The formidable FPGAs (>10k cells) come with the very formidable price (starting at the order of $100 and growing exponentially with the number of cells). Building any of the hyped applications of FPGA will probably cost hundreds if not thousands. With so much fuss about doing the signal processing in FPGAs, FPGA development tools, FPGA student projects, zillions of articles, etc, I am wondering if anyone actually used FPGA to make a sellable product. Of course, there could be the applications like radars or cellular base stations or prototyping CPUs or ASICs. However this is a very special and narrow nitche. Then why all that commotion about FPGAs? Just because it is the cutting edge of the technology? Vladimir Vassilevsky DSP and Mixed Signal Design Consultant http://www.abvolt.com
Anyone actually using FPGA?
Started by ●October 21, 2008
Reply by ●October 21, 20082008-10-21
On Oct 21, 12:33�pm, Vladimir Vassilevsky <antispam_bo...@hotmail.com> wrote:> The formidable FPGAs (>10k cells) come with the very formidable price > (starting at the order of $100 and growing exponentially with the number > of cells). Building any of the hyped applications of FPGA will probably > cost hundreds if not thousands. > > With so much fuss about doing the signal processing in FPGAs, FPGA > development tools, FPGA student projects, zillions of articles, etc, I > am wondering if anyone actually used FPGA to make a sellable product. > > Of course, there could be the applications like radars or cellular base > stations or prototyping CPUs or ASICs. However this is a very special > and narrow nitche. > > Then why all that commotion about FPGAs? Just because it is the cutting > edge of the technology? > > Vladimir Vassilevsky > DSP and Mixed Signal Design Consultanthttp://www.abvolt.comNo, not here. I suspect a lot of what you hear and see has to do with marketing propeganda more than anything. Have you seen the embedded systems design magaize for the last year or so? It would have you believe that the majority of all new designs are 'system on a chip', all in a customizable FPGA. The majority of the designs I have worked on for about the past 8 years have all had a DSP and an FPGA, but the designs were partitioned so that the FPGA was used mostly for implementing various forms of glue logic or for off loading simply state-machine type tasks that could be run concurrently in real time rather than burden software with busy loops and hardware polling. Unless PCB real-estate is so tight that you MUST get everything in one IC (which has never been the case in any application I have worked on), I can't see justifying the cost/
Reply by ●October 21, 20082008-10-21
On Oct 21, 12:33�pm, Vladimir Vassilevsky <antispam_bo...@hotmail.com> wrote:> The formidable FPGAs (>10k cells) come with the very formidable price > (starting at the order of $100 and growing exponentially with the number > of cells). Building any of the hyped applications of FPGA will probably > cost hundreds if not thousands. > > With so much fuss about doing the signal processing in FPGAs, FPGA > development tools, FPGA student projects, zillions of articles, etc, I > am wondering if anyone actually used FPGA to make a sellable product. > > Of course, there could be the applications like radars or cellular base > stations or prototyping CPUs or ASICs. However this is a very special > and narrow nitche. > > Then why all that commotion about FPGAs? Just because it is the cutting > edge of the technology? > > Vladimir Vassilevsky > DSP and Mixed Signal Design Consultanthttp://www.abvolt.comHello Vladimir, A few years ago I did a project for a defense contractor that required an FPGA and I produced multiple copies of the hardware for them. Basically the widgits served as specialized forms of laboratory equipment. I was processing data in the gigahertz range. And $100 for a chip was not even a concern - what I used was quite a bit more. For me the issue was getting ball grid array parts mounted onto a board. After the initial tests were sucessful, we did a redesign that was only for form fitting reasons. What they did after - I'm not privy to. Clay p.s. There was a guy who used to post here a lot, who seem to have almost all of his projects in FPGAs.
Reply by ●October 21, 20082008-10-21
On Oct 21, 12:33�pm, Vladimir Vassilevsky <antispam_bo...@hotmail.com> wrote:> The formidable FPGAs (>10k cells) come with the very formidable price > (starting at the order of $100 and growing exponentially with the number > of cells). Building any of the hyped applications of FPGA will probably > cost hundreds if not thousands. > > With so much fuss about doing the signal processing in FPGAs, FPGA > development tools, FPGA student projects, zillions of articles, etc, I > am wondering if anyone actually used FPGA to make a sellable product. > > Of course, there could be the applications like radars or cellular base > stations or prototyping CPUs or ASICs. However this is a very special > and narrow nitche. > > Then why all that commotion about FPGAs? Just because it is the cutting > edge of the technology? > > Vladimir Vassilevsky > DSP and Mixed Signal Design Consultanthttp://www.abvolt.comQ: "Anyone actually using FPGA?" A: Yes See http://en.wikipedia.org/wiki/Field-programmable_gate_array for a little broader view of applications Ken.
Reply by ●October 21, 20082008-10-21
Noway2 wrote:> On Oct 21, 12:33 pm, Vladimir Vassilevsky <antispam_bo...@hotmail.com> > wrote: > >>The formidable FPGAs (>10k cells) come with the very formidable price >>(starting at the order of $100 and growing exponentially with the number >>of cells). Building any of the hyped applications of FPGA will probably >>cost hundreds if not thousands.>>Then why all that commotion about FPGAs? Just because it is the cutting >>edge of the technology? >> > I suspect a lot of what you hear and see has to do with > marketing propeganda more than anything. Have you seen the embedded > systems design magaize for the last year or so? It would have you > believe that the majority of all new designs are 'system on a chip', > all in a customizable FPGA.Exactly. The amount of hype seems to be disproportionally high compared to the amount of the real applications.> The majority of the designs I have worked on for about the past 8 > years have all had a DSP and an FPGA, but the designs were partitioned > so that the FPGA was used mostly for implementing various forms of > glue logic or for off loading simply state-machine type tasksThis is what is my experience, too. However the above mentioned application usually takes only few hundreds of gates or so. Nothing like the hyped 10k+ cell arrays with the built in multipliers and FFTs. Vladimir Vassilevsky DSP and Mixed Signal Design Consultant http://www.abvolt.com
Reply by ●October 21, 20082008-10-21
Ken Asbury wrote:> On Oct 21, 12:33 pm, Vladimir Vassilevsky <antispam_bo...@hotmail.com> > wrote: > >>The formidable FPGAs (>10k cells) come with the very formidable price >>(starting at the order of $100 and growing exponentially with the number >>of cells). Building any of the hyped applications of FPGA will probably >>cost hundreds if not thousands. >> Q: "Anyone actually using FPGA?" > A: Yes > > See http://en.wikipedia.org/wiki/Field-programmable_gate_array for > a little broader view of applicationsName a particular product. VLV
Reply by ●October 21, 20082008-10-21
Vladimir Vassilevsky <antispam_bogus@hotmail.com> wrote in news:uvnLk.4829$yr3.3451@nlpi068.nbdc.sbc.com:> > The formidable FPGAs (>10k cells) come with the very formidable price > (starting at the order of $100 and growing exponentially with thenumber> of cells). Building any of the hyped applications of FPGA will probably > cost hundreds if not thousands. > > With so much fuss about doing the signal processing in FPGAs, FPGA > development tools, FPGA student projects, zillions of articles, etc, I > am wondering if anyone actually used FPGA to make a sellable product. > > Of course, there could be the applications like radars or cellular base > stations or prototyping CPUs or ASICs. However this is a very special > and narrow nitche. > > Then why all that commotion about FPGAs? Just because it is the cutting > edge of the technology? > > > Vladimir Vassilevsky > DSP and Mixed Signal Design Consultant > http://www.abvolt.com >It can be pretty useful as a step in deployment-- somewhere between having your algorithms in software and on hardware, and can be your first "stand alone" prototype. Not quite the final product, but it has its uses. I think from the FDA perspective, you can treat FPGA as "black box", which can make life easier. -- Scott Reverse name to reply
Reply by ●October 21, 20082008-10-21
Vladimir Vassilevsky wrote:> > The formidable FPGAs (>10k cells) come with the very formidable price > (starting at the order of $100 and growing exponentially with the number > of cells). Building any of the hyped applications of FPGA will probably > cost hundreds if not thousands. > > With so much fuss about doing the signal processing in FPGAs, FPGA > development tools, FPGA student projects, zillions of articles, etc, I > am wondering if anyone actually used FPGA to make a sellable product. > > Of course, there could be the applications like radars or cellular base > stations or prototyping CPUs or ASICs. However this is a very special > and narrow nitche. > > Then why all that commotion about FPGAs? Just because it is the cutting > edge of the technology? > > > Vladimir Vassilevsky > DSP and Mixed Signal Design Consultant > http://www.abvolt.com >One of my clients makes IR imaging systems that make heavy use of the high-end FPGAs for video processing -- but just the imaging portion of the system costs over $50K, and their production numbers are a few hundred a year. I have another client who is currently designing a fairly high-end software defined radio that will use FPGAs to shorten the design cycle. So yes, they do get used. But I suspect that the over-$100 parts are all in one niche market or another -- for that price, if you're going to make it for the consumer market you have a clear and obvious reason to go to an ASIC. I'm more curious about the $20 parts like Xilinx's Spartan. -- Tim Wescott Wescott Design Services http://www.wescottdesign.com Do you need to implement control loops in software? "Applied Control Theory for Embedded Systems" gives you just what it says. See details at http://www.wescottdesign.com/actfes/actfes.html
Reply by ●October 21, 20082008-10-21
On Tue, 21 Oct 2008 11:33:43 -0500, Vladimir Vassilevsky <antispam_bogus@hotmail.com> wrote:> >The formidable FPGAs (>10k cells) come with the very formidable price >(starting at the order of $100 and growing exponentially with the number >of cells). Building any of the hyped applications of FPGA will probably >cost hundreds if not thousands. > >With so much fuss about doing the signal processing in FPGAs, FPGA >development tools, FPGA student projects, zillions of articles, etc, I >am wondering if anyone actually used FPGA to make a sellable product. > >Of course, there could be the applications like radars or cellular base >stations or prototyping CPUs or ASICs. However this is a very special >and narrow nitche. > >Then why all that commotion about FPGAs? Just because it is the cutting >edge of the technology? > > >Vladimir Vassilevsky >DSP and Mixed Signal Design Consultant >http://www.abvolt.comYes, lots of folks use FPGAs in real-live products. Abineau (our company) does pretty much 95% of its work in FPGAs to real-live product targets. You're absolutely right that FPGAs are an expensive solution compared to, say, standardized silicon. The problem is that if you can't find silicon that does what you need, or you can't afford to have a silicon vendor hold your product line hostage (via obsolescence, product shortages, whatever) then you need to do something else. The cost to produce one's own silicon is usually prohibitively high unless one is making very large quantities, and often people want more flexibility (for product roadmap adjustments) than they can get with silicon. So FPGAs provide: 1) Less risk by providing product flexibility, shorter design cycles, and much shorter debug cycles than silicon. 2) Much lower development cost and shorter development cycle than custom silicon. 3) Much greater flexibility than off-the-shelf or custom silicon. 4) Higher part cost than silicon. 5) Higher power consumption than silicon. The advantages in 1-3 often far outweigh the disadvantages in 4 and 5. This does mean that FPGAs are rare in very high volume or low power applications, but there are a lot of other applications out there. Generally our customers make FPGA-based platforms either for internal use or for product runs with volumes in the 10s to 10,000s. The products are typically not consumer items but either professional or safety or whatever applications that can bear the additional cost. Rack mounted equipment is common. Attend an industry show sometime that has a lot of fairly expensive equipment boxes, like NAB or any of the Comm shows. Look at what shows Xilinx and Altera are displaying at. Go to those. You'll see a number of products with FPGAs inside. If you want more evidence, get an annual report from Xilinx or Altera. Somebody's buying those things. Lots of 'em. Eric Jacobsen Minister of Algorithms Abineau Communications http://www.ericjacobsen.org Blog: http://www.dsprelated.com/blogs-1/hf/Eric_Jacobsen.php
Reply by ●October 21, 20082008-10-21
On Oct 22, 1:53�am, Eric Jacobsen <eric.jacob...@ieee.org> wrote:> > If you want more evidence, get an annual report from Xilinx or Altera. > Somebody's buying those things. �Lots of 'em.Its more likely that lots of people are buying handfuls each. Presumably these things are fabbed somewhere like TSMC, in similar processes to advanced graphics chips. If the volumes are big enough its hard to see why a big FPGA should be more expensive than a big graphics chip. I expect volume is the key issue in their pricing. The only thing I can think of made in pretty big numbers, and able to bear the cost, is cellular base stations. I'd guess medical equipment is the next biggest.
