Generally, yes. However, data over mains would also be fine, and there are a number of products that do ethernet protocols over mains.
Generally, yes. However, data over mains would also be fine, and there are a number of products that do ethernet protocols over mains.
*minimum*
Well, it's for home audio systems *starting* at around $30k
I read in sci.electronics.design that Dirk Bruere at Neopax wrote (in ) about 'Ethernet in analog out', on Tue, 11 Oct 2005:
For those, you only need 8 bits. It's the price that matters, not the S/N ratio.
Why do you care that the stream is reliable? However, if you do, why can't you buffer a few seconds? 6.144 Mbps is pretty slow when compared against
802.11g. Setting up QoS parameters on the WLAN can ensure that you don't get blocked when downloading your p*rn... ;)
I think you'll find that reproducing bass down to the required boneshaking 20Hz will be quite expensive just on its own. And multiple 1kW amps are not cheap either, nor the h/w and s/w to tailor the sound to the (large) rooms.
I was meaning stock network-audio products.
Of course the chips are available, I was just pointing out that 16 bit vs 24 can only be noticed in very optimal listening conditions, and
96Khz is just silly.Standard 64 bit words? Nowhere in 802.11* or tcp/ip uses 64 bit words, it's all octet based. You'd want packets of about a K or so, as small packets add overhead. As to timing, I'd probably do it with TCP/IP and a broadcast ping for sync.
I would be tempted to not bother about direct sampling, and go with 320Kbps MP3.
This can be done even on 802.11b.
It will break down unless all the clients on the net know about bandwidth. Doing unthrottled file transfers over such a net will break it.
*minimum*
Yeah. But who wants a system that can actually deafen you when the cat treads on the volume control.
On Tue, 11 Oct 2005 13:19:57 -0700, Bob Monsen wrote: ...
Apparently this guy never borrowed any money. ;-)
Cheers! Rich
this looks roubust, only 16 bit though
martin
Hello Martin,
Thanks! I guess that's it, I just didn't want to fire up the iron and roll my own. But if it has to be done then it has to be done. Maybe someday a product with these shows up at Walmart?
Regards, Joerg
Hello Kevin,
Interesting, I didn't know that one. I'd prefer to buy local though, been burned too many times by bad merchandise, recycled customer returns or what not. So I'll keep an eye out at the Walmarts and Targets.
Philips is kind of skimpy on the specs. Didn't say whether it would play any audio broadcasts from the web which is what I'd need it for.
It is a cool looking unit and that blue display would perfectly match our stereo. Which BTW is a Rio unit that can play MP3 (if we had any, that is...). I never understood why they didn't include wireless LAN instead. They probably would still be in biz if they had.
Regards, Joerg
It depends on how much data gets lost. If you are firing off 2k packets, a lost packet is 2.6ms. Covering up for this kind of drop can be done in software without the listener noticing. Also, if you buffer a bit of data, losses can be retransmitted. Some of the high bandwidth media protocols I've heard about use datagrams rather than a reliable TCP stream.
How does that factor in?
If they are doing playback of recorded media, one can simply buffer enough so the receiver can always recover lost data in time. TCP isn't the best protocol for this, due to its crappy handling of congestion, but others exist, as mentioned above. On a dedicated WLAN, it won't be an issue.
There are fairly cheap products that broadcast video over WLAN (not high quality HDTV, of course).
On a 56M wireless link, a 100ms burst can deliver almost a second of media requiring 6.144Mbps. Now, granted you don't get the full 54Mbps from
802.11g, but it should be able to handle this kind of thing nicely, with a bit left over for other uses, like control.However, using a cat5 gigE link, all of this is moot.
Try XPORT device, it self contained Ethernet port (with build in Web and DHCP !) that transulate TCP/IP packet into serial data output (to PIC RS232) and there is spare IO. I tried that and it worked very well. You have to watchout that RS232 data stream has no intergrity check, where PIC RS232 may goes out of sync.
It should be OK on wired ethernet, although I'm not aware of anyone doing hugely oversampled audio data.
It's quite interesting to see just how much bandwidth is needed just to shove 16 bit linear pcm uncompressed audio @48kHz.sampling over a wireless link ( 4Mbps for the Nordic chipset )
Graham
If it isn't the music will have breaks in it !
Latency is a big issue for real-time audio too.
I'd expect a few seconds delay to be unacceptable to the customer. 100ms is probably OK though.
Graham
Analog FM is limited to 15kHz for one thing.
You would need a special license too.
Graham
Hello Kevin,
That's what I thought when I read about the 40 channels, but upgradeable. Can't use it at all then.
That would be the old trick, trying to push a cash cow. Won't work in the marketplace. At least not at my house ;-)
If that's what they are planning then it will go the way of the Kerbango radio.
That's what I want it for. I'd be willing to let the PC run. I just need something that sends audio from there to the stereo, too far of a stretch for LPFM at legal power levels. Maybe some day I'll just whip up an AM house transmitter. Found an old crystal on an AM frequency that's vacant so there is a start.
A cool thing would be if some little SW in the PC could send the audio back onto WLAN and then something at the stereo could receive it. Who knows, maybe one could even fool this little Philips device by spoofing a URL it likes.
Regards, Joerg
The Philips stuff uses compression. Walkman quality.
Graham
Not really. I want uncompressed wideband audio to get from digital source to power amp using the minimum number of wires. Mains must go to them and so I checked out data over mains, wireless etc and most products with the bandwidth I want provide ethernet functionality.
If scientific reasoning were limited to the logical processes of arithmetic, we should not get very far in our understanding of the physical world. One might as well attempt to grasp the game of poker entirely by the use of the mathematics of probability.
martin
Have something to add? Share your thoughts — no account required.
Ask the community — no account required