Which Linux Distro Are you? - Page 2

http://www.chromium.org/

cheers ...

Here's the demo

http://www.youtube.com/watch?v=w4tyG4upFns

cheers ...

I know I've probably said this. But I right on the verge of giving windows the boot (the dual boot). Vista has been one headache after another. There's only a few features I like about windows 7 it's mainly the features for auto adjusting windows dimensions (like combine win button/w/arrow keys).

One thing I'm looking to do is combine 2 audio inputs into a single virtual audio input. That is take a mic and line in and have a virtual input device that feeds from both them and you can select IT as an input for recording. There's some software for windows (cost money) but works. I haven't a clue in linux. But I am aware that you can do lots of networking mods i.e. a single NIC can be attached multiple virtual adapters (eth0:0 eth0:1) for virtual hosts, packet redundancy etc...

I'm thinking about dual booting with UBuntu just for simple start.

Wait i'm confused I thought Android and Chromium OS were the same. I guess Chromium OS is what I'm interested in now haha. Experiences? I'm skeptical but love Google

Android and Chromium OS are not the same, but I would not be surprised to see both converging sometimes down the road.

I haven't started on Chromium OS - just reading at the moment.

cheers ...

You might need to be more specific, but I suspect the basics of what you are looking for can be found here...

http://www.alsa-project.org/main/ind...hannel_devices

NOTE: You really shouldn't do this for recording purposes unless the two devices are clock sync'd.

If instead you are looking to have it always mix down those sources for recording and these are both on the same device, that is easily done with Jack into Ardour, just have Ardour take both inputs for the input to a track.

Seablade

You do realize you are entering my world now right?

Yeah I understand this.

I'm working on dual boot currently. I use Audacity for windows. But really I'd like to pipe a "mix down" into a virtual device. See, the software I'm writing is very much like a typical audio editor and I've already written code for mixing 44.1 audio (sample addition fed into an audio ring buffer) - I could talk a lot about this. I'm able to stream record directly to the site, but Flash has a limit on the number of inputs you can select at a time.

[quote]You do realize you are entering my world now right?/QUOTE]

haha recording? I plan to do a lot of it. Hopefully higher quality and better lighting then my most recent vid. But also to my site.

http://www.youtube.com/watch?v=TaMJfO7oGyk

I know half the stuff I'm doing can be accomplished a number of different ways. I thought maybe hardware mixers - which would work - but then you have to depend on the hardware for affects and stuff. I had assumed S/PDIF would do a lot of good stuff but had little success with it. Seems reasonable you should be able to hook in a number of devices (me only two), add affects , possibly control them with some keyboard hotkeys, possibly mix them and pipe the audio elsewhere in realtime. Maybe I'm asking for too much and I have to buy those expensive software suites and the knobular sound cards which I can't afford.

Hmm ok well the virtual audio device shouldn't be a problem as I mentioned above.

How many channels can flash handle at once?

Ideally you could use an interface with a large I/O count to pipe things into and out of the computer, both for effects and for your mixdown, but it seems like the primary problem is that you don't have the interface you want really as long as Flash can handle decent I/O for multitracking.

I assume the flash is because you are wanting to record to a website? Might want to take a look at aviary and see what they can do as something worth looking towards, I think they are using Flash, though this does have me wondering if HTML5 would make any difference for this, but I suspect the AUDIO element is likely only for delivery from the website to the computer, not the other way around.

Seablade

You got it. I know all about Aviary, awesome development! I'm wanting to take my site in a different direction though.

Currently flash only has access to one device at a time. And recording via RMTP is limited to only 2 channels at 44.1. It's the only and most versatile option to stream directly to a website. Currently there is no access to the raw audio before it's sent to the server to be recorded (or subscribed too) else one might be able to write their own multi-channel streaming protocol/server. Adobe's API is pretty limiting in this area thus far.

But that's OK though, I can make due until Adobe expands their functionally or if/when HTML5 comes out with advances on the audio element, I will re-write everything.

Asoundrc looks promising. I just got to get a linux environment setup before I can really digg into it. Seems like it might be able to take advantage of S/PDIF; isn't that meant solidly for "large I/O" or am I mistaken?

I use eLive for my main Linux machine, however I have an old pIII Latitude laptop that I got Zenwalk to run on just to see if I could.

You are mistaken. S/PDIF only carries 2 channels of uncompressed audio, it can carry compressed audio (AC3) if it can be encoded and decoded. You might be thinking of ADAT, which can carry up to 8 channels at 48k, but you would still be limited by flash as you have described it.

Well on Linux (And conceivably on OS X or Windows as well, though not currently ported) there is a development to allow Jack and Flash to communicate in development at the moment, not publicly available right now, but drop me a note via PM so I remember to get back to you on this. I need to check with the author as to its capabilities, I know a few people that are using it right now as testers.

EDIT: Just checked with the author, no go on that one sorry.

Hmm that is a big if at this point. I only really know of the effort intended for delivery to the browser, not the other way around, I would have to look into it to see if delivery to the server is being touched on at all.

Seablade

EDIT 2: Come to think of it, I wonder if you might be able to bypass flash for this and utilize netjack instead. Would be interesting to play with at least albiet more difficult to program, and harder to support on all platforms.

hmmm, you'd think an optical connection should be more than capable of having multi-channel support. They need to revamp their protocol. ADAT looks like a dated technology and seems costly.

This sound like magic if it worked Jack looks awesome! Very interesting!

Yeah, maybe one day. Someday.

Yeah I don't want anything you have to download. Again, HTM5 would be ideal if it ever works. I have seen interest in adding elements for streaming cameras/mics all is needed is access to raw audio (and with javascript's multi-threaded support - O BOY!!!). Google gears has the API for low-level audio support, and I have considered that. Google also has said they will make tools for transitioning any gears code to HTML5 and they have a big role in the HTML5 initiative, so that's my hope

update:
So I'm setting up virtual servers. I have installed Ubuntu already. And will will be installing Chromium OS. Virtualbox seems to be a great way to get multiple OS(es) on a computer without repartitioning and dual booting - native too.

Hmm don't confuse the hardware with the protocol... Hardware would be FiberOptics Cable and Connections, but there are many different protocols that can be used over it. Over TOSLINK connections which are a specific connector and cheap fiber you only get S/PDIF(2 Channels Uncompressed, also exists over consumer co-ax connections with RCA) and ADAT(8 Channels Uncompressed). However over SC style connectors and good fiber you move into an entirely different world of which the most common protocol to use for digital audio is MADI(Also a spec exists for this over co-ax), which can do MUCH more, namely at 24/48 it can transmit 64 Channels, at 24/96 32 channels.

Well given the work is open source, you may be able to build upon it to handle input as well.

Random somewhat OT comment... EC2 micro instances rock

Seablade

Shaun-

http://mozillalabs.com/rainbow/2010/...-meet-rainbow/

May be worth keeping an eye on for your work as well. Only Firefox and only nightlies on OS X right now, but it may give you another path to take in the future.

Seablade

Funny you linked me. I didn't know about the rainbow add-on but I've been reading all about the audio synthesis stuff. I installed Firefox 4 beta and checked out most of their audio experiments.
https://wiki.mozilla.org/Audio_Data_API

A lot of simple synthesis examples had bad pops and clicks AND they were multi-threaded too, which was a little discouraging.

Rainbow: being able to record audio locally will rid the problem I have of determining the track start offset times accurately. Plus the container work is done for you, looks too good to be true!

I've already begun moving toward javascript (whenever possible) over flash and eventually I plan to full-on adopt HTML5. Thanks for the link, too bad I don't have a Mac.

I've been using Ubuntu too but unfortunately Adobe discontinued support for Flex3 and up for Linux. So migrating for good is out of the question. I was really hoping to give windows the boot. Still it's useful though. I'm using it to map a network drive which is then piped to a samba share which is mapped to a windows drive - haha - which I had an Amazon S3 account.

Debian squeeze, Ubuntu 10.10 on hdd, on VBox Debian testing, squeeze, LinuxMint, Fedora, Ubuntu & trying to get slack 13.1 configured.

Back at the beginning of time I tried out slackware but didn't "get it", then around '06 I ran debian unstable for a little while(had hardware issues with the stable release and testing, namely ATI) and finally ended up with Kubuntu Dapper Drake on my laptop and Xubuntu on my server machine, this ran quite happily for a while but then the aching for games and such took over and well... windoze won, except on the server which ran happily for three years before it was retired. Last year I got sick and tired of Microsoft bloatware and decided upon linux again, since then I've tried Debian Lenny(a few minor headaches with ATI again), Linux Mint Julia(based off ubuntu Maverick Meerkat) which had another share of ATI trouble(ATI X1800 has no binary driver support in X.org versions newer than 7.3 or 7.4 and kernel newer than 2.26-something), I ended up swapping the X1800 MXM in my laptop for a GeForce 7600go instead and right now LMDE(Linux Mint Debian Edition) based off Debian testing(Wheezy) is flying high on my XI1546 and life's good... I must be a Debian fanboy then and often find myself bringing up the terminal and issuing "sudo apt-get install packagename" rather than using the graphical tools just because it's faster

Linux Mint is a quite good alternative for people who want linux to run with all multimedia support and restricted modules and such right out of the box rather than having to add the restricted repositories and installing the aforementioned stuff themselves, LMDE is a rolling distro that never will become outdated since it's constantly being upgraded, me like

I like OpenSUSE because of yast-curses, but Fedora seems really solid. I think my next built will have fedora core 14 (maybe 15?) with LXDE desktop.