Shop OBEX P1 Docs P2 Docs Learn Events
Just ordered a BE Micro CV for $60, delivered - Page 3 — Parallax Forums

Just ordered a BE Micro CV for $60, delivered

1356

Comments

  • RsadeikaRsadeika Posts: 3,837
    edited 2014-08-10 08:12
    All things considered, I think the DE0-Nano is a better deal. It's a little more expensive and lacks the RAM needed for the character ROM, but it has a faster Cyclone IV chip with a lot more extra logic, and it compiles way faster. Plus, it's easier to get.
    So, will the Be Micro CV still be supported, or does it get pushed to the back burner, meaning the DE0-NANO is the way to go?

    Ray
  • David BetzDavid Betz Posts: 14,516
    edited 2014-08-10 08:13
    Rsadeika wrote: »
    So, will the Be Micro CV still be supported, or does it get pushed to the back burner, meaning the DE0-NANO is the way to go?

    Ray
    It seems like the DE0-Nano is the best way to go for those of us who want to try modifying the Verilog code since it leaves more open space for enhancements. Sounds like the BeMicro CV is pretty much full with just the standard P1 functionality. Of course, I guess we could just use fewer COGs on the BeMicro CV if we run out of space.
  • TorTor Posts: 2,010
    edited 2014-08-10 08:21
    Leon wrote: »
    I couldn't see it on their About page.
    Site is up again, found it in the FAQ section:
    http://www.chip1stop.com/web/USA/en/support.do?cat=000002&itm=000001
    Can anyone use Chip One Stop?
    Chip One Stop is a membership-based online shopping site dedicated for corporations. According to our terms of use, those who can use our website are limited to individuals who belong to a corporation, organization, or association and use this website for business purposes (sole proprietor). Any such person can place an order and receive a quote by free membership registration.
    * Consumers are not allowed to use or register. Please refer here for Chip One Stop's terms of use.
    (Red font added by me)
  • LeonLeon Posts: 7,620
    edited 2014-08-10 08:23
    Just tell them you are a sole proprietor.
  • cgraceycgracey Posts: 14,155
    edited 2014-08-10 12:52
    Rsadeika wrote: »
    So, will the Be Micro CV still be supported, or does it get pushed to the back burner, meaning the DE0-NANO is the way to go?

    Ray

    It'll be supported. It already is, though some tuning is needed. By tomorrow afternoon, it should be all done.
  • jazzedjazzed Posts: 11,803
    edited 2014-08-11 13:31
    Yay.

    Newly arrived BeMicro CV works (using code Chip posted here).
  • Kerry SKerry S Posts: 163
    edited 2014-08-11 14:30
    Arrow has them back in stock! Just ordered mine. Should have it Friday.

    Something to play with and learn Verilog while I wait for the Amazing Parallax FPGA Board.
  • 65816581 Posts: 132
    edited 2014-08-14 07:50
    Hey,

    I got mine today, too (Germany - shipped by arrow from US). Got the Propeller Demo running. Where can I find the source code?
    The board pin mapping file seems to have an error (thats probably why the leds not properly lighted9..

    set_location_assignment PIN_N1 -to user_led_n[0]
    set_location_assignment PIN_N2 -to user_led_n[1]
    set_location_assignment PIN_U1 -to user_led_n[2]
    set_location_assignment PIN_U2 -to user_led_n[3]
    set_location_assignment PIN_W2 -to user_led_n[4]
    set_location_assignment PIN_AA1 -to user_led_n[5]
    set_location_assignment PIN_AA2 -to user_led_n[6]
    set_location_assignment PIN_Y3 -to user_led_n[7]

    should do the job.

    By the way: My board does NOT contain 5CEFA2F23C8 but the 5CEFA2F23I7NFA! Seems like they
    deliver the Board with different configurations? Also I have 1Gib of RAM instead of 128MiB.
  • 65816581 Posts: 132
    edited 2014-08-14 09:47
    6581 wrote: »
    Hey,

    I got mine today, too (Germany - shipped by arrow from US). Got the Propeller Demo running. Where can I find the source code?
    The board pin mapping file seems to have an error (thats probably why the leds not properly lighted9..

    set_location_assignment PIN_N1 -to user_led_n[0]
    set_location_assignment PIN_N2 -to user_led_n[1]
    set_location_assignment PIN_U1 -to user_led_n[2]
    set_location_assignment PIN_U2 -to user_led_n[3]
    set_location_assignment PIN_W2 -to user_led_n[4]
    set_location_assignment PIN_AA1 -to user_led_n[5]
    set_location_assignment PIN_AA2 -to user_led_n[6]
    set_location_assignment PIN_Y3 -to user_led_n[7]

    should do the job.

    By the way: My board does NOT contain 5CEFA2F23C8 but the 5CEFA2F23I7NFA! Seems like they
    deliver the Board with different configurations? Also I have 1Gib of RAM instead of 128MiB.


    Hey! I managed it to wire the SD-Card in the Pin Mapping file to be setup at Prop Pins 10-14!
    Also the LEDs correspond now to the correct leds! This was a bug in the description files provided
    by the vendor.

    Here you go - replace your pin map file with this and re-synthesize the project:
    http://pastebin.com/DCz45BeV
  • AleAle Posts: 2,363
    edited 2014-08-14 10:54
    I also got the I7 part. The DDR§ is a 1GBit part, so stlll is 128 MiBytes :). Just curious, which part do you have (DDR3) ?
  • 65816581 Posts: 132
    edited 2014-08-14 11:12
    I cannot read it, because its too small and there is a sticker on it..
    I now made a custom version of the propeller emulation which supports sd-card, 2 pushbuttons, 3 dip switches :)
  • 65816581 Posts: 132
    edited 2014-08-14 11:12
    I cannot read it, because its too small and there is a sticker on it..
    I now made a custom version of the propeller emulation which supports the on-board sd-card, 2 pushbuttons, 3 dip switches :)
    It's a shame that the user eeprom on the board is only a 24lc01 and not a 256..

    Update: I "piggybacked" the backside of the pin header and adopted it to take the 24lc256 EEPROM
    from a Propeller Board. Now I am re-synthesizing the FPGA code and will test soon!

    PS: Will the source files for the character eeprom and booter be released as well?
    I am very looking forward to program an SD-bootloader!

    I see now that it is released! How do I compile this to get the two hex files (hi and low)?

    Update2: got external eeprom running!

    next step is adding dual port memory in the fpga and a little core which will allow to pause
    the cpu and dump memory etc. for amazing debugging capabilities.

    by the way... we can include the 6502 soft core in it and we have a working prop laptop!
  • Cluso99Cluso99 Posts: 18,069
    edited 2014-08-14 15:22
    All the booter, interpreter and runner, plus char font, log, antilog and sine are all discussed here.
    Shortly I will be posting the files so that $readmemh will load them into RAM (replacing ROM) and removing the scrambling.
    http://forums.parallax.com/showthread.php/156866-Question-for-Chip-re-ROM-code
  • jazzedjazzed Posts: 11,803
    edited 2014-08-14 16:08
    6581 wrote: »
    by the way... we can include the 6502 soft core in it and we have a working prop laptop!


    Sounds like a blast ;-)
  • David BetzDavid Betz Posts: 14,516
    edited 2014-08-14 16:46
    My BeMicroCV came to my house in Bedford, NH today. Unfortunately, I'm in Philadelphia. :-(
  • 65816581 Posts: 132
    edited 2014-08-15 01:20
    Short summary what I have running so far:

    * SD-Card on BeMicro-Board on Propeller Pins 10-14
    * DIP-Switches (3x) wired to Propeller Pins 15-18
    * 2 Push-Buttons on BeMicro-Board wired on Pins 19-20
    * Fixed Pin-Mapping in Top Pin map so that each LED is
    showing the right cog.
    * Replace Hub RAM by Dual-Port RAM.

    What I am working on:
    * Dual-Port RAM based debugger

    What I planned:
    * Try to boost performance
    * Make use of DIRB and OUTB (64 Pins)

    What I am having trouble with:
    * Video via Composite Output doesn't work yet
    * My propeller 2 still didn't arrive ;)
  • 65816581 Posts: 132
    edited 2014-08-15 07:10
    @Chip: I see the PLL is giving a 80 MHz Clock for the "Propeller". Whats the F_MAX possible and what needs to get changed?
  • ozpropdevozpropdev Posts: 2,792
    edited 2014-08-15 07:27
    A bunch of us are already running DE0/DE2 @ 140 MHz.
    pik33 has pushed even harder I think but it becomes a bit unstable. :)
  • Cluso99Cluso99 Posts: 18,069
    edited 2014-08-15 07:40
    ozpropdev, are you getting timing warnings in quartus? I get quite a number even from the original code.
  • ozpropdevozpropdev Posts: 2,792
    edited 2014-08-15 07:58
    Cluso99 wrote: »
    ozpropdev, are you getting timing warnings in quartus? I get quite a number even from the original code.
    All sorts of warnings which to be honest I haven't paid a lot of attention to as I'm happy not to be getting erros now.
    Code runs Ok on both DE0 and DE2, so far no issues. Compiling at the moment.....fingers crossed. :)
    Building a 16 Cog beast right now.....
    Cheers
    Brian
  • 65816581 Posts: 132
    edited 2014-08-15 08:26
    ozpropdev wrote: »
    A bunch of us are already running DE0/DE2 @ 140 MHz.
    pik33 has pushed even harder I think but it becomes a bit unstable. :)

    Ok, can you tell us what you changed? I changed the PLL multiplier to 32 (from 16) and left the divider at 5..
    So serial runs at 230400 baud instead of 115200.. Serial Hello World code works at 160MHz, but math is completely broken.
    I try now to re-synthesize it with custom timing constraints.
  • 65816581 Posts: 132
    edited 2014-08-15 08:27
    ozpropdev wrote: »
    All sorts of warnings which to be honest I haven't paid a lot of attention to as I'm happy not to be getting erros now.
    Code runs Ok on both DE0 and DE2, so far no issues. Compiling at the moment.....fingers crossed. :)
    Building a 16 Cog beast right now.....
    Cheers
    Brian

    Will you share your code? :) Good luck! Already worked on a 64 pin version?
  • ozpropdevozpropdev Posts: 2,792
    edited 2014-08-15 09:14
    6581 wrote: »
    Will you share your code? :) Good luck! Already worked on a 64 pin version?
    Hi 6581,
    Definitely will share code when I get it going properly.
    yes I have made my images 64 IO (even had a PortC 77 IO version for a while) :)
    28/5 multiplier/divider for 140MHz.
    Cheers
    Brian'
  • Cluso99Cluso99 Posts: 18,069
    edited 2014-08-15 13:54
    WTG Brian! Seems you understand verilog - well at least better than some of us.
  • Bob Lawrence (VE1RLL)Bob Lawrence (VE1RLL) Posts: 1,720
    edited 2014-08-15 15:13
    Cluso99 wrote: »
    ozpropdev, are you getting timing warnings in quartus? I get quite a number even from the original code.

    I got the warnings as well with no changes to the code , just compiling. I just ignored them. :)
  • 65816581 Posts: 132
    edited 2014-08-18 06:49
    I am working on a QSF to Image converter, which shows you then directly which pins in your HDL code / pin map
    are physical pins on the BeMicro CV. So far it outputs a graphic which looks like this:

    attachment.php?attachmentid=110462&d=1408369769
    851 x 987 - 780K
  • mindrobotsmindrobots Posts: 6,506
    edited 2014-08-18 07:05
    Wow! That is very impressive!! Sure is easier than going in and updating a picture from the Excel spreadsheets I'm using!

    Will it be able to do the same with a nano or DE2 QSF file? (You knew someone would ask! :smile: )
  • 65816581 Posts: 132
    edited 2014-08-18 07:09
    mindrobots wrote: »
    Wow! That is very impressive!! Sure is easier than going in and updating a picture from the Excel spreadsheets I'm using!

    Will it be able to do the same with a nano or DE2 QSF file? (You knew someone would ask! :smile: )
    Thank you! ;)
    Sure, its a very dirty solution (a python script and an svg-template)..
    The Python Script reads the SVG-Template and the Pin-Assignment and replaces the strings via regex (PIN_XXX) from the SVG
    template and writes them to STDOUT! STDOUT is then the SVG-File with your pin map. Quick and dirty, but it works perfectly.

    Here is how the template looks:
    template.png
    609 x 705 - 529K
  • 65816581 Posts: 132
    edited 2014-08-18 07:21
    Now I also extended it to support the LEDs. This way you see, that my QSF has the LED fixed (the published
    original file has them wrong direction around). Original qsf result also attached now.

    another2.png

    orig.png
    851 x 987 - 789K
    851 x 987 - 788K
Sign In or Register to comment.