C-60 APU and Radeon HD6920

[Raistmer]

With preinstalled OS (Win7 x64 home premium) there is no OpenCL support on GPU (but Direct Copmute 5.0 supported).
I tried to find update on AMD site but highest APU listed there is C-50. So I download "11.12 mobility drivers" for Win7 x64.
Will see if this driver can be used for my hardware config and will it provide OpenCL support on this netbook or not.

[Raistmer]

My hopes are fulfilled, this APU-based netbook can do OpenCL indeed!

WU : PG1327.wu
MB7_win_x86_SSE2_CPU_r390.exe -period_iterations_num 1 -hp :
  Elapsed 1333.506 secs
      CPU 1303.108 secs
MB7_win_x86_SSE3_OpenCL_ATi_HD5_r390.exe -period_iterations_num 1 -hp :
  Elapsed 401.514 secs
      CPU 257.604 secs
MB7_win_x86_SSE3_OpenCL_ATi_LHD4K_r390.exe -period_iterations_num 1 -hp :
  Elapsed 415.554 secs
      CPU 296.839 secs
MB7_win_x86_SSE3_OpenCL_ATi_r390.exe -period_iterations_num 1 -hp :
  Elapsed 359.705 secs
      CPU 245.873 secs
MB7_win_x86_SSSE3x_CPU_r390.exe -period_iterations_num 1 -hp :
  Elapsed 0.047 secs
      CPU 0.016 secs

From the other side it can't do SSSE3 (at least, Intel's flavour one).

Build features: SETI7   Non-graphics   OpenCL   USE_OPENCL_HD5xxx   IPP   AMD specific   USE_SSE3   x86   
     CPUID: AMD C-60 APU with Radeon(tm) HD Graphics

     Cache: L1=64K L2=512K

CPU features: FPU TSC PAE CMPXCHG8B APIC SYSENTER MTRR CMOV/CCMP MMX FXSAVE/FXRSTOR SSE SSE2 HT SSE3
CPU type 0x43
Number of OpenCL platforms:             1


 OpenCL Platform Name:                AMD Accelerated Parallel Processing
Number of devices:             1
  Max compute units:             2
  Max work group size:             256
  Max clock frequency:             275Mhz
  Max memory allocation:          175374336
  Cache type:                None
  Cache line size:             0
  Cache size:                0
  Global memory size:             268435456
  Constant buffer size:             65536
  Max number of constant args:          8
  Local memory type:             Scratchpad
  Local memory size:             32768
  Queue properties:            
    Out-of-Order:             No
  Name:                   Loveland
  Vendor:                Advanced Micro Devices, Inc.
  Driver version:             CAL 1.4.1646 (VM)
  Version:                OpenCL 1.1 AMD-APP (831.4)
  Extensions:                cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_popcnt cl_khr_d3d10_sharing

[Raistmer]

GPU clocks look strange:

[Raistmer]

when GPU idle it rises to 400MHz but with busy GPU it drops to 275MHz



Maybe power usage restriction applies?
it's netbook after all...

[Mike]

GPU temperature seems a little bit high to me.

[Raistmer]

GPU temperature seems a little bit high to me.

It reached 90C during benchmark...

[Raistmer]

Hm... reciving very controversial results in total benchmark currently running on this device.
For example SSE3_INTEL binary runs while SSE2 fails (!).
Mike, could you run similar total benchmark (all available AKV8b2 apps x86+x64) on your FX CPU to compare our findings ?

[Mike]

Hm... reciving very controversial results in total benchmark currently running on this device.
For example SSE3_INTEL binary runs while SSE2 fails (!).
Mike, could you run similar total benchmark (all available AKV8b2 apps x86+x64) on your FX CPU to compare our findings ?

Will do when i´m back from work.
Do you want full PG set or is 1 WU enough.

Mike

[Raistmer]

think 1 WU would be ebough for now.

[Mike]

Finnished my benchrun on the FX CPU with all akv8b2 versions.

I have to say i am surprised.

Only 3 versions are functioning.
SSE3_AMD, SSE3_INTEL and SSE4.1
No x64 version is working.

SSE3_INTEL is faster than SSE3_AMD version 
Will run a full PG bench with both SSE3 versions to make sure.

Mike

[Raistmer]

Fine (if it can be fine at all) you just confirmed my own observations!
Same distribution here (I tried SSE-nly build too also, it works).

EDIT: no, not exactly the same. I have SSE2 not working too and it works on your CPU...
But all x64 fail here too...

[Mike]

Fine (if it can be fine at all) you just confirmed my own observations!
Same distribution here (I tried SSE-nly build too also, it works).

EDIT: no, not exactly the same. I have SSE2 not working too and it works on your CPU...
But all x64 fail here too...

No SSE2 dont work here also.

Only 3 versions are functioning.
SSE3_AMD, SSE3_INTEL and SSE4.1

[Mike]

Here is my full bench SSE3_AMD vs SSE3_Intel.
Intel version is faster on each angle range. 

AK_v8b2_win_SSE3_AMD.exe -verb -nog / PG0009.wu :
AppName: AK_v8b2_win_SSE3_AMD.exe
AppArgs: -verb -nog
TaskName: PG0009.wu
Started at  : 20:28:36.375
Ended at    : 20:35:54.376
    437.939 secs Elapsed
    431.670 secs CPU time

AK_v8b2_win_SSE3_INTEL.exe -verb -nog / PG0009.wu :
AppName: AK_v8b2_win_SSE3_INTEL.exe
AppArgs: -verb -nog
TaskName: PG0009.wu
Started at  : 20:35:58.245
Ended at    : 20:43:11.676
    433.368 secs Elapsed
    426.881 secs CPU time

AK_v8b2_win_SSE3_AMD.exe -verb -nog / PG0395.wu :
AppName: AK_v8b2_win_SSE3_AMD.exe
AppArgs: -verb -nog
TaskName: PG0395.wu
Started at  : 20:43:15.513
Ended at    : 20:50:10.863
    415.288 secs Elapsed
    409.643 secs CPU time

------------
AK_v8b2_win_SSE3_INTEL.exe -verb -nog / PG0395.wu :
AppName: AK_v8b2_win_SSE3_INTEL.exe
AppArgs: -verb -nog
TaskName: PG0395.wu
Started at  : 20:50:14.732
Ended at    : 20:56:58.086
    403.291 secs Elapsed
    397.771 secs CPU time

AK_v8b2_win_SSE3_AMD.exe -verb -nog / PG0444.wu :
AppName: AK_v8b2_win_SSE3_AMD.exe
AppArgs: -verb -nog
TaskName: PG0444.wu
Started at  : 20:57:01.892
Ended at    : 21:02:46.090
    344.136 secs Elapsed
    339.521 secs CPU time

AK_v8b2_win_SSE3_INTEL.exe -verb -nog / PG0444.wu :
AppName: AK_v8b2_win_SSE3_INTEL.exe
AppArgs: -verb -nog
TaskName: PG0444.wu
Started at  : 21:02:49.866
Ended at    : 21:08:18.339
    328.411 secs Elapsed
    324.139 secs CPU time

AK_v8b2_win_SSE3_AMD.exe -verb -nog / PG1327.wu :
AppName: AK_v8b2_win_SSE3_AMD.exe
AppArgs: -verb -nog
TaskName: PG1327.wu
Started at  : 21:08:22.177
Ended at    : 21:13:09.638
    287.399 secs Elapsed
    283.189 secs CPU time

AK_v8b2_win_SSE3_INTEL.exe -verb -nog / PG1327.wu :
AppName: AK_v8b2_win_SSE3_INTEL.exe
AppArgs: -verb -nog
TaskName: PG1327.wu
Started at  : 21:13:13.491
Ended at    : 21:17:48.753
    275.200 secs Elapsed
    270.443 secs CPU time

[Jason G]

Intel version is faster on each angle range. 

That's great news!, probably means the caching & SSE3 implementation is improved enough in that model to negate the need for a special AMD build with the earlier libraries.  That probably means 1 less build already, if we can manage to statically link in the libs needed by the older AMD chips    I think the Instruction set convergence will end up making things substantially easier, so we can get back to optimisation instead of build & library juggling.  Here's hoping 

[Raistmer]

I would not talk about any convergence after these tests done. Looks like divergence much more adequate term...
Look for number of builds that DON'T run on these new AMD chips...
Whole x64 AKv8b2 set fails to even report error...