Overview Features Instructions Performance Forum Downloads Products Reseller Contact

Welcome to the Apollo Forum

This forum is for people interested in the APOLLO CPU.
Please read the forum usage manual.
VISIT APOLLO IRC CHANNEL



All TopicsNewsPerformanceGamesDemosApolloVampireCoffinReleases
Performance and Benchmark Results!

Atari Powapage  1 2 

Olivier Landemarre

Posts 46
14 Apr 2018 12:21


Olivier Landemarre wrote:

Gunnar von Boehn wrote:

 
Olivier Landemarre wrote:

    Kronos is 68K but test are fully coldfire, as tests are inside dynamic library.
   

   
    Ah yes I found the Firebee results.
    And the test code is indeed Coldfire "native/clean"
   
   
 
                Coldfire    68080      Milan-68040
    CPU_BitShift  =  265      198        12
    CPU_RawPower  =  268      200        24
   
    CPU_MEMOREAD  =    83      534        34
    CPU_MEMOWRITE =    47      663        21 
    -----------------------------------------
   

   
    pretty interesting
   
   
   
 

 
  Results for firebee are quite old, I'm going to check if they do some progress since, nobody was able to understand why memory access was so low while processor speed is quite fast if run in cache, it was the same when I was using coldfire evaluation board from freescale, perhaps some wrong cache management.
  On several 68060 systems results could change a lot with configuration, I know a lot of that have used Kronos on CT60 just to fix the configuration up they was able to acheive reference results!
 
 
 
 
 

 
                Coldfire    68080      Milan-68040 CT60/100
    CPU_BitShift  =  265      198        12              199
    CPU_RawPower  =  268      200        24              199
   
    CPU_MEMOREAD  =    83      534        34              101
    CPU_MEMOWRITE =    47      663        21              101
    -----------------------------------------
   

Bitshift - Rawpower test are directly link to frequency you have test on 100Mhz FPGA??? or there is something wrong in Emutos 200 hz Timer?

Memoread-write is quite incredible you have optimized for Kronos?!!! Incredible results, you are incredible! In some month so many improvement!

Very Very nice




Markus (mfro)

Posts 89
14 Apr 2018 16:07


Olivier Landemarre wrote:

Kronos is 68K but test are fully coldfire, as tests are inside dynamic library.

Good to know, I wasn't aware of that.

Memory throughput is indeed the weak point of the MCF5475. I think I read somewhere that the memory controller is a cut-down PowerPC offspring. If you dig in old NXP forum posts, there are lots of people complaining about it with no real solution.

According to the manual, cache settings and memory controller configuration in the FireBee is correct.


Peter Slegg

Posts 15
14 Apr 2018 16:16


Some nice incrememntal improvements.  It would be great to see the last 680x0 features (mmu) implemented so that they Apollo could be used as a replacment on more platforms.  Sooner or later the fpgas will get faster as well so we could see big jumps in performance in the future.


Vojin Vidanovic

Posts 341
14 Apr 2018 16:27


Peter Slegg wrote:

Some nice incrememntal improvements.  It would be great to see the last 680x0 features (mmu) implemented so that they Apollo could be used as a replacment on more platforms. 

Agreed. Hope V4 space will fit at least some sort of Motorola MMU compatibility OR patch existing to be used so, so that Linux and MacOS native box could be opened in similar fashion as Atari. It would make Vamps most interesting "futuristic retro boxes".


posts 24page  1 2