Another FPGA startup met the fate of so many others: Abound Logic is reported to have shut down this week, Wednesday June 2nd, 2010.

Abound logic, previously known as M2000, was founded by three EDA veterans who had previously started Meta Systems. Meta Systems developed the industry’s first emulation system based on custom FPGAs, which was to become the basis for the Abound Logic device after Meta Systems was acquired by Mentor Graphics in May 1996.

Abound logic’s device, the Raptor FPGA, offered in 65nm technology the capacity of 774k CLBs, each containing a 4-input LUT. This made the Raptor the largest FPGA on the market at a time.

Abound logic was in discussions to close another financing round, estimated around $20M, but one of the main tentatively new investor pulled out, which put to rest the last investment opportunities.

Tags:

7 Comments on RIP Abound Logic

  1. Christopher Judge says:

    Olivier,

    What do you think? Was the 65nm a hopeless process or was there an issue on the management of the company itself?

  2. Their 65nm device, although delivered later than expected (nothing unusual), worked perfectly fine. It was the largest device available on the market. Abound planned to go to the next node, but obviously this will not happen now.

    I really cannot comment on the company management for two reasons: first reason is that I don’t have any insight on the management, at least nothing that is not public; second reason is that if I had inside info, I wouldn’t disclose it :)…

    Only think I can add is that this came as a surprise, at least to me, since word on the street a couple of weeks ago was pretty positive about Abound closing its next financing round.

    As I pointed out in the past, I think one of the problem is that having a 2x capacity/density value proposition in FPGA may not be sufficient. It’s relatively easy for Xilinx and Altera to eventually catch up –they just need to go to the next node, which they can do quickly and efficiently, as it is a pretty robust and mature procedure for them. Best case a 2x capacity will get you a narrow high-end market, from which you can live for a while. But it is unclear that it is sufficient to sustain another FPGA vendor. Low power (SiliconBlue) or low-cost FPGA-to-ASIC migration (TierLogic) are differentiations that look more solid.

  3. Christopher Judge says:

    Olivier,

    How can it be possible that the 65nm worked just fine? There were companies that were more than desperate to see Abound Logic succeed with a working chip.
    I know at least one that publicly announced a system called Pico Computing. But then I understand that they had no choice than to swap their FPGA to Altera or Xilinx.
    I also know of other companies that were willing to do whatever it took to have a non Altera-Xilinx chip in their system.

    If what you are saying is true, then the implication is that SW was broken and could not deliver. Should we blame SW for the failure of Abound Logic?

  4. You might have more info than me here! I heard several time that the 65nm device worked OK –but again, with some delays in the delivery. As for the software, it is a point that is overlooked by most FPGA startups. In that very case, Abound Logic was working hard to get its SW to exploit the full potential of its HW, and I believe they were on the right track.

    You are correct to state that some companies were willing to do whatever it takes to have a non Altera/Xilinx chip in their system. This was actually one of the biggest selling point of Abound Logic, e.g., for some sensitive applications in the European community. It looks like that was not enough to convince the investors. At the end, I do not know why one of the main investor pulled out. I’m sure we’ll get more on this at some point.

  5. Christopher Judge says:

    My gut feel tells me that they would have gotten investment money if they could show at least 1 working customer design.
    Based on the evidence I gathered (which I can’t disclose for unknown reasons because we are not harming an existing company) the chip NEVER WORKED.

    Here are the main stakeholders. You can check them out at the still running website. Anybody care to highlight what roll each of these people had in the failure of Abound Logic?

    CEO Frederic Reblewski
    VP of SW Thomas Daniel
    VP of HW and else: Olivier Lepape

  6. Let’s put it this way. The VPs of SW/HW are responsible to deliver their products according to the spec, and on schedule. The responsibility of the CEO includes raising money.

    Looks like in Abound Logic’s case, there was no plan B, or at least no contingency plan for the on-going funding negotiation.

    I wouldn’t be surprise if we see some people in France redevelop the technology within the safety of the main tentative investor’s walls…

  7. Christopher Judge says:

    But who would be the people redeveloping the technology?
    All the parties involved proved unable to scale up to a larger organization. They already screwed up way too many relationships. The stakeholders have no business in the semiconductor industry anymore or in any other high tech industry. They are blacklisted for good.
    There is no plan B. Their technology was broken and stale. There is absolutely nothing positive about an architecture that had too little routing and was barely routable, was too slow and too old.
    They had a chance about 2 years ago. Instead they decided to argue with themselves.
    It kind of reminds me of the French soccer national team.

Leave a Reply