Campfire Audio Andromeda Review: Holy Cow, This is a Dream, Awesomesauce!

I said "990bps should be sufficient to capture 44/16 music losslessly though". 44/16 MUSIC is easily compressible losslessly into FLAC, which tends to achieve about 50% compression and thus well within 990bps. I never made any statement on what Sony was actually doing. Just saying that Sony could easily send 44/16 MUSIC lossless via 990bps if they wanted.

Please read carefully and understand that not all compression is necessarily lossy (or that lossless compression is possible) before you tell me I'm the "one making up definitions". Literacy is your friend.

 

Yup. I shall quote myself: I never made any statement on what Sony was actually doing. Just saying that Sony could easily send 44/16 MUSIC lossless via 990bps if they wanted.

@TomHP: Given your problems with basic reading comprehension and uncontrollable urge to argue small inconsequential points, may I suggest Head-Fi, particularly the Sound Science section for you? You don't seem to provide any value other than stating the obvious, providing a graph or two showing us the obvious, hiding behind a veil of secrecy (that you are employed by some headphone company or work in some research capacity), and being kind of rude without first earning the right to be so.

 

If Sony wanted to, and assuming there are QoS (Quality of Service) provisions in the transmission protocol, then it is definitely possible for Sony to use a FLAC like lossless compression if 990kbps can be maintained, but fall back to a lossy modes, at same bitrate or less, when it cannot. It depends upon the recording. It seems a lot of hot recordings go over 1000kbps for me as well. The data can be buffered to account for momentary bursts.

 

Which ones? Where are your industry insights? Your sole contribution is telling us that moving a headphone on a coupler with an ear will result in changes in dips and peaks, something we all know from Tyll's measurements. Give me a break. Give us real insights without the excuse of "oh that's secret" or bugger off. Specific data, what, where, how, is what it's all about, not generalized data. There's an expectation when you come off as the all-knowing one.

 

You do know that lossless compression or use of a lossless codec allows for perfect reconstruction of a file, right? I understand you're an engineer (the M1 was an interesting headphone at it's price point when it launched) but it seems like you have a fundamental misunderstanding of how software works. It's entirely possible to retain all the original information in a file while reducing file size. Especially in the case of audio, linear prediction is pretty efficient. So yes, it is possible that a redbook file may be lossless (not "effectively" lossless) when streamed at a lower than 1411kbps bitrate provided the decoder can reconstruct all of the original information in it's entirety.

Now, whether or not LDAC is lossless or not is another question. I can say the quality if quite high.

@Marvey I'm certain they're not doing this, but with 909kbps and an efficient enough codec, you can transmit extra information when the necessary bitrate is lower and have enough buffer for spikes.

Edit: also, it is theoretically possible someone could have developed a more efficient lossless codec as well. Again I doubt it was Sony but as long as we're having academic arguments instead of measuring and listening...

 

I'm pretty sure Sony is not either. It would take a gargantuan effort to do so (well actually not, a small team of good coders versed in data transmission and codecs can do it). For all we know, Sony is probably leveraging the ATRACs codecs! Actually ATRAC Advanced Lossless (ALL) can do both lossless and lossy. One stream using ATRAC3, and then another stream with the delta. This could be scalable depending upon bandwidth availability. Scalable on the fly in near real-time too!

Yes, I owned MiniDisc and cassette Walkman. How many Sony fanbois can say that?

 

Now that I think about it, I'm not so sure of Sony not being able to pull this off. If they leverage their ALL codec, they can send a lossy ATRAC type lossy data along with a delta stream that corrects it to get the original 44/16. The delta stream can be sent losslessly compressed if there is enough bandwidth, or lossily compressed or even decimated to fit in the remainder of the 990kbps. This way, up to 90% of of my 44/16 content could make it through unscathed.

Sony can be pretty darn smart sometimes. There's probably a reason why they aren't saying much about how LDAC works.

 

That too.

 

I'll bug the BT guys and Sony about the inner workings of LDAC later. Suffice it to say it's one of the coolest things upstreamed into AOSP for this release and I was very excited when I saw the submission.

I wasn't talking about your subsequent posts. I was clearing up the issue in the post I quoted.

Also, I don't think it's your style of writing that conflicts with the style of the "head honchos." I think you're doing a few things that rub people the wrong way:

• the e-penis waving about your credentials
• you remaining super opaque about the fact that you work at [redacted] as a headphone designer while waving said e-penis (you can't have your cake and eat it, too)
• you making incorrect statements then hand waving about addressing them later

It never works out well to join a community, tell everyone how much of an expert you are, be very vague about why, then drop weird hints about it while being disagreeable in a thread. Marv might have made some incorrect assumptions but you added to the confusion and have been a grumpy pants.

 

Everyone knows I've been in a grumpy pants mood lately with the Jude gaslighting stuff and today with the Amphibianphonics shit.