codec

small_emilien.jpeg

thumbnail_emilien.jpeg

emilien.jpeg

Emilien Gobillot

large_api-video-blog-company.png

medium_api-video-blog-company.png

small_api-video-blog-company.png

thumbnail_api-video-blog-company.png

api-video-blog-company.png

Company News

Api.video started with a mission to empower product builders to effortlessly integrate video into their applications or services. To achieve this, we built and provided a global video backbone with an easy-to-use API for video delivery, transcoding, and live streaming. With this API, our customers are free to upload their videos at will and count on us for them to be quickly available for delivery. However, this becomes a formidable challenge when customers inundate us with their video catalogues at once.

&nbsp;

In tackling this challenge in the past, we initially relied on the most common solution: CPUs. But the limitation with CPUs lies in their inefficiency for video transcoding, particularly with the advent of new video codecs. 

&nbsp;

We had to overcome this challenge and so we started looking at other alternatives: GPUs, FPGA and ASICs.

&nbsp;

## FPGA vs. GPU vs. ASICS: A comparison

When we started with our research on FPGA, GPU and ASICs, we gathered information from technical marketing presentations on some products (i.e Intel flex series cards not yet available on the market at that time) and ran benchmarks in our lab. Mixing real and marketing measures could have been misleading. The Codensity G4 and G5 chips from [Netint](https://netint.com/) turned out to be clear winners by an order of magnitude when it came to power efficiency and performance per dollar.

large_Watt and Cost($) per GigaPixel_s (Lower is better) (1).png

medium_Watt and Cost($) per GigaPixel_s (Lower is better) (1).png

small_Watt and Cost($) per GigaPixel_s (Lower is better) (1).png

thumbnail_Watt and Cost($) per GigaPixel_s (Lower is better) (1).png

Watt and Cost($) per GigaPixel_s (Lower is better) (1).png

So, we decided to continue the theoretical study with real benchmarks with NETINT VPUs (video processing units) and transcoders, to validate this decision.

&nbsp;

## The journey of switching from CPU to ASICs

The decision to choose ASICs for [our video infrastructure](https://api.video/blog/company/api-video-proprietary-infrastructure/) came after rigorous research.

&nbsp;

Our journey began with the conventional use of CPUs. But as the demand for enhanced performance grew, our team delved into an extensive study, exploring and comparing transcoding solutions like FPGA, GPU, and ASICs. 

&nbsp;

Recognizing the need for a solution that aligned with our goals, we benchmarked the [chips from Netint as their product was the clear winner of our study](https://api.video/blog/company/asics-based-video-encoding/). The G4 performances were satisfying. We could have gone into production with the T408 cards based on G4, but NETINT’s Quadra G5-based transcoders were even more appealing with its AV1 support and higher density. The AI chips associated with the G5 on the T1U cards were also giving it some bonus points. 

&nbsp;

We then restarted our benchmarks with the G5-based Quadra T1U cards based on the G5 chips to assess their performance. We also tested different transcoding strategies to get a high transcoding speed and a good density per server. The results matched our expectations and we successfully finalised these benchmarks with the best transcoding strategy adapted to our transcoding video pipeline.

&nbsp;

This decision marked a turning point in our infrastructure, opening doors to greater efficiency. With increased capacity and advanced encoding capabilities, we went into production.

&nbsp;

## How api.video reduced encoding costs with Quadra VPUs 

api.video used the T1U cards based on the G5 chips built by Netint Technologies. These cards are called VPUs (Video Processing Unit). Let's understand the costs that come with them when we consider a 1080p30 video. To simplify the case, I will assume the source video codec is h.264. The output videos after transcoding will be five h.264 renditions for ABR: 1080p30, 720p30, 480p30, 360p30, 240p30.

&nbsp;
 
For a real time transcoding, let's try to sum up the number of MP/s (Megapixel per second) it represents. For the above output, it represents 62 MP/s + 28 MP/s + 12MP/s + 7MP/s + 3MP/s respectively for 1080p30, 720p30, 480p30, 360p30, 240p30 renditions. 

&nbsp;

We need an encoding capacity of 112 MP/s for real time transcoding. Let us assume that the transcoding cost on VPUs is proportional to the number of pixels per frame (whatever the frame resolution) for the sake of simplicity and one VPU of Netint is able to encode 2000 MP/s in h.264. In reality it is a bit more complex but let’s take these as the numbers for a rough idea.

&nbsp;
 
Now, as per the above assumptions, we can encode 2000/112 = 17 minutes of videos (all rendition included) per minute. Below is a table summarizing the minute cost for a server with 4 VPUs. The cost of the server is $10k and the amortization for such a server is 3 years. 

&nbsp;

The electricity cost will be approx. $1000 per year (420w at $0.30 / kWh) at full consumption. Let us see what is the cost of 1 minute of a 1080p30 video transcoded to 5 renditions.

&nbsp;

(The cost is split taking into consideration the average usage of the server per day)



| AVG. USAGE OF THE SERVER IN PERCENTAGE | MINUTES ENCODED OVER THE AMORTIZATION PERIOD | COST PER MINUTE TRANSCODED (MINUTES / COST OF SERVER + ELECTRICITY) IN $ |
| -------------------------------------- | -------------------------------------------- | ------------------------------------------------------------------------ |
| 0,50%                                  | 536112                                       | 0,020                                                                    |
| 5%                                     | 5361120                                      | 0,002                                                                    |
| 25%                                    | 26805600                                     | 0,0004                                                                   |
| 50%                                    | 53611200                                     | 0,0002                                                                   |
| 90%                                    | 96500160                                     | 0,0001                                                                   |
<p style="margin-top: -1em" class="text-AV_Primary_Blue-500/80 pb-5 text-center text-sm">Transcoding cost per minute based upon varying usage percentages</p>



If we use the server at only 0.5% per day (i.e ~7 min per day fully loaded) the cost per minute would be at $0,020.

&nbsp;
 
If on an average, this server is loaded at 25% (6 hours a day fully loaded) the cost per minute would be 1/100 of our previous price.

So, even at 5% utilization, the cost per video is immaterial, allowing us to make the strategic decision to offer transcoding services at no charge to our customers.

&nbsp;

## What about video quality?

In the past, hardware encoders were associated with lower-quality video, but tests performed by us prove that this isn’t the case with the Quadra VPU. Specifically, we encoded a 1080p30 version of the movie Caminandes 2 via our API and compared output quality with a video encoded with FFmpeg and the x264 codec using the following command string.

&nbsp;

**ffmpeg -r 30 -y -i input.mp4 -c:v libx264 -force_key_frames 'expr:gte(t,n_forced*4)' -b:v 4400k -preset medium -c:a copy output.mp4**

&nbsp;

As shown in the table below, [we compared the two videos](https://api.video/blog/product-updates/video-encoding-numbers/) using well-known quality metrics VMAF, PSNR, and SSIM. The scores are all extremely close, with the VPU-encoded video edging FFmpeg in VMAF scoring and the reverse for PSNR and SSIM. Given the closeness of the scores, however, it’s clear that no viewer would notice the difference.


|                  | VMAF (AVG) | PSNR (AVG) | SSIM (AVG) |
| ---------------- | ---------- | ---------- | ---------- |
| api.video (VPUs) | 95.92      | 48.07      | 0.9888     |
| ffmpeg (CPUs)    | 95.33      | 48.69      | 0.990      |
<p style="margin-top: -1em" class="text-AV_Primary_Blue-500/80 pb-5 text-center text-sm">Transcoding cost per minute based upon varying usage percentages</p>

To summarize, api.video builds its own infrastructure so that we can take advantage of the right technology to offer the best to our customers. With these advanced VPUs in our infrastructure, we are easily able to reduce our transcoding TCO. We can pack a lot of VPUs in a single server which increases transcoding capacity and reduces the power consumption and the number of servers by an order of magnitude.

&nbsp;
 
In a nutshell, VPUs help us encode videos at scale without costing us too much. These VPUs are the reason why api.video was able to make the decision to offer [video encoding for free](https://api.video/blog/product-updates/video-encoding-pricing-update/) to our customers.

 &nbsp;

_To start using api.video for your videos, [sign-up](https://dashboard.api.video/register) for a free sandbox account now._

Beyond CPUs: api.video’s tech tale of switching from CPUs to ASICs

large_netint.png

medium_netint.png

small_netint.png

thumbnail_netint.png

netint.png

Normal

In today's world, where environmental responsibility is important, businesses are re-evaluating their impact on the planet. Sustainable practices are not just a choice; they're a responsibility. When it comes to video encoding, a crucial aspect of content delivery in our digital age, there's a technology that aligns perfectly with this responsibility - ASICs. These Application-Specific Integrated Circuits (ASICs) not only promise more efficient video encoding but also contribute majorly to a lower carbon footprint.

&nbsp;

That’s why, api.video has gone forward and switched from CPU based encoding to ASICs based encoding.

&nbsp;

This blog is intended to tell you more about this switch. 

## The rise of ASICs in the video industry

In the past few years, the internet's growth sparked a rise in digital video demand. As platforms like YouTube and Netflix rose, companies leaned more toward CPU-based encoding due to its versatility. However, as streaming services gained popularity, CPU encoding's limitations started getting noticed. 

&nbsp;

Video encoding relying on CPU power is no longer a sustainable approach, as superior alternatives are now available that come without any compromises. A great example of this is Google. Google has developed its own Video Processing Unit (VPU) known as [Argos](https://9to5google.com/2021/04/22/youtube-google-custom-chip/) – a groundbreaking innovation. Through the integration of these VPUs, Google has significantly amplified transcoding capabilities per server, cut down operational expenses, reduced electricity consumption, and ultimately improved the Total Cost of Ownership (TCO). Following in Google's footsteps, Meta has also embraced this transformation. 

&nbsp;

What does this mean? It simply means that ASICs reduces the resource consumption of video encoding (see graph below) and that’s why companies are adopting it. 



MPixel_s per Watt encoded to H.264 (Higher is better)(1).png

Now what is the secret behind these VPUs that are replacing CPUs? VPUs are, in fact, ASICs-based chips specifically engineered for a singular purpose: decoding and encoding video frames. They support various codecs such as h.264, vp9, and AV1, among others. 

&nbsp;

Their power consumption is lower than that of CPUs, and their capacity is substantially greater. Consequently, you can expect VPUs to reduce electricity consumption by 10x to 100x, depending on the codec and the specific VPU in use.

&nbsp;

The best part is that the quality of the video output in VPUs matches that of CPU encoding. (We will delve into the technical details in an upcoming blog post.)

## What made us start ASIC based encoding at api.video

- ASICs also known as VPU (Video Processing Unit) are one order of magnitude more power efficient than CPU based transcoding. 

&nbsp;

- Using dedicated hardware allows us to increase the density which in turn reduces our environmental footprint. As a corollary, it increases our transcoding capacity.

&nbsp;

- Specialised hardware helps improve the TCO (Total Cost of Ownership). When we increase the capacity and density of a server, there is a fewer requirement of servers and less rack space. So it reduces capital expenditure and operational expenditure. 

## What does this mean for you as a customer

In today's environmentally conscious world, the impact of technology on our carbon footprint is a growing concern. This is where ASIC becomes a game-changer. 

&nbsp;

But, you might wonder, what can you, as a customer, gain from embracing ASICs technology?

&nbsp;

- **First and foremost, ASIC brings sustainability to the forefront.** By significantly reducing energy consumption compared to generic hardware, ASIC chips align perfectly with your green initiatives. Your carbon footprint shrinks, making a substantial contribution to a better environment.

&nbsp;

- **Secondly, [you get encoding at ZERO cost ](https://api.video/blog/product-updates/video-encoding-pricing-update/) with api.video.** Yes, it is because of ASICs that we are able to reduce our TCO and offer you video encoding at no cost. This means that if you choose api.video for video streaming, you save on the cost of encoding completely. 

&nbsp;

- **Thirdly, ASIC's efficiency translates to faster video processing.** This means that with ASIC’s optimized infrastructure, your complex videos are encoded faster. 

## Stepping into a greener future with ASICs

To create the environmentally responsible future we desire, changes in video encoding and distribution are imperative. ASICs-powered encoders represent a breakthrough, offering enhanced density, performance, and cost-efficiency. These advancements will ensure video applications to scale economically while significantly reducing their environmental footprint.

&nbsp;

With this initiative, let us pledge to make a greener and a better environment around us!

&nbsp;

_[Sign-up](https://api.video/) with api.video today or [talk to our team](https://meetings.hubspot.com/apivideo/talk-to-us) for more information._


Introducing ASICs-based encoding at api.video

small_ASIC_eco.jpg

thumbnail_ASIC_eco.jpg

ASIC_eco.jpg

Video trends

**The comeback of specialized hardware and its impact on the video industry**

Over the past few decades, the trend in computing has been toward general-purpose computing systems that can run a wide variety of applications. This has led to the widespread use of [CPUs and GPUs](https://www.intel.com/content/www/us/en/products/docs/processors/cpu-vs-gpu.html). However, there has been a recent resurgence of interest in specialized hardware, which is designed specifically to perform one or a few tasks extremely efficiently. In this blog post, we will explore the reasons behind this trend and some of the critical examples of specialized hardware being used in the video industry today.

**Why is specialized hardware making a comeback in computing?**

Specialized hardware in the computing industry refers to computer components or devices designed for a specific purpose, such as application-specific integrated circuits (ASICs)  or systems-on-a-chip (SoCs) . These hardware components are optimized for their specific task, resulting in faster and more efficient performance than general-purpose hardware.

Several factors have contributed to the recent resurgence of specialized hardware in computer science:

* **High efficiency : Why we want it quick and ressource-efficient**

One key driver is the increasing demand for high-performance solutions. Artificial intelligence, machine learning and video processing are domains where CPUs are far from being the right solution. Regarding AI workloads, training, and running models are more efficient on GPU and even better on specialized hardware like Tensor processing units (TPUs). Google has been the first to design TPUs. These ASICs (Application Specific Integrated Circuits) outperform the CPUs.

Why are CPUs outperformed ? In early 2023 commodity servers come with 1 or 2 sockets and maximum 96 cores per socket for the x86 microarchitecture (128 for ARM). On the other hand, a server with a mid range CPUs can receive several ASICs/TPUs on the PCIe ports. Those boosted systems outperforms by a lot the CPUs based system. The TPUv4, in terms of Tflops is 2000x faster than a laptop according to Google.

The benefits of a specialized hardware is that it provides better performance, higher density and a lower energy consumption than the classic CPUs.

In most cases, TPU/ASIC > GPU > CPU on the 3 criteria : performance, density, and energy savings.

* **Total Cost of Ownership (TCO): Why It's Worth the cost**

Specialized hardware exhibits smaller size, higer speed, and greater electrical efficiency. As a result the density increases and the number of computing units decreases. At the end, the TCO is improved. After the switch to ASICs based VCUs (Video coding unit), Youtube is [reported](https://www.semianalysis.com/p/google-new-custom-silicon-replaces) to have saved millions of CPUs ! At this scale you also save data centers 💰

**How specialized hardware is changing the game in the video industry**

In the video industry, one of the most computationally intensive processes is video transcoding, especially for high-quality videos. Transcoding decodes a video from one codec and encodes it to another, such as from H.264 to H.265. This process can also involve changing other video parameters like resolution, bit rate, frame rate, etc. This has led to a need for fast and efficient video transcoding solutions without sacrificing the video quality output.

Transcoding is a service api.video provides in addition to hosting and delivery. Via an API call, a video is uploaded, and then the URL of the video is returned immediately. Note that high resolution and new codecs increase the computational cost.

Similarly to Artificial Intelligence (AI), several types of hardware are commonly used in video transcoding applications:

1. ASICs (Application Specific Integrated Circuits): Video ASICs are specialized hardware accelerators designed specifically for video processing tasks, such as video encoding/ decoding / transcoding. They are sometimes called VPU (Video Processing Unit) or VCU (Video coding Unit). They significantly improve performance over traditional CPUs, particularly for high-resolution or high-frame-rate video. Alphabet, the mother company of Google and YouTube has designed its own ASICs : codename Argos. For a system with 20 VCUs, the TCO has decreased by a 33x factor for VP9 encoding, and throughput increased by a 100x factor!

Facebook/ Meta made the same move to specific hardware. They partnered with silicon vendors to [design custom ASICs](https://engineering.fb.com/2019/03/14/data-center-engineering/accelerating-infrastructure/).

The electric consumption of an ASIC is around a few watts.

2. FPGAs (Field-programmable gate arrays): FPGAs are programmable hardware devices that can be customized for specific applications, including video transcoding. They can provide significant performance improvements over traditional CPUs for certain types of video transcoding tasks, such as simultaneously encoding or decoding multiple video streams. Twitch is using an [FPGA from Xilinx](https://blog.twitch.tv/en/2018/12/19/how-v-p9-delivers-value-for-twitch-s-esports-live-streaming-35db26f6322f/).

The electric consumption is around 25 W for a PCIe card

3. Graphics processing units (GPUs): GPUs are another type of hardware. This hardware is more generic than FPGA and ASIC, but specialized video encoders/decoders are added to these cards. Nvidia was the first company to build a datacenter GPU for transcoding. Intel enters the market in late 2022.

The electric consumption is generally around 70W.

Regarding energy consumption ASICs > (are better than) FPGAs > GPU > CPU. This order is the same for the transcoding speed.

**Reducing the carbon footprint of video transcoding : api.video’s approach**

At api.video we are really focused on the energy consumption of our transcoding pipeline. We want to minimize the ecological impact of video transcoding while keeping the transcoding speed high. The infrastructure team has performed multiple benchmarks, and here are the results.

Those numbers are based on public information (marketing sheet) for the GPUs and FPGAs and real benchmarks for the CPU and ASICs. The quality of the video's output could change the results slightly, but we clearly see a winner regarding power efficiency. In a future blog post, we will share precise benchmarks about the ASICs vs. CPU battle with you. Stay tuned.

Along this article, we have seen the resurgence of dedicated hardware in computer science is driven by a growing need for high-performance computing, particularly in fields like artificial intelligence, machine learning, and video transcoding. Dedicated hardware can provide better performance and energy efficiency than traditional general-purpose computing units, making it an attractive option for many applications. As computing continues to evolve, we will likely see more specialized hardware being developed for specific tasks, driving innovation and progress in the field. 

Our video API allows you to add videos to your products while prioritizing the reduction of your carbon footprint. [Join us](https://api.video/) and take part of the transition to an environmentally responsible video infrastructure.

The comeback of specialized hardware

Frame 427319155.jpg

Modern technology has been helpful for an increasingly fast and simple method of consuming various types of video media, whether it’s a YouTube video, our favorite shows on Netflix, or even catching up with friends across the globe on Zoom. Despite the relative ease of doing anything related to video, a few complex processes take place behind the scenes, away from public view. One such process is video encoding.

Video encoding basically makes the world go around in regards to anything related to video and video content. Thanks to the technologies behind the video encoding process, it has become much easier for people and companies to share video content on the internet. But what exactly is video encoding, and what happens during the process? We’ll take a closer look at this and much more in this handy guide to video encoding.

## What is video encoding?

In a nutshell, video encoding occurs when you take raw video footage, either from a digital camera or an analog film camera, and digitally compress it to make it a more manageable file size. Video encoding happens daily in both professional and personal capacities, and many people don’t realize it.

But, without encoding, services like Netflix and YouTube would not exist – or, at least, not in the same way that it does today. Without it, we couldn’t send videos of our beloved pets to our friends and family as easily as we can today.

## Why is video encoding used?

There are a variety of reasons why video encoding is used, but all factors point to the single function of video encoding – **compression**.

This is done to ensure optimal performance and compatibility with a selected suite of software and hardware, be it for DVD/Blu-ray playback, mobile viewing, online video streaming, or general video editing. A file's video and audio information undergo transformation and compression during the encoding process, all in accordance with the requirements of the selected encoding standard.

On a professional level, the ability to compress raw video into more manageable sizes makes it easier to transmit video content such as training, promotion, presentation, and so on over the internet. 

On a personal level, video encoding makes it much easier to send videos taken from your phone directly to friends and family. Most modern-day personal devices have built-in video encoding capabilities to allow for this to happen without a second thought.

To put things in perspective, uncompressed or unencoded video can have a data rate that’s as low as 3,000 Mbps and as high as 12,000 Mbps, depending on the length and quality of the video recorded. 

**Without encoding, it would take up to days for anyone to not just transmit video, but to be able to consume it, too.**

## Differences between video encoding and video transcoding

You need to be familiar with two more [essential terms related to video encoding](https://api.video/what-is/) – transcoding and codecs.

Transcoding relates to the conversion from one digital encoding format to another. It takes the already encoded video file, decompresses it, and then alters it before re-encoding or compressing it into a different codec. Codecs are what allow encoding to occur in the first place. They are software or hardware that gives us the ability to compress video to make them more easily stored and delivered tightly.

While you may hear “encoding” and “transcoding” used interchangeably, the two terms actually mean different things. Putting it succinctly, encoding is the initial process of compressing a video from a source – such as a camera – into a codec, while transcoding takes an already encoded video file to convert it into a different codec.

## Examples of video codecs

In order to understand video encoding, you must first be familiar with two interrelated terms: [video codecs and video formats](https://api.video/blog/product-updates/every-video-format-codec-and-container-explained/). These also tend to be used interchangeably, but much like the terms encoding and transcoding, they refer to two different things.

Video formats are the type of files in which the compressed video/audio is stored. Also known as a “container”, you can tell what a video format is by looking at the file’s extension — the three to four characters that sit after a period in a file name. Here are some of the most common video formats used today:

* MP4
* FLV
* MOV
* MKV
* LXF
* MXF
* AVI
* WebM
* QT

On the other hand, **the codec is short for “Coder / Decoder” (compression / decompression)** and relates to the video compression algorithms that allow encoding to happen. Over the years, codecs have gone through multiple iterations as it coincides with the increasing sophistication and rise in the quality of filmmaking. 

If you’re a content creator, you might be aware of some of the video codecs that exist today:

- **H.264 – also known as AVC, this is the most widely-used video codec available on every platform and mobile device.**
- H.265 – also known as HEVC, stands for High-Efficiency Video Coding, and it’s slowly overtaking H.264 as the standard. This codec can compress video much more efficiently than its predecessor.
- **VP9 – is an open-source, royalty-free codec developed by Google and is considered the direct competitor of H.265. The video compression codec VP9 was created to reduce file sizes, making streaming videos simpler.**
- **AV1 – is another royalty-free, open-source codec released in 2019 by the non-profit Alliance for Open Media. It was developed to become the successor to Google’s VP9 and contend with HEVC more directly.**

## How do different technologies handle video encoding

As you may have suspected, there are a majority of ways that different technologies handle the video encoding process. Let’s take a look at the most common technologies that use video and how video encoding works on each of them:

### Video editing software

When it comes to video content creation, a majority of people will use one of two video editing software — Adobe Premiere Pro or Apple’s Final Cut Pro. Both of these video editing software handle video encoding in very similar ways: when exporting the finished video.

Depending on which device the video content is recorded on, it will already be encoded and formatted according to the settings of the device. For those using iPhones recording 1080p video at 25 frames per second (FPS), for example, the iPhone will automatically encode the video using H.264 and contain it in a .mov file format. The video editing software will “open up” the container to get to the encoded, compressed video file within so the editing process can begin.

After editing the footage is complete, it will need to be exported, which typically concerns video encoding. The edited raw media taken from a recording device such as a camera will be encoded via the codec of the person’s choosing before being put into a .mp4 container once they’ve hit “export”. As we’ve mentioned before, the most common codec used in video content creation is H.264 for HD videos, with 4K videos using H.265. 

### Websites

Different websites utilize different techniques to handle video encoding – with a majority of the major sites like YouTube, Netflix, and so on using their own technologies to handle the process. A majority of social media websites allow you to upload video files directly onto the site to share, although most of them require the files to be saved in specific containers. 

These websites will take your video file and transcode it before using their own systems and serves to re-encode it to match multiple display resolutions. This process will happen regardless of your video file's original codec and settings. Different websites have different processes, but companies that don’t have the capacity to build their own video encoding systems can use a third-party application programming interface (or API) to handle this process. These APIs act as [dedicated video encoder](https://api.video/what-is/video-encoding/) that allows companies to add on-demand videos to their websites more efficiently.

Using [api.video](https://api.video) as an example, all of the videos submitted to the platform will be transformed into H.264 codec in an MP4 container. This may be done using transmuxing and/or transcoding.

We expect the video bitrate to be lower than 16 Mbps for up to 1080p resolution and 50 Mbps for up to 4K resolution. We accept any bitrates, however, we will scale them down to the limits mentioned because anything above is usually challenging for most viewers’ connection to watch.

### API

The ability to generate, store, and distribute video is made possible via a video API. [api.video](https://api.video)’s solution can be tailored to suit any specific need. 

The backend process of video on demand (VOD) is exceptionally complex, even if you’re a software programmer specializing in video. It is necessary to encode, package, and typically transcode the data before sending it to any of several possible destinations. All of these things, and more, may be done with minimal fuss using api.video.

Our API basically keeps the complexities of streaming media hidden from view, so you can rapidly develop apps with as much control as you need. This saves time and effort so developers can concentrate on making the product.

## Conclusion

As you can see through this video encoding guide, a lot of work goes into making the video you watch and the video you share possible. Without video encoding compressing raw video footage, it could take days to upload and transmit video content over the internet and perhaps even longer to be able to download and enjoy them. 

The best part about encoding at api.video is that is offered at ZERO cost. You will only be charged for delivery and hosting of your videos, and the encoding will be free. Now integrate your videos in minutes with api.video or [contact us](https://api.video/contact/) for more information.

A Guide to Video Encoding & Video Codecs

poster-3.png

Product updates

Once I upload my video, how quickly will it be ready to be viewed?

Our reply to this question has always been “it's really fast.”  And it is really fast. But, the dev team gave me access to the backend database, and pointed me to a few tables that might be helpful (thanks to Thomas Persohn for your patient help!)

Thus armed, I waded into a sea of data to better understand how fast our videos are encoded at api.video

## Encoding

When a video is uploaded to api.video, it is re-encoded to a HLS stream, with several different video dimensions and bitrates.  This allows api.video to deliver properly sized video to all users - no matter the device or network connection. (Read our post on [Responsive video delivery](https://api.video/blog/video-trends/responsive-video-delivery/) to learn more.)

> Summary:  All videos are encoded at 240p, 360p, 480p, 720p, 1080p and 2160p. However, the videos are only encoded up to the initial size of the video (so if the original video is 720p, there will not be 1080p or 2160p versions available). 

**FYI:** For our customers, encoding is absolutely free – we offer adaptive bitrate encoding at zero cost. 

How long does this step take? If a customer uploads a video to our service - they want to know how efficient api.video creates the content for viewing.

### Data in the study

We are constantly making adjustments to our infrastructure to speed the encoding step of our video (and that is one reason this post is dated 2H 2020 - we expect to be faster in early 2021). The last major adjustment was made in mid October, so the data presented in this study from every video uploaded to the api.video service between 15 October and 2 December 2020.

### Parameters that affect encoding

There are 2 major parameters that affect the speed of encoding a new video stream: the output dimensions and the length (duration) of the video.  

#### Dimensions

The output video dimensions have a huge impact on video creation - a 3840 × 2160 (or 4K or 2160p) video has millions more pixels than a 320x240 (240p) video, so it takes a lot longer to create.  In general, the videos exit the encoding process from smallest to largest.

#### Duration

If a video is 30 frames per second, and is 60 seconds long, there are 1800 frames to be created.  If a 30 frame per second video is an hour long 3600s), there are 108,000 frames to be created, so the longer the video - the more time for creation.

## Analysis

To better understand the encoding time, I divided the data by the size of the output video size, and looked at 2 lengths of video  - those under 5 minutes, and videos that are 45-75 minutes long (we are often asked how long a 1 hour video takes to encode).

I calculated encoding time for each bucket of video using percentiles.  

> Quick review - percentiles break the data into 100 evenly sized segments, so the 25th percentile is when 25% of the videos have been encoded. The median is the 50th percentile, and is when 50% of videos have been encoded.  We report here to the 95th percentile for video encoding.

### Short videos - under 5 minutes long

Many of the videos uploaded to api.video are short - under 5 minutes long.  These videos are generally ready to play in a few seconds:



The above chart has a lot of data, the 25, 50, 75, 90 and 95th percentile encoding times for all 6 variants of the video that api.video creates.

The median video is ready to play (at 240p) in 5.6 seconds, but a 720p version will be ready in 15s after encoding begins, and the 4K version will be ready in 37 seconds.

95% of all videos (under 5 minutes) will be playable in 32.6s, and will have 1080p versions ready in just over 2 minutes (140s).

### Long videos - 45 - 75 minutes long

Our most frequently asked question on encoding is "how long for a 1 hour video to be encoded and ready?"  Here are the numbers:


Observant viewers will see there are no 4K videos in this report. At this time, there are very few 4K streams that reach the 1 hour timeframe, and the data was quite noisy.  Hopefully, more of our 'long uploads' will add 4K support, so we can add this metric in our next update.

The median hour long video (red line) will be ready to play (at 240p) in just over 9 minutes (555s), and a 720p version will be ready about 18 minutes after encoding begins (1099s).

95% of hour long videos will have a 240p version ready in 33 minutes, and 720p in 54 (3230s).  the 1080p version will be available for 95% of videos in 108 minutes,

## Conclusion

Short videos will be ready to play - even at high resolution - in under a minute: the median 4K encoding for a short video is 37s. 

Hour long videos take longer to encode, and the median 720p video will be ready in 18 minutes.  95% of 720p videos will be ready to play in 54 minutes.

This is really fast! You can have responsive streaming video ready for your customers in just a few minutes with a simple upload!

That said, our team is working hard in the background to improve our infrastructure to make our encoding speeds even faster! 

We're really excited to have hard numbers to share with you on our encoding speeds, and we cannot wait to re-benchmark this data in 2021 to see how much faster your videos will be encoded as we improve our service.

Do you have questions about api.video or video encoding? Reach out on our [community page](https://community.api.video).


How fast does api.video encode video? 2H 2020 data

gif.jpg

ASICs-based encoding: Building a greener tomorrow through reduced carbon footprint

api-video-blog-product-updates.png

api.video recently announced that it now offers [free encoding for all its users](https://api.video/blog/product-updates/video-encoding-pricing-update/)—a bold move perfectly aligned with our mission to democratize videos. 

&nbsp;

As a next step and to make things more transparent, we decided to take the help of some numbers and explain it to you.

&nbsp;

The natural thought behind video encoding is that it is expensive. That is mainly because encoding workloads demand substantial CPU resources, and CPUs can be quite costly. Consequently, encoding has traditionally incurred high costs and the higher the video resolution, the greater the CPU requirements, directly impacting encoding expenses. It seems logical that pricing should correlate with these factors, which is what experts typically anticipate.

&nbsp;

But that is not always the case. Read on to learn how we, at api.video, are making it possible to scrap video encoding costs for you all. 

## We build our own infrastructure
There are two common routes you can take for video encoding - Central Processing Units (CPUs) or Graphic Processing Units (GPUs). But let us accept that they are both not very affordable. Moreover, datacenter GPUs aren't primarily designed for video encoding but are rather now targeting AI tasks. And considering the popularity of AI, cloud providers are unlikely to reduce GPU prices because they can command higher rates from AI customers. That’s when a new alternative comes in: Application-Specific Integrated Circuits (ASICs). These are specialized chips (Video Processing Units) designed for specific workloads and have been recently in news due to Google developing its own VPU for YouTube called Argos.

&nbsp;

At api.video, we build our own infrastructure that has the most advanced VPUs. With these VPUs, we are able to save power consumption costs since the power consumption cost of VPUs is one order of magnitude lower than CPUs and GPUs. It is also easily possible to pack a lot of them in a single server. This increases their density and therefore, reduces the number of servers. 

&nbsp;

In a nutshell, VPUs help you to encode videos at scale without costing you too much, and that is exactly why we at api.video, made the move to use VPUs for video encoding.  

## Understanding encoding costs on VPUs

api.video uses VPUs built by [Netint Technologies](https://netint.com/). Let's understand the costs with a 1080p30 video. To simplify the case, I will assume the source video codec is h.264. The output videos after transcoding will be five h.264 renditions for ABR: 1080p30, 720p30, 480p30, 360p30, 240p30.

&nbsp;

For a real time transcoding, let's try to sum up the number of MP/s (Megapixel per second) it represents. For the above output, it represents 62 MP/s + 28 MP/s + 12MP/s + 7MP/s + 3MP/s respectively for 1080p30, 720p30, 480p30, 360p30, 240p30 renditions. We need an encoding capacity of 112 MP/s for real time transcoding. Let us assume that the transcoding cost on VPUs is proportional to the number of pixel per frame (whatever the frame resolution) for the sake of simplicity and one VPU of Netint is able to encode 2000 MP/s in h.264. In reality it is a bit more complex but let’s take these as the numbers for a rough idea. 

&nbsp;

Now, as per the above assumptions, we can encode 2000/112 = 17 minutes of videos (all rendition included) per minute. Below is a table summarizing the minute cost for a server with 4 VPUs. The cost of the server is $10k and the amortization for such a server is 3 years. The electricity cost will be approx. $1000 per year (420w at $0.30 / kWh) at full consumption. Let us see what is the cost of 1 minute of a 1080p30 video transcoded to 5 renditions.

&nbsp;

(The cost is split taking into consideration the average usage of the server per day)

&nbsp;

| Avg. usage of the server in percentage | Minutes encoded over the amortization period | Cost per minute transcoded (minutes / cost of server + electricity) in $
| ------------- | ------------- | ------------- |
| 0,50%       | 536112       | 0,020
| 5%   | 5361120        | 0,002
| 25%   | 26805600        | 0,0004
| 50%   | 53611200        | 0,0002
| 90%   | 96500160        | 0,0001

&nbsp;

If we use the server at only 0.5% per day (i.e ~7 min per day fully loaded) the cost per minute would be at $0,020.

&nbsp;

If on an average, this server is loaded at 25% (6 hours a day) the cost per minute would be 1/150 of our previous price.

### What about video quality?
Hardware encoded is always associated with quality losses in the head of many video engineers. But that was probably the case decades ago.

&nbsp;

Not anymore.

&nbsp;

Let’s see how VPUs compete with CPUs in terms of quality by taking a video under [Creative Commons license](https://en.wikipedia.org/wiki/Creative_Commons): Caminandes 2.

&nbsp;

We have uploaded its 1080p30 version via our API and will compare the quality output with a CPU encoded video with this option **-b:v 4400k -preset medium** These parameters generate a rather good quality.

&nbsp;

**ffmpeg -r 30 -y -i input.mp4 -c:v libx264 -force_key_frames 'expr:gte(t,n_forced*4)' -b:v 4400k -preset medium -c:a copy output.mp4**


&nbsp;

Here are the quality metrics (higher is better):

|                  | VMAF (avg) | PSNR (avg) | SSIM (avg) |
| ---------------- | ---------- | ---------- | ---------- |
| api.video (VPUs) | 95.92      | 48,07      | 0.9888     |
| ffmpeg (CPUs)    | 95.33    | 48.69      | 0.990     |


The quality metrics are very closed and report a high score. The VMAF is slightly better through VPU encoding.

&nbsp;

In conclusion, this pricing update reflects our commitment to enhance the video streaming experience for everyone - be it individuals or organizations looking to build their video solutions. We firmly believe that by making video streaming more accessible, we empower creators, businesses, and enthusiasts alike to embark on their video streaming journeys with ease. 

&nbsp;

_To know any other details, [talk to our team ](https://meetings.hubspot.com/apivideo/talk-to-us) and we are here to help you out._
 


Free video encoding: Let's talk numbers

large_Frame 427319242 (1).jpg

medium_Frame 427319242 (1).jpg

small_Frame 427319242 (1).jpg

thumbnail_Frame 427319242 (1).jpg

Frame 427319242 (1).jpg

Huge pricing update: Introducing video encoding at no cost

Blog_Zero Cost.webp

What is the difference between Encoding and Transcoding

Huge pricing update! Introducing video encoding at zero cost

## What is Encoding? The first step.

Encoding is the initial process of converting raw video files into a specific digital format or codec. This process involves compressing and packaging video data, typically to reduce file size while maintaining acceptable quality. Encoding prepares video for storage, transmission, or playback.  

## What is Transcoding? Adapting video for different needs.

Transcoding, on the other hand, involves converting a video from one digital format or codec to another. This process typically occurs after the initial encoding and is used to:

1. Optimize videos for different devices or platforms
2. Adapt to varying network conditions
3. Ensure compatibility with specific playback requirements

## What goes on behind encoding a video?

It's easy to say that encoding is when you go from an _uncompressed source_ to a _compressed source_, but you may ask, what exactly do compressed and uncompressed sources mean?

&nbsp;  

Raw video sources can be uncompressed files directly from a camera sensor, such as those using SDI (Serial Digital Interface) or HDMI (High Definition Multimedia Interface) standards. Analog sources like VHS tapes are also considered uncompressed. The encoding process transforms these into compressed digital formats, making them more manageable for various uses.

&nbsp;

A compressed sources, on the other hand, are video files that have undergone compression during the encoding process. Compression involves reducing the file size by removing unnecessary information or encoding the data in a more efficient format. This process results in a smaller file size, making it more manageable for storage, transmission, and playback.


## Transcoding methods fall into three categories

There are three types of transcoding, depending on what you are trying to do to get a video to suit a particular browser, player, or platform:

1. **Lossless-to-lossless**: Maintains video quality while taking advantage of better hardware or compression algorithms.
2. **Lossless-to-lossy**: Decreases video quality but results in smaller file sizes or improved compatibility.
3. **Lossy-to-lossless**: Prevents further quality loss during conversion but doesn't regain previously lost data.


## The role of codecs for encoding and transcoding

Both encoding and transcoding rely on codecs – algorithms that compress and decompress video files. The term "codec" itself is a combination of "compress" "decompress" or "encode" and "decode". Codecs are essential for shrinking video files or adapting them to meet specific playback requirements.



## What is transmuxing?

Now that you know the difference between encoding and transcoding, let's throw one more term in there - transmuxing. 

Transmuxing is a process that changes the container of a video file without altering the file itself. This method is faster and less computationally intensive than encoding or transcoding. It's useful for quickly making content available across different platforms, but it has limitations:

1. It doesn't allow for creating multiple video resolutions.
2. Choosing an incompatible container may result in playback issues.

&nbsp;

This process is faster than encoding or transcoding and takes less computing power because you only change the container, not the file. 

## Key Takeaways

1. **Encoding**: Compressing raw data from a video camera sensor or analog source.
2. **Transcoding**: Converting an already compressed video file to another format.
3. **Transmuxing**: Changing a video file's container while leaving the content untouched.  

&nbsp;  

Understanding these distinctions is crucial for anyone working with video files, whether for personal projects or professional applications. Each process serves a specific purpose in the video production and distribution workflow, allowing for efficient storage, transmission, and playback across various devices and platforms.  

&nbsp;  

We hope that explains the difference between encoding and transcoding clearly.  

&nbsp;  

_If you are looking for an affordable video API solution that does not burn a hole in your pocket for transcoding and encoding videos, check out our offerings on our [features page](https://api.video/features/) or [sign-up for a free account here](https://dashboard.api.video/register)!_

Encoding-Transcoding.png

Video transcoding: What, why, and how?

small_Blog_Zero Cost.webp

thumbnail_Blog_Zero Cost.webp

thumbnail_Arushi_.webp

Arushi_.webp

Arushi Gupta

large_api-video-blog-video-trends.png

medium_api-video-blog-video-trends.png

small_api-video-blog-video-trends.png

thumbnail_api-video-blog-video-trends.png

api-video-blog-video-trends.png

With video playback technology witnessing quite an evolution because of the increasing popularity of videos, we have the codec comparison of H.264 vs. H.265 coming up quite frequently. You might be wondering, in the battle of H264 vs. H265, what reigns supreme?

## H.264 vs. H.265: What do they mean?

Let’s start from the basics and understand what the two terms – H.264 vs. H.265 – mean.

### What is H.264?

[H.264](https://api.video/what-is/h-264/), also referred to as Advanced Video Coding (AVC), operates by using advanced algorithms, inter-prediction and spatial intra-prediction. These units, termed macroblocks, consist of 16x16 pixel samples which can be further subdivided into sub macroblocks. The H.264 codec is known to achieve significantly [reduced bitrate](https://api.video/blog/video-trends/video-bitrate-importance/) compared to its forerunner and remains in use across various streaming sources.

### What is H.265?

[H.265](https://api.video/what-is/h-265/) represents the advanced iteration of H.264, alternatively also known as High Efficiency Video Coding ([HEVC](https://en.wikipedia.org/wiki/High_Efficiency_Video_Coding)). Diverging from H.264's macroblock approach, H.265 utilizes Coding Tree Units (CTUs) for information processing. While blocks in H.264 were confined to 16x16 sizes to 4x4 sizes, H.265 can use a complete range of blocks from 64x64 pixels all the way to 4x4, resulting in notably more efficient compression. HEVC also demands higher computational power for effective data compression.

## Difference between H.264 vs. H.265: Let’s get into the details

Both H.264 and H.265 can provide high quality and small file size but there is a slight difference between them.


large_Blog_H264_contentImage.png

medium_Blog_H264_contentImage.png

small_Blog_H264_contentImage.png

thumbnail_Blog_H264_contentImage.png

macroblocks-h264-vs-CTUs-h265.png

### 1. Video Quality
H.264 and H.265 codecs exhibit significant disparities in video quality at equivalent bitrates. In the case of H.264, the borders of the blocks are likely to be distorted because each macroblock is fixed, and each macroblock’s data is independent of another. In contrast, H.265 delivers sharper, more detailed images with fewer artifacts. This enhancement arises from its ability to determine CTU (Coding Tree Unit) sizes based on contextual information, making H.265 superior to H.264 in terms of compression and image quality.

### 2. Compression and File Size
The degree to which a codec compresses digital videos directly influences the resulting file size for transmission or streaming. Smaller file sizes equate to lower bandwidth at a certain quality. Since H.265 uses a smaller bitrate to transmit the same quality as H.264, we can conclude that H.265 can export smaller files. Consequently, H.265 takes the lead in scenarios where limited storage capacity or constrained bandwidth is necessary for smaller file sizes.

A corollary of this is that, H.265 also achieves a higher coding efficiency as compared to H.264 at the same quality.

### 3. Compatibility
In the realm of compatibility, H.265 surpasses H.264 but it lags behind it in popularity. While H.264 enjoys support from a larger number of devices and browsers, support for H.265 is limited to a lesser number of devices and browsers. This is mainly due to H.265’s patent issues that are preventing its easy adoption.

### 4. Patent and licensing
The AVC or the H.264 codec has just one patent licensing. On the other hand, HEVC has four: HEVC Advance, MPEG LA, Velos Media, and Technicolor. This fact complicates the adoption of H.265 and makes it more expensive for companies to afford.

## Tabular comparison between H.264 and H.265

It’s no fun for technology-geeks until we talk numbers in the ‘Is H.265 better than H.264’ debate.

## AV1: The game-changer in video technology

AV1 is a royalty-free, open-source video compression standard developed by the [Alliance for Open Media](https://aomedia.org/) (AOMedia), a consortium of technology companies including Google, Mozilla, Netflix, and Amazon. AV1 was designed to provide even better compression efficiency than H.265, and it also addresses the patent and licensing issues associated with it.

&nbsp;

Recently, Apple also announced hardware-accelerated decoder for AV1 as part of an upgraded video engine in the A17 Pro chip for iPhone 15 Pro and iPhone 15 Pro Max.

## H.264 vs. H.265: Which codec is better?

In a nutshell, the H.264 and H.265 codecs are different in terms of video compression efficiency. Each of them have their advantages and disadvantages, and it depends on your video needs as to what you decide to choose.

&nbsp;

If you intend to stream a large amount of videos (like Netflix or Amazon), H.265 may be better for you. If your video needs are limited, H.264 may be your best bet. So, it's worth noting that despite its advancements, HEVC has its share of drawbacks, and so does H.264. Make sure you consider both for a comprehensive perspective.

&nbsp;

So, which codec are you going to choose?


large_h264-vs-h265.png

medium_h264-vs-h265.png

small_h264-vs-h265.png

thumbnail_h264-vs-h265.png

h264-vs-h265.png

## Frequently Asked Questions (FAQs)
### **Does using H.265 (HEVC) require specialized hardware or software?**


While some older hardware and software may not support H.265 natively, most modern devices and software platforms have built-in support for this codec. However, encoding H.265 videos may require specialized hardware or software for optimal performance.

### **What are the implications of using H.265 (HEVC) for live streaming?**

H.265 can significantly improve the quality of live streaming by reducing bandwidth requirements and delivering higher-quality video at lower bitrates. Make sure that your streaming platform and devices support H.265 to avoid compatibility issues, before you choose to opt for it. 


### **Are there any licensing fees associated with using H.265 (HEVC)?**

Yes, H.265 is subject to patent licensing by various organizations, which may entail licensing fees for certain use cases. But many software and hardware manufacturers have negotiated licensing agreements, so end-users may not directly encounter these fees. Do review the licensing terms for your specific use case to ensure compliance.

### **Can I convert H.265 (HEVC) videos to H.264 (AVC) format?**

Yes, it is possible to convert H.265 videos to H.264 format using video conversion software or online converters. Keep in mind that the conversion process may result in some loss of quality, as H.264 is not as efficient at compression as H.265.

&nbsp;

_api.video enables you to quickly and securely deliver on-demand and live-stream videos directly from your website, software, or app. [Sign-up ](https://dashboard.api.video/register)for a free sandbox account to try out api.video’s video API, or [talk to our team](https://meetings.hubspot.com/apivideo/talk-to-us) to get a demo._

What is the difference between H.264 vs. H.265?

large_D.jpg

medium_D.jpg

small_D.jpg

thumbnail_D.jpg

h264-vs-h265.jpg

poster-1_2.png

large_poster-3.png

medium_poster-3.png

small_poster-3.png

thumbnail_poster-3.png

[Video encoding](https://api.video/what-is/video-encoding/) is the process of exporting digital video into a format and specification suitable for playback by a user. 

Each video file contains two elements: 
* Codecs, or “compression-decompression“ algorithms which compress the video
* Formats, the type of file the data is compressed to. 

Imagine a shipping container filled with packages of many types. In this analogy the shipping container is the format, and the codec is the tool that creates the packages and places them in the container.

## Video Codecs


videocontainer-format.png

Video codecs used today include H.264, HEVC, VP9, and AV1. Common formats includes MKV, AVCHD, MP4, and WebM. 
There are pros and cons to each video codec and file format. There is no silver bullet when it comes to streaming content on the web. Certain video codecs and video formats are open source, meaning the source code is available to the user, who is given the rights to study, change, and distribute software to anyone for any purpose. VP9 and AV1 are both open source codecs types. MKV and WebM are also both open source, developed by nonprofit organizations and allowed to be used without a license.

## Video Formats

Video formats are often equated to a file’s extension, for example the .mp4 in Video. mp4. Video formats are much more than that and include a whole package of files, including a files video stream, the audio stream, and any metadata included with the file. All of the data put together, both of the streams and the metadata, is read by a video player to stream video content for playback. A file’s video stream includes the data necessary for motion video playback, while the audio stream includes any data related to sound. Metadata is any data outside of audio and sound, including bitrate, resolution, subtitles, device type, and date of creation.

## Bitrate
Bitrate or bit rate, is the rate at which bits are processed or computed over time. The higher the bitrate, the more data there is, which generally means a higher quality video file. When choosing a bitrate it is important to consider the device your videos will be played on. High bitrate file streams need lots of processing power to encode, and a fast internet connection to download for playback. If you’re having trouble choosing the right bitrate, file format, video codec, and video resolution, maybe it’s time to consider api.video.

With the api.video you can easily and quickly transcode video for your audience no matter the quality, format or bitrate. We wrote this article to help you sort through the hundreds of codecs and formats, and help you take the guesswork out of streaming video content on your site, application, or platform.

## The Most Popular Codec: H.264


h-264.jpg

First let’s talk about H.264, also known as AVC. 

H.264 is the most widely used video codec. Sure, the H.264 codec may be a bit outdated by now, but this type of compression technology is present everywhere, with decoders available for pretty much every platform and mobile device. H.264 likely won’t be the standard for long. 
HEVC, short for High Efficiency Video Coding, or H.265 is slowly becoming ubiquitous thanks to the proliferation of 4K content. If AVC or H.264 was the codec of 1080p content, then HEVC and H.265 is the codec for the 4K era. H.265 compresseses videos far more efficiently (hence the “high efficiency”) when compared to H.264. This means H.265 4K content that is easier to stream and at a higher quality than 4K content compressed using H.264.

## New Codecs: H.265 and VP9


h265.jpeg

H.265 is a licensed codec, meaning you have to pay to use it. Its main competitor, VP9 is a royalty free and open source codec developed by Google. This means that anyone can use the codec to compress their videos license free. Generally, VP9 produces more consistent or more reliable streams, whereas H.265 (HEVC) produces videos of higher quality. 

The differences between the two codecs are due to the strengths and weaknesses of their compression algorithms. VP9 was developed to produce videos with a smaller file size and therefore ones that are easier to stream. H.265 compresses videos using a large variety of processing techniques which ultimately create a compressed video that retains a very high level of detail.

In addition to VP9, there is another open source codec still under development called AV1. AOMedia Video 1 (AV1) is an open source, royalty-free video coding format specifically designed for streaming video content over the internet. AV1 was developed to succeed the VP9 codec and compete directly with HEVC codec on both video quality and file size. AV1 is still a highly experimental codec, but is supported by Firefox, Google Chrome, VLC media player, FFmpeg among others.

## The Most Popular Video Formats
We mentioned this before, but formats are the entire set of storage rules that package a video file’s container, codecs, and metadata. File formats are used so that a variety of devices, platforms, and players can process and playback any number of codecs. Formats are developed to improve image quality, decrease file sizes, allow for universal playback on a variety of devices, and provide special features such as subtitles. Since there are new formats getting developed every day it’s difficult to cover each and everyone. Instead, we’re going to focus on video formats used specifically for online distribution or content streaming.

The most popular video format is MP4, also known as MPEG 4. MP4 is a multimedia container for video, audio, and data, including things like subtitles and still images. Make sure you don’t confuse MPEG 4 and MPEG 4 Audio. MPEG 4 has the file extension .MP4, while MPEG 4 Audio has the file extension .M4A and can only contain audio files. So, why exactly is MP4 so popular? Well it mainly has to do with how ubiquitous it is across a range of devices. Both mobile and desktop devices can playback MP4 video files, so it makes sense that it would be used by most content producers.

## Playback Techniques for Video and Livestreaming
MP4 files are played at a constant bit rate (CBR), meaning the quality of your video stream will remain at the same bitrate no matter the variance in your internet bandwidth. There are other file formats that support adaptive bitrate (ABR), which will adapt the quality of your stream to match the internet speed of your player. One of the most popular formats that supports ABR is HLS..

HLS or HTTP [Live Streaming](https://api.video/product/live-stream/) use adaptive bitrate streaming techniques to enable high quality streaming with small HTTP-based file segments. The key to this format’s popularity is the use of HTTP. HTTP is the universal language protocol for the internet, and allows for reduced infrastructure costs, greater reach, and simple HTML5 player implementation. Another great feature for the HLS format is the M3U8 file descriptor. This file allows the client or device to automatically select the best quality stream at any time, to prevent playback buffering regardless of bandwidth or CPU power.

Now that you know how file formats and codecs work it’s time to get streaming. Visit [this page](https://dashboard.api.video/register) to create a free account, so that you can quickly and easily integrate streaming video content into your application, site, or workflow. Our API can accept pretty much any type of broadcast (even 4K native content), and can quickly transcode and adapt the stream for any type of network, including EDGE. No matter the screen, size, quality, or bitrate, api.video can manage it for you. Our web service is built to answer the most demanding video and streaming needs. If you want to learn more check out our website at [api.video](https://api.video/) or join our [community forum](https://community.api.video/). Good luck and happy streaming!

## You May Also Like...
If you liked this article, you may also be interested in:

* [Create a Live Stream](https://api.video/blog/tutorials/live-stream-tutorial/) - Create your first live stream!
* [Videostreaming with a Raspberry Pi](https://api.video/blog/tutorials/video-streaming-with-a-raspberry-pi/) - Now that you've got a live stream working, have some fun adding one using a Raspberry Pi. 
* [Video Upload](https://api.video/blog/tutorials/video-upload-tutorial/) - Upload your first video. 

Every video format, codec, and container explained

large_FormatsCodecsContainers.png

medium_FormatsCodecsContainers.png

small_FormatsCodecsContainers.png

thumbnail_FormatsCodecsContainers.png

FormatsCodecsContainers.png

Software video codecs are very diverse. Each one is carefully optimized depending on what the codec designers hope to achieve. For example codecs used when capturing video will probably try to retain as much data as possible. However a codec that's going to be used to compress video for conferencing might decrease video quality and increase transmission speed to lower latency. While they may have diverse use cases, one thing software codecs have in common is the design principles applied to their creation. There are five principles applied to every software codec. 

## Principle 1: Maximize allocation of processing resources based on your codec use case 
As I just mentioned, different codecs may hope to achieve different levels of efficiency across their elements based on what they'll be used for. For software codecs, processing resources and available transmission bit rate are what limit the design. Each design is about considering trade-offs, if you put more resources into one area of functionality in a codec, it will become more efficient, but then less resources are available for other areas of functionality. There are four main areas a codec will allocate resources to:

* **Encoded frame rate** - You'd refer to the use case for your codec, say it's video conferencing, and then try to maximize allocation of resources to efficiently achieve the standard 12-15 frames per second that's expected for desktop video conferencing. If you have a unique, efficient way of increasing quality, maybe you would be able to increase the frames per second. At a minimum though, you'd want to achieve the standard.
* **Frame size** - Frame size is what it sounds like, how many pixels wide and tall each frame of your video is. Depending on what level of quality or detail you want to offer with your codec, you'd want to maximize frame size as much as possible. Your use case for the codec would dictate how you choose to maximize frame size. 
* **Peak coded bit rate** - The coded bit rate is how many bits are used per second to represent video. While higher coded bit rates mean the video is less compressed, they can be useful if you want to be able to store or transfer video at a high quality. High bit rate is hard on processors, so it's a decision that must be made carefully. Again, depending on your use case you might want to maximize the peak coded bit rate, or focus your resources on other areas that are more important for your codec. 
* **Video quality** - Video quality is tied to bit rate, so depending on use case, strategies will vary. Quality and computational complexity have a complicated relationship, but within any codec there are many places where you can trade off quality for decreased computational complexity. 

## Principle 2: Minimize latency whenever possible
No matter the codec, the less the delay, the better. Delays are less important for some codecs, but to be avoided at all costs for other codecs - like those used for video conferencing. 

## Principle 3: Minimize compiled code and storage requirements
Not all systems using software video codecs will have a lot of available memory, so codec designs take this into account. Whenever possible, they cut down on the amount of code that needs to be compiled and the amount of storage needed for the video codec to function.

## Principle 4: Make your platform code robust and portable
The more places a codec can easily be used, the greater the chance it will be adopted everywhere. Easy-to-use means the code's been well tested so the codec isn't buggy, and the codec is easy to use on many different platforms. 

## Principle 5: Provide a flexible API
Make your codec developer friendly by providing an easy-to-use, well documented API. 

## Resources 

1. Chapter 13 of Video Codec Design - http://lad.dsc.ufcg.edu.br/mpeg/VCD/Ch13.pdf 
2. Codec - https://en.wikipedia.org/wiki/Codec 
3. A Comprehensive Guide to Video Codec Design - https://www.semiconductorstore.com/blog/2014/A-Comprehensive-Guide-to-Video-Codec-Design/866/
4. Thumbnail by Photo by [Diego PH](https://unsplash.com/@jdiegoph?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText) on Unsplash
  

The five design principles applied to software video codec

large_VideoCodecs (1).png

medium_VideoCodecs (1).png

small_VideoCodecs (1).png

thumbnail_VideoCodecs (1).png

VideoCodecs (1).png

The five design principles applied to every software video codec

video create

small_cedric.webp

thumbnail_cedric.webp

cedric.webp

Cédric Montet

large_api-video-blog-product-updates.png

medium_api-video-blog-product-updates.png

small_api-video-blog-product-updates.png

thumbnail_api-video-blog-product-updates.png

Encoding-giphy.gif

shrink-video.gif

large_EncodingTranscoding.png

medium_EncodingTranscoding.png

small_EncodingTranscoding.png

thumbnail_EncodingTranscoding.png

Have you ever uploaded a video on YouTube and wondered how your video is magically available in 5+ different qualities and 5+ different sizes? It’s not magic ✨ it’s *video transcoding*. 

Well, it’s kind of magic 🌟

In this article, we’ll tell you everything you want to know about video transcoding — from what it is, to its importance in video streaming and [video APIs](https://api.video/). With that said, let’s get started! 🚀

## The difference between video encoding and video transcoding 

Encoding and transcoding sound very similar but, in fact, are two very [different processes](https://api.video/blog/video-trends/what-is-the-difference-between-encoding-and-transcoding/). To learn about video transcoding, let’s first describe video encoding.

### What is video encoding? 

[Video encoding](https://api.video/blog/video-trends/guide-to-video-encoding-video-codec/) digitally compresses raw video from a digital or analog camera. You almost never have to worry about video encoding as your device encodes all videos automatically. The video you recorded of your dog this morning and stored on your phone - it is there on your phone because of video encoding. Similarly, the video that you shared with your friend using WhatsApp wouldn't have been be possible to share without video encoding.

In shirt, encoding is necessary for creating manageable file sizes.

### What is video transcoding?

Video transcoding re-compresses an encoded file which means that video transcoding cannot happen without an initial video encoding.

Transcoding, compresses a file from one digital [codec](https://api.video/blog/product-updates/every-video-format-codec-and-container-explained/) to another. Being able to play videos across a wider range of devices, or to improve playback experience are two of the many reasons that we transcode videos today.

An example of video transcoding is to transcode an encoded AV1 video to an H264 video codec. Why do you need to do that, you may ask? AV1 codec, despite being an efficient codec, is not supported on a wide range of devices. H264 codec on the other hand, is supported on almost all of the devices out there.

## Types of video transcoding

When transcoding a video file, some data is lost, which is what we call lossy—the greater the compression, the smaller the file size and the lower the quality. Let’s now look at three main types of video transcoding.

### 1. Lossy-to-lossy

It occurs when one type of lossy data is converted into a different kind of lossy data. Having a file of lower quality before transcoding means the quality will drop even further during the process. Reducing the bitrate for use on portable devices is one of the justification for this transcoding method.

### 2. Lossless-to-lossy

This transcoding reduces quality loss relative to lossy-to-lossy methods while creating files with manageable sizes for mobile devices. To make good use of this transcoding technique, you will need to save archives of files that have been compressed losslessly.

### 3. Lossless-to-lossless

This type of transcoding involves using improved compression to shrink a file without altering its quality. This transcoding method saves quality when switching between formats, but the generated files are sometimes too large to transfer to mobile devices.

## Why is video transcoding used?

The use of video transcoding is based on requirements. Here we can look at some of the top reasons for using video transcoding:

- **Adaptive bitrate streaming (ABR streaming)** - ABR streaming ensures the best video experience on a wide range of devices depending on your internet connection’s speed and stability. Transcoding is a crucial intermediate step to get the highest quality streaming output based on HLS or DASH. For example, video playback on Youtube could be switching between 1080p and 720p versions of the video stream, depending on your device and connection. This leads to the fewest buffering instances and higher-quality playbacks.

- **Limited storage capacity** - The word cloud is one of the biggest tricks played by marketers. It allows us to imagine a world with an infinite storage capacity for data. This is far from reality. Cloud is some hardware located in a room somewhere. No matter where you store your videos, cloud or not, you are using certain storage space. Transcoding is the key to maximizing your storage capacity, resulting in lower carbon footprints. Yes, actual carbon footprints.

- **Quality of user experience** - Transcoding is about more than just reducing a file size. Transcoding is also used to improve the playback experience.

- **Multiple target devices, browsers, and applications** - Transcoding is used for video compatibility amongst different devices, browsers, and applications.

## Who uses video transcoding?

Many individuals and businesses use video transcoding.

- Video streaming APIs for businesses and developers such as [api.video](https://api.video/). Video transcoding is an integral element of an adaptive video streaming workflow to get video content ready for a delivery protocol — like the current industry standard, HLS or DASH — that can play on as many devices as possible.
- Video-On-Demand (VOD) services such as Netflix, Hulu, and Youtube use video transcoding on large scales.
- Content Management Systems (CMS) targeting multiple devices with a variety of video formats and sizes.
- Video Editors. As a video editor, you may prefer to work on the highest quality video codec, and once you're done editing, you could transcode your video to other codecs for delivering to a wider range of devices.

## Video APIs and video transcoding

As one of the more important users of video transcoding, video APIs are beneficial for businesses, both big and small, to create and host on-demand videos. 

[High-quality video APIs](https://api.video/) such as api.video can change your video into HTTP Live Streaming (HLS) for smooth streaming at high speeds anywhere in the world without you needing to worry about doing the process yourself. For the highest possible video quality on any connection, it handles the nitty-gritty of transcoding in different qualities for ABR streaming over HLS while keeping the best playback quality on a wide range of devices.

## Final thoughts

New gadgets are being introduced to us every day. This means there will be more cameras with greater capturing capacities and there will be more screens with higher display qualities.  At the same time, the video world is booming, and more and more businesses need to reach out to users with all sorts of devices and bandwidth. 

Transcoding is not magic, but it certainly has a magical aspect to it. It makes it possible for us to connect through high-quality videos.

Hopefully, now you know more about transcoding and why it’s used in our industry. 

Would you like to start your video journey today? 🎬 Go ahead and start testing in the [api.video](https://dashboard.api.video/register) dashboard for free.

Transcoding.png

Videos are generally ready to play in a few seconds

screenshot-2020-12-04-at-09-41-12.jpg

how long for a 1 hour video to be encoded and ready?

screenshot-2020-12-04-at-09-42-18.jpg

medium_gif.jpg

small_gif.jpg

thumbnail_gif.jpg

transcoding

video api

large_logo_avatar.webp

medium_logo_avatar.webp

small_logo_avatar.webp

thumbnail_logo_avatar.webp

logo_avatar.webp

api.video

large_transcoding-1-1.png

medium_transcoding-1-1.png

small_transcoding-1-1.png

thumbnail_transcoding-1-1.png

transcoding-1-1.png

large_Transcoding.png

medium_Transcoding.png

small_Transcoding.png

thumbnail_Transcoding.png

Announcements

AI, models, assistants, agents and prompts sound and look magical on social networks. But from the countless conversations we have had with startups and Enterprise customers over the last months, we have realized that the gap between existing video workflow and concrete use cases is real.

&nbsp;

Whether it’s getting transcripts in seconds, changing the language of your video or being able to search for items in your content – the benefits, value or potential savings you see and can get with AI are obvious.

&nbsp;

This is why we are thrilled to unveil how we are bringing AI into video, where we are heading and how you can get more value from it!

## What does the future of video streaming look like?

AI will completely upend video consumption as we know it today. We envision a world where video is not a content piece anymore which you can only start and stop – because AI will make videos discoverable and universally understandable. More than ever it resonates fully with our core mission – connecting people through their cameras.

&nbsp;

Since Day 1, we are building api.video to make your use cases, first, a reality and, then, a success.

&nbsp;

Back in the days, the fragmented video streaming market, the number of services you had to aggregate, the lack of innovation by the incumbents was preventing you from joining our video-first world, while your efforts and budgets were consumed by maintaining mashups instead of focusing on what matters. 

&nbsp;

This is why we have built the video backbone and seeing that, now, you host 2.5 millions videos on api.video and deliver dozens of millions videos every month demonstrate why simplicity, exceptional performances and an E2E approach were key.


&nbsp;

What’s happening today with AI, the efforts it requires to benchmark, setup and customize LLMs, to run and maintain them, the Infrastructure you have to build to execute them before getting any value, remind us of the state of the streaming market when we had launched api.video.

&nbsp;

Helping you to benefit from the AI revolution by abstracting all the underlying complexity became obvious and a natural next step in our evolution.

&nbsp;

To get there and to bring you there, here is what we are building, api.video’s very first – **Video Model Hub**.

&nbsp;

Let’s start with the root principles encompassing our work on this Hub:

&nbsp;

- **Open-source models over private ones.** We are long time believers and contributors of the open source community and we have chosen to run only models that are available publicly.

- **Your data is… your data.** You already trust us to host and deliver sensitive content,and this is exactly why we won’t train our own models or sell the data generated from your videos to anyone.

- **Privacy at the core.** We host and execute the models inside our own infrastructure. Your data remains inside the api.video infrastructure. Your videos or data is not sent to external partner, SaaS providers like OpenAI, or others.

- **We believe that every organization will eventually build their own models to leverage their unique data sets**. Next year, alongside the ability to run public and open source models on api.video, we will offer you the ability to host your own.

- **One API to rule them all.** As we have already replaced a dozen of services and vendors under GET/videos to make your videos available worldwide in seconds, we are abstracting all the complex work of executing LLMs behind clear and straightforward endpoints.

&nbsp;

At a glance, api.video brings to you today the power of AI by:

- Leveraging its years of experience in video management and building infrastructure, with GPUs.
- Combining with the most advanced LLMs available on the market.

so that you get more value from your videos in a few lines of code!

&nbsp;

## Here are the first three new features from the Video Model Hub

### 1. Summarization of long-form videos

Upload a video in English, Spanish, French, Portuguese or Chinese and get a text summary of 120 characters with the 3 key takeaways discussed in the video (within seconds!). Ideal to index or preview long form recorded meetings or courses!

&nbsp;

_The video below gives you a glimpse into AI summarization._


### 2. Transcription of your video in seconds

Get a transcript (in .VTTformat) from your uploaded video with the timecodes and offer the captions to your audiences. The best part? The api.video player supports VTT files so there’s nothing new that you need to build on your side!
This also makes your content more inclusive by making it accessible to people with hearing impairments. 


### 3. Translation of subtitles in 5 different languages (and more to come!)

Is your video in English? You can get Spanish and French subtitles. In Chinese? You will get English subtitles and three more languages. In short, this is the perfect solution to spread your message to your international audience!

&nbsp;

_Check out the video below for a glimpse into AI translation._ 


_Try it for yourself now by uploading a video in your [sandbox account](https://dashboard.api.video/overview)._ 

## What’s next

From video classification, alteration, generation, audio alteration, metadata generation, we have identified and plan to implement more than 200 additional use cases and features within the upcoming months!

&nbsp;

From next year, we will let you bring your own models on the infrastructure to get even better and more personalized outcomes.

&nbsp;

This is a long-journey and we’re only getting started. The whole team is thrilled to be at the forefront of this next chapter!


AI  >< Video: The Video Model Hub unveiled

small_AI thought leadership.png

thumbnail_AI thought leadership.png

AI-thought-leadership.png

Our mission at api.video has always been to turn your use cases into successful realities. We built the video backbone with this goal in mind, and today, you host 2.5 million videos and deliver tens of millions more!

&nbsp;

Helping you harness the power of AI, while simplifying the problems that come along with, was the next direction to take. 

&nbsp;

To achieve this and to support your journey, we bring to you –  **AI features for your videos.**



## What are the AI video features?

### 1. AI video transcription & captions

With AI transcription, you can convert spoken words in your videos into written text, providing an accurate, searchable record of your content. This means you can easily create subtitles, enhancing accessibility for the hearing impaired or getting a written record of your meetings or events. 






### 2. AI video translation

With AI translation, you can translate your videos into multiple languages, ensuring that viewers from different linguistic backgrounds can understand and engage with your content. This can be useful if you’re hosting an international virtual event, creating multilingual educational materials, or expanding your marketing reach.

&nbsp;

Translation will be available for 5 languages – French, English, Spanish, Portuguese and Chinese.


### 3. AI video summarization

Using AI summarization, you can distill lengthy content into concise, digestible summaries, allowing viewers to quickly grasp the key points without watching the entire video with AI video summarization. This can prove invaluable for educational lectures, business presentations, and lengthy seminars where time is of the essence.

## Do you really need this?

Well, the potential [uses of our AI features](https://api.video/use-cases/generative-ai/) are limitless! From e-learning and e-commerce to online streaming, these [AI features](https://api.video/product/ai-features/) are going to be the difference between ordinary and excellent. What can you do really?

&nbsp;

- Summarize educational lectures for a quick review
- Translate virtual events into multiple languages for global reach
- Transcribe video interviews for easy reference
- And so much more!

## More than just encoding, hosting, and delivery

We’ve gone beyond the basics of video hosting and delivery to give you features that are a step up from the ordinary. Now, not only can you ensure that your customers get video delivery at minimal cost, but also that they get the flexibility to offer content  in their preferred language and in a crisp manner. 

&nbsp;

Additionally, when the features are in production, you may also get the option of adding them via your API. This means that you get complete power to add the features you like without having to go through the dashboard.

## Your videos stay private to you

Your data is yours, and we respect that. You already trust us to host and deliver your most sensitive content, which is why we will never train our own models or sell data generated from your videos. Privacy is at the core of everything we do, and that’s why we host and execute the models within our own infrastructure. Your data stays entirely within the api.video ecosystem and is never sent to external partners, SaaS providers like OpenAI or others.

&nbsp;

_Our CEO and founder, Cedric Montet, speaks more about privacy being our topmost concern in his latest [blog](https://api.video/blog/company/video-model-hub-ai/)._

## Give it a try!

See a glimpse how these AI features look like with a free sandbox account! [Upload a video now in your dashboard](https://dashboard.api.video/overview) and witness AI make a difference to your videos. As this is a Beta release, any feedback, questions, doubts on the features are welcome and much appreciated.

Create a sandbox account now 

Lights, Camera, AI Action! Discover our AI features

Introducing AI features to compliment your videos

small_Introducing AI features to compliment your videos.png

thumbnail_Introducing AI features to compliment your videos.png

AI-features-to-compliment-your-videos.png

AI >< Video: The Video Model Hub unveiled

How to integrate AI facial expression detection to control your video – Javascript 

5 AI tools that will save you 10+ hours on video production

Lights, Camera, AI Action! 

In a world where time is money and creativity reigns supreme, AI tools that convert text to video are game-changers! These AI-powered wonders are designed to save your day—and countless hours. Join us as we unveil the top 5 AI video tools that can dramatically cut down your video creation and editing time. From transforming text into captivating visuals to adding that perfect touch of magic to your edits, these tools are set to become your new best friends in the digital playground!

## **1. CapCut**

At number 1, we have Capcut. 

&nbsp;

[CapCut](https://www.capcut.com/) surged in popularity due to its user-friendly interface and powerful editing features, making it a favorite among video creators. Its integration with TikTok and social media platforms is also reason for its widespread adoption.

&nbsp;

The best part about using CapCut is that most of these important features are available for free, which kicks out budget-constraints out of the window. CapCut uses AI to empower users to create stunning videos without needing any fancy editing experience. With mobile and desktop apps, as well as a web version, CapCut can easily be used on the go. 

&nbsp;

CapCut offers a range of AI-driven features that are very, very impressive. With 'Smart Cuts & Trims', CapCut's AI automatically detects shaky parts, long pauses, or unnecessary bits, suggesting cuts to tighten up your video and eliminate boring footage. 'Effortless Audio Magic' helps isolate voices and enhance audio quality, making your message crystal clear even if there’s noisy background interference. 'Instant Captions' auto-generate captions, making your videos more accessible and saving you time and effort. Moreover, if your filming didn't capture the perfect lighting, 'Light Like a Pro' allows CapCut’s AI to adjust lighting in your videos post-production, creating the ideal mood.

What I most love about CapCut AI video generator is its AI-driven background removal feature which eliminates the need for a green screen, allowing me to replace any background with an image or color of my choice.

&nbsp;

Something to note about Capcut is that the free version includes watermarks on exported videos. which can be removed only with a paid subscription. This may potentially pose a barrier for you if you’re a budget-conscious user.

## **2. Runway**

[Runway](https://runwayml.com/) can be called a virtual sorcerer's wand for video editing! It has the potential to be an entire AI video production studio, so it can basically create an entire movie without filming a single frame — if you have the time, that is. Runway can generate videos and animations from simple text prompts, create animations from still images, generate lip-syncing avatars based on a recording of your voice, and even allow you to train your own AI models. 

&nbsp;

Runway might be a little intimidating in the beginning. That’s because it has an array of features that take time to understand, execute and implement, and that could be slightly time-taking and if I can say, complex.

Another aspect to note is that while Runway offers a range of pricing plans, the advanced features that truly set it apart are locked behind higher-tier subscriptions, making it less accessible for small businesses or individual creators on a tight budget.

## **3. Invideo AI**

[Invideo AI](https://invideo.io/ai/) is an excellent tool for generating a starting point for your videos. It helps you to quickly create a basic structure for your videos, which can then be further customized and polished. The  features allow you to input simple prompts, from which the tool generates an initial draft that includes relevant footage, transitions, and effects. 

&nbsp;

A few standout features of Invideo AI are an extensive stock media library that offers a range of options to choose from, AI-powered script generation which creates a complete script for your videos and, voiceover addition. Their AI voiceover generator is particularly good because it has an option to add your own voice which it then uses to narrate scripts in. The narration is also decent, with multiple accents and focussing well on the feeling, intent, and emotion behind each word.

Video on demand

Live streaming

Analytics

Video infrastructure

Video player

AI features

See all features

Online learning & Corporate training

Social & Entertainment

Marketplace & E-commerce

Communication & UGC

Generative AI

Quickstart guides

Documentation

Ecosystem catalog

Clients & SDKs

Community

Blog

Demos

Customer success stories

Help center

Codec