Welcome to Keen Software House Forums! Log in or Sign up to interact with the KSH community.
  1. You are currently browsing our forum as a guest. Create your own forum account to access all forum functionality.

CPU choice i5 9400F or Ryzen 2600 ?

Discussion in 'General' started by mleise, Apr 16, 2019.

Thread Status:
This last post in this thread was made more than 31 days old.
  1. mleise Trainee Engineer

    Messages:
    26
    i5 9400F:
    + can boost clock of an individual core very high
    + offers good performance without overclocking
    - no hyperthreading
    - upgrade probably requires new RAM and mainboard

    Ryzen 5 2600:
    + compatible with next gen of AMD CPUs (upgradable)
    + good workstation performance
    o after market cooler, overclocking and 3200 Mhz DDR4 makes it competitive
    - sup par single-core game performance

    So basically as long as a game used one thread at 100% and some more CPU cores at low utilization, the Intel will be perfect. On the other hand, if the game uses more than 6 threads with a high utilization the AMD will catch up and overtake. As some pointed out, simulating one large ship may favor high single-core performance, while simulating lots of small ships may favor high-thread-count CPUs. Has any of the players or devs done testing with both options? Is there any definite answer?

    Maybe I should formulate it differently: When the FPS drop, and the CPU is the bottleneck, how many CPU cores are actually highly loaded?
     
    Last edited: Apr 16, 2019
  2. Roxette Senior Engineer

    Messages:
    1,396
  3. Ronin1973 Master Engineer

    Messages:
    4,777
    The Ryzen has a larger cache and is unlocked. It also supports hyperthreading.

    Space Engineers uses more than one core but the majority of the action happens on one primary core.

    I think the stock i5 is slightly better in testing than the AMD in gaming. But that margin closes on higher settings where the i5 seems to bottleneck. What I don't see testing for would be games with heavy physics engine usage (like SE). That's where a larger cache will be most helpful as any information that is cached on the chip instead of held in RAM will execute faster.
    https://cpu.userbenchmark.com/Compare/Intel-Core-i5-9400F-vs-AMD-Ryzen-5-2600/m699058vs3955
    https://www.techspot.com/review/1829-intel-core-i5-9400f-vs-amd-ryzen-5-2600x/
     
  4. mleise Trainee Engineer

    Messages:
    26
    Thanks, didn't know about the TechSpot review. Though even without extra OC cooler you'd currently pay 150€ extra for the 2600X setup in the test. (CL15 3400 Mhz DDR4 is quite expensive.) Could bring it down to 40€ extra with CL16 3333 Mhz if I find a mainboard that supports HX433C16PB3K2/16.
     
  5. mojomann71 Senior Engineer

    Messages:
    1,764
    Either one you choose, do not be dishearted if SE still does not render to your liking. A couple of years back I did the same wanting to get the best that I could so I could see SE in "all its Glory" and there was not much of a difference. :) Other games I had greatly improved though!
     
  6. mleise Trainee Engineer

    Messages:
    26
    That was probably before 2017's multi-threaded Havok update. And yeah you really need a recent gfx card for SE before all. The recommended minimum was tested by w4stedspace back then, lol:
     
  7. mleise Trainee Engineer

    Messages:
    26
    So I'm using that i5 9400F + SAPPHIRE Pulse Vega 56 now and it runs super smooth in 1440p @ 60 FPS with everything on high, except for the shadow maps (performance killer).
     
    Last edited: May 15, 2019
Thread Status:
This last post in this thread was made more than 31 days old.