New CRIO in the KOP next year?

[MrForbes]

Originally Posted by oddjob

If it's a lot smaller and lighter, that's a big advantage. The cRIO is a boat anchor and not well suited to a size and weight limited robot where inches and ounces are precious.

I have trouble considering it to be a boat anchor...I have too much old tube gear laying around

The cRio does have an interesting design, not quite how I would do it, but it takes up far less real estate on our electronics board than the power distribution and speed controllers do, as usual. We took the time to figure out how to mount the cRio down low, with the connections easily accessible, and it really wasn't a size/weight issue at all.

[Greg McKaskle]

Originally Posted by squirrel

I have trouble considering it to be a boat anchor...I have too much old tube gear laying around

The cRio does have an interesting design, not quite how I would do it, but it takes up far less real estate on our electronics board than the power distribution and speed controllers do, as usual. We took the time to figure out how to mount the cRio down low, with the connections easily accessible, and it really wasn't a size/weight issue at all.

Glad to hear someone has a normal boat that can't be held in place by 929 grams.

Anyway, to explain what was in the NI booth, we actually had three generations of the cRIO. The drop test was the original generation of modular system. It consists of a cpu module, a backplane module, and the I/O modules. The backplane has the FPGA and comes in four and eight slot sizes. This rectangular packaging allowed the controller to fly to around 4000 ft in a rocket -- slides into the tube like a loaf of bread along with the battery.

The FRC cRIO is called the integrated unit. It is lower weight and lower cost because the cpu and backplane are integrated into one PCB. It wasn't officially released when discussions about its use began.

The third generation on demo is often referred to as the single board RIO. It has the module connectors on the edges to flatten the product from a loaf of bread into a pizza, and in fact doesn't come with a housing since it is designed to be integrated into a cabinet on large machinery, etc. This product was also under development when discussions began, but wouldn't be released early enough, doesn't have an enclosure, etc.

Greg McKaskle

[mwibbels]

Originally Posted by Greg McKaskle

Glad to hear someone has a normal boat that can't be held in place by 929 grams.

Anyway, to explain what was in the NI booth, we actually had three generations of the cRIO. The drop test was the original generation of modular system. It consists of a cpu module, a backplane module, and the I/O modules. The backplane has the FPGA and comes in four and eight slot sizes. This rectangular packaging allowed the controller to fly to around 4000 ft in a rocket -- slides into the tube like a loaf of bread along with the battery.

The FRC cRIO is called the integrated unit. It is lower weight and lower cost because the cpu and backplane are integrated into one PCB. It wasn't officially released when discussions about its use began.

The third generation on demo is often referred to as the single board RIO. It has the module connectors on the edges to flatten the product from a loaf of bread into a pizza, and in fact doesn't come with a housing since it is designed to be integrated into a cabinet on large machinery, etc. This product was also under development when discussions began, but wouldn't be released early enough, doesn't have an enclosure, etc.

Greg McKaskle

Greg, Can you give any further information on the cost targets, the schedule, the size, and the interface options for the 3rd generation cRio? From your description it is not clear to me that it meets Al's desired cheaper, simpler, and smaller goals. Which by the way I share

Thanks,

Mark

[Bob Steele]

I will echo what Dave said earlier. Everyone needs to weigh the costs of all decisions. One good example is this year's playing surface.
Assuming we don't use it again... we can look at the money spent like this:

Cost of a whole field of FRP was something approaching $1500 at undiscounted prices.

If we assume that every regional area probably had at least two or more practice fields.... This is a cost of 40 X 2 X 1500 = $120,000

If 1000 teams each bought 2 sheets at $30 a sheet that would be an additional $60,000.

So in a very conservative estimate... FIRST teams spent $180,000 on FRP this year.

Now the real cost is probably closer to $250,000 just for this material.

And yes... the cost of carpet would have been comparable... but carpet was used many times...in fact the carpet we used under our practice field was from 2 years ago...and you did need carpet under the FRP to match field conditions...

It was an interesting game... but in terms of how costly... the field was high cost.... and not reusable...
so the green cost was also high... FRP is probably partially made from petrochemicals...and what do we do with it now?

These kinds of issues... need to be considered...
Cost is magnified as we get bigger...

If everyone needed an extra Driver's Station this year because they needed a spare... that would be a cost of approximately 1500 X the $162 cost
or $243,000 Even if only 500 teams needed extra driver's stations the cost is $81,000 to teams...

If 1500 FIRST teams could each save even $250 from their cost of entering the competition that would be $375,000

This perhaps is the most telling statement.... FIRST would not get that money...

There is no good answer,
Next year's game should have someone on the GDC that oversees exactly this kind of thing...

both the $$$ hidden costs....
AND the Green costs of the game.
We need to start reusing things....
And we certainly should not be wasting anything...

Perhaps at the end of the season teams could get a rebate on the next year by sending BACK certain unused materials...
This might make more sense than a shopping cart approach to the kit.

just an idea

Anyone want to purchase some used FRP?

[Greg McKaskle]

Greg, Can you give any further information on the cost targets, the schedule, the size, and the interface options for the 3rd generation cRio? From your description it is not clear to me that it meets Al's desired cheaper, simpler, and smaller goals. Which by the way I share

There are currently eight SB-RIO models described on the web site which were announced and shipped this spring. The configurators do not include pricing on the SB-RIO, presumably because they are targeted for OEM usage. If you really want pricing for nonFRC projects, you are supposed to call. My info is from http://zone.ni.com/devzone/cda/tut/p/id/7441.

The original cIO has traditionally been used for prototyping and low volume systems. The SB-RIO is a highly compatible version better suited for integration in higher volume production. If CERN were building dozens more Large Hadron Coliders, perhaps it would be cost effective for them to design a housing and switch to the SB-RIOs . Seriously though, the FedEx fire suppression system project does a better job of demonstrating the relationship between the original and SB-RIO. http://sine.ni.com/cs/app/doc/p/id/cs-11310

As for size -- It is different. The integrated cRIO is a loaf of bread. If you remove the PCB from the housing, flip the I/O connectors to the edges, you have something more like a square-ish pizza. There are photos of SB-RIOs on the above links.

I agree that this may not meets Al's cheaper, simpler, and smaller goals. The SB-RIO was developed independent of FRC, and should not be viewed as an attempted optimization for FRC.

Greg McKaskle

[ZInventor]

as a student from a rookie team, i really hope we do get another CRIO + all stuff to go with, since we don't have the money to buy one (we blew it all on going to Atlanta) and would really like to have at least two (one for the comp bot, one for a practice bot/demo bot)

overall, i like everything about the CRIO other than weight (but i'm not a programmer... other than web)

^^^ same goes for the jaguars..... (oh, and the part about them dying if they get even the slightest bit overloaded... the Victors were much more reliable (i was on another team before starting this one) and MUCH lighter...)

just my $.02

-Z

[EdwardP]

Originally Posted by ZInventor

as a student from a rookie team, i really hope we do get another CRIO + all stuff to go with, since we don't have the money to buy one (we blew it all on going to Atlanta) and would really like to have at least two (one for the comp bot, one for a practice bot/demo bot)

I'm sure you have probably thought of this, but we just didn't ship our CRIO, and brought it with us to our competitions.

[ZInventor]

Originally Posted by EdwardP

I'm sure you have probably thought of this, but we just didn't ship our CRIO, and brought it with us to our competitions.

true, but it's nice to be able to have multiple versions (ie. ne for programmers to work with, and one for the drivers to practice with...

it's also a hassle (easier w/ the CRio than IFI RC) to swap it on and off the bot...

also, it'd be fun to have two bots at off-season demo's for potential sponsors, that way, we could show off what it's all about...


-Z

[AdamHeard]

Originally Posted by ZInventor

true, but it's nice to be able to have multiple versions (ie. ne for programmers to work with, and one for the drivers to practice with...

it's also a hassle (easier w/ the CRio than IFI RC) to swap it on and off the bot...

also, it'd be fun to have two bots at off-season demo's for potential sponsors, that way, we could show off what it's all about...


-Z

I really overestimated how much of a hassle swapping is this season. I saw 254 swap it in/out at vegas and it really didn't take much time at all. Network cord, power input, modules, whatever fasteners, done.