Contact us
|
+1 888 882-1880
M-F 8am to 8pm EST

USING ONLINE VIDEOS: DIY VS. SOCIAL VS. OVP

bsp-admin-1 on July 13, 2011

More and more companies are discovering the diverse potential of online videos: their immense popularity with users, their ability to evoke emotions, excellent marketing opportunities, significant boosts to product sales, and even improved search engine rankings. However, many overlook the unique challenges and hurdles associated with leveraging videos effectively.

Even with a coherent video concept and professionally produced clips, one critical question remains: How should the videos be integrated into the company’s website? There are three primary options.

1. DIY SOLUTIONS

The first option is an individual solution. A (hopefully) skilled employee encodes the videos, programs a Flash player or configures a freely available player, uploads the videos to an HTTP server, and integrates them into the website.

While this approach may work for a small number of videos with limited views, it quickly becomes impractical when handling more videos or higher traffic. This DIY solution can become disorganized, error-prone, maintenance-intensive, and time-consuming.

2. SOCIAL VIDEO SITES

The second option involves using social video sites like YouTube. By uploading videos to these platforms, companies receive embed codes to integrate the videos into their websites. These sites offer relatively robust statistics on video usage, and the videos are discoverable not only on the company website but also via YouTube’s search.

However, this solution comes with significant drawbacks. By uploading videos to YouTube, companies partially relinquish rights, allowing YouTube to market the videos, which could result in competitor advertisements running before the company’s content. Additionally, YouTube retains control over the video’s appearance and technical specifications, leaving companies dependent on the platform’s decisions. In this scenario, companies hand over control to YouTube and must accept whatever changes or limitations come their way.

3. ONLINE VIDEO PLATFORMS (OVPS)

This is where Brightcove comes in as the third and most professional option. An OVP combines the customization of a DIY solution with the technical capabilities of a social video site while offering legal security and complete control.

Of course, an OVP requires an investment—but it’s no different from other professional tools and systems companies routinely use. For a relatively small cost, customers gain access to a comprehensive, well-maintained, and continuously improved system.

KEY ADVANTAGES OF OVPS

  • Reliable delivery: Videos are delivered via Content Delivery Networks (CDNs), ensuring performance without straining the company’s servers.
  • Detailed analytics: OVPs provide detailed statistics on video views, enabling companies to measure effectiveness and gain insights for future productions.
  • Complete control: Videos remain under the company’s full control, with the ability to integrate monetization through connections to ad servers or advertising networks.
  • Optimized workflows: OVPs minimize technical tasks and risks, streamlining video production and distribution processes.

Online video platforms are designed specifically for the professional use of videos. They allow companies to optimize workflows, reduce technical challenges, and use videos on the internet effectively, securely, and cost-efficiently. By investing in an OVP, companies can unlock the full potential of online videos while maintaining complete control and ensuring long-term success.

10 TIPS FOR ENCODING AND HIGH-QUALITY VIDEO PLAYBACK

bsp-admin-1 on June 17, 2011

The community and knowledge team has scoured through search trends on our support site, numerous inquiries to our customer support team, and the forums to find out what kind of topics you are searching for and asking about. One of the trends we found is questions about encoding best practices and tips for high quality video playback.

Encoding is a very broad topic, but we’ve put together a list of the top 10 things to keep in mind when encoding your videos for upload into Brightcove.

1. Encode for Broad Device Support

In order to ensure that you have broad device support, we recommend encoding your videos to h.264. Uploading h.264 videos provides you with more ways to customize your Brightcove content, and they’re delivered with higher quality and less bandwidth than many alternative encodings.

H.264 allows for broad device support, as it provides the best options for mobile video deliver and is the only format that will playback on our HTML5 smart players. Smart players will deliver your video in Flash or HTML5, depending on your viewer’s device capabilities. This enables you to use a single Brightcove player that can deliver video in Flash or HTML5, so you don’t have to create and manage separate players for each viewer environment and your existing players can automatically load in Flash or HTML5 mode without any custom work or additional JavaScript on your part.

2. Use Multi-bitrate Streaming

If you have a diverse audience watching your videos from around the world with big differences in their bandwidth and internet connectivity, using multi-bitrate streaming is a must.

For those with higher bandwidth, a high quality (potentially even your source) video will be delivered automatically. Those viewers with lower bandwidth won’t have to sit though long buffering times, but instead will be able to watch a slightly lower quality video instantly.

On the other hand, if your videos are strictly being viewed, for example, on an internal network with very strong internet connection, you may want to simply upload one high quality video with only one single rendition.

3. Preserve Your Source File as a Rendition

If your source video file is in h.264 format, you can choose to keep your original source file as an available rendition. This option lets you retain an h.264 master that can be at an even higher level of quality than Brightcove’s highest-quality rendition.

In addition, when you select this option, your h.264 source video is available immediately, as soon as the upload is complete, and you don’t have to wait for the video to be transcoded before it is available in the Media module and in your players.

4. Capture Videos at a Constant Frame Rate

To avoid stuttering during playback, record video at a constant frame rate of 25-30 frames per second, and shoot film at constant frame rate of approximately 24 frames per second.

5. Consider the Content You’re Uploading

Factual or news type videos typically require lesser quality than an action packed longer-form video or an engaging nature film which will require much higher quality.

If you are producing screencast videos with very little action at all, other than the occasional change of a slide, be sure to export the video at h.264 at the same aspect ratio as your recording and use some of these techniques for configuring your player.

One-pass vs. Two-pass Encoding

In general, two-pass produces a better transcode than one-pass. However, a two-pass encode does take up a lot more time to perform. For videos that have a lot of motion, we recommend taking the extra time to export as two-pass. You may also want to perform a comparison test to determine if there is a noticeable difference and keep an eye out during any transitions or high motion areas. If you do not see any noticeable difference, stick with the one-pass and save some time.

6. Upload the Best Quality Source File

While Brightcove can certainly preserve the quality of your source file and create multiple different levels of renditions, we can’t create renditions that are higher quality than the source file you provide us with. We have a list of source file recommendations and export settings to help you create the best quality source file to upload into your Brightcove account.

7. Don’t Ignore Audio Quality

Most people focus on having the highest-possible image quality in their videos and ignore audio completely. However, viewers of your videos are much more likely to drop off if the audio quality drops out or is choppy than if the video is slightly lower quality.

For the most part, if a user can’t hear what is being said in a video, there is no purpose in continuing to watch it. Make sure to use a good microphone when recording your videos and consider our recommended sound settings.

Audio Setting Best Practices

  • Codec. Select “AAC” when encoding h.264 video.
  • Sample rate. When in doubt, choose 44.1 kHz or 48 kHz.
  • Bit rate. Use 128-160 kbps, or 192 kbps+ for content that’s heavy on audio.

8. Deinterlace Analog Videos

If you are working with content that was shot on tape, then we highly recommend you to select the Deinterlace Source Video checkbox when exporting to avoid any sort of interlacing effects on your Brightcove content. If your content was shot to a digital format and is not currently interlaced, then you do not need to deinterlace.

9. Modify Your Transcode Settings

If you have a professional or enterprise edition account, you have the option to modify your transcode settings directly in the Brightcove Studio. By default you have six renditions available in your transcode profile. However, if you have a particularly diverse customer base viewing your videos on different devices and connections speeds, you may want to add additional renditions (up to 10). Brightcove’s default transcoding settings provide a good baseline set of renditions that should meet the needs of most publishers and viewers.

10. Allow Video Smoothing

Brightcove players can use video smoothing to improve the perceived quality of video playback. There’s a trade-off, however, between the added quality you might get by using video smoothing and the additional CPU burden that video smoothing imposes on the client.

With higher video bitrates, the benefits of video smoothing are less noticeable. Viewers, especially those with less powerful computers, may perceive a choppier video quality due to frame dropping, which may outweigh the benefits of video smoothing.

By default, Brightcove uses video smoothing for videos with a bit rate less than 950 kbps, and does not use video smoothing for videos with a bit rate greater than or equal to 950 kbps.

You can override the default video smoothing behavior by including the optional videoSmoothing configuration parameter in your player publishing code and setting it to either “true” (video smoothing always used) or “false” (video smoothing never used).

ENCODING SETTINGS FOR QUALITY HD VIDEO DELIVERY

bsp-admin-1 on January 20, 2011

Brightcove enables publishers to stream HD content to the web. But you need to ensure you’re encoding your source files correctly to take advantage of our high-quality streaming capabilities. For those of us who think “4:2:2 pulldown” sounds like a foreign language, it’s helpful to have a simple explanation of how to deliver true high definition video streaming.

Step 1: Know How It Works

Brightcove uses multi-bitrate streaming technology to give viewers the best video quality their Internet speed can handle. This means we create up to six renditions of your video at varying resolutions and bitrates for a wide variety of internet connections, from blazing-fast T3 office lines to sometimes-spotty 3G mobile connections. Our video players detect viewers’ internet speeds and serve them the appropriate rendition of your video.

If you want to use Brightcove’s default settings, simply upload a file at the highest resolution and bitrate you have. The default settings support just about all video codecs and containers in use today, and your highest-quality rendition will be a very respectable 1280×960 (1280×720 for 16:9 formats) at about 1.8 Mbps. That said, if you’re willing to take a few extra steps you’ll be set up to deliver true HD video content through Brightcove.

Step 2: Upload in H.264 and Preserve the Source File as a Rendition

Since Brightcove’s transcoding process limits the highest-quality rendition to 1280×960 at 1.8 Mbps, the key to squeezing the best possible quality out of Brightcove is to upload your video in a format we can deliver through our players without requiring any transcoding. This format is h.264.

H.264 video can play on PCs, iOS devices, and Android devices. Given its broad compatibility across devices and operating systems, it is Brightcove’s preferred video format. As such, Brightcove gives you the option to preserve an H.264 source and add it to the video’s list of available renditions. The end result is Brightcove’s six default renditions, plus your source file just as you encoded it on your machine.

Step 3: Encode Your Source File in Web-Friendly HD

The key considerations when encoding a video for Brightcove are quality and playback accessibility. There is no reason to include a 10 Mbps source rendition when few, if any, end users will have sufficient Internet speeds to stream 10 Mbps video. Similarly, a 2 Mbps source file will be virtually indistinguishable from Brightcove’s 1.8 Mbps rendition.

Video content at 1920×1080 requires exceptionally high bitrates (between 6-8 Mbps and up) to display clearly, so it’s best to stick to a 1280×960 or 1280×720 resolution for your videos. Most viewers on screens smaller than 35″ won’t be able to tell the difference.

Finally, you need to decide on an encoding bitrate. We suggest a bitrate between 3 and 6 Mbps. Which side of that range you lean to depends on whether you’d rather sacrifice a bit of quality for greater accessibility or vice versa. Just remember this: If a viewer doesn’t have a strong enough internet connection to view your source rendition, it’s not the end of the world. Brightcove will be able to serve them one of the lower renditions our encoding engine created.

HOW TO ENCODE VIDEO FOR MOBILE USE

bsp-admin-1 on September 30, 2010

There are hundreds of mobile devices out there, and it’s basically impossible to support all of them. But the good news is that mobile devices are getting better.

Modern smartphones can actually play high-quality video, and smartphone use is increasing. That’s not to say that 3GP is over, or that everyone has a smartphone. But smartphone use is growing, and not surprisingly, smartphone users are more likely to watch video on their phones.

So if you want to support 90%+ of mobile devices, you need at least two video types: 3GP + MPEG-4 for less sophisticated devices and h.264 + MP4 for smartphones. That’s good news, really. One output video can cover all of your smartphone users: iPhone/iPad/iPod, Android, and (for the most part) Blackberry. You can include PSP, PS3, and Xbox 360 for good measure.

Of course, while one universal smartphone output can take care of most smartphone users, you can do better with multiple mobile outputs. For example, the iPad has a native resolution of 1024×768—five times higher than the 480×320 on earlier iPhones. So if you encode your video at 480×320, you’ll be missing out on the near high-def capabilities of the iPad.

Fortunately, you can target mobile devices well using a handful of standard encoding profiles. Start with the Universal Smartphone Profile for wide compatibility. Then, add in an Advanced Smartphone Profile version for the more advanced devices and round out your mobile list with a legacy profile for widest compatibility (either our Legacy Smartphone Profile below or even a 3GP video for even wider compatibility).

Note that the following defaults are the starting point for these profiles. Brightcove Zencoder uses these settings by default, but you can replicate them easily enough in whatever encoding tool you’re using.

Defaults

  • Video: h.264, Level 3.0
  • Baseline profile audio: AAC, 1-2 channels

1. Universal Smartphone Profile

This is a great starting profile for wide compatibility with modern smartphones. Plays on just about everything, though it doesn’t take advantage of the higher resolutions and codec complexity possible on the newest crop of devices.

Plays On

  • iOS: iPhone, iPad, Apple TV, iPod Touch, iPod Classic, iPod 5.5G
  • Blackberry: Bold 9000, Curve 8910, 8900, 8520, Pearl 9XXX, Storm, Storm 2, Torch, Tour, Bold 9650 + 9700
  • Android: All (?)
  • Other: PSP (3.30+), PS3, Xbox 360, web

Doesn’t Play On

  • iPod 5G
  • PSP (pre-3.30)
  • Blackberry Curve 9330, 9300, 8530, 83XX
  • Pearl 8XXX, 88XX

Settings

Defaults, plus:

  • Audio_bitrate: 128 (or less)
  • Audio_sample_rate: 44100 (or less)
  • Size: 480×320
  • Max_frame_rate: 30
  • Video_bitrate: 1500 (or less)

1b. Universal Smartphone Profile B: Higher Resolution

This profile plays better on iPhone 4g, iPad, Apple TV, new iPod Touch, Droid, PS3, and Xbox, by increasing the video resolution. The extra pixels are wasted on older iPhones though, and make for a video that won’t play on Blackberry and some Android phones.

Plays On

Everything above, minus Blackberry and maybe weaker Android devices.

Settings

Universal Smartphone Profile (above), plus:

  • Size: 640×480

2. Advanced Smartphone Profile

Newer iOS devices allow higher resolutions and higher encoding complexity (which means better compression). In particular, iPad and Apple TV users shouldn’t have to watch 480×320 video on their beautiful screens, so it makes sense to provide a higher quality version if you want to provide a good experience to these users.

Plays On

  • iOS: iPhone 4G, iPad, Apple TV*, newer iPod Touch
  • Android: Nexus One, Droid, maybe others (Note: Some users report trouble with 720p video)
  • Other: PS3, web

Doesn’t Play On

  • iOS: iPod 5G/5.5G/Classic, iPhone 3GS and before, older iPod Touch PSP, old Apple TV*
  • Blackberry: all
  • Android: others
  • Other: PSP, PS3, Xbox 360, web

Settings

Defaults, plus:

  • H264_profile: main
  • H264_level: 3.1
  • Audio_bitrate: 160 (or less)
  • Audio_sample_rate: 48000
  • Size: 1280×720 (max) or 960×640 (iPhone 4 native)
  • Max_frame_rate: 30
  • Video_bitrate: 5000 (or less)

*2b. Advanced Smartphone Profile B: with Old Apple TV Compatibility

To support older Apple TV devices, use the Advanced Smartphone Profile setting, plus one of the following.

Settings

Advanced Smartphone Profile (above), plus either one of the following:

  • Size: 960×540
  • Max_frame_rate: 24

3. Legacy Smartphone Profile

This profile plays on the last major set of H.264-based mobile devices: notably, older iPods and some Blackberries. The tradeoff is significantly smaller video: 320×240, at no more than 768kbps.

Plays On

Everything above, plus:

  • iPod 5G, PSP (pre 3.30)
  • Blackberry Curve 9330, 9300, 8530, 83XX
  • Pearl 8XXX, 88XX

Settings

Defaults, plus:

  • Audio_bitrate: 128 (or less)
  • Audio_sample_rate: 44100 (or less)
  • Size: 320×240
  • Max_frame_rate: 30
  • Video_bitrate: 768 (or less)
  • H264_level: 1.3

4. Legacy 3GP Profile A and B

Finally, a 3GP profile or two will extend support to many remaining mobile devices. Notably, you can use these on most of the same devices supported above under the Legacy Smartphone Profile. So if you’re encoding a 3GP video at 320×240, you might not need to encode another H.264 video at 320×240. Note that 3GP video support is still in beta at Zencoder. Finally, note that these videos will look terrible, but that’s the cost of supporting 3GP phones.

Plays On

Hard to say. There are thousands of types of 3GP devices, and each one is a little different. Consider these a starting point.

 Profile AProfile B
Format3gp3gp
Video_codecmpeg4mpeg4
Size320×240176×144
Aspect_modepadpad
Frame_rate155
Upscaletruetrue
Video_bitrate19252
Bitrate_cap19258
Buffer_sizeN/A16
Audio_bitrate2416
Audio_channels11
Audio_sample_rate1600016000

Summary

If you want to create mobile video, start with the Universal Smartphone Profile. For better quality, supplement this with Advanced Smartphone Profile video. For wider compatibility, add a Legacy profile or two using either MP4 or 3GP. It only takes 1-3 profiles to support most mobile devices.

Edits

Older iPhone/iPod devices ask for the “H.264 Baseline Low Complexity” profile. “Low Complexity” isn’t an H.264 standard; it actually just means “only 1 reference frame.” The jury is out on how much Apple devices really enforce this, but for true compatibility, you should probably use Baseline profile and limit reference frames to 1. You can do this at Zencoder with the new h264_reference_frames setting.

November 23, 2010: A few people have asked about Palm Pre video. The published specs for Palm Pre are very similar to other smartphones:

  • 480×320 native resolution (with 640×480 supported)
  • H.264, H.263, or MPEG-4 video
  • MP3 and AAC audio (along with a few other codecs)

If these specs are accurate and comprehensive, then the Universal and the Legacy profiles above should work on Palm Pre.

January 24, 2011: In order to deliver 3GP video as an RTMP stream it needs to be “hinted.” Add "hint": 1 to your API request to enable it.

DOING Y COMBINATOR IN YOUR 30S

bsp-admin-1 on September 21, 2010

The stereotype of a Y Combinator founder goes something like this.

Programmer. Smart and driven. Went to MIT or Stanford, or else skipped college altogether. Able to live on a ridiculously low salary. Enjoys working 90 hour weeks.

And of course: Young. Early-to-mid 20s.

I almost didn’t apply for Y Combinator last year because I thought I was too old. I was 31 at the time. The application asks for age, and I figured they would just ignore anyone who wasn’t in their 20s. But friends of mine made it in to the summer class, and they were older than I was, so I thought I’d try.

Granted, it can be more difficult to do YC when you’re advanced in years, like myself. I have a wife, two kids, and a mortgage. But there are advantages to being a bit older too. Since applications are open for the Winter 2011 class right now, I thought I’d share my experience.

Here is a quick guide to doing Y Combinator in your 30s.

Does YC Accept Founders Older Than 30?

Yes. And according to Paul Graham, age is neither a penalty nor a bonus. He writes:

“I don’t actually know the numbers. We don’t keep track. But I know there haven’t been any with founders in their 50s, and only 2 or 3 with founders in their 40s or their teens. Most founders are in their 20s or 30s. Completely guessing, I’d say 15-20% have founders in their 30s.”

That was in 2009, so the numbers may have changed.

Basically, the age distribution of YC companies is pretty close to the age distribution of applicants. More YC founders are 25 than 35, but more 25-year-olds apply than 35-year-olds.

Paul has an essay where he says the ideal range to start a startup is 22-38. As far as I know, that isn’t a rule, just a suggestion. YC has accepted plenty of people younger than 22, and at least a few over 38.

I’m Over 30 and Want to Start/grow a Startup. Should I Apply to YC?

Sure. If you can be in the Bay Area for 4+ months (3 isn’t enough if you need to raise money), and if you think you’d benefit from some excellent mentoring, then take a shot.

How Can You Afford to Do YC When You’re So Old?

It isn’t easy. A team of three only gets $20K. For most people, that’s not enough to live on in the Bay Area. When you’re 21, you solve this problem by living cheaply. Cheap apartment, no office, eat inexpensive food, don’t travel, don’t buy stuff. A low burn rate is a huge advantage for a startup, and if you can keep that up after YC, you’re a step ahead of the old guys.

Living cheaply is a bit harder to do when you’re older; when you’re 31, you’re more likely to have a high burn rate (due to family, mortgage, or more expensive lifestyle, etc.). But at 31, you have some advantages, too, like savings.

If you’re a good programmer at 31, you’ve probably made a reasonable salary for at least a few years. I supplemented our meager YC funding with savings. I also had run a software consulting shop leading up to YC, so I even had a bit of income in during my YC months (as my employees finished up a few projects).

What Happens When Your Savings Runs Out?

You raise money, or get profitable. If your entire team can live on $5K/month, getting profitable is much easier than if you need $25K/month. So if your expenses are high, then you’re often forced to raise money rather than getting profitable immediately.

Personally, I love the idea of running lean, getting profitable, and avoiding big funding. It isn’t for everyone, but it resonates with me. But it’s really, really hard to do, especially when you have a team that needs near-market salaries. So paradoxically, raising money is the easier option for most of us.

But Raising Money is Still Hard. Isn’t That Risky?

Yes. It is risky to drop everything with just a few months of runway and hope to raise money quickly. If you can’t take that risk, don’t do YC, or do a startup that doesn’t need investment. Raising money is hard, uncertain, and can kill a startup. But running out of money is guaranteed to kill a startup, so raising money is often a good idea.

The good news is that Y Combinator does an amazing job of preparing founders to raise money. You get great mentoring (by YC itself, by YC alum, and even by unaffiliated angel investors), a small PR boost, and the opportunity to pitch 100 quality investors at Demo Day. No guarantees that you’ll raise money, but a lot of YC founders do.

Is It Easier or Harder to Raise Money in Your 30s?

I don’t think it makes a big difference. At the end of the day, investors are looking for strong founders and a giant opportunity. Young founders have some things going for them, and older founders have others.

What About Co-founders?

This is important. If you’re 31 and have enough saved up to work on your app for 6 months, but your co-founders don’t, that’s a problem. To some extent, your company’s runway is only as long as the shortest co-founder runway. If your runways are uneven, be sure you have a clear understanding up-front of what happens when the first runway runs out.

What About Spouse/kids/mortgage?

This is very situation-dependent. I have friends who could never do what I do, because their home lives wouldn’t support it, or they couldn’t swing it financially. But I have a very supportive wife, and a little bit of savings. This made YC possible for me.

Aren’t You Supposed to Pay Yourself Like $30K/year as a Startup Founder?

Yes and no. I think the rule of thumb is, “pay yourself as little as possible,” which equates to, “pay enough that finances won’t keep you from focusing on your startup.”

Investors understand this. If you have kids and your real burn rate is $90K/year, pay yourself $90K/year. If you pay yourself $60K, you’ll have trouble sticking with the startup for the long-term. But if you only need $30K/year and you pay yourself $60K, you’re doing your startup a disservice.

If that sounds unfair (“Why should I only take $30K when the founder down the block is taking $90K?”) remember that if you take a low salary, you have an immediate advantage over the founder who needs a big paycheck. All else being equal, your startup is less likely to die than the expensive startup down the block—at least early on. If you have a team of 15 and VC funding, by all means, take a real salary.

Do People Look At You Funny Because You’re So Old?

No. Believe it or not, many startups have founders in their 30s or 40s. I even hear that there are people in their 50s doing startups. Shocking, I know.

Actually, it turns out that the 22-year-old founder/prodigy is the exception, not the rule. Based on a study of 549 entrepreneurs, the average age of a founder starting a “high-growth company” is 40, not 25. This is partly true because most of us aren’t prodigies, so we need a decade or two of experience and failure before we start to figure things out. But I think the biggest reason is simple: Founders don’t stop with one startup. If you enjoy starting a company at 25, there is a good chance you’ll enjoy starting one at 40 too.

REFERENCE IDS: HOW TO ASSIGN CONTENT TO VIDEO PLAYERS

bsp-admin-1 on July 9, 2010

Publishing videos via the Quick Publish tool in the Media module is fast and easy. But for those of you with large content libraries and a large number of video players throughout your web properties, there’s a trick that can save you time. This post will show you how to use the Reference ID feature of the Studio to simplify the process of programmatically assigning content to players.

PROGRAMMATICALLY ASSIGNING CONTENT TO PLAYERS

You may have noticed that all Brightcove publishing code follows a standard format. What this means is that you can programmatically swap out a few ID fields and change what the embed code brings up:
  • PlayerID parameter. Determines which video player is displayed.
  • @videoPlayer parameter. Determines which video is displayed (single video players only).
  • @playlistTabs parameter. Determines which playlists are displayed (tabbed playlist players only).
  • @playlistCombo parameter. Determines which playlists are displayed (players with dropdown playlist selection only).

Next, let’s look at how to use the Reference ID to simplify the process of programmatically assigning content.

USING THE REFERENCE ID IN PLACE OF ID NUMBERS

All videos, playlists, and players are assigned a unique ID number that’s used for identification purposes, specifically with the parameters I mentioned above. The Reference ID allows you to add another unique ID to use in place of that number. This can dramatically simplify the process of adding a new player to your site, since you won’t need to constantly refer back to the studio to get the ID numbers.

The Reference ID can also be handy if you’re managing your videos and web pages in a content management system (CMS). You can assign a Reference ID to a video that matches the ID used by your CMS, so you can easily match up the same info in the CMS and the studio.

Remember: You can also assign a Reference ID to playlists and assign playlists to your video players via the embed code instead of going through the Studio. Just be sure to keep all your Reference IDs unique so that Brightcove knows which video or playlist to display.

HOW TO EDIT THE REFERENCE ID

This step is simple: Just select your video in the Media Module, click on Edit at the bottom of your screen, and check the box next to “Reference ID.” A small text box will pop up, allowing you to enter a reference ID.

VP8, WEBM, AND THE FUTURE OF HTML5 VIDEO

bsp-admin-1 on May 19, 2010

In 2010, Google announced a new open-source video project: WebM, which uses the VP8 codec. This is a big deal for digital video. Brightcove has launched VP8 transcoding support as well, and we’re committed to offering the best VP8 transcoding on the market.

WHAT IS VP8? WEBM?

VP8 is a video codec, like h.264, VP6, Theora, etc.

WebM is two things: a container format (based on Matroska) and the name for an open-source video project. Presumably, WebM could contain other open video and audio codecs down the road.

DOES BRIGHTCOVE SUPPORT VP8?

Yes. We’ve worked with Google to build VP8/WebM support into Brightcove.

WHERE DOES VP8 COME FROM?

Before Google bought them, On2 was one of the few companies in the world that made proprietary video codecs. Most video and audio codecs are industry standards, combining patents from a few dozen companies.

MP3 audio, for instance, contains 100+ patents. Anyone can read the MP3 spec, but to implement an MP3 encoder, or to decode or encode MP3 audio, you have to license these patents. Conveniently, the patents are grouped together into a patent pool, so you only have to sign one license, not 50, in order to do something with MP3.

On2 was different. They were a commercial company that created their own video codecs, like VP6, VP7, and VP8. And their codecs were pretty good; VP6 became a big deal on the web when Macromedia licensed it for Flash 7 and 8.

At the time, it was a major step forward from h.263, the Flash 6 video codec. It was even almost as good as h.264 for a little while, before h.264 encoder implementations matured. Overall, though, h.264 is a better codec than VP6. So when Flash 9 got h.264, the need for VP6 was diminished a bit. Not that it went away; it’s still the second most watched video codec on the web.

WHAT’S THE DEAL WITH HTML5 VIDEO?

HTML5 includes a <video> tag, like an <img> tag, which browsers will use to natively play video, without requiring a plugin (e.g., Flash Player or Sliverlight). But it doesn’t specify a standard codec.

Firefox and Opera backed Theora, an open source codec, but a generation or two behind the best video codecs available today. Theora is actually based on On2 VP3, which On2 released under a BSD-like license several years ago.

Apple and Microsoft won’t include Theora support in their browsers because it’s worse than h.264, and because they’re concerned about submarine patents. Safari and IE backed h.264, which is the best codec on the market, but not free. H.264 requires a license from MPEG-LA (who manages the patent pool).

The terms aren’t currently onerous; h.264 is free or cheap for most users, at least in 2010. But that could change down the road, and even if it doesn’t, the folks behind Firefox and Opera don’t want to use a patent encumbered codec for philosophical and pragmatic reasons.

At the moment, HTML5 video is at an impasse. Firefox and Opera won’t support h.264, and IE and Safari won’t support Theora (Google Chrome, incidentally, supports both).

WHY DOES VP8 MATTER?

VP8 is a good video codec, and it is open. That makes it unique in the world. Theora is open, but not a great codec. H.264 is a great codec, but patent encumbered.

Firefox, Chrome, Opera, and Adobe currently support VP8, and Safari and IE may at some point. So at best, VP8 will completely solve the HTML5 video problem. At worst, it is major step forward for open video. And either way, a high quality open-source video codec is a Good Thing for everyone, even h.264 users.

HOW GOOD IS VP8?

It’s really early, but it appears to be good. H.264 is probably still better, but VP8 is better than most other codecs (including Theora, VP6, MPEG-4, etc.).

Also, VP8 will get better as its encoders get better. Even with the same codec, the difference between one encoder and another can be huge. When h.264 was first released, it wasn’t that much better than VP6 or MPEG-4 ASP (e.g., DivX and XviD), because the first h.264 encoders were rushed to market. But five years later, h.264 encoders have gotten significantly better.

Give VP8 a year (let alone five), and it’s going to get better and better.

WHAT DOES THIS MEAN FOR H.264?

Competition is good. H.264 has liberalized their licensing terms a bit in the last year, and part of this is probably due to pressure from Firefox and Opera. VP8 will only accelerate this; a credible competitor to h.264 will encourage the h.264 patent holders to offer favorable licensing terms.

IS H.264 GOING AWAY?

No. A lot of companies are committed to h.264, including Apple, and many devices (like the iPhone) include hardware-based h.264 decoders. Beyond that, h.264 is still the best codec on the market, so bandwidth-conscious businesses will continue to use it.

WHAT AUDIO CODEC DO I USE WITH WEBM/VP8?

Ogg Vorbis. Unlike Theora, which is quite a bit behind h.264 in quality, Vorbis is actually quite good as an audio codec. It’s competitive with MP3 and AAC.

HOW DO I PLAY VP8 VIDEO?

Download Chromium, Firefox, or another browser. Other players will pop up soon.

HOW DO I ADD VP8 TO MY WEBSITE?

Checkout our video.js project, an open-source universal video player. Video.js provides a single embed code for HTML5 video, Flash video, and iPhone and other mobile devices.

VAST: WHY VIDEO AD RESPONSE STANDARDS ARE IMPORTANT

bsp-admin-1 on April 26, 2010

This post will cover what VAST is, why this standard is so important, and what’s happening with VAST 2.

WHAT IS VAST?

VAST, or Video Ad Serving Template, is the standardized way to deliver an ad response to a video player. This is done with XML which contains information about the type of ad, where the ad creative is located, events to fire when certain things happen, and much more. The VAST specification explains in detail what all can be done.

VAST was created by the Interactive Advertising Bureau (IAB). The IAB put together this specification by talking with a lot of ad server companies, video platforms, ad agencies, and other people interested in getting a standard format. Brightcove was one of the companies who helped with this specification, and we used our knowledge from the Brightcove ad XML to suggest changes to the specification.

WHY IS VAST IMPORTANT?

Having this standard ad response allows for ad servers to work very differently. Instead of spending a lot of time worrying about the differences between different video players and the types and formats of responses they are expecting, the ad server can ideally return the same ad XML to all places.

More importantly, ad servers often talk to each other, grabbing ads from different places as needed and requested. This is the real power of VAST, which is that it allows third-party ad serving to happen easily. This makes things like the real-time bartering of video ads possible.

WHAT IS HAPPENING WITH VAST 2?

The first version, VAST 1, had some support in the industry, but VAST 2 has completely overtaken it. Many ad servers, like DoubleClick’s platform have announced supported for VAST 2. And a growing number of video players, including Brightcove’s player, allow its use.

Besides the wider support, VAST 2 also has a lot more of the elements and attributes that are needed for ad responses. There are a number of limitations in the VAST 1 specification, which does not allow ad creatives in certain places, misses some events that people wanted to know about, and has elements that need to be clarified.

While VAST 2 is now recommended, Brightcove implements both VAST 1 and VAST 2 specifications. We are also continuously adding new features to the player to allow for more of the VAST 2 elements to be used.

HOW TO ENCODE H.264 USING MEDIACODER

bsp-admin-1 on August 18, 2009

For some of our publishers and partners, there exists very valid reasons to use their own external media transcoding workflow which excludes the use of the BC3 Server-Side Transcoding functionality.

Those that need to really turn the screw on control of end quality and size, you’ll need to really dig deep in the more advanced levers available in the encoding process. Things like the Profile, Level, and Predictor Frames are just some of the things you need to tweak to squeeze the most from your encoding.

For 99% of us, this is just too much specialized detail to get our heads around. Luckily the internet is here to save us as a gateway to the real experts out there.

Here’s something to try.

How to Get MediaCoder

Download MediaCoder, an open sourced GUI tool that allows you to transcode your videos to h.264 free of charge. They have a command line tool for a cost if you like it enough.

How to Set Up MediaCoder

Set up MediaCoder following these instructions.

File menu –> Revert All Settings

Audio Tab

Select what makes the most sense for you. Remember that audio takes up space and the majority of your audio will be consumed and limited by the quality of PC speakers, so don’t feel you need to up much here.

Video Tab

  • Mode: Bitrate-based
  • Bitrate: 500 kbps (start with this and vary it to see how things improve)
  • Format: h.264
  • Container: MP4
  • Source: (Auto-Select)
  • Encoder: (Auto-Select)

x264 Tab

  • Profile: High
  • Predictor Frames: 6
  • Motion Est. Mode: Normal
  • Motion Est. Range: 16
  • Level: 5.1
  • B-Frames: 5
  • Subpel Refinement: 7

MediaCoder Preference Window

Click the “Advanced” button –> MediaCoder Preferences Window

Video Encoders

|—–x264

  • Bitstream profile = Baseline (highlight and select “Baseline” in the drop down box below)
  • Level of bitstream = 51 (enter value in the box below and click apply button)
  • Frames used as predictors in B and P frames = 6 (enter value in the box below and click apply button)
  • Maximum B frames between I and P frames = 5 (enter value in the box below and click apply button)
  • B-frames decision = true (click “Yes” radio button below)
  • B-frames used as for predicting = true (click “Yes” radio button below)
  • Use CABAC = true (click “Yes” radio button below)
  • Direct macroblocks prediction = Auto (highlight and click “Auto” in the drop down box below)
  • Use weighted prediction for B-frames = true (click “Yes” radio button below)
  • Adaptive spatial transform size = true (click “Yes” radio button below)
  • Fullpixel motion estimation algorithm = Hexagonal (highlight and click “Hexagonal” in the drop down box below)
  • Subpel refinement quality = 7 (enter value in the box below and click apply button)
  • Mixed reference frame = true (click “Yes” radio button below)
  • Rate-distortion optimization of B-frames macroblock type = true (click “Yes” radio button below)
  • Refine motion vectors use in bidirectional macroblocks = false (click “No” radio button below)
  • Rate-distortion optimal quantization = Enable during all mode decisions (highlight and click “Enable during all mode decisions” in the box below)
  • Performs early skip detection in P-frames = false (click “No” radio button below)
  • Threads (0 for auto) = 6 (enter value in the box below and click apply button)

Alpha and Beta Deblocking

The only settings left under the MediaCoder Preferences window is Alpha and Beta deblocking. For more information, please read How To Use Mpeg4 AVC Deblocking Effectively (Small FAQ).

Deciding what values to use is a personal choice. In this example, we’ll be using more detailed look and occasional block setting of -2:-1.

  • Alpha parameter of deblocking filter = -2 (enter value in the box below and click apply button)
  • Beta parameter of deblocking filter = -1 (enter value in the box below and click apply button)

Click the “X” button in the upper right-hand corner of the MediaCoder Preferences window.

Muxer Tab

  • Radio button: Enable
  • Multiplexer: MP4Box

Check Mediacoder Output

Run your videos through and see the output. Remember, the better your source file the better any transcode output will be.

You should see much less blockiness or quality degradation during fast moving, complex scenes (look at the waves in transition).

This file is ready for upload and immediate playback by any Brightcove player as well as the iPhone (thanks to the Baseline setting).