PCB 布局是一个完全不同的领域，其他类型的 EDA 软件（例如 IC 设计）很快就会变得非常复杂。我会避免这些，尽管我知道一些聪明的人总是在研究自己的 PCB 自动布线器……这当然是一项具有挑战性的任务，而自动布线器是没有单一解决方案的问题之一。对于自动布线器，存在各种各样的权衡，并且您通常可以将它们相互替代。弄清楚用户应该能够调整/限制什么、调整多少，以及当路线“足够好”时，它就不是黑与白。白色的。
不管怎样，也许里面的东西会给你一些可以咀嚼一段时间的东西。祝你好运。
LTSpice is a free but very capable & popular device/circuit simulation package (it also does schematic capture & has a waveform viewer). It's gonna be tough to outdo that one, but playing with it might give you some ideas.
One thing that comes to mind... DSP is an area where math, electrical engineering & software all come together. It's beautiful, it can be tough, but it's extremely useful to know and it's very widely applicable.
When you're doing DSP work, you're very often working with sets of sampled data, trying to improve or refine the algorithms/filters/electronics/etc... I've used a pretty simple but useful program called ScopeDSP for this, and I've often thought about "rolling my own", but I've never had the time. There is another program from the same company, ScopeFIR, which is used for designing & analyzing Finite Impulse Response (FIR) digital filters. The programs aren't terribly complex (no pun intended), and with a little DSP knowledge, you could try something along those lines.
PCB Layout is a whole different bailiwick, and other kinds of EDA software (e.g., IC design) can get very hairy very quickly. I'd avoid those, although I know a couple smart guys who are always working on their own PCB autorouters... certainly a challenging undertaking, and autorouters are one of those problems that don't have a single solution. With autorouters, there are all sorts of tradeoffs, and you can usually play them off of each other. Figuring out what the user should be able to tweak/constrain, by what amount, and when the route is "good enough", it's not black & white.
Anyway, maybe something in there will give you something to chew on for a while. Good luck.
I agree with Johan -- please consider improving software already started elsewhere, rather than creating the 101st partially-implemented variant of an idea that already has 100 partially-implemented variants floating around.
Push the state of the art forward.
Here are some projects that could use some software help that are much closer to the electronic hardware than most software projects:
RepRap (can this machine be re-programmed to make printed circuit boards?)
Gwyddion (scanning probe microscope software)
"Open Programmer" and "TekBots Universal Programmer" (can these be merged?)
SDCC, the Small Device C Compiler
"Quite Universal Circuit Simulator" and "DigitalSimulator" and "ngspice"
发布评论
评论（3）
一些现有的项目怎么样：
How about to some existing projects like:
LTSpice 是一个免费但功能强大的工具。流行的设备/电路仿真包（它还可以捕获原理图并具有波形查看器）。想要超越它是很困难的，但是使用它可能会给你一些想法。
我想到一件事......DSP 是一个数学、电气工程和应用领域。软件全部整合在一起。它很美丽，也可能很困难，但了解它非常有用，而且应用非常广泛。
当您进行 DSP 工作时，您经常会处理采样数据集，尝试改进或细化算法/滤波器/电子器件/等...我使用了一个非常简单但有用的程序，称为 ScopeFIR，用于设计与分析有限脉冲响应 (FIR) 数字滤波器。这些程序并不是非常复杂（没有双关语），只要有一点 DSP 知识，您就可以按照这些思路尝试一些东西。
PCB 布局是一个完全不同的领域，其他类型的 EDA 软件（例如 IC 设计）很快就会变得非常复杂。我会避免这些，尽管我知道一些聪明的人总是在研究自己的 PCB 自动布线器……这当然是一项具有挑战性的任务，而自动布线器是没有单一解决方案的问题之一。对于自动布线器，存在各种各样的权衡，并且您通常可以将它们相互替代。弄清楚用户应该能够调整/限制什么、调整多少，以及当路线“足够好”时，它就不是黑与白。白色的。
不管怎样，也许里面的东西会给你一些可以咀嚼一段时间的东西。祝你好运。
LTSpice is a free but very capable & popular device/circuit simulation package (it also does schematic capture & has a waveform viewer). It's gonna be tough to outdo that one, but playing with it might give you some ideas.
One thing that comes to mind... DSP is an area where math, electrical engineering & software all come together. It's beautiful, it can be tough, but it's extremely useful to know and it's very widely applicable.
When you're doing DSP work, you're very often working with sets of sampled data, trying to improve or refine the algorithms/filters/electronics/etc... I've used a pretty simple but useful program called ScopeDSP for this, and I've often thought about "rolling my own", but I've never had the time. There is another program from the same company, ScopeFIR, which is used for designing & analyzing Finite Impulse Response (FIR) digital filters. The programs aren't terribly complex (no pun intended), and with a little DSP knowledge, you could try something along those lines.
PCB Layout is a whole different bailiwick, and other kinds of EDA software (e.g., IC design) can get very hairy very quickly. I'd avoid those, although I know a couple smart guys who are always working on their own PCB autorouters... certainly a challenging undertaking, and autorouters are one of those problems that don't have a single solution. With autorouters, there are all sorts of tradeoffs, and you can usually play them off of each other. Figuring out what the user should be able to tweak/constrain, by what amount, and when the route is "good enough", it's not black & white.
Anyway, maybe something in there will give you something to chew on for a while. Good luck.
我同意 Johan 的观点——请考虑改进已经在其他地方开始的软件，而不是为已经有 100 个部分实现的想法创建第 101 个部分实现的变体。
推动最先进的技术向前发展。
以下是一些可以使用比大多数软件项目更接近电子硬件的软件帮助的项目：
I agree with Johan -- please consider improving software already started elsewhere, rather than creating the 101st partially-implemented variant of an idea that already has 100 partially-implemented variants floating around.
Push the state of the art forward.
Here are some projects that could use some software help that are much closer to the electronic hardware than most software projects: