Reduced Tonnage RUB/SH 3 Commander

[Enfilade]

This isn't a criticism, just a question. I have no problem with Rub reducing the stock tonnage of the C2/C3 from the stock game's 6-7k to 4+k if this is the realistic tonnage variation. Presently running Rub 1.43 (no chance to install 1.44 yet as still on patrol). But noticed the C2 and the C3 have just about the same tonnage values now. And they're pretty big suckers - especially up close! Wondering what source you're using for these dramatically reduced tonnages? Gotta sink a lot of C3's to make much of a splash..no pun intended....

[Kobal2]

There may also be another side effect - the renown gained for a patrol depends on the tonnage sunk. Lower the tonnage, you also lower the renown. Or maybe Beery adjusted the renown values accordingly ?

[Wolfram]

I like the tonnage adjustments

Large Mechants were ~17,000 at one time now they down to ~12,000 using the RUb (with IUB ovelay on it)

:lurk:

[Enfilade]

It's not that I dislike the tonnage adjustments... I was just curious to know what information they were based on. Tonnage variations may also be a result of changes made in SH3 Commander and I believe these will vary according to year. I'm an enthusiastic user of both RuB and SH3 Commander... they make a good game great.

[BladeHeart]

I was wondering the same.

If the tonnage is reduced and as a result the renown is reduced then surely the promotion levels, etc should also be adjusted.
Unless of course the renown remained the same. :hmm:

Maybe Beery could give his views on the matter? :up:

:lurk:

[Wolfram]

It's not that I dislike the tonnage adjustments... I was just curious to know what information they were based on. Tonnage variations may also be a result of changes made in SH3 Commander and I believe these will vary according to year. I'm an enthusiastic user of both RuB and SH3 Commander... they make a good game great.

Good points :know: maybe Beery will catch this thread and answer but I meant I liked the fact the ships (tonnes) seem more realistic. Large Merchants always mystified me why they appeared bow to stern shorter and lesser keel draft than a light Cruiser but had about 50 to 60% more tonnage...

have Fun!

:lurk:

[Observer]

This isn't a criticism, just a question. I have no problem with Rub reducing the stock tonnage of the C2/C3 from the stock game's 6-7k to 4+k if this is the realistic tonnage variation. Presently running Rub 1.43 (no chance to install 1.44 yet as still on patrol). But noticed the C2 and the C3 have just about the same tonnage values now. And they're pretty big suckers - especially up close! Wondering what source you're using for these dramatically reduced tonnages? Gotta sink a lot of C3's to make much of a splash..no pun intended....

You get this mod with SH3 Commander. I suppose since I did it, I'll explain how it works.

The data for this is from over 3000 actual ship sinkings that happened during the course of the war. This data was broken up by year (mostly..some years are combined), and then analyzed using some statistical tools for mean and standard deviation. I generically placed ships in classes such as Large Merchant, Medium Merchant, Small Merchant and Tramp Steamer (by the way I did a mod to get rid of the C2, C3, T2 and T3 names which applied to specific classes of merchant ships and saw limited action only during the later part of the war, at least from a actual ship sunk perspective). I then randomized the tonnage values based on the mean and standard deviation. If the standard deviation was large for a given class of ships it could result in tonnages being very close to each other (i.e. C3 and C2 which I referred to as Large and Medium merchants). This isn't unrealistic because many Uboat Captain's had difficulty estimating ship tonnages unless they could find the ship in the Lloyd's registry.

Just as an aside, the same database used for tonnages was also used as data for the improved convoy mod.

Hope that's the answer you're looking for. This varies in SH3 Commander by month, so some months you may get a 3500 gross ton C2 (medium merchant) and another month you may get a 5500 gross ton medium merchant. It adds to the excitement and means you never can count on a given class of ships getting you an exact 100% predictable tonnage. Every time. Now there are some surprises, and some disappointments especially when you spend 3 torpedoes taking down a merchant of only 3500 tons.

As another aside, the renown values for ships have not been altered. You should find, on average, the results from a patrol more closely match historical sinkings from Uboats during the war on a tonnage basis.

[Observer]

One other note, just for clarity. This database represents ships sunk by Uboats only (including mines). I didn't include ships sunk by aircraft or surface action. The sample size in this case is large enough where I don't think it adversely affected the statistical analysis.

[Enfilade]

Observer, many thanks for the informative post. Just what I was looking for. I like the variations possible depending on the month too.

[Kalach]

I think the reduced & random tonnages are great :up:
Especially when you sink something a lot bigger than you expected it to be :arrgh!:
I have noticed that getting medals for your crew is a lot harder through. The commander will be able to rack-up his medals quickly with some good patrols, but your crew getting medals is very slow.
I'm currently in my 25th patrol in a IID in '43 and have only 2 iron crosses between my whole crew :(

[Beery]

There are two issues here:

Firstly the large merchant and the large transport were both reduced because they had enormous tonnages - well outside of the normal range for WW2 ships. This is a slam dunk - the earlier tonnage values on these were certainly too heavy. I feel they're within acceptable tolerances now.

The more difficult issue is the general tonnage mod reductions that came with Observer's random tonnage mod. I've actually done some research on this today, and I'm beginning to think that the mod may be a bit too conservative. When you calculate the random tonnage mod's tonnages and extrapolate them to careers they turn out to be very light (i.e. the ratio of of ships sunk to tonnage doesn't match up well with historical figures for the famous aces, whereas it should fall within a reasonable range). As a double check, I also took my longest career which used the standard SH3 tonnage figures and compared it to a similar-length and similar-tonnage historical career, and the number of ships sunk in both careers were almost identical, although the particular historical career that best matched up with mine (Schepke's) was very light in terms of tonnage per ship sunk. 18 out of the top 25 commanders got between 20 and 27 ships sunk, with tonnage of between 115,000 and 180,000 tons, for an average sunk ship size of 6276. My long career has an average sunk ship size of 4158, and that's with the old higher ship sizes - if that career had used the random tonnage mod it would have been well out of the range that we would have expected for a real career. Schepke has an average of 4213 (of the top aces, his is probably the smallest tonnage per ship sunk). The largest tonnage per ship sunk (of the top aces) seems to be Ernst Kals with an average ship size of 7282, which also argues for bigger (not smaller) ship sizes. I would have to say that the C2 is probably representative of the most sunk ship in the game - it's probably above the average sunk ship size. The standard game's C2 is 6200 tons (just 76 tons lighter than the average historical tonnage for the above listed commanders). The random tonnage mod has the C2 at between 4100 and 5200 tons. I feel a better range for this ship type would be 5200 to 7200 (perhaps even higher). That's just my gut feeling on the issue. If players are using the realistic career length in SH3 Commander, there is virtually no chance of amassing tonnage totals anywhere near the top aces. I feel the possibility has to be there. It must be at least possible to equal Ernst Kals's score at the time he made that score (July 1941 to December 1942), but with SH3 Commander at present, this is only possible if you generally sink larger tankers or liners. The chances of getting tankers on virtually every mission are very low indeed, and liners are very hard to find, let alone sink.

I think perhaps where the discrepancy comes in may be in using an average for all ships sunk by U-boats. The less successful commanders probably sank smaller ships on average, and this may be skewing the averages a bit. The game tends to simulate the very successful careers better than it does unsuccessful ones. Another thing that plays into it is the fact that we tend to be able to sink more ships than real commanders could. Our torpedoes are far less prone to failure. This results in a larger number of ships being sunk. Observer solved this problem by lowering tonnages, but there's a side effect, in that real commanders got less ships but more tonnage than we get when using the random tonnage mod.

I'm not sure what's to be done about this. Certainly the ratio of ships to tonnage has to fit with the historical tonnage to ship ratio. I'll have to think on this a bit.

[CCIP]

Hmm, looking over my WaW career, I've sunk 43 ships for just over 250,000t now. That puts the average exactly where it's ought to be as you say Beery, and I've been using Observer's mod all along.

Luck, perhaps? :hmm:

[Observer]

I'm afraid I have to disagree Beery.

It is an error to look at just the best performing aces as a method to evaluate the correctness of the tonnage mod. From a statistical standpoint, you have to look at the aggregate of the data in order to get a representative sample. In this case, using Uboat.net for a data source, I can see the aces who sunk over 100k tons of allied shipping had on average a tonnage of 5,527 gross weight tons. These were the best performing Uboat Captains.

Also sucessful were the Captains sinking over 50k tons of allied shipping. These men had an average of 4991 gross weight tons. These men, on average were very sucessful. Of course this completely disregards every Uboat Captain sinking less than 50k tons, but there were many who only completed a few patrols before being lost, and some only sank 1 or 2 ships. I don't have the statistics at hand or the time to compile them, so I can only assert this without data to back it up. I think a spot check of uboat.net will reveal the truth to this statement though.

By comparison, the C2 merchant, which makes up a fair amount of pre-1942 convoys has an average tonnage, in the tonnage mod from 1939 - 1942, of 4,516 gross weight tons. By comparison, the C3 merchant has an average tonnage, in the tonnage mod from 1939 - 1942, of 6,618 gross weight tons. If you consider in IC (Improved Convoys) that you will see about 24 C2 and C3 merchants, on average, this equates to about 6 C3 merchants and 18 C2 merchants. Using the above average tonnage values from 1939 - 1942 this gives an average convoy tonnage of 5,042 gross weight tons. This is to the lower end (near the 50k aces), but still only 500 tons off of the best performing aces.

Further, the best performing aces should be, from a statistical standpoint, rare. These aces are at the extreme of the bell curve (you could argue that the 50k aces are near the extreme as well) with the majority of the Uboat Captain's never performing anywhere near this level. From a statistical standpoint, this is comparing the performance of the P95 Uboat Captians and wondering why everyone didn't performe this well. Clearly these men were very special and acheived something special in the process - performance that should be difficult to match, not routine.

Finally, this discounts the presence of Liberty and Victory ships (more so Liberty ships than Victory). These ships had very little variation in tonnage, and because of it, they are not altered in the random tonnage mod. These ships are heavier than those in the random tonnage mod, and would be cherry picked by SH3 skippers if given the opportunity. These merchants will also tend to move tonnage averages up rather than down. Of course this depends on how many patrols a person runs and the time in the war they start the career.

In conclusion, I stick by the values as represented in the random tonnage mod. While not perfect, they are close enough that I don't think many would be able to notice a difference of a few hunderd tons either way. Similarly to CCIP, my WaW career has sunk 46 ships for almost 270k tons. This averages about 5,800 gross weight tons, and is consistent with Uboat aces scoring over 100k tons.

[panthercules]

For what it's worth (not much I'm sure), if you're looking for data from game careers my first/best career had my kaleun sinking 85 ships for a total of 437,782 tons over 17 patrols (Sep '39 - Jan '42), for an average of 5,150 tons/sinking - I was playing stock at first and then various RUB versions up to about 1.42 - never used any other mods that appeared to affect ship tonnages, and didn't do anything like harbor hunting or tanker farming or whatever to get any really big ships.

In my second, rather less successful career, I sank 9 ships totalling 57,930 tons, over 6 patrols, for an average of 6,436 tons/sinking - using RUB 1.42-1.43 (first 4 patrols were with a Type IIA, so I was only shooting at the larger targets first and didn't have many fish left or a deck gun for shooting the smaller targets, which probably skewed the average to the high side a bit).

I like the variability of the tonnages to some degree - it was pretty weird that first career seeing how many ships I sank that were all within like 2 or 3 tons of each other - certainly didn't seem very realistic.

[Observer]

For what it's worth (not much I'm sure), if you're looking for data from game careers my first/best career had my kaleun sinking 85 ships for a total of 437,782 tons over 17 patrols (Sep '39 - Jan '42), for an average of 5,150 tons/sinking - I was playing stock at first and then various RUB versions up to about 1.42 - never used any other mods that appeared to affect ship tonnages, and didn't do anything like harbor hunting or tanker farming or whatever to get any really big ships.

In my second, rather less successful career, I sank 9 ships totalling 57,930 tons, over 6 patrols, for an average of 6,436 tons/sinking - using RUB 1.42-1.43 (first 4 patrols were with a Type IIA, so I was only shooting at the larger targets first and didn't have many fish left or a deck gun for shooting the smaller targets, which probably skewed the average to the high side a bit).

I like the variability of the tonnages to some degree - it was pretty weird that first career seeing how many ships I sank that were all within like 2 or 3 tons of each other - certainly didn't seem very realistic.

RUb does not provide random merchant tonnage. This is only provided using SH3 Commander. If you did not use Commander, then your number reflect stock SH3 tonnage values.

The first career looks about right. The second seems high compared to what I posted above. My guess is you sank mostly C2s during that career.

[panthercules]

RUb does not provide random merchant tonnage. This is only provided using SH3 Commander. If you did not use Commander, then your number reflect stock SH3 tonnage values.

The first career looks about right. The second seems high compared to what I posted above. My guess is you sank mostly C2s during that career.

Yeah - either C2s or C3s mostly - with my Type IIA in my second career, I was pretty particular about target selection 'cause I knew I was only going to get a few shots each patrol, whereas with my Type VIIB in my first career I wouldn't hesitate to go after the smaller ships too and could even deck gun a few of those on the way home whereas I didn't have that option with my Type IIA.

I tried SH3 Commander for the first time very briefly a couple of days ago, but for some reason I started getting CTDs again with it, so I stripped everything off and started over again with just RUB1.44. (Something about my machine seems to have problems with some of this stuff and not others - I had a devil of a time with CTDs with RUB 1.41 or 1.42, but not versions before or after, and then again with SH3 Commander for some reason, but not RUB1.44) I'm sorry to hear that the merchant tonnage thing is in there rather than RUB, 'cause it did seem like a pretty cool addition. Is that particular feature maybe available as a standalone mod that I could load in on top of RUB 1.44?

[Beery]

Further, the best performing aces should be, from a statistical standpoint, rare. These aces are at the extreme of the bell curve (you could argue that the 50k aces are near the extreme as well) with the majority of the Uboat Captain's never performing anywhere near this level...

The problem is that while that's true in real life, it's certainly not true in the game. We often get large tonnage scores. It is almost impossible in this game to simulate the low to middle areas of the bell curve. This is not a simple U-boat simulation - if it was we'd generally get very low tonnage scores and many patrols with no successes. This is a U-boat ace simulation.

[CCIP]

This is a U-boat ace simulation.

There's truth to that. I think part of it is the extreme aggressiveness of players. I never thought of myself as a reckless player, but keeping up with my WaW career, where I keep the immersion high with a variety of things (including keeping detailed logs), I keep noticing how from a real-world standpoint - I think a captain of my character could only be called two things: both incredibly lucky and near-suicidally aggressive.

Part of it is that we're just playing a game, I think, and try as we might, even with a lot of the things modded down and even using only realistic tactics - we're really playing with way more confidence and risk than most real skippers would have been capable of :hmm:

I still haven't had a career that lasted longer than a half-dozen patrols and didn't produce an ace captain.

[joea]

Don't forget, all of us have "died" many times in this game and others (like flight sims) but we don't come back as rookies. We still benefit from our mistakes as it were. Beery said it right, this and a lot of games are "ace" simulations, though I did have one carreer with very low tonnage.

[rudewarrior]

I've noticed in this thread there are some interesting discussions on uboat sinking statistics, dud torpedos, training, etc. I have also read several threads on the "torpedo crisis."

One comment I have is that, although we don't die and are able to restart, I also know that the sub commanders had training before they put to sea.

Second, how exactly is a successful sub commander defined? With all the dud torpedos, we are unable to truly assess whether or not commanders were successful at hitting ships with torps, be they dud or not. Also, sometimes I kind of think about my tonnage per torpedo fired ratio, and wonder what it would be for real commmanders. If that could be determined, maybe we could analyze "success" from a different viewpoint.

Not trying to hijack this thread, just trying to stimulate some thought. :D

Comments?

[Observer]

The problem is that while that's true in real life, it's certainly not true in the game. We often get large tonnage scores. It is almost impossible in this game to simulate the low to middle areas of the bell curve. This is not a simple U-boat simulation - if it was we'd generally get very low tonnage scores and many patrols with no successes. This is a U-boat ace simulation.

That's an interesting point. Based on a variety of factors others have mentioned previously, I must agree to a point, though it is not impossible to get a very low tonnage career, it is just more unlikely. In my mind the AI and living conditions are two of the most important considerations and they are impossible to model. Even I know the AI will depart 15 minutes after they've been unable to detect my boat. That leads to some very bad behaviors, but also greatly reduces some of the uncertainty. That would actually be a nice parameter to randomize.:hmm:

Even so, when comparing aces with more than 50k tonnage sunk in a career to the tonnage averages for a "typical convoy" in SH3, it's pretty darn close. Comparison of some average SH3 careers are also pretty close (at least mine and CCIPs) to that of the aces.

The SH3 torpedoes are much more reliable than German torpedoes historically. We certainly don't have to contend with depth keeping issues for instance. When using manual targeting the average number of torpedoes to sink a ship seems about right (no statistics to back this up though...that's a research project by itself). This gets skewed when using the weapons officer for solutions, so I don't know that number of torpedoes per kill is really a better metric.

[Beery]

...Even I know the AI will depart 15 minutes after they've been unable to detect my boat. That leads to some very bad behaviors, but also greatly reduces some of the uncertainty. That would actually be a nice parameter to randomize.:hmm:

Great idea! We could easily do this. What parameters do you envision? Currently, RUb has a 40 minute departure delay, but we could easily randomize it so that it went from 15 to 45. After ships lost contact, what would be a reasonable time delay for them to give up the search? We have to be careful not to have too big a delay because the escorts get too far from the convoy and they don't catch up, which leaves the convoy more open to attack.

Even so, when comparing aces with more than 50k tonnage sunk in a career to the tonnage averages for a "typical convoy" in SH3, it's pretty darn close. Comparison of some average SH3 careers are also pretty close (at least mine and CCIPs) to that of the aces...

I've come around to your position on this. Possibly my test tonnage scores are because I'm being too conservative.

[Observer]

...Even I know the AI will depart 15 minutes after they've been unable to detect my boat. That leads to some very bad behaviors, but also greatly reduces some of the uncertainty. That would actually be a nice parameter to randomize.:hmm:

Great idea! We could easily do this. What parameters do you envision? Currently, RUb has a 40 minute departure delay, but we could easily randomize it so that it went from 15 to 45. After ships lost contact, what would be a reasonable time delay for them to give up the search? We have to be careful not to have too big a delay because the escorts get too far from the convoy and they don't catch up, which leaves the convoy more open to attack.

Is it really 40 min in RUb? That explains quite a few things...

I'd guess the lost contact time parameter should vary from about 30 min as a minimum to 75 or 90 min as a max. Just to mix things up, this could be varied every time the player starts SH3 using SH3 Commander by maybe using 6 different variants in the random folders for SH3 Commander (10 minute increments).

Even so, when comparing aces with more than 50k tonnage sunk in a career to the tonnage averages for a "typical convoy" in SH3, it's pretty darn close. Comparison of some average SH3 careers are also pretty close (at least mine and CCIPs) to that of the aces...

I've come around to your position on this. Possibly my test tonnage scores are because I'm being too conservative.

Maybe we're too aggressive. I don't know.:hmm:

My big gripe is with crew experience. My last patrol sank 28k total tonnage and my crew got 6 experience points. In order for the "retirement" feature to work well in SH3 Commander I'd think the experience needs to be double to triple that or the officers will never get a promotion during a realistic career. The other option is to lower the promotion criteria (maybe 450 instead of 500 for example). I thought I'd found the parameter but it didn't seem to work the way I had expected.

As an alternative I think it might be nice to be able to adjust experience scores, just for the officers really, using the crew manager in SH3 Commander. It's possible to edit the files manually, but it's not easy.

[Beery]

Is it really 40 min in RUb? That explains quite a few things...

I'd guess the lost contact time parameter should vary from about 30 min as a minimum to 75 or 90 min as a max...

The only problem with that is that after 40 or so minutes the escorts lose track of the convoy and never catch up. This allows you to do an 'end-around' and go to town on an unescorted convoy. This is even worse than the original problem of 15 minutes not being enough to get you far enough from the convoy.

Even so, when comparing aces with more than 50k tonnage sunk in a career to the tonnage averages for a "typical convoy" in SH3, it's pretty darn close. Comparison of some average SH3 careers are also pretty close (at least mine and CCIPs) to that of the aces...

I've come around to your position on this. Possibly my test tonnage scores are because I'm being too conservative.

Maybe we're too aggressive. I don't know.:hmm:

Well, in the upcoming SH3 Commander all aspects of the randomization will be fully customizable. So we'll be able to adjust the file to be very conservative or very aggressive.

My big gripe is with crew experience. My last patrol sank 28k total tonnage and my crew got 6 experience points. In order for the "retirement" feature to work well in SH3 Commander I'd think the experience needs to be double to triple that or the officers will never get a promotion during a realistic career. The other option is to lower the promotion criteria (maybe 450 instead of 500 for example). I thought I'd found the parameter but it didn't seem to work the way I had expected.

In reality, promotions came very slowly indeed. Probably the promotion system we use is much more liberal than real life.

As an alternative I think it might be nice to be able to adjust experience scores, just for the officers really, using the crew manager in SH3 Commander. It's possible to edit the files manually, but it's not easy.

Options are always a good thing.

[Observer]

I've no knowledge on the rate of promotions in the KM. My thought process is that if crew members are transferred off of the boat, they should be replaced by one of equivalent rank (or at least close in rank). That's fine. In fact that's great! If you have to spend renown to "buy" a replacement of similar rank then it's a problem.

It's interesting the escorts lose the convoy. I suppose the upper limit should be set at the max the escorts can stay with the convoy. I thought there was a 20 km limit to calls for help from the merchants. One other thing I've noticed is that only one or two (haven't seen three yet) will stay behind at the last known position. For many of the IC convoys that leaves one to three escorts with the convoy. Maybe with this in mind a longer time isn't a bad thing, just as long as the warships are within the 20 km radius.

Thoughts?

[panthercules]

I've no knowledge on the rate of promotions in the KM. My thought process is that if crew members are transferred off of the boat, they should be replaced by one of equivalent rank (or at least close in rank).

I don't have any data on this either, but I would have thought that it would be more likely that if one of your crew was transferred off your boat you would get a relatively green replacement - after all, what would really be the point to a "lateral" move or exchange of crewmen in terms of overall U-bootwaffe personnel policy/efficiency?

What would be really cool though is if for every relatively senior crewman you had taken away from you in a transfer you would be able to promote one of your other crewmen to step up and take his place - just think how proud Dieter would be at becoming a brevet Oberstabsbootsmann :P

[CCIP]

I actually run my own system of transfers, where every patrol I transfer an average 10% (sometimes more, sometimes less) of crewmen who have served for 5 or more patrols, and I generally replace them with junior crew of the same class. For officers, I tend to be a bit more sophisticated and consider things like their rank and function more than anything else.

From what I've read, this is generally applicable for real u-boats. For instance in my WaW career, in 8 patrols I've transferred off a total of 15 crew, 4 of which were officers. (My current crew is 50).

I also play a little game on patrol where I 'rate' my crew's performance and give them 'points' for any distinctions or write little stories about their screwups. Which I also use in deciding transfers. Silly roleplay, I know, but sure adds to immersion :)

[Beery]

I've no knowledge on the rate of promotions in the KM. My thought process is that if crew members are transferred off of the boat, they should be replaced by one of equivalent rank (or at least close in rank). That's fine. In fact that's great! If you have to spend renown to "buy" a replacement of similar rank then it's a problem.

I may be mistaken in this, and Jaesen can correct me if I'm wrong, but I'm pretty sure there's no way to get equivalent replacements without spending renown to do it. I think the promotion issue will be optional - you're never going to be forced to lose crewmen, but if you choose to promote someone, you will have to accept a loss in renown if you replace the crewman with a man of equal rank. The only way to get free crewmen is to get someone who costs nothing, and those guys have no experience. A big part of the problem with the replacement system as it is, is that there is too little crew turnover, so replacements are virtually never used. The new crew turnover systems Jaesen is implementing right now (promotions, transfers and casualties) will make replacements become a much more active part of the game. The downside of that is that it will use up renown if you choose to implement the feature.

It's interesting the escorts lose the convoy. I suppose the upper limit should be set at the max the escorts can stay with the convoy. I thought there was a 20 km limit to calls for help from the merchants. One other thing I've noticed is that only one or two (haven't seen three yet) will stay behind at the last known position. For many of the IC convoys that leaves one to three escorts with the convoy. Maybe with this in mind a longer time isn't a bad thing, just as long as the warships are within the 20 km radius.

Basically the problem is that ships guarding a convoy have standing orders to cruise at the convoy's speed, so if they get stuck for 45 minutes on a search routine, after it's over they steam along oblivious to the convoy, constantly lagging behind it. Unless they actually spot you with their sonar, radar, hydrophone or AI eyeball, they won't do anything even if they seem to be able to clearly see ships being hit by torpedoes. After 45 minutes they are only perhaps 8km away, but even at 8km, the chances of spotting a periscope is very small, and sonar and hydrophones have no chance at that range. The only chance an escort has of spotting the periscope is by radar, and I'm not sure how far out even that goes, or how reliable it is.

[Observer]

I've no knowledge on the rate of promotions in the KM. My thought process is that if crew members are transferred off of the boat, they should be replaced by one of equivalent rank (or at least close in rank). That's fine. In fact that's great! If you have to spend renown to "buy" a replacement of similar rank then it's a problem.

I may be mistaken in this, and Jaesen can correct me if I'm wrong, but I'm pretty sure there's no way to get equivalent replacements without spending renown to do it. I think the promotion issue will be optional - you're never going to be forced to lose crewmen, but if you choose to promote someone, you will have to accept a loss in renown if you replace the crewman with a man of equal rank. The only way to get free crewmen is to get someone who costs nothing, and those guys have no experience. A big part of the problem with the replacement system as it is, is that there is too little crew turnover, so replacements are virtually never used. The new crew turnover systems Jaesen is implementing right now (promotions, transfers and casualties) will make replacements become a much more active part of the game. The downside of that is that it will use up renown if you choose to implement the feature.

I can think of a couple of ways to do it, but perhaps the easiest is to change the renown requirement to "purchase" new crew members. It would force the user to go to the "barracks" to look for new crew members. What I'm thinking is perhaps a very nominal renown (0, 5, or 10) requirement, but the higher ranking officers would only have one specialty (i.e. watch, repair, torpedo). I'd think the same should be true for the warrant officers as well. I'll have a look at the files to see if it's possible.

It's interesting the escorts lose the convoy. I suppose the upper limit should be set at the max the escorts can stay with the convoy. I thought there was a 20 km limit to calls for help from the merchants. One other thing I've noticed is that only one or two (haven't seen three yet) will stay behind at the last known position. For many of the IC convoys that leaves one to three escorts with the convoy. Maybe with this in mind a longer time isn't a bad thing, just as long as the warships are within the 20 km radius.

Basically the problem is that ships guarding a convoy have standing orders to cruise at the convoy's speed, so if they get stuck for 45 minutes on a search routine, after it's over they steam along oblivious to the convoy, constantly lagging behind it. Unless they actually spot you with their sonar, radar, hydrophone or AI eyeball, they won't do anything even if they seem to be able to clearly see ships being hit by torpedoes. After 45 minutes they are only perhaps 8km away, but even at 8km, the chances of spotting a periscope is very small, and sonar and hydrophones have no chance at that range. The only chance an escort has of spotting the periscope is by radar, and I'm not sure how far out even that goes, or how reliable it is.

Interesting. I know while running RUb 1.43 I've seen escorts steam off at full speed to rejoin the convoy. It's usually one of the things I'm listening for when avoiding escorts. As soon as they kick in high speed (as reported by my SO) I know it's safe to come to PD. Maybe there's a range on this as well. For instance if the convoy is more than x km away, the escorts won't rejoin at max speed. I can think of at least one instance where I noticed an escort lagging way behind. At the time I wrote it off to the heavy seas.

[Beery]

...I would have thought that it would be more likely that if one of your crew was transferred off your boat you would get a relatively green replacement - after all, what would really be the point to a "lateral" move or exchange of crewmen in terms of overall U-bootwaffe personnel policy/efficiency?

What would be really cool though is if for every relatively senior crewman you had taken away from you in a transfer you would be able to promote one of your other crewmen to step up and take his place - just think how proud Dieter would be at becoming a brevet Oberstabsbootsmann :P

Exactly. The important thing to realise in these situations is that by the time you get a crewman who is to be promoted you'll also have other crewmen who have risen in the ranks, so you're likely to have a man already in the crew who is ready to take the vacant position, and you'll have a guy who can take his position too. After promoting from within the boat, the only person you're likely to need will be a raw recruit. This will cost in terms of renown for NCOs and officer recruits, but it will be much less of a cost than if you had to buy a very qualified officer or NCO replacement. Because of this, I think the replacement system will work much better than the standard system, in that there will be a real reason to get a raw recruit.

[Beery]

I can think of a couple of ways to do it, but perhaps the easiest is to change the renown requirement to "purchase" new crew members. It would force the user to go to the "barracks" to look for new crew members. What I'm thinking is perhaps a very nominal renown (0, 5, or 10) requirement, but the higher ranking officers would only have one specialty (i.e. watch, repair, torpedo). I'd think the same should be true for the warrant officers as well. I'll have a look at the files to see if it's possible.

The problem with reducing renown requirements is that it rewards lazyness. If you're not adversely affected by buying a high-level crewman there's no incentive to get your current crew promoted to take the place of a lost crewman, so you'll just buy a high level replacement instead of solving the problem with planning. This is very similar to the fatigue system, which, in the standard game, rewards the player for a lack of long-term strategy. Currently the replacement system is broken because there are virtually no crew losses so you simply don't need replacements at all. A system whereby high-level recruits were cheap would result in a crippled replacement system (this time because of a lack of incentive to keep replacement costs down). Only by keeping recruits expensive AND implementing transfers and promotions off the boat can we have a system that rewards forethought and punishes lazyness. Having said that, I'm not looking to impose a system that punishes players, after all, the player only has a certain amount of control over promotions, but one can (to a certain extent) 'groom' crewmen to replace crewmen who are likely to go. What I'm hoping for is a system that provides incentives to careful play, so that it makes the game richer for players.

In short, the replacement system is an area of the game where we can probably institute a positive 'work for reward' system, so that the game can become more interesting in terms of decisions made when in port.

[Observer]

I can think of a couple of ways to do it, but perhaps the easiest is to change the renown requirement to "purchase" new crew members. It would force the user to go to the "barracks" to look for new crew members. What I'm thinking is perhaps a very nominal renown (0, 5, or 10) requirement, but the higher ranking officers would only have one specialty (i.e. watch, repair, torpedo). I'd think the same should be true for the warrant officers as well. I'll have a look at the files to see if it's possible.

The problem with reducing renown requirements is that it rewards lazyness. If you're not adversely affected by buying a high-level crewman there's no incentive to get your current crew promoted to take the place of a lost crewman, so you'll just buy a high level replacement instead of solving the problem with planning. This is very similar to the fatigue system, which, in the standard game, rewards the player for a lack of long-term strategy. Currently the replacement system is broken because there are virtually no crew losses so you simply don't need replacements at all. A system whereby high-level recruits were cheap would result in a crippled replacement system (this time because of a lack of incentive to keep replacement costs down). Only by keeping recruits expensive AND implementing transfers and promotions off the boat can we have a system that rewards forethought and punishes lazyness. Having said that, I'm not looking to impose a system that punishes players, after all, the player only has a certain amount of control over promotions, but one can (to a certain extent) 'groom' crewmen to replace crewmen who are likely to go. What I'm hoping for is a system that provides incentives to careful play, so that it makes the game richer for players.

In short, the replacement system is an area of the game where we can probably institute a positive 'work for reward' system, so that the game can become more interesting in terms of decisions made when in port.

Can't do it anyways. I think the requirements are hardcoded.

[Observer]

...I would have thought that it would be more likely that if one of your crew was transferred off your boat you would get a relatively green replacement - after all, what would really be the point to a "lateral" move or exchange of crewmen in terms of overall U-bootwaffe personnel policy/efficiency?

What would be really cool though is if for every relatively senior crewman you had taken away from you in a transfer you would be able to promote one of your other crewmen to step up and take his place - just think how proud Dieter would be at becoming a brevet Oberstabsbootsmann :P

Exactly. The important thing to realise in these situations is that by the time you get a crewman who is to be promoted you'll also have other crewmen who have risen in the ranks, so you're likely to have a man already in the crew who is ready to take the vacant position, and you'll have a guy who can take his position too. After promoting from within the boat, the only person you're likely to need will be a raw recruit. This will cost in terms of renown for NCOs and officer recruits, but it will be much less of a cost than if you had to buy a very qualified officer or NCO replacement. Because of this, I think the replacement system will work much better than the standard system, in that there will be a real reason to get a raw recruit.

I actually rather like this idea as long as junior crew members can get promoted at a reasonable rate. As it is 100 experience points for the next rank for officers means you'll always have a crew of the most junior officers regardless of qualifications at the experience rates I'm currently seeing (4 - 12 points per patrol). I think this requirement should get changed to something like 25 or 30 points of experience (maybe a bit more like 40). This would translate to about 5 to 8 good patrols before the next rank with no guarantee they wouldn't get transferred before the next rank anyways. Then you could pick up the lowest ranking officer in the barracks without a renown penalty and it would still require the player to do some long term planning for qualifications.

[Beery]

I actually rather like this idea as long as junior crew members can get promoted at a reasonable rate. As it is 100 experience points for the next rank for officers means you'll always have a crew of the most junior officers regardless of qualifications at the experience rates I'm currently seeing (4 - 12 points per patrol). I think this requirement should get changed to something like 25 or 30 points of experience (maybe a bit more like 40). This would translate to about 5 to 8 good patrols before the next rank with no guarantee they wouldn't get transferred before the next rank anyways. Then you could pick up the lowest ranking officer in the barracks without a renown penalty and it would still require the player to do some long term planning for qualifications.

Yes. There will probably be some need to fine tune the promotions system. Once the system is working we'll see what's needed in terms of tuning the experience points to make them work efficiently.

[JScones]

...Even I know the AI will depart 15 minutes after they've been unable to detect my boat. That leads to some very bad behaviors, but also greatly reduces some of the uncertainty. That would actually be a nice parameter to randomize.:hmm:

Great idea! We could easily do this. What parameters do you envision? Currently, RUb has a 40 minute departure delay, but we could easily randomize it so that it went from 15 to 45. After ships lost contact, what would be a reasonable time delay for them to give up the search? We have to be careful not to have too big a delay because the escorts get too far from the convoy and they don't catch up, which leaves the convoy more open to attack.
I've just added this feature to SH3 Commander. I've used 15 and 45 minutes as the boundaries, but the values are stored in a new cfg file, so are fully editable.

[pampanito]

May I comment again on the issue of tonnage, the WaW campaign shows that SH3 players are perhaps sinking too many big ships, just look at the latest two periods covered by WaW:

Period June/August 1941
Real U-boat sinkings: 119 ships / 524,044 tons (average 4403)
Wolves at War (SH3): 88 ships / 534,377 tons (average 6072)

Period September/November 1941
Real U-boat sinkings: 100 ships / 459,079 tons (average 4590)
Wolves at War (SH3): 151 ships / 991,493 tons (average 6566)

For the whole war, U-boats destroyed 2918 ships for 14,879,472 tons, that's an average of just over 5000 tons per ship (5099 to be exact).

I agree that some commanders racked big tonnages, but if we look at Kals in U-130 we see that his impressive total is owed, in part, to a successful attack in unusual circumstances (the sinking of three anchored troop transports on 12 November 1942, EDWARD RUTLEDGE, TASKER H. BLISS and HUGH L. SCOTT, for a total of 34,407 tons) and also to the fact that his big U-boat was sent three times to the US coast/Caribbean during the 'Happy times' of 1942, bagging no less than seven big tankers. Even so, Kals' successes included two ships of barely 1500 tons, FRISCO and FIDRA, a size which is smaller than even the coastal merchants of SH3.

The Random tonnage mod works OK for me, with one 'but': the randomization seems too small, often just one or two tons between ship and ship, I think it should be at least a hundred. I have just sunk three C2 in my latest patrol, November/December 1941; awarded tonnages:
5159, 5160, 5161 !
In another patrol I sank just two C2; each had the same tonnage (5041).
Although I am afraid this small random margin is not the fault of the Random mod, but of the way the SH3 engine calculates the tonnage.

And I have yet to sink a ship in the range 4000-4900...

One last comment: if something seems too high (at least for the period 40/42) is the frequency with which convoys are attacked, too many convoys versus single ships. In WaW, 90% of the sinkings reported are from convoys! Add 'veteran' and 'elite' SH3 players, and each convoy battle ends in multiple sinkings of big ships.

BTW, a big THANK YOU to Beery, Observer and all other modders, who are still trying to make this game better and better!

[Beery]

As I said before, I fear the overall tonnage sunk by all boats is not going to be truly representative of the aces. The game tends to simulate the aces much better than it simulates KptLt. Hans Sixpack who went on 6 patrols and sank little more than a few smaller ships. I suspect that if we could break down the numbers we would find a marked difference in tonnage per ship between aces and ordinary commanders.

Anyway, the next version of Jaesen's SH3 Commander will have a new random tonnage feature which is based on Observer's mod. The feature is fully user-configurable, so the player can adjust the base tonnage figure for each ship type as well as the tonnage range (%+/- for all ships), and how much the ships increase in tonnage as the war goes on.

[Beery]

The Random tonnage mod works OK for me, with one 'but': the randomization seems too small, often just one or two tons between ship and ship, I think it should be at least a hundred. I have just sunk three C2 in my latest patrol, November/December 1941; awarded tonnages:
5159, 5160, 5161 !
In another patrol I sank just two C2; each had the same tonnage (5041).
Although I am afraid this small random margin is not the fault of the Random mod, but of the way the SH3 engine calculates the tonnage.

It is. The problem is, however we randomize the tonnage, if you sink two of the same ship type in a patrol, you will get the same tonnage (plus or minus the tiny one or two tons of randomization that already exists in the game). There is no known way to make a bigger difference between two or more ships of the same type sunk in one patrol.

[iambecomelife]

The Random tonnage mod works OK for me, with one 'but': the randomization seems too small, often just one or two tons between ship and ship, I think it should be at least a hundred. I have just sunk three C2 in my latest patrol, November/December 1941; awarded tonnages:
5159, 5160, 5161 !
In another patrol I sank just two C2; each had the same tonnage (5041).
Although I am afraid this small random margin is not the fault of the Random mod, but of the way the SH3 engine calculates the tonnage.

It is. The problem is, however we randomize the tonnage, if you sink two of the same ship type in a patrol, you will get the same tonnage (plus or minus the tiny one or two tons of randomization that already exists in the game). There is no known way to make a bigger difference between two or more ships of the same type sunk in one patrol.

I'm glad you're bringing this up, since it's been driving me crazy. Part of the appeal of "Aces of the Deep" was only having a rough idea of how much each vessel weighed; with SH3 it's entirely predictable. It makes sense considering that almost all of the ingame ships are supposed to be mass produced classes but it also reveals that they may have overestimated how common these ships were. Even the "aces" often had patrols with nothing but 3000-5000 tonners, as opposed to the relatively large t-2's and c-3's we always get. I'm afraid that even if we DO find out what file is responsible for tonnage variation, altering it will affect all ships including liberty ships and escorts.

[Beery]

We are planning, in SH3 Commander, to have a much greater range of tonnage variation. In the proposed mod as it is currently, the ships will vary up to 30% (+10% and -20% from their base values), and the ships will gain about 2% in tonnage each year. These settings will probably change, and are fully player-configurable.

One question I have is regarding the mass-produced ships. Were tonnage claims based on the official tonnage of the ship? Were they based on the tonnage plus cargo? If the Victory and Liberty ships were all produced based on the same plans, the tonnage would be identical unless they included the cargo weight. If these ships are all the same, and if the cargo was not a factor in tonnage claims, there's not much point in including these in the tonnage randomization file.

[thasaint]

observer:

where can i find the mod that changes the ship names to small medium large? and perhaps this could be added as one of the optional things in SH3 commander, i never even knew of it till i read this post

[pampanito]

If the Victory and Liberty ships were all produced based on the same plans, the tonnage would be identical unless they included the cargo weight.

Exactly! 90% of the Liberty ships were registered at 7176 tons (a few 7177), the other 10% built at an specific yard were 7191 tons. The Oceans were registered with minimal differences, between 7172 and 7175, and the Forts between 7130 and 7135. Nobody should protest if all were given 7176 tons in SH3.

[iambecomelife]

We are planning, in SH3 Commander, to have a much greater range of tonnage variation. In the proposed mod as it is currently, the ships will vary up to 30% (+10% and -20% from their base values), and the ships will gain about 2% in tonnage each year. These settings will probably change, and are fully player-configurable.

One question I have is regarding the mass-produced ships. Were tonnage claims based on the official tonnage of the ship? Were they based on the tonnage plus cargo? If the Victory and Liberty ships were all produced based on the same plans, the tonnage would be identical unless they included the cargo weight. If these ships are all the same, and if the cargo was not a factor in tonnage claims, there's not much point in including these in the tonnage randomization file.

According to David Irving's book on PQ-17, tonnage was calculated based on GRT (Gross Registered Tonnage). Thus, the weight remained constant, with liberties always being about 7200 tons, for instance. It wouldn't matter if you sank one in ballast or loaded.

[Beery]

So does that count for the Victory Ships too? Currently I have both Liberty Ships and Victory Ships set up with a lot of variation.

[pampanito]

So does that count for the Victory Ships too? Currently I have both Liberty Ships and Victory Ships set up with a lot of variation.

Perhaps Victory Ships should not be included in SH3 at all. They entered the war very late, and mainly in the Pacific.
Just three of them were lost in action, all to Japanese kamikazes in April 1945: CANADA VICTORY, HOBBS VICTORY, LOGAN VICTORY. Tonnages given as 7608, 7607 and 7607 respectively.

[Observer]

So does that count for the Victory Ships too? Currently I have both Liberty Ships and Victory Ships set up with a lot of variation.

Victory and Liberty ships should have a tonnage values which should not vary more than a few GWT.

[Observer]

observer:

where can i find the mod that changes the ship names to small medium large? and perhaps this could be added as one of the optional things in SH3 commander, i never even knew of it till i read this post

Sorry it's taken a while to reply. I've been out of town for a few days.

The mod can be found on http://u-boot.realsimulation.com/

Look for the file "Random Tonnage and Names Mod"

[Observer]

So does that count for the Victory Ships too? Currently I have both Liberty Ships and Victory Ships set up with a lot of variation.

Perhaps Victory Ships should not be included in SH3 at all. They entered the war very late, and mainly in the Pacific.
Just three of them were lost in action, all to Japanese kamikazes in April 1945: CANADA VICTORY, HOBBS VICTORY, LOGAN VICTORY. Tonnages given as 7608, 7607 and 7607 respectively.

If you are using Improved Convoys, Victory Ships are included, but only in a very limited approach based on a mathematical curve with a small percentage probability of appearance in the fall of 1944 if I remember correctly. If you want a more accurate number, I'd have to go back and check my data on that to be sure.