Log from #neuros IRC channel

nerochiaro: sry, thought you was afk. My question is what I should do with the NDA?

I just put my name there and email it back?

Darchon: fill in all the blanks, sign it and email it back. i would do that

<crweb>

and read it 2-3 times

[08:15]

may1937: you there ?

[08:19]

how do I actually sign it?

[08:19]

Darchon: you print it, then sign it

[08:20]

Darchon: then snail mail it back (maybe scan it and email the scan if you're in a hurry. or fax)

[08:20]

I can fax it

[08:21]

where can i find the fax number?

[08:22]

<crweb>

wow.. i fell asleep at 03:00 and woke up like 10 min ago

[08:22]

Darchon: good question, i have no idea

[08:22]

<crweb>

thats almost 12 hours

[08:23]

crweb: man, that's sleep finally catching up on you

[08:23]

just means u needed it

[08:23]

Darchon: email Joe or support@neurosaudio.com and they will tell you i think

[08:25]

nerochiaro: ok, will do

[08:27]

crweb: do you know what the equivalent of cat /proc/bus/usb/devices on the OSD ?

[08:27]

<crweb>

nerochiaro: if you mount sysfs, its /sys/bus/

[08:27]

it's mounted by default

[08:28]

<crweb>

should be /sys/bus/usb then

[08:28]

<crweb>

its a dir structure

[08:28]

yep, not as good as the one in proc/bus/usb/devices but it will do

[08:29]

thanks

[08:29]

<crweb>

not as good.. heh

[08:38]

hi all

hey turran

nerochiaro: I don't think I can pm cause I don't have a registered nick

gl2tosl2: ah ok. well, just tell me. i keep forgetting :)

nerochiaro: hey =), can you give me your email? and JoeBorn's if you have

[08:40]

my real name is leif

[08:41]

turran: nerochiaro at neuros dot us and jborn at neurosaudio dot com

gl2tosl2: ok, thanks

hi all

morning

JoeBorn, hi

nerochiaro, thanks =)

[08:46]

hey JoeBorn

[08:46]

JoeBorn: there's an email in ML that might require some of your PR skillz ;)

[08:47]

morning JoeBorn

[08:47]

JoeBorn: to what number can I fax the NDA?

[08:48]

oh boy!

[08:48]

* JoeBorn fears that Steve Jobs may have declared a jihad against Neuros

?

Darchon, try 707-220-2263

JoeBorn: send me some AK47 and i can help defend

JoeBorn: tnx. US country code is 81, right?

no, just 1

I guess we invented the modern phone system :)

lol

ok

[08:52]

JoeBorn: there's an italian dead inventor that is rolling in his grave right now, though

[08:54]

<crweb>

nerochiaro: they teach us Alexander bell invented the telephone

[08:54]

<crweb>

and at the time he was a US citizen

[08:55]

<crweb>

and he was scottish

[08:55]

* crweb back to paying attention to professor..

crweb: they at least have the decency to teach us that there's controversy on who invented it. here's some more details: en.wikipedia.org

crweb: not that i personally care much, it's just to note it's not that clear cut

<crweb>

we all know the US invented even the wheel...

and hot water

[08:59]

<crweb>

nerochiaro: eh, thats how things are done in the US patent system

[08:59]

<test>

test

[08:59]

<crweb>

nerochiaro: look at microsoft

[09:00]

<crweb>

founded by theives

[09:00]

crweb: in all patent offices in the world the situation isn't much better

[09:00]

crweb: at least apperently

[09:28]

JoeBornio

[09:36]

<crweb>

wow, benchmark wise, wavpack sucks compaired to the competition

[09:36]

crweb: decoding or encoding ?

<crweb>

both

<crweb>

almost 50% slower than flac

what settings are you using on eacg?

<crweb>

daurnimator: settings don't matter for playback

its accepted to be at least 5% lower in resources than flac

crweb: sure they do - what is the quality of the input files

<crweb>

decoding is an algorithm

are the input files highly compressed, merely stored, or somewhere inbetween?

so decoding time is a function of how much compression you apply '

?

yeah

<crweb>

its not

is for everything AFAIK

even zip/rar

<crweb>

encoding time, is a function of how much compression

[09:40]

if the decoding algo is the same, i don't see how it would matter on decoding

[09:40]

i know for a fact that settings you encode wavpack at affect the decode speed of it (as well as encode speed)

[09:40]

eg, -hh doubles encode AND decode resources needed

<crweb>

yeah, thats cause wavpack sucks. ;)

it probably means that it will decode using a different algorithm

<crweb>

encoding has predictions, file scanning, etc

<crweb>

you don't have to do predictions and so much read ahead for decoding

<crweb>

you just take the block, and decompress it

[09:42]

<crweb>

there are several instances where wavpack decoding is slower than encoding however

[09:42]

<crweb>

so, something strange there

[09:43]

see: wiki.hydrogenaudio.org

[09:43]

<crweb>

flac, apple, and monkey are pretty much constant no matter what quality

<crweb>

his comparisions are very "general"

look at uclc.info then :P

<crweb>

i just mean he says "very fast"

<crweb>

comparied to what standard?

also, www.bobulous.org.uk

<crweb>

i run very fast compared to my 3 year old cousin

[09:46]

compared to other lossless codecs - baduh

[09:46]

baduh ?

[09:47]

nerochiaro: sound you make when something should have been obvious

[09:47]

daurnimator: you have a weird language down under ;)

see en.wikipedia.org

<crweb>

but its really not obious.

its often prefixed with "ba" :P

<crweb>

if you compare them to 1 slow one, they are all fast

[09:49]

crweb: if flac is "fast" and wavpack is "very fast" - surely you can conclude that wavpack is faster than flac.....

[09:49]

<crweb>

maybe

[09:49]

<crweb>

but who says they were directly compared to eachother?

thats what the table will be telling you

otherwise why would you have a table

<crweb>

unless flac was compared to monkey, and wavpack comparied to standard wav

<crweb>

he doesn't really say

i'd guess that the time to encode for each on a similar setting (from pcm data) was timed

<crweb>

but he doesn't say.

<crweb>

anyone can make any table, say anything

and things with low results were ranked "high"

crweb: hydrogen audio is one of the main places for audio discussion on the web

<crweb>

so?

their wiki should be taken as gospel

<crweb>

so you are saying the author doesn't have a perticular format he likes better?

i kinda trust them too, but they usually provide very scientifc and detailed data when they do benchmarks

there is data there to make your own decsion!!!

<crweb>

data != very fast, fast, slow

[09:53]

data in the table

[09:53]

sorry to interrupt this fistfight. anyone experienced with wireless, iwconfig and stuff ?

[09:54]

<crweb>

i use it

[09:54]

nerochiaro: kinda

[09:55]

i'm trying to setup this wifi thing on OSD. so far so good however when i try to set essid with iwconfig it says " SET failed on device rausb0 ; Network is down"

[09:55]

<crweb>

the name of the device is rausb0?

ya

sure - ralink usb device

i just don't uderstand the network is down thing

iwconfig rausb0 essid network_name ?

it's down, i know it. i'm trying to set it up before bringing it up

<crweb>

its just telling you that it didn't come up

JoshMalone: aye

Oh - ralink controllers...

...need to be "UP" before configuring

ifconfig rausb0 up; iwconfig ...

JoshMalone: so iwconfig up before anything ?

<crweb>

iwconfig rausb0 essid whatever

ifconfig rausb0 up

<crweb>

then setup network interface

<crweb>

really?

I hafta do that for my laptop

and then ifup the thing

<crweb>

JoshMalone: what distro?

ubuntu

<crweb>

strange..

but the built-in scripts work fine

<crweb>

i know sometimes i have to have the network settings setup

JoshMalone: i did ifconfig up it, ifconfig shows it then. but still same thing on iwconfig

hmm....

<crweb>

nerochiaro: its just a notification to tell you that because it didn't connect, the network is down

still says set didn't work

crweb: i don't think so. the "SET failed" isn't normal

crweb: doesn't do that on my PC

yeah - that's a driver BUG for sure

<crweb>

nerochiaro: right, the SET failed, is the error

<crweb>

nerochiaro: it won't let you set essid

JoshMalone: works on my PC, same driver

<crweb>

nerochiaro: so it couldn't connect, so network is down

right

driver doesnt work on ARM then

endianness issues or timing

<crweb>

endianness is the same

JoshMalone: i need to fiddle with the build, i saw some arm options. gotta check

???same btwn ARM and PC?

<crweb>

arm9 can be either

but linux uses LE

<crweb>

ARM is bi-endian

but it's determined by the OS and can't be both @ same time

<crweb>

so is mips, ia64, some powerpc, and Alpha

<crweb>

JoshMalone: so if Linux uses LE, why can't linux on arm use LE?

wait - maybe Ive got it wrong LE vs BE, but it's normally opposite of PC

whichever it is....

<crweb>

JoshMalone: not with arm.

<crweb>

arm can be either, whatever you set it to

[10:03]

<crweb>

i don't know which one the osd does.

[10:03]

<crweb>

en.wikipedia.org

[10:04]

i know it's not network byte order

[10:05]

<crweb>

JoshMalone: the osd is LE

[10:05]

well - I remember having to deal with this on the Xscales....

[10:05]

but anyway - the driver doesn't work on ARM apparently; and I know that's not unusual

[10:06]

* JoshMalone goes back to hacking bash scripts

[10:07]

<crweb>

JoshMalone: xscale is bi-endian too. Who ever's you were using must have been booting it into BE

[10:07]

JoshMalone: well, i'll figure it out, or ask the folks who wrote the driver. thanks

[10:07]

<crweb>

nerochiaro: hasn't this driver worked before?

[10:08]

crweb: it was another driver, for the older version of the same chipset. the hardware i got has a newer chipset version, so i'm fiddling with the new driver

[10:09]

<crweb>

what card is it?

[10:09]

<crweb>

is it based on rt2400 or rt2500?

[10:09]

rt73

[10:10]

it cross compiles cleanly. just has these errors

[10:10]

<crweb>

did you build wireless support into the kernel?

aye, or it would fail in insmod

<crweb>

802.11 stack and such?

i added the wireless extension

CONFIG_NET_RADIO i think

<crweb>

you need the firmware loaded

aw, true. let's see how to do that

<crweb>

you'll find a copy of the firmware in /lib/firmware/<kernel>/rt73.bin

<crweb>

on your ubuntu

[10:13]

i already have the one that comes with the driver, i trust that better

[10:13]

but i'm not sure how to load it

[10:13]

<crweb>

yeah.. i've got no idea

<crweb>

only exp i had was, i had to copy my firmware to /lib/firmware once

i'll find it, don't worry. however i'm not sure why on PC i didn't have to do it

<crweb>

that should fixe the SET error though