Issue metadata
Sign in to add a comment
|
Dropped Frames in Fullscreen on YouTube (and other video streaming services)
Reported by
kwip...@gmail.com,
May 18 2017
|
||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36 Example URL: https://productforums.google.com/forum/#!topic/chrome/m78yLHuz08Q Steps to reproduce the problem: 1.) Have a dual monitors, and have Monitor A, duplicated to Monitor B. 2.) Have a game running in the background in Monitor A. (I am using Dark Souls 3 in windowed mode at 1280x720 resolution) 3.) Open up Chrome and play a video in full-screen on Monitor A, and make sure you click on the "Stats for Nerds" so you can see the dropped frames. What is the expected behavior? Uhh... no dropped frames? What went wrong? I get dropped frames and audio/video sync issues... only in fullscreen during Youtube video playback. Did this work before? Yes 49 Is it a problem with Flash or HTML5? HTML5 Does this work in other browsers? Yes Chrome version: 58.0.3029.110 Channel: stable OS Version: 10.0 Flash Version: Contents of chrome://gpu: Graphics Feature Status Canvas: Hardware accelerated Flash: Hardware accelerated Flash Stage3D: Hardware accelerated Flash Stage3D Baseline profile: Hardware accelerated Compositing: Hardware accelerated Multiple Raster Threads: Enabled Native GpuMemoryBuffers: Software only. Hardware acceleration disabled Rasterization: Hardware accelerated Video Decode: Hardware accelerated Video Encode: Hardware accelerated VPx Video Decode: Hardware accelerated WebGL: Hardware accelerated WebGL2: Hardware accelerated Driver Bug Workarounds clear_uniforms_before_first_program_use decode_encode_srgb_for_generatemipmap disable_discard_framebuffer disable_dxgi_zero_copy_video disable_framebuffer_cmaa exit_on_context_lost force_cube_complete scalarize_vec_and_mat_constructor_args texsubimage_faster_than_teximage Problems Detected Some drivers are unable to reset the D3D device in the GPU process sandbox Applied Workarounds: exit_on_context_lost TexSubImage is faster for full uploads on ANGLE Applied Workarounds: texsubimage_faster_than_teximage 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 ANGLE crash on glReadPixels from incomplete cube map texture: 518889 Applied Workarounds: force_cube_complete Framebuffer discarding can hurt performance on non-tilers: 570897 Applied Workarounds: disable_discard_framebuffer Limited enabling of Chromium GL_INTEL_framebuffer_CMAA: 535198 Applied Workarounds: disable_framebuffer_cmaa Zero-copy NV12 video displays incorrect colors on NVIDIA drivers.: 635319 Applied Workarounds: disable_dxgi_zero_copy_video Disable KHR_blend_equation_advanced until cc shaders are updated: 661715 Decode and Encode before generateMipmap for srgb format textures on Windows: 634519 Applied Workarounds: decode_encode_srgb_for_generatemipmap Native GpuMemoryBuffers have been disabled, either via about:flags or command line. Disabled Features: native_gpu_memory_buffers Version Information Data exported 5/18/2017, 6:08:21 AM Chrome version Chrome/58.0.3029.110 Operating system Windows NT 10.0.15063 Software rendering list version 12.20 Driver bug list version 9.36 ANGLE commit id 461d9a3060e3 2D graphics backend Skia/58 4c81ba6ba3a3270db809bf7d4c3bc782694a56a4 Command Line Args Files (x86)\Google\Chrome\Application\chrome.exe" --flag-switches-begin --flag-switches-end Driver Information Initialization time 84 In-process GPU false Passthrough Command Decoder false Sandboxed false GPU0 VENDOR = 0x10de, DEVICE= 0x1c03 Optimus false Optimus false AMD switchable false Desktop compositing Aero Glass Diagonal Monitor Size of \\.\DISPLAY1 23.9" Diagonal Monitor Size of \\.\DISPLAY1 21.4" Driver vendor NVIDIA Driver version 22.21.13.8219 Driver date 5-10-2017 Pixel shader version 5.0 Vertex shader version 5.0 Max. MSAA samples 8 Machine model name Machine model version GL_VENDOR Google Inc. GL_RENDERER ANGLE (NVIDIA GeForce GTX 1060 6GB Direct3D11 vs_5_0 ps_5_0) GL_VERSION OpenGL ES 3.0 (ANGLE 2.1.0.461d9a3060e3) GL_EXTENSIONS GL_ANGLE_client_arrays GL_ANGLE_depth_texture GL_ANGLE_framebuffer_blit GL_ANGLE_framebuffer_multisample GL_ANGLE_instanced_arrays GL_ANGLE_lossy_etc_decode GL_ANGLE_pack_reverse_row_order GL_ANGLE_request_extension GL_ANGLE_robust_client_memory GL_ANGLE_texture_compression_dxt3 GL_ANGLE_texture_compression_dxt5 GL_ANGLE_texture_usage GL_ANGLE_translated_shader_source GL_CHROMIUM_bind_generates_resource GL_CHROMIUM_bind_uniform_location GL_CHROMIUM_copy_compressed_texture GL_CHROMIUM_copy_texture GL_CHROMIUM_sync_query GL_EXT_blend_minmax GL_EXT_color_buffer_float 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_frag_depth GL_EXT_map_buffer_range GL_EXT_occlusion_query_boolean GL_EXT_read_format_bgra GL_EXT_robustness GL_EXT_sRGB GL_EXT_shader_texture_lod GL_EXT_texture_compression_dxt1 GL_EXT_texture_compression_s3tc_srgb GL_EXT_texture_filter_anisotropic GL_EXT_texture_format_BGRA8888 GL_EXT_texture_norm16 GL_EXT_texture_rg GL_EXT_texture_storage GL_EXT_unpack_subimage GL_KHR_debug GL_NV_EGL_stream_consumer_external GL_NV_fence GL_NV_pack_subimage GL_NV_pixel_buffer_object GL_OES_EGL_image GL_OES_EGL_image_external GL_OES_EGL_image_external_essl3 GL_OES_compressed_ETC1_RGB8_texture GL_OES_depth32 GL_OES_element_index_uint GL_OES_get_program_binary GL_OES_mapbuffer GL_OES_packed_depth_stencil GL_OES_rgb8_rgba8 GL_OES_standard_derivatives 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 Disabled Extensions GL_KHR_blend_equation_advanced GL_KHR_blend_equation_advanced_coherent Window system binding vendor Google Inc. (adapter LUID: 00000000000091cd) Window system binding version 1.4 (ANGLE 2.1.0.461d9a3060e3) Window system binding extensions EGL_EXT_create_context_robustness EGL_ANGLE_d3d_share_handle_client_buffer EGL_ANGLE_d3d_texture_client_buffer EGL_ANGLE_surface_d3d_texture_2d_share_handle EGL_ANGLE_query_surface_pointer EGL_ANGLE_window_fixed_size EGL_ANGLE_keyed_mutex EGL_ANGLE_surface_orientation EGL_ANGLE_direct_composition EGL_NV_post_sub_buffer EGL_KHR_create_context EGL_EXT_device_query EGL_KHR_image EGL_KHR_image_base EGL_KHR_gl_texture_2D_image EGL_KHR_gl_texture_cubemap_image EGL_KHR_gl_renderbuffer_image EGL_KHR_get_all_proc_addresses EGL_KHR_stream EGL_KHR_stream_consumer_gltexture EGL_NV_stream_consumer_gltexture_yuv EGL_ANGLE_flexible_surface_compatibility EGL_ANGLE_stream_producer_d3d_texture_nv12 EGL_ANGLE_create_context_webgl_compatibility EGL_CHROMIUM_create_context_bind_generates_resource EGL_EXT_pixel_format_float EGL_ANGLE_display_texture_share_group EGL_ANGLE_create_context_client_arrays Direct rendering Yes Reset notification strategy 0x8252 GPU process crash count 0 Compositor Information Tile Update Mode One-copy Partial Raster Enabled GpuMemoryBuffers Status ATC Software only ATCIA Software only DXT1 Software only DXT5 Software only ETC1 Software only R_8 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_8888 Software only YVU_420 Software only YUV_420_BIPLANAR Software only UYVY_422 Software only Diagnostics 0 b3DAccelerationEnabled true b3DAccelerationExists true bAGPEnabled true bAGPExistenceValid true bAGPExists true bCanRenderWindow true bDDAccelerationEnabled true bDriverBeta false bDriverDebug false bDriverSigned false bDriverSignedValid false bNoHardware false dwBpp 32 dwDDIVersion 12 dwHeight 1080 dwRefreshRate 59 dwWHQLLevel 0 dwWidth 1920 iAdapter 0 lDriverSize 885216 lMiniVddSize 0 szAGPStatusEnglish Enabled szAGPStatusLocalized Enabled szChipType GeForce GTX 1060 6GB szD3DStatusEnglish Enabled szD3DStatusLocalized Enabled szDACType Integrated RAMDAC szDDIVersionEnglish 12 szDDIVersionLocalized 12 szDDStatusEnglish Enabled szDDStatusLocalized Enabled szDXVAHDEnglish Supported szDXVAModes szDescription NVIDIA GeForce GTX 1060 6GB szDeviceId 0x1C03 szDeviceIdentifier {D7B71E3E-5F43-11CF-C957-8C121BC2DB35} szDeviceName \\.\DISPLAY1 szDisplayMemoryEnglish 14235 MB szDisplayMemoryLocalized 14235 MB szDisplayModeEnglish 1920 x 1080 (32 bit) (59Hz) szDisplayModeLocalized 1920 x 1080 (32 bit) (59Hz) szDriverAssemblyVersion 22.21.13.8219 szDriverAttributes Final Retail szDriverDateEnglish 5/9/2017 7:00:00 PM szDriverDateLocalized 5/9/2017 19:00:00 szDriverLanguageEnglish English szDriverLanguageLocalized English szDriverModelEnglish WDDM 2.2 szDriverModelLocalized WDDM 2.2 szDriverName C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_c018c05275eddc7f\nvldumdx.dll,C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_c018c05275eddc7f\nvldumdx.dll,C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_c018c05275eddc7f\nvldumdx.dll,C:\WINDOWS\System32\DriverStore\FileRepository\nv_dispi.inf_amd64_c018c05275eddc7f\nvldumdx.dll szDriverNodeStrongName oem15.inf:0f066de38c1ebff8:Section088:22.21.13.8219:pci\ven_10de&dev_1c03 szDriverSignDate Unknown szDriverVersion 22.21.0013.8219 szKeyDeviceID Enum\PCI\VEN_10DE&DEV_1C03&SUBSYS_32811462&REV_A1 szKeyDeviceKey \Registry\Machine\System\CurrentControlSet\Control\Video\{14A6DE4E-47F9-4794-A302-686FB2BB714B}\0000 szManufacturer NVIDIA szMiniVdd unknown szMiniVddDateEnglish Unknown szMiniVddDateLocalized unknown szMonitorMaxRes Unknown szMonitorName Generic PnP Monitor szNotesEnglish No problems found. szNotesLocalized No problems found. szOverlayEnglish Supported szRankOfInstalledDriver 00D12001 szRegHelpText Unknown szRevision Unknown szRevisionId 0x00A1 szSubSysId 0x32811462 szTestResultD3D7English Not run szTestResultD3D7Localized Not run szTestResultD3D8English Not run szTestResultD3D8Localized Not run szTestResultD3D9English Not run szTestResultD3D9Localized Not run szTestResultDDEnglish Not run szTestResultDDLocalized Not run szVdd unknown szVendorId 0x10DE Log Messages GpuProcessHostUIShim: The GPU process exited normally. Everything is okay. This here provides a more detailed description of the problem. https://productforums.google.com/forum/#!topic/chrome/m78yLHuz08Q
,
May 19 2017
I just did the same test in the latest version of Microsoft Edge. Here are the results at the end of the video... Video ID:rql_F8H3h9E Dimensions:854 x 480 Resolution:1920 x 1080@60 Optimal Resolution:1920 x 1080@60 Volume:100% Stream Host:r3---sn-vgqs7nls Stream Type:https CPN:bpgqxJ8v8nRuGpTZ Mime Type:video/webm; codecs="vp9" DASH:yes (303/251) Connection Speed:30437 Kbps Buffer Health:0.0 s Network Activity:0 KB Dropped Frames:0/8432
,
May 19 2017
As you can see in the last line... No dropped frames.
,
May 23 2017
So... does nobody have any ideas on how to fix this?
,
May 23 2017
Here is a recording I made of the issue I am having. https://www.youtube.com/watch?v=AJ2gdvzpM_8#t=62.8934
,
May 23 2017
Is that the actual recording, or a mispaste? Also, could you follow the steps at https://www.chromium.org/developers/how-tos/submitting-a-performance-bug to record a trace when youtube is dropping frames? I probably only need a couple of seconds of trace while it's happening. Then you can attach the trace to this bug. Also, could you see how it works if you run Chrome with the --disable-accelerated-video-decode command-line flag? That might be a workaround.
,
May 24 2017
Is this on creator's edition or anniversary update?
,
May 24 2017
The about:gpu says 10.0.15063, which is creators update.
,
May 24 2017
Here is that trace you wanted. I hope I did this right.
,
May 24 2017
I just tried it using the command line flag --disable-accelerated-video-decode and I still have the same issue with dropped frames, like before.
,
May 24 2017
,
May 25 2017
Thanks for the trace. It looks like raw presents to the screen are taking 50ms, which is an incredibly long time. I guess running the game is taking a lot of GPU away from Chrome. Could you try running chrome with the --disable-gpu-vsync command-line flag? It's possible that would help, though I'm not too optimistic. Have you tried it with only one monitor connected, or with it set to extend the desktop onto the other monitor rather than mirroring it?
,
May 25 2017
I will try this out with the command and let you know shortly. For your other question, I have tried this with just ONE monitor and it works fine. I have even tried this with the extended desktop, again the video plays back just fine. It only happens when I clone the display... and only with every version of Google Chrome that is ABOVE version 50.
,
May 25 2017
I tried the command "--disable-gpu-vsync" and it did not work. :( I still had the same issue as before.
,
May 25 2017
Is it possible that you can try this out on your computer to see if you run into the same bug?
,
May 25 2017
Err I made an error in my response above. I should have said. "...and only with every version of Google Chrome that is version 50 and ABOVE."
,
May 25 2017
Good point - Chrome 50 is approximately when we enabled DirectComposition, so the --disable-direct-composition command-line flag should fix this for you. I'm working on reproducing this on my system, but I need to get the right combination of hardware set up. Hopefully this doesn't need Creators Update to reproduce.
,
May 25 2017
Let me try that now. I will let you know if this works or not.
,
May 25 2017
IT WORKED! IT ACTUALLY WORKED! THANK YOU! What does this command do exactly?
,
May 25 2017
I don't think this will be a creators update issue, for I was having this issue long before the creators update was released to the public. I was able to reproduce this issue on my friends computer and he is using an Nvidia GTX 1050Ti GPU. I wonder if this issue persists on an AMD GPU as well. Sadly, I don't have an AMD GPU to test this out with.
,
May 25 2017
Ok, good to hear that it works. On windows 10 we normally use the DirectComposition API to draw to the screen because it reduces power consumption a little. It looks like there's a bug in DirectComposition for your case, so that flag disables it and uses the older DXGI_SWAP_EFFECT_SEQUENTIAL mechanism for drawing to the screen
,
May 25 2017
Sweet! I really hope that this bug can be fixed so I won't need to use the command-line anymore, till then, I'll use the command line as a workaround.
,
May 25 2017
Hmmm. I might be having a new issue now. Ever since issuing the command --disable-direct-composition to my browser. It would appear that at random, the whole browser becomes unresponsive. I can't click on anything within the browser window, and I can't close it by normal means. I have to resort to using the Task Manager to force close it. I am unsure exactly what triggers it either. It just randomly happens while I am using it.
,
May 26 2017
I haven't seen those hangs before. Are there entries in chrome://crashes/ corresponding to them. If so, could you copy the list here?
,
May 26 2017
Sadly. The only entries I see are from May 4th. There aren't any here that were generated recently.
,
Jun 17 2017
Interesting. While I am using the command "--disable-direct-composition". If I right click on the bookmark bar and select any of the options "Edit", "Add Page, Add Folder", "Rename". The whole browser just locks up and I have to use the Task Manager to force close the browser.
,
Jul 29 2017
Been having this issue for months now and would like to share some additional details that might help. 1) It doesn't need to be a game, any app that uses D3D acceleration can cause it, MPC-HC player with a paused x264 video in the background is enough to cause the issue. 2) Having had to face this issue for a long time the workaround I've found was to press windows key twice so that the start menu would open and close. This would cause the taskbar to overlay the video and make it stop dropping frames for as long as it stayed showing on top. 3) --disable-direct-composition flag also seems to fix the issue for me. Glad to see the issue finally getting some attention, should have probably have reported it myself as well but was convinced it was an NVIDIA driver issue :\ I've also attached my chrome://gpu in case it's relevant.
,
Aug 9 2017
,
Sep 14 2017
Been having this issue for months on a GTX 970 and a core i7 4790k. I even switched over to Waterfox since I couldn't find a solution until I found this. Using the Mario Skate benchmark video provided by OP, I was dropping 1/3 of the frames while duplicating my monitors and on fullscreen on the latest version of chrome, and about 1/30th of the frames while reproducing the problem on Canary (still noticeable, but not nearly as bad). When extending my monitors, the amount of dropped frames dropped drastically, to about 1/30th of the frames dropped on chrome and 1/60th on canary. I tried the --disable-direct-composition flag and reproduced the problem and 1 minute into the video I've only dropped 7 frames (compared to 1100 without the flag), so I can confirm that the flag does serve as somewhat of a workaround to the issue, albeit the video still drops some frames. It's important to note that on Waterfox, both on extended screens and duplicated screens, the amount of frames dropped is a steady 0 at all times. Thank you for being in touch and transparent with the community, chrome devs, and here's to hoping that the bug with nvidia cards gets resolved. P.S. This is exclusively a Windows issue, I'm dualbooting El Capitan and I can confirm that I get 0 frames dropped on Chrome El Capitan
,
Sep 14 2017
+vmiura can you help triage this now that jbauman@ has left?
,
Oct 2 2017
As per c#29, this bug still repro. I don't have a dual monitor can repro this issue thoug. give to vmiura@ to start rolling ball.
,
Oct 3 2017
I'm still having this issue myself. Though I have another issue were the flag isn't recognized on initial load. This means I have to remove the flag, start it normally, end it, re-add the flag for this to work. Will there be a more permanent long term fix for this at all?
,
Oct 26 2017
Same issue for me. Same duplicated dual monitor setup. Works perfectly in Edge. vp9 codec in chrome, avc1 in edge.
,
Dec 2 2017
Has the --disable-direct-composition option been removed? Having it on seems to no longer fix the issue and I'm forced to keep my taskbar above the video so it remains playable otherwise it stutters with dropped frames and eventually just stops playing.
,
Dec 2 2017
Don't hold hope about this being fixed, no-one seems interested in fixing it, despite it being reproducible. I'm now finding this bug has been forked into other chromium based browsers now, like Vivaldi.
,
Dec 3 2017
I know this bug will likely never get fixes, however I did had some hope the existing workaround would continue to work. Although given the release of Firefox Quantum maybe it's time to reevaluate it again to see if it's good enough to switch over.
,
Dec 5 2017
I'm also having this issue. It's incredibly annoying since my second "monitor" is actually my amplifier which is receiving audio from my GTX970. Having the screen extended, my mouse cursor and windows can basically get lost on the amplifier part of the desktop. So I have to switch between extended and duplicating monitor mode between general use and Youtube watching. It's horrendous watching 60fps videos on youtube otherwise!
,
Dec 5 2017
Just reporting back that at least so far Firefox Quantum has proven a competent replacement for Chrome, so depending on your needs changing to it could be considered a "permanent" solution to this issue, which is what I've done.
,
Dec 10 2017
Can confirm I'm having these atrocious frame drops since the latest "Creators" Windows 10 update on my Core i7 2620M. Tried 6 or 7 different drivers but all the same. In Edge everything works just fine. Chrome is the new IE, it seems.
,
Jan 11 2018
I have a similar problem on my Gigabyte Aero 15 when I used the laptop with external monitor and windows 10 display mode set to duplicate. My workaround is to configure Windows 10 to use the second monitor only. Also, I have configured the laptop to use Intel HD Graphic 630 only to reproduce the scenario even though the laptop also has NVidia Geforce 1060 build in. Also, I won't have any issue if I use extended desktop mode. Lastly, Firefox Quantum and Microsoft Edge browser work fine.
,
Jan 11 2018
The most annoying part is this is reproducible, the cause is known, but there is no effort being made to fix it. Or at least we can't see any work being made towards it. Worse yet, it's being forked into every browser based on this. I'm honestly annoyed I have to use Firefox to get video to play properly in full screen. Comeon, you know the cause and reason, if you aren't going to look into a fix at least create a blacklist rule for this super small power efficiency for these GPU's in a multi monitor setup. Make it so the browser is usable please.
,
Jan 11 2018
I suspect many of the comments here are related to issue 775898, which last we heard Microsoft has a fix for, but hasn't detailed when they'll be releasing it :( See c#116, there are some comments indicating that installing https://support.microsoft.com/en-us/help/4051963/windows-10-update-kb4051963 helps.
,
Jan 11 2018
,
Jan 12 2018
Already on Windows 10 Build 16299.192 and the bug is still there with Chrome Version 65.0.3318.0 (Official Build) canary (64-bit) so doubt the update will do anything.
,
Jan 12 2018
I've read trough Bug #775898 thread and two important details: * --disable-direct-composition is working again in v64 (beta) or up * current version 63 can be fixed by setting compatibility to windows 8
,
Feb 12 2018
I've set the compatibility mode to windows 8 and got the following results: Chrome(Version 63.0.3239.132 (Official Build) (64-bit)): https://i.imgur.com/es21Hac.png Firefox(Quantum 58.0.2 (64-bit)v): https://i.imgur.com/Yz6arEd.png While chrome is now playable, firefox is still significantly outperforming chrome. |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by dalecur...@chromium.org
, May 18 2017