Proof of Transcoding as a New Consensus Mechanism

Vivid Labs Team
VideoCoin
Published in
6 min readOct 2, 2019

The term ‘Proof of Transcoding’ sounds confusing if you’re not familiar with blockchain technology or video transcoding. Actually…. it’s complicated even if you ARE familiar with the term. However, if you have an understanding of both of these two different technologies, proof of transcoding makes a lot of sense for an actual use case of blockchain technology.

You may be thinking, what does video transcoding have to do with blockchain technology? Wait a minute, what is transcoding? Does video need to be decentralized too? How would a consensus mechanism technically work as proof of transcoding? If you finish this blog post, you will have a general understanding of the answers to all of these questions.

What is transcoding?

Transcoding is a process¹ that allows you to watch a video across many different platforms and devices.

What was the last video you watched?

What device did you watch it on?

If it was a computer, what browser were you using?

On the surface, it does not seem like it’s a complicated process to make this happen. Isn’t the video just a single file format? However, this is not possible, because, behind the scenes, each device has its codec, colorspace, and resolution requirements.

Transcoding is the process that takes a digital source video file and converts it into different file formats to match different device requirements. Today some of the most utilized codecs² are H265, H264, VP9, and VP8 because they achieve high degrees of compression.

High-quality compression is essential because it transforms large video files into smaller files while preserving a majority of the quality of the original file. Smaller video files require less storage, which makes it easier to distribute them across many platforms, and makes the end-user playback experience seamless.

While transcoding successfully makes large files smaller, this process requires a significant amount of computer processing power and time. Also, you can’t just transcode a file one time. You need to repeat the process to accommodate ALL of the different file formats for distribution.

Transcoding a video file that has been pre-recorded is a labor-intensive process, which means transcoding video files recorded in real-time to multiple high compression codecs is exponentially more complicated.

Would video streaming benefit from being decentralized?

The modern cloud video infrastructure powering video applications on the internet consists of a standard set of three technology building blocks: transcoding, storage, and distribution. The video streaming process is distributed between these three blocks of infrastructure.

It makes sense to distribute different parts of this complex process across multiple resources. In other words, make this process decentralized to maximize efficiency.

Transcoding — The process of converting video files to fit the format requirements of different video players and devices. Modern video infrastructure makes use of codecs like H265, H264, VP9, and VP8 to achieve high degrees of compression, but it comes at the cost of high CPU requirements.

Storage — Video files are occupying more storage space due to increased resolution. 4K videos are the favored file format for high-end devices. Consumers view videos on a staggering number of devices and screens, and encoding for each device is compute-intensive. Since transcoding large video files is compute-intensive, videos are typically transcoded once and stored in a storage network for later consumption.

Distribution — To successfully provide a good user experience that avoids long buffering times, videos typically are cached geographically close to the end-consumer. This quality user experience is achieved by distributing chunks of video over the internet using servers that are near the consumer. Current video infrastructure uses content delivery networks to implement this function.

How would a consensus mechanism technically work as proof of transcoding?

If you’re familiar with the mechanics of bitcoin, you know that the consensus mechanism is³ called proof of work (PoW). It requires a participant node to prove that the work done and submitted by the worker qualifies to receive the right to add new transactions to the blockchain.

Another popular consensus mechanism is proof of stake (PoS)⁴, which is used by Ethereum. It involves allocation of responsibility in maintaining the public ledger to a participant node in proportion to the number of virtual currency tokens held by it.

The consensus mechanism for proof of transcoding would utilize decentralized versions of the three technology building blocks of video infrastructure. It would create a new type of video miner. These miners would compete with other miners to gain incentives by contributing CPU and/or GPU cycles for the three blocks of video infrastructure, which are for transcoding, disk space for video storage, and internet bandwidth for content distribution network services.

How could proof of transcoding improve video streaming?

As explained above, large scale video processing is a complex distributed systems problem. Transcoding videos to match several devices, resolutions, and codecs require a lot of computing resources. While Moore’s law⁵ has been lowering prices of individual CPUs every day, the computing capacity of CPUs has yet to catch up. Thus, the necessary power needed to transcode video is still needed.

The consensus mechanism proof of transcoding would give video miners the chance to run on data center servers, household desktop computers, or mobile phones, creating a powerful network of computers that would otherwise go unused. It would incentivize miners to apply any unused resources to provide as much computing, storage, and bandwidth currently available. These extra resources would help overcome the gap of Moore’s law for live streaming video.

Future of video streaming and proof of transcoding

Ultimately, blockchain technology will help improve many different types of distributed system difficulties. Proof of transcoding has the potential to be a game-changer for the quality and quantity of video streaming and particularly live streaming.

¹ “What is Transcoding? — Definition from Techopedia.” https://www.techopedia.com/definition/973/transcoding. Accessed 9 Sep. 2019.

² “Codec: What It Is, Why You Need One, and How to Get It.” 21 May. 2019, https://www.lifewire.com/what-exactly-is-odec-2483426. Accessed 9 Sep. 2019.

³ “Proof-of-Work, Explained | Cointelegraph.” 17 Jan. 2018, https://cointelegraph.com/explained/proof-of-work-explained. Accessed 9 Sep. 2019.

⁴ “Proof of Stake (PoS) Definition — Investopedia.” 11 Aug. 2019, https://www.investopedia.com/terms/p/proof-stake-pos.asp. Accessed 9 Sep. 2019.

⁵ “Moore’s Law Definition — Investopedia.” 4 Sep. 2019, https://www.investopedia.com/terms/m/mooreslaw.asp. Accessed 9 Sep. 2019.

--

--

Vivid Labs Team
VideoCoin

Creators of VIVID, the next generation NFT publishing platform that allows anyone to create, manage, and sell multimedia NFTs.