TDC innacuracies.

[Zivs]

Alright I have made another thread about this before but I believe I have finally found the issue.

I found a T2 tanker cruising along at a whopping 13 kts (love the speed changes btw). Spent half an hour save scumming this poor bloke trying to figure out why sometimes my shots were dead perfect and others were scraping by aft of the ship especially on smaller targets. Finally think I found what may have been the issue and as others have said before it has to do with salvo setting. I have attached a video with 3 tests. the measurements and all are near perfect, as I have spent half an hour acquainted with my new T2 friend.

What is demonstrated in the video:
Game state 1-
T2 with a perfect 90 AOB at 1600-1700m. The distance is not the issue. I was able to hit him using settings from 1300-2100 meters in previous tests. The hit is solid albeit not as accurate as I would like especially if gunning for a small ship.

Game state 2-
Same exact setup. This time I manually set the salvo to 0 degrees. It's a different hit and more accurately reflects the errors that I probably put in with an imperfect solution which is good enough for me.

Game state 3-
An exaggerated salvo spread of 15 degrees. I wanted to exacerbate this issue to show that even on a single torpedo launch, salvo is still being factored into the firing solution. This shot misses WILDLY aft.

Why all of this matters? As I alluded to earlier it matters for hitting smaller targets. Hitting slightly aft on a massive T2 of 152m isnt a big deal. Hitting, but more like missing aft on a isles corvette of 50m is a big issue.

Another issue I have observed is although I may disable salvo on the TDC or on the reciever panel, after some time in the periscope it will revert all of my settings back to what the AI wants and in the case of this T2 it is a salvo spread of 8 degrees.

The video is rough, I just wanted to quickly showcase what is happening and what I think may be the root cause of why sometimes you will have an incredibly solid hit, and other times your torp will take a vacation to the Marianas trench.

I will be adding more observations and videos in this thread. If there are any TDC whizzes who are interesting in finding out the inner workings and quirks in the game, feel free to add me on discord to discuss. I suppose I will be doing lots of reading to make sure I understand functions to get the most accurate hits possible.

Discord: zivs.

https://youtu.be/7CQ-aJqox-E

[Zivs]

Did some more testing. Figured out what I think is a bandaid to the salvo issue. Flood 2 torp tubes and set dispersion to 0m.

I also installed the stable 2022 version and did some comparisons between aiming in beta.
In this video it starts off with 2022. I use only the periscope to collect the data and hit from 2.9km pretty much dead center. Clip 2 is a night surface attack in beta from 1km and its a narrow miss. Clip 3 is a submerged day attack using the map to back up my distances from 3.6km. This is a miss as well. I close to 2500m to barely hit the rudder of this corvette. All of these tests between the beta and stable were on flower corvettes.

https://youtu.be/k4fjnZwPOJ4

In this short clip I pull of a periscope surface shot from 5km. I use the periscope to help get a higher vantage point of this Isles corvette which is the smallest (I think?) ship in the game, barring fishing boats.

https://youtu.be/7bKFAU7Rg7o

Amusingly enough, the impact seemed to make the corvette stop due to collision. The torp didn't detonate because it wasn't warmed. I also made the mistake of not having the periscope centered which would have probably made for a cleaner hit.

[tequilatibbs]

I dont use salvo (or at least havent yet), however I do know that it was tied up in the very particular way data must be entered.

This video is a explanation of the historical system :
https://youtu.be/1ik1Z5bNdlU?si=8HGeCPiX0j0_7wFb&t=3900
You may find some useful info there.

I caution against assuming bugs if you havent watched this video as 99.9% of peoples "bug" declarations with this tdc turn out to be historical features. Not saying it isnt a bug though.

Other thing is I dont think we can rely on ships to maintain their speed. I chronometered a merchant in perfectly calm weather and each time produced a different result ranging from 5.8-7.7kn. I dont know if this was a bug but obviously it will effect shots, given range and target size.

[Zivs]

Originally posted by tequilatibbs:

I dont use salvo (or at least havent yet), however I do know that it was tied up in the very particular way data must be entered.

This video is a explanation of the historical system :
https://youtu.be/1ik1Z5bNdlU?si=8HGeCPiX0j0_7wFb&t=3900
You may find some useful info there.

I caution against assuming bugs if you havent watched this video as 99.9% of peoples "bug" declarations with this tdc turn out to be historical features. Not saying it isnt a bug though.

Other thing is I dont think we can rely on ships to maintain their speed. I chronometered a merchant in perfectly calm weather and each time produced a different result ranging from 5.8-7.7kn. I dont know if this was a bug but obviously it will effect shots, given range and target size.

Now I will watch this video as I intend to learn as much as I can about the TDC. However, this game is made for arguably all skill levels. In stable you could nail shots simply by having course (now AOB), speed and distance and having those punched in. Good to go, shot lands if data is good. It was even confirmed by ruby that you should never need to touch the TDC if you have no desire to. In the videos I have posted above, I have good data, backed up with map and perfect AOB and the torps are not landing as intended or missing altogether. A shot from 1km misses in beta whereas a 5km shot in stable lands.

There are speed changes and course changes it seems like but a shot from 1km should not miss to un-alerted ships. Very unlikely there is a speed or course change in the 3-5 min of in game time from gathering data to firing.

[tequilatibbs]

Originally posted by Zivs:

Now I will watch this video as I intend to learn as much as I can about the TDC. However, this game is made for arguably all skill levels. In stable you could nail shots simply by having course (now AOB), speed and distance and having those punched in. Good to go, shot lands if data is good. It was even confirmed by ruby that you should never need to touch the TDC if you have no desire to. In the videos I have posted above, I have good data, backed up with map and perfect AOB and the torps are not landing as intended or missing altogether. A shot from 1km misses in beta whereas a 5km shot in stable lands.

The tdc functions exactly the same whether you go fully manual unlocked or fully automated (crew solution) and everything in between.

As I say, I don't have much experience with the salvo setting, I just know two things:

1). You must enter data in a particular order, and the salvo system was for that,
2). The german tdc salvo setting was for hitting a target with one torpedo with imperfect data. So if you have a four shot spread, expect ONE torpedo to hit based on target length.

I assume if your setting up a salvo but only firing one torp in that salvo, itll create a bad solution

[Zivs]

Originally posted by tequilatibbs:

Originally posted by Zivs:

Now I will watch this video as I intend to learn as much as I can about the TDC. However, this game is made for arguably all skill levels. In stable you could nail shots simply by having course (now AOB), speed and distance and having those punched in. Good to go, shot lands if data is good. It was even confirmed by ruby that you should never need to touch the TDC if you have no desire to. In the videos I have posted above, I have good data, backed up with map and perfect AOB and the torps are not landing as intended or missing altogether. A shot from 1km misses in beta whereas a 5km shot in stable lands.

The tdc functions exactly the same whether you go fully manual unlocked or fully automated (crew solution) and everything in between.

As I say, I don't have much experience with the salvo setting, I just know two things:

1). You must enter data in a particular order, and the salvo system was for that,
2). The german tdc salvo setting was for hitting a target with one torpedo with imperfect data. So if you have a four shot spread, expect ONE torpedo to hit based on target length.

I assume if your setting up a salvo but only firing one torp in that salvo, itll create a bad solution

I think you may be misunderstanding me. I am firing one torp and not a salvo. The first video demonstrates that the salvo setting is being used for just one torp. The AI is setting up the salvo and setting it to 8 degrees. I have not changed this or entered this in, it is automatically doing it. I have seen it automatically put in a salvo dispersion of over 10 degrees. This appears to be causing the single torp to deviate due to the automatically inputted salvo dispersion. If you are not babysitting the reciever panel in the bow it will re-input any salvo dispersion that the AI is erroneously wanting.

The comment I made directly under the post is just highlighting the simplicity between stable and beta. The exact same methods are used in both videos on collecting ship data, as I attack ships the same every time with the scenario permitting. Identify, speed, course (now AOB), distance. These videos do not have me messing with the TDC in the beta version. Stable= long range 3+ km shots landing. Beta= 1km shots missing. Exact same methods used.

[ждун]

The most important factors for accuracy is distance and the quality of speed estimation.
The better the speed estimation and the shorter the distance, the more likely you get a good hit.

Now regarding long shots, the quality of speed estimation becomes very important. At long distances only little errors in that lead to large errors at the other end of the solution. If you get significant daviations in speed estimations (+/- 1kn) you would likely get errors of like 300metersleft/right when firing from longer distances. So why there are sometimes such inaccuracies?

I have an idea.

Taking accurate speed measurement requires you to be not moving and not turning.
Perhaps some commanders don't know that, and measure speed when they are moving getting inacurate speed measurements and wondering why they are missing.
But even if you are not moving, speed estimation may become inacurate, because the boat rocks in the waves and it slightly changes its orientation. You see it clearly when time compression is on. This small changes of the boats course, may cause inaccurate speed measurements even when sitting still with stopped engines.

[Zivs]

Originally posted by ждун:

The most important factors for accuracy is distance and the quality of speed estimation.
The better the speed estimation and the shorter the distance, the more likely you get a good hit.

Now regarding long shots, the quality of speed estimation becomes very important. At long distances only little errors in that lead to large errors at the other end of the solution. If you get significant daviations in speed estimations (+/- 1kn) you would likely get errors of like 300metersleft/right when firing from longer distances. So why there are sometimes such inaccuracies?

I have an idea.

Taking accurate speed measurement requires you to be not moving and not turning.
Perhaps some commanders don't know that, and measure speed when they are moving getting inacurate speed measurements and wondering why they are missing.
But even if you are not moving, speed estimation may become inacurate, because the boat rocks in the waves and it slightly changes its orientation. You see it clearly when time compression is on. This small changes of the boats course, may cause inaccurate speed measurements even when sitting still with stopped engines.

I thought that may be the issue as well too. The 1km night shot I keep referring to from my video was done at 5 knots speed. Just enough to keep the boat straight. The boat can yaw substantially in rougher seas if completely stopped.

2:14 timestamp
https://youtu.be/k4fjnZwPOJ4

[tequilatibbs]

Originally posted by Zivs:

I think you may be misunderstanding me. I am firing one torp and not a salvo. The first video demonstrates that the salvo setting is being used for just one torp. The AI is setting up the salvo and setting it to 8 degrees. I have not changed this or entered this in, it is automatically doing it. I have seen it automatically put in a salvo dispersion of over 10 degrees. This appears to be causing the single torp to deviate due to the automatically inputted salvo dispersion. If you are not babysitting the reciever panel in the bow it will re-input any salvo dispersion that the AI is erroneously wanting.

Ok, my apologies. Last thing, do you manually set the control box to "tube 1" - no salvo?

I always check this box is at the settings i want.

If you want everything done automatically then it should all be set to off

[Zivs]

Originally posted by tequilatibbs:

Originally posted by Zivs:

I think you may be misunderstanding me. I am firing one torp and not a salvo. The first video demonstrates that the salvo setting is being used for just one torp. The AI is setting up the salvo and setting it to 8 degrees. I have not changed this or entered this in, it is automatically doing it. I have seen it automatically put in a salvo dispersion of over 10 degrees. This appears to be causing the single torp to deviate due to the automatically inputted salvo dispersion. If you are not babysitting the reciever panel in the bow it will re-input any salvo dispersion that the AI is erroneously wanting.

Ok, my apologies. Last thing, do you manually set the control box to "tube 1" - no salvo?

I always check this box is at the settings i want.

If you want everything done automatically then it should all be set to off

It is indeed set to off which should be the 'auto' position.

And just for anyone elses information who may be reading this thread, I am not throwing my arms up in the air yelling BUG BUG BUG! I am attempting to get to the root of this cause to give some helpful feedback to the devs. In my eyes, things were working as intended, if not a little to well in the stable version. Now from what I can see there is some sort of error. It is my intention to help get this fixed, because as it stands, this is not friendly to a beginner or casual player who does not wish to crank knobs. I myself am mostly a casually obsessive player, who likes elements of realism but does not wish to do the dreaded maths, calculations, or overly complicated strategies. I enjoy the TDC in the game because its cool to look at and figure out how it works. The options available are pretty damn cool! Does not mean I want to crank it for every shot.

As also mentioned feel free to add me on discord to figure this out!

Discord: zivs.

[tequilatibbs]

This is how im understanding it:

Settings in control box to off = auto
you manually set a gyro angle at the gyro box
you only fire one torpedo
Your torp misses because of the gyro angle you set

Is that correct? (apologies if im getting the wrong end of the stick)

[Zivs]

Originally posted by tequilatibbs:

This is how im understanding it:

Settings in control box to off = auto
you manually set a gyro angle at the gyro box
you only fire one torpedo
Your torp misses because of the gyro angle you set

Is that correct? (apologies if im getting the wrong end of the stick)

Negative. In auto/off the salvo spread is set to 8 degrees (single torp still). Torp follows the auto salvo dispersion which should not be there in the first place. Manually setting salvo degrees to 0 seems to rectify the issue a little more but there are still inaccuracies as opposed to stable.

[tequilatibbs]

Originally posted by Zivs:

Negative. In auto/off the salvo spread is set to 8 degrees (single torp still). Torp follows the auto salvo dispersion which should not be there in the first place. Manually setting salvo degrees to 0 seems to rectify the issue a little more but there are still inaccuracies as opposed to stable.

Ok, hope the issue is resolved for you

[Urfisch]

Interesting thread. But the thread title is misleading. From what I know, the tdc is as accurate as the original device :) so, all errors are likely human here. If you find an evidence and reproducible error, please send a report of your gamestate using F11 in the game.

[Zivs]

Originally posted by FreekOly (aka Urfisch):

Interesting thread. But the thread title is misleading. From what I know, the tdc is as accurate as the original device :) so, all errors are likely human here. If you find an evidence and reproducible error, please send a report of your gamestate using F11 in the game.

I will gladly accept user error as the cause, but up until this point no one has been able to correct whatever my error is. I'm not being combative, just looking for correction. I have found one issue, in regards to the salvo bug that seems to be a partial factor.

In regards to the game state, can the game states be re-winded? The devs would have to re-watch my games and my torp misses.

Keep in mind the TDC is not perfect yet. I have gyro angles (set by the AI/game) set to -1501 degrees. Currently I am looking at a gyro angle of 428.2 degrees on the receiver panel, while the gyro on the TDC is saying 68. The maths check out for what it is trying to convey, but regardless there is no such thing as 428 degrees or -1501 degrees. Whether this bug is something deeper in the TDC modeling or just a superficial "lol 428 is actually 68, same diff", I don't know.

[Battleshipfree99]

AFAIK, OP is suggesting a default salvo angle of 0 so players won't launch a wide spread accidentally? Then it's not a BUG, nor "inaccuracy", it's merely a suggestion for qualify of life improvement.