[mpugsley]

Originally Posted by jasonb885

Sigh.

I don't care how realistic a bug is.

If something doesn't work the way it was obviously _intended_ to work, the game is bugged.

Fair enough. That seems a reasonable usuage of the term "bug". But even so, and assuming that the game isn't working as intended, it only follows that the game is bugged. It doesn't follow that every bug (this bug in particular) is bad and needs to be fixed. I'm not saying that you were saying that the bug was bad. Your point may have just been to insist that we call a bug a bug. In which case, I completely agree. I just wanted to argue that Beery is at least this much right: we can treat the bug like a feature. If the developers have unintentionally made the game more realistic than it would have been had their intentions been realized, we should celebrate the event, not mourn it.

Originally Posted by Ariel/User 1834

Edit - Also, it seems unlikely to me that the U-boat crews would allow their battery to discharge completly. Not only would this badly damage their battery but it would also leave them defensless for a considerable amount of time. I say again, your test is inconclusive.

This is the best argument I've seen yet for the claim that Beery's test is a good one. (At least that it sufficiently approximatese a good one to be a good one.) As I understand it, the argument in the thread (recently) has been whether it should take 3-5 hours or 13 hours (as Beery found) to fully recharge the battery. (For the moment, let's count 98% as full. It's close enough. There is still a huge difference between whether the 98% [or so] should take 3-5 hours or 13 hours.)

Beery says 13 hours is reasonable. Lots of other people are unhappy because this seems too long. Data is cited to support the claim that the entire battery should be recharged within 5 hours (i.e. 0% to either 98% or 100% in 5 hours).

What Beery's test shows is that a 5 hour recharge (from 2% to 55%) allows the XXI to run at 25m and at 5 knots for 72 hours. (I am curious if it really stayed at 5 knots the whole time or whether the battery ran out after 72 hours, in which case, the sub would be moving slower than 5 knots during the final hours.) This is consistent with the idea that the sub must recharge every 72 hours (for 5 hours) before being able to repeat the routine.

The claim that no sub crew would ever fully discharge the battery strongly supports Beery's conclusion because the 5 hour recharge would be applied well before the battery ran out. Thus, the 5 hour recharge (even if it were to a full 100%) wouldn't be a recharge of the entire battery. Thus a 13 hour recharge of the whole battery (or at least, 98% of it) is quite plausible.

What matters is this: a 5 hour recharge gives you that portion of the battery which is used during the 2-3 days of submerged travel (at 4-8 knots). That portion of the battery would be less than the entire battery since no crew would fully discharge the battery. Thus it should certainly take longer than 5 hours to recharge the entire battery. How much longer depends on how much of the battery is used in 2-3 days of travel (which is regenerated in 5 hours).

I think a more interesting version of Beery's test wouldn't have the boat operate from 0% to a percentage reached after 5 hours (which has turned out to be 55%). Rather, we should have the boat start at 98% (which is the game's max after a little use), and then go for 2-3 days. Stop there. Assume that that is where the U-boot crew would say "The battery has had enough. We could push it further, but we would damage it. It is time to recharge." Then recharge back to 98%. See if it takes 5 hours. If so, then we have a fairly reasonable model of the operational data provided to us by Uboat.net.

In other words, test the 'top' half of the battery, which is the part that would be used. The results should be about the same.

Of course even if the test verifies the data from Uboat.net, the test won't be completely conclusive. There are multiple ways of achieving the same operational results.

A)The XXI might use the top 60% of the battery in 2-3 days of submerged travel at 4-8 knots, leaving the battery at 40%. Further assume that it is important to take the battery no lower than 40%. This would explain why the crew chooses to charge the battery after 2-3 days. It isn't empty, but it's at 40%, which is the minimum safe charge (as far as not damaging the battery goes). Let us further assume that a 5 hour charge is sufficient to get the crew back to where they started, namely at 100% so that they can repeat this routine (i.e. travel submerged at 4-8 knots for 2-3 days and recharge to their original charge after 5 hours). The conclusion would be that the 5 hour charge is sufficient for charging the top 60% of the battery.

B)The XXI might use the top 70% of the battery in 2-3 days of submerged travel at 4-8 knots, leaving the battery at 30%. Further assume that it is important to take the battery no lower than 30%. This would explain why the crew chooses to charge the battery after 2-3 days. It isn't empty, but it's at 30%, which is the minimum safe charge (as far as not damaging the battery goes). Let us further assume that a 5 hour charge is sufficient to get the crew back to where they started, namely at 100% so that they can repeat this routine (i.e. travel submerged at 4-8 knots for 2-3 days and recharge to their original charge after 5 hours). The conclusion would be that the 5 hour charge is sufficient for charging the top 70% of the battery.

In either case, the 5 hour charge is *not* of the whole battery, precisely for the reason stated by Ariel, namely that no U-boot crew would *try* to charge the whole battery. The crew would never entirely discharge the entire battery. The 5 hours is sufficient for completely recharging the operational portion of the battery. So in a sense, the 5 hour recharge is complete, but it is only the complete recharge of that portion of the battery which is safe to use (without damaging the battery).

The point of my examples A and B is to show that we don't yet know exactly how much of the battery would be used by the XXI in 2-3 days of sumberged travel at 4-8 knots.

Conclusions:
If we put our faith in this one line from Uboat.net, then we can be confidant that the XXI is (roughly of course) operationally realistic. We cannot be sure that the battery needle accurately represents what the battery would be like after x hours of travel/recharge. We don't know how far it is to go down after 2-3 days of submerged travel at 4-8 knots, and so, we don't know how much it's supposed to go up after 5 hours of recharge. All we know is that it goes down to some degree (presumably not all the way down), and then, after 5 hours of recharge, it goes all the way back up.

How much does this further worry about accuracy matter? So long as we're in the territory of 'reasonable' numbers and the operational results are (roughly) accurate, then shouldn't we be happy? I'd say so, although I concede that representational accuracy does matter some.

Our situation is like the man in Plato's dialogue "The Theatetus". (I think it was "The Theatetus". I'm too lazy to look it up.) Anyways, there's a man who wants to go to the city of Larissa. He believes (falsely) that Larissa is the north. It isn't. It's to the south. But he's also confused about which way is north. He thinks north is south and south, north. So he ends up travelling south, thinking that (a) he's headed north and (b) Larissa is to the north. He ends up actually getting to Larissa with these two false beliefs. His beliefs are false. And it would be 'better' if he had true beliefs. But he did, after all, get to Larissa. So is it really such a big deal that he had false beliefs?

Similarly, it would be nice to have our battery work perfectly--max charge should decrease with not only the first recharge, but presumably every recharge (and in a realistic way), and recharge and discharge rates should be perfectly accurate (which would include variable rates, depending on the current state of charge). Our battery surely isn't perfect. (Hopefully it's closer to being correct than the beliefs of the man who was going to Larissa.) But it does (at least it seems to me) get the operational data correct. It's not at all clear to me that we need more than that (or better, it's not clear to me that it's worth much work to get the 'perfectly accurate' battery representation which successfully models the operational data, when the present model (although not perfectly accurate) seems to do a fairly good job of successfully modelling the operational data.

To compare this issue with others and perhaps bring a bit of acceptance to the 'bug', 'feature', whatever: Why are we so content with the other Uboots? Even if they are operationally reasonable, they may be completely wrong. They might 'work' only because the discharge and recharge rates are *both* wrong (like the beliefs of the man travelling to Larissa). Additionally, we're allowed to go near 0% without damaging our batteries. Isn't that extremely unrealistic? I haven't seen lots of angst about those issues, but worry seems to be equally well warranted about them (if not more warranted).