NVIDIA GeForce GTX TITAN X OpenCL performance leaks out

Published: 13th Mar 2015, 07:23 GMT

While Jen Hsun is almost done rehearsing his speech for GTC about Maxwell GM200, we are looking for clues about Big Maxwell’s performance.

With no surprise we found TITAN X in CompuBench database. It will probably be removed by the time most of you will see this post, so we took few screengrabs, and posted the full opencl info dump, just in case.

According to new information from CompuBench, TITAN X has 24 compute units, which basically means it has 3072 CUDA cores. The software also reports on maximum clock, in other words boost clock, of 1076 MHz.

TITAN X outperforms the competition in every benchmark. Just remember that CompuBench top list is using median value from all entries, so some cards can actually perform better (for instance, if someone is using overclocking).

I’m leaving the interpretation to you, OpenCL benchmark has no value for gamers, but some of you might still find it interesting.

NVIDIA GeForce GTX TITAN X performance in CompuBench NVIDIA GeForce GTX TITAN X Video Composition NVIDIA GeForce GTX TITAN X Particle Simulation 64k NVIDIA GeForce GTX TITAN X Face Detection NVIDIA GeForce GTX TITAN X Face Detection Optical Flow NVIDIA GeForce GTX TITAN X Face Detection Ocean Surface Simulation

NVIDIA GeForce GTX TITAN X OpenCL
NameValue
address_bits32
compiler_available1
double_fp_configCL_FP_DENORM CL_FP_INF_NAN CL_FP_ROUND_TO_NEAREST CL_FP_ROUND_TO_ZERO CL_FP_ROUND_TO_INF CL_FP_FMA
driver_version347.84
endian_little1
error_correction_support0
execution_capabilitiesCL_EXEC_KERNEL
extensionscl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_d3d9_sharing cl_nv_d3d10_sharing cl_khr_d3d10_sharing cl_nv_d3d11_sharing cl_nv_copy_opts 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_fp64
global_mem_cache_size393216
global_mem_cache_typeCL_READ_WRITE_CACHE
global_mem_cacheline_size128
global_mem_size0
host_unified_memory0
image_support1
image2d_max_height32768
image2d_max_width32768
image3d_max_depth4096
image3d_max_height4096
image3d_max_width4096
local_mem_size49151
local_mem_typeCL_LOCAL
max_clock_frequency1076    <- IMPORTANT
max_compute_units24   <- IMPORTANT (24 * 128 = 3072)
max_constant_args9
max_constant_buffer_size65536
max_mem_alloc_size3.22E+09
max_parameter_size4352
max_read_image_args256
max_samplers32
max_work_group_size1024
max_work_item_dimensions3
max_work_item_sizes1024 1024 64
max_write_image_args16
mem_base_addr_align4096
min_data_type_align_size128
nameGeForce GTX TITAN X
native_vector_width_char1
native_vector_width_double1
native_vector_width_float1
native_vector_width_half0
native_vector_width_int1
native_vector_width_long1
native_vector_width_short1
opencl_c_versionOpenCL C 1.1
preferred_vector_width_char1
preferred_vector_width_double1
preferred_vector_width_float1
preferred_vector_width_half0
preferred_vector_width_int1
preferred_vector_width_long1
preferred_vector_width_short1
profileFULL_PROFILE
profiling_timer_resolution1000
queue_propertiesCL_QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE CL_QUEUE_PROFILING_ENABLE
single_fp_configCL_FP_DENORM CL_FP_INF_NAN CL_FP_ROUND_TO_NEAREST CL_FP_ROUND_TO_ZERO CL_FP_ROUND_TO_INF CL_FP_FMA
typeCL_DEVICE_TYPE_GPU
vendorNVIDIA Corporation
versionOpenCL 1.1 CUDA

Source: CompuBench


by WhyCry

Previous Post
Inno3D Announces iChill GeForce GTX 980 and GTX 970
Next Post
NVIDIA GeForce GTX TITAN X pictured with GM200-400 GPU






Back to Top ↑