Joshua_S

Members
  • Content Count

    3
  • Joined

  • Last visited

About Joshua_S

  • Rank
    Newbie

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Hi guys, I am trying to properly set up the set_output_delay for a source synchronous output from the FPGA to the LTC2000A DAC, see: http://www.kynix.com/Parts/138283/LTC2000ACY-11%23PBF.html I have 2 DACs so there is DAC1 and DAC2. I am using a select_io_wizard to set up the output for each DAC. Here is the datasheet for the LTC2000A-11 DAC, see: http://cds.linear.com/docs/en/datasheet/2000afb.pdf On page 19 is the timing diagram, where I have set up the DAC to operate as in Figure 2. (DCKI_Q = 0, DCK_TADJ = 000) From the timing table on page 8, for t11 and t12, setup time t11 = 570, and hold time t12 is -170. Because the hold time specified in the LTC2000A datasheet is negative, it looks like the data can be released before the clock edge? I am using a Virtex 7 at 375MHz, xc7vx485tffg1761-2 (active) The 375MHz clock comes from a high speed ADC, which goes to a MMCM to generate the FPGA system 375MHz clock, where clk_in1 of the clk_wiz_inst is the output of the selectio_wiz input IP that forwards the clock coming from the high speed ADC. The 375MHz system clock clocks the selectio_wizard output IP for the DACs. The CLK_375M_OBJ is the source clock for the forwarded clock.The create_clock and create_generated_clock statements are following the timing reports. In addition, see the attached 1, I think this shows where the data needs to be valid at the DAC inputs. If the data is held longer, after the clock edge, that would still meet the requirements. Am I doing this correctly? Would it be better to just put in zero or a small positive value for the hold time? I'm sure the DAC would not mind if the data was held longer than specified, right? Thanks all.
  2. Hello guys, Some time ago I've watched a talk of the nand2tetris course/project. If you've never heard of it you can check it out here: http://www.nand2tetris.org/. Basically it is a course designed to build a computer from the very ground up. It starts with logic gates and goes all the way up to programming a small game project (hence the name nand2tetris). I've always been curious about doing something with FPGA hardware, but never had any idea of a feasible, yet interesting-result-yielding project. Well, as you can probably guess, up until now that is. During my university days I attended an FPGA workshop and I've read some stuff about the hardware components and the available course material such as Basic Introduction and Design flow of Programmable Logic Device FPGA. If you need to pick up something relating to the PLD or FPGA, you can check it out here: http://www.apogeeweb.net/article/67.html. So I think I kind of have an idea about the difficulty of the project. But since everything I've been reading just made me more excited, I decided I absolutely want to give this a shot. Now there's a lot of FPGA information out there, so I'm surely still missing a lot of important information, but I would like to get started and think the best way to learn is to actually experiment with a real FPGA instead of wasting too much time with HDL simulations only to become used to functions that aren't going to synthesize on the board anyway. So I now would like to ask you about some things I'm still unsure about and would like to have clarified before buying an expensive development kit. I've read several articles about Altera and Xilinx and right now my choice would be a Spartan 3E Starter Board - this one to be exact: https://store.digilentinc.com/spartan-3e-starter-board-limited-time/ The main questions I'm having right now: Is there a general reason that would argue against getting the Spartan E3 board? I actually have no idea how powerful an FPGA really is, but assuming it's running on Cyclone II hardware, it probably should run on a Spartan 3 as well? Or is that in itself already a stupid question, as FPGA comparison doesn't work so easily? In general, who is more newbie-friendly, Altera or Xilinx? (I've worked with VHDL before, which I think is Xilinx, right? Altera's Quartus is probably very similar?) From how I understand the FPGA toolchain, in the above linked offer there should be everything included to get me going. right? I also have this second candidate: http://www.digilentinc.com/Products/Detail.cfm?NavPath=2,400,836&Prod=ATLYS Maybe I'm an idiot, but as I read the offer, *only* the board is sold? I.e., in contrast to the first offer, I would need additional stuff to get something running on the board? It also strikes me as a mayor drawback that there is not really an output option besides a few LEDs. which for me would be very annoying; I'm already unhappy about the board linked before *only* having a small display (which in itself, however, is awesome to have of course). This might also be a stupid question: I remember from back in the FPGA workshop I mentioned that getting a number-display to run is not that hard at all. How much more effort is it to get something displayed on a screen (either an attached one or a PC screen accessed via one of the available ports)? And as a final question: I've also seen some very much smaller boards than the ones linked here, that are also much cheaper. Are the boards I'm looking it overkill for what I'd like to do with them? Or does actually the contrary hold, and such a project wouldn't even run on one of the smaller boards available? Regards, Joshua