Do you guys think we would be better off with alot more drive train diversity, or do you like everybody having complex movement?
5 Likes
Swerve meta best meta. it makes FRC FAR more exciting for everyone involved
50 Likes
Letās be clear though, there has pretty much always been one dominant drive train. Before swerve, it was tank. There has never really been a diversity of drive trains.
32 Likes
Iām newer to FRC, but from what Iāve seen swerve drive makes every game a degree more exciting
12 Likes
The ubiquity of swerve makes it much more expensive to play the sport for the average team, which I think is a bad effect much larger than any coolness factor to everyone running swerve drives.
That being said I donāt think you can put this cat back in the bag very easily - I think the sport really needs to fully commit and find ways to make swerve drive more accessible - like putting it as an option in the kit of parts potentially.
40 Likes
Overall yes.
However, if it is to be the continued dominant meta and nearly required to win events or even be picked for playoffs: Iāll like it better when it is included at least as an option in the KoP or FRC includes it in a recommended budget for teams so it is immediately factored into new teams budgeting. Swerve is still expensive and we need to raise the floor for all.
32 Likes
As a driver I love it I have driven some of our old tank bots for outreach and they are just kind of boring to drive. There is something about swerving around on the field and avoiding other robots thatās just so much fun.
9 Likes
Itās getting better, but until we gather a critical mass of swerve knowledge there is still a lot of pain on the technical side. The games are fun to watch with so many on swerve, but the amount of technical debt teams took on is crazy.
There are a lot of āswerve code broke. Helpā threads (and the mechanical equivalents) on this forum. Teams treat is as a black box to increase competiveness, questions later. This concerns me as the teams that donāt frequent CD see swerve as needed, too much can go wrong and it can be a hotbed for frustration.
So fun to watch? Yes.
Do I like it? Overall yes.
Does it make FRC unique? Yes.
Good for FRC? Probably not for the bottom 10% of teams on several metrics.
16 Likes
Swerve makes matches exciting, but it creates a norm where they are required to win events. Sadly, not every team can afford them.
In a theoretical world, Iād love to see a game with more diverse drive trains. As an engineer, I like to marvel at unique and innovative solutions teams come up with rather than seeing the same swerve on every bot.
3 Likes
I like what it has done to the game visually. I like the increased opportunity for teams to take on technical challenges. I like that it makes automated movement easier to reason about (at the cost of more mechanical and ālower-level controlsā complexity).
I do think it will be important for the GDC to work toward games where simpler (or simply lower-performing) drivetrains have a significant role to play.
Alternatively, effort toward a swerve kitbot might be important. Some elements of this:
- Easy assembly
- Cheap
- Trivial-as-possible wiring
- FRC-supplied software that ājust worksā
These are not easy things to achieve though.
The fundamental issue will always remain that when you multiple the degrees of freedom of the drivetrain by 4, you induce that much more opportunities for failures and bugs. Most ābottom 1/3rdā teams do not have the capacity to absorb that extra work. So some supplier or HQ has to soak it up.
15 Likes
More if you add the cost of brushless motors.
Unfortunately, the arms race is a topic thatās been beaten to death on CD. Itās one that, long term, FIRST is going to have to deal with: If the floor to be competitive rises every year (whether in cost or just technical ability), sustaining any team becomes harder, as does recruiting rookie teams.
1 Like
This is a huge lift that relies on a manufacturer to engineer themselves out of many common pitfalls.
Not that it canāt be done, it is just a big project for a niche hobby market.
I would love to see it too, but the teams that need it are not the ones that make the COTS suppliers their margins.
Most of the current issues we have are people not following directions closely or underestimating how important some details are. This happed in the tank era too. I am getting tired to of telling filks to double/triple check magnets and if the CAN ID s listed in code correspond to the cirrect physical corners.
That and drive ratios.
FRC is an information war and there will always be a bottom 10 percent.
1 Like
I have clearly seen the difference that additional wonder and awe can inspire in students, parents, mentors, and sponsors. Anything that increases inspiration is a win in my book. Spend the money and get students the cool shiz.
As a programming mentor, the swerve meta is awesome. Some of the automation is phenomenal. I feel like itās empowered my students to tackle more diverse problems than ever before.
8 Likes
Swerve meta is annoying because iād love to treat it like a magical black box that teleports the robot from point A to point B in any orientation like everyone else does, but as a controls mentor I donāt get that luxury.
Also Iām not particularly interested about the technical details of a good swerve control implementation.
Trying to get YAGSL+Pathplanner to black box mode as effectively as I can so I can ignore it and do more fun stuff.
Isnāt the answer here that the kitbot should have swerve, and there should be a kitbot swerve code template?
I mean, they did it with tank. The problem is that the opt out voucher would go from $450 to $800-1000, and the registration fee would probably increase. But maybe the volume of supplying kitbot swerves would lower that swerveās price?
Or, can you imagine, the kitbot being a āpick your modelā deal, where the suppliers supplied the modules at cost?
Just spitballing. I see the problems involved as well. Asking a rookie to jump straight into swerve is scary, and ANY increase in the $ required is almost a non-starter.
10 Likes
The long-term financial investment required for teams to switch to swerve is less than it looks like. Rules changes over the last half decade mean that reuse of major mechanisms is no longer defined the way it once was, and teams can more easily (legally) reuse swerve modules than they could reuse drive bases in the past. Modules are also quite durable; I would be surprised if teams were not able to maintain and reuse modules like the SDS Mk4 for four or more seasons; remember that gearboxes of old were expensive, easy to break/wear out, and likely to not fit your needs the following season anyway. So it becomes more of a ābuy once, cry onceā approach to funding. Not an easy hump for new and low resource teams to get over, but once they have done so it is less of a burden in the long run.
I agree with @Tom_Line, I think FIRST needs to figure out how to add swerve hardware and software to the future KOP for rookies.
6 Likes
In fact I think Iād compare the Swerve Revolution to the Smartphone Revolution. Pre-iPhone, there was a plethora of partially functional communication options that were distinguished by features, format and pricing. Post-iPhone, everyone got a black brick in their pocket after complaining about the price, but now I only need to replace my phone once every four years unless I drop it, and it does all the things. Swerve modules are approaching a level of consistent performance, ease of assembly and maintenance, and durability that makes all robots perform at a close-to-equal level in terms of basic functionality.
8 Likes
One thing that I have not given much thought but would be cool, is if with different drivetrains you would have different ābuffsā or āde-buffsā kind of like nhrl. So say if you used tank drive you would have the normal frame parameters that we have now but say if you had an āinnovativeā type of drivetrain you could get cut bumpers (rip) or you could make your robot taller, also you may be able to add more drive motors.
On a side note I do not think this will ever be added to frc I think the swerve meta is way to much fun. these changes would be way to complicated, I just think its fun to think about.) 
2 Likes
This part here. Expensive in dollars, expensive in expertise even if you have the dollars. And when tank drives have been nerfed by the propulsion motor rule so they canāt keep up with something that can put full power down in any direction all the time, not swerving is expensive in Ls and short Saturdays.
Until they figure out something in the Kickoff kit, Iām going to be grumpy about it.
10 Likes
Swerve is very, very cool. I do kind of miss the era of custom-developing drivetrains from 2 perspectives. As a machinist, I found the drive rails incredibly fun to make. As a pit scouter, it was nice hearing āWe made a custom 6-wheel shifting WCD (or something)ā instead of just āmk4i, L2ā all the time.
I also think people underestimate the dev time required for even COTS swerve (let alone custom like we did in 2023). I was talking to a team at one of our competitions this year who was not a rookie but essentially built the Everybot every year. They were ~6 people and 1 mentor, and they lacked a lot of resources (IK and money, mainly) but wanted to buy swerve to gain a competitive advantage. I advocated that they put their resources elsewhere as they work on training software people for the workload of swerve. I donāt really know if they listened.
In this vein, I think putting swerve in the KoP would be harmful, as unless there is extensive code that already works, I think you will see a lot more low-resource teams/rookies dead on the field because they couldnāt get it working than we are used to. This would lead to a lot of demotivating, feels bad moments.
4 Likes