ZO Analysis

dndn Posts: 18 Baller
edited September 2010 in Technique & Theory
<p>
<span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt">I apologize in advance for the length of this post - I hope you'll find it worthwhile. </span>
</p>
<p>
<span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt"></span><span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt">When a skier rounds the buoy and loads the line, the boat slows in response.  ZO senses the slowing within 0.05 second and starts to increase engine throttle.  However, the fastest V8 engine throttle response I could find a spec on is 0.12 second (BMW) and I bet most throttle response times are a little longer.  By the time the increased throttle starts to reverse the slowing, I've observed the boat slowing by 0.4 to 0.7mph, depending on the skier load.  Then ZO will increase the speed by roughly the same amount for a similar duration to bring the average speed back to 34.2mph, again depending on the skier load and settings.  </span>
</p>
<p>
<span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt"></span><span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt">From an engineer's perspective, this is a classic problem in which ZO forms a feedback control loop which responds based upon the present speed error, the accumulation of past errors, and the prediction of future errors based on the current rate of change, such that the average speed is 34.2 and the skier doesn't get too badly manhandled.  The ABC123 settings are the simplified inputs the skier can use to shape the response to some extent. </span>
</p>
<p>
<span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt"></span><span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt">Here's my analysis of what this means to the skier, based on what info I can glean from the lousy ZO literature and my own skiing experience:</span>
</p>
<p>
<span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt">  </span><span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt">1.  The throttle response time constant is inescapable!  No matter how many times ZO samples the speed per second, the throttle response will limit how fast the boat can respond.  Several time constants will elapse before the engine starts to reverse the slowing.  A heavy skier will slow the boat more than a light skier.  A hard pull will slow the boat more than a light pull.  A heavy boat will slow less than a light boat (momentum).  A high rpm & lower prop pitch will slow less than lower rpm & higher prop pitch (I notice the 2010 Nautique has gone to higher rpm & lower prop pitch, and is heavier as well!)  Throttle response varies from model to model and even from boat to boat of the same model because throttle response is not that tightly controlled, sadly.   The differences are real and noticeable!  ABC123 settings have no effect on the throttle response time. </span>
</p>
<p>
<span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt"></span><span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt">2.  For every fraction of a mph and fraction of a second the boat spends going less than 34.2mph, ZO will have to make up for it later by going faster.  The more you load the boat, for whatever reason, the more ZO will extract in payment later.  This is why heavier skiers complain more about ZO than lighter skiers - they experience greater extremes in boat speed.  ZO makes you reap exactly what you sow - it isn't concerned about fairness.</span>
</p>
<p>
<span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt"></span><span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt">3. ABC controls how much time the boat takes to get back to 34.2mph.  The C setting gets you back to 34.2 the fastest, so it feels 'strong'.  B allows more time, and A allows the most.  This makes C feel a little more responsive to me, but to some it feels a little too stiff or unforgiving.   123 controls how hard the boat responds during the time the boat is going faster than 34.2.  3 will respond stronger (not faster in terms of absolute mph - what in engineering is termed 'slew rate', not amplitude) but with a shorter duration, 2 is not as hard but the duration is longer, 1 is softest and longest. </span><span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt"><span> </span></span>
</p>
<p>
<span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt"><span></span>(I think my points 1. and 2. are right on, but I'll admit my assessment of ABC123 could be off a little because I don't have a way to precisely measure the ZO response.)</span>
</p>
<p>
<span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt"></span><span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt">It's interesting to compare ZO and PP.  Oversimplifying:  With ZO, if you slow the boat down to 33.5mph for 0.6 second, it will speed the boat up later to 34.9mph for 0.6 second, so you have to deal with both slow and fast extremes and relatively rapid transitions between those extremes, but the boat averages 34.2mph.  PP comes into the course at say 35mph.  You slow the boat down to 34mph for a second or more, and PP gets it back to 35 soon enough that you still average 34.2mph, but you only have to deal with what seems like the slowing, and the transistion between the two different extremes is much slower than with ZO, so the pull seems softer.  Also, ZO is a closed loop system:  it will always respond with as much force as necessary to insure an average speed of 34.2.  PP is predictive system:  if you pull harder than the settings have been programmed to expect, you'll just slow the boat down - this is another reason PP can seem much less severe. </span>
</p>
<p>
<span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt"></span><span style="font-family: 'Trebuchet MS','sans-serif'; color: #062971; font-size: 8.5pt">As an engineer with a lot of experience designing both types of systems, I think there is a lot of room for improvement in ZO, and I hope the ZO engineers feel the same way & strive to make those improvements.  </span>
</p>
«13

Comments

  • 94009400 Posts: 597 Crazy Baller
    Good stuff...thanks for taking the time and effort. The more we understand, the more we can make adjustments and recommendations.
  • scokescoke Posts: 644 Solid Baller
    <p>
     
    </p>
    <p>
    Very interesting and well thought out.
    </p>
    <p>
     I've looked at the Zero Off process a little bit and read some things but maybe take another look at your third point (3.) again as you might have the letters classified improperly especially regarding "strong" and "time".
    </p>
    <p>
     
    </p>
    <p>
    scoke.
    </p>
    <p>
     
    </p>
  • eleeskieleeski Posts: 3,819 Infinite Pandas
    <p>
    Great analysis.
    </p>
    <p>
    If the time is to be 16.95 ZO is great. My inconsistent pulls make PP Classic run a tad slow ("if you pull harder than the settings have been programmed to expect, you'll just slow the boat down"). Stargazer hammers me at 5 ball to make up - offering the worst feel. The true improvement with ZO might  be to allow some deviation in times by capping the throttle response so it doesn't hammer too much on an unexpected overpull.
    </p>
    <p>
    Each .01 is about 6" over the length of the course. Tournament times with PP Classic seemed to average 16.99. 16.95 happened occasionally and anything faster was very rare. That was proper - a driver should not be adding weight to compensate for that one overpull. But the result was a soft time. And so often I sneak around 6 ball with inches to spare - I needed that soft time! ZO does not allow that extra time (distance). So maybe my buoy count did suffer with ZO. But at least when I overpull, the rest of the course stays proper. Survive the mistake and you are back to everything as it should be - and the time will be perfect to keep the Powers That Be happy.
    </p>
    <p>
    PP Classic feels great. But when I fall it always gives a hot time and when I run the pass it always is a bit slow. I try to mix up my training so the target times are hot (before tournaments) or soft (when working at my limits) to offset this time factor.
    </p>
    <p>
    ZO I just ski. Maybe when I get more comfortable with it I'll play with different number/letter settings. But for now it is just about making the ski work with a consistent pull - and the same pull I will get in a tournament. I love ZO for that. And I have enough time behind it that I ski as well behind ZO as PP Classic.
    </p>
    <p>
    Plus ZO is fantastic for tricks.
    </p>
    <p>
    Eric
    </p>
  • Jody_SealJody_Seal Posts: 2,633 Mega Baller
    edited September 2010
    <p>
    For the numerous (20) tournaments that I drove this last year including three in Australia I have found that ZO errors on the fast (but other drivers have not noticed this) I have found mostly 16.94, 16.07 far more than actual. Also 16.96 and 16.09 hardly ever occure as mostly the .01 on the fast or actual. I have asked Joe Grogan About that and he Said they have it set up to error on the fast with their target being actual.
    </p>
    <p>
    One thing that  dn did not include in his description and analysis is the length of time and effect of how the throttle responds comming off the acceleration phase through The abc/123 process, This to will have impact on the skier as well as response ramping up.  
    </p>
    <p>
    Again still would like option D a RPM baised control.
    </p>
    <p>
    As far as Star Gazer Eric you more than likely do not have it set up correctly or refuse to take the time to do so. We have three Syetems running on our lake that get as good times as ZO through out the course with out gassing at five ball, Yes their was some initial fiddle factor and a learning curve but now we just set the crew weight speed and let it know if we have a light skier or normal skier 1.77, 7.13 and 16.95 with every thing in between being spot on. Also one boat is a 07 206 the other is a 04 196 and a 02 1:1 MasterCraft 197.
    </p>
    <p>
    Also Dn thanks for the post, again more info and knowledge is what is needed in this sport.
    </p>
    Hobby Boats can be expensive when the hobbyist is limited on their own skill and expertise.


  • Than_BoganThan_Bogan Posts: 6,359 Mega Baller
    <p>
    dn -- Thanks for this!  As a software engineer with a general interest (but not true expertise) in control systems, I am also facinated by ZO.  Your thoughts are mostly similar to mine, but I think you've exposed some possibilities that I hadn't thought of -- in particular that ZO might actually allow you pull it off further than PP in some cases, and that the "making up for it" phase is what many of us feel as unduly harsh driving.  This may mean that ZO actually does a <em>worse</em> job of maintaining speed, if I define that as maximum instantaneous deviation from desired.
    </p>
    <p>
    Have you considered videotaping the digital speedo and then trying to plot it out?  This is a lot of manual work and I've been too lazy to do it.  But it seems like it could be really interesting.  It might finally tell us what ABC123 <em>actually</em> do.
    </p>
    <p>
    Is the 50 ms response time of ZO out of their literature?  I assume this is mainly from the satellite.
    </p>
    <p>
    120 ms throttle response seems slow.  Even a cheap 50MHz ARM processor can do 6 million cycles in that much time.  Is this delay more related to mechanical systems?  I.e. how long it takes to actually get more fuel into the cylinders?
    </p>
    <p>
    It's worth nothing that 170ms is "only" about 5% of the time between balls, although more than 10% of the time from buoy to centerline -- and that fraction increases as the line gets shorter and the pull phase is faster.  So this delay is not quite trivial.
    </p>
    <p>
    You assume (or I think you do) that ZO is purely reactive.  That is probably true, but I think it's a mistake and one of the key weaknesses of the current algorithm.  The events of the course are periodic, so predicting the next moment at which the skier will load the rope may be worth attempting.  This won't be precise, so we must be careful not to just assume it and do a "pre-emptive strike."  But if we get a 170ms head-start, we can begin with a much softer correction.  And of course we maintain the reactive element so that we back off if we don't then detect the expected slow-down.  I am confident that the best human drivers anticipated the pull, so I see no reason ZO shouldn't do so as well.
    </p>
    <p>
    The same is true on the other end -- you can make a reasonable prediction of when the skier will release and the boat will begin to accelerate.  Thus it is possible to also anticipate the reduction of throttle.
    </p>
    <p>
    Secondly, I believe the ZO algorithim should (and possibly already does) use the derivatives of the speed as inputs into the control system.  If the boat is being decelerated rapidly, that tells you something different about what you need to do and --  importantly -- how soon you need to do it.  (The second derivative may provide additional predictive capability, although I'd have to think about that more.)  As I have been yammering on about on this very forum, I am optimistic that this can eventually eliminate the need for ABC123 entirely.  Obviously these are artificial parameters and it would be better if the system just did the right thing and skiers didn't have to think about this stuff.
    </p>
    <p>
    Bottom line, I firmly agree with you that the ZO algorithm can be better, and I also believe it will be better.  If for some reason they are reading this, I am willing to volunteer some time and my MIT/CS masters degree to help them!
    </p>
    Nathaniel Bogan -- GUT Padawan
  • HO 410HO 410 Posts: 351 Baller
    Here is a link to the RevX fact sheet. https://backup.filesanywhere.com/fs/v.aspx?v=8c6d6b865c66a7b7b4a4
    Nikon D80, 50mm f 1.8, Tokina 12-24mm... Sorry, wrong forum. Josh T.
  • ktm300ktm300 Posts: 399 Baller
    Zero competition often means zero improvement.
  • LazLaz Posts: 320 Baller
    <p>
    Does ZO know where you are in the course?  PPSG expects at least one magnet so it knows exactly when the boat goes through the gates.  My understanding is that this allows it to gas the engine in anticipation of the pull which, I would imagine, would lead to a more consistent speed.
    </p>
    <p>
    On the other hand, ZO can't know exactly where you are because GPS absolute positioning is not nearly as accurate as relative position.  So it must just be responding to the boat slowing down.
    </p>
    <p>
     Is this correct?
    </p>
  • GMCGMC Posts: 99 Baller
    For what it's worth I have a few comments to add. There are some smart guys on here and I'm not sure I'm one of them! Jody seems to have the most insight on how the ZO system is truly designed so he can probably answer the question I will ultimately ask.

    dn - I think you're spot on with points 1 & 2 and I can't explain #3. I can make sense of ABC123 to myself in my head but I try not to confuse anyone else with my explanations. I may be about to as I'm thinking out loud here... Maybe A, B, and C control setpoint overshoot differently as well? As you indicated, to get out the gates at an average of 34.2 and the skier pulls the boat down to X mph less than that for Y seconds you have to make up for that with excess speed for corresponding time. The integral of speed error over the course has to equal close to zero to get 16.95. So maybe A overshoots a lot for a short amount of time and C overshoots a little for a long time? I'm picturing a sine wave in my head where set speed is the X and speed error is the Y. The area above has to equal the area below at the exit gates. Then there is the question of ramp rates and delays but I don't want to go there. That is a large part of ABC123 and I don't have enough info to say.

    All the talk is about throttle response, which is governed by the physics of the throttle blade moving, allowing more air into the intake, and then the engine injesting that additional air. This is relatively slow. The blade moves instantly with ETC, but the size of the throttle body, the manifold volume, pressure difference across the throttle body, and lots of other physical attributes all play a role. What about spark and fuel? Does ZO add power to the system just by opening the throttle or does it move those too? If you're not at the knock limit, you can add power really quickly by adding spark (the fastest ECM loop rates I've seen are in the order of 5 msec).

    So Jody, is ZO a fully integrated power controller that manipulates spark and fuel in addition to air or does it just move the blade to let in more air like PP did? If that's the case, a video of the tach through the course might not be the whole story - and is probably filtered heavily anyway.
  • Than_BoganThan_Bogan Posts: 6,359 Mega Baller
    <p>
    <em>On the other hand, ZO can't know exactly where you are because GPS absolute positioning is not nearly as accurate as relative position.</em>
    </p>
    <p>
    I agree with this premise. 
    </p>
    <p>
    <em>So it must just be responding to the boat slowing down. </em>
    </p>
    <p>
    I do not agree that this is implied.  It may be true at the gate, but thereafter it has its own history of speed that it can use to make a very good prediction of the next cycle.
    </p>
    <p>
    Consider this:  If you had a blind-fold on as a human driver -- and your only job was the throttle -- could you anticipate when to add or decrease throttle?  Yes, because of the repeating pattern.
    </p>
    Nathaniel Bogan -- GUT Padawan
  • scokescoke Posts: 644 Solid Baller
    <p>
     
    </p>
    <p>
    this is going to end badly.
    </p>
    <p>
     
    </p>
  • Than_BoganThan_Bogan Posts: 6,359 Mega Baller
    <p>
    Hm, perhaps I should mention:  DO NOT DRIVE A BOAT BLIND-FOLDED! <img src="/vanillaforum/js/tinymce/jscripts/tiny_mce/plugins/emotions/images/smiley-smile.gif" border="0" alt="Smile" title="Smile" />
    </p>
    <p>
    (It's a thought experiment.)
    </p>
    Nathaniel Bogan -- GUT Padawan
  • Jody_SealJody_Seal Posts: 2,633 Mega Baller
    <p>
    GMC to my knowledge the ZO unit only opperates the butterfly valve on the valve body their is again to what I know no closed loop that takes in account injector pulse, duration, nor is their any direct communication controling timing, the motor just responds to what the ZO tells the throttle body to do, as did DBW Perfect Pass. I think it would get pretty confusing for the programer that designs the operational control of the motor to involve operation of the speed control also in that fashion. Keep in mind that is for the MIT guy's as I am purely a gear head in all this. Also the slalom mode is not like some suspect dumb timing and direct speed control. This  program has different controling gains through out the course it knows where it is to some degree, It comes in nearly a half mile over set speed and has some ramp up gains that are different from start of the course to the last timing point. This as was splained to me at a test last Dec with Joe and Nautiques Engineers.
    </p>
    <p>
     
    </p>
    Hobby Boats can be expensive when the hobbyist is limited on their own skill and expertise.


  • dndn Posts: 18 Baller
    edited September 2010
    <p class="MsoNormal">
    <font face="Calibri" size="3" color="#000000">Thanks for the comments – I’ll make a few specific replies:</font>
    </p>
    <p class="MsoNormal">
    <font face="Calibri" size="3" color="#000000">I agree that the ZO “making up for it” phase is what makes it harder to ski behind.<span>  </span>ZO probably has a lower standard deviation from 34.2 than PP, but to achieve that it accelerates and decelerates in a disruptive fashion.  PP has bigger excursions around 34.2 and they last longer, but that's easier for a skier to deal with, because a skier response time is about 240ms!  We don't react well to things that change faster than that.</font>
    </p>
    <p class="MsoNormal">
    <font size="3"><font color="#000000"><font face="Calibri">I don’t know what ZO update rate is specifically, but most GPS units update 20 times/second, hence I’m assuming 50ms.<span>  </span>The 120ms throttle response was for a BMW S62 5L engine used in M5’s and Z8’s.<span>  </span>That is actually the response time from the accelerator pedal to ECM to DC servo motor that controls the fuel injection.<span>  </span>As soon as that fuel enters the cylinder the engine will respond in under 4ms at 3600rpm, and then it takes some time to rev up the rpm. <span>  </span>If you’ve driven a car like that, you’ll swear there is no lag, but that’s because the human response is considerably slower.<span> </span></font></font></font>
    </p>
    <p class="MsoNormal">
    <font size="3"><font color="#000000"><font face="Calibri"><span></span></font></font></font><font size="3"><font color="#000000"><font face="Calibri"><span></span></font></font></font><font face="Calibri" size="3" color="#000000">I too have been too lazy to video the gauges and try and correlate what the boat does with what the skier experiences.<span>  </span>I bet ZO already has this reporting function built-in to communicate via USB, but I’d be surprised if ZO would ever make it available to us.</font>
    </p>
    <p class="MsoNormal">
    <font face="Calibri" size="3" color="#000000">I do agree ZO is just reacting.<span>  I don’t see any indication ZO applies the knowledge of where it is in the course or what the skier history is to anticipate any kind of pull – too bad it doesn’t.  </span>For example, if the skier misses a pull, the engine is not throttled in any kind of anticipatory fashion.<span>  </span>Adding this predictive type of response is something I think ZO should do to reduce the slowdown and speedup that is so disruptive to most skiers.  I don't think it would be hard to add - I hope ZO agrees.  One other option would be to keep the reactive approach but make the throttle response MUCH faster so that in reaction mode, the skier can't reduce the boat speed so much as they can now.  Then less 'payback' would be required after the pull.  I'll bet ECM could be designed to respond faster, but since existing ECM is designed to work well with a human in the automotive feedback loop, ECM should be re-designed for the way ZO is trying to use it.  Given all the emissions features etc. ECM is required to control these days, that may be too much to ask, however!  If that can't be done, the anticipatory approach will likely work better. </font>
    </p>
    <p class="MsoNormal">
    <font size="3"><font color="#000000"><font face="Calibri">Thanks for the fact sheet HO410.<span>  </span>I think my description of ABC fits reasonably well with ZO’s description.<span>  </span>My 123 description seems to vary from theirs, but my experience seems at odds with what ZO claims.<span>  </span>Since in all cases, be it a heavy or a light skier, skier load slows the boat enough that ZO has to respond ASAP to prevent the error from getting too large.<span>  </span>To say it takes a light load or heavy load to initiate an rpm response makes no sense in my observations.<span>  </span>I chalk up the ZO statement on 123 to tech-to-journalistic translation error – something I see a lot of in almost all technical publications.<span>  </span></font></font></font>
    </p>
    <p class="MsoNormal">
    <font face="Calibri" size="3" color="#000000">David Nelson</font>
    </p>
  • GMCGMC Posts: 99 Baller
    Jody - good enough for me. Those algos do exist but they are overkill for slalom speed control so I didn't think ZO actually used something like that. And don't underestimate your gear head knowledge!!

    What I do know is that I like C1 at 36 and B1 at 34. Back to figuring out which ski I want to ride.
  • my position on PP vs. ZO is i prefer ZO.  that being said i don't know why other than i ski better behind ZO. boats.  you all seem much more knowledgable than me an di have a question.  at what point does the boat make up the time if the skier pulling is the reason the apeed control system would ever be going less than 34.2 MPH?  the only itme i hear a ZO boat surging with higher RPM is when the skier is behind the boat.  if it knows it is going slow would it not stop the RPM surge until it has made up the time.  this would potentially have the boat slowing when the skier is turning.  not an ideal time, leads to slack in the line.  what does everyone think of the idea of the boat speeding up as the skier is not loading the line?
  • jordanskisjordanskis Posts: 12
    edited September 2010
    <p>
     
    </p>
    <p>
    Okay...first let me declare: I am not a software engineer, and I am only a decent(for an old guy that doesn't train) course skier...that beig said, it seems to me that what is most important is where/when the acelleration hits.
    </p>
    <p>
     In the middle of a turn bad, in the pre-turn would be very bad. So we want it to hit somewhere after the turn has been completed to directly behind the boat.
    </p>
    <p>
    Sooooo, wouldn't an input for rope position allow for a system that would consistently work for a given skier??? Okay, so now somebody needs a position sensor in the pylon....go get 'em guys!
    </p>
  • dndn Posts: 18 Baller
    <p>
    Good question skiergeeg
    </p>
    <p>
    I observe the rpm surge as soon as the skier starts to load, usually just after they get both hands back on the handle.  The rpm surge gets the boat back to 34.2mph prior to the first wake, and the higher speed payback starts immediately or soon after.  On something like an A1 setting the correction continues into the preturn quite a ways.  On C3 it ends sooner.  It never lasts into the turn.
    </p>
    <p>
    On C and especially C3, the rapid/strong correction causes me to be way early & wide - then I usually waste it by standing too tall in the turn.  I find I'm better to be just in time than early, so I do better with B2, but I keep trying other settings because I'm never satisfied.
    </p>
    <p>
    To ski ZO well, you need to try and stay invisible, meaning:  load progressively and get an early edge change.  By loading only lightly early on you won't be slowing the boat a lot.  Then ZO will only have to engage lightly.  As your load increases toward the wake, ZO is already engaged and the correction will also be progressive, hence nearly proportional!  That's not disruptive.  If I could do that I'd be skiing like Andy Mapple, for whom ZO was optimized in the first place!  Only in my dreams.  It sounds like you're skiing closer to that ideal than most of us! 
    </p>
    <p>
     
    </p>
  • HortonHorton Posts: 27,225 Administrator
    <p class="MsoNormal">
    <font face="Times New Roman" size="3" color="#000000">I had a long conversation about ZO with Will Bush at the Diablo ProAm. I wish I had a tape of the conversation as he explained a lot of this and most of it went over my head.</font>
    </p>
    <font face="Times New Roman" size="3" color="#000000"> </font>
    <p class="MsoNormal">
    <font face="Times New Roman" size="3" color="#000000">As for the above posts I have a few questions:</font>
    </p>
    <p class="MsoNormal">
    <font face="Times New Roman" size="3" color="#000000">Would we want the boat to react faster? </font>
    </p>
    <p class="MsoNormal">
    <font face="Times New Roman" size="3" color="#000000">If the boat kept a .1 MPH or .25 MPH tolerance would that be better?</font>
    </p>
    <p class="MsoNormal">
    <font face="Times New Roman" size="3" color="#000000"><span> I think we need some about of give and take. </span></font>
    </p>
    <font face="Times New Roman" size="3" color="#000000"> </font>
    <p class="MsoNormal">
    <font face="Times New Roman" size="3" color="#000000">My feeling is that ZO does a great job of executing its programming. I do not love the programming. I presume that the delay of the engine is accounted for in the programming.</font>
    </p>
    <p class="MsoNormal">
    <font face="Times New Roman" size="3" color="#000000"></font>
    </p>
    <p class="MsoNormal">
    <font size="3"><font color="#000000"><font face="Times New Roman">In some boats (Prop – engine- other factor) some settings feel great and other boats feel totally foreign. <span>In the end I want all boats to feel the same and that is the problem now.</span></font></font></font>
    </p>
    Join us at the 2019 NorCal MasterCraft BallOfSpray Cash Prize


    Support BallOfSpray by supporting the companies that support BallOfSpray

    Barts ★ Connelly ★  DBSkis ★ Goode ★ Hobe Lake ★ HO Syndicate 

    MasterCraft ★ Masterline ★ MOB ★ Performance Ski and Surf ★ Reflex ★ Radar ★ Stokes

  • ToddLToddL Posts: 2,788 Mega Baller
    edited September 2010
    <p>
    DN / GMC,
    </p>
    <p>
    The area of the curve under the line (skier's force pulling speed down) has to equal the area of the curve above the line (the throttle force pulling the speed up) so that they cancel each other out with ideal speed between any two buoys.    This is an absolute with ZO since it must provide ideal times at each buoy.
    </p>
    <p>
    The goals are simple:

    1) Ski with as minimal skier force to minimize the amount of reaction the throttle has to use to compensate.
    </p>
    <p>
    2) Attempt to best match your force curve with the right letter/number combination so as to further reduce the area of both curves.
    </p>
    <p>


    Thus...

    A) If you hit the boat hard out of the buoy, you want the throttle to react sooner and harder so that the net impact of your pull is neutral to the overall boat speed.   
    </p>
    <p>
    B) If you are soft and pull long out of the buoy, then you want the throttle to react lighter and later so that the net impact of your pull is neutral to the overall boat speed.   
    </p>
    <p>
     
    </p>
    <p>
    Consider how your pull style changes from starting pass to final pass and why your letter/number setting might need to change, too.
    </p>
    <p>
    Regards,

    Todd
    </p>
    -- The future of skiing depends upon welcoming novice skiers regardless of age to our sport.
  • HortonHorton Posts: 27,225 Administrator
    <p>
    Todd,
    </p>
    <p>
    I am not sure if this is out side your point but the letters are different curves and as a skier that is a huge deal. Darn I will I could repeat what Will told me. Should of had a tape runnning.
    </p>
    <p>
     
    </p>
    Join us at the 2019 NorCal MasterCraft BallOfSpray Cash Prize


    Support BallOfSpray by supporting the companies that support BallOfSpray

    Barts ★ Connelly ★  DBSkis ★ Goode ★ Hobe Lake ★ HO Syndicate 

    MasterCraft ★ Masterline ★ MOB ★ Performance Ski and Surf ★ Reflex ★ Radar ★ Stokes

  • dndn Posts: 18 Baller
    <p>
    Yeah, I'd love to hear what Will Bush said too!  Write him a letter & see if he'll write it down.
    </p>
    <p>
    You asked if a faster response from the boat would help things - YES.  When the boat takes 200ms to respond, you'll slow the boat say 0.5mph before ZO & the engine starts to reverse the loss.  If it could respond in 2ms, you'd only have slowed the boat 0.005mph before it starts to respond.  You wouldn't notice that little of a slowdown.  As far as you'd notice, the boat would respond equally and proportionally to your pull in real time.  There would be no 'payback' later!  The boat would always go 34.2mph and we'd have to find something else to complain about.
    </p>
    <p>
    The problem, as I mentioned earlier, is that for a variety of reasons we're unlikely to ever get the response much faster than what it already is.  Besides the built in limitations of physics, the ECM has to worry about emissions, fuel efficiency, engine stress & wear, etc. and then there are limitations on how fast GPS can update too.  The next best approach is to anticipate the skier pull, with the overall goal to be the same as in the first case:  keep the slowdown minimal so the payback can be minimized.
    </p>
    <p>
    If ZO can be improved at some point, great.  Until then, work on becoming 'invisible'.
    </p>
    <p>
    good comments, Todd.
    </p>
    <p>
    David Nelson
    </p>
  • eleeskieleeski Posts: 3,819 Infinite Pandas
    <p>
    Doesn't ZO have accelerometers that measure speed changes before GPS inputs would show a deviation?
    </p>
    <p>
    Horton poses the critical question - how much speed variation do we want? Does a large variation lead to a softer feel? In jumping, isn't the widest acceptable variation going to give the longest jumps (as long as the fastest portion is right at the ramp?)?
    </p>
    <p>
    For me ZO feels the best when the speed is 16.94. It does seem to miss hot by .01 frequently.
    </p>
    <p>
    Eric
    </p>
  • skibugskibug Posts: 2,029
    <p>
    Eric,
    </p>
    <p>
    I believe you are correct about the accelerometers, I heard / read that in another thead.
    </p>
    Bob Grizzi
  • Jody_SealJody_Seal Posts: 2,633 Mega Baller
    Their are three Accelerometers in the head unit.
    Hobby Boats can be expensive when the hobbyist is limited on their own skill and expertise.


  • scokescoke Posts: 644 Solid Baller
    edited September 2010
    <p>
     
    </p>
    <p>
    Quite a bit of talk in this thread but the problem is most of the solutions we have already crossed those bridges, ie, been there done that. We aren't going back.
    </p>
    <p>
     examples:
    </p>
    <p>
    <em>We want the speed of the system dip only 0.1mph</em>. Been there. The first few iterations, L-N maybe, where so harsh very few people could ski them. The speed barely changed as the skier loaded. Held the 0.1mph very closely. This was the fall of 2007. Mortals could barely run passes. That's when P then P101 version spun out.
    </p>
    <p>
    <em>We only want one setting.</em> Been there. early versions had 1 setting and everyone complained and wanted options.
    </p>
    <p>
    <em>We have 3 settings but now we want more.</em> Ok, then Rev "X" came out, later version Q officially. People complaining that it's too hard to figure out the settings.
    </p>
    <p>
     <em>Oh now we want the system to have a 4th setting "D" like SG/PP. </em>B2 is the most benign middle ground setting of them all. If you don't use or try the other settings, you aren't actually benefitting from the system and more than likely fighting it.
    </p>
    <p>
     
    </p>
    <p>
     
    </p>
    <p>
    1-2-3 are trigger points on intensity of engaging.
    </p>
    <p>
    A-B-C are location letters of the system engaging.
    </p>
    <p>
     I could be wrong but mark it, the Rev Q we have will be the same for the remainder of 2010 and 2011. No changes.
    </p>
    <p>
     
    </p>
  • Than_BoganThan_Bogan Posts: 6,359 Mega Baller
    <p>
    scoke, these are valid and important points, but they do not imply the system cannot be improved.
    </p>
    <p>
    I view the settings as a crutch -- just like when you need real crutches it's a LOT better than the alternative.  But if you actually didn't need the crutch, it would be better.
    </p>
    <p>
    Overall, I think ZO is the best we've had in this sport so far.  But I still think it can be better.
    </p>
    Nathaniel Bogan -- GUT Padawan
  • scokescoke Posts: 644 Solid Baller
    <p>
     
    </p>
    <p>
    Than, agreed about making things better but at a certain point we have to ask "why?" what level is "good enough"?
    </p>
    <p>
    I am a realist, ZO won't be changing anytime soon. Why should it.
    </p>
    <p>
     
    </p>
  • HortonHorton Posts: 27,225 Administrator
    I sent Will an email. He is traveling but I think at some point he will write something for us.
    Join us at the 2019 NorCal MasterCraft BallOfSpray Cash Prize


    Support BallOfSpray by supporting the companies that support BallOfSpray

    Barts ★ Connelly ★  DBSkis ★ Goode ★ Hobe Lake ★ HO Syndicate 

    MasterCraft ★ Masterline ★ MOB ★ Performance Ski and Surf ★ Reflex ★ Radar ★ Stokes

  • DirtDirt Posts: 1,584 Open or 55K Rated Skier
    I think ZO is probably fine if you have the ability to train behind it often. I think it is pretty drastic when you practice behind PP and ski tournaments behind ZO. Weighing 215 lbs and not the best technique may have something to do with it. It is difficult to get a good gate and 1 ball. This is not too critical until the line gets short. Any other "big boned" skiers notice the same thing? 
    I learned everything I know not to do from Horton

Leave a Comment

BoldItalicStrikethroughOrdered listUnordered list
Emoji
Image
Align leftAlign centerAlign rightToggle HTML viewToggle full pageToggle lights
Drop image/file