April fool or real?

[Owen Smith]

Read the fine print. In a later post BigShot says his 256kbps AAC for transparency was mono and you'd have to double the bit rate for encoding a stereo track. I'm not sure whether you can go above 320kbps with AAC, certainly a fair number of codecs and transports impose that limit for various reasons (I think it goes back to Dolby Digital on 35mm film using just one side gaps between the sprocket holes gets 320kbps because that's what the optical squares can encode).

Anyway, even if you could use 512kbps AAC I can't see the point. You might as well add another couple of hundred kbps and use FLAC.

-- 
Owen Smith <owen.smith at cantab.net>
Cambridge, UK

On 7 Apr 2017, at 15:50, RS <richard22j at zoho.com> wrote:

>> From: Jim web
>> Sent: Thursday, April 6, 2017 13:50
> 
>> I've now had a reply from someone-who-should-know and they said that ATUI
>> as yet the BBC is only committed to a 4 week trial, and may then decide
>> what next.
> 
>> But it is, indeed, a real test project, not an April Fool. So I'll see if I
>> can set up things to do some measurements to help assess the tests.
> 
> The web page says it is there for 6 months.  However it is only one short piece of choral music.  I wonder what the BBC is hoping to get from it.  The suggestion that anyone is going to hear a difference with laptop speakers seems a bit odd.  Like Jim I would have thought it was necessary to be able to download it so that it could be played on better equipment than the browser was running on.  Also there doesn't seem to be anything to compare the FLAC with.
> 
> If the trial is to be of any use I would have expected it to include music which was difficult to encode and compress.  It is not my field of expertise but intuitively I would have expected that to be fast transients such as percussion, piano, pizzicato violin, harp and instruments with lots of overtones.  David Lake mentions audience applause.  I would have expected choral music to be easier to compress than these.
> 
> I came across this forum post a few weeks ago.
> http://www.head-fi.org/t/728636/public-listening-test-of-codecs-mp3-aac-mp4-ogg-vorbis-and-opus
> I thought post #2 by bigshot was interesting.  He says for each codec the point of transparency should be found.  This is the bit rate at which artefacts from the most difficult sounds to encode and compress cease to be perceptible, so there is no benefit from increasing the bit rate further or going lossless.  He says from his tests that is 256kbit/s VBR AAC .  If he is right there should be no audible difference between 320kbit/s AAC and FLAC.
> 
> When are we going to get 256kbit/s AC3 TV sound on the iPlayer to match the HD satellite broadcasts?
> 
> 
> 
> _______________________________________________
> get_iplayer mailing list
> get_iplayer at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/get_iplayer