How slow is too slow?

I’m going to pick on Tristan Lall from this thread: http://www.chiefdelphi.com/forums/showpost.php?p=293502&postcount=3

I’m going to go out on a limb and say that 5 ft/sec is nowhere near blindingly slow. Sometimes, it isn’t even slow at all.

In 2002 (Zone Zeal), I was on 340. Our top speed was 6 ft/sec. Over 50% of the time we were the first person to the goals. Only a few times were all the goals taken before we got there (and then most of the time we got them back).

In 2004 (last year, First Frenzy), I was on 1405. Our top speed was 8 ft/sec, and when we didn’t snag our hook on the goal, we were always the first team hanging.

Dr. Joe’s old advice to rookie teams was to make your robot go about as fast as you walk. At least for me that is 5-6 ft/sec. His newer advice for top teams is about 10 ft/sec.

I don’t think that 5ft/sec is anywhere near blindingly slow. Yes it is slow if you want to traverse the 48’ field 10 times per match. But for most teams that only need to stay on 1 side, or only move across a few times, 5 ft/sec is quite all right.

So, given the games in recent history, what’s the slowest top speed you would consider for next year?

I wouldn’t make any decision about that until I saw the game, as we all know if the field is big and open, speed makes a difference…but if there are a lot of objects and manuevering to do i’d lower the top speed then…

Again it all comes down to what the game design crew comes up with…

Lets put it this way, every year there has been at least one very good strategy that did not need a fast robot.

2001, perhaps the year that had the biggest point advantage for being fast, required a lot of finesse balancing the bridge and most teams did not gain much by being ultra fast.

2002, It didn’t matter if you got the goals first, only if you had them at the end.

2003, vision was the limiting factor in clearing out all the bins from your opponents area, not speed.

2004, you could very easily herd balls and cap and then hang, without ever traversing the field. If you held the balls internally, you didn’t really have to be fast at all.

In every single one of the gearbox threads, someone asks “how fast is it?” That means that someone has ben thinking about desired speed. Why not think of it from a strategic viewpoint (note the forum) rather then purely a gearbox viewpoint?

That comment struck me too. Last year our bot went around 4’/s and that was good for the space we had to work with. I think I would have made it 5-6’/s in hindsight. But blinding slow to me would be 2-3 or less. Also our chassis was somewhat fragile to high speed impact - as Tytus rammed a goal during testing at high speed (around 8’/s) and put a 2" dent in the front of it. So we had to slow her down or we would have to put it on the rack after every match. I think if you want to go faster than 6-7/fps you better build it tough, cuz thats quite an impact you’re going to make on the field and other bots.

Although this year we do plan to design for 9’/s :wink:

Our team has always made relatively slow robots.

Our winning robot from last year never went faster than 4ft/s. Yet it still achieved its goals consistently. Does it really matter whether you go 5ft/s or 15ft/s so long as you do what you do really well?

Ideally, you want something that can go 10+ f/s in high, and something like 4-5 f/s in low. The only example I can think of at the moment is my first year, 2002. My team, 810, designed a wheel/tread system (all credit goes to M. Krass for that), which was, if I remember correctly, 11 f/s in high, and 4 fs/ in low. In high, we could consistently be the first team to the goals, and in low, we could push almost any other team we were up against. Even more important, being rookies, we hadn’t quite designed a hook that fit (weight requirements), so the extra speed was crucial in getting around the goals to push. I’d go as far as to say we could’ve even beaten 71 with that robot, but unfortunately, I never got the chance to go up against them, and that robot has since been destroyed (I wasn’t exactly thrilled when I saw that).

However, 2004 my team had another problem. We had a single speed transmission, but the sprocket ratio on the wheels was off, so we had speed, but no power. We then were unable to turn. I’m not an ME, so I’m not sure if this is related to transmissions at all, but remember that you will need a good amount of power to turn any robot, because the carpet is a very sticky surface, so don’t make your transmissions based on speed alone.

For 2002, note that the winners of a lot of events grabbed onto the goals first and didn’t let go…speed was a factor cough71cough

Also 2003, unless your machine had a combination of speed and strength if you watch a lot of the matches as the year went on, a majority of bins stayed on the side of the field they originally fell towards. If those bins fell against you, you better have been fast enough to get as many of them over to the other side as possible

I definately think speed was a major factor in those 2 games…

One qualification to that statement: if a single ratio is all that you have, and you desire controllability and pushing ability at low speed, perhaps 5 fps is an acceptable top speed–but for a dual-ratio transmission, 5 fps (in high) is indeed rather slow, and doesn’t take advantage of the potential to position yourself on the field at will. Even for a single-ratio design, 6-9 fps is more common, and significantly speedier (though of course sacrificing the pushing abilities).

In 2004, a high top speed wasn’t used frequently, and the obstruction in the centre did limit the usefulness of much faster designs. (Not to say that it was totally useless, but 12+ fps wasn’t hardly a priority.)

2003, however, is a better example of the benefits of speed: while you didn’t necessarily have to arrive first at any particular position to win, it was often necessary to cross the field quickly to lend aid to your partner, or defend a stack, or gain momentum for a charge up the ramp. A low ratio is great for pushing matches, but with such a large, open field, a robot that could position itself at will was at a clear advantage.

In 2002, speed was a factor, but so was pushing. If ever there was a game for Blizzard 5’s three-motor, two-speed transmission, geared to 12 fps in high, that was it. It’s a toss-up.

So, for a dual-ratio design (like the one from the linked post), when you have a low gear of 3 fps, you won’t derive any of the benefits of high speed by using a 5 fps top gear, because many robots will still be doing circles around you. In that sense, it is indeed much too slow, and likely a sub-optimal use of your dual-speed capabilities. For a single-speed transmission, a clear decision on speed vs. torque has to be made. In that regard, if faced with a 2004- or 2002-style game, you might want to use something slow and strong; alternatively, with a 2003-, 2001- or 2000-style game, something a little faster might be in order.

In essence, I would say that for a robot designed for the 2004 game, 7 fps would have to suffice for a middle-of-the-road single-speed gearbox, while 4 fps and (a rather quick, but still controllable) 12 fps would be best for a dual. For the 2003 game, I would tend to go even higher, with 9 fps for the single, and 5 and 13 fps for the dual.

One last item. I’m used to using multi-motor designs which can still push formidably at high speed. (Recall Blizzard 4 in 2003, which had a single-speed, 14 fps design, and could still handily out-push and outmanoeuvre [Canadian spelling :cool: ] most of the opposition.) I realize that many (indeed most) teams don’t go that route, and therefore are more used to strategizing with slower robots.

The Martians’ zippy little chihuahua of a robot was extremely – and annoyingly – fast. Its defensive ability helped take its alliance all the way to the top. Speed coupled with agility was definitely useful in the 2004 game.

Well, I didn’t want to bring these up, because they’re not representative of the typical strategies and designs employed by most teams, but I definitely see the advantages of this sort of thing. In addition to the Martian robot, 188’s 2004 robot was also extremely fast and powerful. The Martians’ gearbox was number one on my list of things that I wish I’d had a look at, because it was the only one out there that appeared to have comparable speed and power. (Was it a 3-motor, 2-speed design?)

My design (on Woburn’s Blizzard 5) incorporated the six biggest motors in the kit, and a two-speed transmission good for 16 fps in high (really!) and 4 fps in low. The 16 was a little too fast (at the edge of controllability), but in the 2003 game, it would have been amazing as-is. We’re always ready for the last war, after all.

have any of you played with or against team 60 in the last few years? and tell me that speed isnt a factor or even important… or intimidating? haha

particularly 2002(OMG) 2003 and 2004

tho we all do remember 2002 71 BEAST (Still had a very important speed factor involved on those back wheels) and the match they finally had at national semi finals 2002

Maybe speed alone isnt super important always… but if you come up with something dominating once u get to where u wanna go then u want all the speed u can to get there to do it first

Osc’s 2 Cents

That’s not a byproduct of speed. That’s a byproduct of fantastic drivers. 254 is the same way.

The speed helps, but without a good driver, speed means nothing, because you can’t control it.

2001…then entire game had to do with the speed with which you did something. If you got from point a to point b faster you had more time to work with balancing or whatever.

2002…dj already made my point

2003, you may not have won, but if you knocked the bins over quickly (quickly…nice pun) you shifted the game to one side dramatically. The rest of the time the other team had to play catch up…

2004…defensively, speed was a huge factor (who did 71 pick first in atl? oh oh 494 that’s right)

I think some people underestimate what being the first to the punch can do for you. Personally I am not one that likes having to play catch up…and i don’t believe this stuff from certain ppl that the students can’t handle the speed (with enough practice they should be fine)…btw did anyone see 25 in 2003 fly around, and 60/254 in 2004…? “speed is king” imho…

If you were referring to my point, I wasn’t saying that some people just cant drive fast. I’d hazard a guess that maybe 5% of all FIRST teams get a significant amount of practice time in before ship. The rest either havent finished their robots, or barely finished.

Once regionals roll around, the teams that practiced are going to do well and get more practice. The teams that didn’t finish the bot are going to use the practice rounds to finish it, and maybe even some/most of the qualifying rounds. Those that finished the bot have time in the practice and qualifying rounds to learn to drive the bot.

So basically, for the average FIRST team, it takes AT LEAST one regional to become familiar with the robot, some maybe even two. Most teams attend one regional (and perhaps nationals) some attend two (nats, possibly) and hardly any attend 3 regionals (nats as a possibility again).

So unless you’ve had that practice time before build ends, or you have a second bot, or an old one that handles similarly, most teams dont have much practice at all coming into their first, or even second event, which makes having a very fast robot tough on the drivers.

No I wasn’t…George1083 and I have had debates on this topic many times into the wee hours of the night/next morning…that is who I was referring to, and did not want to say his name, but I will if I need to clarify. :cool:

Despite what time you do or don’t have I believe that working until the last minute is really quite pointless unless you have practiced what you are going to do on the field…otherwise ppl are just guessing what they think they can/can’t do (but that is another debate altogether). Make time to practice for the drivers, otherwise all the engineering and work put into a robot is not wisely used if the product does not preform on the field.

Our new driver/drive team got experience and even though we didn’t have such a stellar machine he/they drove well enough for us to be a part of the winning alliance in the Peachtree Regional. I don’t mean to toot my team’s proverbial horn, but that is the team I know best right now.

Ok im done.

I do think anything over 8 to 10 fps is too fast.

I formed this opinion after years of watching drivers do their thing. When machines went faster than about 10 fps, the drivers got very sloppy; they just weren’t able to accomplish their tasks consistantly. Many times, they didn’t use their top speed at all!

A slower top speed helps to keep your motors cool and gives you a bit more pushing force in addition to helping your drivers out.

Being from the same team as Tristan, I may be a bit bias. I did see a lot of games in that I was an announcer for 4 Regionals, Championships and an observer at 1 regional. Speed is good and speed is bad. If you can’t control the robot at any speed then it is not good.

What I do agree with is that having only a 2 ft/sec difference isn’t really worth the time to build or the weight that is generated. If you added 4 (double) to 6 ft/sec then the advantages could be seem and measured. IMHO our robot was a bit too fast for our driver (sorry Jon) but if he had more time to practice maybe it wouldn’t have been. I did see a few matches that the dash from 1 end to the other was the reason that they won the match.

Right speed with right driver = success

Designing a FIRST robot is all about trade-offs. Picture this pre-build season conversation, with a “Robot Dealer”…

**RD: **Hey you, strategy guy.
**ME: **Who me?
**RD: **Yeah you. Do you want a robot that can go 10+ ft/sec?
**ME: **Well yeah. Who doesn’t? The extra speed can always come in handy. But wait a second, this is a FIRST robot. There must be a catch…
RD: Okay, maybe there’s a catch. The one robot I had in mind has a one speed transmission.
ME: Hmm, no deal. I’d have to give up a lot of pushing power to be geared to go at that speed. Most FIRST games, require more torque than that gearbox would probably provide.
**RD: **HM, so you’re one of those smart strategy guys. Okay, I’ll give you a two speed gearbox. 10 ft/sec in high, 2 ft/sec in low…
**ME: **Wow, that’ll let me be really fast and give me the ability to be a strong pusher I like that. But hold on. That sounds like a mighty big reduction. How much does that thing way?
**RD: **Fine. It weighs a lot. But I’ve got another one that weighs less and plays in the 4-10 ft/sec range.
**ME: **That sounds a lot better, but my drivers are pretty green. I don’t know if they can handle that much speed.

This conversation could go on for a while…

The point is, speed is always an asset in FIRST game. You just can’t give up too much to achieve it. I’ve seen many teams slave away in an effort to try and build the fastest robot. A lot of the time it’s not worth it.

Remember, on a 48 ft field, there’s not many times when you’ll need to go much faster than 8 ft/sec. On the other hand, the ability move at 12 ft/sec, can be a huge asset. The decision you have to make, is whether or not it’s worth the effort and weight to build in a function that may not be used very readily. This depends on your team’s capabilities.

When it comes to the issue of control, I have seen some very fast robots which moved across the field with lots of precision. Team 25 in 2003 really sticks out in my mind. I’ve also seen some really fast robots slam into a lot of obstacles, and look like they’ve been driven by a drunk. More often than not, anything more than 8 ft/sec is uncontrollable. But with a nice PID algorithm and skilled, well practiced drivers, it is possible to control.

Would I build a robot that could only go 5 ft/sec? Probably not. I think it’s very doable to build a simple two-speed tranny that puts you in the 4-10 range. This is a great range for most games. Remember, if your driver isn’t comfortable with top speed, s/he can always pull back on the throttle. All that being said, there are many possible game designs that would cause me to go significantly faster or slower.

if there is sufficient practice among drivers, (usually by the time nationals rolls around…hehehe) a high top speed, if done within the limits of the team in question, done comfortably, the benefits are immense and immediately measurable. There was a direct happy finish for our team in Atlanta more than once due to our high gear.
Check out the final seconds of Match 53 at http://www.soap108.com/2004/movies/arc/index.cfm

where we streak across the field and prevent 1272 from capping.while that could have been done with 12 seconds left with a lower top speed, the fact is we spent another good 5 seconds manuvering with our grabbers, forcing the 2X ball to drop. Without that extra time, it is very likely that we wouldnt have succeeded. Thanks FEDS :wink:

I agree, 16 fps (give or take) was too much for that game, especially with the configuration of that robot. Not tipping every match required great care on the part of the drive team, and hard turning manoeuvres and dime-stops with the arms up were perilous (we never tipped ourselves, by the way). It was quite hard to control. Mind you, the freakishly high speed was useful once in a while.

Archimedes Match 53.

That being said, what we needed last year was a smaller gap between gear speeds; 16 is too high and 4 is too low. The truth is that everything is changed by a shifting tranny; that is, George1083 is absolutely right (with regard to his last comment), but only in the case of single-speed gearboxes. The truth is, limiting yourself to 10 fps (well, 10 isn’t that bad, but 8 for example) in your high gear ratio is a severe limitation of your ability to take control of the field and in its nature limits possible strategies.

In any case, as the bottom line has always been, you can’t tack on a speed to shoot for. We’ve tried to many a time on CD. As was demonstrated in this thread already, 639 and 60 were both winning robots with different philosophies. Setting a ‘magic number’ is just plain silly, especially for teams with the means to build really advanced stuff.