OpenRISC ASIC chip fundraiser

in News, open source, site by the machinegeek | 12 comments

asic_block
January 18 is OpenRISC day and the OpenRISC project currently is conducting a fundraiser.

We are raising $250,000 to build the worlds first truly open source system on chip processor. Help make this happen by donating to the project today.

Money raised will be used to fabricate a batch of OpenRISC SOC ASIC chips that will be available for sale to individuals and product manufacturers.

The processor will be affordable, and competitive in mobile and embedded applications. The 32 bit RISC chip will run at an estimated 250Mhz, support Video, DDR2 Memory, Gigabit Ethernet, USB, Audio, Serial, PCI and run the Linux operating system. Individual processors will be sold for an estimated $5 per unit.

Visit the OpenRISCday webpage for more info or to donate. To read more on the OpenRISC ASIC chip see their page at Opencores, or download an 8 page PDF of the chip’s proposed specs.

Olof Kindgren via the contact form.

This entry was posted in News, open source, site and tagged , , .

Comments

  1. Kris Lee says:

    Why not DDR3?

  2. fonz says:

    apart from the feel-good of being open source what would this offer?
    Something like an Allwinner A10 or the numerous other Cortex A8, are 5x the speed, use ddr3 and
    have a gpu and is about the same price afaict

    • Tsvetan says:

      like any open source technology this offers manufaturer independance
      now you can buy A10 but in 6-12 month there will be A20 and you will not be able to buy A10 in another 12 months you will be able to buy only A30 instead of A20 and so on
      Allwinner do not have even proper documentation and datasheet so any comparison is unappropiate
      ARM licensees start from $2 millions
      here you will have access to everything which gives you *freedom*
      sure the specs are not great but I guess this is just the start and Olimex will definitely make OLinuXino with this SoC to support the open technology with low cost development board

  3. Michal says:

    You probably don’t understand it. The allwinner and cortex a8 is proprietary technology.

    OpenRISC is open. As is you can download the sources and compile your own cpu.

    • fonz says:

      I understand but I just don’t see the point ARM sell an A8 design to anyone willing to pay so it has become the standard and pretty much all manufacturers have their version at a competitive price

      sure you can “compile” you own OpenRISC and then what? put it in an FPGA, 1/10 performance, 10x power, 10x price of an ARM

      or you get some chip build, but there’s a reason they need $250K it is extremely expensive to get started, I’m surprised they can get a mask set for that price

  4. Alan says:

    All that speed, and the USB is still only 1.1 compatible. No high-speed USB 2.0, no USB 3 speeds. DDR2, not DDR 3… No, this will not be the “fastest” uber-chip out there.
    Going for a low pin-count, hopefully it won’t be a BGA either.
    Big hopes: underclocking, disable segments as required, and go for lowest power consumption.

  5. Tiersten says:

    Everybody seems to be missing the point. Everything is available as source from the CPU core itself to every peripheral attached to it. If you just licensed the latest/greatest ARM core from ARM then you’d need to pay a large sum of money and then end up with something you can’t redistribute in source form. If you just want ARM then buy one from the many manufacturers that offer one.

    Those specific peripherals are selected by that is what is available as a freely usable core with source from opencores. There is a USB 2.0 controller at opencores but it isn’t complete and hasn’t been tested in hardware at all.

    As for whether this will actually be a useful product or not? I’ve no idea. I’ve seen other ASICs out there using freely available CPU cores such as the LEON SPARC core so it isn’t too outlandish to suggest a general purpose OpenRISC based SoC.

    • fonz says:

      I think everyone gets the point, it is open source. but open source for something like a CPU isn’t nearly as useful as open source software where everyone can produce it for “free”

    • Alan says:

      Economy Of Scale rears its ugly head…
      Say you make 500,000 (half a million) chips. It costs you $1million, plus anoth $500,000 in IP license fees. Throw in distribution etc. and you need to charge at least $4 per chip in order to show a profit. Opn source cuts out the IP fees, dropping the profit point to $3 per chip.

      Problem is – the ARM companies have a supported product, and their product runs are much bigger than half a million. Once the process is set up, they could produce 50 million chips for (say) $3 million. Even throwing in their IP License fess, they can make a profit by selling those chips at a dollar.

      Open source is nice, but hardware overheads are unlikely to let it be profitable. License the design (free) to TI or NXP and it might be a different story.

  6. David says:

    Who cares if this is “Open Source”? It would cost you hundreds of thousands of dollars to reproduce it if you tried to contribute by making even the slightest change to the design. Compare this to Open Source “Soft Cores” on programmable logic (e.g. FPGA’s); that’s something you can change then realize for next to nothing – contributing back in the process of-course…

    Anyway – if you’re going to bother making a SoC in silicon these days, please include SDRAM support and an MMU. VGA out is also sorely missed on the likes like ARM SoC’s. VGA is still very viable today and will remain so for the foreseeable future Wold-wide. The fact is that the majority of people on Earth (outside the first-World) are far more likely to have affordable access to a VGA compatible display device than an HDMI compatible device, especially HDMI with HDCP abstraction.

  7. Chuckt says:

    Why not put this project up on Kickstarter? I want to be guaranteed something in return.

    • Berry says:

      Yeah. At least it will require making a viable project. For the time being they’ve got $20k / 2years, so that 250k goal will be reached by 2036.

Leave a Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Notify me of followup comments via e-mail. You can also subscribe without commenting.