Jump to content

timofonic

Members
  • Posts

    82
  • Joined

  • Last visited

Posts posted by timofonic

  1. Hello.

    I would like to know if the XCore from XMOS can be an interesting platform for MIDI and music hardware projects, so know the advantages and disadvantages compared to other chips like FPGA and microcontrollers like PIC or Atmel AVR. A fast looking make it a nice platform for many applications, but I would like to know the opinion by people from the MIDIbox community (specially developers).

    XMOS has developed a family of embedded multi-core multi-threaded processors which resonate strongly with the Transputer and INMOS, named as XCore.

    Transputer was a pioneering concurrent computing microprocessor design of the 1980s from INMOS, a British semiconductor company based in Bristol. Many considered in the 80s to be the future of computing. Despite being highly influential for news computer architecture ideas, sadly it did not live to the expectation.

    XMOS is a 2005 founded fabless semiconductor company that manufactures multi-core multi-threaded processors designed to execute several real-time tasks, DSP, and control flow all at once. It's a British company based in Bristol. The XMOS name is a loosely reference to Inmos.

    The mentioned advantages of the XCore chips from XMOS are mainly the cheaper cost compared to other solutions like FPGA, easier to program and an high level of parallelism even between different chips connected by chain.

    The XCore family are 32-bit processors, that runs up to 8 concurrent threads per. It was available as of June 2009 running at 400 MHz. Each thread can run at up to 100 MHz; four threads follow each other through the pipeline, resulting in a top speed of 400 MIPS if at least four threads are active. The 400 MIPS of each core is equally distributed over all active threads. This allows the use of extra threads in order to hide latency.

    There are actually available in two forms, the one core XCore XS1-L1 and the quad core XCore XS1-G4. The first ones has one core, the second has four (quad core).

    The languages to be programmed can be C, C++ and computer language based on C (plus some similarities with occam) but specific for XMOS chips called XC targetted to fully utilize the capabilities of XCore chips like real time and parallelism.

    About for MIDIbox-like projects, some people did some easy stuff by porting code to the processor. An example is the XC-1 SID emulator demo.

    Here you can find a list of current projects.

    In this youtube video, David May (CTO XMOS and was the chief architect of Transputer from Inmos ) explains the communicating between XMOS proccessors.

    Another interesting video is this XMOS XK1 modular development board overview. It features David May (XMOS CTO) and Ali Dixon (XMOS Co-Founder) discussing some of the features of the recent XMOS XK1 development board.

    XCore Exchange was launched in December 2009 and is a site to enable and encourage innovative and entrepreneurial discussion and collaboration.

    A community about the XMOS processors is XMOSLinkers.org.

    Regards.

  2. How about your programming skills? I could give you the changed builder "as-is" for a proper integration into the sidplay project.

    Me? I even can't code a line of basic. I want to learn computer programming soon, but I think it's very difficult to me and not sure when being able to do something really useful.

    You can ask Raph, a canadian geek that does DIY hardware stuff related to retrocomputing (and a few other stuff related to microcontrollers) and maintains MikMod. Maybe you can convince him for sending a proper patch to sidplay2 developers or making some kind of fork if they don't reply.

    I find the idea of playing SID tunes from C64 with MIDIbox very interesting, but I have no skills.

  3. Thanks! ;)

    Sadly, I didn't found the sizes I need on those shops.

    Luckily I found an ebay selled from USA with a lot of sizes and able do request a custom ebay item with all stuff I want. I'm afraid about the shippment cost being too high so I'll ask him as he said me, in the worst case I may find another shop or sending the package to some friend.

    I found a seller from Germany that is certainly a shop. I'll ask him about informing me in english and if possible to have a custom item for me.

    http://stores.ebay.es/fa-ars

    EDIT: Bad news, the owner of the shop replied me with an unintiligible english (worser than mine) and saying like only having the sizes of the shops...

    Hello we only the ESD bag which has we also with us in shop to have www.fa-ars.de and then on shop of company acre

    It could be nice if someone translates that cryptic message, I only understand certain meaning of it...

  4. Hello,

    I want buy certain electronic components and metallized shielding bags in different sizes with zip-lock and reusable. Here in Spain is impossible to buy those bags in quantities less than 100 for certain sizes.

    Can anyone say me about great electronics shops in Germany that sells online? Not problem if only in deutsch, I may use an online translator or some friend helping me with the page.

    Of course, I want to buy midibox boards. If some nameized seller of MIDIBOX PCBs can sell me those metallized shielding bags too, I can buy it from him.

    Best regards,

    timofonic

  5. The expensive solution would require seperate CAN tranceivers for each PIC. But the diode/pull-up resistor approach works extremely stable as I noticed, and I think that tranceivers are only required for higher distances (e.g. > 1m or so..), or if you are planning to access the master/slaves from another CAN node like from a PC with CAN interface card (I've prepared this multimaster capability in the protocol).

    Maximum distance: 40 meters

    Best Regards, Thorsten.

    Does the CAN interface provide advantages against a MIDI interface? It will be used for interconnecting all the MIDIbox platform modules in the future? You can have a big rack of different MIDIbox modules and connected to a nice panel on your desk using that CAN stuff, so not worring about putting an extremelly giant box on your precious desk space.

    Are there a good list of CAN interface cards for PC? What ones are Linux-compatible? And BSD? I found the driver stuff seems a bit messy, too many projects and it seems none is standarised.

    I found the following URLs:

    http://canfestival.sourceforge.net

    http://www.comedi.org

    http://www.port.de/engl/canprod/sw_linux.html

    http://www.vscp.org

    http://www.ocera.org

    http://www.can-wiki.info

    http://caraca.sourceforge.net

    http://canpie.microcontrol.net

    http://sourceforge.net/projects/cantalope/

    http://sourceforge.net/projects/fttcan4rtai/

    http://sourceforge.net/projects/hms30c7202/

    http://cantrace.sourceforge.net

    http://ar.linux.it/software/index.html#ocan

    http://sourceforge.net/projects/rtcan/

  6. I'm up for spanish translation. I can try to ask a friend for brazillian/portuguese too but not sure.

    I want to help in the english article. Do you give a permiss to use content from your wiki and putting it on Wikipedia? For this is needed to relicense it to  GNU Free Documentation License.

    If you agree, you have three options:

    1. Putting the content I need to use for the wikipedia article in a different license with the following license: http://en.wikipedia.org/wiki/Wikipedia:Text_of_the_GNU_Free_Documentation_License#How_to_use_this_License_for_your_documents

    2. Relicense your entire site and wiki to GFDL...

    3. You can do something like this: http://portal.wikinerds.org/gfdl-nsk-2005sep

  7. Hello,

    I'm a big fan of MIDIBOX and hoping to soon having the enough money for mounting a MIDIBOX SID.

    I'm addict to the wiki stuff and I have some experience on different wikis. I occasionally do some stuff on wikipedia and knowing the dokuwiki and tikiwiki syntax. I like to creating new articles and editing existing ones, my english is a bit poor but I try to write better on the wiki than forums or IRC/IM.

    I think uccaps.de information must be on a wiki, this will help a lot the editing of the information by other people and improving it. Of course TK can review those changes too. An idea could be merging wiki.midibox.org with uccaps.de into one page because practically all content from ucapps is about midibox. The actual page could be redirected to the wiki.

    TK: What do you think? I want to help in the wiki stuff, it seems it needs a bit maintaining on certain stuff and I  ask permission and your opinion before doing something. I don't want to annoy with stuff you can consider wrong by some reason I don't know. If you agree, I will help all possible on this effort and helping to maintain the wiki.

  8. This is a bit off topic but :

    How is it possible to open the SNES without the special nintendo screwdriver ?

    Thanks !!!

    http://www.tototek.com/phpBB2/viewtopic.php?p=281

    An useful thing to know if you want to open you Nintendo console without (!) a gamebit screwdriver...!

    You can use some plastic pens! Carefully heat up the tip som the pen, and as fast as you can put it over the screw! It worked for me!

    I did that too and it works ;)

  9. You missunderstandme completely!

    I just only said is one of the possible uses, I mean the wii controller can emulate a theremin-like thing. The possibilities of the wii controller are nearly infinite, please look videos about wii music..

    http://en.wikipedia.org/wiki/Wii_Music

    http://media.wii.ign.com/media/827/827335/vids_1.html

    It can even simulating drums and there was a demostration by proffesional musicians. The wii controller is really very impressive!

  10. Maybe is not exactly bluetooth, but some low-latency derivated. The device is the most precise thing I seen, it seems to be done with some kind of black magic by Nintendo, too good to be true. I didn't see latency on the presentations, it runs perfect!!!  :o

    I'm sure the Wii controller will be reverse engineered and electronic geeks over the world will use it for their DIY projects!!! ;)

  11. Sorry for refloating, but I have some ideas...

    I'm sure some people from here know about Wii, the new super-innovative (and obscure in specs) videogame device from Nintendo, mostly because of the controller. The controller has full spatial X Y Z detection, so it seems very interesting for a lot of uses like gaming, simulating real objects interaction and a lot more.

    Nintendo used gyroscopes and technology from Analog Devices for their revolutionary controller, and this kind of device can have *A LOT* interesting uses in the music world.

    Because the controller will probably be bluetooth 2.0, with a few of reverse engineering it will can be used for non-Wii purposes. Think of controlling a MIDIBOX with the Wii-controller in a matter of hyper-theremin in steroids, converting MIDIBOX on a real-time device for orchestal purposes or a great interface for creating weird and creative stuff, easier usage. Imagine controlling MIDI stuff in a like a orchest director, like a theremin user and an infinite number of possibilities!

    Because Wii stuff will be mass produced, those Wii controllers will probably be very cheap. I consider this very accesible for most users and I'm sure there are cheap  bluetooth 2.0 stuff for use on DIY projects ;)

    Wii is still not released, but I think is something to consider for the future. What do you think, guys? ;)

  12. Hello,

    I'm not sure if others said this idea, but here I go:

    I see the MIDIBOX project has many possibilities and is underrated and underused. I think it must use some manners for incentiving the creation of audio content with MIDIBOX.

    I think of a MIDIBOX compo, here are the rules I think for it:

    - A year of lifetime, in the end of the year the 50 best compositions will get some very interesting gifts (so sponsors is very needed for this). The minimun number of compositions for the compo are 1000 compositions, if that number is not reached, the compositions will be merged to the next year compo.

    - Only MIDIBOX synths can be used, not filters using computers programs or any kind of non-MIDIBOX synths (hardware of software). Midified stuff probably will be rejected. Original MIDI files and needed stuff (like sysex patchs) for reproducing them on other MIDIBOXes must be sended.

    - There will be a category for each synth. Music using various synths at once will be on a category named "misc".

    - Each category will win the same stuff. After that, there will be a "best of all" awards for the best projects of all categories, with a lot more important (in terms of price or interest  and oriented for MIDIBOX users) gift.

    - All projects must have a Creative Commons license and must allow to being sampled and modified, in a GPL or BSD form (this must be considered, but those projects must be some kind of inspiration for future projects, so they must be copyleft and allowing modification and sampling).

    - The number of participations is unlimited.

    What do you think about this idea? First is needed some sponsors for the project, like some geeky universities, experimental music groups and studies...

    Best regards,

    timofonic

×
×
  • Create New...