There's a lot of autorouter hate in the electronics community which I think has a basis in truth, but is in danger of losing sight of the big picture.
Autorouters can be terrible. They mainly are. The reasons are partially with optimisation algorithms (although these are getting better), but primarily with poorly defined constraints.
Currently a schematic holds the physical connections and maybe a little bit of extra data - some pins might be "power" pins, or maybe a net or two is defined as a high current net - but broadly the circuit metadata which is highly relevant for routing is held in the designers head.
When you lay out a board you're always thinking about high current and high frequency (and especially high current high frequency) paths. You're looking for paths to ground, you're isolating sensitive analog regions from noisy digital regions, you're making sure the ground plane is uninterrupted. You're basically making thousands of tiny tradeoffs based on an understanding of what the board should be doing. If a computer doesn't have that knowledge, of course it can't effectively lay out a board.
If we can improve schematic annotation and part representation to properly hold this information then there's absolutely no reason autorouters can't be better than manual layout. If we can combine it with simulation then autorouting should really be the only logical choice. If you can almost completely avoid EMC testing by carefully defining all the high frequency nets and letting the autorouter find a design which is optimised for RF performance, then why wouldn't you? An EMC chamber costs thousands of dollars a day - you can easily spend 10s of thousands just on the testing alone, not to mention redesign costs.
There are tons of rule of thumb routing design choices made that lead to being able to route incredibly complex PCB's which becomes a topological problem. Rules that would need to be either programmed into such a system, or somehow potentially AI could use a huge corpus of board designs and schematics to make a better auto router.
My point is that as they stand currently, autorouters are absolute trash and you should not be depending on them.
The only real place they work is when you have a trivial routing situation but it would be incredibly tedious to route by hand, like a bus with 100+ traces or something.
I'm not depending on an auto-router, which is why I went through the exertion of adding the word "rough" in my post. Auto-routers are a very nice time saver to get something going quickly.
Also, you have strictly no idea what kind of design I'm working on and therefore no idea what constraints - or lack thereof - I have to deal with when I design boards.
Finally, I've heard so many boomer EDA folks spew out this lame argument ... it's the same old tiresome song from folks who use to claim compilers could never beat hand-crafted assembly. They're very seldom heard from these days.
If you have design requirements that warrant using an auto-router, I don't think you'd be posting on HN asking for it in an open source CAD package.
No, they're not for getting anything going quickly. Route it yourself so your boards don't look like absolute garbage. Unlike a compiled programming language, aesthetics matter, and you get no optimization from an autorouted board unless you're doing something like routing the exact same trace a hundred times over for some super simple parallel bus.
Which, again, I'm pretty sure you'd know what you were doing enough to not ask about an autorouter feature on HN if that were the case.
Autorouters can be terrible. They mainly are. The reasons are partially with optimisation algorithms (although these are getting better), but primarily with poorly defined constraints.
Currently a schematic holds the physical connections and maybe a little bit of extra data - some pins might be "power" pins, or maybe a net or two is defined as a high current net - but broadly the circuit metadata which is highly relevant for routing is held in the designers head.
When you lay out a board you're always thinking about high current and high frequency (and especially high current high frequency) paths. You're looking for paths to ground, you're isolating sensitive analog regions from noisy digital regions, you're making sure the ground plane is uninterrupted. You're basically making thousands of tiny tradeoffs based on an understanding of what the board should be doing. If a computer doesn't have that knowledge, of course it can't effectively lay out a board.
If we can improve schematic annotation and part representation to properly hold this information then there's absolutely no reason autorouters can't be better than manual layout. If we can combine it with simulation then autorouting should really be the only logical choice. If you can almost completely avoid EMC testing by carefully defining all the high frequency nets and letting the autorouter find a design which is optimised for RF performance, then why wouldn't you? An EMC chamber costs thousands of dollars a day - you can easily spend 10s of thousands just on the testing alone, not to mention redesign costs.