*jOi~

Woo! wiki problems are all fixed now.. Just hit 100 chipsets! Seemed to go from 50 => 100 in no time at all. Now it's just a matter of time before I start making 'bleeps' and 'bloops' and 'chjdigeajifossszzzzzzz' again.. Yes, I love FM.

I had to wait overnight for things to be updated, but now I can login to the wiki with 'Joi' as the username - no issues. I did however still have to login to the forum with username '*jOi~', yet it says 'Hello Joi' up the top. Is this because my username can't be changed but my nick can? And the wiki uses your nick as login? Hrmm.. These questions aren't terribly important, but thanks for the suggestion as I can now use both. Trent

Thanks for the suggestion. I deleted all temporary internet files and cookies, which allowed me to access the wiki. I then tried to login and again I entered my username and password and then "permission denied" again. I think I'll try changing my username now (can I do that?). The wiki is so great - I surprise myself that I've never logged in before... hah! No harm in maturity though, the stars and tilda are getting a little young for me.

Hi folks, not sure if this is the right place to post this, but I am currently unable to access the wiki (and add my name to the MBFM bulk order) Last night I looked at the page, hit login, entered my username etc. and clicked ok. I'm now getting the following message::: Warning: preg_grep() [function.preg-grep]: Compilation failed: nothing to repeat at offset 20 in /home/midibox/public_html/dokuwiki/inc/auth.php on line 259 Warning: Invalid argument supplied for foreach() in /home/midibox/public_html/dokuwiki/inc/auth.php on line 261 Warning: preg_grep() [function.preg-grep]: Compilation failed: nothing to repeat at offset 18 in /home/midibox/public_html/dokuwiki/inc/auth.php on line 283 Warning: Invalid argument supplied for foreach() in /home/midibox/public_html/dokuwiki/inc/auth.php on line 285 Warning: Cannot modify header information - headers already sent by (output started at /home/midibox/public_html/dokuwiki/inc/auth.php:259) in /home/midibox/public_html/dokuwiki/inc/actions.php on line 91 I haven't previously attempted to log in to the wiki, and whenever i visit any page of the wiki the same thing happens. It does however still say "logged in as *jOi~" in the bottom message bar, just that "Permission Denied" is always there. I'm on OSX 10.4.11, and have tried with Safari and Firefox to the same conclusion. Any help would be fantastic

This sounds great - I've been wanting to make a MBFM for a while now (1/2 rack enclosure), but I could never get around to finding the old soundcards. Unfortunately I'm getting a 'you do not have permission' on every part of the wiki... I would love to order 2 chipsets if someone could help add my name - or better yet suggest what might be wrong when I try to access the wiki (I will do a search now...)

I won't say much here, and don't want to cause anybody any strife. I would however like to say a few words in response to you Smash I read your post and saw some very interesting points, visited your links and opened my eyes that little bit more. The way you have attacked and labeled the folks at monome as 'clueless', 'dollarblinded' or 'lying', is hasty, unprovoked and inconsiderate. In my discourse with Mr. Crabtree, he has been very helpful and gracious, originally pointing me to this very community some 2 years ago suggesting it as a good starting point into controlling computer music. They are very kind and thoughtful people, and whilst being 'kind and thoughtful' is no rebuttal to your harsh words, it does go to show the thought they have put into the project. Whilst you may present 'facts' and 'proof' that opposes their hopes and aims, I see it as admirable that they have done quite extensive research (extensive, yes) and believe their method of manufacture is the best way for them to move forward both ethically and commercially. Thoughtful people, attempting to do their best for the community both world and local, are held in high esteem in my book. Re: Buttons. I see your point, though for a small company with large profits, it seems a little self-defeating to go through these processes. Sometimes corners have to be cut, and solid design must be trusted? I'm not sure how I would deal with this issue, however I wouldn't let it stop me from producing a product that would further artistic endeavours. I don't mean to step on toes with this post - I merely want to suggest that we should be more gratuitous and in general, more kind to those whose projects we are not totally informed about. Slinging around words as were stated above would be enough to throw off some who may well own a device and are visiting the site because they have become interested in the idea. With kind wishes, Trent

Just a humble opinion, but isn't the FM already pretty simplified. All of the sound is generated on chip, and it seems that as is would be a great audio source for a modular. I guess the complicated bit is the controls you use. If you want to control pitch etc. I believe it would be a case of requiring an AIN module that you could assign to certain parameters in the FM program. I haven't built one.. so I don't really know.. Trent

Hi everybody, just wondering if anyone has any thoughts where i could find some high-output (200+mcd)bi-colour LEDs. I'm using pads from <monome.org> which require 3mm (T1) size LEDs. The trick is to have access to 3 different colours for more visual correlation between buttons rather than simply memory. Example - I want to be able to load a set in ableton and use 3 buttons - 1 for record (Red), Mute (Yellow) and Solo (Green). I'll program all of the MIDI code in MaxMSP. I'm happy to spend ~1USD per Dual LED. If anyone has suggestions, I would be ever-so-grateful. Trent

just been playing with the code, and it works well - few problems though which i think stem from the lack of debouncing. i tried to migrate the code into C but alas it hasn't worked as seemlessly as hoped (i have a much easier time working with C code). hardware was 8x16 switch matrix using 2xDOUT and 1xDIN i assume you left the debouncing out, as it would consume too much processor power for all the buttons, however on second thought, maybe my problem is not debouncing... i will get back to you about it. nonetheless, it works *thumbs up* t

hey qbas ive been trying to write the code to expand the examples to 16x16 and been unsuccessful so far. looking at your code though, it seems we approached it quite differently > i will try and test it out tomorrow. i've only got 8x12 matrix at the moment, but theoretically it should work on smaller matrices yes...? it's too late in the evening right now, but hopefully some success soon... trent

wow! this box looks incredible - all the best wishes for the gig in ~1 week!!! would really love to see some higher resolution photos of the thing, and maybe even a short video of the whole setup in action. how are you using the soundcard, and what in/out do you have available on the box? I really love the idea of having the sound output inside the midi controller - that is so cool. only problem for me is i only have firewire cards (2 cables anyway!)... trent

i have seen that page before, though somehow i always manage to lose the bookmark.. thanks for the redirection. i am trying to dim each LED individually, so I would be incredibly interested in the solution you speak of - please keep me/us posted. as an aside i'd just like to say how much of a kick i get out of having my questions answered by mr. Henke himself.. thanks trent

friends, i've been searching this forum far and wide, and whilst i can find a huge number of references to using PWM to control the brightness of LEDs, i haven't been able to find anything beyond "it's possible", and "i used it in my box". i was curious to know if anyone had any sample code as to how it is done in the firmware/application, and what the potential number of PWM'd LEDs I could hope to get from one core. originally i had planned to run a matrix, though after various experiments, im much more worried about maximum brightness, and the prospect of building a ridiculous number of DOUT modules is actually kind of endearing. I would like to do ~200 LEDs, and am quite prepared to chain a number of core modules together to do this. any thoughts? trent

ergg... apologies but in fact, the magic number is '21', and so 2*0x0b-1 should be right. Thanks for picking me up Therezin.. I wish you were right though, and I'd been careless in my decimal > hex conversion, but alas the PIC is programmed correctly and yet I can't get applications to upload correctly. Running MIOS Studio via Max/MSP w/ the serial object (and MIDI Yoke), and whilst I had no troubles uploading MIOS to both PICs, I could only get my app to upload correctly to the chip at 38400. Nevertheless, I have successfully got the 38400 PIC running my led/button matrix program (c/o Wilba) without any serial cable or MIDI port. Sometimes it is easy to forget the small goals when you fall short of the large ones, but I have just proved to myself that it is possible to run a MBHP with only a PIC burner and a cheap/small USB chip. This is still a Max/MSP reliant implementation. Developments soon..

success!.. well the beginnings of.. after ~6hrs of study i know a huge amount more about the beauty of PICs and how they function, assembler is beginning to make sense, and I discovered the all important meaning of 2*0x21-1.... the simple answer = (2 * 33) - 1 = 65 how does 65 give a baudrate of 38400? only the PIC datasheet knows the answer.. '79' gives 31250 '65' gives 38400 '21' gives 115200 my new line is therefore: 2*0x0b-1 i finally used MPLAB (which i installed ca. 12mths ago) and compiled my code. my brand new PIC burner has 2 fresh PICs for me now - one with the standard TO-COM 38400 baud (which I know is working), and a second with TO-USB 115200 with which I will now experiment. Next tasks are not the easiest - trying to decipher the way Max/MSP handles SysEx commands, so I can upload MIOS - the joys of uncharted waters with a lack of available drivers.. I'll keep you updated trent

thanks for the direction stryd... i get the feeling that i will figure this out, but after the first half hour of staring at that datasheet, im glad there is no deadline hanging over my head to get it finished. the answer is there, i can see it, i just read it, all i need now is to understand it. cheers

Hi folks, I've had a little box here working w/ a FT232 USB chip for a while now, and it's about time I delve a little further into the possibilities of using MIDI over USB. I'm planning to write up a driver using either chuck, or a small MaxMSP application, though this is probably still a few months off. My question is something that I have touched upon in previous questions, but never got a full answer. What I am looking for is the point in the bootloader firmware that I can modify to allow baudrates other than MIDI and To-COM. My plan is to alter the To-COM value and for it to become To-USB running at 115.2 kbps. Whilst I've been learning a lot about java (and similar languages) I'm still useless working with assembler, and staring at hex values continues to mystify me. If anyone could point to the register that must be changed, I will experiment with the value and investigate the effects of different baudrates - see if i can cause problems w/ overloading the I/O buffer. Thanks for the help friends Trent Edit: I believe this is the spot, but I have little clue how to interpret the hex values, and what they should become.. movlw (1 << TXEN) | (1 << BRGH) movwf TXSTA IFCLR MIOS_BOX_CFG0, MIOS_BOX_CFG0_TO_HOST, goto Init_USART_MIDIBaudrate Init_USART_RS232Baudrate movlw 2*0x21-1 ; set baudrate to 38400 baud IFSET MIOS_BOX_CFG0, MIOS_BOX_CFG0_20MHz, movlw 0x21-1 goto Init_USART_Cont Init_USART_MIDIBaudrate movlw 8*0x0a-1 ; set baudrate to 31250 baud IFSET MIOS_BOX_CFG0, MIOS_BOX_CFG0_20MHz, movlw 4*0x0a-1 ;; goto Init_USART_Cont

yes yes.. documentation all in good time. i'll still ironing out a few mechanical issues at the moment, but after i knock the next few things on the head, some simple documents should be doable. and thanks for the name clarification... ... thanks wilba! you've been a wonderful help with the programming, as whilst im becomming quite efficient in max/msp, i am still rather useless when looking at assembler. cheers trent

hey everyone, just writing to say that after about 6months with this half finished project lying around the house, i finally have my midibox reliably talking to my computer through USB. I had started with severe problems where large numbers of messages would be lost and it was useless.. prior efforts used the COM port though on my laptop that ancient technology is no longer available... as much as everyone says how cheap MIDI to USB adapters are, routing the PIC directly to a USB module actually makes the core module to be much simpler, and a breakout board for the USB module can be got very cheaply. the ease of use is also wonderful, as the box is powered through USB and requires only one capable for everything. the idea of having to carry around adapters and midi cables and the such is just a pain, and i like my boxes small so putting the adapter inside the box was unwanted. my midibox is a 8x8 button and led matrix programmed by Jason W from around here (sorry forgot your username). the code is minimal and simply does what it needs to with nothing else. i'm using Duplo to create the button caps - they are lit by the red leds that are inside. communication is made through FTDI FT-232 chip with the USB-serial driver. I somehow wired the SMD chip with a cheap dick smith (RS) soldering iron, and it still works. I think usb parts cost ~$20AUD and i saved about $5 by not buying the chips and jacks for midi (mind you, no midi cables too)... as you can see not terribly more expensive. I use Max/MSP to interpret the serial data and split it up into midi bytes. I can then route these messages to a midiout object to send midi to any application on my computer (coreMIDI on my iBook). Typically I only use max for my music anyway, but occassionally I will want to use Live or obviously sysEx and the such. It took some guess work to get the messages right, but it is now working. pictures are coming soon, but i just wanted to put this out there so people know that it *is* possible to use USB for their midibox, so long as you have a program to interpret the serial data (as I could not find a USB-midi driver that worked). I hope this encourages people to try this out. ...next task is to alter the code to allow transfer speeds of 115k/s rather than 38.4k/s of serial or 31.2k/s of midi... when trying to address a large number of buttons and leds at the same time this could be handy.. trent

grepo, thanks for the those couple of links - i think i have been very close, but never really gave it the time to just sit down and slog through all of the literature. i've been in touch with brian crabtree (the head guy at monome) quite a bit, and he's given me a whole bunch of great info and insights. i actually have my pics flashed with the serial baudrate (meant for the to-COM module), but I guess there still maybe some kind of clash there - i will have a better look today and perhaps some success... thanks for the kick in the bum to get me thinking about this again. ill be sure to post a bunch of pics and info if i can get this little device to work... im sure some guys around here will have a right laugh at how i soldered that tiny little SMD with a $30 dick smith soldering iron and terribly large diameter solder.. thanks, and i hope it gives you hope that it may be possible to use this chip - perhaps at even higher baudrates (with less latency??) than the standard midi in/outs. trent

I can't speak for the whole world, but I have had no luck with this chip. I've built the sample circuit out of the datasheet on two separate occasions and had the same problem: Firstly, I could not find any driver that would work with the usb data and convert it into midi, so my solution was to use Mas/MSP. The problem I experienced here was that little pieces of the midi messages would get lost. This meant that not every message was captured and it was unusable for a piece of performance gear. I have recently returned from overseas and am just getting back into my midibox, which I am determined to run through USB so hopefully I can work out these problems and not have to purchase a usb interface. I have a sneaking suspicion it has to do with the signal not being filtered well enough, so perhaps that is the next point of attack for me. If you do try with this chip, I hope you have better luck, but for me I could not get it to work. I need to learn how to program code, and take the information directly from the serial port. good luck trent

Hey stryd! I'd love to go in with you on the deal, but unfortunately i'm headed overseas for ~3months.. so it probably won't work out :-\ Alas, I hope you find a few to chip in and soften the burden of shipping... but even so, I seem to remember paying $8 for a quarter pound roll ??? I guess i'm just a sucker who needs to shop around a bit more... Cheers mate, Trent

Sorry about my delayed response guys - I've been deep in study for my end of year university exams...going to be a doctor soon... does that scare you? anyway - the only markings on the pots are as follows: ALPS (logo) JAPAN 015M 10KohmB plus the terminals are numbered 1, 2 and 3 The guys that I've sold some to so far have payed ~$1/pot Shipping to the UK was about $17 and I would assume to be comparable to european countries, though If you're interested shoot me an email or PM and I'll try and work out something more exact. I'm off on holidays soon though, so if anyone is desperate please get in touch ASAP or you may have to wait a couple months... ironically I'll be in europe at that time - only a shame my backpack is full to the brim already. If anyone has a floor for me to sleep on i'll love you forever! haha cheers Trent

Well I can say a huge thanks to Wilba for helping me out with that code! It's working on a grass roots level, and after I figure out how it works I'm sure it'll do more than I think. I've got the hardware here to do a hacked monome, with my switch matrix just needing to be attached to the registers, along with figuring out how to integrate that scanning matrix demo on the ucapps page with this wonderful little led matrix handler. When I get some time (Uni exams this week) I'll try to put together a program that allows full control. I do just want to say though, that Brian Crabtree (the mastermind behind the monome) has been a wonderful help to me, and I encourage you to think about buying one for the look and feel alone! I'm just waiting for the 16x16 to come out... trent

billmac: why would you spend $50 when you can slave away over a super tiny SMD chip, trying to solder each tiny pin to a wire because you're too cheap to pay for a PCB!! ::)haha seriously though, if we can get it working, the options of direct serial transmission of the midi data mean things like baud rates of 115000kbps (~4x faster) and up to 500mA direct filtered bus power! anywho... i will continue to try and get everything to talk to each other.