100% CPU load in Jitsi Meet conference

jochen01

VB-74672

Hi,

I noticed, that Jitsi Meet video conference in Vivaldi raises CPU usage to about 450%. Same conference with same settings in firefox uses just about 80% CPU.

I already enabled the experimental settings dealing with HW acceleration. From vivaldi://gpu it seems to be quite o.k.:

Graphics Feature Status
Canvas: Hardware accelerated
Compositing: Hardware accelerated
Multiple Raster Threads: Enabled
Out-of-process Rasterization: Hardware accelerated
OpenGL: Enabled
Hardware Protected Video Decode: Hardware accelerated
Rasterization: Hardware accelerated on all pages
Skia Renderer: Enabled
Video Decode: Hardware accelerated
Vulkan: Disabled
WebGL: Hardware accelerated
WebGL2: Hardware accelerated

How can I fix it?
It's Vivaldi 3.4.2066.106 (Stable channel) (64-Bit) for Linux on Intel Core i5 notebook (kUbuntu 20.04 if that matters).

--
ModEdit: Added VB#

Pathduck

@jochen01 Hi, how are you getting these numbers? The Vivaldi task manager gives CPU usage as accumulated over the logical processors/threads. So if you have 6 logical processors 450% would be 75%.

The Firefox task manager does not give processor % (AFAIK) only a value of low/medium/high.

I already enabled the experimental settings dealing with HW acceleration.

What setting would this be? In vivaldi://experiments/ there is no setting related to Hw acceleration.

jochen01

@Pathduck said in 100% CPU load in Jitsi Meet conference:

@jochen01 Hi, how are you getting these numbers? The Vivaldi task manager gives CPU usage as accumulated over the logical processors/threads. So if you have 6 logical processors 450% would be 75%.

Accumulated load over all cores. If you count the sum of all cores as 100% the load is about 80..90%. Far too high for low resolution video decoded by hardware.

The Firefox task manager does not give processor % (AFAIK) only a value of low/medium/high.

For measurement of CPU load I don't use tools of the browser. For that I usually use htop.

What setting would this be? In vivaldi://experiments/ there is no setting related to Hw acceleration.

See settings in vivaldi://flags (or chrome://flags). There I've set

Override software rendering list --> enabled
GPU rasterization --> enabled
Out of process rasterization --> enabled
Hardware-accelerated video decode --> enabled
WebGL Draft Extensions --> enabled
YUV Decoding for WebP --> enabled

That's a try to improve (lower) CPU usage. No success so far.

mib2berlin

@jochen01
Hi, does your laptop have a discrete GPU or Intel integrated?
I have an old i7 with HD4000 GPU and HW acceleration is working.
I would not play with Flags until it is clear your card is supported at all.
Please set to defaults and post output of vivaldi://gpu marked with </> button for better readability.

Cheers, mib

TbGbe · Win: Snapshot Vivaldi 7.4.3671.3

@jochen01 said in 100% CPU load in Jitsi Meet conference:

Override software rendering list --> enabled
Hardware-accelerated video decode --> enabled

If this means that it was disabled, perhaps there was a reason your GPU was in the list?

jochen01

@mib2berlin It does not have a dedicated graphics chip. So, just Intel CPU graphics. lshw output (just display):

       *-display
             description: VGA compatible controller
             product: UHD Graphics
             vendor: Intel Corporation
             physical id: 2
             bus info: pci@0000:00:02.0
             version: 02
             width: 64 bits
             clock: 33MHz
             capabilities: vga_controller bus_master cap_list rom
             configuration: driver=i915 latency=0
             resources: irq:127 memory:8a000000-8affffff memory:80000000-87ffffff ioport:5000(size=64) memory:c0000-dfff

Output of vivaldi://gpu:

 
Graphics Feature Status
Canvas: Hardware accelerated
Compositing: Hardware accelerated
Multiple Raster Threads: Enabled
Out-of-process Rasterization: Disabled
OpenGL: Enabled
Hardware Protected Video Decode: Unavailable
Rasterization: Software only. Hardware acceleration disabled
Skia Renderer: Enabled
Video Decode: Software only. Hardware acceleration disabled
Vulkan: Disabled
WebGL: Hardware accelerated
WebGL2: Hardware accelerated
Driver Bug Workarounds
adjust_src_dst_region_for_blitframebuffer
clear_uniforms_before_first_program_use
decode_encode_srgb_for_generatemipmap
exit_on_context_lost
msaa_is_slow
scalarize_vec_and_mat_constructor_args
disabled_extension_GL_KHR_blend_equation_advanced
disabled_extension_GL_KHR_blend_equation_advanced_coherent
Problems Detected
Protected video decoding with swap chain is for certain Intel and AMD GPUs on Windows: 1093625
Disabled Features: protected_video_decode
Clear uniforms before first program use on all platforms: 124764, 349137
Applied Workarounds: clear_uniforms_before_first_program_use
Always rewrite vec/mat constructors to be consistent: 398694
Applied Workarounds: scalarize_vec_and_mat_constructor_args
On Intel GPUs MSAA performance is not acceptable for GPU rasterization: 527565
Applied Workarounds: msaa_is_slow
adjust src/dst region if blitting pixels outside framebuffer on Linux Intel: 664740
Applied Workarounds: adjust_src_dst_region_for_blitframebuffer
Disable KHR_blend_equation_advanced until cc shaders are updated: 661715
Applied Workarounds: disable(GL_KHR_blend_equation_advanced), disable(GL_KHR_blend_equation_advanced_coherent)
Decode and Encode before generateMipmap for srgb format textures on Linux Mesa ANGLE path: 634519
Applied Workarounds: decode_encode_srgb_for_generatemipmap
Some drivers can't recover after OUT_OF_MEM and context lost: 893177
Applied Workarounds: exit_on_context_lost
Accelerated video decode has been disabled, either via blocklist, about:flags or the command line.
Disabled Features: video_decode
ANGLE Features
disable_program_caching_for_transform_feedback (Frontend workarounds): Disabled: IsAndroid() && isQualcomm
On some GPUs, program binaries don't contain transform feedback varyings
lose_context_on_out_of_memory (Frontend workarounds): Enabled: true
Some users rely on a lost context notification if a GL_OUT_OF_MEMORY error occurs
scalarize_vec_and_mat_constructor_args (Frontend workarounds) 398694: Enabled: true
Always rewrite vec/mat constructors to be consistent
sync_framebuffer_bindings_on_tex_image (Frontend workarounds): Disabled: IsWindows() && isIntel
On some drivers TexImage sometimes seems to interact with the Framebuffer
add_and_true_to_loop_condition (OpenGL workarounds): Disabled: IsApple() && isIntel
Calculation of loop conditions in for and while loop has bug
adjust_src_dst_region_for_blitframebuffer (OpenGL workarounds) 830046: Enabled: IsLinux() || (IsAndroid() && isNvidia) || (IsWindows() && isNvidia)
Many platforms have issues with blitFramebuffer when the parameters are large.
allow_clear_for_robust_resource_init (OpenGL workarounds) 883276: Disabled: IsApple()
Using glClear for robust resource initialization is buggy on some drivers and leads to texture corruption. Default to data uploads except on MacOS where it is very slow.
always_call_use_program_after_link (OpenGL workarounds) 110263: Enabled: true
Always call useProgram after a successful link to avoid a driver bug
avoid_1_bit_alpha_texture_formats (OpenGL workarounds): Disabled: functions->standard == STANDARD_GL_DESKTOP && isAMD
Issue with 1-bit alpha framebuffer formats
avoid_dxt1_srgb_texture_format (OpenGL workarounds): Disabled: IsWindows() && isIntel
Replaces DXT1 sRGB with DXT1 sRGB Alpha as a driver bug workaround.
clamp_array_access (OpenGL workarounds) anglebug:2978: Disabled: IsAndroid() || isAMD || !functions->hasExtension("GL_KHR_robust_buffer_access_behavior")
Clamp uniform array access to avoid reading invalid memory.
clamp_frag_depth (OpenGL workarounds): Disabled: isNvidia
gl_FragDepth is not clamped correctly when rendering to a floating point depth buffer
clamp_msc_rate (OpenGL workarounds) 1042393: Disabled: IsLinux() && IsWayland()
Some drivers return bogus values for GetMscRate, so we clamp it to 30Hz
clamp_point_size (OpenGL workarounds): Disabled: IsAndroid() || isNvidia
The point size range reported from the API is inconsistent with the actual behavior
clear_to_zero_or_one_broken (OpenGL workarounds) 710443: Disabled: IsApple() && isIntel && GetMacOSVersion() < OSVersion(10, 12, 6)
Clears when the clear color is all zeros or ones do not work.
clip_src_region_for_blitframebuffer (OpenGL workarounds) 830046: Disabled: IsApple() || (IsLinux() && isAMD)
Issues with blitFramebuffer when the parameters don't match the framebuffer size.
decode_encode_srgb_for_generatemipmap (OpenGL workarounds) anglebug:4646: Disabled: IsApple() && functions->standard == STANDARD_GL_DESKTOP
Decode and encode before generateMipmap for srgb format textures.
disable_blend_func_extended (OpenGL workarounds) anglebug:1085: Enabled: isAMD || isIntel
ARB_blend_func_extended does not pass the tests
disable_draw_buffers_indexed (OpenGL workarounds): Disabled: IsWindows() && isAMD
Disable OES_draw_buffers_indexed extension.
disable_gpu_switching_support (OpenGL workarounds) 1091824: Disabled: isDualGPUMacWithNVIDIA
Disable GPU switching support (use only the low-power GPU) on older MacBook Pros.
disable_native_parallel_compile (OpenGL workarounds) 1094869: Disabled: isTSANBuild && IsLinux() && isNvidia
Do not use native KHR_parallel_shader_compile even when available.
disable_semaphore_fd (OpenGL workarounds) 1046462: Disabled: IsLinux() && isAMD && isMesa && mesaVersion < (std::array<int, 3>{19, 3, 5})
Disable GL_EXT_semaphore_fd extension
disable_timestamp_queries (OpenGL workarounds) 811661: Disabled: IsLinux() && isVMWare
Disable GL_EXT_disjoint_timer_query extension
disable_worker_contexts (OpenGL workarounds) 849576: Disabled: (IsWindows() && (isIntel || isAMD)) || (IsLinux() && isNvidia)
Some tests have been seen to fail using worker contexts
do_while_glsl_causes_gpu_hang (OpenGL workarounds) 644669: Disabled: IsApple() && functions->standard == STANDARD_GL_DESKTOP && GetMacOSVersion() < OSVersion(10, 11, 0)
Some GLSL constructs involving do-while loops cause GPU hangs
does_srgb_clears_on_linear_framebuffer_attachments (OpenGL workarounds): Enabled: functions->standard == STANDARD_GL_DESKTOP && (isIntel || isAMD)
Issue clearing framebuffers with linear attachments when GL_FRAMEBUFFER_SRGB is enabled
dont_initialize_uninitialized_locals (OpenGL workarounds) anglebug:2046: Disabled: IsAndroid() && isQualcomm
Initializing uninitialized locals caused odd behavior in a few WebGL 2 tests
dont_relink_programs_in_parallel (OpenGL workarounds) anglebug:3045: Disabled: IsAndroid() || (IsWindows() && isIntel)
Relinking a program in parallel is buggy
dont_use_loops_to_initialize_variables (OpenGL workarounds) 809422: Disabled: (IsAndroid() && isQualcomm) || (isIntel && IsApple())
For loops used to initialize variables hit native GLSL compiler bugs
emulate_abs_int_function (OpenGL workarounds) 642227: Disabled: IsApple() && isIntel
abs(i) where i is an integer returns unexpected result
emulate_atan_2_float (OpenGL workarounds) 672380: Disabled: isNvidia
atan(y, x) may return a wrong answer
emulate_copyteximage2d_from_renderbuffers (OpenGL workarounds) anglebug:4674: Disabled: IsApple() && functions->standard == STANDARD_GL_ES && !(isAMD && IsWindows())
CopyTexImage2D spuriously returns errors on iOS when copying from renderbuffers.
emulate_isnan_float (OpenGL workarounds) 650547: Disabled: isIntel && IsApple() && IsSkylake(device) && GetMacOSVersion() < OSVersion(10, 13, 2)
Using isnan() on highp float will get wrong answer
emulate_max_vertex_attrib_stride (OpenGL workarounds) anglebug:1936: Disabled: IsLinux() && functions->standard == STANDARD_GL_DESKTOP && isAMD
Some drivers return 0 when MAX_VERTEX_ATTRIB_STRIED queried
emulate_pack_skip_rows_and_pack_skip_pixels (OpenGL workarounds) anglebug:4849: Disabled: IsApple() && (isAMD || isIntel || isNvidia)
GL_PACK_SKIP_ROWS and GL_PACK_SKIP_PIXELS are ignored in Apple's OpenGL driver.
emulate_primitive_restart_fixed_index (OpenGL workarounds) anglebug:3997: Disabled: functions->standard == STANDARD_GL_DESKTOP && functions->isAtLeastGL(gl::Version(3, 1)) && !functions->isAtLeastGL(gl::Version(4, 3))
When GL_PRIMITIVE_RESTART_FIXED_INDEX is not available, emulate it with GL_PRIMITIVE_RESTART and glPrimitiveRestartIndex.
finish_does_not_cause_queries_to_be_available (OpenGL workarounds): Disabled: functions->standard == STANDARD_GL_DESKTOP && isNvidia
glFinish doesn't cause all queries to report available result
flush_before_delete_texture_if_copied_to (OpenGL workarounds) anglebug:4267: Disabled: IsApple() && isIntel
Some drivers track CopyTex{Sub}Image texture dependencies incorrectly. Flush before glDeleteTextures in this case
initialize_current_vertex_attributes (OpenGL workarounds): Disabled: isNvidia
During initialization, assign the current vertex attributes to the spec-mandated defaults
max_3d_array_texture_size_1024 (OpenGL workarounds) 927470: Disabled: limitMaxTextureSize
Limit max 3d texture size and max array texture layers to 1024 to avoid system hang
max_msaa_sample_count_4 (OpenGL workarounds) 797243: Disabled: IsAndroid() || (IsApple() && (isIntel || isAMD || isNvidia))
Various rendering bugs have been observed when using higher MSAA counts
max_texture_size_limit_4096 (OpenGL workarounds) 927470: Disabled: IsAndroid() || limitMaxTextureSize
Limit max texture size to 4096 to avoid frequent out-of-memory errors
pack_last_row_separately_for_padding_inclusion (OpenGL workarounds) anglebug:1512: Disabled: IsApple() || isNvidia
When uploading textures from an pack buffer, some drivers count an extra row padding
pack_overlapping_rows_separately_pack_buffer (OpenGL workarounds): Disabled: isNvidia
In the case of packing to a pixel pack buffer, pack overlapping rows row by row
pre_add_texel_fetch_offsets (OpenGL workarounds) 642605: Disabled: IsApple() && isIntel
Intel Mac drivers mistakenly consider the parameter position of nagative vaule as invalid even if the sum of position and offset is in range, so we need to add workarounds by rewriting texelFetchOffset(sampler, position, lod, offset) into texelFetch(sampler, position + offset, lod).
query_counter_bits_generates_errors (OpenGL workarounds) anglebug:3027: Disabled: IsNexus5X(vendor, device)
Drivers generate errors when querying the number of bits in timer queries
read_pixels_using_implementation_color_read_format (OpenGL workarounds) anglebug:4214: Disabled: functions->standard == STANDARD_GL_ES && functions->isAtLeastGLES(gl::Version(3, 1)) && functions->hasGLESExtension("GL_EXT_texture_norm16")
Quite some OpenGL ES drivers don't implement readPixels for RGBA/UNSIGNED_SHORT from EXT_texture_norm16 correctly
reapply_ubo_bindings_after_using_binary_program (OpenGL workarounds) anglebug:1637: Disabled: isAMD || IsAndroid()
Some drivers forget about UBO bindings when using program binaries
regenerate_struct_names (OpenGL workarounds) 403957: Disabled: IsApple()
All Mac drivers do not handle struct scopes correctly. This workaround overwrites a structname with a unique prefix.
remove_dynamic_indexing_of_swizzled_vector (OpenGL workarounds) 709351: Disabled: IsApple() || IsAndroid() || IsWindows()
Dynamic indexing of swizzled l-values doesn't work correctly on various platforms.
remove_invarient_and_centroid_for_essl3 (OpenGL workarounds): Disabled: functions->isAtMostGL(gl::Version(4, 1)) || (functions->standard == STANDARD_GL_DESKTOP && isAMD)
Fix spec difference between GLSL 4.1 or lower and ESSL3
reset_teximage2d_base_level (OpenGL workarounds) 705865: Disabled: IsApple() && isIntel && GetMacOSVersion() >= OSVersion(10, 12, 4)
Reset texture base level before calling glTexImage2D to work around pixel comparison failure.
reset_texture_generates_errors (OpenGL workarounds) anglebug:3859: Disabled: IsApple() || (IsWindows() && isAMD)
Calling glTexImage2D with zero size generates errors.
rewrite_float_unary_minus_operator (OpenGL workarounds) 308366: Disabled: IsApple() && isIntel && GetMacOSVersion() < OSVersion(10, 12, 0)
Using '-<float>' will get wrong answer
rewrite_repeated_assign_to_swizzled (OpenGL workarounds): Disabled: isNvidia
Repeated assignment to swizzled values inside a GLSL user-defined function have incorrect results
rewrite_row_major_matrices (OpenGL workarounds) anglebug:2273: Disabled: false
Rewrite row major matrices in shaders as column major as a driver bug workaround
rewrite_vector_scalar_arithmetic (OpenGL workarounds) 772651: Disabled: isNvidia
Certain types of GLSL arithmetic ops mixing vectors and scalars may be executed incorrectly
rgb_dxt1_textures_sample_zero_alpha (OpenGL workarounds) anglebug:3729: Disabled: IsApple()
Sampling BLACK texels from RGB DXT1 textures returns transparent black on Mac.
rgba4_is_not_supported_for_color_rendering (OpenGL workarounds): Enabled: functions->standard == STANDARD_GL_DESKTOP && isIntel
GL_RGBA4 is not color renderable
set_primitive_restart_fixed_index_for_draw_arrays (OpenGL workarounds) anglebug:3997: Disabled: features->emulatePrimitiveRestartFixedIndex.enabled && IsApple() && isIntel
Some drivers discard vertex data in DrawArrays calls when the fixed primitive restart index is within the number of primitives being drawn.
unfold_short_circuits (OpenGL workarounds) anglebug:482: Disabled: IsApple()
Mac incorrectly executes both sides of && and || expressions when they should short-circuit.
unpack_last_row_separately_for_padding_inclusion (OpenGL workarounds) anglebug:1512: Disabled: IsApple() || isNvidia
When uploading textures from an unpack buffer, some drivers count an extra row padding
unpack_overlapping_rows_separately_unpack_buffer (OpenGL workarounds): Disabled: isNvidia
In the case of unpacking from a pixel unpack buffer, unpack overlapping rows row by row
unsized_srgb_read_pixels_doesnt_transform (OpenGL workarounds) 565179: Disabled: IsAndroid() && isQualcomm
Drivers returning raw sRGB values instead of linearized values when calling glReadPixels on unsized sRGB texture formats
use_unused_blocks_with_standard_or_shared_layout (OpenGL workarounds): Disabled: (IsApple() && functions->standard == STANDARD_GL_DESKTOP) || (IsLinux() && isAMD)
Unused std140 or shared uniform blocks will be treated as inactive
vertex_id_does_not_include_base_vertex (OpenGL workarounds): Disabled: IsApple() && isAMD
gl_VertexID in GLSL vertex shader doesn't include base vertex value
Version Information
Data exported	2020-11-26T11:04:01.886Z
Chrome version	Chrome/86.0.4240.198
Operating system	Linux 5.4.0-54-generic
Software rendering list URL	https://chromium.googlesource.com/chromium/src/+/d8a506935fc2273cfbac5e5b629d74917d9119c7/gpu/config/software_rendering_list.json
Driver bug list URL	https://chromium.googlesource.com/chromium/src/+/d8a506935fc2273cfbac5e5b629d74917d9119c7/gpu/config/gpu_driver_bug_list.json
ANGLE commit id	unknown hash
2D graphics backend	Skia/86 b939c288f3d6479d88d1444fafbe7441d11348aa
Command Line	/usr/bin/vivaldi-stable --disable-renderer-accessibility --save-page-as-mhtml --flag-switches-begin --flag-switches-end
Driver Information
Initialization time	95
In-process GPU	false
Passthrough Command Decoder	true
Sandboxed	false
GPU0	VENDOR= 0x8086 [Google Inc.], DEVICE=0x9b41 [ANGLE (Intel, Mesa Intel(R) UHD Graphics (CML GT2), OpenGL 4.6 core)] *ACTIVE*
Optimus	false
AMD switchable	false
Driver vendor	ANGLE
Driver version	2.1.0.unknown
GPU CUDA compute capability major version	0
Pixel shader version	1.00
Vertex shader version	1.00
Max. MSAA samples	16
Machine model name	
Machine model version	
GL_VENDOR	Google Inc.
GL_RENDERER	ANGLE (Intel, Mesa Intel(R) UHD Graphics (CML GT2), OpenGL 4.6 core)
GL_VERSION	OpenGL ES 2.0.0 (ANGLE 2.1.0.unknown hash)
GL_EXTENSIONS	GL_ANGLE_base_vertex_base_instance GL_ANGLE_client_arrays GL_ANGLE_depth_texture GL_ANGLE_explicit_context GL_ANGLE_explicit_context_gles1 GL_ANGLE_framebuffer_blit GL_ANGLE_framebuffer_multisample GL_ANGLE_instanced_arrays GL_ANGLE_memory_size GL_ANGLE_multi_draw GL_ANGLE_multiview_multisample GL_ANGLE_program_cache_control GL_ANGLE_provoking_vertex GL_ANGLE_request_extension GL_ANGLE_robust_client_memory GL_ANGLE_texture_compression_dxt3 GL_ANGLE_texture_compression_dxt5 GL_ANGLE_texture_external_update GL_ANGLE_texture_rectangle GL_ANGLE_translated_shader_source GL_APPLE_clip_distance GL_ARB_sync GL_CHROMIUM_bind_generates_resource GL_CHROMIUM_bind_uniform_location GL_CHROMIUM_color_buffer_float_rgb GL_CHROMIUM_color_buffer_float_rgba GL_CHROMIUM_copy_texture GL_CHROMIUM_lose_context GL_CHROMIUM_sync_query GL_EXT_blend_minmax GL_EXT_color_buffer_half_float GL_EXT_debug_marker GL_EXT_discard_framebuffer GL_EXT_disjoint_timer_query GL_EXT_draw_buffers GL_EXT_draw_elements_base_vertex GL_EXT_float_blend GL_EXT_frag_depth GL_EXT_gpu_shader5 GL_EXT_instanced_arrays GL_EXT_map_buffer_range GL_EXT_multisample_compatibility GL_EXT_occlusion_query_boolean GL_EXT_read_format_bgra GL_EXT_robustness GL_EXT_sRGB GL_EXT_sRGB_write_control GL_EXT_shader_texture_lod GL_EXT_texture_compression_bptc GL_EXT_texture_compression_dxt1 GL_EXT_texture_compression_rgtc GL_EXT_texture_compression_s3tc_srgb GL_EXT_texture_cube_map_array GL_EXT_texture_filter_anisotropic GL_EXT_texture_format_BGRA8888 GL_EXT_texture_rg GL_EXT_texture_sRGB_decode GL_EXT_texture_storage GL_EXT_unpack_subimage GL_KHR_debug GL_KHR_parallel_shader_compile GL_KHR_texture_compression_astc_ldr GL_KHR_texture_compression_astc_sliced_3d GL_NV_depth_buffer_float2 GL_NV_fence GL_NV_pack_subimage GL_NV_pixel_buffer_object GL_NV_read_depth GL_NV_read_stencil GL_NV_shader_noperspective_interpolation GL_OES_compressed_EAC_R11_signed_texture GL_OES_compressed_EAC_R11_unsigned_texture GL_OES_compressed_EAC_RG11_signed_texture GL_OES_compressed_EAC_RG11_unsigned_texture GL_OES_compressed_ETC1_RGB8_texture GL_OES_compressed_ETC2_RGB8_texture GL_OES_compressed_ETC2_RGBA8_texture GL_OES_compressed_ETC2_punchthroughA_RGBA8_texture GL_OES_compressed_ETC2_punchthroughA_sRGB8_alpha_texture GL_OES_compressed_ETC2_sRGB8_alpha8_texture GL_OES_compressed_ETC2_sRGB8_texture GL_OES_depth24 GL_OES_depth32 GL_OES_depth_texture GL_OES_draw_elements_base_vertex GL_OES_element_index_uint GL_OES_fbo_render_mipmap GL_OES_get_program_binary GL_OES_mapbuffer GL_OES_packed_depth_stencil GL_OES_rgb8_rgba8 GL_OES_standard_derivatives GL_OES_surfaceless_context GL_OES_texture_3D GL_OES_texture_border_clamp GL_OES_texture_cube_map_array GL_OES_texture_float GL_OES_texture_float_linear GL_OES_texture_half_float GL_OES_texture_half_float_linear GL_OES_texture_npot GL_OES_vertex_array_object GL_WEBGL_video_texture
Disabled Extensions	GL_KHR_blend_equation_advanced GL_KHR_blend_equation_advanced_coherent
Disabled WebGL Extensions	
Window system binding vendor	
Window system binding version	1.5 (ANGLE 2.1.0.unknown hash)
Window system binding extensions	EGL_EXT_create_context_robustness EGL_KHR_create_context EGL_KHR_get_all_proc_addresses EGL_ANGLE_create_context_webgl_compatibility EGL_CHROMIUM_create_context_bind_generates_resource EGL_CHROMIUM_sync_control EGL_ANGLE_sync_control_rate EGL_EXT_pixel_format_float EGL_KHR_surfaceless_context EGL_ANGLE_display_texture_share_group EGL_ANGLE_display_semaphore_share_group EGL_ANGLE_create_context_client_arrays EGL_ANGLE_program_cache_control EGL_ANGLE_robust_resource_initialization EGL_ANGLE_create_context_extensions_enabled EGL_ANDROID_blob_cache EGL_ANDROID_recordable EGL_ANGLE_create_context_backwards_compatible EGL_KHR_create_context_no_error EGL_NOK_texture_from_pixmap
Window manager	KWin
XDG_CURRENT_DESKTOP	KDE
Compositing manager	Yes
System visual ID	0
RGBA visual ID	0
Direct rendering version	unknown
Reset notification strategy	0x8252
GPU process crash count	0
gfx::BufferFormats supported for allocation and texturing	R_8: not supported, R_16: not supported, RG_88: not supported, BGR_565: not supported, RGBA_4444: not supported, RGBX_8888: not supported, RGBA_8888: not supported, BGRX_8888: not supported, BGRA_1010102: not supported, RGBA_1010102: not supported, BGRA_8888: not supported, RGBA_F16: not supported, YVU_420: not supported, YUV_420_BIPLANAR: not supported, P010: not supported
Compositor Information
Tile Update Mode	One-copy
Partial Raster	Enabled
GpuMemoryBuffers Status
R_8	Software only
R_16	Software only
RG_88	Software only
BGR_565	Software only
RGBA_4444	Software only
RGBX_8888	Software only
RGBA_8888	Software only
BGRX_8888	Software only
BGRA_1010102	Software only
RGBA_1010102	Software only
BGRA_8888	Software only
RGBA_F16	Software only
YVU_420	Software only
YUV_420_BIPLANAR	Software only
P010	Software only
Display(s) Information
Info	Display[1881264395124802] bounds=[0,0 1920x1080], workarea=[0,0 1920x1080], scale=1, rotation=0, panel_rotation=0 external.
Color space (all)	{primaries:BT709, transfer:IEC61966_2_1, matrix:RGB, range:FULL}
Buffer format (all)	BGRA_8888
SDR white level in nits	100
Bits per color component	8
Bits per pixel	24
Refresh Rate in Hz	120
Video Acceleration Information
Vulkan Information
Device Performance Information
Log Messages
[133688:133688:1126/120339.556621:ERROR:sandbox_linux.cc(374)] : InitializeSandbox() called with multiple threads in process gpu-process.

But I cannot believe that the internal GPU of an Intel CPU might not be supported.

npro

@jochen01 said in 100% CPU load in Jitsi Meet conference:

Same conference with same settings in firefox uses just about 80% CPU.

If you count the sum of all cores as 100% the load is about 80..90%. Far too high for low resolution video decoded by hardware.

I don't get the purpose of this thread, to make it not use 85% but 80% of the CPU? Clearly an old i5 of a notebook is not strong enough, its integrated gpu so old that was dropped in Chromium. The "bad" (but expected) performance is due to the machine, not the browser.

You could post your inxi -Fmxxzy --no-host for us to see what kind of machine it is.

mib2berlin

@jochen01
Hehe no, but Google don´t trust the Intel driver, I guess.
Now you can try to enable video decode with the flag:
vivaldi://flags/#enable-accelerated-video-decode
and check.
I guess the compression is high for these video conference tools.
I have no idea if the Rasterization is involved, too.

Cheers, mib
EDIT: Forgot to mention you may have to start Vivaldi with --ignore-gpu-blocklist
to enable it. Cant test at moment.

jochen01

@npro You don't get the purpose of this thread? Maybe I can help.
A simple video conference uses about 85% CPU. I don't want it to consume 5% less but about 80% less. The load should be near 10...15%.
O.k. for you if you consider it an "old i5". Probably you find yourself quite alone with this opinion. CPU is Intel Core i5-10210U. Too old for you?
Driver for Intel GPU is used since H265 at full hd does not cause noticeable load. So it is decoded in hardware by the "old i5".

Output of "inxi -Fmxxz --no-host" (only relevant data):

System:    Kernel: 5.4.0-54-generic x86_64 bits: 64 compiler: gcc v: 9.3.0 Desktop: KDE Plasma 5.18.5 tk: Qt 5.12.8
           wm: kwin_x11 dm: SDDM Distro: Ubuntu 20.04.1 LTS (Focal Fossa)
Machine:   Type: Laptop System: TUXEDO product: InfinityBook_S_14_v5 v: N/A serial: <filter> Chassis: Notebook
           type: 10 serial: <filter> 
           Mobo: TUXEDO model: L140CU serial: <filter> UEFI: INSYDE v: 1.07.07RTR3 date: 02/17/2020
Memory:    RAM: total: 15.37 GiB used: 2.83 GiB (18.4%)
           RAM Report: permissions: Unable to run dmidecode. Root privileges required.
CPU:       Topology: Quad Core model: Intel Core i5-10210U bits: 64 type: MT MCP arch: Kaby Lake rev: C
           L2 cache: 6144 KiB
           flags: avx avx2 lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx bogomips: 33599
           Speed: 1312 MHz min/max: 400/4200 MHz Core speeds (MHz): 1: 1312 2: 2404 3: 1986 4: 1458 5: 1393 6: 1260
           7: 1827 8: 1692
Graphics:  Device-1: Intel UHD Graphics vendor: CLEVO/KAPOK driver: i915 v: kernel bus ID: 00:02.0
           chip ID: 8086:9b41
           Display: x11 server: X.Org 1.20.8 driver: modesetting unloaded: fbdev,vesa compositor: kwin_x11 tty: N/A
           OpenGL: renderer: Mesa Intel UHD Graphics (CML GT2) v: 4.6 Mesa 20.0.8 direct render: Yes
Audio:     Device-1: Intel vendor: CLEVO/KAPOK driver: snd_hda_intel v: kernel bus ID: 00:1f.3 chip ID: 8086:02c8
           Sound Server: ALSA v: k5.4.0-54-generic
Network:   Device-1: Intel Wi-Fi 6 AX200 driver: iwlwifi v: kernel port: 5040 bus ID: 06:00.0 chip ID: 8086:2723
           IF: wlp6s0 state: up mac: <filter> 
Drives:    Local Storage: total: 465.76 GiB used: 179.91 GiB (38.6%) 
           ID-1: /dev/sda vendor: Samsung model: SSD 860 EVO M.2 500GB size: 465.76 GiB speed: 6.0 Gb/s 
           serial: <filter> 
Sensors:   System Temperatures: cpu: 46.0 C mobo: N/A 
           Fan Speeds (RPM): N/A

jochen01

@mib2berlin Tried it, does not really help.
But I noticed that switching off the camera (built in) helps a lot (CPU load drops to about 40%). It's not the camera itself, because if showing live video from the camera with vlc the CPU stays near idle.

Zalex108

@jochen01

Hi,

Try on a New 's Profile, if not yet:

" Right side" |

Rename it as Tests or something related, and use it as a Clean Profile, to test without custom configurations that can alter the original operation.

Since the result, check:

Active Experiments
Ad Blocker
Clean Cache
Site Data
Extensions - Tip

Or fill in a Bug Report and please, add the VB# to the topic for follow up.

--
"Off Topic Tip"
Follow the Signature's Backup | Reset link.
Take the opportunity to start a Backup plan and even create a Template Profile.

Windows 7 (x64)
Vivaldi Back up | Reset

npro

@jochen01 Much better now, we can keep that as a system spec reference for feature issues, as CoC also requires. Indeed now we can see it's not an old notebook and it's a tuxedo laptop, nice .The point still remains though, if this program makes Firefox consume 80% and Vivaldi 80%-90% then it's not Vivaldi's fault but the program's. A simple duckduckgoing shows numerous reports about its high CPU usage, even in their own github https://github.com/jitsi/jitsi-meet/issues/2596

And another one, random, fwiw: https://bugzilla.mozilla.org/show_bug.cgi?id=1247244

For future reference: I think one thing we only found out later is that the Firefox performance on the integrated Intel graphic cards is really bad. Although it shouldn't be like that on MacBook Pro's it helps to switch to the discrete Nvidia or AMD card instead.

So, you have to study those, unfortunately

jochen01

@Zalex108 It happens in a clean profile too. Filled a bug report, but don't know what VB# means. Issue key: VB-74672.

Zalex108

@jochen01 said in 100% CPU load in Jitsi Meet conference:

@Zalex108 It happens in a clean profile too. Filled a bug report, but don't know what VB# means. Issue key: VB-74672.

That exactly, - Vivaldi Bug Number | VB-74672.

Then,
After those other checks, seems that a waiting time will be necessary.

You can eventually, check on the Vivaldi SnapShot version.

npro

@jochen01 Another good idea is to always check vanilla Chromium on the same version, instead of Firefox, as it's the base for all Chromium-based browsers like Vivaldi. The above github link is about Chrome after all.

jochen01

@npro Uups. Now I understand why it has been difficult to convince you that here may be a problem. It's been a typo.

Vivaldi: CPU almost 100% (85..95%); all cores
Firefox: CPU about 20%
Base load from other software/system included but has been constant during the test.

npro

@jochen01 Right, this shouldn't happen in the first place, but anyway, see the above post for comparisons.

jochen01

@npro Unfortunately chromium is only available as a snap package for Ubuntu 20.x. That the reason for not trying that.
But I have installed Google Chrome to compare the browsers. Chrom is about the same as Firefox: 20..30% load.
Version info for Chrome: 87.0.4280.66

jochen01

@npro said in 100% CPU load in Jitsi Meet conference:

A simple duckduckgoing shows numerous reports about its high CPU usage, even in their own github https://github.com/jitsi/jitsi-meet/issues/2596

Cisco WebEX (we use it in the office) shows the same problem.

npro

@jochen01 said in 100% CPU load in Jitsi Meet conference:

@npro Unfortunately chromium is only available as a snap package for Ubuntu 20.x. That the reason for not trying that.
But I have installed Google Chrome to compare the browsers. Chrom is about the same as Firefox: 20..30% load.
Version info for Chrome: 87.0.4280.66

Even that is not comparable:
@npro said in 100% CPU load in Jitsi Meet conference:

@jochen01 Another good idea is to always check vanilla Chromium on the same version,

so you should compare Vivaldi (86) to Chrome 86, not 87.

P.S. Rhetorical question: Why you still on Ubuntu... couldn't you be using at least... Mint?