Overview Features Coding Performance Forum Downloads Products OrderV4 Contact

Welcome to the Apollo Forum

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



All TopicsNewsPerformanceGamesDemosApolloVampireAROSWorkbenchATARIReleases
Performance and Benchmark Results!

APOLLO SILVER2 Candidate Minibench Result

Gunnar von Boehn
(Apollo Team Member)
Posts 5413
20 Feb 2016 13:20


APOLLO SILVER2 candidate Minibench result
 
 

  -----------------------------------------------------------
  Processor & Memory Performance Bench v6.09
  -----------------------------------------------------------
  -----------------------------------------------------------
  Comparing different CPU functions:
  Results are in million instructions per sec.
  Higher value is faster.
  ALU-Instructions 16MB 64KB
  -----------------------------------------------------------
  nop        142.7 142.7
  add-reg    142.8 142.7
  add-im16  142.8 142.6
  shift-reg  142.8 142.7
  shift-im8  142.8 142.6
  and-reg    142.7 142.7
  and-im16.  142.8 142.6
  mul-reg    40.4  40.4
  mul-im16    40.4  40.4
  div-reg      2.4  2.4
  -----------------------------------------------------------
  ALU-Workloads 16MB 64KB
  -----------------------------------------------------------
  Workload AAAA  150.7 150.5
  Workload LA    150.7 150.5
  Workload LAA    150.6 150.4
  Workload LAAA  150.7 150.5
  Workload LAAAA  150.7 150.5
  Workload LLA    116.3 116.1
  Workload LLAA  150.7 150.5
  Workload LLAAA  127.1 127.0
  Workload LLAAAA 150.7 150.5
  Workload LAALA  150.7 150.5
  WORKLOAD addq  150.7 150.4
  WORKLOAD add.w  150.7 150.5
  WORKLOAD add.l  150.7 150.5
  WORKLOAD lea    150.7 150.5
  Work 26        150.7 150.5
  Work 2266      150.7 150.5
  Work 4x2 4x6    150.7 150.5
  Work 8x2 8x6    150.7 150.5
  WORK memadd      81.3  81.3
  WORK memadd++    79.8  79.7
  -----------------------------------------------------------
  ALU-Addressmodes 16MB 64KB
  -----------------------------------------------------------
  LOAD ea-im(r)    79.7 79.7
  LOAD ea-(r)++    77.5 77.5
  LOAD ea-im(r)+A  77.5 77.5
  LOAD ea-(r,r)    77.5 77.4
  write ea-im(r)  78.5 78.5
  write ea-(r)++  76.3 76.2
  write ea-im(r)A  76.2 76.2
  write ea-(r,r)  76.3 76.2
  cache read odd  79.8 79.7
  cache write odd  41.7 41.7
  -----------------------------------------------------------
  ALU-Branches 16MB 64KB
  -----------------------------------------------------------
  gosub chain-1    5.2  5.2
  gosub chain-4    7.7  7.7
  gosub chain-8    6.7  6.7
  gosub chain-16  6.3  6.3
  gosub chain-32  6.2  6.2
  BCC IF(FALSE)  39.1 39.1
  BCC IF(TRUE)    39.1 39.1
  BCC IF(MIXA)    39.1 39.1
  BCC IF(MIXB)    39.1 39.1
  goto-bra        35.7 35.6
  goto-bra + 2A  25.1 25.1
  goto-bra + 4A  19.3 19.3
  -----------------------------------------------------------
  ALU-Latencies 16MB 64KB
  -----------------------------------------------------------
  add-reg-2loop    42.3  42.3
  add-reg-4loop    67.8  67.8
  add-reg-6loop    84.8  84.7
  add-reg-8loop    96.9  96.9
  add-reg-16loop  123.3 123.3
  add-reg-32loop  142.8 142.7
  add-reg-64loop  155.0 155.0
  add-reg-128loop 162.0 161.9
  alu_latency+0    78.2  78.2
  alu_latency+1  135.6 135.5
  alu_latency+2  135.6 135.5
  cache_latency+0  71.4  71.3
  cache_latency+1  50.2  50.2
  cache_latency+2  38.7  38.7
  alu->ea_lat+0    71.3  71.3
  alu->ea_lat+1    38.7  38.7
  alu->ea_lat+2    38.7  38.7
  -----------------------------------------------------------
  Measuring memory latency:
  Result is million random accesses per sec.
  Higher value is faster.
  Memory Latency 16MB
  -----------------------------------------------------------
  rand-read 2MB  2.9
  rand-read 4MB  2.9
  rand-read 8MB  2.9
  rand-read 16MB  2.9
  -----------------------------------------------------------
  Measuring memory throughput:
  Results are in MB/sec. Higher value is faster.
  Memory 2 Memory
  Alignment 0-0 16MB 64KB 4KB 1KB
  -----------------------------------------------------------
  libc memcpy  77.2  77.1  77.1  76.0
  read 8        28.2  28.2  28.2  28.1
  read 8x4      56.4  56.4  56.3  55.8
  read 32      112.9 112.9 112.8 111.7
  read 32x4    221.1 220.8 219.4 213.9
  write 8      28.1  28.1  28.1  28.1
  write 8x4    37.4  37.4  37.2  37.2
  write 32    112.6 112.6 112.7 112.8
  write 32x4  222.9 222.9 222.4 220.7
  copy 8        31.8  31.8  31.8  31.7
  copy 8x4      62.3  62.3  62.2  61.8
  copy 32      108.1 108.0 107.6 106.5
  copy 32x4    136.9 164.1 167.4 163.9
  -----------------------------------------------------------
  Cache 2 Cache
  Alignment 0-0 16MB 64KB 4KB 1KB
  -----------------------------------------------------------
  libc memcpy  77.2  77.2 134.7 132.1
  read 8        28.2  28.2  28.2  28.2
  read 8x4      56.5  56.4  56.4  56.2
  read 32      112.9 112.8 113.0 113.2
  read 32x4    221.1 220.6 225.4 224.2
  write 8      28.1  28.1  28.1  28.1
  write 8x4    37.4  37.4  37.4  37.2
  write 32    112.6 112.6 112.7 112.9
  write 32x4  222.9 222.9 222.5 220.9
  copy 8        31.8  31.8  33.8  33.8
  copy 8x4      62.4  62.3  67.3  66.8
  copy 32      108.1 107.9 135.2 134.0
  copy 32x4    136.6 162.3 267.1 259.2
  -----------------------------------------------------------
 


posts 1