Jump to content

901234

Members
  • Content Сount

    205
  • Joined

  • Last visited

  • Battles

    9582
  • Clan

    [O7]

Community Reputation

66 Good

About 901234

  • Rank
    Master Chief Petty Officer
  • Insignia

Recent Profile Visitors

1,541 profile views
  1. 901234

    -7% dispersion vs Sigma?

    Reverse engineered code shows that wows dispersion distribution is a truncated normal distribution although the method used to generate it is close to what you describe. Vertical dispersion also depends on other parameters such as: delim, zero radius, delim radius, and max radius [which can only be seen through datamining] and unexpectedly the maximum range of the guns. For more details read this thread:
  2. 901234

    Grosser Kurfürst being replaced

    The only differences between Iowa and FdG vertical dispersion are a result of range and shell ballistic differences. Both of which become less apparent if you use the "impact normal" projection which controls for the effect of ballistics. Having floatier shells probably does not contribute to the accuracy problem nearly as much as you think it does. Factors that are likely more important: USN BBs have better sigma compared to the less accurate BBs that were mentioned T9+ USN BBs have an upgrade that reduces horizontal dispersion by 11% [which also factors into vertical dispersion] USN BBs can have very long firing ranges [which slightly improves vertical dispersion] Higher tier [8+] USN BBs tend to have good shell volume which may anecdotally appear to have better dispersion since you will land more hits In addition, do note that ballistic [or angle of fall] effects on vertical dispersion need to take into account the type of target you are hitting. If the target is far wider than it is tall, then [all else equal] floatier shells will perform better. However, if the target is taller than it is wide [e.g. citadels if deck penetrations are not possible] then [all else equal] shells with a flatter trajectory will perform better. Additional Note: Ships like Kronshtadt and Roma also have worse base dispersion compared to their counterparts [which will also worsen vertical dispersion]
  3. 901234

    External Ballistics

    Is this referring to shell flight behavior in wows?
  4. 901234

    Another Russian Bias Rant

    By the way, vertical dispersion is dependent on max range. So for a more fair comparison [unless you want to compare against stock Vladivostok] you should go into ship modules and select the upgraded GFCS module for Vladivostok. If you apply the upgraded GFCS, the dispersion difference narrows considerably.
  5. 901234

    fletcher main gun range

    It probably should be made visible. However given that it only applies to very few ships and that you generally don't want to run double range on most ships means that it probably won't be addressed any time soon - even though fixing it probably isn't that hard.
  6. 901234

    fletcher main gun range

    It's probably for balancing reasons - even though all have the same shells and thus the same maximum shell ranges
  7. 901234

    fletcher main gun range

    This is from reverse engineering the wows ballistics model. The other sources probably just applied the multipliers without accounting for the theoretical max range. I made a post regarding that here: Link to website used to generate charts: https://jcw780.github.io/wows_ballistics/ Maximum shell ranges for Smolensk (AP) and Harugumo (AP) of ~28km and ~19.5km(I have an image associated w/ this but I can't upload it for some reason) You can replicate this by going to the linked website, picking a ship, selecting the shell of choice (this can make a difference depending on the ship), then go into settings and set the maximum elevation to 90 degrees. Then look for the maximum range that the curves reach - this is roughly the maximum range (probably ~.1km accuracy)
  8. 901234

    fletcher main gun range

    This is because the Fletcher HE shells' physical max range is 16.3km. The game caps out the range to maximum which can be achieved by shells. This is rarely met as the shells max range is usually much higher than the achievable gun range, however you have found one of the few exceptions. Another example is Yamato's HE. [Edit: Physical/Shell max range refers to the highest range the shell can achieve at optimal elevation - also note that Fletcher's guns in game can elevate up to 85 degrees so it is not constrained by that (keep in mind that optimal elevation is around 40-50 degrees)]
  9. Autobounce: identical [start] 50, [always] 65 degrees Fuse Time: identical 0.022s Material [assuming it's fusing threshold]: identical 37mm Krupp: Riga: 2800; Petropavlovsk: 2950 Though for practical purposes, I included a penetration chart. Edit: Other differences: Mass: Riga 170kg; Petropavlovsk 207kg
  10. 901234

    Champagne Shells - What am I doing wrong?

    I would assume that at those ranges citadel overpenetration becomes less of a problem. In addition, as Champagne shells have shallow arcs - even at those ranges - and even for maximizing hit chance [as opposed going for citadels] you should probably still try to hit the sides.
  11. 901234

    Champagne Shells - What am I doing wrong?

    Aiming in the water would work, however the aiming area where you can get pens / citadels is not very large [especially at close range] so it might be a bit of a gamble.
  12. 901234

    Champagne Shells - What am I doing wrong?

    If you were doing fine with French 380's then roughly 3-4km further would be needed. Reference target is a Chapaev [100mm belt w/ ~19m beam] The conclusion here isn't wrong but technically the penetration itself is not why shells are more likely to over-penetrate. From what others have gathered, post-penetration velocity decreases less for effective armor thicknesses that are significantly lower than the shell's penetration. As battleship shells tend to have much higher penetration than the thicknesses cruiser belts at that tier, the main difference is caused by the fact that Champagne's shell impact velocity is much higher compared to even the French 380's [Richelieu, Jean Bart etc.] which are already faster than a lot of the shells of BBs at that tier.
  13. Testing with those is meant to identify if maxDist has an effect w/in ship lines or between ships with the same shells but different ranges and not for comparisons between ship lines (or otherwise for identifying the behavior of the "V/H ratio function"). Also the new RU heavy cruiser line is suspected within datamining circles to have different vertical dispersion even if most players do not notice/complain about it - partially because cruisers in general tend to have much better dispersion compared to BBs anyways. In addition, I forgot to mention that the Slava (not surprising to anyone) does also have [if the hypothesis is correct] incredible vertical dispersion parameters: radiusOnZero radiusOnDelim radiusOnMax delim Slava 0.1 0.25 0.45 0.3 That sin isn't a vertical to horizontal ratio, it's more like the perpendicular [of the impact trajectory - which equals the scaled horizontal] to vertical in this hypothesis. Apologies for the lack of clarification.
  14. To be honest, I thought that the reference range was ideal distance [30km for main battery] - There is no way main battery max range affects vertical right? However, I heard from some people that reverse engineered in game functions that take in these parameters use maxDist instead. Also, effect of range modifiers is uncertain so for all we know it's the same for each ship regardless of upgrades. Also the differences aren't too significant for those ships so you would likely need extensive testing to ascertain a statistically significant difference. In all likelihood, it makes a lot more sense to test ships w/ faster rof with higher predicted differences to get a good sample size faster instead of something like a BB. My approximation would roughly be like: ratio(distance) * horizontal / sin(impact angle) = vertical so to find the ratio it would be: ratio(distance) = vertical * sin(impact angle) / horizontal where ratio(distance) is function noted in my previous comment
  15. There is a hypothesis w/ some data [though not high enough resolution to know for sure] and some outputs from reverse engineered functions backing it within that the following parameters [found within gameparams.data]: radiusOnZero, radiusOnDelim, radiusOnMax, delim, and maxDist are used to form a piecewise linear function that produces the vertical to horizontal ratio. The hypothetical function is something along the lines of: when: 0 <= dist <= delim * maxDist: scalar(dist) = (radiusOnDelim - radiusOnMax) / (delim * maxDist) * dist + radiusOnZero when: delim * maxDist <= dist <= maxDist: scalar(dist) = (radiusOnMax - radiusOnDelim) / (maxDist * (1 - delim)) * (dist - delim * maxDist) + radiusOnDelim when: maxDist < dist scalar(dist) = radiusOnMax Note: maxDist is standard max range These values [and hypothetical function] do corroborate your finding that IJN ships have higher vertical dispersion radiusOnZero radiusOnDelim radiusOnMax delim Normal 0.2 0.5 0.6 0.5 IJN BB 0.2 0.6 0.8 0.5 RU BB 0.25 0.4 0.75 0.6 RU CA [new] 0.25 0.4 0.95 0.6 However, briefly looking through the analysis I am not certain that the tan approximation lines up with predictions. Can you also do one that computes the 'vertical' component by multiplying by sin instead because I also suspect that the actual dispersion is calculated by modifying angle of launch instead of computing point of impact on an aiming plane. Also, for all those who are willing to collect / analyze replay data regarding this, I have been informed that @JustDodge - along with others who I believe are from other servers - has made a replay parser that shows shell landing coordinates. Edit: this is the parser https://github.com/Monstrofil/replays_unpack
×