Back to my point you need to speak to your RSM to get more information. I am not sure where people think VEXForum is the place to get definite information regarding RECF operation issues. Sure team members may not think to reach out to RSM, but the responsible adult for each team would know that.
And in this case, reprogramming doesn’t damage the device.
As a competition, all information relevant to competing teams should be public so no team may gain a competitive advantage using privately obtained information.
The forums are a public venue, private conversations with your RSM is not.
The official q/a is public for this exact reason.
RSMs are told to redirect teams to post on the public q/as when they have a question about rules.
Information about what is included in registration can give teams a competitive edge, so I would expect this information to be able to be obtained in a public platform instead of a private setting.
Its not unreasonable to assume the public forums in which the CEO of REC regularly visits is a good place.
Honestly, I am very surprised at the negative comments on this thread. The REC Foundation and VEX Robotics worked very hard to provide an exciting and challenging new program for both High School and University teams that wanted more flexibility to build multiple robots with custom parts and experiment with AI. This is the PILOT YEAR!!! We will learn and make adjustments after this first season. If you feel this program doesn’t have value for your organization, I get it. Don’t sign up. But we have a waiting list for organizations that do want to participate. This is not a failure just because not everyone on this forum agrees with how we are launching the PILOT program. Constructive feedback is always welcome but this is not some out of touch money grab and the RECF is not some incapable organization because we are trying to launch a new program without a sponsor in the middle of the pandemic.
The RECF is providing the same hardware bundle to all teams. We pre ordered the hardware to make sure all pilot teams had what is needed for this first season. The costs with shipping, operations, storage, etc is around $1300. The RECF is charging approximately $700 in registration fees: $200 towards IT / System costs to launch the new program and $500 to administrative costs for this first year. We have one staff person assigned to VAIC and we have to train RSMs to support EPs. This is a real cost. The RECF is also paying $500 per team to VEX for the VAIC software. The software took many man years to develop and I am very happy to pay VEX to use their software.
I met with multiple Universities in many countries to discuss the need for VAIC. The feedback I received is that we are providing a great learning opportunity at a very affordable price point. We opened this to High School teams because of the feedback that there are many high school students that will benefit from this opportunity. From our very first communication, we let all teams know the cost will be $2500. This is not a surprise price that we just dropped.
I hope this provides some clarity.
Regards,
Dan
Today we leaned the reality of bringing a product to market,
- There is fixed costs for the hardware
a) Supply chains have issues. While they could support this via China mfg, it’s hard to get that in three days
b) It cost money to put products on shelves. You think @Grant_Cox has a Mohawk, that’s just longer hair from him pulling it out because of long storage times of products he’s paid for sitting around, - There is fixed costs for the software
a) These are often called NRE (non recurring engineering costs. We see that in the cost of VEXIQ parts with design time, creation of injection molds, testing, etc. Companies recover that cost across the “profit lifetime” of the product. Remember the “cost of money”, you don’t want to wait decades to get your money back.
b) Open source is fun, but this first release needs to be dead solid. We all watched a release of V5 development that crashed and burned. Nobody at VEX/RECF wants to do that again. - There is a recurring cost for software
a) Software changes, we’ve watched multiple releases of VEX 0.5, VEXIQ, V5 firmware. That isn’t a free effort, that shows up as a cost. Presently the “cost” is bundled into the price of VEXIQ and V5 brains. So we are paying a NRE an a Upgrade cost to each brain to keep the software pipleine growing. We should expect the cost of VAIC to not drop next season as the software improves. - Marketing – RECF has assigned a few people to this program to roll it out and they need to get paid. They do a ton of coordination to make all if this work, and the last time I checked they don’t eat “weenie beans” every day.
- Risk – This is a risk move by both RECF and VEX. VEX put money up front to build this, RECF needs to make this a flying program. If it crashes it
a) Paints a bad coating on other programs
b) Shakes the relationship with VEX for future long reach endeavors.
Most of you go “Buuuut Raspberrry Pi!” And lets take that apart for a few seconds.
- Raspberry Pi is supported by a foundation of lots of people (including me) that gives them money to do what they do. Marketing, Software and Hardware design are things the foundation pays for. Yes, the software is helped by people that contribute to the ecosystem, but the bytes to metal part is paid for by the foundation.
- A small part of the cost of the PI is returned to the foundation, but if you look at the supply chain (Sony->AdaFruit->Foster) there is profits at two levels. While hardware is becoming super cheap, both Adafruit and Sony are making a profit. And I don’t begrudge either one of them for doing that.
Let’s circle back to VAIC. It’s a new program, new start up costs, new risks. You want to do this, sign up. If it’s too risky then as @DanMantz said, don’t do it.
Let’s remember some key points.
- VEX is a for profit company. They have lots of people that they employ, both the management and the employees would like them to stay in business. So they need to make a profit.
- RECF is a non profit company, but that does not mean they don’t need the money. Like VEX, they have employees and the employees would like them to stay in business. So they need to make a profit. WAIT FOSTER they are a non profit. Exactly, but the “profit” is what they pour into the next big thing. In this case, they didn’t up the overall cost of VRC by $5 to cover this, they put the burden on VAIC. (And I’ll guess that this was a much talked about item, but they came down on “user pays”)
You have been afforded great transparency. Applause is needed, rocks are stupid.
I’m not doing VAIC this season, I don’t have the University / High School support I need. I could maybe wrangle one (Deltech + HS) or a second (UoD + HS) this season. I’m also about future building, if I have HS roboteers and University Robtoteeers, I can become a mini pipeline. I can’t get traction to support split teams. But in this year of crazy, I can pass on being a pilot.
Thanks for wading through this…
So, I agree with most of the post, insofar as “the costs are what they are, the RECF isn’t just Scrooge McDuck trying to fill their money bin”. Of course, just because a price is the accurate price (insofar as it represents the cost of the product) that doesn’t mean the price is worth it, but apparently there is demand. Whatever, if the only issue was cost, people can make their own decisions. Complaining on the forum isn’t going to magically bring the very real costs down.
How does making it open source make it less solid? I assume that VEX is using version control. Just… post the code on GitHub. There is zero reason that making it open source for people to look at (I’m not talking community contributions, I’m just talking “here is the code, learn from it, if you want to try to modify it cool, but you’re on your own”) would make it more or less solid. Just post it online, say “here’s the latest version of the code, it is what it is, learn from it if you want.”
There are two reasons I can think of why they are keeping this closed source. One, VEX is licensing some of their code in turn from some third party, and as part of that they can’t release it. OK, I guess, makes sense.
Option two is VEX owns all the code, and wants to keep it proprietary as some competitive advantage because otherwise FRC or whatever can have the same thing, and then the super cool auto vision sensor thing isn’t a unique selling point, or they plan to also license this code out for like private actual real world use or something. At which point as I see it, VEX the business is really really getting in the way of the nonprofit education mission of RECF. I don’t see how the inclusion of these sensors adds meaningfully to the programming learning experience, since you’re not learning machine vision, this is more like being handed the answers on the test.
I dunno. I got my first internship and most of my initial programming experience from what I learned doing VEX in high school back on the Cortex. And it seems like a lot of the recent moves have been “let’s just give the end user magic black boxes to they don’t have to understand what’s going on, we do the thinking for them.”
Technically, I don’t think anything is stopping you from writing your own code. The Jetson Nano is, after all, a developer kit. But I have to agree that I dislike the black box approach, especially when those black boxes are vision sensors. Vex did handle the PID control on the motors pretty well though.
Thank you Dan and Foster for the clarifications and good points. Firstly, I want to say that, at least for my posts, the intention was never to attack the RECF or VEX for how VAIC is going, but rather to try to get more information on what is being provided and to also provide constructive feedback where I can. I apologize if it came across in a different way than I intended.
I want to comment on one thing:
I agree with you here, though I want to make clear that I am not saying “VAIC is a failure if the RECF doesn’t do X”; rather, I am saying “I think that the RECF should do X, because I think it will help the program better fulfill its goals”. I understand that I don’t have the same perspective on the program as yourself, but that is precisely why I feel the need to provide such feedback. Also, if any of these points are already things that the RECF is planning for the future of VAIC but are just not feasible in the pilot year, that makes complete sense.
I appreciate that the RECF needs money to cover the cost of the program and that the hardware and software have costs, and I thank you for providing the breakdown for these additional costs. While I personally can’t justify advising any team I’m affiliated with to join the pilot (for a multitude of reasons that go beyond the pricetag), for those who are still considering, I think the one thing that’s still missing from this thread is a confirmation of what comes in the box. So in that vein, is the (educated guess) hardware list in the original question correct?
I’d also like to address this:
I disagree with the notion (which one might from Foster’s post) that Open Source is inherently less reliable (as I type this response in an open-source browser running on an open-source operating system); moreover, the V5 issues had nothing to do with open-source. In fact, I can guarantee that the Jetson Nano firmware image will include many open-source components (e.g. the Linux kernel), including likely ones used in VEX’s software package (I bet OpenCV is used in there somewhere).
But at the same time, there are legitimate reasons why VEX might not want to release everything in an open-source, modifiable system; notably, to paraphrase Levi, they would be creating a nightmare for their support teams if making modifications to said code became a common approach. This is especially a valid concern for the pilot phase, where the goal should be proving and troubleshooting the model for the competition, rather than providing as much breadth as possible.
That being said, this is all kind of missing the point of why the licensing and documentation nature of the code was brought up in the first place, namely that making the vision system opaque creates a barrier to learning, and given that teams are required to purchase the software, there is a fairly large sunk-cost reason to use it. Moreover, there are valid concerns from teams that the system will not be adequate, and the closed nature means there’s no way to make incremental improvements on it. At the same time however, as I said earlier, I recognize that this is a pilot year and that the program will not be ideal immediately; my goal here is to ensure that these issues are addressed for the “non-pilot” seasons.
Lastly, I want to address this:
I completely agree with the latter response here. The reason why this thread exists (and why, when I did reach out to our RECF rep, I posted the information I received here), is to ensure that everyone in the community is properly informed and can raise valid concerns. As I said before, I’m not trying to bash the RECF; rather, I’m trying to ensure that information is known and legitimate concerns are heard.
Dan, Levi and others, thank you for taking the time to read and respond to this thread.
Hi, I wasn’t saying that Open Source was a problem, but the initial release uses VEX internal API’s . I hope that the software ends up Open Source, I’m a huge fan and have submitted things that were accepted. I was trying to say that VEX has an interest in keeping things locked down and working well.
BTW, non-pilot seasons means the first season goes well. Many of us are looking forward to season 2!
Thanks!
Ah, thanks for the clarification, but I’d also like to clarify that my concerns are regarding the software on the Jetson Nano; after all, that is the software which is actually implementing the vision system.
Aside
If you’re referring to the parallel situation of VEXlink, then yeah I completely understand and support the work VEX has done there, and the reasoning why it currently is only supported in VEXcode (which is closed-source), and not yet in the Partner SDK used by PROS and others.
So there are a few 1000 lines of code going out here, nobody yet know what it does, but we all know we will blame Levi for making it 
Still way to early in the season, remember “Easy to tell Pioneers, they have arrows in their backs”
Hi Nick,
I do want to clarify that I can tell you are providing constructive feedback to be helpful and I personally appreciate it. And I appreciate your respectful tone. Thank you.
To everyone: Again, I think it is important to remember that this is a Pilot program and that we will learn from this first year and will make adjustments to make VAIC better.
Thanks,
Dan
I can’t tell if this is in jest or not, but the only “blame” that’s useful in software dev is git blame. I’m here to point out ideas, not to point fingers. Cheers.
Thank you Dan, and I appreciate the work that’s being put in all around.
Happy to help out. Not often we get to interact with a CEO on an open forum. “Hey Mark Z” can i get you into a conversation"
RECF transparency is @DanMantz best trait since RECF started. +1
Ahhh time does dull memory. Let me refer you back to the V5 initial code release. That didn’t go over well with the masses.
I apologize for using @levipope in this thread, his willingness to stand up and talk about software is appreciated. Sorry Levi for the slur.
Foster
Let say people already have tools that do the job. Then, if manufacturers want to successfully sell them new tools, they need to consider {value proposition/quality vs price} and {supply and demand} curves in order to plan R&D budget and bring competitive products to the market. Expensive products will not sell well if they offer no significant value over the competition or what people already have (unless there is a monopoly and buyers have no other choices).
My guess is that the eventual high volume target market that VEX had in mind for their VAIC hardware were High School labs, where they may not have mentors with @tabor473 or @nickmertin level of expertise in AI, and advanced stuff (like on-field GPS, depth vision, robot-to-robot comm, etc…) needs to work out of the box. That must have determined system architecture requirements and R&D budget.
When they were estimating costs and modeling supply and demand, they may have counted on sponsors offsetting the costs in the first inaugural season and they, obviously, had no idea at a time of the initial planning how everything will be turned upside down with the disastrous coronavirus response.
More than six months into pandemic there are still huge uncertainties everywhere. We don’t know much about long term coronavirus health impacts, we cannot predict how virus will evolve in the Fall, and we cannot be certain that people will be disciplined in following any social distancing, masking, and vaccination guidelines. Judging by the first few school reopenings that we’ve already seen - teen and tween behavior may be a source of major risk in spreading the virus.
Even if there is a small chance that herd immunity could happen sooner than expected, it would be foolish to count on it. Time after time covid numbers had turned out worse than expected. Given all that uncertainty and downside potential, it is prudent to increase the safety margins and, for the schools, that could mean to preemptively cancel all optional activities or even cut in-person classes and labs. Many schools already did that and it feels likely that lots more will have no other choice but to do the same in the near future. For VEX that could be translating into collapsing hardware demand and revenues for the next six to nine months.
I wouldn’t be that categorical. Although it would be super nice to have open hardware and software, which could easily connect to and communicate with the third party products. I do believe, that it is a better business strategy in the long run, because it leverages community contributions and gives potential for more diverse customer base than just school labs and robotic teams. But we are not really qualified or informed enough to give advice to people running IFI.
As as a consumer I could definitely say that I would be buying much more of their products if I knew that I could easily re-purpose, re-program, and inter-connect them with third party sensors and micro-controllers. For example, I love EDR structural components, sensors, and 393 motors because it is so easy to connect them to a cheap controller like Arduino or Node MCU, but anything V5, in my opinion, is not worth the money for an average hobbyist because of its closedness.
After all, people who could be the best VEX champions/advocates are current or future engineers who are very good at comparing costs of the potential solutions. This topic is a good illustration that closed hardware and software model just don’t carry as much of the perceived value anymore and don’t seem to have enough flexibility to rapidly adapt to the unexpected conditions.
The bottom line is that VEX and RECF need to keep their business financially afloat while the usual demand is not there. Their bet is that they will still have enough demand at the current price point. If not - they could either decrease the price in the hope to increase the volume, or somehow increase the value of the offering, for example, by opening their source code or interfaces. But that may be conflicting with the business model for the future revenues.
Could there be any other options to make everyone involved with VAIC happy this season?
To be clear, the reason I said speak to your RSM was to get as much information you need to make a decision about to participate or not. As a pilot year, lots of unknowns and decisions were being made about specifics of what would be in the kits. Once finalized, they get published with more details. If you needed to make a decision quickly because your school/organization needed it for the purchase order, then work with your RSM.
Most schools would love to participate in VAIC but when you go to get funding some teams need to know what’s in the kit to be able make the decision for the school. Not all schools can just drop 2.5k and not know what is in the kits. You have to be able to tell the district or whomever you go for funding what it is for. Every school doesn’t have unyielding support especially in these times with funding. It is not like it is anyone’s fault but as you said the list isn’t finalized which is cool and all that you have someone who knows that info and sees the inside works. In the end some schools need to know what’s in that list and see if they would rather wait a year or get in on a ground breaking competition. You shouldn’t judge for situations you don’t understand. If some school is trying to judge risk and reward they need every part of information because I doubt a refund would be an option. If they were to keep the equipment what benefit could they use it for and if they can use these parts for other purposes and still use it for something. I get that they could simply practice for next year and get familiar with the equipment for next year but sometimes you need a little more incentive considering the main focus would probably just be VRC or VexU rather than AI. We know the cost and that worlds will be extra now all that’s left is what comes in the kit and the sooner the better. If the kit isn’t finalized then what is finalized could be confirmed because I was digging for a while to find what comes in it and couldn’t find but what was assumed in this thread. An official statement would put all of this to rest. If the list isn’t finalized I doubt the RSM would be first to know.
I am not sure what you are talking about - it is not judging to give an expedited route to get firm answers for the scenario you describe. If there is insufficient information to make an informed decision, then don’t opt-in. Risk-reward balance will look different from organization to organization.
My original point about getting information expedited through RSM vs VEXforum is that RSM is more likely to get you the information sooner than waiting for a post on VEXforum.
As it stands, it seems they have the number of interested teams for the pilot.