I wouldn’t say it’s a question of should or shouldn’t as much as it is “if people are buying subsystems designed for this year’s game this year, odds are they’ll be interested in buying subsystems designed for 2019’s game next year”.
Do you think the GDC will allow that to happen in 2019?
I don’t know, unfortunately I’m not the GDC, but I sure hope so!
I don’t think I would want the GDC’s job… they have to deal with all of us!
I think they will. I like to think the products are helping far more people than they possibly could hurt (which I’m certain is zero anyway).
That being said…
If they ban game specific mechanisms, then I’ll shift more towards things like elevators that are universal (that being said the claw can pick up a 2014 track ball, so it’s pretty darn universal).
If they ban products launched after kickoff, that would put quite a damper on all existing suppliers that routinely don’t stock items until after kickoff. This mostly punishes teams, and I’m not really sure what the upside is.
VexPro and AndyMark both announced their full product line before kickoff, even if some items weren’t already stocked.
For this season, sure. In previous seasons both have launched products after kickoff (as they should) without any issue.
Announced and released are a very subtle practical difference as well.
The difference being you can announce mechanisms covering a wide range of previous years challenges, and only release the ones relevant to the actual game?
The Greyt claw would cover tubes and a lot off ball sizes. It wouldn’t be too hard to look at past challenges and whip up handful of mechanisms that could cover 80% of them somewhat effectively. It’s not like veteran teams don’t do exactly that basing current solutions off previous designs. You’re just making that experience available to teams that haven’t been around for 10 years or don’t spend all their free browsing other teams’ public CAD.
I agree with this. Our intake this year is almost a literal copy of 1114’s in 2015, because we saw the cubes and thought “hey, this is just a fancy tote!”. Any mechanism that is announced pre-season with relevant design information (like CAD files and the like) and not reconfigured on Day 3 to suit the game exactly is exactly the kind of thing I’d like to see more of.
It’s funny, I came back to this thread tonight after a conversation with a student today reminded me of the debate about COTS components.
The student wanted to convert the team’s cheap, manual Grizzly Mill into a full-scale CNC machine as a summer project. After watching a video and some online discussion about how to do it, we estimated it would take a 2-person team about 3 weeks (100 hours per person) to get the design worked, parts machined, machine assembled and tweaked before relying on it for precision machining in the shop.
We then realized that instead, the student could just go work as a cashier at fast food restaurant for the same number of hours and earn the cost of just buying a new CNC machine.
Both methods dedicate the same amount of time and in both cases you end up with a precise, working CNC machine. However, I wonder which method would have made the student more fulfilled and provided a better experience?
I suppose the pro-COTS folks here would have gone for the fast food job…
Note: I’m not meaning to belittle anyone working in the fast food industry, I’m just picking the job as an example. I applaud any student, etc, working over the summer to help pay for college, hobbies, etc. I worked over school break in fast food as a student myself
I see where you’re coming from but i think you’re missing one key point. Your example ignores the fact that FRC is limited to a six week build season. From my experience the availability of COTS solutions doesn’t lead to teams not working on their robots, it simply allows them to spend their time designing new things instead of reinventing the wheel to align with their manufacturing resources.
In 2015 as a student on a team with limited manufacturing resources the release of the vexpro elevator gusset kit allowed us to redirect our time towards our intake and can grabber leading to a successful and fun season. While I have no doubt we would have developed a workable solution without COTS gussets I feel the team learned more by focusing on game specific tasks.
100 hours is about 3 weeks of work, even allowing for some part-time work. Just for reference.
To put his point bluntly, build vs buy can take the same amount of time and effort (even if the effort itself is slightly different). But who learns more? Who can, if necessary, do the maintenance/troubleshooting without needing to call the tech support?
And that point is applicable to both FRC AND the Real World. I speak from experience on that.
Your students would learn more from the fast food job. Manual Mills make for terrible CNC machines. I’ve never seen a manual mill conversion that doesn’t suck (and I’ve worked on a couple). There’s a reason why almost everyone you see online with a converted grizzly/x2/x3 mill either stops using it after a couple of months or ends up buying a proper CNC machine. A converted grizzly mill would be a terrible machine for an FRC team.
If you really want to convert something because you believe that waisting valuable time is a “learning experience” buy an old CNC mill and retrofit it with modern controls. There are plenty of them out there, I bought a huge 5 ton CNC mill for $500. Even after the retrofit I’d bet it would still cost less than a grizzly conversion.
Maybe next time you need to write a disclaimer that you’re not belittling something at the bottom of your post, you should just delete the whole part you need to disclaim.
I don’t appreciate your sarcastic (or worse, condescending) tone throughout the entire post. And I’m not even the target. If I were you, I’d take your own advise and heavily edit your post. Start by removing “that’s cute”–you’re saying that about an estimate, so how about pointing out where the estimate went wrong?
mman already stated exactly what you were stating, and did it much more constructively.
I hear a lot of discussion around what the students on a team would find most rewarding or engaging. Why on earth would we try to generalize that? I remember a NE team at DCMP last year that was something like 80% programming students and maybe 2 or 3 mechanical students. A team like that might *love *the ability to buy a lift or gripper! I’m sure the programming students would be very rewarded by getting to work with the interesting controls that surround a lift mechanism that they might not have gotten without a COTS solution.
Personally, I dislike COTS items like this, a complete bolt on solution. This is different than small components like gearboxes, wheels, structural members, etc. Those still require work to configure and optimize, the Kitbot is a starting point, a way to at least get you to the event and on the field. Things like the Greyt claw & lift, the REV lift, etc short circuit the learning process, as a current college student I can tell you that looking at the worked out solution and understanding it is not the same as solving the problem on your own and getting help when you are stuck.
When I was a student we built everything custom, it often broke at competition, and while I enjoyed fixing it in a rush, it was a miserable experience for most of our team. The students who graduated from that era have stayed in stem in spite of their experiences, they weren’t inspired by it. Since I graduated the team has adopted a new philosophy and uses a lot of cots components, mostly things from McMaster Carr, Lowe’s, and Home Depot. Linear elevator? Built our own out of garage door track and rollers, worked great, and parts were available at most hardware stores. Ratcheting mechanism to stop the climber from back-driving, broached a nut for the shaft and bolted a Gearwrench on. My experience in three years of Co-ops has been that this is how a lot of engineers do their job, it’s about finding something close to what you need and tweaking it from there, this is an important skill that complete cots solutions deprive the students of.
(The real issue is how do we help a team that only cares about one aspect [Programming] do more of what they care about while not creating an environment where a [Robot] team can’t compete/win, or vise versa)
What is the origin of the price limit on COTS items? I’m not sure I ever learned the purpose.
Hypothetical to those that like COTS items like this: If price limit did not exist, should Snow Problem sell their Ri3D robot (all pieces, painted that pretty blue color and everything) as an upgrade kit alternative to AM14U3?
I am not giving my own opinion here, but if it’s a question about where the line is drawn, where are you comfortable drawing it, and why?
Currently we live in a free-market FIRST economy with some regulation on COTS price and availability. Genuine question for discussion.
It would be literally impossible to assemble the elevator without gaining a full understanding of all bearing interfaces and the various forms of power transmission.
The claw has adjustable width/travel and doesn’t ship with wheels or a suggested configuration. It would be impossible to make a functional claw out of it without gaining a good understanding of your own specific configuration.
Both are currently far from a bolt on solution.