DIL module for P2?

Hi,
to use P2 in maker projects it is necessary to have a module with essential addons for an attractive price. Dual inline modules are very handy. I think, they are used also widely, because pcb space is expensive.
When do you plan a DIL module for a competitive price? What add ons will it have? (SD slot at least, I think?)
Regards Christof

Comments

  • jmgjmg Posts: 13,922
    Have you seen this ?
    http://forums.parallax.com/discussion/168645/p2d2-an-open-hardware-reference-design-for-the-p2-cpu/p1

    That has 0.1" headers, and I have a variant PCB design that has a Raspberry Pi pin map on the 40 pin headers.
    Yes, that has SD slot, and choices for Oscillator sources, and Peter added a Si5351A footprint, for very flexible clocking choices.
  • I’ll be doing at least two pcbs...

    A 1.1”sq pcb with 0.050” pitch headers around all 4 edges. There will be NO regulators OR Flash OR SD.

    A 1.1x1.4” pcb with 0.050” pitch headers around 3 sides and regulators and SD on the fourth side.

    I’ll probably do a large base board to break out these boards with various connectors.
    My Prop boards: P8XBlade2, RamBlade, CpuBlade, TriBlade
    Prop OS (also see Sphinx, PropDos, PropCmd, Spinix)
    Website: www.clusos.com
    Prop Tools (Index) , Emulators (Index) , ZiCog (Z80)
  • My choice would be the P2D2 PCB - I already have a lot of projects in mind for something like that P2D2 Hopefully they will go into production.
  • I know Peter will be doing a revised P2D2 if he hasn’t already done so. I’ll let Peter comment further.
    My Prop boards: P8XBlade2, RamBlade, CpuBlade, TriBlade
    Prop OS (also see Sphinx, PropDos, PropCmd, Spinix)
    Website: www.clusos.com
    Prop Tools (Index) , Emulators (Index) , ZiCog (Z80)
  • There will also be a choice of Parallax boards and modules for P2 once the chip status moves from "Engineering Sample" to full production.

    Official announcements will follow once testing has been completed with the Eval boards.


  • JRetSapDoogJRetSapDoog Posts: 795
    edited 2019-08-13 - 00:38:06
    Back during P2-Hot, a board layout was published for a module. That was an encouraging sign for the community. But due to the challenges P2-Hot faced, it was re-spun into the current cool chip (like a phoenix rising to live again), and the module never came to be. So, it's understandable that Parallax has focused on getting the P2 ready for production. But with the latest test results being so good, maybe it's not too presumptive to start sending some signals again (maybe they just did) as to their intentions in the module space (though maybe they'll wait for the packaged samples to arrive first, hopefully this month).

    Anyway, I like what Peter has done with the P2D2 and was hoping that Parallax would partner somehow with him on that (beyond whatever cooperation they've already had). I thought that Peter's form-factor was good enough that I built a P1 module for a project using the same DIL double header layout. And it plugs in nicely with a base/carrier board (as the module is not too big (or small)).

    At any rate, there almost for sure needs to be a Parallax manufactured or sponsored module board that can plug into a carrier. And they know that, but are just waiting for the appropriate time to target their resources. Good days are ahead. Things are really starting to move now. And other members will do their own modules, too, as Parallax doesn't want to make lots of different ones (kind of like the Apple product approach).
  • Dear Parallax;

    I’ve been a very good boy this year. I’ve washed behind my ears and ate my veggies every day. So, for Chipmas, could I please have a P2-Flip? I know its a lot of pins, but my breadboard has room.

    Your friend,

    Roark in Deep South Texas
  • I agree. Good days ahead! (Now I just gotta pry P2 time back in to the amount I want.)

    Honestly, I want Parallax to make a deluxe P2 board like the PPDB. I used the crap out of mine. Best ever.

    That and a proto board type thing one can plug a P2 module into would be fab!

    Do not taunt Happy Fun Ball! @opengeekorg ---> Be Excellent To One Another SKYPE = acuity_doug
    Parallax colors simplified: https://forums.parallax.com/discussion/123709/commented-graphics-demo-spin<br>
  • jmgjmg Posts: 13,922
    JRoark wrote: »
    ... could I please have a P2-Flip? I know its a lot of pins, but my breadboard has room.
    The P2D2 has 0.1" 40 pin headers so that can go into a bread board.
    If you want a single row PCB, something has to give.
    I toyed with a DIP64-900 footprint, but you cannot connect all P2 io pins to that, and that board is quite a bit longer than P2D2 form factor.

  • can one still get sockets for the 68000 series?

    That where the largest DILs I ever used

    Mike
    I am just another Code Monkey.
    A determined coder can write COBOL programs in any language. -- Author unknown.
    Press any key to continue, any other key to quit

    The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this post are to be interpreted as described in RFC 2119.
  • jmgjmg Posts: 13,922
    msrobots wrote: »
    can one still get sockets for the 68000 series?
    That 0.9" wide one was what I chose, but they are 'fading from view'.
    They are stocked at Digikey and Mouser, in modest volumes in machine screw.

    There are also dual-wipe 64 sockets out there, but in 1.778mm & that's no longer bread-board compliant....

  • jmg wrote: »
    msrobots wrote: »
    can one still get sockets for the 68000 series?
    That 0.9" wide one was what I chose, but they are 'fading from view'.
    They are stocked at Digikey and Mouser, in modest volumes in machine screw.

    There are also dual-wipe 64 sockets out there, but in 1.778mm & that's no longer bread-board compliant....

    With an MOQ of 102 from Mouser for the machine pin sockets; they aren't exactly hobbyist friendly quantities.
  • jmgjmg Posts: 13,922
    AJL wrote: »
    With an MOQ of 102 from Mouser for the machine pin sockets; they aren't exactly hobbyist friendly quantities.
    Mouser show 184 in stock of 110-93-964-41-001000
    However, the price is an indication of how many people still use these...

  • Why do you need a DIP socket? A pair of SIP sockets would work just as well and might be easier to source. I use SIP sockets with the FLiP. One huge advantage is that I can place other components between the SIPs without interference from a DIP's crossbars.

    -Phil
    “Perfection is achieved not when there is nothing more to add, but when there is nothing left to take away. -Antoine de Saint-Exupery
  • Why do you need a DIP socket? A pair of SIP sockets would work just as well and might be easier to source. I use SIP sockets with the FLiP. One huge advantage is that I can place other components between the SIPs without interference from a DIP's crossbars.

    -Phil

    Brilliant...and perfect timing for my current project :cool:
    Failure is not an option...it's bundled with the software.
  • VonSzarvas wrote: »
    There will also be a choice of Parallax boards and modules for P2 once the chip status moves from "Engineering Sample" to full production.

    Official announcements will follow once testing has been completed with the Eval boards.


    Thanks for the answers! I am looking forward to the offers. I think a breadboard friendly version with usb-programming interface, power regulator and SD-slot for a good price would be attractive, even if it has a reduced number of pins available. A good price seems to be very important, as there seem to be not too many maker projects, that force you to use a P2. (Actually, I do not yet see anything that can't be done with others using DMA, Interrupts and/or preemptive multitasking. I am still wondering, what could be done with the cordic, which sounds interesting. But if you don't do assembler, it will just be hidden by the compiler to do some math? Do the Dacs have 16bit 41kHz CD audio quality? As far as I have understood, the ADCs have about 12bit usable resolution.)
  • jmgjmg Posts: 13,922
    Do the Dacs have 16bit 41kHz CD audio quality? As far as I have understood, the ADCs have about 12bit usable resolution.)
    The P2 DACs are power supply referenced, so the power/gnd supply noise is likely to be one limiting factor.
    Super-low-noise regulators exist, but they cost much more than i2s DACs !
    The P2 DACs have a 16b dither mode, that would be audio-useful, but anyone chasing 'CD quality' top end, would be best using external i2s DACs / headphone drivers.
  • 16bit, 44.1KHz is passé. Why bother?
    Failure is not an option...it's bundled with the software.
  • Well, with the exception of peak dynamic range possible, it exceeds human hearing.

    A very few, young people can hear 22khz and maybe a bit higher. Everyone else won't really benefit.

    This is why:



    It is damaging, but people can enjoy a 120db performance. 16 bits = roughly 96db (and yes, that is a quibble)
    Do not taunt Happy Fun Ball! @opengeekorg ---> Be Excellent To One Another SKYPE = acuity_doug
    Parallax colors simplified: https://forums.parallax.com/discussion/123709/commented-graphics-demo-spin<br>
  • Christof Eb.Christof Eb. Posts: 318
    edited 2019-08-14 - 05:43:47
    I am well aware, that 16bits is not the best you can do these days. The question was about what is possible without external hardware. Cogs own multiplier is limited to 16*16bits too, if you exeed this, P2 is somewhat single core.
    The "singing monks" of P1 are a very impressive demo, what you can do with about 12bits. 12bits are quite good, if the signal always has the same full loudness.
    I was shocked seeing the price of the P2ES board. I understand, that they are very expensive to make. I doubt, that the costs are covered.

    But on the other side there are esp32 boards out there with a lot of power for a low price even, if you don t use bt and wlan. An other side is blocked by the Linux boards.
    Alltogether I think a DIL board for about 25Euro with Usb and SD can be attractive.
  • MJBMJB Posts: 1,119
    I am well aware, that 16bits is not the best you can do these days. The question was about what is possible without external hardware. Cogs own multiplier is limited to 16*16bits too, if you exeed this, P2 is somewhat single core....

    what do you mean by this?

    every COG can use the CORDIC multiplier at the same time ...
    and each of those is pipelined in itself.
  • MJB wrote: »
    I am well aware, that 16bits is not the best you can do these days. The question was about what is possible without external hardware. Cogs own multiplier is limited to 16*16bits too, if you exeed this, P2 is somewhat single core....

    what do you mean by this?

    every COG can use the CORDIC multiplier at the same time ...
    and each of those is pipelined in itself.

    I thought, that instead of 2 clocks for every cog instruction including 16*16bit cog multiply, you need 55 clocks plus delay for the start. If it is really true, that the cordic can do maths for each cog in parallel (has it got a higher clock speed?), then a very clever compiler might use the 54 stage pipeline, if all operations are independant.
  • potatoheadpotatohead Posts: 9,786
    edited 2019-08-15 - 15:47:29
    They are. And it just has a deep enough pipe. Cogs cannot make simultaneous requests, due to the CORDIC being a HUB device. That is why it can work like it does.
    Do not taunt Happy Fun Ball! @opengeekorg ---> Be Excellent To One Another SKYPE = acuity_doug
    Parallax colors simplified: https://forums.parallax.com/discussion/123709/commented-graphics-demo-spin<br>
Sign In or Register to comment.