Share this
Optimizing LL-HLS: The Impacts of GOP size on Viewing Experience
by Negar Hajihoseini on September 14, 2021
In the previous blog, we’ve covered the 4 key factors affecting the quality of low latency streaming experience when utilising Apple's LL-HLS protocol. In this blog, we will take a dive into the importance of Group of Pictures (GOP) and its impacts on the overall viewing experience.
As a refresher, the GOP size, or size of your Group of Pictures is one of the main encoding parameters that have a direct impact on video bitrate, video quality and an indirect impact on end-to-end latency. It determines how often a keyframe (or IDR frame) will be available. In LL-HLS, the player requires a keyframe to start decoding, meaning it can start the playback only at GOP boundaries. Longer GOPs cause higher start-up delay and higher latency.
THIS IS A SNIPPET FROM OUR “How to optimize LL-HLS for low latency streaming” BLOG WHICH YOU CAN VIEW HERE.
Impact of GOP size on video bitrate
Apple’s recommended GOP size is 2 seconds. Typical LL-HLS implementations support LL-HLS with 3-second end-to-end latency when the GOP is set to 1 second. However, small GOP sizes come at the cost of higher bandwidth consumption. The smaller the GOP size is, the more frequent the keyframes would be. Depending on the video, keyframes can be 10 times larger than P frames and small keyframe intervals will increase the video bitrate and hence the bandwidth consumption.
In this table, you can see how video bitrate changes with different GOP sizes. For having a comprehensive view, four different types of videos have been tested:
- A movie (Tears of Steal)
- An animation (Big Buck Bunny)
- A bike race TV program
- A static screen streaming
GOP/BitRate Kbps | 0.5 Sec | 1 Sec | 2 Sec | 3 Sec | 6 sec | 10 sec |
Tears of Steel | 4758 | 4060 | 3770 | 3699 | 3554 | 3554 |
Big Buck Bunny | 8105 | 7094 | 6741 | 6687 | 6607 | 6593 |
Bike Race | 6261 | 5504 | 5175 | 5059 | 4966 | 4915 |
Static Video (screen streaming) | 1934 | 1161 | 775 | 652 | 504 | 453 |
Table 1. Bitrate changes on different GOP sizes
For this calculation, the parameter factor CRF (Constant Rate Factor) is kept the same for all GOP sizes forcing the encoder to keep the same video quality in all GOP sizes. As we can see in higher GOP sizes, we can keep the same video quality while using less bandwidth.
Bitrate reduction in large GOP size can differ depending on the type of video. For example, in the static video type (e.g. screen streaming), we have up to 70% reduction in bandwidth consumption from GOP 0.5 seconds to GOP 10 seconds. For other videos, we still have up to a 20% reduction in video bitrate.
Impact of GOP size on video quality
GOP size also has an impact on the video quality. The larger the GOP size is, the higher the video quality will be. Because for the same bitrate we can put more details in the P frames when the GOP size is larger.
We studied how the GOP size affects the video quality. To measure the video quality, we use the VMAF metric. Below is a brief explanation of VMAF.
What is VMAF?
Video Multimethod Assessment Fusion (VMAF) is a video quality metric designed by Netflix consolidating four different metrics:
- Visual Information Fidelity (VIF): considers fidelity loss at four different spatial scales
- Detail Loss Metric (DLM): measures detail loss and impairments which distract viewer attention
- Mean Co-Located Pixel Difference (MCPD): measures the temporal difference between frames on the luminance component
- Anti-noise signal-to-noise ratio (AN-SNR)
VMAF score is ranged between 0 and 100 (100 being identical to the reference video). 6 VMAF points represent a noticeable difference. The VMAF default model is used in this test.
In the table below we depicted how GOP size affects the video quality in different video types. For each encoded video, a VMAF score in comparison to the reference video has been calculated. Depending on the video type, the VMAF score drop in lower GOP sizes is different. Except for the static video streaming, in all the rest of the videos, there is a significant VMAF drop between GOP 10 sec and GOP 0.5 sec. Big Buck Bunny drops by 8 points between GOP 10 sec and GOP 1 sec, which is a noticeable quality degradation.
Please note that in this test the aim was to see the impact of different GOP sizes in the VMAF scores. All videos are encoded at max bitrate 4Mbps. There could be the case that the chosen 4Mpbs is not the highest VMAF scored bitrate for its resolution, but matching the highest VMAF score for each resolution in different videos is out of the scope of this test.
GOP/VMAF | 0.5 Sec | 1 Sec | 2 Sec | 3 Sec | 6 sec | 10 sec |
Tears of Steel | 79.80846 | 83.0559 | 84.5657 | 85.0617 | 85.6186 | 85.7711 |
Big Buck Bunny | 49.5474 | 56.6863 | 61.0857 | 62.0536 | 63.8682 | 64.7515 |
Bike Race | 59.6006 | 64.5885 | 67.2823 | 67.7979 | 68.4479 | 68.9312 |
Static Video (screen streaming) | 95.7727 | 95.7692 | 95.8438 | 95.8537 | 95.8863 | 95.8856 |
Table 2. VMAF scores for different GOP sizes
Based on the VMAF points we see that for some types of videos such as static screen streaming, the quality does not improve that much with large GOP size while you still gain a huge reduction in the bandwidth consumption in large GOP sizes (Table 1). On the other hand, for another type of video such as Big Buck Bunny, the video quality improves up to 15 VMAF points (GOP 10 seconds with respect to GOP 0.5 seconds) which is a considerable amount since every 6 VMAF points is a visually noticeable difference. We also have another pattern for the Tears of Steel video where the VMAF improvement is below 6 VMAF points (between GOP 1sec and GOP 10 sec). In this case, you still have ~20% bitrate reduction in the largest GOP size (Table1).
Impact of GOP size on zapping time and latency
In LL-HLS live streaming, if we increase the GOP size to decrease the bandwidth consumption and increase the video quality, we need to sacrifice the short zapping time and/or the latency.
The player requires a keyframe to start decoding, meaning that a large GOP will impact the zapping time and latency of the stream. It can either wait for the following GOP, implying a long startup time and low latency or it can start playback of the current GOP, implying short startup times, but potential latencies of up to the GOP size. Having large GOPs with only one keyframe every 6 seconds, for example, will mean that the player can start playback on a position once every six seconds. This doesn’t mean your zapping time will be six seconds, but it might require your player to start at a higher latency. With the 6 seconds example, starting playback immediately implies that the average additional latency at the start will be 3 seconds, and in the worst case it can reach up to 6 seconds.
Figure 1. Smaller GOP sizes enable shorter latencies.
In the next blog, we will provide 4 different optimizations for the best low latency streaming experience when utilizing LL-HLS protocol. You can download the complete version of this topic in our “HOW TO OPTIMIZE LL-HLS FOR LOW LATENCY STREAMING” blog here.
Any questions left? Contact our THEO experts
Share this
- THEOplayer (45)
- online streaming (40)
- live streaming (35)
- low latency (32)
- video streaming (32)
- HESP (24)
- HLS (21)
- new features (21)
- THEO Technologies (20)
- SDK (19)
- THEOlive (17)
- best video player (17)
- html5 player (16)
- LL-HLS (15)
- cross-platform (15)
- online video (15)
- SmartTV (12)
- delivering content (12)
- MPEG-DASH (11)
- Tizen (11)
- latency (11)
- partnership (11)
- Samsung (10)
- awards (10)
- content monetisation (10)
- innovation (10)
- Big Screen (9)
- CDN (9)
- High Efficiency Streaming Protocol (9)
- fast zapping (9)
- video codec (9)
- SSAI (8)
- Ultra Low Latency (8)
- WebOS (8)
- advertising (8)
- viewers expercience (8)
- "content delivery" (7)
- Adobe flash (7)
- LG (7)
- Online Advertising (7)
- Streaming Media Readers' Choice Awards (7)
- html5 (7)
- low bandwidth (7)
- Apple (6)
- CMAF (6)
- Efficiency (6)
- Events (6)
- drm (6)
- interactive video (6)
- sports streaming (6)
- video content (6)
- viewer experience (6)
- ABR (5)
- Bandwidth Usage (5)
- Deloitte (5)
- HTTP (5)
- ad revenue (5)
- adaptive bitrate (5)
- nomination (5)
- reduce buffering (5)
- release (5)
- roku (5)
- sports betting (5)
- video monetization (5)
- AV1 (4)
- DVR (4)
- Encoding (4)
- THEO Technologies Partner Success Team (4)
- Update (4)
- case study (4)
- client-side ad insertion (4)
- content encryption (4)
- content protection (4)
- fast 50 (4)
- google (4)
- monetization (4)
- nab show (4)
- streaming media west (4)
- support matrix (4)
- AES-128 (3)
- Chrome (3)
- Cost Efficient (3)
- H.265 (3)
- HESP Alliance (3)
- HEVC (3)
- IBC (3)
- IBC trade show (3)
- THEOplayer Partner Success Team (3)
- VMAP (3)
- VOD (3)
- Year Award (3)
- content integration (3)
- customer case (3)
- customise feature (3)
- dynamic ad insertion (3)
- scalable (3)
- server-side ad insertion (3)
- video (3)
- video trends (3)
- webRTC (3)
- "network api" (2)
- Amino Technologies (2)
- Android TV (2)
- CSI Awards (2)
- Encryption (2)
- FireTV (2)
- H.264 (2)
- LHLS (2)
- LL-DASH (2)
- MPEG (2)
- Microsoft Silverlight (2)
- NAB (2)
- OMID (2)
- Press Release (2)
- React Native SDK (2)
- Start-Up Times (2)
- UI (2)
- VAST (2)
- VP9 (2)
- VPAID (2)
- VPAID2.0 (2)
- ad block detection (2)
- ad blocking (2)
- adobe (2)
- ads in HTML5 (2)
- analytics (2)
- android (2)
- captions (2)
- chromecast (2)
- chromecast support (2)
- clipping (2)
- closed captions (2)
- deloitte rising star (2)
- fast500 (2)
- frame accurate clipping (2)
- frame accurate seeking (2)
- metadata (2)
- multiple audio (2)
- playback speed (2)
- plugin-free (2)
- pricing (2)
- seamless transition (2)
- server-side ad replacement (2)
- subtitles (2)
- video publishers (2)
- viewer engagement (2)
- wowza (2)
- "smooth playback" (1)
- 360 Video (1)
- AOM (1)
- API (1)
- BVE (1)
- Best of Show (1)
- CEA-608 (1)
- CEA-708 (1)
- CORS (1)
- DIY (1)
- Edge (1)
- FCC (1)
- HLS stream (1)
- Hudl (1)
- LCEVC (1)
- Microsoft Azure Media Services (1)
- Monoscopic (1)
- NAB Show 2016 (1)
- NPM (1)
- NetOn.Live (1)
- OTT (1)
- Periscope (1)
- React Native (1)
- Real-time (1)
- SGAI (1)
- SIMID (1)
- Scale Up of the Year award (1)
- Seeking (1)
- Stereoscopic (1)
- Swisscom (1)
- TVB Europe (1)
- Tech Startup Day (1)
- Telenet (1)
- Uncategorized (1)
- University of Manitoba (1)
- User Interface (1)
- VR (1)
- VR180 (1)
- Vivaldi support (1)
- Vualto (1)
- adblock detection (1)
- apple tv (1)
- audio (1)
- autoplay (1)
- cloud (1)
- company news (1)
- facebook html5 (1)
- faster ABR (1)
- fmp4 (1)
- hiring (1)
- iGameMedia (1)
- iOS (1)
- iOS SDK (1)
- iPadOS (1)
- id3 (1)
- language localisation (1)
- micro moments (1)
- mobile ad (1)
- nagasoft (1)
- new web browser (1)
- offline playback (1)
- preloading (1)
- program-date-time (1)
- server-guided ad insertion (1)
- stream problems (1)
- streaming media east (1)
- support organization (1)
- thumbnails (1)
- use case (1)
- video clipping (1)
- video recording (1)
- video trends in 2016 (1)
- visibility (1)
- vulnerabilities (1)
- zero-day exploit (1)
- August 2024 (1)
- July 2024 (1)
- January 2024 (1)
- December 2023 (2)
- September 2023 (1)
- July 2023 (2)
- June 2023 (1)
- April 2023 (4)
- March 2023 (2)
- December 2022 (1)
- September 2022 (4)
- July 2022 (2)
- June 2022 (3)
- April 2022 (3)
- March 2022 (1)
- February 2022 (1)
- January 2022 (1)
- November 2021 (1)
- October 2021 (3)
- September 2021 (3)
- August 2021 (1)
- July 2021 (1)
- June 2021 (1)
- May 2021 (8)
- April 2021 (4)
- March 2021 (6)
- February 2021 (10)
- January 2021 (4)
- December 2020 (1)
- November 2020 (1)
- October 2020 (1)
- September 2020 (3)
- August 2020 (1)
- July 2020 (3)
- June 2020 (3)
- May 2020 (1)
- April 2020 (3)
- March 2020 (4)
- February 2020 (1)
- January 2020 (3)
- December 2019 (4)
- November 2019 (4)
- October 2019 (1)
- September 2019 (4)
- August 2019 (2)
- June 2019 (1)
- December 2018 (1)
- November 2018 (3)
- October 2018 (1)
- August 2018 (4)
- July 2018 (2)
- June 2018 (2)
- April 2018 (1)
- March 2018 (3)
- February 2018 (2)
- January 2018 (2)
- December 2017 (1)
- November 2017 (1)
- October 2017 (1)
- September 2017 (2)
- August 2017 (3)
- May 2017 (3)
- April 2017 (1)
- March 2017 (1)
- February 2017 (1)
- December 2016 (1)
- November 2016 (3)
- October 2016 (2)
- September 2016 (4)
- August 2016 (3)
- July 2016 (1)
- May 2016 (2)
- April 2016 (4)
- March 2016 (2)
- February 2016 (4)
- January 2016 (2)
- December 2015 (1)
- November 2015 (2)
- October 2015 (5)
- August 2015 (3)
- July 2015 (1)
- May 2015 (1)
- March 2015 (2)
- January 2015 (2)
- September 2014 (1)
- August 2014 (1)