Site Home Archive Home FAQ Home How to search the Archive How to Navigate the Archive

Compare FPGA features and resources

Threads starting:

Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Custom Search

On Wed, 23 Feb 2005 23:27:08 -0800, Rob Barris <rbarris@mac.com> wrote: > >I have a simple binary FSK demodulator circuit based on the old 74HC297 >DPLL, a divide-by-4 counter, and a flip flop. I'm interested in seeing >how many such circuits I could pack into a cheap FPGA like one of these: > >http://www.fpga4fun.com/board_pluto.html > (uses Altera EP1K10TC100) > >http://www.altera.com/products/devices/acex/acx-index.html > >but I'm not sure where to start. I'm just trying to figure out a >ballpark number, i.e. "one copy will just barely fit" or "seems like you >should be able to get five copies in there with room to spare". > >Is this a common kind of problem - translating old-school discrete chip >designs into FPGA? I did a lot of googling for resources along this >vein but didn't come up with much. > >At the low price of the fpga4fun board, it's almost an impulse buy, but >I wanted to ask the newsgroup first and mull it over. > >Rob 297 contains approximately 50 flops and associated logic. As a very rough approximation, you should be able to implement one 297 and your divide by 4 counter in 100-150 LEs. A 10K100 has ~5K LEs so you're looking at ~ 30 of your blocks. But of course it's late and it's been a while since I got any coffee in my system...

On Wed, 23 Feb 2005 23:27:08 -0800, Rob Barris <rbarris@mac.com> wrote: > >I have a simple binary FSK demodulator circuit based on the old 74HC297 >DPLL, a divide-by-4 counter, and a flip flop. I'm interested in seeing >how many such circuits I could pack into a cheap FPGA like one of these: > >http://www.fpga4fun.com/board_pluto.html > (uses Altera EP1K10TC100) > >http://www.altera.com/products/devices/acex/acx-index.html > >but I'm not sure where to start. I'm just trying to figure out a >ballpark number, i.e. "one copy will just barely fit" or "seems like you >should be able to get five copies in there with room to spare". > >Is this a common kind of problem - translating old-school discrete chip >designs into FPGA? I did a lot of googling for resources along this >vein but didn't come up with much. > >At the low price of the fpga4fun board, it's almost an impulse buy, but >I wanted to ask the newsgroup first and mull it over. Beware of performing a naive (i.e. literal) translation of your SSI design into a programmable part. Often designs using individual chips do things which don't work very well in programmable logic. Some of these are: 1. Gated clocks (to implement clock enables). 2. Using async resets on flip flops and counters to implement logic functions. 3. Hazard coverage. (Usually the tools will optimise out the "unnecessary" logic.) It's usually a trivial matter to turn these into a more FPGA or CPLD friendly design. This shouldn't affect the gate count / logic cell count much though. Regards, Allan

>I have a simple binary FSK demodulator circuit based on the old 74HC297 >DPLL, a divide-by-4 counter, and a flip flop. I'm interested in seeing >how many such circuits I could pack into a cheap FPGA like one of these: >but I'm not sure where to start. I'm just trying to figure out a >ballpark number, i.e. "one copy will just barely fit" or "seems like you >should be able to get five copies in there with room to spare". > >Is this a common kind of problem - translating old-school discrete chip >designs into FPGA? I did a lot of googling for resources along this >vein but didn't come up with much. For a rough cut, count FFs. THen divide that into the number from the data sheet. Most modern FPGAs have a basic unit that is a clump of logic and a FF. The "clump of logic" is typically a 4 input LUT (Look Up Table). It can implement any function of 4 inputs. If your design/problem doesn't have a lot of logic between the FFs, there is a good chance that all or most of it will fit in that 4 input LUT. So counting FFs works out to be a good rough estimate. -- The suespammers.org mail server is located in California. So are all my other mailboxes. Please do not send unsolicited bulk e-mail or unsolicited commercial e-mail to my suespammers.org address or any of my other addresses. These are my opinions, not necessarily my employer's. I hate spam.

Hello friends I am implementing a design which has about 250 32-bits unsigned numbers which have to be added to give the final outcome, since it is a very large number of elelments I think implementing this operation on FPGA would be very slow (250 cascade adders), I am thinking in using Carry-Save adders scheme to avoid the long propagation times due to the carry, but I am not sure now if it would really help. I need to carry out this additions in the shortest possible time, Carry-Save arithmetic would be helpful? is there any other scheme I can use to speed up the addition? (at some point I will need to increase the number of quantities to be added far above 250). Many Thanks

About the original message multiple additions, I forgot to add that I'm currentily using a Virtex II Pro 2vp100 device. Thanks

On 24 Feb 2005 03:30:50 -0800, "blackduck" <rg66@le.ac.uk> wrote: >Hello friends > >I am implementing a design which has about 250 32-bits unsigned numbers >which have to be added to give the final outcome, since it is a very >large number of elelments I think implementing this operation on FPGA >would be very slow (250 cascade adders), I am thinking in using >Carry-Save adders scheme to avoid the long propagation times due to the >carry, but I am not sure now if it would really help. I need to carry >out this additions in the shortest possible time, Carry-Save arithmetic >would be helpful? is there any other scheme I can use to speed up the >addition? (at some point I will need to increase the number of >quantities to be added far above 250). > >Many Thanks How fast do you need the result and how fast are these numbers coming in ? You're presenting your problem as if someone is giving you all these 250 numbers in cycle n and wants the sum in cycle n+1. If the numbers are coming one at a time and you need the sum after number 250, the problem is quite different. Please tell us little bit more on your data (speed, format etc).

blackduck wrote: > About the original message multiple additions, I forgot to add that I'm > currentily using a Virtex II Pro 2vp100 device. > > Thanks [...] > I need to carry out this additions in the shortest possible time, Hi BlackDuck, Is pipelining an option ? always adding 2 32..38 bit numbers (depending of the stage within your pipeline) leads to 7 Pipeline-Stages to add 256 numbers !!! - 100 MHz - one Result on every clock - with an inital latency of 7 clocks should be possible in a Virtex2Pro... Cheers Jochen

Hi All, Not sure who can do this, but I think this extra info for FPGA-FAQ entry 0044 will help. For the SuSE distro, adding the two include directories to CFLAGS works if you place them at the start. So, it might be clearer to have this in the FAQ: * on Debian, just run make. On SuSE: edit makefile -change KERNEL_DIR: KERNEL_DIR = /usr/src/linux/ -modify CFLAGS: i) delete all -I flags ii) add to start of CFLAGS -I/usr/src/linux/include iii) add next in CFAGS -I/usr/src/linux/include/asm-i386/mach-default/ I originally placed them elsewhere and compilation failed. Hope this helps someone Andy

john, Thanks a lot of unbiased information. It really is helpful to me.

Many thanks for your response, Actually the problem is as you said, 250 filters are giving me at time n 250 different values, then i need to get the addition of those 250 values at time n+1, and this new value conforms an input for a comparator. Each data is a 32 bits unsigned integer, which is generated by a lowpass filter at 300MHz, is in fact the impulse response of such a filter.

Thanks Jochen, Well, I can use the arrangement you propose to speed up the process, which solves part of the problem, about the pipeline, I am not sure if it will work since the design uses 250 filters which gives the 250 different values which have to be added to generate a single value. As you said, each two of them can be added in 7 stages to generate the final addition, but as I know a very little about pipeline I cannot imagine how to implement it. Can you explain me a little how the implementation would look like please?

well you make a additionner that add 2 number you put 125 additionner in parrallel to add your 250 number next level you use 63 additionner to add the 125 result of the first level and you do that till you have only 1 result so in final you will have 125 + 63+ 32+16+8+4+2+1=251 adders and don't forget that ouput of each level is 1 bit more than input but another thing is that your input don't arrived all at the same cycle clock (or does virtex 2P have 8000 input pin??) so you could make an accumulator imagine one data arrive at each cycle clock, you add your new data to accumulator untill you have added your 250 data( = 250clock cycle) "blackduck" <rg66@le.ac.uk> a écrit dans le message de news: 1109249301.040661.131480@l41g2000cwc.googlegroups.com... > Thanks Jochen, > > Well, I can use the arrangement you propose to speed up the process, > which solves part of the problem, about the pipeline, I am not sure if > it will work since the design uses 250 filters which gives the 250 > different values which have to be added to generate a single value. As > you said, each two of them can be added in 7 stages to generate the > final addition, but as I know a very little about pipeline I cannot > imagine how to implement it. Can you explain me a little how the > implementation would look like please? >

So your sample frequency is what?? 300MHz?? "blackduck" <rg66@le.ac.uk> a écrit dans le message de news: 1109248428.297754.45730@g14g2000cwa.googlegroups.com... > Many thanks for your response, > > Actually the problem is as you said, 250 filters are giving me at time > n 250 different values, then i need to get the addition of those 250 > values at time n+1, and this new value conforms an input for a > comparator. Each data is a 32 bits unsigned integer, which is generated > by a lowpass filter at 300MHz, is in fact the impulse response of such > a filter. >

"willie CHEN" <changewhere@126.com> wrote in message news:cvjr1m$1tb$1@mail.cn99.com... > Hi, > In my project,there's a xilinx IP core. I want to use the synplify7.7 to > synthesize it, but there's a warning when synthesize . > The warning is : > @W: CD280 :"G:\project\itu656_dec\itu656_Decoder.vhd":29:10:29:19|Unbound > component counter_11 mapped to black box > @W: CD280 :"G:\project\itu656_dec\itu656_Decoder.vhd":37:10:37:18|Unbound > component counter_4 mapped to black box > These warnings can be ignored. The netlist (edf, ngo, etc) for the core will be picked up when you run ngdbuild. "-sd" option of ngdbuild may be helpful. HTH, Jim jimwu88NOOOSPAM@yahoo.com (remove capital letters) http://www.geocities.com/jimwu88/chips

statepenn99 wrote: > LSI: Reasonable NRE and pretty low minimum volumes. Logic density is > adequate, but it has very little embedded memory. I would be taking a > hard look at LSI, if I didn't need so much RAM. Was that the new LSI RapidChip Integrator2? It has max. 5.6Mgates and 8-9Mbits of SRAM. That SRAM amount is comparable to EP2S180 (9.3Mbit). --Kim

"statepenn99" <statepenn99@yahoo.com> writes: > EasyPath: > Hardcopy II: > LSI: Does anybody have any experience with the NEC ISSP? How it compares to RapidChip? Petter -- A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing on usenet and in e-mail?

rickman wrote: > > I have a known-good initial bitstream in a configuration Flash (the > > 'boot' image). After that I disable it and I read data from a memory > > card, including the other bitstreams. > > Do you require the FPGA to be running when you load the other bitstreams. It is the only way I will be able to load bitstreams from the memory card and do a couple of other things at the same time. > APP290 does not apply to S3 yet. What docs talk about PR in S3s? At least the datasheet mentions it (regarding the SelectMAP interface). Also XAPP462 refers to it in the bitgen generation section. An there is the #18416 Answer Record... Cheers. -- PabloBleyerKocik /"But what... is it good for?" pbleyer / -- 1968 Engineer at IBM's Advanced Computing @embedded.cl / Systems Division, commenting on the microchip

Hello Peter. Peter Alfke wrote: > Do I get a ticket for the Love Boat also? > Outside cabin, next to the swimming pool, good food, babes, sunshine, > booze... > Man, I might not even miss this newsgroup after a while... > Peter Alfke I was going to suggest this also, but I couldn't remember any alter ego of you in Altera. But if you don't mind a m=E9nage =E0 trois... sure, I can add you to the list :D Thanks for your good humor. Regards. PS: Perhaps you can help me with my Spartan-3 partial reconfiguration issue above... ;) -- /"It would appear that we have reached the limits of PabloBleyerKocik/ what it is possible to achieve with computer technology, pbleyer / although one should be careful with such statements, as @embedded.cl / they tend to sound pretty silly in 5 years."-J.von Neumann

newman5382 wrote: > "nospam" <nospam@nospam.invalid> wrote in message > news:dq2n015ge9pgefno5jl0rlod7d7aao9m74@4ax.com... > > Alex <uksb@greenbank.org> wrote: > > > >>Alex <uksb@greenbank.org> wrote: > >>[ SNIP original request ] > >> > >>Thank you for all of the replies. I've ordered directly > >>from Xilinx and I'm looking forward to my new toy > >>arriving. > > > > Did you or anyone else get the now bundled EDK evaluation (and 6.3i ISE > > evaluation I presume because the 6.3i EDK appears to need it)? > > > > I have not got my EDK either. The Spartan III starter kit came with the ISE > eval CD's I emailed the online store on Feb 07, 2005 and got the same > answer as Carsten did on Jan 30, 2005 : > > "quote" > The EDK eval is being sent to all who have purchased the Spartan 3 > Starter kit and will begin shipping within the next couple of weeks. > > "unquote" > > I replied : > > On or before Jan 30, 2005, you gave the same exact answer to another > customer.My question is, has your schedule slipped. > > > The answer on Feb 08, 2005 was : > The only information I have at this time is what was provided to us by our > Marketing group. If possible, please send me your ship to address and ship > to address of the other customer.I will then forward that information > directly to the department responsible for the eval. > > I replied back on Feb 08, 2005 with the information requested. I have yet > to receive an answer back as of Feb 10, 2005. > > - Newman I heard the EDK 6_3 Eval CD's have started to ship. - Newman

"KCL" <kclo4_NO_SPAM_@free.fr> wrote in message news:421dd0cd$0$846$8fcfb975@news.wanadoo.fr... > well you make a additionner that add 2 number > you put 125 additionner in parrallel to add your 250 number > next level you use 63 additionner to add the 125 result of the first level > and you do that till you have only 1 result > so in final you will have 125 + 63+ 32+16+8+4+2+1=251 adders > and don't forget that ouput of each level is 1 bit more than input > > but another thing is that your input don't arrived all at the same cycle > clock (or does virtex 2P have 8000 input pin??) so you could make an > accumulator > imagine one data arrive at each cycle clock, you add your new data to > accumulator untill you have added your 250 data( = 250clock cycle) Might be worth looking at the filters as well - what are they? If they're an FIR filter array then is it possible to do some of the additions in the MAC array? Like you I suspect that there is some data folding somewhere - at the moment this device has an 80 Giga samples/second input - should be able to make the coffee as well...

Hi KCL, The design only receives 250 bits as inputs which are the incoming impulses to lowpass filters (250), these filters produce an output which has to be added to get the total response, then this total response in sent to a comparator, indication if the total response is greater to some value. Then in the first clock cycle, the comparator has zero as input from the filters, the second clock cycle, the comparator receives the total response from the filters and compares this against a value, and this process should be repeated each clock cycle. The device is able to run at 300 MHz, but the time required to get the total response is of course slower than this (plus internal delays), therefore I am trying to carry this process as fast as possible. The design has 250 bits as inputs and 1 bit output. Sorry if initially sounded as a 8000 input design. Thanks

Hi everyone.. I wonder how to build a prescaled counter! That is a counter that can be prescaled by fx. a factor 1.000.000 or less! Any nice ideas how to do this? Thanks Preben

Hi All, I'm kinda new in the business so this question may be very rookie. Anyways, I was wondering if all synchronous designs need to be state machine based designed or the synchronous behaviour can be modelled using multiple clocked processes like always (in Verilog) blocks activated by clock to do the same thing. Thanks morpheus

Hi, When using Xilinx, the SRL16 is a very good candidate for prescaler implementation. Göran Preben Holm wrote: > Hi everyone.. > > I wonder how to build a prescaled counter! That is a counter that can be > prescaled by fx. a factor 1.000.000 or less! > > Any nice ideas how to do this? > > > Thanks > Preben

I finally got around to watching the "Implementing Multi-Processor Systems in FPGAs" TechOnLine Webcast: http://seminar2.techonline.com/s/altera_feb1005 http://seminar2.techonline.com/~additionalresources/altera_feb1005/altera_Feb10_slides_edited.pdf Based on this seminar and my own imagination, I can envision quite a lot of potential usage models. However, I was wondering to what level folks are _actually_ using multiple soft-core processors in an FPGA for their commercial, academic, and/or personal projects right now. What is the overall architecture--how many processors are used? How do the processors coordinate their activities? How is data processing distributed across them? Is the code/data stored in on-chip memory or externally? Thanks. Paul

Site Home Archive Home FAQ Home How to search the Archive How to Navigate the Archive

Compare FPGA features and resources

Threads starting:

Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Custom Search