I can see this affecting Limelight stock given the relatively small company size and hindering future runs of Gloworms given that it is just someone in their college dorm.
I do board level electronics design in my day job. We are seeing long lead times (over a year) on a few semiconductor devices of all types; processors, FPGAs, linear devices, discrete devices. We are also seeing similarly long lead times on some passive devices; resistors, capacitors. It’s not a lot of parts but if you can’t get a $0.02 resistor, you are not shipping. Finding alternative parts can be difficult because other people are also looking for the same alternatives. We also have a lot of parts qualified for operation at temperatures well above their rated maximum operating temperature.
Good news, all rookie FRC teams will be receiving first generation Jaguar motor controllers in their kits!
It’s more like 12+ months in advance. Single source items are problematic and that’s a very large portion of the semiconductor market. Fab capacity is maxed out!
Yes. This is actually going to hit us hard. I know we are in the minority for anyone using a Jetson Xavier NX (current prices are around $1500 for what was a sub $500 dev kit) but this will and is genuinely going to impact us.
I’ve long been critical that the rules for FRC in regards to cost accounting make absolutely no sense but this is one of those things that is going to impact teams using any sensor or embedded system close to the magical and arbitrary $500 limit that they have imposed.
I’m sitting on an email to the teams account to ask what they plan to do and questioning a solution but I’m holding off on sending it at the moment. My emails seem to get ignored, made fun of behind my back (which gets back to me, HQ leaks as much if not more than than the EWCP slack does to Adam Heard), or they respond asking me more questions but then stop when it turns out that I actually might be competent and not just 5 squirrels in a trench coat.
These shortages are going to impact supplies for several years. This isn’t going to be a short term thing for a single season.
What could FIRST do to help mitigate the effects of this shortage?
(not disagreeing with you or saying FIRST cant do anything, just genuinely curious)
This post was a wild ride.
Don’t underestimate the competency of 5 squirrels in a trenchcoat. How did they get there? Where did they acquire the trenchcoat? These are questions nobody ever stops to consider.
Seems that increasing the $500 part limit or adding an exception for electronic components would solve Marshall’s issue in particular. This already existed to some extent in the 2020 rules - there was an exception for all Rockwell Automation components given out as any part of any year’s FIRST Choice and an exception for all IMUs in the total BOM. I don’t think it would be unreasonable to expand that list slightly and apply it to the single part limit rather than the BOM to include things like coprocessors and some more types of sensors in order to accommodate price increases. Alternatively, the definition of FMV could be adjusted to be the lowest MSRP a part has had in its history or the price a team acquired a part for, provided they bought it from a VENDOR. I’m not quite sure how that would work for any new products that come out mid-shortage that launch at a high price, but I suppose we can cross that bridge when we get to it.
That would have my vote.
This won’t work in the long term.
It fails to address the inflation problem that is going to happen over the next few years. Prices are going to go up and they likely aren’t going to come back down. So this rule change kicks the can for anything that was created before now-ish but it won’t last beyond a season if the trend continues and lots of component suppliers seem to think it is going to. I can’t tell you what NVIDIA is going to do with their dev kit prices but prices across the board are increasing.
As-is the amount has long since not tracked with inflation from when it introduced and more than a few supplier provided parts were above the amount in terms of FMV.
I’m not sure I understand what a $500 part in FRC buys a team that a $1000 or $2000 part doesn’t? What is this limit designed to stop exactly? Ohh no! They purchased a more reliable gearbox or a better computer!!! If they have the budget to spend on those things to go on a robot then good chance that budget was going to be used to just buy those $500 versions and use them as wear items or, more likely, it isn’t going to help them substantially more anyway because the value of the part isn’t the determining factor.
Aside from that, the way the rule is written, it’s more than ambiguous when it comes to certain types of kit items and how to break down the costs. The blue box I think for FMV is massive and longer than a single page, IIRC.
I think this rule exists to make people feel better that teams don’t spend hundreds of thousands on FRC robots but the amount of iteration and engineering time that some teams spend, I’m guessing the true value of an FRC robot is shockingly high.
It’s predominantly going to impact the semi-conductor industry for a few years right now but that has some impact on supplies in secondary markets too. There are also other things suffering from shortages right now.
My preferred solution is to just make the cost cap vanish and address the “out of control spending” in other ways, which to me is far more productive but it’s not likely to happen.
More realistically, carve out an exemption for any sensors, embedded systems, and/or electronic items.
Increasing the allowed component costs mitigates this issue for the teams that can afford the bump. What will this do to low/mid level funded teams that have to look hard at the price increase in Raspberry Pi’s?
Is the price cap for FRC a determining factor for the funding for those teams? I don’t think so. You are welcome to prove me wrong.
I get it. The goal should be to somehow create equity on the competitive level but the cost caps don’t do that. They never have.
Those teams are going to have to deal with that limitation in their funding regardless of the cost cap for an individual part being $500 or $5000. The same as everyone else is going to be dealing with it. The price limit that FRC has in their rules has absolutely no impact on what the Pi folks decide to price their goods at.
One way to lower costs for FRC participation would be divorce the KOP costs from the registration costs. I think a lot of folks believe that the KOP is largely donated… so being a bit more up front with teams about those costs and enabling teams to make choices with their funding is something that would help, I think.
Hey, just want to say: l love this question, thanks for asking it. It’s important to make “FIRST” into a stronger networked community of problem-solvers (us), rather than a black box in the Manchester millyard (them).
Ideally, FIRST (likely though a sponsor) secures stock of those components at a discount, or ahead of any price increase, and expands the FIRST Choice program to include more parts teams will actually use. If they can’t do it with Pis, get another Linux SBC and attempt to have our community standardize on that. This doesn’t just go for Pis, they should do it with anything that teams use often and could face price issues in the coming years. It’s not entirely feasible for everything, and I’m not sure if it should even be FIRST’s responsibility to secure and distribute parts for teams in this way, but it would sure be nice if FIRST could use their status and preexisting systems to help teams through this.
At the same time, Pis are reusable. If their price goes up, it’s not going to affect what you already have. The Pi Foundation has stuck pretty hard to the $35 price point, so it’ll be interesting to see what they do if a Pi 5 is coming up soon, though I doubt we’ll see it for a while.
if hq can’t even demonstrate whatever “gracious professionalism” means what does that even mean for the rest of the program.
is it just a meaningless phrase people use to bludgeon out things they don’t want to hear?
Meh. Making fun of me hardly qualifies as lacking gracious professionalism. Probably the opposite is true.
“Five squirrels in a trench coat” is my new favorite euphemism for “incompetence”.