As it turns out, either I have been very unlucky or the X3 live is actually pretty bad.

At my first band rehearsal with it, I couldn't press the Stomp or Mod buttons without it changing patches. That is very very bad because those switches are only meant to turn on or off an effect in the chain, but changing patches has a small silent delay (unusable during a song) and the next patch may very well be too loud or quiet, and certainly not what you want.

So the good people at Thomann were kind enough to take it back and send me a new one.

I just tried it today, and it has its own problems. First the display has one full row and one full column of dead pixels. That would be annoying on a notebook, but on such a tiny display (what, 320x240) that is quite a lot of missing information. But on top of that the sound just cut out randomly, and would come back when switching patches.

I am really unimpressed with Line 6 now. They have obviously rushed their product in time for the christmas rush without appropriate testing.

They really shot themselves in the foot.

People like me who are having a bad time with the first generation of the product are unlikely to trust them for giggable gear, so they're losing a share of the market.

Many of the defective units are likely to be in stock in a number of stores, so unless they do a product recall any customer has a chance of getting a faulty product.

Previous satisfied Line 6 may very well become cautious on the back of the rumour that Line 6 have just "lost it", or are going downhill.

I'm not sure what to do next but I am inclined to ask for my money back and buy either a Boss GT-8 or simply get back to my Pod Pro+FCB1010 setup (which works, after all, just fine.)

I just got my new Line 6 Pod X3 Live yesterday. There are few user reviews so far so I thought I'd make one.

First the unit is physically impressive; it looks very well-built and sturdy, more so than my Pod Pro. It is also quite heavy (I would estimate about 15 pounds, or 7kg). I was surprised that the power transformer is not built-in, but it's not a wall-wart either. It's one of those transformers with leads either side. Not too terribly inconvenient, but a standard IEC socket on the back of the machine would have been nice (as on my Behringer FCB1010.)

I plugged the X3 in the power amp in of my Fender "The Twin" and just played throught the presets.

Wow! It's a lot of fun. Some nice sounds, some very nice AND crazy sounds (now, that, I like.) The interface had me baffled a little at the beginning, but a quick glance at the manual goes a long way. The expression pedal is excellent and feels right, better than my FCB1010, or my wahs (CryBaby and PW-10).

Switching between patches seems instant (a very good thing.)

The display is a little small for stage use, but the is a "Big User" (uh?) mode, which only displays the patch number in big letters on the display. Quite usable, I'd say, but time will tell.

Right, gotta go and play some more now!

In my quest for simple, robust and easy physical interaction computing for dance and music, I have now decided to try using a BTarduino with the ADXL311 2-axis accelerometer from Analog Devices. The ADXL311 is in the same family as the ADXL330 used in the Wii Remote (apparently).

So, first things first. I ordered a BT arduino from PCB Europe about a month ago. A few days ago I ordered two units of the ADXL311 at around 9 quid each from RS electronics.

Now, let's try to get them to work together.

The Arduino site says that the BT arduino can be powered between 1.2V and 5.5V. First attempt, with 2x AA is working fine. And that is with the really rubbish  AA batteries I bought in a pound shop for £1. Success.

That should also be enough juice to power the ADXL311. Apparently it needs between 2.7V and 5.25V according to the spec sheet from AD, and it consumes 400microamps. Cool, that ain't gonna drain my batteries in ten minutes like the Wi-MiniDig.

Docs say I must also connect a bias resistor between 50k Ohm and 2M Ohm. I'll get to that later.

Ok, now how do I get the BT arduino to talk to Windows XP? First I "added new device" in the windows Bluetooth control panel. It showed up as "BTarduino" and the default passkey "12345".

Note to self: change the name and passkey later, I intend to you use several BTarduinos.

Then downloaded version 0009 of the arduino software. It runs.

First thing to do: go under tools and select the correct microcontroller, the atmega168.

Ok, now what? Let's read this official BT arduino tutorial.

Ok, there's a bit I don't understand. This says "the ATmega168 comes preloaded with a bootloader that allows you to upload sketches to the board via bluetooth.". Does it mean I don't have to upload a bootloader to my arduino?

Ok, if that's true, I should be able to just upload the code to blink an LED. Let's try that.

It looks like there's some code on the arduino already that does blink the LED. Will have to try something else.

In any case, I need to select the port number in the arduino IDE. Hmm, three COM ports come up: 27, 29 and 30. From the bluetooth properties, I find that BTarduino is number 29. I suspect this is that same problem I've had in the past. It seems that whenever I add a bluetooth device it creates several port number for each device (especially mobile phones.) High numbers seem to be problematic, so I've often had to resort to deleting most port entries and redoing them. Note to Microsoft: change this. It sucks big time. Mac OS X seems to have this sussed out, with name BT ports.

Anyway: to try to solve things, changed the port number to COM2, and changed the speed to 115600.

Ok, next step: upload the Firmata sketch so that I can read from within PD. I get this error:

"avrdude: stk500_getsync(): not in sync: resp=0x00
avrdude: stk500_disable(): protocol error, expect=0x14, resp=0x51"

I'm assuming the software sees the connection but can't send for some reason. Let's investigate.

UPDATE 10.NOV.2007
-------------------------------

I discovered that you have to change the transmission speed in the Firmata script too. I had never really looked at it in detail. But after changing it in the script, it worked.

I never got the ADXL311's to work, however. The connectors (what do you call those, anyway) are too small to solder, and I have a feeling that the heat from the soldering iron fried them. Poo. Next I intend to try the SparkFun Lilypad accelerometers, if only because you don't need a microscope and coffee withdrawal to connect them. But also because they look cool.

I've got this HP dv8263ea laptop, and I've set it up as dual boot XP and Debian machine.

I've installed Debian Etch (kernel 2.6.18) and many things work. However, there are a few things that are still not working right, so I'll keep track of my changes and discoveries here.

First of all, it was quite a pain to set up the wireless card, an Intel 3945abg. I had to install the intel proprietary drivers and regulatory daemon. It now works. Except two main things.

1) he driver/daemon/whatever should load at boot-up; it doesn't. I have to start it manually with the 'ipw3945' command. No big deal, though. Then I have to connect using wlassistant, which does see my wi-fi network. Let's call this issue solved.

2) I have an internal intel sound card, and an cardbus Hammerfall HDSP (which is a great multi-channel card.) I can't recall exactly what I had to do to install the Hammerfall but a lot of it installed with the default installation. However there are a few things that don't work:

- somehow the alsa sequencer is not installed or not configured properly. For instance, the 'aplaymidi' command returns:
    "ALSA lib seq_hw.c:457:(snd_seq_hw_open) open /dev/snd/seq failed: No such file or directory
     Cannot open sequencer - No such file or directory".

Why is that? I don't quite understand what to do to fix that, and the ALSA help is, well, not helpful.
 
2b) I can live without midi, but jack is having a bit of trouble with the card too. First of all, it won't start unless I tell it to use all 18 inputs and 18 outputs. Fine, let's pretend I'm using all 18.  Secondly, I don't seem to be able to figure out reasonable low-latency settings. I'll dig deeper, but from what I gather, people are able to go as low as a few ms, whereas even 12ms is giving me tons of xruns, and that's with running jack at a priority of 70.

Oh well, let's keep digging.