Hi everyone, I would like to count very fast (100MHz) using an 19-bit counter in spartan 3 - any ideas how to make such a fast counter! One idea is to use the Chinese Remainder Theorem to count, but that make's a "hit-rate" of only 60% or so, if the timing is restricted. The other Idea is to make a pipelined counter, but that make's the whole design a lot more complicated. Any other Ideas? I use the Spartan 3 digilent starter board! Thanks in advance! / Preben
Fast counting in Spartan 3
Started by ●February 13, 2005
Reply by ●February 13, 20052005-02-13
Preben, have you done the timing analysis on a straightforward implementation? I have run longer counters at higher frequency in Spartan 3...There is a built-in carry structure! Is your counter loadable? if not, your counter might be implemented as a ripple ( or a partially-ripple) counter, then you can count several hundred MHz ( use the Q as clock source for the next slice, or next group of 4 flip-flops.) If you do this, you have to wait out the ripple delay before you use or decode the counter output. Thre are many ways to achieve 100 MHz, or 2x, or 4x that performance, it all depends on your real requirements. Peter Alfke, Xilinx Applications.
Reply by ●February 13, 20052005-02-13
"Preben" <64bitNOspamNO@mailme.dk> wrote in message news:420fab6f$0$12714$edfadb0f@dread14.news.tele.dk...> Hi everyone, > > I would like to count very fast (100MHz) using an 19-bit counter in > spartan 3 - any ideas how to make such a fast counter! > One idea is to use the Chinese Remainder Theorem to count, but that make's > a "hit-rate" of only 60% or so, if the timing is restricted. > > The other Idea is to make a pipelined counter, but that make's the whole > design a lot more complicated. >!00 MHz does not sound that fast for a Spartan III. Did you try a HDL synthesys and P&R with time constraint?> Any other Ideas?Perhaps an 8 bit prescaler with a pipelined clock enable to an 11 bit counter. 20 bit counter using block RAM http://www.synplicity.com/literature/syndicated/pdf/spartan_3_v3_i3.pdf Have fun -Newman> > > I use the Spartan 3 digilent starter board! > > > Thanks in advance! > > / Preben
Reply by ●February 13, 20052005-02-13
"Preben" <64bitNOspamNO@mailme.dk> schrieb im Newsbeitrag news:420fab6f$0$12714$edfadb0f@dread14.news.tele.dk...> Hi everyone, > > I would like to count very fast (100MHz) using an 19-bit counter in > spartan 3 - any ideas how to make such a fast counter!Just a straight forward approach. Been there, done this recently. It was a 32 Bit counter, running at 130 MHz and still having some timing margin left. Regards Falk
Reply by ●February 13, 20052005-02-13
Preben <64bitNOspamNO@mailme.dk> wrote:> Hi everyone,> I would like to count very fast (100MHz) using an 19-bit counter in > spartan 3 - any ideas how to make such a fast counter! > One idea is to use the Chinese Remainder Theorem to count, but that > make's a "hit-rate" of only 60% or so, if the timing is restricted.Spartan 3, 100MHz and ... fast? I think you are kidding. Did you try to implement the counter? I guess it will run much faster Bye -- Uwe Bonnes bon@elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------
Reply by ●February 13, 20052005-02-13
Like everyone say, 100MHz is trivial. If you had to go much faster there are many clever ways (depending on how you can compromise). Here are two O(1) counters: 1. Linear feedback shift register. Of course, when you really need the count the conversion back is expensive. 2. Carry-save addition. Maintain the count y as x1-x2, where x1 and x2 are two n-bit integers. To increase y, assign (in parallel) x1 <= x2 ^ ~x1; x2 <= (x2 | ~x1) << 1; This time the conversion back is cheap & trivial. For how and why this works, see page 32 of Knuth's MMIX introduction: http://www-cs-faculty.standford.edu/~knuth/mmix-doc.ps.gz I think I've seen Peter pull some really clever counter tricks (though more FPGA specific). Tommy Uwe Bonnes wrote:> Preben <64bitNOspamNO@mailme.dk> wrote: > >>Hi everyone, > > >>I would like to count very fast (100MHz) using an 19-bit counter in >>spartan 3 - any ideas how to make such a fast counter! >>One idea is to use the Chinese Remainder Theorem to count, but that >>make's a "hit-rate" of only 60% or so, if the timing is restricted. > > > Spartan 3, 100MHz and ... fast? I think you are kidding. Did you try to > implement the counter? I guess it will run much faster > > Bye >
Reply by ●February 13, 20052005-02-13
"Tommy Thorn" <foobar@nowhere.void> wrote in message news:GdTPd.6680$m31.81473@typhoon.sonic.net...> > This time the conversion back is cheap & trivial. For how and why this > works, see page 32 of Knuth's MMIX introduction: > http://www-cs-faculty.standford.edu/~knuth/mmix-doc.ps.gz >I got a 401 on that.
Reply by ●February 13, 20052005-02-13
Tommy Thorn wrote:> Like everyone say, 100MHz is trivial. If you had to go much faster there > are many clever ways (depending on how you can compromise). Here are > two O(1) counters: > > 1. Linear feedback shift register. Of course, when you really need the > count the conversion back is expensive. > > 2. Carry-save addition. Maintain the count y as x1-x2, where x1 and x2 > are two n-bit integers. To increase y, assign (in parallel) > > x1 <= x2 ^ ~x1; > x2 <= (x2 | ~x1) << 1; > > This time the conversion back is cheap & trivial. For how and why this > works, see page 32 of Knuth's MMIX introduction: > http://www-cs-faculty.standford.edu/~knuth/mmix-doc.ps.gzTranscriping URL's by hand and mispelling, grief: http://www-cs-faculty.stanford.edu/~knuth/mmix-doc.ps.gz Sorry. Tommy
Reply by ●February 14, 20052005-02-14
On Sun, 13 Feb 2005 18:02:25 -0800, "Pete Fraser" <pfraser@covad.net> wrote:> >"Tommy Thorn" <foobar@nowhere.void> wrote in message >news:GdTPd.6680$m31.81473@typhoon.sonic.net... > >> >> This time the conversion back is cheap & trivial. For how and why this >> works, see page 32 of Knuth's MMIX introduction: >> http://www-cs-faculty.standford.edu/~knuth/mmix-doc.ps.gz >> > >I got a 401 on that. >Just remove the extra "d" in stanford. I am curious how it got there though. I can't imagine anyone actually typing that url.
Reply by ●February 14, 20052005-02-14
Maybe you've got one of those fancy new layout keyboards, but on my QWERTY version D is next to F. Perhaps the perp pressed two adjacent keys at the same time when they meant to press just oner. Not exactly CSI is it? ;-) Cheers, Syms. "mk" <kal*@dspia.*comdelete> wrote in message news:92f011hp912jnbn8lo06sq1tu4sapga197@4ax.com...> Just remove the extra "d" in stanford. I am curious how it got there > though. I can't imagine anyone actually typing that url. >
