Let me start by saying thanks to those of you experienced and enthusiastic EE Times members that have helped to make getting into a new electronics project (such as FPGAs) especially thought provoking at a high level for a relative newbie such as myself.
That said, I was hoping to tap that experience to garner a recommendation for a hardware purchase. I am interested in an FPGA development board that can double as a learning tool, but most importantly stay relevant as I ramp up to more involved problems very soon. After a (hopefully) quick USB snooping/hacking beginning project, I am interested in implementing computer vision algorithms into hardware (motion detection, blob detection, pattern recognition, etc. or at least preprocessing of the source image prior to these functions in software), initially using HDMI- and VGA-based inputs and outputs.
I have narrowed it down to two offerings from Digilent (although if someone knows of a perfect fit that I missed, I am definitely interested in hearing it) which are under my budgeted $400 USD. Initially, the Spartan-6 based Atlys board caught my eye, already having dual HDMI inputs and outputs, and being explicitly used in an HDMI white paper from Xilinx that should speed my development. But then the 7th generation capacity and new Vivado Design Suite compatibility of the Artix-7 based Nexys 4 board is also intriguing.
Repetition of some high points of each for comparison: Atlys Board: IC: Xilinx Spartan-6 LX45 FPGA 6,822 slices 2.1Mbits of fast block RAM 4 clock tiles, 6 phased-locked loops 58 DSP slices 500MHz+ clock speeds RAM: 128Mbyte x 16-bit wide DDR2 Included Ports: Two HDMI video input ports & two HDMI output ports 10/100/1000 Ethernet PHY General Use Expansion I/O's: 48 Nexys 4 Board: IC: Xilinx Artix-7 XC7A100T 15,850 logic slices 4,860 Kbits of fast block RAM 6 clock tiles, 6 phase-locked loops 240 DSP slices 450MHz+ clock speeds RAM: 16Mbyte CellularRAM Included Ports: 12-bit VGA output 10/100 Ethernet PHY General Use Expansion I/O's: 40
Which of these two do you think is the best choice for computer vision-related tasks? From my limited knowledge, it would seem that the decision boils down to the plentiful memory of the Atlys versus the plentiful logic capacity of the Nexys 4 (and it's added bonus of being "newer" and thus further from eventual obsolescence, assuming it's possible to implement HDMI and/or VGA ports on the general I/O's of the Nexys). Which is more likely to bottleneck a video processing application, memory or capacity?
I don't have the experience to weigh the relative needs of such an application, but if I had to choose today I suppose I'd go with the Atlys only because it has a proven video tutorial available, but I would regret not being able to begin the learning experience with the "new" Vivado toolset. What do you think?
What are the engineering and design challenges in creating successful IoT devices? These devices are usually small, resource-constrained electronics designed to sense, collect, send, and/or interpret data. Some of the devices need to be smart enough to act upon data in real time, 24/7. Are the design challenges the same as with embedded systems, but with a little developer- and IT-skills added in? What do engineers need to know? Rick Merritt talks with two experts about the tools and best options for designing IoT devices in 2016. Specifically the guests will discuss sensors, security, and lessons from IoT deployments.