hardware.

    Nothing wrong with my IQ kid, and I don't need to talk like a brat to prove it. If anything you're only making yourself sound like the dime a dozen intellectual wannabe teenagers on the net talking that way.
     
    Dent1 says thanks.
  25. crazyeyesreaper

    crazyeyesreaper Chief Broken Rig

    Joined:
    Mar 25, 2009
    Messages:
    8,162 (4.00/day)
    Thanks Received:
    2,805
    Location:
    04578
    How are my statements blanket statements? everyone runs the regular benchmark few run the CPU bench, the graphics benchmark is essentially worthless yes X GPU can render 60-100 fps in the game, Will it once battle joins? no it wont. CPU bench is included for that reason the bench isnt even the worst case senario anymore with the push to 40 units vs 40 units.
     

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

Share This Page