**FIRST EMAIL**/Future of FRC - Sneak Peek

[T3_1565]

I hope its not NXT based as well... I kinda like the RC the way it is! If it ends up being like Lego Mindstorm I'm going to be angry

Although I highly doubt they would do that to us...

I'M AFRAID OF CHANGE!!! DON'T DO IT!!!

[whytheheckme]

I really wouldn't mind change, as long as the new interface is highly customizable. Lots of inputs and outputs, of all kinds, and lots of onboard features. Language doesn't bother me too much.

In my mind, the current RC seems quite outdated. It looks the same as it did in 2001. Wireless, deployable sensor capabilities (bluetooth), direct support for Serial and USB devices, and higher bandwidth inputs and outputs are all things that are here and now, and would be a great teaching utility, not to mention it'd open lots of opportunities. Having the bandwidth of 802.xx and allowing for video feeds, off-board processing, real-time 3D modeling, etc, would be AWESOME, and really would not cost that much. If the price of the KOP had to go up $500 or $1000 for a super-advanced control system, I think it would be well worth the cost.

Jacob

[JesseK]

Originally Posted by whytheheckme

If the price of the KOP had to go up $500 or $1000 for a super-advanced control system, I think it would be well worth the cost.

Jacob

I disagree. There should be better technology at a lower cost for larger volumes of processing than there was a few years ago. Hence, there should be almost no reason for the KOP cost to go up just because a new RC has a few more widgets. We've already figured out some pretty advanced things to do for control of an FRC bot with a simple processor. Aside from 80-dimension closed-loop control systems whose backend processing most students can't comprehend until a couple of years into college, what else is there to do with an RC? There are only so many ways to tell a machine to move.

I know I exaggerated a bit, but NASA put a man on the moon with less processing than some FRC bots. Think about that for a minute.

[Adam Y.]

Originally Posted by JesseK

I disagree. There should be better technology at a lower cost for larger volumes of processing than there was a few years ago. Hence, there should be almost no reason for the KOP cost to go up just because a new RC has a few more widgets.

The increase in cost would be from making sure the thing works not the processing power. NASA puts stuff into space all the time with processing power less than computers from the mid 1990's and yet those processors cost more than your entire networth. Of course NASA is a really bad example of the cost inflation because getting electronics to work in space is not trivial.

I guess I'm joking, I hope, since there's really no way for Lego, NXT or Robolab to be part of FRC. But nothing would really surprise with FIRST right now.

Why though? If done correctly you could theoretically come out with a great program. Out of the four kits the controller using the most advanced processor is the NXT.

[whytheheckme]

Originally Posted by JesseK

I disagree. There should be better technology at a lower cost for larger volumes of processing than there was a few years ago. Hence, there should be almost no reason for the KOP cost to go up just because a new RC has a few more widgets. We've already figured out some pretty advanced things to do for control of an FRC bot with a simple processor. Aside from 80-dimension closed-loop control systems whose backend processing most students can't comprehend until a couple of years into college, what else is there to do with an RC? There are only so many ways to tell a machine to move.

I know I exaggerated a bit, but NASA put a man on the moon with less processing than some FRC bots. Think about that for a minute.

I guess I was talking about packaging costs. For instance, the IFI RC currently uses the PIC18F8722, which costs $8.36 apiece (bulk) from Digikey. The reason that the RC costs $449.95 from IFI is all of the design and testing costs and packaging costs associated with making the product.

Designing an interface with all of the features that I mentioned, plus more, would have huge overhead. Because the FIRST market is limited, the cost per unit would be high. Seeing as the Lego® NXT system is $250, assuming worst case scenario that the controller only is $150 dollars of that cost, that is still really expensive for something with such a broad market. Imagine if something of similar or greater complexity was designed for a market of 2000 units per year, instead of hundreds of thousands of units per year. Instead of $150 per unit, the cost could easily be near $1000 per unit, compared to the current RC, which is $449.95.

Could they build it for 450 bux a unit? Yeah. All I'm saying is that I wouldn't mind paying more for the features. If 200 bux meant the ability for use of bluetooth devices onboard, increased wireless bandwidth, and a USB controller, that's the best 200 dollars that our team has ever spent.

Jacob

[Racer26]

I'm going to venture out on a limb and suggest that IFI would have a VERY hard time trying to make a patent on 802.11 as a means of controlling robots enforceable. It depends entirely upon your definition of a robot, and anyone else doing it, could simply say it was communicating to another computer (RC is a computer, which controls the robot). Problem solved.

[Corey Balint]

Originally Posted by 1075guy

I'm going to venture out on a limb and suggest that IFI would have a VERY hard time trying to make a patent on 802.11 as a means of controlling robots enforceable. It depends entirely upon your definition of a robot, and anyone else doing it, could simply say it was communicating to another computer (RC is a computer, which controls the robot). Problem solved.

I think IFI, if they wanted to, would have been smart enough to make it more specific and figure out any loopholes that they might have in their patent application.
If they wanted to block out FIRST from using the technology they helped develop/produce, I'm sure they would have taken every necessary step and made sure everything was very in depth in description.

[Brandon Holley]

Originally Posted by 1075guy

I'm going to venture out on a limb and suggest that IFI would have a VERY hard time trying to make a patent on 802.11 as a means of controlling robots enforceable. It depends entirely upon your definition of a robot, and anyone else doing it, could simply say it was communicating to another computer (RC is a computer, which controls the robot). Problem solved.

Patents are pretty complicated documents. I have my name on a couple, and it confused the heck out of me the few times I was involved. There is a lot of room in the patent to cover up the "loopholes", not to say that there aren't any, but it appears that IFI has 3 patents on this kind of thing, so I am willing to bet they have a lot of those holes covered up.