Tornado-Cat

Boats and foils comparison

Recommended Posts

The cups can read the same, but there is a bunch of voodoo between the compass speedo and wind direction that convert apparent wind back into true wind. If I remember, these were never done by a single formula, but each vendor had his own way of fudging it back to believable numbers.  I think the biggest problem was the “upwash correction” which is the amount the rig bends the wind before it gets to the mast.  On the wings, the upwash was dependent on the lift coefficient, which in our case went up and down something like 150%.     This means that TWD & TWS can be consistent within each program( and they have to be to make any testing valid) but may not correlate between programs.  I’m sure there are people here who know all about this stuff.

When the boats  are sailing side by side you would expect to see very similar numbers.  This never happened on Saturday.

SHC

Share this post


Link to post
Share on other sites
47 minutes ago, Steve Clark said:

The cups can read the same, but there is a bunch of voodoo between the compass speedo and wind direction that convert apparent wind back into true wind. If I remember, these were never done by a single formula, but each vendor had his own way of fudging it back to believable numbers.  I think the biggest problem was the “upwash correction” which is the amount the rig bends the wind before it gets to the mast.  On the wings, the upwash was dependent on the lift coefficient, which in our case went up and down something like 150%.     This means that TWD & TWS can be consistent within each program( and they have to be to make any testing valid) but may not correlate between programs.  I’m sure there are people here who know all about this stuff.

When the boats  are sailing side by side you would expect to see very similar numbers.  This never happened on Saturday.

SHC

I'm not 100%, sure but I have a feeling that the instruments providing data for virtual eye and broadcast are supplied equipment potentially running the same software with identical settings. If that's true, then the differences we see in TWS numbers may be caused by differences in sail design, size and trim. That would be an interesting challenge to calculate back from errors in TWS / TWA traces.

Share this post


Link to post
Share on other sites

great screenshot of how low the surface area is for AM. Fast but not enough life in light air

Screen_Shot_2021-01-16_at_6_05.15_PM.png

Share this post


Link to post
Share on other sites
2 hours ago, WindySurfer said:

After jerk come jolt and jounce, but after that we find some very silly physicists who liked rice crispees.  The 7th, 8th, and 9th derivatives of position are snap crackle and pop.

2nd derivative is otherwise known as acceleration. Of course, integral is actually what matters to get to the finish line first!

Share this post


Link to post
Share on other sites

LR data. @X35 GER50, note the wind strength was different between two weeks, so vmg/tws might be better. 

 

image.png.d9c6eb96c168d167c22211240489ca4c.pngimage.png.705f4900c39d35fec54ac5b818adcd97.png

image.png.a3395902f22102dffad467321442861f.png

  • Like 1

Share this post


Link to post
Share on other sites
8 hours ago, dorox said:

LR data. @X35 GER50, note the wind strength was different between two weeks, so vmg/tws might be better. 

 

image.png.d9c6eb96c168d167c22211240489ca4c.pngimage.png.705f4900c39d35fec54ac5b818adcd97.png

image.png.a3395902f22102dffad467321442861f.png

Thanks Dorox! Can you add to the comparison the data from today?

 

Share this post


Link to post
Share on other sites
21 hours ago, X35 GER50 said:

Thanks Dorox! Can you add to the comparison the data from today?

 

I haven't implemented the height analysis in my code yet. Do we know where they measure it? Bow?

What's the theory behind correlating height with vmg? I think the height numbers could depend a lot on pitch and heel angles and the optimal height may be different for each boat.

Share this post


Link to post
Share on other sites

I've done the height analysis as well, but I don't know how accurate these numbers are. I doubt LR sails 70 cm higher than INEOS.

1062213235_uwheight.png.89446ce51b60ff5c251f7f88e8884d98.png  534154244_dwheight.png.50702030fc9dd2f7f963101c4455161a.png

Height during tacks:

height2.png.b0c679768dc14669a5cadc1fa623ccb3.png

Share this post


Link to post
Share on other sites
33 minutes ago, erdb said:

I've done the height analysis as well, but I don't know how accurate these numbers are. I doubt LR sails 70 cm higher than INEOS.

1062213235_uwheight.png.89446ce51b60ff5c251f7f88e8884d98.png  534154244_dwheight.png.50702030fc9dd2f7f963101c4455161a.png

Height during tacks:

height2.png.b0c679768dc14669a5cadc1fa623ccb3.png

thanks erbd for looking at that. woul dbe nice to have LR race 6 vs LR race 8. But the numbers seems the same. Probably they have a different offset in the sensor placement.

Share this post


Link to post
Share on other sites
On 1/17/2021 at 3:52 AM, erdb said:

I'm not 100%, sure but I have a feeling that the instruments providing data for virtual eye and broadcast are supplied equipment potentially running the same software with identical settings. If that's true, then the differences we see in TWS numbers may be caused by differences in sail design, size and trim. That would be an interesting challenge to calculate back from errors in TWS / TWA traces.

Erdb,

Thanks for the very interesting performance analyses that I discover only now - great job!

Years ago we did a study for TackTick on the influence of the sail on AWA & AWS as measured in the top of the mast. There certainly is a noteworthy influence, and sail trim affects too like Steve Clark is saying, but could their sail trim be that much different? I doubt it.

 

AWA corr1.png

AWAcorr2.png

  • Like 1

Share this post


Link to post
Share on other sites
On 1/17/2021 at 4:10 AM, zillafreak said:

great screenshot of how low the surface area is for AM. Fast but not enough life in light air

Screen_Shot_2021-01-16_at_6_05.15_PM.png

Small wetted area means less drag in most conditions, but not all. Because lift force must be pretty equal for same conditions for all the boats, (weight being same and presumably heeling force from sailplan also very similar in identical conditions), small area must mean higher lift coefficient. That results earlier cavitation (lower boatspeed when cavitation begins being an issue), which is the opposite of being fast.

It's therefore not just the takeoff condition when small area is a hindrance.

 

 

Share this post


Link to post
Share on other sites

Is their any collective sense here about who has used the most of their sail allocations? Is saving fresh sails for last a factor? 

Share this post


Link to post
Share on other sites
On 1/16/2021 at 7:58 PM, Steve Clark said:

The cups can read the same, but there is a bunch of voodoo between the compass speedo and wind direction that convert apparent wind back into true wind. If I remember, these were never done by a single formula, but each vendor had his own way of fudging it back to believable numbers.  I think the biggest problem was the “upwash correction” which is the amount the rig bends the wind before it gets to the mast.  On the wings, the upwash was dependent on the lift coefficient, which in our case went up and down something like 150%.     This means that TWD & TWS can be consistent within each program( and they have to be to make any testing valid) but may not correlate between programs.  I’m sure there are people here who know all about this stuff.

When the boats  are sailing side by side you would expect to see very similar numbers.  This never happened on Saturday.

SHC

 

I might be wrong, and not that it matters much, but i thought "upwash" was reserved for the effect that occurs when the apparent wind is from behind, and the wind flows up and over the eased mainsail from behind, causing a zone of anomalously high wind speed, and change of direction near the sensor.

when the apparent wind is forward - as it always is on these boats, the airflow deflects to flow around the rig well ahead of the rig which causes an error in direction, but not so much in speed.

But one thing I read somewhere about the TWD/TWS calculation for these boats is that they have to use a tidal current model, which would be a whole new source of error.

The reason is that they are mostly using SOG, not BSP, as it's nearly impossible to measure BSP on boats like this. If SOG and COG are used in the wind calculation instead of BSP and HDG.., the calculation yields wind referred to the earth, which is usually called ground wind: ground wind speed and ground wind direction. By convention, TWS and TWD are referred to the water, which moves with respect to the earth. Supposedly, they are taking the calculated ground wind.., adding the model current - so changing from a coordinate system fixed to the earth, to one fixed to the water - and getting TWD/TWS.

I did see this interesting article about a Doppler device being used by INEOS to measure actual BSP during development., but I'm not sure it's on their race foils, or if any of the other boats are using it.

https://insight.nortekgroup.com/ineos

 

 

 

Share this post


Link to post
Share on other sites

The data that Dorox is pulling, does it include very-precise GPS coordinates at 10 times per sec? The (freely available) Race Data sets during the past 2 AC’s did include that data, the mark positions too. It made for extremely accurate COG and VMG calcs. 

Share this post


Link to post
Share on other sites
1 hour ago, Stingray~ said:

The data that Dorox is pulling, does it include very-precise GPS coordinates at 10 times per sec? The (freely available) Race Data sets during the past 2 AC’s did include that data, the mark positions too. It made for extremely accurate COG and VMG calcs. 

I haven't seen GPS coordinates so far. At least not in the files I'm able to get based on his instructions. There may be a way, because the virtual eye display needs those coordinates, too. Maybe Dorox could figure out how to pull those as well. 

I was also wondering about what technology the umpires use this time. I remember in SF and Bermuda there was an additional system on top of GPS that allowed getting boat and camera positions with 1-2 inch accuracy both for the umpires and for Stan Honey's graphics. I wonder if there's something like that now. The on-screen graphics did seem a bit jumpier, especially during the ACWS. 

 

Share this post


Link to post
Share on other sites
18 minutes ago, erdb said:

I haven't seen GPS coordinates so far. At least not in the files I'm able to get based on his instructions. There may be a way, because the virtual eye display needs those coordinates, too. Maybe Dorox could figure out how to pull those as well. 

 

Here’s hoping, that data is basically the dynamite. 

Share this post


Link to post
Share on other sites
20 minutes ago, erdb said:

I was also wondering about what technology the umpires use this time. I remember in SF and Bermuda there was an additional system on top of GPS that allowed getting boat and camera positions with 1-2 inch accuracy both for the umpires and for Stan Honey's graphics. I wonder if there's something like that now.

The ‘projection arrows’ were cool too, would have been very illustrative in for example the close-call on the last leg of ITA/GBR. 

Share this post


Link to post
Share on other sites
10 minutes ago, Stingray~ said:

The ‘projection arrows’ were cool too, would have been very illustrative in for example the close-call on the last leg of ITA/GBR. 

Funny, I was just looking at that. The boat images don't scale accurately on Virtual Eye when you zoom in, but the little circles showing the tracks of the boats seem to be 3 seconds apart. I scaled the image based on the circles and the speeds of the boats at that time and draw the dimensions of the boats on the image (with yellow lines of course). It seems the closest they got was 31.5m, which at those speeds was about 1.4 seconds. That was the true delta in that race. If LR gets there 1 second earlier, they win and we get to watch another race the next day in 20 kts of wind...:angry:

cross2.thumb.JPG.8ddfd5bba51546e1e06a49c0110bf3d4.JPG

  • Like 3

Share this post


Link to post
Share on other sites
1 hour ago, erdb said:

Funny, I was just looking at that. The boat images don't scale accurately on Virtual Eye when you zoom in, but the little circles showing the tracks of the boats seem to be 3 seconds apart. I scaled the image based on the circles and the speeds of the boats at that time and draw the dimensions of the boats on the image (with yellow lines of course). It seems the closest they got was 31.5m, which at those speeds was about 1.4 seconds. That was the true delta in that race. If LR gets there 1 second earlier, they win and we get to watch another race the next day in 20 kts of wind...:angry:

cross2.thumb.JPG.8ddfd5bba51546e1e06a49c0110bf3d4.JPG

Nice work! 
 

Still wish we had better data though, you’d be even sharper! 

Share this post


Link to post
Share on other sites
4 hours ago, erdb said:

I haven't seen GPS coordinates so far. At least not in the files I'm able to get based on his instructions. There may be a way, because the virtual eye display needs those coordinates, too. Maybe Dorox could figure out how to pull those as well. 

I was also wondering about what technology the umpires use this time. I remember in SF and Bermuda there was an additional system on top of GPS that allowed getting boat and camera positions with 1-2 inch accuracy both for the umpires and for Stan Honey's graphics. I wonder if there's something like that now. The on-screen graphics did seem a bit jumpier, especially during the ACWS. 

 

In the JSON file the positional data is:

> coordIntep
        > xCerp
        > yCerp
 

  • Like 1

Share this post


Link to post
Share on other sites
46 minutes ago, MaxHugen said:

In the JSON file the positional data is:

> coordIntep
        > xCerp
        > yCerp
 

Ha, you're right! Somehow I've missed that. It's not Lat/Lon, it's converted to meters, but we have positions with 20Hz frequency! I don't really know what to do with it, but it's cool. Let me know if you have any ideas.

1832642958_Pradar8.thumb.png.c7573b6097a5004fe2815d67c970d0e3.png

 

 

 

  • Like 3

Share this post


Link to post
Share on other sites
5 minutes ago, erdb said:

Ha, you're right! Somehow I've missed that. It's not Lat/Lon, it's converted to meters, but we have positions with 20Hz frequency! I don't really know what to do with it, but it's cool. Let me know if you have any ideas.

I had thought a while back to look at the distance made good during tacks/gybes? 

Deciding the period to examine may be a little tricky, eg they may bear away a bit for a few secs to pick up speed before/after the maneuver, so that time period should probably be included.

Share this post


Link to post
Share on other sites
3 minutes ago, MaxHugen said:

I had thought a while back to look at the distance made good during tacks/gybes? 

Deciding the period to examine may be a little tricky, eg they may bear away a bit for a few secs to pick up speed before/after the maneuver, so that time period should probably be included.

I've been doing that already using the TWA, speed and VMG data:

vmgs2.png.37171b3b659f7fcb51c115f2ae65c4b3.png

In a way it's more accurate, because it corrects for wind shifts (as long as the TWA data is accurate). From the GPS tracks, we could get radius of tacks and jibes by each boat and  G forces, but it's not that easy. We could also get minimum distances between boats. That'd be relatively simple.

Share this post


Link to post
Share on other sites

Here is the distance between INEOS and LR during the last race:

distance1.thumb.png.66880bc59d7b2f5ba820810a134f4f17.png

The min distance between the two was 19.24 m at the start. The closest they got during that last cross was 30.9 m. Pretty close to what I got from the scaled screenshot. Of course, this is the distance between the GPS antennas or whatever reference point they calculate from that on the boat.

Share this post


Link to post
Share on other sites
On 1/24/2021 at 5:41 AM, erdb said:

I haven't implemented the height analysis in my code yet. Do we know where they measure it? Bow?

What's the theory behind correlating height with vmg? I think the height numbers could depend a lot on pitch and heel angles and the optimal height may be different for each boat.

I thought the intent was to map height to boat speed rather than VMG to see what speed the boats took off

Share this post


Link to post
Share on other sites

I made another web app with boat tracks and data side-by-side. It is better to use it on a PC, because map zoom doesn't seem to work on mobile.

https://ac36.herokuapp.com/map

So far it has only race 8 from Prada cup. You can scroll the slider at the bottom to change the time and see the race progress.

It seems like LR got really unlucky with wind shifts at the last upwind.

The boats are shown roughly to scale, but I didn't have a good image for AC75 from above. If someone wants to draw it, it would be really cool. It has to be .png with transparent background.

image.png.f518cf3e24c4cd88a841be6ee4db879e.png

 

image.png

  • Like 12

Share this post


Link to post
Share on other sites
1 hour ago, dorox said:

I made another web app with boat tracks and data side-by-side. It is better to use it on a PC, because map zoom doesn't seem to work on mobile.

https://ac36.herokuapp.com/map

So far it has only race 8 from Prada cup. You can scroll the slider at the bottom to change the time and see the race progress.

It seems like LR got really unlucky with wind shifts at the last upwind.

The boats are shown roughly to scale, but I didn't have a good image for AC75 from above. If someone wants to draw it, it would be really cool. It has to be .png with transparent background.

image.png.f518cf3e24c4cd88a841be6ee4db879e.png

 

image.png

This is awesome. I'm sure @MaxHugen could give you some beautiful boat images - maybe even team specific ;)

Share this post


Link to post
Share on other sites

A little video I made... setting up for more detailed discussion on individual teams. I hope it's a good baseline for the 'design problem' discussion. 

Thanks to all the anarchists, especially this thread!

 

  • Like 6

Share this post


Link to post
Share on other sites

Good again @Mozzy Sails

One comment is that you forgot that the foils also have to have a void for this supplied transponder for the race committee usage

And it did remind of one question that has been at the back of my mind for a long time: why did they insist that the foils were symmetrical?

  • Like 2

Share this post


Link to post
Share on other sites
23 minutes ago, enigmatically2 said:

Good again @Mozzy Sails

One comment is that you forgot that the foils also have to have a void for this supplied transponder for the race committee usage

And it did remind of one question that has been at the back of my mind for a long time: why did they insist that the foils were symmetrical?

Since the sections are symmetrical, how much downforce do the foils produce when they are being hauled up out of the water?  How much drag is this?  Do the teams account for this when retracting foils?

Share this post


Link to post
Share on other sites
1 hour ago, Amati said:

Since the sections are symmetrical, how much downforce do the foils produce when they are being hauled up out of the water?  How much drag is this?  Do the teams account for this when retracting foils?

If they normally supply lift, why would they produce downforce when coming up?

Share this post


Link to post
Share on other sites
14 hours ago, MaxHugen said:

In the JSON file the positional data is:

> coordIntep
        > xCerp
        > yCerp
 

Great find! Nice, nice, nice

Are mark positions available too? 

Share this post


Link to post
Share on other sites
18 minutes ago, enigmatically2 said:

If they normally supply lift, why would they produce downforce when coming up?

They are symmetrical sections?  Or, when folks are talking symmetry they’re talking about bilateral symmetry?  At any rate, when the windward foils come up, there’s  a fair amount of water coming off, at least.

On the nomenclature side, are the foils that come off the side of the boats officially foils, or are they called struts, and only the two foils on the ends are foils?

Share this post


Link to post
Share on other sites
19 minutes ago, Amati said:

They are symmetrical sections?  Or, when folks are talking symmetry they’re talking about bilateral symmetry?  At any rate, when the windward foils come up, there’s  a fair amount of water coming off, at least.

On the nomenclature side, are the foils that come off the side of the boats officially foils, or are they called struts, and only the two foils on the ends are foils?

Only the bits that stick out of the arm are foils, and they have to be symmetrical about the arm. So they both produce lift (unless flaps are lived to do otherwise)

Share this post


Link to post
Share on other sites
2 hours ago, enigmatically2 said:

Good again @Mozzy Sails

One comment is that you forgot that the foils also have to have a void for this supplied transponder for the race committee usage

And it did remind of one question that has been at the back of my mind for a long time: why did they insist that the foils were symmetrical?

I think to give the foil arm manufactures (one design supplied equipment) some chance of predicating loads. Any asymmetry in load either side of the arm can only come from flap position, which is again limited by flap size (and competitors willingness to have that much drag).

  • Like 1

Share this post


Link to post
Share on other sites
10 minutes ago, enigmatically2 said:

Only the bits that stick out of the arm are foils, and they have to be symmetrical about the arm. So they both produce lift (unless flaps are lived to do otherwise)

Arms.  OK.  And the sections of the foils can be anything?  So the arms are streamlined, and  lift neutral, even if aoa changes?

Share this post


Link to post
Share on other sites
17 minutes ago, Mozzy Sails said:

I think to give the foil arm manufactures (one design supplied equipment) some chance of predicating loads. Any asymmetry in load either side of the arm can only come from flap position, which is again limited by flap size (and competitors willingness to have that much drag).

How much does load change going through waves?  (Does wave circulation change aoa as the arm goes through or meets a wave?). Are the arms a particular section?  Are the horizontal stripes on the arms anti ventilation fences?

Share this post


Link to post
Share on other sites
44 minutes ago, Amati said:

They are symmetrical sections?  Or, when folks are talking symmetry they’re talking about bilateral symmetry?  At any rate, when the windward foils come up, there’s  a fair amount of water coming off, at least.

On the nomenclature side, are the foils that come off the side of the boats officially foils, or are they called struts, and only the two foils on the ends are foils?

Most of us refer to foil arms and foil wings. The arms do have different trailing fairings across the different boats and there’s been some speculation that those designs might be inducing ‘crabbing.’

Share this post


Link to post
Share on other sites
5 minutes ago, Stingray~ said:

Most of us refer to foil arms and foil wings. The arms do have different trailing fairings across the different boats and there’s been some speculation that those designs might be inducing ‘crabbing.’

Like Laser crabbing?  I’m trying to picture that.....

Share this post


Link to post
Share on other sites
2 minutes ago, Amati said:

Like Laser crabbing?  I’m trying to picture that.....

Where the boat (and therefore rig too) are at a lower AWA than the hull and its course.. We’ve seen glimpses of it but the ‘upwind drift’ could have been tidal. 

  • Like 1

Share this post


Link to post
Share on other sites
15 hours ago, MaxHugen said:

In the JSON file the positional data is:

> coordIntep
        > xCerp
        > yCerp
 

Can you please show an example of this data? If it’s not lat/long to the nth decimal but meters instead, then presumable it is X and Y using some reference grid? 

Share this post


Link to post
Share on other sites
3 minutes ago, Stingray~ said:

Can you please show an example of this data? If it’s not lat/long to the nth decimal but meters instead, then presumable it is X and Y using some reference grid? 

the data is in UTM format for zone 60S. If I understand correctly, these are x, y coordinates in meters relative to a specific point. Mark data is there too, but is arguably harder to pull out. There is also wind data for buoys and committee boat and race boundaries.

7 minutes ago, Stingray~ said:

Where the boat (and therefore rig too) are at a lower AWA than the hull and its course.. We’ve seen glimpses of it but the ‘upwind drift’ could have been tidal. 

You can actually get the leeway angle from the available GPS velocity vector and heading. Alternatively, you can see it for yourself here https://ac36.herokuapp.com/map. The boat heading data is used to rotate the boat icon independently of GPS. So if you zoom in to see the boat icons, you'll notice their heading is not always aligned with GPS track. This is especially pronounced on downwind legs.

image.thumb.png.c810c13d2378c5372a892ecfab94b9b4.png

  • Like 3

Share this post


Link to post
Share on other sites
19 minutes ago, dorox said:

the data is in UTM format for zone 60S. If I understand correctly, these are x, y coordinates in meters relative to a specific point. Mark data is there too, but is arguably harder to pull out. There is also wind data for buoys and committee boat and race boundaries.

You can actually get the leeway angle from the available GPS velocity vector and heading. Alternatively, you can see it for yourself here https://ac36.herokuapp.com/map. The boat heading data is used to rotate the boat icon independently of GPS. So if you zoom in to see the boat icons, you'll notice their heading is not always aligned with GPS track. This is especially pronounced on downwind legs.

image.thumb.png.c810c13d2378c5372a892ecfab94b9b4.png

Awesome! 

Share this post


Link to post
Share on other sites
21 minutes ago, dorox said:

Mark data is there too, but is arguably harder to pull out.

Would make for brilliant Course-VMG calcs if based on mid-gate reference points, and/or based on the mark ultimately rounded so as not to skew the last part of the leg’s VMG. 

Share this post


Link to post
Share on other sites
2 hours ago, dorox said:

the data is in UTM format for zone 60S. If I understand correctly, these are x, y coordinates in meters relative to a specific point. Mark data is there too, but is arguably harder to pull out. There is also wind data for buoys and committee boat and race boundaries.

You can actually get the leeway angle from the available GPS velocity vector and heading. Alternatively, you can see it for yourself here https://ac36.herokuapp.com/map. The boat heading data is used to rotate the boat icon independently of GPS. So if you zoom in to see the boat icons, you'll notice their heading is not always aligned with GPS track. This is especially pronounced on downwind legs.

 

Interesting! So how did you figure out what the reference point is? You were somehow able to fit the tracks on the map. Really nice work!

The leeway is interesting. I would have thought they get the heading from the GPS, but if it's separate as you say, I may be able to analyze the leeway angle for each boat the same way I analyze all the other parameters. 

 

Share this post


Link to post
Share on other sites
1 hour ago, Stingray~ said:

Would make for brilliant Course-VMG calcs if based on mid-gate reference points, and/or based on the mark ultimately rounded so as not to skew the last part of the leg’s VMG. 

The problem is that often times the left and right side of the course have very different wind direction and speed. If you just use one buoy somewhere on the course, it won't really tell the story. I guess the question is, what you'd like to know. If you want to compare boat performance, then the local TWA, TWS measured by the boats is better (as long as they reasonably accurate) to get a VMG against local wind direction. If you want to evaluate how well the teams position themselves on the course, you could use the axis of the course or buoy data or just use the average TWA from the boats. That's what I'm using for those "on favored tacks" graphs (I should come up with a better name for those).

Share this post


Link to post
Share on other sites

Video of ETNZ practice today, January 26, shows them sailing with both foils down.  They seemed to be doing as many maneuvers as they could manage and still stay on foils.  This makes sense as they have just shipped the smallest set of foils seen to date.

I had been wondering about using both foils for trying to get out of the water. 

SHc

 

Share this post


Link to post
Share on other sites
1 hour ago, erdb said:

The problem is that often times the left and right side of the course have very different wind direction and speed. If you just use one buoy somewhere on the course, it won't really tell the story. I guess the question is, what you'd like to know. If you want to compare boat performance, then the local TWA, TWS measured by the boats is better (as long as they reasonably accurate) to get a VMG against local wind direction. If you want to evaluate how well the teams position themselves on the course, you could use the axis of the course or buoy data or just use the average TWA from the boats. That's what I'm using for those "on favored tacks" graphs (I should come up with a better name for those).

Yes, exactly, but need it be an either or VMG measurement? 
 

Perfomance-wise, VMG to the Wind is great but tactics-wise, VMG to the course (mark or mid-gate) is brilliant too. 

Share this post


Link to post
Share on other sites
32 minutes ago, Steve Clark said:

Video of ETNZ practice today, January 26, shows them sailing with both foils down.  They seemed to be doing as many maneuvers as they could manage and still stay on foils.  This makes sense as they have just shipped the smallest set of foils seen to date.

I had been wondering about using both foils for trying to get out of the water. 

SHc

 

I'm wondering if they are practicing "stay up on manoeuvres at all costs" moves. Even if you slow drastically through a tack on both foils, the drag on two would still be no where near drag on even a small hull kiss. And there's stability gained on exit that would aid precise sail trimming.

  • Like 1

Share this post


Link to post
Share on other sites
42 minutes ago, Stingray~ said:

Yes, exactly, but need it be an either or VMG measurement? 
 

Perfomance-wise, VMG to the Wind is great but tactics-wise, VMG to the course (mark or mid-gate) is brilliant too. 

Certainly both are interesting. The problem is that if a boat has a better performance/VMG to begin with, they'll look good on the VMG to course, too, even if they aren't following the best tactics.  Something about how a fast boat makes the tactician seem like a genius.

That's why I came up with that method of analyzing tactical choices in degrees x seconds. Basically, for every second a boat sails on the favored tack, the angle of the shift in degrees is added to their "on favored tack" total. For every second they are on the wrong tack, the same number is deducted. This gives a number that's more or less isolated from boat performance. I've never seen anything like this used before, so I'm sure there are a lot of issues with it, but so far, it seemed to mirror what was happening in the race and matched the commentary from NO and KR.

  • Like 1

Share this post


Link to post
Share on other sites
2 minutes ago, erdb said:

Certainly both are interesting. The problem is that if a boat has a better performance/VMG to begin with, they'll look good on the VMG to course, too, even if they aren't following the best tactics.  Something about how a fast boat makes the tactician seem like a genius.

That's why I came up with that method of analyzing tactical choices in degrees x seconds. Basically, for every second a boat sails on the favored tack, the angle of the shift in degrees is added to their "on favored tack" total. For every second they are on the wrong tack, the same number is deducted. This gives a number that's more or less isolated from boat performance. I've never seen anything like this used before, so I'm sure there are a lot of issues with it, but so far, it seemed to mirror what was happening in the race and matched the commentary from NO and KR.

I absolutely love that measure and yes, it’s novel and very clever! 

Share this post


Link to post
Share on other sites

TE posed a question on FB today asking (am not sure why, but) if there has been noticeable difference between when JS is on the helm and Bruni doing flight control (starboard) versus on the opposite side? Are you set up well enough to differentiate? 
 

in old-school something like 

select C-VMG where 

boat=‘ITA’ and 

Race = 2021011502

group by direction, tack

Share this post


Link to post
Share on other sites
11 hours ago, dorox said:

I made another web app with boat tracks and data side-by-side. It is better to use it on a PC, because map zoom doesn't seem to work on mobile.

https://ac36.herokuapp.com/map

So far it has only race 8 from Prada cup. You can scroll the slider at the bottom to change the time and see the race progress.

It seems like LR got really unlucky with wind shifts at the last upwind.

The boats are shown roughly to scale, but I didn't have a good image for AC75 from above. If someone wants to draw it, it would be really cool. It has to be .png with transparent background.

image.png.f518cf3e24c4cd88a841be6ee4db879e.png

 

image.png

Awesome job mate! I had some suggestions - but when I went to your app you'd already implemented them - the slider and sync'd chart!

Sent you an SVG file of boats by email before realising I could have attached it here. :D

Is animation also on the to-do list?

  • Like 1

Share this post


Link to post
Share on other sites

You guys are amazing !!

As long as you do not provide spinbot with some stats that he can use to prove AM was fastest even though they did not win the cup?

  • Like 1

Share this post


Link to post
Share on other sites
12 hours ago, dorox said:

...

It seems like LR got really unlucky with wind shifts at the last upwind....

This is something I've found in my club racing to be a really unfair part of the whole sport of sailing.  The better sailors get a disproportionate amount of the lucky wind shifts!

  • Like 10

Share this post


Link to post
Share on other sites
1 hour ago, Stingray~ said:

Yes, exactly, but need it be an either or VMG measurement? 
 

Perfomance-wise, VMG to the Wind is great but tactics-wise, VMG to the course (mark or mid-gate) is brilliant too. 

That's something I've thought about too. Especially as they get closer to the gates, VMG to the mark can be quite different to VMG to the course as a whole.

No doubt the teams know every optimum angle to sail from every metre of the course! :)

Share this post


Link to post
Share on other sites
10 minutes ago, MaxHugen said:

Awesome job mate!

+10! Awesome is right 

Share this post


Link to post
Share on other sites
4 minutes ago, MaxHugen said:

That's something I've thought about too. Especially as they get closer to the gates, VMG to the mark can be quite different to VMG to the course as a whole.

No doubt the teams know every optimum angle to sail from every metre of the course! :)

Damn, I ran out of Likes today! :) 

Share this post


Link to post
Share on other sites
9 hours ago, Mikko Brummer said:

Setting up Luna Rossa for an aero-simulation, to compare the double skin to a small wing mast+regular sail. . Original model by LCurreriDesign at https://www.turbosquid.com/Search/3D-Models/ac75

LR1.png

LR3.png

LR2.png

LR4.png

Look forward to seeing your sim results.  Been dying to see some airflow sims especially over the hulls, hull/sail intersection, etc. (there's a long list of wishes actually)  ;)

Share this post


Link to post
Share on other sites
8 hours ago, Mozzy Sails said:

A little video I made... setting up for more detailed discussion on individual teams. I hope it's a good baseline for the 'design problem' discussion. 

Thanks to all the anarchists, especially this thread!

 

Mozzy...this makes so much of the analysis accessible...thank you!!!

 

Share this post


Link to post
Share on other sites
4 minutes ago, Couta said:

Mozzy...this makes so much of the analysis accessible...thank you!!!

 

+10... thank you!!!

Share this post


Link to post
Share on other sites
2 hours ago, Steve Clark said:

Video of ETNZ practice today, January 26, shows them sailing with both foils down.  They seemed to be doing as many maneuvers as they could manage and still stay on foils.  This makes sense as they have just shipped the smallest set of foils seen to date.

I had been wondering about using both foils for trying to get out of the water. 

SHc

Yes, we've seen them doing this quite some weeks ago too. Taking off and staying on foils must be quite a challenge for all teams in the light stuff.

Haven't yet seen any video of a team trying to take off with both foils down, I guess at lower boat speeds the drag is too much of an acceleration penalty.

Interestingly, although opportunities for measurement with any accuracy are very limited, I haven't yet found any significant difference in boat speed required to take off, despite substantial differences in foil area. Perhaps parasitic drag evens things out?

Share this post


Link to post
Share on other sites

Personally I think there is some sandbagging going on???

Share this post


Link to post
Share on other sites
6 hours ago, MaxHugen said:

Look forward to seeing your sim results.  Been dying to see some airflow sims especially over the hulls, hull/sail intersection, etc. (there's a long list of wishes actually)  ;)

Pls. present your list and I will see what I can do. I realize I posted this in the wrong thread, it should have been in the twin-skin setup. Anybody knows how it can be transferred, or maybe I will just continue posting there when I get some results. 

Share this post


Link to post
Share on other sites
6 minutes ago, Mikko Brummer said:

Pls. present your list and I will see what I can do. I realize I posted this in the wrong thread, it should have been in the twin-skin setup. Anybody knows how it can be transferred, or maybe I will just continue posting there when I get some results. 

No probs, see you on the other side topic. :)

Share this post


Link to post
Share on other sites
8 hours ago, erdb said:

Certainly both are interesting. The problem is that if a boat has a better performance/VMG to begin with, they'll look good on the VMG to course, too, even if they aren't following the best tactics.  Something about how a fast boat makes the tactician seem like a genius.

That's why I came up with that method of analyzing tactical choices in degrees x seconds. Basically, for every second a boat sails on the favored tack, the angle of the shift in degrees is added to their "on favored tack" total. For every second they are on the wrong tack, the same number is deducted. This gives a number that's more or less isolated from boat performance. I've never seen anything like this used before, so I'm sure there are a lot of issues with it, but so far, it seemed to mirror what was happening in the race and matched the commentary from NO and KR.

I think it is a good idea but may not be the whole story for these boats. With displacement sailing boats  (at least upwind) they quickly get to hull speed and any more wind doesn't make a great deal of difference. With these boats it does, so they may go a lot quicker. So a smart tactician might (and I believe) has give up a lift or header for more breeze. Especially in the light stuff when it can make the difference between foiling or not of course

Share this post


Link to post
Share on other sites

More wind doesn't make the boats go much faster, but it does affect the angles they sail

Share this post


Link to post
Share on other sites
On 1/17/2021 at 2:52 AM, erdb said:

I'm not 100%, sure but I have a feeling that the instruments providing data for virtual eye and broadcast are supplied equipment potentially running the same software with identical settings. If that's true, then the differences we see in TWS numbers may be caused by differences in sail design, size and trim. That would be an interesting challenge to calculate back from errors in TWS / TWA traces.

Good illustration with 2 boats as close as it can be, slight difference in TWD but big one on TWS. Notice that delta is usually consistent, one boat is measuring more than the other most of the time and that seldom revert.

TWS.JPG.a7e918e8d97734afd8e5953f63669b93.JPG

Share this post


Link to post
Share on other sites
On 1/13/2021 at 2:00 AM, erdb said:

I'm still going through

257831033_uwheel.png.8d85db92a5fad945e431d55ad2832e5b.png  254502351_uwpitch.png.026c11450e4e3cc5cfbb364710abc023.png

Downwind:

1697625613_dwheel.png.d8c2f2b9045657a3df39d0af7bb0f3bb.png  11468258_dwpitch.png.a4b0b53df23d7eb51a0a835b1581329d.png

AM heels more to windward and pitches more forward both upwind and downwind.

Up- / downwind foil cant angles:

788743555_uwfcant.png.a59fe4ab23525bb4619cd3a33f0ee30f.png   62521715_dwfcant.png.0f1d7f0992c9b95a2f3c067023ffbd35.png

 

Have you produced these upwind and downwind pitch histogram plots for GBR and ETNZ?

Also, would be interesting to plot pitch V speed for each team

Share this post


Link to post
Share on other sites

 A little late asking this question but can anyone explain the thought process behind Patriot being so bow heavy??   How does Patriot get a speed advantage by heavy a heavy bow but light stern???? I am not understanding the major keel kink.

Share this post


Link to post
Share on other sites
16 hours ago, enigmatically2 said:

I think it is a good idea but may not be the whole story for these boats. With displacement sailing boats  (at least upwind) they quickly get to hull speed and any more wind doesn't make a great deal of difference. With these boats it does, so they may go a lot quicker. So a smart tactician might (and I believe) has give up a lift or header for more breeze. Especially in the light stuff when it can make the difference between foiling or not of course

 

15 hours ago, kiwin said:

More wind doesn't make the boats go much faster, but it does affect the angles they sail

All good points, but we can only play with data we have. The TWS data doesn't seem reliable, the difference between boats just seem too big. If it was accurate, we could compare which boat sailed in more pressure in general. On the other hand, if a boat sails in better pressure on average than the other, it may show up as having better average TWA or VMG. So yeah, it's possible that those differences between boats are partially due to putting the boat in better wind, and not the boats' performance. We would need a lot more data to know for sure.

Share this post


Link to post
Share on other sites
7 hours ago, Mozzy Sails said:

Have you produced these upwind and downwind pitch histogram plots for GBR and ETNZ?

Also, would be interesting to plot pitch V speed for each team

Yes, I have some of the ACWS races as well. This is race 8 ETNZ vs INEOS. TWS was around 11-12 kts.

615793510_uwpitch.png.e506ecc583b44b8d695394e9533f49f5.png  1795997828_uwheel.png.650f09c332435e422182d07a77189cfb.png1959303851_uwcant.png.03e371b9ecad6f3587d1887d075a14c8.png

 

253436765_dwpitch.png.21a6d14274757f8e6191da352778c74f.png  419088567_dwheel.png.d83b7c4c211c2b3c063895f1ab3e7a9a.png

220833531_dwcant.png.a5664ef89f645336522046c7cf470150.png

Share this post


Link to post
Share on other sites
34 minutes ago, dorox said:

Update: now all races are available at https://ac36.herokuapp.com/map

Thanks to @MaxHugen for boat images. If you zoom in, you should see schematic AC75s drawn to scale.

Looks great! Although I'm a bit disappointed that the sails are not trimmed according to AWA and AWS. :P

So this was THE cross:

cross-dx.thumb.png.f7855661702920cbfdb9a99a4b4234b4.png

May have been a little less than a boat length based on this, but it also depends on where you put the GPS location reference point on the boat. 

Share this post


Link to post
Share on other sites
1 hour ago, erdb said:

Looks great! Although I'm a bit disappointed that the sails are not trimmed according to AWA and AWS. :P

So this was THE cross:

cross-dx.thumb.png.f7855661702920cbfdb9a99a4b4234b4.png

May have been a little less than a boat length based on this, but it also depends on where you put the GPS location reference point on the boat. 

I'll have to send @dorox imgs with the sails on the other tack... might be an issue trying to swap between the images though? :)

Share this post


Link to post
Share on other sites
3 hours ago, erdb said:

Looks great! Although I'm a bit disappointed that the sails are not trimmed according to AWA and AWS. :P

So this was THE cross:

cross-dx.thumb.png.f7855661702920cbfdb9a99a4b4234b4.png

May have been a little less than a boat length based on this, but it also depends on where you put the GPS location reference point on the boat. 

I understand that they didn't actually make contact, and were not overlapped. But this is not the cross. This is after LR turned, before the closest approach, and before the cross. At the closest approach, they were well within one boat length.

Share this post


Link to post
Share on other sites
24 minutes ago, nroose said:

I understand that they didn't actually make contact, and were not overlapped. But this is not the cross. This is after LR turned, before the closest approach, and before the cross. At the closest approach, they were well within one boat length.

On the Dorox app, this was the closest between the two. Check out yourself if you can make them any closer. https://ac36.herokuapp.com/map

My GPS based minimum distance calculation was also ~30 m. LR didn't turn to avoid. They turned as a last resort trying to get a penalty on INEOS. It almost worked, they were just 1 sec late.

Share this post


Link to post
Share on other sites

 

Share this post


Link to post
Share on other sites
5 hours ago, erdb said:

Yes, I have some of the ACWS races as well. This is race 8 ETNZ vs INEOS. TWS was around 11-12 kts.

Thanks you, that dispels one theory and reinforces another

Share this post


Link to post
Share on other sites

@dorox, your data is great - thanks a bunch for compiling it. Would it be possible to add a CMG plot by integrating the VMG curves?  And if I’m able to add one more request, is it possible to colour key the map plots so that you can identify speed along the track e.g if Ineos is blue then where the line is pale blue speed is <20kts, dark blue >40kts and shades in between.  

  • Like 1

Share this post


Link to post
Share on other sites

CMG is quite simple to add, just one question: should it be derived from wind VMG, or course VMG, or simply the distance sailed from GPS?

Re colour gradients: this is doable, but in a form of scatter (colour-coded dots along the boat track)

1 hour ago, sosoomii said:

@dorox, your data is great - thanks a bunch for compiling it. Would it be possible to add a CMG plot by integrating the VMG curves?  And if I’m able to add one more request, is it possible to colour key the map plots so that you can identify speed along the track e.g if Ineos is blue then where the line is pale blue speed is <20kts, dark blue >40kts and shades in between.  

 

Share this post


Link to post
Share on other sites
2 hours ago, dorox said:

CMG is quite simple to add, just one question: should it be derived from wind VMG, or course VMG, or simply the distance sailed from GPS?

Re colour gradients: this is doable, but in a form of scatter (colour-coded dots along the boat track)

 

This app is amazing! I'm really spending too much time trying to analyse how they compare to each other! :D That 36th America's Cup presented by PRADA dashboard website is also pretty amazing - I'll try to follow it along a live a race next time

I also have some suggestions to make if that's alright... 

In the map app, would it be possible to add a buttons next to time stamp to be able to step through data points slowly?
image.png.be78b9d2c780619bb276fade65ac0c8c.png

 

The other questions is whether it would be possible to display values for the current data time point over the chart below the map?

image.png.1d7eb972b7fd6e1a66a0e23cb3f7d7ea.png

 

Thank you again for putting this together - it's almost information overload :D

  • Like 2

Share this post


Link to post
Share on other sites
5 minutes ago, Enzedel92 said:

So after all of this who has the fastest boat?

Have to say NZ

(otherwise some posters will have a major tantrum)

  • Like 1

Share this post


Link to post
Share on other sites