View Full Version : To autocrew or not to autocrew, that is the question.
somedude88
03-27-18, 12:41 PM
Do you autocrew any or all stations? I try to man the sonar to understand how to use it but the workload is just off the charts!
Also what does autocrewing fire control do? Autocrewing that station doesn't seem to do anything g.
The stations I don't auto crew are ESM and Radio. The radio shack belongs to ET's and RM's and, BTW, keep your hands off my freaking gear. :up:
In DW, you're playing as the OOD. That means you're over-seeing the watch standers in Control. You get to make the call when and how to maneuver the boat so those guys can do their job.
I'll turn off the sonar auto crew if I really want to nail down a contact early or I'll turn off the FT/tracking auto crew if a track is starting to turn sour but most of the time I let them do the heavy lifting.
In the end, it comes down to your style. :salute: Playing as a laid-back OOD on the mid watch and hanging out near the QMOW or FTOW is just as valid as playing like a micro manager who is bugging everyone to polish their sneakers and get a hair cut. :yeah:
First of all, whatever your understanding level and experience in the game permits you to do. Whatever your playstyle is.
The torpedo fire control autocrew should enter apropriate torpedo parameters. run-to-enable, floor and ceiling depths... that sort of thing. It really isn't the most complicated stuf you have to do. Prevent your torps from running in the floor and not enable after passing the contact. Search depth can help with popping the torpedo up to the other side of the layer and stay stealthy for a bit longer. (torpedo transit at the depth you fired them at) I'm not sure the firecontrol autocrew is really smart either.
I autocrew mostly the tedious tasks. Radar plotting and sonar intercept. TMA on the OH-Perry because it is just practically impossible to do it manually there. TMA on the subs in target rich environments only. But I do want to fiddle with that myself mostly. TMA Autocrew seems to cheat too much. It knows too much with too little information. So I autocrew it as little as possible. I'm not sure what autocrew the helo has, but I think I only let the mad/esm, radar and pilots autocrew.
Sonar autocrew in the subs basically only scan the bearings for contact and mark anything it comes across. Even if it is a false signal or your own noise. TMA autocrew drops it if it is a mirror contact. Generally the sonar crew spam many contact reports (that TMA dropped previously) and doesn't help clarity or situational awareness. Also, I find searching the faint signals, classifying and figuring out who's where on what array the fun part. So I do that manually.
Gray Lensman
03-27-18, 06:43 PM
AutoTMA is flawed per the RedBook in the download section here: http://www.subsim.com/radioroom/downloads.php?do=file&id=5187
In fact pg. 22, has a specific topic "IX. TMA - THE FLAWS OF AUTOTMA (aTMA).
It does work somewhat competently against strictly AI opponents, which if that's all you will ever play against, you will probably think that you're a great sub commander.
Playing against a live opponent that's reasonably competent in manual TMA and knows that you are relying on AutoTMA (aTMA) will probably result in your losing 9 times out of 10. He will use those known flaws against you and your sub's (aTMA) will NOT get a valid solution.
TMA Autocrew seems to cheat too much. It knows too much with too little information.
Ding! :D That isn't cheating, that's an exploit. :yeah:
There is no way I can pick up a weak contact before the auto crew unless I get lucky.
:Kaleun_Cheers:
Gray Lensman
03-27-18, 07:22 PM
Ding! :D That isn't cheating, that's an exploit. :yeah:
There is no way I can pick up a weak contact before the auto crew unless I get lucky.
:Kaleun_Cheers:
That is the one reason to use it, INITIALLY. I totally suck at finding weak contacts on the waterfall. Might be due to some slight color blindness. Once it finds a bearing for you, take over manually, because from that point on, it's flawed in its' following TMA execution steps. See the Redbook referenced above.
Gray, I would argue that your auto TMA is only as good as the way you drive the boat.
:Kaleun_Cheers:
For example, if you hold the same course and speed during a two hour leg- of course you're going to get a crappy solution from TMA. TMA relies on how your bearing information changes over time. Many times, its up to you to force that change in the raw data. If you start out in a lag situation, change over to lead (or at least a different lag solution) and watch how smarter your auto crew gets. :D
Most players in discussion autocrew ON/OFF focus on sonar and TMA. For me most important is countermeasures setting: always manual. It allows you to launch Chaffs and Flares which is more effective than in auto mode.
As MP game host usually settings looks:
(Auto.manual means each player have controll over this setting independently)
TMA - auto/manual
Active intercept - auto/manual
ESM - auto/manual
Surf countermeasures - auto/manual
Link data - on/off
All rest - manual
Quick launch for weapon or helo - disabled
Quick repairs - disabled
Engage with menu (from map) - disabled
Show Dead Platforms - enabled
Discussion about auto TMA is usually conducted by people who play only with or only without it. Both settings have props and cons. Auto TMA very often gives you quite good range, course and speed of target in first solution. But after time error can be bigger and even solutions aren't somtimes updated correctly. It happens in more complex MP scenarios.
In my opinion best Auto TMA profits have players who knows how to play with manual TMA. They knows how important is changing course and speed own subamrine. Most "toxic" in manual TMA is style of playing this guys who wanna resolve target solution as perfect as mathematical equation :D
Gray Lensman
03-27-18, 09:02 PM
Gray, I would argue that your auto TMA is only as good as the way you drive the boat.
:Kaleun_Cheers:
For example, if you hold the same course and speed during a two hour leg- of course you're going to get a crappy solution from TMA. TMA relies on how your bearing information changes over time. Many times, its up to you to force that change in the raw data. If you start out in a lag situation, change over to lead (or at least a different lag solution) and watch how smarter your auto crew gets. :D
I'll stick to RedBook information. It does in fact suggest course changes every 10 minutes accompanied with Speed changes to screw up the opponents TMA (especially AutoTMA). Now, if you use those same course changes to keep your course relatively perpendicular to the target LOS, then you will also help your own TMA gather better raw data since you are maximizing speed "across the LOS".
Another thing the RedBook points out is that if you have a bearing on a target, he almost always will have one on you, even AutoTMA, or you should at least assume so.
This being a pretty good assumption, you really don't want to purposefully place your ownship in a LEAD situation in regards to your target. due to the advantage it gives his Torp weapons if he decides to solution fire or snapshot your bearing. Basically, it will take you much longer to evade/turn out of the torps' lead targeting course since you are heading into it already.
Besides, attempting TMA on a LEAD LOS on a "single array" is generally useless, read the RedBook to understand why.
Dual array is different. Once converted to a master contact it represents "truth" on the enemy solution, no matter LEAD or LAG. It only takes 2 such readings to determine a reasonably accurate solution, if the target does not change course or speed of course. Basically, the last two tick marks of the ruler placed on the intersecting dual/master "hits" gives you a pretty good solution without all the other dots lining up. A third "hit" (using the last 3 ticks of the ruler) pretty well eliminates "eyeball" error on the TMA display.
Gray Lensman
03-27-18, 09:23 PM
<cut>
Most "toxic" in manual TMA is style of playing this guys who wanna resolve target solution as perfect as mathematical equation :D
Yep, that's why I was pointing out dual array/master contact needing only 2 or 3 "truth" hits to line up the last 2 or 3 corresponding ladder links and forget about the rest of the ladder and previous hits.
For clarification, by "truth" hits, I don't mean turning on Truth in game. I mean the intersection of both array bearings is a "true" reading where the bearings cross in regards to range.
Yep, that's why I was pointing out dual array/master contact needing only 2 or 3 "truth" hits to line up the last 2 or 3 corresponding ladder links and forget about the rest of the ladder and previous hits.
I disagree because previous legs are impotrant for better solutonas and predicting speed (even without DEMON work). Also merged contacts from different sensors gives you "not clear" picture.
I had in mind taht you needs some experience for estimating which legs/dots are important for good solution.
https://i.imgur.com/V5BB41L.png
This screen is from my video.
https://youtu.be/v01LWOWTjyM?t=24m30s
I'll stick to RedBook information. It does in fact suggest course changes every 10 minutes accompanied with Speed changes to screw up the opponents TMA (especially AutoTMA). Now, if you use those same course changes to keep your course relatively perpendicular to the target LOS, then you will also help your own TMA gather better raw data since you are maximizing speed "across the LOS".
Another thing the RedBook points out is that if you have a bearing on a target, he almost always will have one on you, even AutoTMA, or you should at least assume so.
This being a pretty good assumption, you really don't want to purposefully place your ownship in a LEAD situation in regards to your target. due to the advantage it gives his Torp weapons if he decides to solution fire or snapshot your bearing. Basically, it will take you much longer to evade/turn out of the torps' lead targeting course since you are heading into it already.
Besides, attempting TMA on a LEAD LOS on a "single array" is generally useless, read the RedBook to understand why.
Dual array is different. Once converted to a master contact it represents "truth" on the enemy solution, no matter LEAD or LAG. It only takes 2 such readings to determine a reasonably accurate solution, if the target does not change course or speed of course. Basically, the last two tick marks of the ruler placed on the intersecting dual/master "hits" gives you a pretty good solution without all the other dots lining up. A third "hit" (using the last 3 ticks of the ruler) pretty well eliminates "eyeball" error on the TMA display.
The red book is good, just trust me on this: I've had more real world experience with TMA (both underway and in attack trainers) than most folks would find rational or even sane. :doh: :D
TMA is tricky. You need to understand the theory, the geometry, and the math. You also have to realize that TMA is ALWAYS in the past. At best, it will always be an approximation to what's really going on. It doesn't really matter if you have a really expensive state-of-the-art sonar dome and processors when you're using them in a highly imperfect medium like an ocean.
At some point, you have to drop the textbooks and the theory and go with the "zen". :03:
Let me try a quick example to prove what I'm saying. Let's say you could plot a TMA solution that's so good you can dial in your contact to within 10 yards of where they really are, within one degree of their true course and within one knot of their true speed. Let's call that a perfect solution. :up:
What do you think is probably going to happen as soon as you launch that fish?
:yep:
Yep, that perfect solution is no longer perfect and your target will be doing everything they can to stop being a target before your fish goes active.
That's why its important to know the theory and geometry BUT its also important to recognized things like "windage" "ballparks" and "guess work". :D
Yeah, yeah, yeah.. I know. What the heck does this have to do with turning the auto crew off or on? :o
Recall that DW was based off a training program for midshipmen ("Harpoon"). Part of what Harpoon taught was how to maneuver a ship and launch weps. There was also another sneaky lesson: learn when to trust the enlisted crew. :yeah:
Gray Lensman
03-27-18, 11:26 PM
I disagree because previous legs are impotrant for better solutonas and predicting speed (even without DEMON work). Also merged contacts from different sensors gives you "not clear" picture.
I had in mind taht you needs some experience for estimating which legs/dots are important for good solution.
<cut>
edit> Nice tutorial video against a straight running surface ship. Your use of the Narrowband screen was great. I'd love to see one of your manual TMA videos against a maneuvering enemy sub at depth however. Do you have a YouTube channel? If so, could you link it?
Straight out of the RedBook pg 14
VI. TMA - TRACKING ON A DUAL ARRAY
1.6.1 For tracking targets on a dual array, the techniques to acquire an accurate solution discussed up to this point need not apply. A dual array contact, or master contact, represents truth at all times. As such, this is the most desirable of all the TMA techniques as it is an instantaneous, accurate solution and cannot be foiled by target zigs. It is still prudent to keep the target on your beam, in a lag line of sight, for evasion purposes (see Chapter 3: Tactics).
1.6.2 It doesn't matter how much your target zigs, you will always know exactly where he is by ensuring the two most recent tick marks of the ruler fall on the two most recent intersections of your dual array bearings. It is not imperative that you get the "straight dot stack" here, you know where he is at and you know his course, and thus his range, from the two most recent intersections of your dual array bearings. Just ensure that the two most recent tick marks of the ruler fall on the two most recent bearing lines; forcing the top two dots to be in a straight line on the zero error line.
1.6.3 Dual Array bearings are typically purple and white lines. The purple lines represent bearings that are being attained from your towed array and the white lines represent bearings being attained from your spherical array. Additionally, you can have dual array contact with your hull array and either your spherical or towed array - though rare. In fact, if your hull array holds the target, then your spherical and towed array most certainly hold the target as well, assuming you have the target exposed to these arrays. Thus, you could have a three array contact. Same principles still apply - where the bearing lines of the various arrays intersect is where the target was at the time those bearing lines were laid down. Refer to Figure 1.6.3 as an illustration.
See the RedBook pg 13 for the Figure 1.6.3 referred to above. It basically shows the dual array hits as a sequence of triangulations which when connected together form a pretty accurate solution based on just the last two or three hits of the TMA sequence.
The above makes perfect sense to me. Now what happens after Torps are in the water is a different discussion, but the TMA solution offered above by dual-array contacts on just the last two or three hits is perfectly valid.
FPSchazly
03-28-18, 08:05 AM
That Red Book quote with the two latest ticks is quite interesting, I may have to test that at some point. The towed array can have decent bearing error that produces a sloppy looking dot stack in the TMA, I can't imagine that the red book scenario would be true for that. Does anyone know if the towed array bearing error is a stock game thing or if RA added it? Regardless of whether the Red Book thing is correct or not, two points always make a straight line so I'm always skeptical of those kind of situations :D
To throw in my $0.02 to the original question, I like doing manual TMA as a "time killer" and it kind of makes the game like a puzzle game. Auto TMA is good for very busy areas or evading torpedoes.
Gray Lensman
03-28-18, 10:52 AM
That Red Book quote with the two latest ticks is quite interesting, I may have to test that at some point. The towed array can have decent bearing error that produces a sloppy looking dot stack in the TMA, I can't imagine that the red book scenario would be true for that. Does anyone know if the towed array bearing error is a stock game thing or if RA added it? Regardless of whether the Red Book thing is correct or not, two points always make a straight line so I'm always skeptical of those kind of situations :D
I guess it would depend on the amount of error.
More speed across the line of sight "generating a wider bearing fan" might mitigate this error for purposes of the above "last" 2 or 3 tick mark setups. Might be interesting to test it out against the game "truth" display to see how much error is present.
In regards to testing above, I've been trying to test the 2-3 tick method in a modified one on one scenario between 2 688i class subs, but trying to get a dual array contact on a 688i before it gets a shot off and necessitating evasion is extremely difficult. I had already tried a one on one 688i vs Akula, but the Akula missiles were a distraction then. Might have to change the test to 688i vs Victor I for dual array tracking purposes or go against a noisy surface ship to eliminate the "evasion distraction".
To throw in my $0.02 to the original question, I like doing manual TMA as a "time killer" and it kind of makes the game like a puzzle game. Auto TMA is good for very busy areas or evading torpedoes.Agree.
I use all auto-crew!
I'm the Captain, god dammit!!! :)
Ding! :D That isn't cheating, that's an exploit. :yeah:
There is no way I can pick up a weak contact before the auto crew unless I get lucky.
:Kaleun_Cheers:Developers exploiting the use of game internal state-variables that you do not have is cheating!
Auto-TMA doesn't detect (weak) contacts at all. (Sonar and ESM autocrew detects). My problem with it is it knows through divine (developer-) intelligence without you changing course that the mirror contact on the Towed Array doesn't have an actual soundsource. It is dropped immediately. And it seems to know the target course from a few lines of bearing before you even made a turn to alter the relative motion between you and the contact. Clearly Auto-TMA knows things it shouldn't.
edit> Nice tutorial video against a straight running surface ship. Your use of the Narrowband screen was great. I'd love to see one of your manual TMA videos against a maneuvering enemy sub at depth however. Do you have a YouTube channel? If so, could you link it?
Thx but it wasn't straight running surface ship. This ship has made turn couple minutes ago.
I have YT channel but im not experienced video maker :)
My DW playlist (https://www.youtube.com/playlist?list=PLKfuYesPGTfOQio1aa6XFI-RsPi_C2IJT)
BTW Red book was writen to Sub Command. This game had much easier TMA because legs were always exactly where they should to be. In DW (with RA mod) situation is little bit different.
See the RedBook pg 13 for the Figure 1.6.3 referred to above. It basically shows the dual array hits as a sequence of triangulations which when connected together form a pretty accurate solution based on just the last two or three hits of the TMA sequence.
The above makes perfect sense to me. Now what happens after Torps are in the water is a different discussion, but the TMA solution offered above by dual-array contacts on just the last two or three hits is perfectly valid.
Now you know what to do and how to do so... please show me this on my previous screenshot :)
https://i.imgur.com/V5BB41L.png
You have above dual and even triple sensor legs on one tracked target :)
FPSchazly
03-28-18, 07:40 PM
BTW Red book was writen to Sub Command. This game had much easier TMA because legs were always exactly where they should to be. In DW (with RA mod) situation is little bit different.
I think we have our answer. :up:
I just made scenario for TMA training in "real" mission. Main feature is that primary target all the time makes low random course changes.
This scenario probably will be very easy with auto TMA :)
This is video from my game:
https://youtu.be/EV2cIqJRf8g
Link to scenario is description (under video)
P.S. I forgot to uncheck "read only" from replay.dat file so this video haven't replay at end :(
Gray Lensman
03-28-18, 08:51 PM
I think we have our answer. :up:
Of course it is dated being written during Sub Command, but the principles are still valid regarding master contacts. The DW towed array "bearing errors" will just cause a little more uncertainty to the solution probably meaning 3 or more such hits will be necessary. I am also using RW 1.44.
O.T. How the heck do posters insert images on this site? or more precisely, what site do they use to provide the url link asked for when you click "insert image"?
Subsim is user "un-friendly" in this case :)
Try ro use external images host services like imgur or similar.
Gray Lensman
03-28-18, 10:32 PM
I just made scenario for TMA training in "real" mission. Main feature is that primary target all the time makes low random course changes.
This scenario probably will be very easy with auto TMA :)
This is video from my game:
<cut>
Link to scenario is description (under video)
P.S. I forgot to uncheck "read only" from replay.dat file so this video haven't replay at end :(
Enjoyable and informative video, BUT it still doesn't actually disprove the validity of the earlier discussed technique of using the last 2 or 3 master contact hits.
Admitedly, this will be hard to show, but since conditions change between the time of the weapon release and the actual Torp target acquisition (all that being variable), I'm more interested in the immediate TMA solution (Bearing, Speed, Course and Range) that you utilized for weapon firing (not the Torp settings), compared to the instantaneous "Show Truth" information (Bearing, Speed, Course and Range). Especially if you can manage to just use the last 2 or 3 master contact hits to get the immediate TMA solution.
Subsim is user "un-friendly" in this case :)
Try ro use external images host services like imgur or similar.
Thanks. Here's Figure 1.6.3 of the RedBook that I was referring to earlier.
https://imgur.com/uRQMFerhttps://i.imgur.com/uRQMFer.jpg
Figure 1.6.3 Depicts a dual array contact on target in lead LOS. Note that no matter what the LOS, or what the target is doing, a dual array contact always represents truth. Also note that the same solution would be attained with only two sets of bearing lines instead of the three sets that are present.
Now with the discussion deviating to the inaccuracy of the towed array introduced in DW vice SubSim. It would be interesting to prove or disprove the validity of the dual contact accuracy vs the longer TMA solutions that your videos are showing. I would believe that more than likely three sets may actually be required now in DW due to the induced towed array error.
https://imgur.com/uRQMFer
Developers exploiting the use of game internal state-variables that you do not have is cheating!
Wellllllll............ Um. :)
This might get me shot at dawn, but there was a way back in the old days to perk up that waterfall display. :03: I'm pretty sure everyone in the old community knew about it, but if you had a CRT monitor all you had to do was crank up the gamma slider on the monitor and you'd gain some dB in the waterfall. It looked nasty and you wouldn't gain much but the data was there. :up:
And yeah, of course I tried it on a flat screen but you can't see it anymore. :arrgh!: LCD and LED monitors don't process brightness and gamma like the old CRTs.
The truth is: most players doesn't know how to use manual TMA. So if you wanna play more than once in 2 month, you should give a choice.
Auto TMA isn't perfect. Its true that computer use "mystical knowledge" for predicting course and speed but in other hand, auto TMA very often gives you:
- too big error for contacts tracked long time.
- dont updates solutions (even with strong SNR)
- first solution can be very accurate 50%
- first solution can be terrible unaccurate 50%
In my opinion players should learn how to play with auto TMA. Its like "area of enemy presence" in Harpoon games on CMANO. Game draws you large figure where is enemy ship. In DW auto TMA players should have in mind "virtual area" just like in Harpoon or CMANO.
Now with the discussion deviating to the inaccuracy of the towed array introduced in DW vice SubSim. It would be interesting to prove or disprove the validity of the dual contact accuracy vs the longer TMA solutions that your videos are showing. I would believe that more than likely three sets may actually be required now in DW due to the induced towed array error.
Ok I will try to make video which will show how much this "trick" was effective only in SC (and 688 H/K). But i think you should have more wider perspective on TMA work. Of course merging contacts from different sensors in one Master gives you more informations. But not in that way as you think (not via crossing lines as 100% location of target). I have observe that many players are focusing on dots like students on mathematical riddle :)
Its wrong! You should have more resoluteness in TMA work. When i see screens on subsim where someone tries to cross lines in one point and speed of his sub is only 1-3 kts i know - he do it wrong.
BTW im not TMA specialist. My games are in 99% with auto TMA. I just know how it works - not only from "red/blue books" but mainly from single player testing in DW.
I think FPSchazly is much better in TMA because he likes this work - I dont :P
Gray Lensman
03-29-18, 12:37 AM
The truth is: most players doesn't know how to use manual TMA. So if you wanna play more than once in 2 month, you should give a choice.
Auto TMA isn't perfect. Its true that computer use "mystical knowledge" for predicting course and speed but in other hand, auto TMA very often gives you:
- too big error for contacts tracked long time.
- dont updates solutions (even with strong SNR)
- first solution can be very accurate 50%
- first solution can be terrible unaccurate 50%
In my opinion players should learn how to play with auto TMA. Its like "area of enemy presence" in Harpoon games on CMANO. Game draws you large figure where is enemy ship. In DW auto TMA players should have in mind "virtual area" just like in Harpoon or CMANO.
Thing is, if I wanted to play like that, I would have bought Cold Waters, but I prefer the technical challenge of playing with mostly manual TMA. Obviously, the only way to learn it is to practice, practice, practice.
O.T. > I have all the time in the world since I'm retired now. (Literally, 12+ hours each day) It's definitely fun and interesting to me, since 45 years ago I was a young sailor on a U.S. DDG (USS Lawrence DDG-4 to be specific). A few years later, I was stationed at Charleston, S.C. Weapons station servicing Boomers on the USS Hunley AS-31. (1976-1977 I believe) I got to know a lot of Boomer crew members and had several of them for neighbors in military housing on the base.
Ok I will try to make video which will show how much this "trick" was effective only in SC (and 688 H/K). But i think you should have more wider perspective on TMA work. Of course merging contacts from different sensors in one Master gives you more informations. But not in that way as you think (not via crossing lines as 100% location of target). I have observe that many players are focusing on dots like students on mathematical riddle :)
Its wrong! You should have more resoluteness in TMA work. When i see screens on subsim where someone tries to cross lines in one point and speed of his sub is only 1-3 kts i know - he do it wrong.
I try to work with 10 knots across the LOS preferably in a LAG situation to increase the bearing fan. The only time I'm slower than that is near the surface.
BTW im not TMA specialist. My games are in 99% with auto TMA. I just know how it works - not only from "red/blue books" but mainly from single player testing in DW.
I think FPSchazly is much better in TMA because he likes this work - I dont :PActually, you're both pretty good, though you could add some voice to your videos for improvement. I have both you and FPSchazly's YouTube channels bookmarked for reference. :)
...This game had much easier TMA because legs were always exactly where they should to be. In DW (with RA mod) situation is little bit different....
You have above dual and even triple sensor legs on one tracked target :)p7p8, be aware that you may have some missunderstanding of the word "leg" commonly used when talking about TMA (or navigation in general). Often a "leg" in navigation refers to the path of a contact between course (or speed) changes or between waypoints. You seem to mean lines of bearings with "(sensor) legs". That is a significant difference you need to be aware of.
Wellllllll............ Um. :)
This might get me shot at dawn, but there was a way back in the old days to perk up that waterfall display. :03: I'm pretty sure everyone in the old community knew about it, but if you had a CRT monitor all you had to do was crank up the gamma slider on the monitor and you'd gain some dB in the waterfall. It looked nasty and you wouldn't gain much but the data was there. :up:
And yeah, of course I tried it on a flat screen but you can't see it anymore. :arrgh!: LCD and LED monitors don't process brightness and gamma like the old CRTs.We seem to be talking about 2 totally different things. Me: contact solutions, You: signal strength.
@Pisces, you have right. I had in mind line between sensor and tracked contact
Thanks :salute:
vBulletin® v3.8.11, Copyright ©2000-2025, vBulletin Solutions Inc.