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
"Ed McGettigan" <ed.mcgettigan@xilinx.com> schrieb im Newsbeitrag news:4342A6AA.6020500@xilinx.com... > Antti Lukats wrote: > > "Ed McGettigan" <ed.mcgettigan@xilinx.com> schrieb im Newsbeitrag > > > >>We have improved the VGA quality on the upcoming ML405 and ML410 boards > >>that will be released early next year. > >> > >>Ed > > > > > > Hi Ed, > > > > your comment about 405 "to be released early next year" (2006) sounds like > > firm indication that Xilinx has serious problems with 4FX? The ML405 was > > announced no later than jan 2005 (maybe earlier), now you are saying that it > > is coming sometime in 2006 ? Why announce products that are 'maybe' coming > > more than a year later? I would have expect ML405 to be available by now. > > Well if there is really an issue with 4FX that would explain why no FX > > boards are available from Xilinx online shop. Just wondering. > > > > Right now almost all of the FX20 silicon is being shipped to customers > on a priority basis. We do have fully functional ML405 and ML410 boards > in house that we are now making available to our processor specialist > FAEs and available on a limited loaner basis to customers and this will > increase over coming months. > > It takes 3-4 months from this point before we can complete all of the > rest of the material in order to turn it from early access board that > requires a knowledgeable and trained person to use it into a general > availability product with full manufacturing tests, documentation, > reference designs and packaging. Also, our internal processes has > certain requirements that must be met before we place boards in the Xilinx > Online Store, this should happen 1-2 months after the boards are > available from our distributors. > > We haven't formally announced the ML405 and ML410, but we have shown > them at trade shows and forums such as this as an upcoming development > vehicle for Virtex-4. We do this so that our customers and partners can > understand what our plans our for support collateral so that they can > develop their own product development and test roadmaps. > > Ed Ok that explains it a bit. I expected that Xilinx internally should have ml405 boards. So the deal is that the boards are there, but are just not going to be made available until someday in 2006. I still dont see a point to have the ml405 being mentioned in Xilinx publications in JAN 2005, and then ship sometime in 2006 AnttiArticle: 90076
Subhasri krishnan wrote: > Hi all, > I have a custom designed hardware.... when i ran the xilinx iMPACT to > configure the device, it ended up detecting the FPGA and a 101 unknown > devices and then just stopped dead? has anyone ever had a similar > problem? Please do help. Sounds like your JTAG chain is screwed up. -aArticle: 90077
vssumesh wrote: > The problem is i want that in a single chip. How can i link those huge > control signals out of FPGA. > But i am still wondering why the ISE is not working with my design. Probably because your expectations are not inline with the device capability? You're trying to create a huge combinatorial mux. It's not at all a surprise that you're not meeting your timing requirements. -aArticle: 90078
In the interrupt routine you need to clear the interrupt. Check to see that you are doing that. GeorgeArticle: 90079
CMOS wrote: > please someone let me know the effects of having "ibuf_lvcmos33" for > each input , "obuf_lvcmos33" for each output and "bufg" in some entity > declarations. The ibuf and obufs tell the tools what sort of I/O structure is to be used. There's some detail on this in the device datasheets. > In addition please let me know the effect of mappimg clk to clock in > the following entity declaration. > > entity test is port (clock : in std_logic ) > > end clock ; > > architecture arch_test of test is > > signal clk : std_logic ; > clk <= clock; > > end architecture arch_test ; > > in the remainder of the definition, only "clk" is used. "clock" is > never used. Why cant we just use "clock". Perhaps there's a language barrier here, but I have no idea what you're asking. I suspect that Symon feels the same way. -aArticle: 90080
Bill wrote: > Is the following a good way to avoid meta stability problems? ... > At least one of the signals a,b or c could end up in a meta stabel > condition. ... > It then takes the median of the 3 samples. ... [VHDL stuff] > q <= (a and b) or (a and c) or (b and c); You ask about metastability, which is a problem down to the transistor and parasitic capacitance level, and then suggest some VHDL. The actual behavior of this function is unknown until you know what logic gates and transistors (and other physical circuit items) your VHDL will produce after synthesis, optimization and layout, etc. Depending on the actual realization of your majority function, it could actually worsen the resolution time of the circuit over that of just using a closely spaced register pair delay on the same clock. IMHO. YMMV. -- rhn A.T nicholson d.O.t C-o-MArticle: 90081
Hi I know that the CoolRunner2 CPLD TQ144, XC2C256-7TQ144I is not designed to be radiation tollerent, but... I would be interested if anyone reading this has ever radiated a CPLD or know of some CPLD radiation data on the web. I have google searched but found very little. Most ics like 74hc04 parts can take about 10,000 or more rads of total dose (Co60). I hope to do some total dose testing this month. Also as it has a reprogramable memory, it will be suseptible to SEU. I wont be able to test for that.Article: 90082
please tell me what should i do now to set it right? should i read something or should i check for something? why does it detect so many devices when i have only one FPGA in there? ...thanksArticle: 90083
Subhasri krishnan wrote: > please tell me what should i do now to set it right? should i read > something or should i check for something? why does it detect so many > devices when i have only one FPGA in there? > ...thanks > Scanning the JTAG chain for devices is done by reseting the TAP controllers which loads the devices IDCODE into the TAP output register and then shifting the register out and waiting until you get a set of 32 ones (zeros??) indicating the end of the chain. The most likely cause is that you have connected the TDO cable pin to something other than the TDO pin of the last device in the chain that is generating bad data back to the cable. EdArticle: 90084
Subhasri krishnan: > please tell me what should i do now to set it right? should i read > something or should i check for something? why does it detect so many > devices when i have only one FPGA in there? > ...thanks I've had problems with too many detected devices when the cable frequency was too high. You can change it in the cable setup settings. If you use a Parallel Cable IV and it is properly detected, iMPACT usually sets it to a default of 5MHz. Try lowering the frequency and see if it gets any better. cu, SeanArticle: 90085
(Not Real) 'Bob', If you want an answer to this, why haven't you already contacted our Xilinx Aerospace/Defense Field Applications Engineers? We have plenty of data that we are willing to share with (Real) 'customers'. If you need a contact, I can send their name to you once you email me your (Real) 'Bob' email address (and real affiliation). It is Xilinx' interest to be sure that any aerospace/defense application (as well as any other) of our products uses all of the best information that we can provide. Austin Bob wrote: > Hi I know that the CoolRunner2 CPLD TQ144, XC2C256-7TQ144I is not > designed to be radiation tollerent, but... > > I would be interested if anyone reading this has ever radiated a CPLD > or know of some CPLD radiation data on the web. > I have google searched but found very little. > Most ics like 74hc04 parts can take about 10,000 or more rads of total > dose (Co60). > I hope to do some total dose testing this month. > Also as it has a reprogramable memory, it will be suseptible to SEU. > I wont be able to test for that. >Article: 90086
Hi Austin, First of all, thanks for having been polite and courteous to everyone while I was on paternity leave. Second, to reiterate, I do _not_ work for Altera. I did, but I don't, and haven't done so for several years. Thus, any reference to Altera with 'you' when replying to my stuff is wrong. > By the way, if Altera doesn't use "faulty bits" why do you have: > ... > 9 Patents (plus) for them? Yep, and for you too. Not sure whether you actually had to use Altera's cross-licensed patents to make EasyPath see the light of day, but it surely must have saved on attorney fees, and $DEITY knows all of us should avoid feeding those critters. > Your use of laser frapped fuses to replace bad columns is identical to > EasyPath (we just avoid the defects, the same as you). Those patent cross-licensing deals are good, aren't they? To be honest, I always understood that EasyPath devices were carefully selected defective parts where the defect happened not to interfere with the specific user's design. Thanks for clarifying this. You just can't trust Marketing people these days. > Meanwhile, 'EasyPath' remains easy, and now includes the standard option > of being able to change the logic (LUTs may be reprogrammed) or change > the IO (strength). Just like with an FPGA... Ok, but now I'm thoroughly confused. Can you tell me, on a hardware level, what is the difference between your standard XC4Vxxx offering and its accompanying EasyPath device? > Every try to ECO an ASIC? Yep, and it's something best avoided. I FIBbed a few dies. Not fun, and I was a lucky bastard for getting two of the bloody things to work that way. Then it was bow-in-shame time for requesting a new mask from the boss... Best regards, BenArticle: 90087
>> We have improved the VGA quality on the upcoming ML405 and ML410 boards >> that will be released early next year. > > I have done the "tie all grounds together" mistake myself, so I know > how you feel. > Jecel, could you give me some advice on this ? Same application; maybe using TI THS8133. I feel I'm going to do the same mistake again :-) Splitting grounds didn't always give me the best results. But I'm not so expert in high speed layout.Article: 90088
Sorry I don't use (real) info on newsgroups to keep spam to a minimum. I will email you with my (real) contact info. However I have tried before and was told that there was no info on the CoolRunner 2 as it is not intended for radiation and there are other parts that are designed for radiation. However that was last year. I have also searched on the NASA and ERRIC web sites and was unsuccessful in finding data. On Tue, 04 Oct 2005 11:55:47 -0700, Austin Lesea <austin@xilinx.com> wrote: >(Not Real) 'Bob', > >If you want an answer to this, why haven't you already contacted our >Xilinx Aerospace/Defense Field Applications Engineers? > >We have plenty of data that we are willing to share with (Real) 'customers'. > >If you need a contact, I can send their name to you once you email me >your (Real) 'Bob' email address (and real affiliation). > >It is Xilinx' interest to be sure that any aerospace/defense application >(as well as any other) of our products uses all of the best information >that we can provide. > >Austin > > > >Bob wrote: > >> Hi I know that the CoolRunner2 CPLD TQ144, XC2C256-7TQ144I is not >> designed to be radiation tollerent, but... >> >> I would be interested if anyone reading this has ever radiated a CPLD >> or know of some CPLD radiation data on the web. >> I have google searched but found very little. >> Most ics like 74hc04 parts can take about 10,000 or more rads of total >> dose (Co60). >> I hope to do some total dose testing this month. >> Also as it has a reprogramable memory, it will be suseptible to SEU. >> I wont be able to test for that. >>Article: 90089
Hi, i spend one week to figure out how this combination could work - with no success. currently we install the rh ws 3.0 into a separate directory ; chroot to it and start the software, that works I think something with the X libraries doensīt work with current amd 64 port of debian, ise produces a segmentation fault at startup. does anyone have some ideas / suggestions? regards, thomas BTW: I know that Xilinx says:RH WS3.0 is the ONLY supported distribution. System is dual opteron (2600MHz), 8GB mem, SATA etc... I īd like to use debian for a lot of reasons....Article: 90090
"Metastability" is a popular word to scare inexperienced designers. If you run a 1.8 MHz clock (even with a similar asynchronous data rate), your chance of having a 3 ns extra metastable delay is once per billion years (at 24 MHz it would be only 5 million years). For every additional ns of acceptable settling time, the mean-time-between-failure increases at least a million times. (see XAPP094 on the Xilinx website) The probability of your flip-flop failing during the life of this universe (even if you do nothing) with more than 10 ns of unaccounted-for delay is so minute it is practically zero. There are more important things to worry about, forget metastability... Peter Alfke, Xilinx Applications (who actually has created quantitative data about metastability)Article: 90091
I have had some problems similar to that in that I would get a strange number of devices rather than the six I have on the board. The Xilinx people pointed out the signal integrity issues I designed. Mostly the TCK line was the culprit. How is your TCK line routed? Is there a series resistance? Or perhaps you put an end Thevenin termination at the end? How long are the lines? > Hi all, > I have a custom designed hardware.... when i ran the xilinx iMPACT to > configure the device, it ended up detecting the FPGA and a 101 unknown > devices and then just stopped dead? has anyone ever had a similar > problem? Please do help. > Thanks >Article: 90092
Hello, Here is my problem : I want to have a combinatorial block with 4 inputs : - 2 vectors A and B - 2 control signal 'passthru' & 'sel' that produces either A+B (if passthru=0), A (if sel=0) or B (if sel=1). And all that in a signle layer of logic. In VHDL: process(a,b,passthru,sel) begin if passthru = '0' then q <= a + b; else if sel = '0' then q <= a; else q <= b; end if; end if; end process; I know it's possible : - using a(n), b(n), not passthru, sel as F1 F2 F3 F4 - using MULT_AND to stop carry propagation and I could do it for just mux(A, A+B) by describing it as bp <= b when passthru='0' else (others=>'0'); q <= a + bp; (and that must be a + bp, when putting bp + a that doesnt work, it produces two level!) Does any one know the good 'template' to use ? SylvainArticle: 90093
Ben, See below, -snip- > First of all, thanks for having been polite and courteous to everyone while > I was on paternity leave. Congratulations. > Second, to reiterate, I do _not_ work for Altera. I did, but I don't, and > haven't done so for several years. Thus, any reference to Altera with 'you' > when replying to my stuff is wrong. I will remember that in future. >>By the way, if Altera doesn't use "faulty bits" why do you have: >>... >>9 Patents (plus) for them? > > Yep, and for you too. We do not use any laser, or other non-volatile methods to improve, or increase die yield (yet). Not sure whether you actually had to use Altera's > cross-licensed patents to make EasyPath see the light of day, Nope. but it surely > must have saved on attorney fees, and $DEITY knows all of us should avoid > feeding those critters. Altera's patent portfolio is quite respectable. So is ours. Due to the last legal settlement, I can not comment on any of it. >>Your use of laser frapped fuses to replace bad columns is identical to >>EasyPath (we just avoid the defects, the same as you). > > Those patent cross-licensing deals are good, aren't they? Again, no comment. I am all for avoiding those 'critters' as well... > To be honest, I always understood that EasyPath devices were carefully > selected defective parts where the defect happened not to interfere with > the specific user's design. Thanks for clarifying this. You just can't > trust Marketing people these days. Who at Xilinx every represented EasyPath as any other than what you described above? I understand Altera tells everyone EasyPath are "defective" parts. They also fail to mention that ALL of their parts use redundancy to repair defective portions of their FPGAs. At least our non-EasyPath FPGAs really have no defects whatsoever. EasyPath is tested to the customer's pattern, and represents an extremely high test coverage for their specific application. One advantage of EasyPath is that our method of testing provides a much higher quality part than ASIC testing is capable of. >>Meanwhile, 'EasyPath' remains easy, and now includes the standard option >>of being able to change the logic (LUTs may be reprogrammed) or change >>the IO (strength). > > Just like with an FPGA... Ok, but now I'm thoroughly confused. Can you tell > me, on a hardware level, what is the difference between your standard > XC4Vxxx offering and its accompanying EasyPath device? Absolutely nothing at all. No difference (in silicon). The EasyPath component is just tested (and marked) for a specific bitstream (or two). A customer can have a bistream for in-house manufacturing test, and another one for the application, or any two of their choice as an option. Try that with an ASIC sometimes! >>Every try to ECO an ASIC? > > Yep, and it's something best avoided. I FIBbed a few dies. Not fun, and I > was a lucky bastard for getting two of the bloody things to work that way. > Then it was bow-in-shame time for requesting a new mask from the boss... Then you can imagine what fun it is to do backside fib'ing of flip chip die. But, one advantage of "hardtocopy" is that Altera has to do the fib'ing when something doesn't work (not the customer). We just change bits and the customer keeps shipping, uninterrupted.Article: 90094
(Real) Bob, Thanks for sending to me directly. There is an aerospace/defense FAE who will contact you shortly to provide you with what you need (if we have what you need). I know we had Coolrunner II in the LANSCE facility in May of this year, as it was my group that set it up, and was there for the testing. (surreal) Austin Bob wrote: > Sorry I don't use (real) info on newsgroups to keep spam to a minimum. > I will email you with my (real) contact info. > However I have tried before and was told that there was no info on the > CoolRunner 2 as it is not intended for radiation and there are other > parts that are designed for radiation. However that was last year. > I have also searched on the NASA and ERRIC web sites and was > unsuccessful in finding data. > > > On Tue, 04 Oct 2005 11:55:47 -0700, Austin Lesea <austin@xilinx.com> > wrote: > > >>(Not Real) 'Bob', >> >>If you want an answer to this, why haven't you already contacted our >>Xilinx Aerospace/Defense Field Applications Engineers? >> >>We have plenty of data that we are willing to share with (Real) 'customers'. >> >>If you need a contact, I can send their name to you once you email me >>your (Real) 'Bob' email address (and real affiliation). >> >>It is Xilinx' interest to be sure that any aerospace/defense application >>(as well as any other) of our products uses all of the best information >>that we can provide. >> >>Austin >> >> >> >>Bob wrote: >> >> >>>Hi I know that the CoolRunner2 CPLD TQ144, XC2C256-7TQ144I is not >>>designed to be radiation tollerent, but... >>> >>>I would be interested if anyone reading this has ever radiated a CPLD >>>or know of some CPLD radiation data on the web. >>>I have google searched but found very little. >>>Most ics like 74hc04 parts can take about 10,000 or more rads of total >>>dose (Co60). >>>I hope to do some total dose testing this month. >>>Also as it has a reprogramable memory, it will be suseptible to SEU. >>>I wont be able to test for that. >>> > >Article: 90095
In verilog I'd try to use: q <= (~sel | passthru ? A : 0) + (sel | passthru ? B : 0); The trouble is the MULT_AND doesn't like the OR functions for the "enable." If the sel and passthru are redefined to the control terms above, the single layer of logic for A and B would work. If you need sel, passthru, A, and B to all be single levels of logic, you can't do it because the MULT_AND wants two inputs: one of your vectors and one control term. q <= (useA ? A : 0) + (useB ? B : 0); // 1 level "Sylvain Munaut" <com.246tNt@tnt> wrote in message news:4342e651$0$25209$ba620e4c@news.skynet.be... > Hello, > > Here is my problem : > > I want to have a combinatorial block with 4 inputs : > - 2 vectors A and B > - 2 control signal 'passthru' & 'sel' > > that produces either A+B (if passthru=0), > A (if sel=0) or B (if sel=1). And all that in a > signle layer of logic. > > > In VHDL: > > process(a,b,passthru,sel) > begin > if passthru = '0' then > q <= a + b; > else > if sel = '0' then > q <= a; > else > q <= b; > end if; > end if; > end process; > > I know it's possible : > - using a(n), b(n), not passthru, sel as F1 F2 F3 F4 > - using MULT_AND to stop carry propagation > > and I could do it for just mux(A, A+B) by describing it as > > bp <= b when passthru='0' else (others=>'0'); > q <= a + bp; > > (and that must be a + bp, when putting bp + a that doesnt work, it > produces two level!) > > > Does any one know the good 'template' to use ? > > > SylvainArticle: 90096
John_H wrote: > In verilog I'd try to use: > q <= (~sel | passthru ? A : 0) + (sel | passthru ? B : 0); > > The trouble is the MULT_AND doesn't like the OR functions for the "enable." > If the sel and passthru are redefined to the control terms above, the single > layer of logic for A and B would work. If you need sel, passthru, A, and B > to all be single levels of logic, you can't do it because the MULT_AND wants > two inputs: one of your vectors and one control term. > > q <= (useA ? A : 0) + (useB ? B : 0); // 1 level > Actually, I don't see how to fit your representation (useA useB) into a single level ... What I wanted the tool to do is : (pt_n is not passthru) . cout . _|_ . ,--/___\ CYMUX . | | | . ______ | | | . a(b) ---------| | | | | ____ XORCY . b(n) ------x--| LUT4 |-x----------\\ \_____ . pt_n ---x--|--| | | x----//___/ . sel ---|--|--|______| | | . | | / | . | | __ / | . | '------| \ _/ cin . '---------|__/ . MULT_AND SylvainArticle: 90097
Peter Alfke wrote: > "Metastability" is a popular word to scare inexperienced designers. > > If you run a 1.8 MHz clock (even with a similar asynchronous data > rate), your chance of having a 3 ns extra metastable delay is once > per billion years (at 24 MHz it would be only 5 million years). > For every additional ns of acceptable settling time, the > mean-time-between-failure increases at least a million times. This certainly depends on the logic family, which the OP didn't specify. I have seen IO equipment, constructed with older slower logic families, fail due to improper synchronization at clock rates of much less than 24 MHz. Even today, with fast submicron CMOS, you can still create reliability problems with worse case register placements or interconnect routing. IMHO. YMMV. -- rhn A.T nicholson d.O.t C-o-MArticle: 90098
Austin Lesea <austin@xilinx.com> writes: >> To be honest, I always understood that EasyPath devices were carefully >> selected defective parts where the defect happened not to interfere with >> the specific user's design. This was my understanding as well. I think most people who take the time to read about it come to this conclusion. > Who at Xilinx every represented EasyPath as any other than what you > described above? Well, to be honest, going from the marketing materials alone, I wasn't quite sure if it was a new flavor of peanut butter or a boiled cabbage accelerator. But Xilinx's marketing department isn't really any worse than any other silicon valley company in that regard. ;) > I understand Altera tells everyone EasyPath are "defective" parts. > They also fail to mention that ALL of their parts use redundancy to > repair defective portions of their FPGAs. Interesting, I didn't know that. > At least our non-EasyPath FPGAs really have no defects whatsoever. Er, really? I thought you had to pay extra for the 100% tested ones. Does Xilinx really test every net on every chip for (say) stuck-at faults before shipping? - a -- PGP/GPG: 5C9F F366 C9CF 2145 E770 B1B8 EFB1 462D A146 C380Article: 90099
Hi Austin, > Who at Xilinx every represented EasyPath as any other than what you > described above? I unfortunately don't get to talk to Xilinx people very often nowadays, which is a bit of a pity because I used to get along well with the whole Xilinx FAE team in Belgium - including Marc Defossez. > I understand Altera tells everyone EasyPath are "defective" parts. Well, they were presented to me as 'possibly partially defective parts, but happening to work OK with the specific customer design'. > They also fail to mention that ALL of their parts > use redundancy to repair defective portions of their FPGAs. Not quite true. When the EP20K1500E and its slightly less bulky brethren came out Altera quite proudly presented the redundancy as a yield improvement technique. The redundancy 'feature' was also present in Stratix customer presentations. I don't remember about Stratix II presentations though. > At least our non-EasyPath FPGAs really have no defects whatsoever. Oh, once Altera's repair department is done with a die, so does a Stratix (II), which is 100% tested as well. > EasyPath is tested to the customer's pattern, and represents an > extremely high test coverage for their specific application. So, can I then summarize that EasyPath is basically a standard Virtex II/4 but with less time on the testbed (only the cells and routing used by the customer are tested) in order to reduce cost? Best regards, Ben
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