The new version is a huge improvement! I am loving the less-punishing windward performance in light airs, as well as the visual fix to t'gallants set over reefed topsails. The ship drifting without any canvas set and the ability to scud under bare poles is also a huge deal. Here are some suggestions for tweaks in case you ever get back to tinkering with the game:

* The 'foul wind' disabling Travel Acceleration is often more trouble than its worth, since it prevents players from waiting out a contrary wind easily. You can also get your nose 'stuck' in the wind too easily this way. What if Travel Acceleration only turned out due to foul wind when the auto-adjusting sails feature was turned on? That would give players the option of using acceleration regardless of wind direction.

* The ship should heel more in bad weather under bare poles, in my opinion. Perhaps a hacky solution to add virtual canvas to induce heel when no other sails are set?

* When the ship is wallowing under bare poles, it should eventually orient itself beam-on to the wind. In the game the bow falls off. Is there any way to adjust this?

* Another potential hacky improvement would be bumping up the ship's leeway when no topsails are set. Right now we can sail to windward at 2 kts like a schooner by furling the square sails and using fore-and-aft canvas alone, which isn't realistic.

* The auto sail adjustments are a little aggressive in higher winds for my taste. What if the adjustments were more conservative than strictly necessary, giving the manual control an edge on performance?

* Your fore course blows away in conditions where the game says you should have it set.

* IMHO the main t'gallant staysail and flying jib should be doused before 20 kts of wind. The headsails should also need to be reefed much earlier. The auto adjustments have you flying a full size jib in a near-gale, which is excessive. Adding some randomness to topgallant damage would be nice as well, since I feel a bit too bulletproof flying them at 28 kts.

* Just a question, does apparent wind matter for sail damage?

Thanks for the feedback!

On the first point, there is a feature which mitigates this a little bit, which is that when you are in travel acceleration, you can rotate instantly by clicking on the compass. When not in travel acceleration, you can do it by control+clicking.

More heeling under bare poles - I'm worried that this would be too much, without the ability to strike the topgallantmasts. For the orientation under bare poles, I don't know. I feel this should depend on how far forward the bare poles are -- I think right now there is no wind drag modelled on the hull, which is maybe also relevant.

Sailing to windward - what do you mean by windward? It's not clear to my why you couldn't in fact sail a ship like a schooner. I suppose the likeliest things are drag from the unused spars, plus staysails not being as efficient (or as big?) as a proper fore-and-aft rig, so these would be the things to model.

I agree the damage stuff is currently not entirely logical. Apparent wind does matter for the sails.

That's right, in reality a ship can't make useful headway with just the fore and aft canvas, at least not upwind in ordinary conditions. The ratio of sail area to windage (rig and hull) becomes so small that the leeway angle will eat up all the gains.

Actually in the older version, the behavior was just about perfect. If you tried to sail upwind into the 'no go zone' the ship would naturally fall off again.

Did you make a wide variety of changes to lift settings in this version? Or just change some of the existing variables in the in-game Physics menu as a default? I've half a mind to do some experimenting on my own and try to find a happy medium (because the old version truly was punishing with regards to upwind performance).

I looked at this and you're quite right. In a fresh breeze, under all fore-and-aft sails, I can do 3 and a bit knots four points from the wind, which seems way too good. I will see if I can make some time to mess with it over the holiday. I can try some combination of making the fore-and-aft sails less efficient (essentially making them smaller) and increasing the drag on the masts and rigging.

It was a long time ago, but I believe most of the tweaks to aerodynamics are things that are exposed. In the physics menu on the left, wind lift coeff. and drag coeff. should have straight linear effects on the lift and drag forces. Lift is zero at angle 0 (the wind is parallel to the yard), reaches a maximum at some angle (currently 30 degrees), and falls to 0 when it is perpendicular to the yard. I think it's modelled just as a sin curve, except you can use the max lift angle slider to shift where the peak is.  Lastly "griping" on the right isn't aerodynamics, but controls how far forward the lift force on the hull acts. If you put it higher, the ship gets less stable around the yaw axis, meaning that tacking is easier, but on the other hand balancing the sails is more important; and vice versa.

Edit - I'm pretty sure the aerodynamic sliders don't do anything to the drag on the masts etc.

I have been looking at this. If you want to mess with it, I have uploaded a file painted_TEST_VERSION which has two extra physics sliders. I would be very happy to hear what you find is most realistic.

One is a multiplier for the wind force on the masts and rigging etc. I have found a couple of (not great) sources describing ships making 11 knots under bare poles in a hurricane. The 0.5a build would already do this. The default in the test build is to increase this drag by a factor of 1.5;  you could partially justify this by saying that the 11 knots is presumably with no topgallant masts (or topmasts). Like this, the force on the bare poles is about equivalent to half the force on maintopsail.

The other is a multiplier for the lift force on all the fore-and-aft sails. It doesn't affect the drag. I have set this to 0.6.

These two seem to make it much more difficult to make any progress to windward under only fore and aft sails, while keeping the staysails still a little bit useful, and not changing the usual performance too much. But I haven't tested tacking etc. properly, and it's possible that there are still some of the effects I noticed while tweaking it, that when sailing with the wind on the beam you can go faster by taking in most sails behind the mainmast. Maybe this is realistic, or maybe it needs some tweaking of the hydrodynamics.

So I just logged in to see the test version! But haven't tried it yet. I did come here to report my experiences with tweaked settings in the September 2019 version (release) version that I still have installed on  my old PC.

As you can see I am sailing close-hauled in 15 kts, all sail set. The Global Physics values for everything but skin drag have been adjusted upwards, but less aggressively than in the current release build. As a result the ship can make decent progress on a bowline, tacks in 3.5-.4.5 minutes. But most importantly if you try to sail too close to the wind, the leeway angle increases rapidly and eats up the gains. 

The problem with being able to sail like a schooner with square sails furled DOES appear when the settings are tweaked like this. 'Schooner sailing' was impossible in the September 2019 version, due primarily to the lack of leeway with that sailplan.

I of course realize that I should be doing these experiments on the new build. I'll try that next.

OK, here is a better benchmark. By decreasing the Wind Lift Coeff and Water Lift Coeff, the ship can no longer sail too close to the wind with all canvas set. The screenshot shows how leeway will punish you if you try. 

However, tacking is still a bit too torturous, at over 4 minutes. This time isn't necessarily excessive, but you barely have enough inertia to get head to wind, and would probably fail in light airs. What settings should I adjust to make the ship keep forward momentum for longer?

I will try the new test build later.

To reduce speed loss, you can limit one or another kind of drag.

Good tests are to get the right speed first right before the wind, and then with the wind on the beam. Before the wind, you are only dealing with being pushed by aerodynamic drag and slowed down by hydrodynamic skin drag, so it's only two things to balance with each other. With the wind on the beam, you are only being pushed by aerodynamic lift, but now you will be seeing some form drag.  (The "form drag" should come in to play when you have some angle of attack with the water.)

The water lift coefficient controls how big a side force you get from an angle of attack with the water, and the "griping" slider controls how far forward is the point where this lift acts on the hull.

Your new build is already an excellent result! In the screenshot below you can see how the leeway punishes you now when you try to pinch too close. This is the behavior I was reaching for in my earlier tests.

Better yet, when you try to sail with just F&A sails, the ship falls off. Ideal behavior there. I haven't tested tacking yet, so that will have to come in a later post.

I can also report much better behavior with regards to heeling under bare poles. Furthermore the ship seems to oscillate between close-hauled and a broach reach when wallowing with no canvas set. This seems more or less realistic.

However, tacking is probably a bit too hard right now. It took me about 4 minutes in 17 kts of breeze (10 kts is an important testing benchmark which I haven't tried yet). The ship loses all headway by the time it is in the eye of the wind, and loses a bit of ground overall throughout the evolution.

Is there some way to make this easier without sacrificing the progress made with the other scenarios? I would also say that ships take a long long time to coast to a halt (no canvas set) in this game (such as when you are roleplaying anchoring). So there probably isn't much scope for increasing inertia.

This is what I see tacking, in the test build in the world scenario (17kt wind):

I start with the wind 1pt free, going at 7kts. I order the helm hard over, and ease the headsails and take in [edit: haul in] the spanker once I am clearly turning. I am pointing at the wind after 60s, and still going at 1 or 2 knots.

If I start 1/2 a point closer to the wind, going at 5kts, then this takes 120s, and for the last minute I am nearly motionless or moving backwards, being pushed around just by the foresails.

Maybe this is as it should be - to tack snappily, you need to build up some speed. I haven't tried it in a lighter wind.

I also observe the following. Start the world scenario, set the waves to 0 and bring the wind right on the beam, with physics time acceleration on high. You will reach 9.7kts. Take in all sail abaft the mainmast, and you slow down to 9.3kts.

So far so good. But now do the same thing, starting one point closer to the wind. You will reach 7.1kts. Take in all sail abaft the mainmast, and you speed up to 7.3kts.

What is happening is that with the sails set in the second case you need one spoke of weather helm to balance them, and the drag caused by the rudder slows you down. With the sails taken in, you have a tiny bit of lee helm, and essentially no drag from the rudder. This is deliberate behaviour, sort of, because I wanted drag on the rudder to be a consideration in your sail plan. Possibly it is too strong right now - but I'm not sure how to tell. Reducing drag from the rudder would also make tacking a little bit easier.

Edit: Substantially decreasing the drag from the rudder doesn't change this, so my explanation above is wrong. It may be something to do with the extra leeway, with more sails set, causing more drag; but that doesn't work by itself, because removing sails forward would also decrease the leeway. It should be the torque from the sails that is the problem.

Testing light winds performance (10 kts):

On a broad reach the ship is really fast, around 6.5 kts. I would note that the weather-related UI predicts only 3-4 kts, which isn't for this ship or anything. But it is still a big difference.

When tacking, I started the evolution at 4 kts. It took me 5 minutes to get into the eye of the wind, then I got bored of waiting and pressed the time acceleration button. The ship eventually tacked but made plenty of sternway. Maaaaybe realistic but also pretty brutal for gameplay.

I think we have a funny situation where speed and windward performance are still way better than the release build, but tacking is still difficult. 

Maybe it would also be worth increasing the rotational pressures of sails that are flat aback? Falling off (after mainsail haul) is a bit slow I think.

In the new build I have tested tacking at 25 kts, 17 kts, 10 kts. Resulting times are around 2 min, 2.5 min and 5+ min. This feels pretty plausible and good for gameplay. Tacking in 7 kts of wind is barely possible, and takes a long time but I didn't time it. Tacking in 5 kts does not seem doable.

Overall, seems very well-balanced at the moment.