Has Tidal died?

Streaming is very much the present and the technology of pushing files around networks pre-dates the CD. I have been streaming only for several years, although I do use CD playback in the car. What I don't do is stream music I do not own a copy of, such as that available from a paying a subscription service. There are many ways to stream many things. Some actually work all the time and do not involve a pay per play model and a remote server.

Hi Simon 

Just wondering, as you will be very clued up on these things, if you know why Apple TV always seems to just work, streams seamlessly and never falters (with both music and video), yet the likes of Tidal really seem to struggle. I suffer dropouts with Tidal, not as bad as some (or prior to using our VM router in modem mode) but still very frustrating nonetheless.

As mentioned, our network is set-up with our Virgin router on modem mode and with an Asus router handling the all the Wi-Fi duties and since doing so things have improved a lot with Tidal but we do still suffer dropouts.

Thanks in advance 

Gingerbeard 

In fairness gingerbread I think we all believe that tidal is a massive multi corp with everything it needs. Realistically I would guess its trying for an organic growth strategy, trying to keep ahead of the curve, bringing on server space as demand increases etc.

Apple has circa 145 billion in the bank and no doubt has on shore and off shore server farm space that is able to keep up with the unbelievable numbers it deals with. Its had around 100 billion downloads from its app store since 2008, these numbers are eye watering and they have massive infrastructure to deal with it.

Also tidal is endorsed by a bunch of total douche bags, so  I don't know what anyone is expecting really.

The chief douche bag is also constantly leaking all over the professional managers; hence more CEO changes than toilet rolls in the exec karzy.   

One obvious answer to that question is the statement at the beginning!

Streaming from your own storage, or playing CDs etc, means it works every time, when you want it, no hiccups or stuttering. 

whilst online streaming would seem a reasonable way of listening to see if you like something, or maybe to listen to some piece of music that you know is only of transient interest, buying a copy, whatever the format, ensures you can have it to listen to when you want, without paying for it over agin (as in subscription), it won't disappear because the service has gone bankrupt, or removed to create more storage for something more popular.  

Hi Gingerbeard, AppleTV, at least my version of it, works by downloading the file locally to itself and then playing from that file. It can start playing from that file while its downloading and can cope with quite varying throughputs at the expense of latency. This is different from web streaming which uses a different technique which in fact is pretty identical to local UPnP media streaming and requires a more consistent data throughput, but requires less memory on the client, and is more realtime for lower bandwidth connections.

Simon

That is a fair point Garryi. Although, that said, I would still expect a company that is providing a streaming service to at the very least be able to supply this without the constant dropouts. After all this renders their product useless. I do understand that there are likely to be teething problems, but this has already been going on for sometime and lets not forget that they are offering a premium product, at a premium price. Perhaps some of Jay-Z's many millions should go on purchasing more servers! 

I actually really like the product and when it works well, I could easily see myself abandoning CD's altogether. But it would need to be infinitely more reliable and stable before I did so. 

Thanks Simon, I wasn't aware of that and that is interesting to know. It must really be quite something, to see an idea and technology that you worked on in its infancy grow and start to become mainstream. A bit like watching your baby come of age  

Based on what Deezer stated in its last annual statement, the cost of sales was around 90%, the bulk of which was music Co payouts. On that basis I'd say it was the music industry killing streaming. 

Tidal's dropping out multiple times per track here in Singapore today, it's been really unlistenable for me.

If this happens even a few hours a week when I want to listen, I'll be sticking with Spotify for discovery along with hi Res download services

Last night was pretty bad. We had friends for dinner and it's somewhat embarrassing when your expensive hi fi cuts out all evening. 

When Tidal arrived on the NDS I felt a sense of completeness. Finally, I owned a digital hub which (along with the record player) provided all my musical needs at superb sound quality. I could continue to buy the music I love and enjoy new music streamed into my home. However, continuous dropouts will limit my use of Tidal. I had hoped Qobuz would be next and that I could swap back to it if Tidal prove themselves unable or unwilling to sort out their IT issues. Sadly, financial problems at Qobuz make that unlikely.

Keith

Worth a try is changing DNS server....to public DNS from google.

For full instructions follow this link (is that allowed to put here?) https://developers.google.com/speed/public-dns/

Be careful though, to write down your original settings!

Good Luck!

Rik

And why on earth would that make any difference on TCP latency and throughput issues? At the TCP level things like DNS are meaningless abstractions..... The real time media transfer uses transport level IPv4 addresses, not application level URLs.....

For the best web browsing experience however I would always advise using your ISP's DNS. This will normally be assigned to you when your internet router makes its PPP connection to your ISP... and best not fiddle around with

Simon

Using a public DNS is likely to add a whole bunch more problems unless you know exactly what you're doing. Not for the faint of heart, and I can't see what advantage it would give in this case.

CD-quality music streaming presents a unique set of challenges. With tv streaming people accept a lag time for the programme to download sufficiently to start playing; systems such as Sky+ calculate the speed of the connection and don't let you start watching until you are at the point that the whole thing will have downloaded before you get to the end. It's a simple of way of (mostly) avoiding pauses in playback. Apple tv does that too, and will also degrade picture quality if the connection slows - many people don't notice, but colour blocks get simplified and resolution gets coarser. Also, there seems to be a general assumption that tv file sizes must be vastly bigger because it's video, but actually they're not that much bigger than music, if at all.

In CD-quality music streaming we want it to start almost instantly, we accept no degrading of quality and immediately notice even the tiniest pause in playback. It's an incredibly unforgiving scenario, and the stress on the end-to-end delivery from the server to your streamer is huge.

I have a few medium-term concerns about the market . . . that ADSL will rarely be consistent enough in the real world, that this generation of streamers can't buffer well enough and that the supply-side is not yet developed enough. But time will likely resolve all of these. Our household has recently moved to Virgin's fibre optic broadband and there's a significant improvement in, well, everything; the next generation of streamers are going to be designed to handle cloud services in a way the current one isn't, and we can already see some signs of the supply-side maturing, although it's very early days.

I'm now at risk of veering into my day job . . . but if you research how innovations are adopted, there are definite patterns that are fairly consistent (read, for example, Crossing the Chasm by Geoffrey Moore). Generally, the very first people to adopt are 'innovators' who accept (or even enjoy) the limitations and glitches of a new technology because they love new things and want to be first. But with regards to this forum, I suspect there are some people who find themselves in the 'innovator' position on hi res music, but are perhaps not natural neophytes, ie they wouldn't generally be among the first to adopt a new technology. In effect, they're more naturally in the 'majority' group, who expect all the problems to be ironed out before they adopt. It's a classical business dilemma: how can a company like Tidal skip the normal stages of maturing to meet the needs of a market that isn't used to being at the vanguard? It'll be fascinating to see how this plays out. My prediction (and you can read Fast Second by Geroski and Markides for the rationale) is that Spotify and others will let the minnows solve all the problems and then jump in themselves.

And the more that use the service, the more challenging it is at the supplier end to meet the conditions as described by Solid Air. So as popularity increases there is a risk of degrading of service (And maybe the complaints about download performance on the Qobuz receivership thread might relate to a last-minute rush overloading their capability).

Simply another reason for my personal dis-inclination to use online streaming for anything other than sampling of music I want to know if I like.

@Innocent - yes, me too at present. I'm not inclined to rely on any one service yet for more than 'extra' listening beyond my own albums, either for music I may want to buy or that other people in the household enjoy - the latter being a huge downside for me as my wife likes Abba :-( and my elder son likes hardcore rap . . . although I do find myself getting to like some of that.

It's a curious feature of digital start-ups that, while their customers/revenues increase (or decrease) in a smooth curve, their costs don't. This is because each new subscriber adds a fixed, small revenue increase, but costs are for software development, licenses, staff hiring, buildings, servers and so on, which are generally expensive, inflexible and take time to implement. So there's an investment lumpiness and a lag which can create huge pressure just to meet service obligations, even though the fundamentals of the business are sound. If you invest too far ahead you hit cash flow problems and if you invest too far behind you face potential customer dissatisfaction. It's like trying to drive down a road when your only options are full acceleration or full braking. You can easily look at all such services (Netflix, Spotify, Beats, Qobuz etc) and see where they've bumped into the sides. Whether that was terminal or not depends on other factors, such as pocket depth, marketing skill and customer expectations, but none of them escape without some bruising here and there.

Solid Air - I suspect you will find that most streaming services are not on-premise deployments that your posting suggests (with the initial capital costs you outline) but are either using outsourced data centres or the cloud (like Amazon or Microsoft Azure) where their costs are subscription based and computing resources can also be scaled to user load. Funding mechanisms like leasing will also spread other capital costs.

These are not the fundamental costs that cause the problems for startups - look at the financial figures that Deezer published. These show that the biggest slice of the costs are payments to the records companies! 

Anyone know how the royalties on streaming compare to the royalties for radio droadcasting of music? It certainly shouldn't be more for streaming, and arguably should be less as the audience is smaller.  If it is disproportionately higher, then either it is true, as some have suggested, that the record companies are trying to strangle streaming, or, perhaps more likely, the record companies simply see it as a new growth phenomenon that they can milk - and, it seems by some of what has been sail, are milking it to death, oblivious to their own stupidity.

@Kevo  - yes, if they're wise they will steer as much cost towards 'variable' as possible, including using  services as as Azure and AWS, contractors, agile methodology, etc. And you're right about royalties, which are likely to be in line with usage, at least in general.

But there will still be considerable cost lumpiness due to thresholds, license costs, poor creditworthiness, marketing campaign costs and the simple fact that investment in new platforms and infrastructure comes ahead of realising the value of what they built. That even assumes no false starts and dead-ends, which are all too common.