Forum Replies Created
-
Posts
-
Actually… it’s wrong. It fails to consider the temporal domain, and that at frequencies closer to Nyquist it’s possible for many people to hear about 1/5th of the details that whatever sampling rate provides. That is to say there’s still about 5 times more detail that gets lost in comparison, at the extreme high frequency. I can dig up the tests again if anyone is interested, but I’m pretty sure Leif has seen them too.
The thing is, it’s a very minute difference and they had to create a specially designed headphone amp and custom modify top-end Grado headphones to make the test possible. And also it should be noted that 192kHz sampling rate is very close to the maximum capability of the ear to discern differences in the temporal domain.
That being said, when applied to FM you have two phenomena.
-1- there’s the L-R signal which is multiplied upwards so this temporal inaccuracy will be more prevalent than it would normally be with a broadband stereo signal of the same frequency response.
-2- 99.999% of receivers alone would throw any perception of temporal loss right out the window because they themselves are not high quality enough for it to make a difference, and this says nothing of the audio amps, speakers, headphones, etc… that most people use which are equally or even more crappy.
As Leif pointed out, bad filters in your ADC/DAC are of MUCH more concern, and I agree 100%. Filters are by far the #1 reason that crappy DACs sound like crap these days. Be concerned about two things… noise and signal transparency. You can’t go wrong when you have very low noise, and near perfectly reproduced signals. There’s no concern at all to be had with 192kHz PCM and for all intensive purposes you will have a near perfect reproduction of anything that’s input or output.
My suggestion through all of this is to go with a provider that allows you to stream in realtime, without going through any websites or popups or logins or advertisements or any other crap like that.
Bandwidth and dedicated boxes are cheap enough now where you can completely pay off your box with only 2-3 very very small hosting clients of your own, and then you can start making money from your box and still have way more than enough bandwidth for yourself.
I’ve seen Atom 330 dedis for about $30 EU a month, with unmetered 100 mb/s connections on em. A few friends of mine have them, it’s not hard to pay off that bill by sharing a little bit of that bandwidth. 8)
And also use this:
http://www.thesycon.de/eng/latency_check.shtml
and disable any device drivers that make that spike up. that can help a lot.Vorbis?
Why not use FLAC? There’s already a standardized way to wrap it in an Ogg container (aka OggFLAC) and even though it’s VBR I don’t see why it can’t fit into your fixed latency picture.
But I’m not saying don’t include Vorbis too. For lower bitrates (like 300-400 kbps) it beats layer-2, at least in audio quality, by a long shot. 🙂
Although v3.93.1 is no slouch either. In fact it’s my current encoder of choice for CBR or streaming of CBR.
Since the only time I ever save CBR now is for 320kbps DJ masters, I just use v3.98b anyways, or mainly -V 0 -q 0 😉 But yeah, for streaming I’m using v3.93.1.
With XP SP1 it’s not a bad idea to firewall the box from your own network too if you can, except for a few ports it needs like ARPA, PING, DNS, and your remote software. Otherwise you probably want something half decent to firewall it from your LAN, not just WAN.
Also see what other CPUs can run in your mainboard, you might be able to update to a Pentium D which could probably work half decent. The thing is even the Pentium D is past demand and return of positive revenue, not even being made anymore, so the price of those is actually going UP now. And for quite a while already. You may even consider a new mainboard with matching memory and CPU to go with it, which you can get something workable for only $250-350 USD with speed that annihilates your P4. 🙂
The only other solutions that will get you into the ballpark of Breakaway (but still not as clean) cost in the range of 2-3 grand minimum. To get into the range (still not as clean) as BBP, you’re talking over 10 grand. Partly because BBP is essentially a $10,000+ processor (it still beats processors in the 13-15 grand range), it just costs less. 🙂
So if you have BBP, you shouldn’t have to prep your tracks before hand at all. As long as the input levels are reasonable, depending on the preset. Many of the presets offer a huge amount of variability in the input levels they work great with – that is to say you can sometimes input a -30dBLKFS or -10dBLKFS level signal and get the same exact sound from both. Just depends on the preset. 🙂 Some are not meant to be as aggressive in the long term.
Mostly because of all the other stuff that gets slapped on and plugged into the mainboard, not even to mention the quality of the power supply being used.
Devices that use their own powering from a wallwart, preferably a high frequency switching design (very few converters do that, but more are starting to) the cleaner the power source is. Getting the electronics outside of the massive RFI of the mainboard itself is most of the battle though.
Another decent onboard audio brand that you don’t see too much of anymore is C-Media.
http://www.cmedia.com.tw/
The ASUS Xonar D2 has a C-Media chipset for instance.But as Leif said… fine for a portable demo, but wouldn’t run a station with onboard audio. 🙂 It is inherently filthy business.
After re-reading your post, I noticed that it wasn’t that it didn’t come up, but that you couldn’t type into it. Actually… you’re totally right. Now that is odd. (pun intended)
Leif, I’m confirming this bug.
Let’s see…
Latest version of Breakaway
(don’t worry, just because it’s not quite public yet, NOTHING has changed in it with regard to Winamp plugins, I probably don’t have to ask Leif cos he would probably tell me about a mistake somewhere… aaaaaand… I’ve seen his Winamp loading code, and it’s a FULL implementation. Fake Winamp window for track titles and all.)
Ok, so let’s install EdCast into the Breakaway folder. Done. It installs:
.
.Plugins
And we’re going to add this optional file to:
.Plugins
Ok, so now that that’s settled… Enable Encoders in Breakaway, add the dsp_edcast.dll to the list, make sure it’s Enabled, click OK, click Reload, and:
Let’s check the about to see what version it is:
It’s the latest version, ok. Let’s try to configure the encoder we just added:
Well, it appears to be working fine.
Try exactly what I just did above, and see if you have the same results. 8)
[quote author=”TDCat”]Frequency Cutoff: 15kHz (may not be needed with 256kbps really)[/quote]
It should never be needed, or the encoder is a total piece of junk. Encoders have low-pass filtering built into them, which is usually automatically set at the ideal frequency for the rest of the settings you are using.I’ll tune in a bit later, glad to see you are re-purposing your old hardware. 8)
There certainly could be a mobile product from Leif at some point down the road, as the power of those machines increases. But I’m sure it won’t be anything more advanced than Breakaway Audio Enhancer version, if even that powerful. Because with increased power, comes increased requirement for battery, running the battery down a lot faster even if there is the power there.
@ Porting it… it’s a matter of maintaining everything, as Leif said. Leif could probably do a faster and better job of porting it than anyone, partly because he made it… and make it easier to maintain the codebase too. But as he said… barely even enough time to deal with it right now with just Windows. Claesson Edwards Audio is not the only thing he has on his plate right now.
Technically I’m not sure if it’s even possible, but it might be. wxWidgets (among others) makes it pretty easy to port from MFC and GDI, which would be the main concern for Leif.
Also, I should note that OSX is, and has been for almost 10 years now, on the way OUT for video and audio production as a whole. The film audio production industry is really the only of the last people that are still getting ripped off a bit by Digidesign. Everyone else has started moving on on droves to cheaper, better (software & hardware & support), and faster workflows…
The stability of Linux for some stuff is over-rated too… Especially if you have your device on the network and you’ll need to be updating its kernel automatically. And the stability of Windows is highly under-rated by most people as well. Just because they are great at ruining the stability of THEIR windows machines doesn’t mean that Windows is inherently unstable. I personally managed over 200 machines at my last "day job" and had them up for a year at a time for almost 5 years running, only rebooting once a year because of XP having a bug where it won’t make sound after 365 days of uptime.
[quote author=”X-Hosted”]Dont think that audiostocker does a lot, the built in processing (AGC) and clipper in SAM Broadcaster does most of the magic 😉
However, i do expect BW Broadcast to sound better.[/quote]
It’s not even a comparison, BBP annihilates it.
And re: the AGC in SAM4’s audio path… it can be disabled independently of the 5-band processor. That happens to be what could probably be called SoundSolution v1.9 😉
