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
Hi Guys, I'm using the Spartan-3E 1600E Microblaze development kit. What I'm basically doing is writing to the on board DAC (Linear Tech LTC2624 Quad DAC), and then reading from the on board ADC (Linear Tech LTC6912 Dual A/D) whose analog input is connected to the DAC analog output. The data written to the DAC and the data read from the ADC is then made available to other modules within my design for analysis/comparison and is also made available to my LCD driver module so as to display the data on the LCD. The problem that I am having is that no data is read back from the ADC (ADC data displayed on LCD is all 0s), however the DAC was successfully written to (DAC output voltage changed correctly). Now initially I thought I must be reading from the ADC incorrectly so I simply brought out debug signals to a header on the board (J4) so that they could be viewed with an oscilloscope. The debug signals were connected to the SPI interface signals TCK, MOSI, MISO, and ADCconv. After bringing out these debug signals, which should not effect the design, the design then worked and ADC data was being read back correctly. Now when any combination of the ucf contraints that assign the debug signals to the header pins are commented out and the design is re-implemented (not res-synthesised) the design does not work. Does anyone have any ideas about this? Any help would be greatly appreciated. Regards, James --------------------------------------- Posted through http://www.FPGARelated.comArticle: 152676
jleslie48 <jon@jonathanleslie.com> wrote: > the xilinx says it has 500,000 gates, In the olden days (before BRAM and embedded multipliers) there was some use for this number, though very little. A little less useful than MIPS (Meaningless Indicator of Processor Speed) is of serial CPU performance. But now it includes the logic in BRAMs and multipliers, and anything else that they add on. Often your design doesn't use these additional resources in the exact proportion that they are supplied, and so the extras go to waste. (Like CPU cycles while waiting for I/O.) > the altera says it has > 22,320 Logic elements (LEs) > 594 Embedded memory (Kbits) > 66 Embedded 18 x 18 multipliers > 4 General-purpose PLLs This allows you to separately determine the logic, block memory, and block multplier resources. You alse need to know the size of an LE to compare with anything other than another device in the same family. > so are these two fpga's comparable in size/ computing power/ ability > to support the same VHDL or what? Is a 200MHz Pentium faster than a 250MHz PowerPC 620? --glenArticle: 152677
In article <_9GdnRsyfMSQXR3TnZ2dnUVZ_rydnZ2d@giganews.com>, James <james.gahan@n_o_s_p_a_m.analog.com> wrote: >Hi Guys, > >I'm using the Spartan-3E 1600E Microblaze development kit. >What I'm basically doing is writing to the on board DAC (Linear Tech >LTC2624 Quad DAC), and then reading from the on board ADC (Linear Tech >LTC6912 Dual A/D) whose analog input is connected to the DAC analog output. >The data written to the DAC and the data read from the ADC is then made >available to other modules within my design for analysis/comparison and is >also made available to my LCD driver module so as to display the data on >the LCD. > >The problem that I am having is that no data is read back from the ADC (ADC >data displayed on LCD is all 0s), however the DAC was successfully written >to (DAC output voltage changed correctly). Now initially I thought I must >be reading from the ADC incorrectly so I simply brought out debug signals >to a header on the board (J4) so that they could be viewed with an >oscilloscope. The debug signals were connected to the SPI interface signals >TCK, MOSI, MISO, and ADCconv. After bringing out these debug signals, which >should not effect the design, the design then worked and ADC data was being >read back correctly. > >Now when any combination of the ucf contraints that assign the debug >signals to the header pins are commented out and the design is >re-implemented (not res-synthesised) the design does not work. Timing constraints on the SPI input? Do they exist? Do they pass timing analysis - for both implementations (with and without the debug connections)? Depending on how/where you brought out the debug connections, it may affect timing greatly. If you're not having the tool check the timing, you're just rolling the dice. Regards, MarkArticle: 152678
"fpgaiua" <fpga@n_o_s_p_a_m.i.ua> wrote in message news:j-mdnQ-uufQUDx3TnZ2dnUVZ_sadnZ2d@giganews.com... > Good day, > I'm working now with Spartan3 PCI project using the pci32tlite core. Small question: is the pci core from opencores works fine?Article: 152679
>In article <_9GdnRsyfMSQXR3TnZ2dnUVZ_rydnZ2d@giganews.com>, Thanks for the help. As you could probably guess I'm new to setting timing constraints and analysing timing results. Is there any literature or documentation that you could recommend that might help me? Regards, James --------------------------------------- Posted through http://www.FPGARelated.comArticle: 152680
>>In article <_9GdnRsyfMSQXR3TnZ2dnUVZ_rydnZ2d@giganews.com>, > >Thanks for the help. > >As you could probably guess I'm new to setting timing constraints and >analysing timing results. >Is there any literature or documentation that you could recommend that >might help me? > >Regards, >James > http://forums.xilinx.com/t5/PLD-Blog/Timing-Constraints-Part-1-of-5/ba-p/57594 --------------------------------------- Posted through http://www.FPGARelated.comArticle: 152681
Scott Michel <scooter.phd@gmail.com> sent: |--------------------------------------------------------------------------| |"On Sep 14, 2:06 pm, glen herrmannsfeldt <g...@ugcs.caltech.edu> wrote: | |> > 6. A new language with APL-like semantics would allow programmers to| |> > state their wishes at a high enough level for compilers to determine | |> > the low-level method of execution that best matches the particular | |> > hardware that is available to execute it. | |> | |> APL hasn't been popular over the years, and it could have done | |> most of this for a long time. On the other hand, you might look | |> at the ZPL language. Not as high-level, but maybe more practical." | |--------------------------------------------------------------------------| Languages can allow compilers to optimize in a way which compilers do not bother to do. That was true of APL and other languages. |----------------------------------------------------------------| |"ACM killed off SIGAPL about 5-6 years ago. Sorry to see it go."| |----------------------------------------------------------------| SIGAPL exists. In what way was SIGAPL "killed off"? |-----------------------------------------------------------------------| |"Have a look at the DARPA HPCS languages, notably, Chapel, Fortress and| |X10. Not entirely sure about their respective statuses, but they were | |an attempt in the HPC arena to raise the level of abstraction. | | | | | |-scooter" | |-----------------------------------------------------------------------| Someone of Oracle authored an article in "The Journal of the ACM" or in "The Communications of the ACM", mainly re replacing efficient Fortran code with fewer lines of Fortran with "only" a drop in performance of circa 100%. Fortress and circa two other languages (maybe Chapel and Fortress) were briefly mentioned. In a manuscript by myself which is currently under review, I mention Fortress (and Fortran). In some drafts of the manuscript I mentioned part of Fortress which is less abstract for efficiency (or maybe at a higher level and less efficient). I edited that observation out before submitting the manuscript to the journal because of a space constraint and the focus of the manuscript is something else. The edited Fortress section still shows that the idea that the notion that high abstraction is deployed in Fortress is laughable. Sincerely, Paul Colin GlosterArticle: 152682
On 2011-09-27, scrts <hidden@email.com> wrote: > "fpgaiua" <fpga@n_o_s_p_a_m.i.ua> wrote in message > news:j-mdnQ-uufQUDx3TnZ2dnUVZ_sadnZ2d@giganews.com... >> Good day, >> I'm working now with Spartan3 PCI project using the pci32tlite core. > > Small question: is the pci core from opencores works fine? We used it in an Ethernet NIC project a long time ago and it worked well. Unfortunately it had severe performance issues with repeated burst reads and writes. The problem was that the FIFO:s in the PCI bridge didn't implement any sort of high watermark/low watermark system. So as soon as the FIFO was full you could only fill the FIFO with a transaction that was one word long. At this point the performance of the system could not be recovered if the injection rate did not decrease significantly. More information about our experiences can be found here: ihttp://liu.diva-portal.org/smash/get/diva2:22746/FULLTEXT01 regards /AndreasArticle: 152683
I want to do something like this signal slv : std_logic_vector(7 downto 0); signal sl : std_logic; signal res : std_logic; res <= slv = x"01" and sl = '1'; But I get an error. I guess its something to do with them being different types. But I would of thought that the two terms would both resolve to 1 or 0 and then I could just "and" them. Jon --------------------------------------- Posted through http://www.FPGARelated.comArticle: 152684
On Sep 30, 11:24=A0am, "maxascent" <maxascent@n_o_s_p_a_m.n_o_s_p_a_m.yahoo.co.uk> wrote: > I want to do something like this > > signal slv : std_logic_vector(7 downto 0); > signal sl =A0: std_logic; > signal res : std_logic; > > res <=3D slv =3D x"01" and sl =3D '1'; > > But I get an error. I guess its something to do with them being different > types. But I would of thought that the two terms would both resolve to 1 = or > 0 and then I could just "and" them. > No, (slv =3D x"01") resolves to type boolean as does (sl =3D '1'). I have a function called to_std_logic that takes a boolean and returns a std_logic for just such occasions. So then you have res <=3D to_std_logic(slv =3D x"01" and sl =3D '1'); Kevin JenningsArticle: 152685
Ok thanks, I thought there would be some kind of conversion function required. Jon --------------------------------------- Posted through http://www.FPGARelated.comArticle: 152686
OK, well I have updated to v16.5. Nice improvement in the colors and such. But I can't see the vias as being isolated from the copper pours.Article: 152687
Brad Smallridge wrote: > I downloaded the design files and opened file 4199_SP605_brd_091909.brd > with Allegro's Summer '09 viewer, that is, v15.7. <snip> > These inner plane layers display as completely filled areas with > no regard to vias. <snip> >Is it possible to get these via connections to show properly? With v15.7 : Turn off everything else: display->color and visibility->global visibility->all invisible Turn on the etch for the layer you want to see: Visibility Tab-> 07_Pwr2 etch checkbox Enable display of Thermal Pads: Setup->Drawing Options->Display->Thermal Pads This should show you a circle/circle-with-spokes/solid connection for each via. Note that this setting will "hide" the normal colored-if-connected display of vias whenever the etch display is turned on. > > Another issue is the plot. I can plot to my printer, that's good. But when I > plot to a file, the file has a .PLT extension. IIRC, PrimoPDF worked for me a few years back to print to a pdf. HTH, BrianArticle: 152688
> Enable display of Thermal Pads: > Setup->Drawing Options->Display->Thermal Pads Beautiful. Well worth reinstalling the 15.7 viewer again. I uninstalled the 16.5 but maybe I didn't have to do that. I did not see any similar option on the 16.5 viewer. > Note that this setting will "hide" the normal colored-if-connected > display of vias whenever the etch display is turned on. I don't follow this. > IIRC, PrimoPDF worked for me a few years back to print to a pdf. I will try this if I can't find something else. I'd like to be able to edit the images in something easier, like a JPEG. My backup is to print onto paper and rescan. Thank you so much Brian, Brad Smallridge Ai VisionArticle: 152689
> My workaround was to "idle" the chip with an "almost empty" bitstream first ISE 13.2 says "The design is empty. No processing will be done" Moreover, what will happen during dynamic logic "tear down"? Will you have uncontrolled inputs (randomly uncontrolled drivers)?Article: 152690
Brad Smallridge wrote: > I uninstalled the 16.5 but maybe I didn't have to do that. They seem to co-exist OK, the default 16.5 install directory differed from 15.7's when I ran the 16.5 installer today. > I did not see any similar option on the 16.5 viewer. Looks like it has been moved to: Setup->Design Parameters->Display->Enhanced Display Modes->Thermal Pads > > > Note that this setting will "hide" the normal colored-if-connected > > display of vias whenever the etch display is turned on. > > I don't follow this. > The boards I used to review from the local layout houses were generally set up with different net color assignments for the power nets ( instead of the all-white theme of that SP605 board ), so displaying just the vias on a power plane would show a constellation of differently colored donuts for the power vias, and light gray for everything else. Particularly on power planes that haven't been poured yet, this makes it very easy to eyeball split plane cuts. But when you have both vias _and_ the layer etch turned on, _and_ that "thermal pads" option is true, this colored-donut view is hidden by the thermal relief rendering. [ It's been a few years since I've actively used that viewer to review design files, so my description above might be slightly mangled ] To see this effect, try using the Display->Assign Colors in 16.5 to give some of the power nets a different color on that SP605 board. BrianArticle: 152691
On Sep 30, 11:45=A0am, "maxascent" <maxascent@n_o_s_p_a_m.n_o_s_p_a_m.yahoo.co.uk> wrote: > Ok thanks, I thought there would be some kind of conversion function > required. > > Jon =A0 =A0 =A0 =A0 > > --------------------------------------- =A0 =A0 =A0 =A0 > Posted throughhttp://www.FPGARelated.com KJ's reply hints at a pretty fundamental concept in VHDL: the language- level logic is purposefully different from the digital logic used to describe circuits (true/false vs. hi/lo/x/z/...). For example, I would imagine KJ's "to_std_logic" function to include a statement like: output_logic <=3D '1' when (input_boolean =3D true) else '0'; Just something to keep in mind. MBArticle: 152692
> > > IIRC, PrimoPDF worked for me a few years back to print to a pdf. > > I will try this if I can't find something else. I'd like to be able > to edit the images in something easier, like a JPEG. My backup is to > print onto paper and rescan. Looking around the 16.5 menus, I spotted the following command: File->Capture Canvas Image that can save the current board view to a JPEG file. BrianArticle: 152693
> ISE 13.2 says "The design is empty. No processing will be done" To generate an empty bitstream, open 'fpga_editor', save an empty design (.ncd) and then run 'bitgen' on it.Article: 152694
I've seen lots of messages a while ago about how ISE is going downhill... Which version of ISE would people recommend for fairly simple VHDL projects using Spartan-3 and Spartan-6? I'm currently developing with an old S3A prototype board, with a view to change to S6 for the next iteration. I've used ISE 10.1 in the past with no issues, but would prefer to upgrade now to ease any pain changing from S3 to S6.Article: 152695
Cant say I have ever had a problem with ISE. I use 13.2 with a Virtex 5 device. I guess if you are going to use S6 then you would need to use a newer version. Jon --------------------------------------- Posted through http://www.FPGARelated.comArticle: 152696
The trouble I am having is that I have been using Verilog for a number of years but now need to use VHDL. As Verilog lets you do whatever you like, I have been finding VHDL quite hard going. But I am slowly beginning to get the hang of it. Jon --------------------------------------- Posted through http://www.FPGARelated.comArticle: 152697
It really depends what you are doing which is best. ISE12.4 was very good but had some changes particularly in IP starting to move to AXI bus interface. Prior to that 12.2 was also good. We are now running 13.2 a lot currently and many things are fixed, or much better, in the IP side. Generally the change from PLB busses to AXI has meant a lot of change recently and that's never good for stability but that should get better with more releases. I expect 13.3 will be better again but wait and see. John Adair Enterpoint Ltd. - Home of Raggedstone2. The Spartan-6 PCIe Development Board. On Oct 3, 11:24=A0am, Mike Harrison <m...@whitewing.co.uk> wrote: > I've seen lots of messages a while ago about how ISE is going downhill... > Which version of ISE would people recommend for fairly =A0simple VHDL pro= jects using Spartan-3 and > Spartan-6? > I'm currently developing with an old S3A prototype board, with a view to = change to S6 for the next > iteration. > I've used ISE 10.1 in the past with no issues, but would prefer to upgrad= e now to ease any pain > changing from S3 to S6.Article: 152698
Could someone please try the following: In ISE 13.2, open one of the 9572XL schematic example projects (jc2_sch, jc2_sver or jc2_svhd) and run the Floorplan IO - Pre-Synthesis process. Does the design in the floorplanner resemble the schematic *at all*? -aArticle: 152699
John Adair wrote: > It really depends what you are doing which is best. ISE12.4 was very > good but had some changes particularly in IP starting to move to AXI > bus interface. Prior to that 12.2 was also good. We are now running > 13.2 a lot currently and many things are fixed, or much better, in the > IP side. > > Generally the change from PLB busses to AXI has meant a lot of change > recently and that's never good for stability but that should get > better with more releases. I expect 13.3 will be better again but wait > and see. > > John Adair > Enterpoint Ltd. - Home of Raggedstone2. The Spartan-6 PCIe Development > Board. > > On Oct 3, 11:24 am, Mike Harrison <m...@whitewing.co.uk> wrote: >> I've seen lots of messages a while ago about how ISE is going downhill... >> Which version of ISE would people recommend for fairly simple VHDL projects using Spartan-3 and >> Spartan-6? >> I'm currently developing with an old S3A prototype board, with a view to change to S6 for the next >> iteration. >> I've used ISE 10.1 in the past with no issues, but would prefer to upgrade now to ease any pain >> changing from S3 to S6. > I think the bus changes would affect EDK projects more than the "VHDL projects" noted in the OP. However I would think that 12.4 should be a minimum revision for the Spartan 6 if you want to use MIG and get the design to meet timing at the high end of its performance range. If you use ISIM for simulation, there have also been improvements in release 12.x, but I don't use newer versions so I can't comment on ISIM there. Modelsim XE is no longer available with the newer versions, so if you want to use ModelSim you may need to buy a seat of PE. I found that ISIM is easy enough to get used to, and a full seat of ISIM is at least as good as ModelSim XE, (but still far off from ModelSim PE). It's hard to say which version of ISE is most "stable" because they all have quirks, so once you learn to work around them all version become stable enough to get the job done. One thing to watch for in versions 12.x and newer is the "new parsers" for V6 and S6 HDL design. While these have the ability to generate significantly smaller and faster results, they have some bugs to be worked out still. Version 11.5 still uses the old parsers for V6 and S6. If you find you have issues with some structures in the new parsers, you can always try to go back to 11.5 for a sanity check. One would hope that future releases will fix most of these bugs. In the meantime you may need to do post-translate simulation to make sure the synthesis didn't break the design. -- Gabor
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