2012 Team Update 4

[Phyrxes]

Originally Posted by waialua359

Reminds us of VEX season when we stick our robot almost off the blue/red tile, only to have a little zip tie touch it.

I am now waiting for the "magical zip tie" to get ruled an illegal traction device...

On a more serious note can we actually get some ruling on rule G33 that actually explains what they really want?

[TD912]

The alternate "Bondé" radio is bigger, heavier, and has external antennas, compared to the more compact radio with internal antennas currently being used. I'm guessing the DIR-825 simply has better reception with the larger antennas than the compact DAP-1522.

I don't think anybody expected for a situation to arise where the current radio bugs out when there are over 60 APs in one place drowning out a single FIRST field AP. I'm imagining FIRST does test for some radio interference, but not extreme interference coming from over 60 WiFi APs at once.

I don't think there will be any major issues with the current radio, but it's good to see FIRST is thinking about these things.

The current radios will be set to only use Wireless N over the 5GHz band, which means there shouldn't be any interference from all the old APs on 2.4GHz.

The only way I can see this becoming an issue is if the venue recently upgraded their entire wifi network and installed tons of brand-new, high-end, 2.4/5Ghz dual-band, Wireless N APs relatively close together.

[EricH]

Originally Posted by waialua359

Reminds us of VEX season when we stick our robot almost off the blue/red tile, only to have a little zip tie touch it.

Remember 2005, Glenn? With those loading zones?

[linuxboy]

Originally Posted by TD912

The alternate "Bondé" radio is bigger, heavier, and has external antennas, compared to the more compact radio with internal antennas currently being used. I'm guessing the DIR-825 simply has better reception with the larger antennas than the compact DAP-1522.

I don't think anybody expected for a situation to arise where the current radio bugs out when there are over 60 APs in one place drowning out a single FIRST field AP. I'm imagining FIRST does test for some radio interference, but not extreme interference coming from over 60 WiFi APs at once.

I don't think there will be any major issues with the current radio, but it's good to see FIRST is thinking about these things.

The current radios will be set to only use Wireless N over the 5GHz band, which means there shouldn't be any interference from all the old APs on 2.4GHz.

The only way I can see this becoming an issue is if the venue recently upgraded their entire wifi network and installed tons of brand-new, high-end, 2.4/5Ghz dual-band, Wireless N APs relatively close together.

The Field Access Point has both 2.4Ghz and 5Ghz antennas (3 of each), so I imagine that before going to Operation Bondé, FIRST would try switching the frequency, in case the venue has a lot of 5 Ghz APs (I obviously don't know, but I think that Operation Bondé would be avoided if at all possible). Does anyone know about a regional last year that had an issue with too many APs that would cause FIRST to start sending better radios, or does it seem like they are preempting the problem?

[waialua359]

Originally Posted by EricH

Remember 2005, Glenn? With those loading zones?

Not really. All I remember is you guys won that year at CMP against Gila Monsters and Poofs.

[Aren Siekmeier]

I just don't understand why we are using wifi at all. Sure switching to a bigger, more powerful controller is great (despite all the overhead of using an FPGA and not really taking advantage of it), but when all the wireless connections needed are known, specifically, FMS to 6 and only 6 robots, why do we need a protocol that is inherently built for arbitrary connectivity? Isn't this just asking for problems? And it gains us nothing. I think it would make a lot of sense to stick with a proprietary radio connection, and if you must use IP, establish a forward of the wired network run by the FMS over this radio link. I may be oversimplifying things, but the current situation is hardly simple.

So stick with the cRio and fancy shmancy Driver Station laptops, but stick with a wireless connection that works and is not so heavily dependent on a friendly environment.

[AdamHeard]

Originally Posted by ratdude747

but then the bridge would likely tip and no longer be within 5 degrees of horizontal.

Not necessarily, depends on when it's used and what other actions are performed.

[theprgramerdude]

Originally Posted by compwiztobe

I just don't understand why we are using wifi at all. Sure switching to a bigger, more powerful controller is great (despite all the overhead of using an FPGA and not really taking advantage of it), but when all the wireless connections needed are known, specifically, FMS to 6 and only 6 robots, why do we need a protocol that is inherently built for arbitrary connectivity? Isn't this just asking for problems? And it gains us nothing. I think it would make a lot of sense to stick with a proprietary radio connection, and if you must use IP, establish a forward of the wired network run by the FMS over this radio link. I may be oversimplifying things, but the current situation is hardly simple.

So stick with the cRio and fancy shmancy Driver Station laptops, but stick with a wireless connection that works and is not so heavily dependent on a friendly environment.

I can't tell if this is a pointless rant or if you're being serious.



What wireless protocol and frequency would you use? 2.4Ghz? 5Ghz? Something other than wi-fi? There isn't much to switch to that isn't regulated.

[Aren Siekmeier]

Originally Posted by theprgramerdude

What wireless protocol and frequency would you use? 2.4Ghz? 5Ghz? Something other than wi-fi? There isn't much to switch to that isn't regulated.

The frequency is only part of the problem. Sure there is a lot of noise at 2.4 GHz because of the prevalence of Wifi, but I feel like most of the problem is the protocol. There are numerous hobby/RC bands just outside of 2.4 GHz that have no trouble maintaining a connection, even with several hundred controller/vehicle pairs in a smallish area, and I feel it's mostly because the nodes know exactly who are they are trying to talk to and don't have to deal with traffic from a gazillion other nodes (or if they do, because it's not fully scalable like IP, it doesn't have to look through and potentially forward every packet coming in, it can just ignore most of them). FIRST itself had a band for its own competition in 2008 and prior. In terms of regulations, for short range broadcasting I'm fairly certain there are a lot of bands where you don't need to tell anyone about what you're doing (I'm talking something like less than a few hundred meters). I'm sure there are some Ham radio buffs around here somewhere who know all about this.

[Pat Fairbank]

Originally Posted by linuxboy

The Field Access Point has both 2.4Ghz and 5Ghz antennas (3 of each), so I imagine that before going to Operation Bondé, FIRST would try switching the frequency, in case the venue has a lot of 5 Ghz APs (I obviously don't know, but I think that Operation Bondé would be avoided if at all possible). Does anyone know about a regional last year that had an issue with too many APs that would cause FIRST to start sending better radios, or does it seem like they are preempting the problem?

I have a post from last year explaining the problem. The issue manifested at FLR and Florida, and matches couldn't be run until they were able to get the venues to shut down some of the APs.

[Koko Ed]

Originally Posted by Pat Fairbank

I have a post from last year explaining the problem. The issue manifested at FLR and Florida, and matches couldn't be run until they were able to get the venues to shut down some of the APs.

Ah the memories (Actually more like AHHHHHHHH! THE MEMORIES! as they drag you off to the padded room in a straightjacket).

[AlexD744]

As stated above, this issue did occur in Florida last year. Considering that the regional is on UCF's campus, a campus with over 50,000 students, it's not a surprise that there was a lot of wifi in the area. Teams would spend over 5 minutes trying to connect to the field only to be told to try again next time because it was taking too long. Once a team did connect, however, their chances of being found the next time increased exponentially, at least that is what was explained to me. I believe that is why the problem only really occurred on Thursday, because by Friday everyone had been able to connect at least once. At our next regional the FTA wanted everyone to connect to the field at least once on Thursday whether or not they were inspected, probably for this reason.

[pfreivald]

Originally Posted by Paul Copioli

This update is actually pretty important. The blue box at the end has really clarified the "stick something out the bottom of your robot to help balancing" as being perfectly legal. The Q&A responses this week made it look like it would be illegal.

Now, if you are on the bridge and you stick something down to the carpet it can be argued via the blue box clarification that your bumpers are perfectly legal as measured from the bridge that you are sitting on. Big clarification.

I think you're misreading that. I think your configuration would be illegal because if the bot were placed flat on the bridge or carpet your bumpers would not be entirely in the zone.

[Aren Siekmeier]

Originally Posted by pfreivald

I think you're misreading that. I think your configuration would be illegal because if the bot were placed flat on the bridge or carpet your bumpers would not be entirely in the zone.

But I think his point is that the robot is already flat on the bridge, the or only requiring one of them to make it legal. And of course what is "flat?"

[Greg McKaskle]

Not that I know all of the details on the radio issue, but it sounds to me like they have identified a firmware issue with the model that they were using. It sounds like working with the radio providers, they have identified a model without the issue. Rather than replace all the radios for a rare complication, they have a plan to deal with a known issue if and when it occurs.

Again, I'm not an expert, but I assume that proprietary radio bands, especially ones with lots of bandwidth are susceptible to implementations issues in firmware and logistical surprises too.

There is nothing about the new system that precludes a special band or radio, but using off-the-shelf products, teams can pretty easily have their own N speed setups in their school or shop.

I'm also curious what you mean by FPGA overhead?

Greg McKaskle