Login

russian armor

TDs vs mediums in Live and in numbers.

31 Jul 2017, 14:18 PM
#21
avatar of vietnamabc

Posts: 1063

DPS for tanks would only make sense if the state machine worked as follows:
1. Enemy spotted
2. Reload shot
3. Fire
4. Go to 2

Instead, it works like:
1. Enemy spotted
2. Fire
3. Reload
4. Go to 2

Basically, DPS completely ignores alpha damage, which is the most crucial part of the equation.

Moreover, DPS completely masks the effects of what happens when you increase the amount of tanks dealing damage. When you have 2 tanks opening up at once, the DPS doesn't simply just double; it goes even beyond that.

For instance, it takes an AT gun 4 shots to kill a T34. That means 3 reload cycles. If you put 2 AT guns on the work, that's still 4 shots, but that takes only one 1 reload cycle (that occurs in parallel between the 2 AT guns). That's 3 times faster, and a DPS calculation would completely ignore that.

Alpha damage is what allows Panther spam to be amazing despite the fact that singular Panthers are shit.

You can get away with DPS calculations for infantry, simply because the reload time is small, and the duration of combat is usually long (especially when cover is involved).

Yup I'll take alpha over DPS any days, which is why Jpz IV and Firefly is the most common TDs, followed by Stugs and M10.
My priority: accuracy>>alpha>>>cost>>>DPS. This is why Jpz IV and Firefly is the most feared TD, Jpz because of the awesome camo shot and Firefly due to lol alpha strike.

31 Jul 2017, 21:17 PM
#22
avatar of Vipper

Posts: 13496 | Subs: 1

M4A2 Sherman Firefly

• 17 pounder accuracy from 0.08/0.07/0.05 to 0.055/0.045/0.4(!) typo probably meant 0.04
• Moving accuracy from 0.75 to 0.5
• Tulips no longer cancels move commands upon stun.
Tulips now deal 33% damage versus infantry.


the effect is chance to damage:



sherman_firefly_76mm_mp 1,76 1,71 1,66 1,61 1,56 1,39 1,10
sherman_firefly_76mm_mp 1,33 1,28 1,22 1,17 1,12 1,09 0,97


and the effect in DPS (tullips no included here):


sherman_firefly_76mm_mp 19,39 19,39 19,39 19,39 19,39 19,39 19,39
sherman_firefly_76mm_mp patch 19,39 19,39 19,39 19,39 19,39 19,39 18,77


Note that the changes have almost no affect in chance to damage when unit stationary due to x1.1 accuracy from commander and the long range of "mid range". The unit still piratically has around 100% to damage a medium tank at range 60 when stationary. The vet 3 accuracy buff also has almost no effect in the units performance when stationary.


Changes mostly affect the accuracy on move.
2 Aug 2017, 14:06 PM
#23
avatar of Vipper

Posts: 13496 | Subs: 1

Comparison of the Elephant prepatch post patch.

Chance to damage (hit and penetrate):

elefant_88mm_gun_mp 1,10 1,02 0,95 0,87 0,80 0,71 0,63 0,55



Modified DPS:


elefant_88mm_gun_mp 36,57 36,57 34,48 31,61 28,73 25,86 22,99 20,11
elefant_88mm_gun_mp patch 32,00 32,00 30,17 27,66 25,14 22,63 20,11 17,6


Note that Su-76 by vet 2 has more DPS at range 60 than the Elephant in the same range.
2 Aug 2017, 18:09 PM
#24
avatar of elchino7
Senior Moderator Badge

Posts: 8154 | Subs: 2

Since you are taking the time to do this charts, wouldn't it be better to replace DPS for TTK ?

Not sure where you are taking your data from, but from an easier/faster POV just take the whole RoF cycle and multiply it based on the number of shots required to kill and then a secondary based on the "expected" amount of shots required due to scatter rounded up. So if it's 3.5, it would be 4.

Again, DPS doesn't tell me anything. Take for example the elefant stats (live). You'll expect that at min range, the Elefant would take 640HP/36.57DPS (17.5s) to kill a tank which in fact should be half due to alpha damage/reload cycle.

Now compare what it's easier to analyse:

Elefant kills medium 640HP tanks at close range on: 8.75s of been spotted or 36.57DPS.

If you ask me i would plot something like:
A)ROF B)Number of shots required to kill C)Accuracy D)Round: B/C value

Then you have:
(A*B) - A = Ideal TTK
(A*B/C) - A = Avg TTK based on acc
(A*D) - A = Realistic Avg TTK based on acc

First is the potential with no miss chances. Second one takes into account fraction number of shots opportunities. Third one is more of a nitpick cause you can't really have "half" shots. If >.5 then it means on average you are requiring an extra shot.

AND this is only taking into account a 1v1 case scenario. Whenever you add more vehicles, as Smith explained, you have more issues.
2 Aug 2017, 18:29 PM
#25
avatar of Esxile

Posts: 3602 | Subs: 1

jump backJump back to quoted post2 Aug 2017, 14:06 PMVipper
Comparison of the Elephant prepatch post patch.

Chance to damage (hit and penetrate):

elefant_88mm_gun_mp 1,10 1,02 0,95 0,87 0,80 0,71 0,63 0,55



Modified DPS:


elefant_88mm_gun_mp 36,57 36,57 34,48 31,61 28,73 25,86 22,99 20,11
elefant_88mm_gun_mp patch 32,00 32,00 30,17 27,66 25,14 22,63 20,11 17,6


Note that Su-76 by vet 2 has more DPS at range 60 than the Elephant in the same range.


Are you sincerely comparaing a SU-76 with an Elefant?
Put the 2 at 60 ange firing at each other and tell us who win in 100% of the cases.
2 Aug 2017, 18:49 PM
#26
avatar of Vipper

Posts: 13496 | Subs: 1

Since you are taking the time to do this charts, wouldn't it be better to replace DPS for TTK ?


I find chance to damage which I include more interesting than DPS.

When I do used DPS I am usually comparing prepatch with patch for the same gun which mean that extra reload circle has less impact in estimating the effects of the change.

jump backJump back to quoted post2 Aug 2017, 18:29 PMEsxile


Are you sincerely comparaing a SU-76 with an Elefant?
Put the 2 at 60 ange firing at each other and tell us who win in 100% of the cases.

If that is what you understand from the sentence:
"Note that Su-76 by vet 2 has more DPS at range 60 than the Elephant in the same range."
I am sorry but I can not really help you.
2 Aug 2017, 19:02 PM
#27
avatar of Mr.Smith

Posts: 2636 | Subs: 17

Here's a thread where you can get the stats for time to kill that you get 50% of the time (also taking variance and scatter shots into account).

There has to be something seriously wrong about your calculations if you get that SU76 has better DPS than an Elefant.

https://www.coh2.org/topic/56469/tank-destroyer-time-to-kill-stats
2 Aug 2017, 19:13 PM
#28
avatar of Esxile

Posts: 3602 | Subs: 1

jump backJump back to quoted post2 Aug 2017, 18:49 PMVipper

If that is what you understand from the sentence:
"Note that Su-76 by vet 2 has more DPS at range 60 than the Elephant in the same range."
I am sorry but I can not really help you.


It's your Ostheer's fanboyism that can't be helped. TD vs mediums in Live and in numbers but forgetting half of them.

What do you want to proof with your numbers if you don't even take the armor of each unit in consideration. What the purpose of saying a SU-76 vet2 deal more raw damage than a Elefant (true or not). If the idea is using them to shot cans on your backyard and see which get the biggest hole, yes maybe it as a value. In Coh2 sorry but it doesn't.
2 Aug 2017, 21:22 PM
#29
avatar of LoopDloop

Posts: 3053

A couple flaws about TD's "DPS": a lot of TDs are used like snipers, shooting once then pulling back until they are sure to be reloaded for a variety of reasons, and penetration isn't guaranteed on all targets/with all weapons. SU-76's chance to pen a kt, jt/ele, even panther, is drastically different than a JTs chance to pen a cromwell or E8. Similarly, a stug's chance to pen a IS-2 is much different than a firefly's chance to pen a P4.

Mediums also aren't the only tanks on the field. Ele and Jt aren't meant to counter shermans, and SU-76 isn't meant to counter Panthers or KTs
2 Aug 2017, 21:55 PM
#30
avatar of JohnSmith

Posts: 1273

I think we all tried to reason with Vipper and we all failed to break through. If she must continue to do biased, mathematically and statistically incorrect charts and continue ignoring any input, we should just let her do it..
2 Aug 2017, 23:43 PM
#31
avatar of LoopDloop

Posts: 3053

I just played a game against a JT and DPS doesn't mean shit to that thing, since it's alpha does half of any tank usf has access to. DPS doesn't take into account range or health and armor either, which it has plenty of. And people bitch about the new jackson with its 600 hp and 60 range lol.
3 Aug 2017, 06:28 AM
#32
avatar of Vipper

Posts: 13496 | Subs: 1

Here's a thread where you can get the stats for time to kill that you get 50% of the time (also taking variance and scatter shots into account).

There has to be something seriously wrong about your calculations if you get that SU76 has better DPS than an Elefant.

https://www.coh2.org/topic/56469/tank-destroyer-time-to-kill-stats

You probably need to have closer look to what I wrote:

A VET 2 SU-76 has more DPS (at range 60)vs medium tanks than Elephant post patch has at the same range.

If you think my number are wrong feel free to do your own calculations.
3 Aug 2017, 06:37 AM
#33
avatar of Vipper

Posts: 13496 | Subs: 1

jump backJump back to quoted post2 Aug 2017, 19:13 PMEsxile


It's your Ostheer's fanboyism that can't be helped. TD vs mediums in Live and in numbers but forgetting half of them.

What do you want to proof with your numbers if you don't even take the armor of each unit in consideration. What the purpose of saying a SU-76 vet2 deal more raw damage than a Elefant (true or not). If the idea is using them to shot cans on your backyard and see which get the biggest hole, yes maybe it as a value. In Coh2 sorry but it doesn't.


The TD that are not here are Puma and M10 because they are "flanker" tds.

You can interpret the sentence as you wish, I simply made a note.

The armor of each unit is of less importance than you think if this units can snipe medium tanks 20 units away from the max range of medium tanks. If in your opinion armor is so relevant pls feel free to include them in this thread.
3 Aug 2017, 06:46 AM
#34
avatar of Vipper

Posts: 13496 | Subs: 1

A couple flaws about TD's "DPS": a lot of TDs are used like snipers, shooting once then pulling back until they are sure to be reloaded for a variety of reasons, and penetration isn't guaranteed on all targets/with all weapons. SU-76's chance to pen a kt, jt/ele, even panther, is drastically different than a JTs chance to pen a cromwell or E8. Similarly, a stug's chance to pen a IS-2 is much different than a firefly's chance to pen a P4.

Mediums also aren't the only tanks on the field. Ele and Jt aren't meant to counter shermans, and SU-76 isn't meant to counter Panthers or KTs

Read closer, this thread is about TDs vs mediums and it uses Ostheer PZIV and Cromwell as base reference.

Elephant and JT, are according to patch-notes, being nerfed because they are too effective vs mediums and TDs.
3 Aug 2017, 07:43 AM
#35
avatar of LoopDloop

Posts: 3053

jump backJump back to quoted post3 Aug 2017, 06:46 AMVipper

Read closer, this thread is about TDs vs mediums and it uses Ostheer PZIV and Cromwell as base reference.

Elephant and JT, are according to patch-notes, being nerfed because they are too effective vs mediums and TDs.

I understand that, I'm just pointing out that it's silly to only talk about dps against mediums.
3 Aug 2017, 08:36 AM
#36
avatar of Vipper

Posts: 13496 | Subs: 1


I understand that, I'm just pointing out that it's silly to only talk about dps against mediums.

Not if the TDs performance vs mediums is a big part of the scope:
Elephant's, Jt's, Fireflies's, M36's have been changed with their performance vs medium in mind.
Stug's and JP changes also has affected their performance vs mediums.

This thread has highlighted the strong indication that increasing accuracy for m36 (in version 1) was in the wrong direction and that the new accuracy values in ver 1.1 where not fixing the issue.

It has also highlighted that the changes to firefly will have very little impact vs mediums.
3 Aug 2017, 16:15 PM
#37
avatar of JohnSmith

Posts: 1273

You really should read up on statistical analysis. Have a look at t-tests, right now your calculations are wrong which leads to wrong conclusions. I told you that earlier in the thread. Calculating percentage difference will only reinforce your own biased hypothesis. DPS calculations also ignore a large part of the game, as others explained earlier.
3 Aug 2017, 16:19 PM
#38
avatar of Vipper

Posts: 13496 | Subs: 1

You really should read up on statistical analysis. Have a look at t-tests, right now your calculations are wrong which leads to wrong conclusions. I told you that earlier in the thread. Calculating percentage difference will only reinforce your own biased hypothesis. DPS calculations also ignore a large part of the game, as others explained earlier.

Feel free to provide the correct calculation and your conclusions and even your suggestions.

Now I have responded to you for the next couple of days, so bb once more.
3 Aug 2017, 16:21 PM
#39
avatar of JohnSmith

Posts: 1273

I gave you an example of a t-test and conclusion earlier in the thread, and even contacted you privately. You can even run a t-tests on the data you provided using the info I sent, and I did one for the first examples you provided. Just read up on t-test, they show exactly why percentage difference calculation for your type if data is a wrong approach. Alternatively you can Google it. Even MrSmith gave you a thread to build your work on.

No need to be rude 'mate if you disagree with people who want to help. We already know that you're biased, so there's no need to demonstrate more of that either :p
2 users are browsing this thread: 2 guests

Ladders Top 10

  • #
    Steam Alias
    W
    L
    %
    Streak
Data provided by Relic Relic Entertainment

Replay highlight

VS
  • U.S. Forces flag cblanco ★
  • The British Forces flag 보드카 중대
  • Oberkommando West flag VonManteuffel
  • Ostheer flag Heartless Jäger
uploaded by XXxxHeartlessxxXX

Board Info

298 users are online: 298 guests
5 posts in the last 24h
17 posts in the last week
29 posts in the last month
Registered members: 49921
Welcome our newest member, Tiley47430
Most online: 2043 users on 29 Oct 2023, 01:04 AM