BBC Blog Network

User Profile

BtEtta

Listed below are comments made by BtEtta between Friday, 10th October 2008 and Thursday, 7th January 2010 (14 Hours Ago) .

«Newer
Older»

Round up: Thursday 7 January 2010 5:06pm on 07 Jan 2010 Viewer00, I read it as:

The difference between 'a' and 'b' — where 'a' is the old encoder, and 'b' the new — is smaller than the difference between 'x' and 'y'; 'x' being a show that encodes well, and 'y' being one that particularly taxes an encoder. The idea presumably being that any perceived change in quality will be more likely, statistically, to be the result of content, than of encoder choice.

Nothing the BBC say or do now (justified or not, I don't get the HD channel) will stop the people with a hate-on though.
BBC Mobile: Carousel of Content 5:13pm on 08 Dec 2009 What would be nice is if you made use of the X-Forwarded-For header supplied by the Opera Mini/Opera Turbo proxies so we don't get forwarded to the international version because of an apparently Norwegian IP; I was wondering why I couldn't see the changes.

Also, the prev/next buttons don't get added for the Opera 10 Symbian beta because (from what I can tell from a remote debug session, which I'm very thankful to you for giving me the first excuse to try that) you simply don't supply the <script> tags. I presume this is what you mean by limited range of handsets, as my (painfully slow, but still competent) Webkit based Nokia browser does get given the buttons, as does the desktop version of Opera slightly confusingly?
Round Up: Internet Radio 7:03pm on 30 Nov 2009 Bit disappointing that the codec review got some basic things wrong:

"Originally designed as a lossless codec, the developers of Vorbis switched to make this codec lossy at bitrates lower than 96kbps."

Wait what?! Utter nonsense; lossy at all bitrates. Hence why FLAC was brought under the Xiph.org banner to fill the lossless role. Development was galvanised after Fraunhofer started waving their MP3 related patents around in much the same way as PNG being a response to Unisys's patents covering GIF images. Forced to agree that it's been superseded in quality by AAC+ (though not so much against regular AAC) and that its hardware support is disappointing — although the current drive to make Vorbis (and Theora for video) standards for the <video> and <audio> tags in HTML 5 is certainly a more promising development.

"Encoder efficiency is worst out of others compared in this document. MP3 needs high bitrates to come close to AAC quality at lower bitrates. MP3 technically requires a commercial license, though there are some less than efficient 'open source' solutions."

LAME is as I understand it held to be easily the equal in quality to the Fraunhofer codec and has been for some time. Certainly its quality as an encoder (especially since it started life as a series of quality-related patches to the ISO reference encoder) has never been something to be dismissed.
RealMedia - an update 2:59pm on 10 Oct 2009 Also, why did the preview show single line breaks with a clear space between paragraphs when the posted version doesn't.

Please imagine I intended for my paragraphs to be spaced like this, and the preview assured me that was what would happen.
RealMedia - an update 2:56pm on 10 Oct 2009 @Slated
You're missing the difference between encryption in general, and DRM specifically.
With DRM systems the intended recipient is also the potential attacker. If you want to stop someone from ripping a video/audio stream but still allow them to watch it in certain players then you have to supply the decryption key — but still hide it somewhere. This is of course doomed to fail: CSS, AACS, CPPM, Microsoft's Janus — they've all been broken partially or totally (though I believe the most recent iterations of Janus remain secure for the time being). Open source software would be useless in this regard because if you can look at the source you can either see exactly where the key is stored, or better yet compile a new version that helpfully dumps the key in a location of your choosing.
With GnuPG and the like the person having the decryption key is not going to also be an attacker; so the system remains secure despite the code being open source.
DRM is a dead-end though (for exactly the above reason), and Real is an outdated format killed off by its horrible, bloated excuse for a player/plugin. The BBC has been the last place I've seen such a stream for a very long time; I'll grant you I don't know the situation for Internet radio hardware though. A poll of the top 9 best-selling devices on Amazon however has them all claiming support for WMA — one notably claimed no support for Real formats.
Also, because this has been bugging me throughout the comments. Flash is not a streaming format in itself. RTMP was developed for Flash, but within that protocol both MP3 and AAC are valid; AAC I believe can be either raw, or within an MPEG-4 Part 12 based container (MP4, or 3GP mostly). Outside of RTMP content can also be streamed over HTTP, and there are a few more codecs beyond MP3 and AAC that are supported in the full Flash Player. It's worse frankly than calling Ogg a codec. :P
BBC iPlayer & N95: Update 9:49pm on 10 Oct 2008 The above comment is simply not true.

Having recently upgraded from an E65 to the N96 I can confirm that while the Real Player application can indeed still be configured for always use your 3G connection if it is launched via the iPlayer application it will use the WiFi connection. When I've used Real Player on my E65 it always used its configured connection (and annoyingly could not be configured to always ask) rather than what the launching process was using.

In fact, the task list will not list a separate process for Real Player when launched from iPlayer, suggesting the difference is in the ability to embed Real Player within your own application.

I'd guess this feature was introduced with S60 Feature Pack 2. Only three devices currently available use this latest version. The 6220 Classic has no WiFi, so that's out. The N78 might be a better proposition, but seems to lack (from what I can tell) hardware accelerated video, this coupled with the highly limited built in storage makes a very good case for no official support.

The point is not whether it works, just that the BBC don't want take the risk (and quite frankly, it *would* happen sooner rather than later) of someone blaming them for a sky-high bill due to 3G usage.

«Newer
Older»

^ Back to top

BBC MMVI

The BBC is not responsible for the content of external internet sites

BBC.co.uk

BBC Blog Network

Find and talk to the BBC's bloggers.

User Profile

BtEtta