[Notice of Holidays]
Please note that our staff will not be available from 2024/Dec/28 - 2025/Jan/5 JST.
The Live2D Creators Forum will remain open during the holidays.
Submissions that require a response will be answered after we reopen for business.
Please note that it will take longer than usual to respond to your request.
5.0 beta UI Upscale causes sever fps drop from high res monitors in physics window
Something that I have run into with multiple monitors at different resolutions is that with the ui upscaling the preview in the physics window drops frames a noticeable amount.
In my case, I have a widescreen (5120x2160), a 4k monitor, and a spare 1440p monitor. Across all of them, the frame drop becomes progressively worse to the point where single frames are seen.
Currently what I have found is that the only solution is to turn off the upscaling however the whole point of the ui upscaling is so that you can see/read what you are doing in the software and without it it is harder to navigate through to the program.
In the videos attached, the first half is with the ui scaling and the last without. For heavier model files it becomes extremely apparent.
https://drive.google.com/file/d/1--YK-Ch8GDWSnDYEykwCWKAgHPPhECT4/view?usp=sharinghttps://drive.google.com/file/d/1-9xq1ncwe4Bo4iW6Keaq_4crBjKZe0Nb/view?usp=sharing
1 ·
Comments
specs:
- Live2D Cubism 5.0 beta1
- OS: Windows 11
- GPU: RTX 4080
- CPU: Ryzen 9 5900X
- Memory: 64GB
Still experiencing severe lag after running live2d from "CubismEditor5_nod3d.bat"
It does not seem to fix the issue
@cillia
Thank you for always using our products.
Thank you for reporting the problem.
We will investigate further.
Please disable scaling for the time being.
Best regards.
@cillia
Thank you for always using our products.
Thank you for reporting the problem.
- Please let me know the model number of the monitor you are using and the resolution and magnification settings.
(e.g., one laptop monitor 120% & one monitor 3840x2160 100% )
- Please try to see if the problem occurs even in an environment with only one monitor.
- Does the problem still occur when the resolution is set to FHD?
- Drag the width of the parameter display area in the physics window to narrow it so that it cannot be seen at all. Please try to see if the problem occurs even in that state.
- Update your graphics driver to the latest version.
- GPU information is displayed in the log when Cubism Editor is started. Please check if the information of your graphic board is displayed without problems.
Best regards.
I tried setting my main monitor to 1080p, and 100% scaling, but there still are frame drops. I think maybe just slightly better. (still worse in comparison to 4.2)
https://drive.google.com/file/d/10IwqZDE5acOKLG7JstIxlfPDWOadL0OW/view?usp=sharing
Here is testing of the physics parameter display at 1440p 125% magnification. The showing of the parameters does seem to greatly affect performance. However, even with the parameters hidden, the frame rate is still lower than when running Live2D version 4.2.
https://drive.google.com/file/d/1P-TDjxvsGcNQ-62xyQuAubXcfifICzkm/view?usp=sharing
Graphic drivers are fully updated, and the log is showing the information correctly.
I have not yet tried it with just 1 monitor on. I will see if I can test that later today.
At the least, it is clear that the visual processing of the parameters is one aspect of the slowdown. (But not the only cause)
With my monitor I cannot set things to FHD since it is not an option that is available within the display settings. Any options within FHD range will crop the display leaving black bars along the sides.
Here is my test for 5.0, 5120 x 2160p, 150% magnification.
https://drive.google.com/file/d/1-A0hYnoNvFvWya7CLgncf3WDhYSNbkav/view?usp=sharing
Here is what Live2D runs like on 4.2. During this test I had my normal setup up of two monitors, the main monitor listed above and an alt monitor to my right set to vertical orientation at 2160x3840p
https://drive.google.com/file/d/1-DkGb_ouxw-Blwa5TXj6hMbIdExUdH3r/view?usp=sharing
@indigo_cho
We apologize for the inconvenience.
There was a report that this bug was improved in version beta2 released yesterday.
Please try beta2.
Best regards.
If I changed my resolution to 1080p, the average FPS I get is 20 ~ 30 FPS for a moderately big project.
But if I use the native 4K resolution, the average FPS is cut by half, only ~10 FPS average.
There's something wrong with the Hi-DPI mode here. I never had this sluggish performance in 4.2 version. And yes I tried to update beta2, while there's an improvement compared to D3D mode but in general it still runs worse than 4.2.
My log:
[INFO ] ======================================================================
[INFO ]
[INFO ] Live2D Cubism Editor ver5.0.00 beta2 [ 500000004 / 2023/08/28]
[INFO ]
[INFO ] ======================================================================
[INFO ] ======================================================================
[INFO ] Tue Aug 29 14:21:33 ICT 2023
[INFO ] BASE : jar:file:/C:/Program%20Files/Live2D%20Cubism%205.0%20beta/app/lib/Live2D_Cubism.jar!/res/ @IRes_Design
[ERROR] INFO: jpen-2-3-64 couldn't be loaded
[ERROR] INFO: no suitable JNI library found
[INFO ] java2d graphics pipeline[ sun.java2d.opengl = NotSet ]
[INFO ] java2d graphics pipeline[ sun.java2d.d3d = false ]
[INFO ] java2d graphics pipeline[ sun.java2d.noddraw = NotSet ]
[INFO ] java2d graphics pipeline[ sun.java2d.metal = NotSet ]
[INFO ] application ui scale : 100.00%
@CubismAppBackupManager
[INFO ] ---------- OpenGL Environment ----------
[INFO ] Version : 4.6.0 NVIDIA 536.40
[INFO ] Vendor : NVIDIA Corporation
[INFO ] Renderer : NVIDIA GeForce RTX 3060 Laptop GPU/PCIe/SSE2
[INFO ] Extensions : GL_AMD_multi_draw_indirect GL_AMD_seamless_cubemap_per_texture GL_AMD_vertex_shader_viewport_index GL_AMD_vertex_shader_layer GL_ARB_arrays_of_arrays GL_ARB_base_instance GL_ARB_bindless_texture GL_ARB_blend_func_extended GL_ARB_buffer_storage GL_ARB_clear_buffer_object GL_ARB_clear_texture GL_ARB_clip_control GL_ARB_color_buffer_float GL_ARB_compressed_texture_pixel_storage GL_ARB_conservative_depth GL_ARB_compute_shader GL_ARB_compute_variable_group_size GL_ARB_conditional_render_inverted GL_ARB_copy_buffer GL_ARB_copy_image GL_ARB_cull_distance GL_ARB_debug_output GL_ARB_depth_buffer_float GL_ARB_depth_clamp GL_ARB_depth_texture GL_ARB_derivative_control GL_ARB_direct_state_access GL_ARB_draw_buffers GL_ARB_draw_buffers_blend GL_ARB_draw_indirect GL_ARB_draw_elements_base_vertex GL_ARB_draw_instanced GL_ARB_enhanced_layouts GL_ARB_ES2_compatibility GL_ARB_ES3_compatibility GL_ARB_ES3_1_compatibility GL_ARB_ES3_2_compatibility GL_ARB_explicit_attrib_location GL_ARB_explicit_uniform_location GL_ARB_fragment_coord_conventions GL_ARB_fragment_layer_viewport GL_ARB_fragment_program GL_ARB_fragment_program_shadow GL_ARB_fragment_shader GL_ARB_fragment_shader_interlock GL_ARB_framebuffer_no_attachments GL_ARB_framebuffer_object GL_ARB_framebuffer_sRGB GL_ARB_geometry_shader4 GL_ARB_get_program_binary GL_ARB_get_texture_sub_image GL_ARB_gl_spirv GL_ARB_gpu_shader5 GL_ARB_gpu_shader_fp64 GL_ARB_gpu_shader_int64 GL_ARB_half_float_pixel GL_ARB_half_float_vertex GL_ARB_imaging GL_ARB_indirect_parameters GL_ARB_instanced_arrays GL_ARB_internalformat_query GL_ARB_internalformat_query2 GL_ARB_invalidate_subdata GL_ARB_map_buffer_alignment GL_ARB_map_buffer_range GL_ARB_multi_bind GL_ARB_multi_draw_indirect GL_ARB_multisample GL_ARB_multitexture GL_ARB_occlusion_query GL_ARB_occlusion_query2 GL_ARB_parallel_shader_compile GL_ARB_pipeline_statistics_query GL_ARB_pixel_buffer_object GL_ARB_point_parameters GL_ARB_point_sprite GL_ARB_polygon_offset_clamp GL_ARB_post_depth_coverage GL_ARB_program_interface_query GL_ARB_provoking_vertex GL_ARB_query_buffer_object GL_ARB_robust_buffer_access_behavior GL_ARB_robustness GL_ARB_sample_locations GL_ARB_sample_shading GL_ARB_sampler_objects GL_ARB_seamless_cube_map GL_ARB_seamless_cubemap_per_texture GL_ARB_separate_shader_objects GL_ARB_shader_atomic_counter_ops GL_ARB_shader_atomic_counters GL_ARB_shader_ballot GL_ARB_shader_bit_encoding GL_ARB_shader_clock GL_ARB_shader_draw_parameters GL_ARB_shader_group_vote GL_ARB_shader_image_load_store GL_ARB_shader_image_size GL_ARB_shader_objects GL_ARB_shader_precision GL_ARB_shader_storage_buffer_object GL_ARB_shader_subroutine GL_ARB_shader_texture_image_samples GL_ARB_shader_texture_lod GL_ARB_shading_language_100 GL_ARB_shader_viewport_layer_array GL_ARB_shading_language_420pack GL_ARB_shading_language_include GL_ARB_shading_language_packing GL_ARB_shadow GL_ARB_sparse_buffer GL_ARB_sparse_texture GL_ARB_sparse_texture2 GL_ARB_sparse_texture_clamp GL_ARB_spirv_extensions GL_ARB_stencil_texturing GL_ARB_sync GL_ARB_tessellation_shader GL_ARB_texture_barrier GL_ARB_texture_border_clamp GL_ARB_texture_buffer_object GL_ARB_texture_buffer_object_rgb32 GL_ARB_texture_buffer_range GL_ARB_texture_compression GL_ARB_texture_compression_bptc GL_ARB_texture_compression_rgtc GL_ARB_texture_cube_map GL_ARB_texture_cube_map_array GL_ARB_texture_env_add GL_ARB_texture_env_combine GL_ARB_texture_env_crossbar GL_ARB_texture_env_dot3 GL_ARB_texture_filter_anisotropic GL_ARB_texture_filter_minmax GL_ARB_texture_float GL_ARB_texture_gather GL_ARB_texture_mirror_clamp_to_edge GL_ARB_texture_mirrored_repeat GL_ARB_texture_multisample GL_ARB_texture_non_power_of_two GL_ARB_texture_query_levels GL_ARB_texture_query_lod GL_ARB_texture_rectangle GL_ARB_texture_rg GL_ARB_texture_rgb10_a2ui GL_ARB_texture_stencil8 GL_ARB_texture_storage GL_ARB_texture_storage_multisample GL_ARB_texture_swizzle GL_ARB_texture_view GL_ARB_timer_query GL_ARB_transform_feedback2 GL_ARB_transform_feedback3 GL_ARB_transform_feedback_instanced GL_ARB_transform_feedback_overflow_query GL_ARB_transpose_matrix GL_ARB_uniform_buffer_object GL_ARB_vertex_array_bgra GL_ARB_vertex_array_object GL_ARB_vertex_attrib_64bit GL_ARB_vertex_attrib_binding GL_ARB_vertex_buffer_object GL_ARB_vertex_program GL_ARB_vertex_shader GL_ARB_vertex_type_10f_11f_11f_rev GL_ARB_vertex_type_2_10_10_10_rev GL_ARB_viewport_array GL_ARB_window_pos GL_ATI_draw_buffers GL_ATI_texture_float GL_ATI_texture_mirror_once GL_S3_s3tc GL_EXT_texture_env_add GL_EXT_abgr GL_EXT_bgra GL_EXT_bindable_uniform GL_EXT_blend_color GL_EXT_blend_equation_separate GL_EXT_blend_func_separate GL_EXT_blend_minmax GL_EXT_blend_subtract GL_EXT_compiled_vertex_array GL_EXT_Cg_shader GL_EXT_depth_bounds_test GL_EXT_direct_state_access GL_EXT_draw_buffers2 GL_EXT_draw_instanced GL_EXT_draw_range_elements GL_EXT_fog_coord GL_EXT_framebuffer_blit GL_EXT_framebuffer_multisample GL_EXTX_framebuffer_mixed_formats GL_EXT_framebuffer_multisample_blit_scaled GL_EXT_framebuffer_object GL_EXT_framebuffer_sRGB GL_EXT_geometry_shader4 GL_EXT_gpu_program_parameters GL_EXT_gpu_shader4 GL_EXT_multi_draw_arrays GL_EXT_multiview_texture_multisample GL_EXT_multiview_timer_query GL_EXT_packed_depth_stencil GL_EXT_packed_float GL_EXT_packed_pixels GL_EXT_pixel_buffer_object GL_EXT_point_parameters GL_EXT_polygon_offset_clamp GL_EXT_post_depth_coverage GL_EXT_provoking_vertex GL_EXT_raster_multisample GL_EXT_rescale_normal GL_EXT_secondary_color GL_EXT_separate_shader_objects GL_EXT_separate_specular_color GL_EXT_shader_image_load_formatted GL_EXT_shader_image_load_store GL_EXT_shader_integer_mix GL_EXT_shadow_funcs GL_EXT_sparse_texture2 GL_EXT_stencil_two_side GL_EXT_stencil_wrap GL_EXT_texture3D GL_EXT_texture_array GL_EXT_texture_buffer_object GL_EXT_texture_compression_dxt1 GL_EXT_texture_compression_latc GL_EXT_texture_compression_rgtc GL_EXT_texture_compression_s3tc GL_EXT_texture_cube_map GL_EXT_texture_edge_clamp GL_EXT_texture_env_combine GL_EXT_texture_env_dot3 GL_EXT_texture_filter_anisotropic GL_EXT_texture_filter_minmax GL_EXT_texture_integer GL_EXT_texture_lod GL_EXT_texture_lod_bias GL_EXT_texture_mirror_clamp GL_EXT_texture_object GL_EXT_texture_shadow_lod GL_EXT_texture_shared_exponent GL_EXT_texture_sRGB GL_EXT_texture_sRGB_R8 GL_EXT_texture_sRGB_decode GL_EXT_texture_storage GL_EXT_texture_swizzle GL_EXT_timer_query GL_EXT_transform_feedback2 GL_EXT_vertex_array GL_EXT_vertex_array_bgra GL_EXT_vertex_attrib_64bit GL_EXT_window_rectangles GL_EXT_import_sync_object GL_IBM_rasterpos_clip GL_IBM_texture_mirrored_repeat GL_KHR_context_flush_control GL_KHR_debug GL_EXT_memory_object GL_EXT_memory_object_win32 GL_NV_memory_object_sparse GL_EXT_win32_keyed_mutex GL_KHR_parallel_shader_compile GL_KHR_no_error GL_KHR_robust_buffer_access_behavior GL_KHR_robustness GL_EXT_semaphore GL_EXT_semaphore_win32 GL_NV_timeline_semaphore GL_KHR_shader_subgroup GL_KTX_buffer_region GL_NV_alpha_to_coverage_dither_control GL_NV_bindless_multi_draw_indirect GL_NV_bindless_multi_draw_indirect_count GL_NV_bindless_texture GL_NV_blend_equation_advanced GL_NV_blend_equation_advanced_coherent GL_NVX_blend_equation_advanced_multi_draw_buffers GL_NV_blend_minmax_factor GL_NV_blend_square GL_NV_clip_space_w_scaling GL_NV_command_list GL_NV_compute_program5 GL_NV_compute_shader_derivatives GL_NV_conditional_render GL_NV_conservative_raster GL_NV_conservative_raster_dilate GL_NV_conservative_raster_pre_snap GL_NV_conservative_raster_pre_snap_triangles GL_NV_conservative_raster_underestimation GL_NV_copy_depth_to_color GL_NV_copy_image GL_NV_depth_buffer_float GL_NV_depth_clamp GL_NV_draw_texture GL_NV_draw_vulkan_image GL_NV_ES1_1_compatibility GL_NV_ES3_1_compatibility GL_NV_explicit_multisample GL_NV_feature_query GL_NV_fence GL_NV_fill_rectangle GL_NV_float_buffer GL_NV_fog_distance GL_NV_fragment_coverage_to_color GL_NV_fragment_program GL_NV_fragment_program_option GL_NV_fragment_program2 GL_NV_fragment_shader_barycentric GL_NV_fragment_shader_interlock GL_NV_framebuffer_mixed_samples GL_NV_framebuffer_multisample_coverage GL_NV_geometry_shader4 GL_NV_geometry_shader_passthrough GL_NV_gpu_program4 GL_NV_internalformat_sample_query GL_NV_gpu_program4_1 GL_NV_gpu_program5 GL_NV_gpu_program5_mem_extended GL_NV_gpu_program_fp64 GL_NV_gpu_shader5 GL_NV_half_float GL_NV_light_max_exponent GL_NV_memory_attachment GL_NV_mesh_shader GL_NV_multisample_coverage GL_NV_multisample_filter_hint GL_NV_occlusion_query GL_NV_packed_depth_stencil GL_NV_parameter_buffer_object GL_NV_parameter_buffer_object2 GL_NV_path_rendering GL_NV_path_rendering_shared_edge GL_NV_pixel_data_range GL_NV_point_sprite GL_NV_primitive_restart GL_NV_primitive_shading_rate GL_NV_query_resource GL_NV_query_resource_tag GL_NV_register_combiners GL_NV_register_combiners2 GL_NV_representative_fragment_test GL_NV_sample_locations GL_NV_sample_mask_override_coverage GL_NV_scissor_exclusive GL_NV_shader_atomic_counters GL_NV_shader_atomic_float GL_NV_shader_atomic_float64 GL_NV_shader_atomic_fp16_vector GL_NV_shader_atomic_int64 GL_NV_shader_buffer_load GL_NV_shader_storage_buffer_object GL_NV_shader_subgroup_partitioned GL_NV_shader_texture_footprint GL_NV_shading_rate_image GL_NV_stereo_view_rendering GL_NV_texgen_reflection GL_NV_texture_barrier GL_NV_texture_compression_vtc GL_NV_texture_dirty_tile_map GL_NV_texture_env_combine4 GL_NV_texture_multisample GL_NV_texture_rectangle GL_NV_texture_rectangle_compressed GL_NV_texture_shader GL_NV_texture_shader2 GL_NV_texture_shader3 GL_NV_transform_feedback GL_NV_transform_feedback2 GL_NV_uniform_buffer_unified_memory GL_NV_uniform_buffer_std430_layout GL_NV_vertex_array_range GL_NV_vertex_array_range2 GL_NV_vertex_attrib_integer_64bit GL_NV_vertex_buffer_unified_memory GL_NV_vertex_program GL_NV_vertex_program1_1 GL_NV_vertex_program2 GL_NV_vertex_program2_option GL_NV_vertex_program3 GL_NV_viewport_array2 GL_NV_viewport_swizzle GL_NVX_conditional_render GL_NVX_linked_gpu_multicast GL_NV_gpu_multicast GL_NVX_gpu_multicast2 GL_NVX_progress_fence GL_NVX_gpu_memory_info GL_NVX_multigpu_info GL_NVX_nvenc_interop GL_NV_shader_thread_group GL_NV_shader_thread_shuffle GL_KHR_blend_equation_advanced GL_KHR_blend_equation_advanced_coherent GL_OVR_multiview GL_OVR_multiview2 GL_SGIS_generate_mipmap GL_SGIS_texture_lod GL_SGIX_depth_texture GL_SGIX_shadow GL_SUN_slice_accum GL_WIN_swap_hint WGL_EXT_swap_control
[INFO ] ----------------------------------------
[INFO ] calc screen size @UtGui : 0.042 msec
@CubismAppBackupManager
[INFO ] check update @AppCtrl_System : 501.943 msec
[INFO ] reset
[INFO ] reset
I found setting override for OS to handle the application scaling may have improved the Editor performance for me.
It's the setting that I always turned on prior 5.0 to make Live2D Editor scale appropriately on Hi-DPI mode. It's on the application Compatibility properties tab
I've noticed quite an improvement where the Editor doesn't feel as sluggish as before.
I'll be using Niziiro Mao model as an example. For the test I'll just spin the Angle XY parameters around.
This is the average stats with the override OFF.
And this one with override ON.
I hope this finding will be helpful.
We apologize for any inconvenience caused.
We have released version 5.0.00, how is its performance?
I would appreciate any feedback on whether it has improved or if there has been no change at all.
Best regards.
At this point, I actually can't tell if the UI scaling itself is causing slowdowns after the update, it honestly just seems to be just 5.0 itself.
https://drive.google.com/file/d/1Q-whFsowqYIRjPDjz2hKCeXcOaXlHVgz/view?usp=sharing
This is a video comparison. First is the model in 4.2, second is the model in 5.0 with UI scaling enabled. Third is 5.0 with UI scaling disabled.
Here is another recording, running 5.0 using the d3d.bat file. First is with UI scaling off, second is with UI scaling on.
https://drive.google.com/file/d/13QqqgNZ0aXE5nrPdttIL0_X24naPl4FA/view?usp=sharing
thank you for your reply.
I am relieved that it seems to have improved significantly.
I think it can be further improved by hiding the parameter palette in the physics settings, so if you find it a little slow in the future, please try it.
If you notice a significant performance deterioration, please report back.
Thank you!
The frame drops in 5.0 are still quite noticeable, to the extent where I am currently continuing to use 4.2, as it runs smoother. It's I'd say about the same amount of frame drops as Ran had mentioned. (~10+ fps loss) Between 5.0 beta and 5.0 public release, what has improved is the parameter list processing in the physics window. Adjusting the parameter palette now no longer has a large effect on the frame rate of the program compared to before, which is good. However, the frame drops/loss in the program as well as in the physics window are still much worse than 4.2. Unfortunately doing the override as Ran mentioned did not improve performance, and it also results in blurry UI (which is why I didn't use that setting in the past as well)
However, without UI scaling, the performance has become much more consistent. Before, even after disabling UI scaling, there was still quite a lot of performance lost between 4.2 and 5.0. Now that seems to be better. (though there is still CPU spikes happening in the physics window which did not occur before 5.0)
I'm hoping there's still something that can be done to fix the UI scaling lag with a future update. It's quite straining on the eyes to have to work with smaller UI and text size than normal.
Thank you for always using our products.
The cause of this issue is currently unknown.
We'll do some research, but it's likely to take some time.
I replied the same thing regarding CPU spikes.
We apologize for any inconvenience caused.
Best regards.