ChrisWerb

Well, I've obviously been hiding under a rock for the last decade (sadly, not far from the truth!). I wasn't aware there was so much out there.

OK, I did some more testing. 

Five three tank platoons of Russian tanks from the T-62 to the Armata. All ammo removed (even MG and smoke discharger grenades). Shooter is a Leo2A6 with DM53. Shot placement to the lower hull aligned with the centre of the turret to hit the carousel (except for the T62 when I aimed into the front hull from the side due to its different ammo storage arrangements). I did this test three times and got interesting results each time. I'm posting the AAR from the last one. All tanks either blew up, roman candled or burned (of which in earlier tests some subsequently exploded), (though sometimes this took more than one shot) except one Armata which would not ignite regardless of where in the carousel area I hit it. This time I'm including the scenario so you can repeat the test if you wish to.

Disclaimer: This post is not intended as a criticism of or harrassment of the SB team or any member thereof, or in any way a statement of disastisfaction with SB. 

Exploding_ammoless_vehicles_4832_110219HP-Z4401308.aar Exploding ammoless vehicles.sce_4_11-02-19_13_08_17.htm Exploding ammoless vehicles.sce

OK, if we're talking Armata, if aiming frontally, I find aiming for the gun with sabot will often result in gun disablement regardless of Afganit intercept. From the front quarter, the crew module and ammo carousel both become vulnerable, although the latter has less protection. From the side, if not hull down, my first choice is the ammo carousel, followed by the crew compartment. shots into frontal portion of the turret, parallel or just below the level of the main gun will almost always take out the gun. However, if you are close enough, a hit into the turret bustle, is extremely likely to take out the turret drive, stabilisation, GAS, and fire control systems, plus other electronics. This produced identical results for me six out of six times. As you would expect, the vehicle is very vulnerable from the rear - again my first choice would be ammo carousel - preferably through that into the crew compartment. If hull down, turret bustle again. From directly behind, turret bustle through the autoloader into the gun breech. I just ran an ambush scenario in a German valley and killed ten Armatas without loss with six Leo2A6s, but it took a lot of trial and error (mostly the latter) to achieve that result. It was a lot of fun too!

When I'm pitching SB to friends, a huge selling point, to me, is that SB is an actual military training system that is available in a very slightly attenuated version to civilians. You could argue that Harpoon/CMANO/CMO is the opposite, having been adopted by militaries subsequent to its civilian inception. What I'd like to know is are there any other simulations out there with real civilian/military crossover in usage? The only one I can think of might be TACOPS. I'm sure you can come up with other examples.

On 10/30/2019 at 5:39 AM, Volcano said:

One thing we already do have is that certain ammo storage areas are designated as 'stored ammo' (ie. not ready ammo). Once stored ammo is all used up, then those areas are ignored (they will not trigger explosions anymore). So in that respect, we do model a real difference between stored and ready ammo in regards to vulnerability.

 

Also, if you want to have a less vulnerable T-72B1, or less vulnerable Leopard 2, you can choose to remove all the stored maingun ammo, which is something that is considered in real life situations. (Or at least that is how the feature is supposed to work anyway.)

 

Just saying.

I've definitely had vehicles (T-90s and Armatas) from which I have removed all ammo explode on many occasions during testing (almost always, given the correct shot placement). I'm not asking for any changes to how damage is modelled, just giving my findings. Those AARs I posted on the earlier thread had all vehicles devoid of ammunition (to stop them shooting back, without disabling the gun  ) and plenty of those blew up.

@Volcano

I have read and re-read your posts and really do appreciate your explanations. I fully accept that you have applied the damage model that exists with SB diligently.  I can see an additional problem from your point of view. Let's say there are critical components in the centre of the T-14 other than in the key systems indicated in your diagram - if you fixed that for the T-14 (which would take military espionage at this stage) you would have to do the same for every vehicle in the game and that's just not do-able and would not be a profitable use of your time. There are other anomalies that are not worth fixing. For example, quite early on I realised I was getting ammo explosions in target vehicles from which I had removed all ammunition. That started me thinking about how that could be overcome and I again realised that fixing that would be a whole world of work for someone as the likelihood of an explosion, its magnitude and subsequent effects would depend on the number and type of each round and where it was stored at the moment the vehicle was hit - that would take a lot of modelling for no real training benefit. It's obviously very easy to get into diminishing returns in simulation.

I did not know we did not have a model for spalling and look forward to that being added at some point in the future. On another positive note, you will remember that damage to helicopters was a bit of a bugbear of mine. Up to recently, you could hit even unarmoured helicopters over 100 times with 20mm AP-T or APDS-T and they would be unaffected - KETF had very little effect on them. That damage model was rectified in spades and shooting at helicopters is now very realistic and satisfying. So, I hope, at some point in the future, the new explosive/fragmentation model can be adapted and give more intuitively likely results with lateral centre mass hits on the Armata. Thank you for your time in looking into this - it was much appreciated!

If you review my tests, I spread the shots around a bit (added to ballistic dispersal). I also fired one round per target vehicle, so cumulative damage or shots going through the same hole repeatedly did not figure in my tests, nor would it in real life  The customer, reviewing hits, would at some point notice that a high proportion, aimed for centre mass from the side, did no damage. I noticed it in game play as did others. I just decided to try to work out what was happening by doing some testing.

Also, when considering how full up the T14 turret is, remember that schematics online don't show wiring, junction boxes and a lot of miscellaneous things. Again, the sabot is not a drill - it doesn't have a perfectly linear effect but also a cone of displaced metal fragments travelling at very high speed and dispersing from its point of penetration. 

Volcano, I LOVE the T-14 and all the new Russian vehicles in Steel Beasts. I just want to see realistic outcomes. That's all.

OK, I tried to do a 100 shot test, but for some reason I still ran out of ammo on unlimited at 48 shot. This is M829A3 from the side at around 1400-1500 metres, mostly into the turret, but I tried to vary my aim a bit. Afganit was turned off.

48 hits, of which two grazed, 30 produced no apparent damage of which two were grazing shots at 02:28 and 10.31. So lets say 28 good penetrating hits of which none caused damage or 58.33%.

Now think of it this way. I would guess the T14 was introduced mainly to offer military customers something interesting and state of the art in the inventory of their most likely peer OPFOR to shoot at. If you were an instructor, how would you explain to your trainee that well over half of penetrating hits to the turret and mid section of the hull, fired from the flank, had no effect whatsoever on the target? 

Armata test 1500m side on M829A3 Afganit off.sce_10_10-29-19_18_03_07.htm Armata_test_1500m_side_on_M829A3_Afganit_off_10056_102919HP-Z4401803.aar

Sorry Volcano, I don't like to argue, but I think you're wrong. Nils can see something odd going on and I can't understand why you cant. c. 20% (17% in one test and 24%, admittedly with a caveat in another) of rounds going through the centre of the tank are doing no damage at all. if you look at available schematics of the T-14 and apply your knowledge of other tanks, you will see that there is a lot going on in that turret which has intentionally been kept as small as possible to present as small a target as possible. There is a massive breech, training and elevating gear, autoloader components and the hull penetrating parts of optronics, plus the wiring for the turret and the RWS and the hull penetrating part thereof. The upper part of the ammo carousel is there too in one picture. It's a lot of stuff in a confined space. As you know and APFSDS projectile is not a drill that neatly bores a hole each side and there is a limited amount a spall liner can do against a massive lump of tungsten or DU crashing through relatively thick (compared to APC) armour and taking quite a bit with it at very high speed.

The test I did vs the T-90 got realistic results - the T-90s suffered far higher crew casualties, but actually blew up much less, due to the ammo storage arrangements and shot placement relative to both - both realistic results and full credit to the SB team for accurate modelling.

PS. In the last test I didn't notice that I'd gotten the Afganit removed on some vehicles. That doesn't invalidate the fact that penetrators are going through T-14s and not causing damage, but it does invalidate the relative one shot kill numbers between the T-14 and T-90, although I suspect, not by much.

One thing that would be neat would be the ability to tell AI gunners in your platoon to prioritise their aim to centre mass (default), turret, or hull. That would help in combating Armata. It's basically a heavyweight fighter with a glass jaw, as Ssnake already (sort of) pointed out.

11 minutes ago, Ssnake said:

We will certainly have a critical view at the component damage likelihoods in the Armata. I'm not saying that everything is perfectly fine as it is. 24% of shots passing through the vehicle and not causing any damage at all is not exactly what I'd like to see either. But we have to ask all these questions to make sure that there really is something that needs to be done. Wouldn't be the first time that someone is frustrated by a single game session and then starts blaming game design rather than accepting bad luck.

I'll happily own up to being that person. 

1 hour ago, Raven434th said:

This is why I questioned their damage model...that would IMO have been a damaging shot,and I see ALOT of these when engaging the armata.Now to be clear...THAT reported a "no damage" shot correct?

I'd have to go back through all eight of the tests I posted to confirm that one instance. You can see lots of them. For the last T-14 test above (the one with 50 target vehicles) open the AAR I uploaded in game and the report in the same post. Look at the report and note the timestamps of those hits that only have mobility damage (which was because I had already disabled the tracks) then set the AAR to real world and events and keep clicking until you hit those time stamps. There are 12 examples in that one test, which is 24% of all hits obtained (two of which are are explicable as peripheral).

32 minutes ago, Assassin 7 said:

Correct, but there is no damage at all being recorded even though there are several shots like the ones I posted above

Yes, and in a comparatively high instance of cases. I think it was 5 ex 30 hits in one case and 12 ex 50 in another. I believe that's statistically significant and probably also significant that it happens with the T-14 but not with the T-90. However, I'll happily run the tests again if more data is needed. There is something peculiar happening, but, as Ssnake said, it's down to the damage model they use and it can't be tweaked without invoking the Law of Unintended Consequences - I'm still interested to hear why it differs for the two tanks though.

That's one reason why I said I didn't think Afganit made a lot of difference from the side. It may well have frontally. 

Basically, that was an interesting exercise, but it comes down to what Ssnake said (although it's rather more than a handful and strange (to me) that this problem exists with the T-14 but not the T-90).

"That a handful of isolated rounds don't cause any damage - well, that's also to be expected if you have an energy based non-deterministic damage model. Of course we could dial up the component failure likelihoods to eleven, but then you no longer have a non-deterministic damage model. You can't have chance, and then reject the possibility of luck. Maybe we can dial down the chances of "no-damage" hits a little, but it's not going to be a dramatic effect. "

What is going to be interesting to see in real life is what effect having a tank that goes so far to protect its crew has on crew morale and thus on real life outcomes. I suspect, not much from a tactical standpoint for the following reasons (this is just my opinion, sorry!):

1. Those directing tank actions, down to tank commander level, are always going to balance the risk to their vehicle against the worth of the mission. If they are likely to save the crew, but lose the (very expensive and scarce) vehicle they are unlikely to make a different choice than if they were likely to lose both.

2. I can't think of an example in history of AFV crews exhibiting unreasonably risk-averse behaviour, so it is difficult to come up with outcomes which would have differed had the crews felt protected by a tank which was otherwise somewhat fragile and easily mission killed.

What do you think?

I tested this extensively and it's not a repeatable issue - adjustment works perfectly. I'm guessing I had a duff gunner on two consecutive uses of the same scenario. Please delete this one and accept my apologies.

Same tests with T-90 (no AVEPS fitted, side on, 2km, aim point just below centre of turret, 50 x DM53 one per target vehicle). There were no solid hits not resulting in damage. Two appear as non (mobility) damage causing on the report of which one at 1:11 was through a road wheel and went through where I think the torsion bars would be on the vehicle, so may have (remote possibility) caused a mobility kill or suspension damage (much more likely) which would not register because the vehicle was already immobilised. The other one at at 14:55 hit road wheels on either side - again, there's a very remote possibility that might have immobilised. Also of note, this test produced 44 one shot kills, vs 30 for the Armata test with Afganit on, though I don't think Afganit made much if any difference to the outcomes.  Crew casualties for the T-90 were 110 (one hundred and ten) and zero (yes, zero!) for the Armata. That separate crew module clearly works.

Armata test, Leo2A6 side on DM53 2km targets T90.sce_9_10-27-19_13_32_26.htm Armata_test,_Leo2A6_side_on_DM53_2km__targets_T90_9420_102719HP-Z4401332.aar

OK, I did the test again. 2km range, same aim as before, just below the turret and upping the number of Armatas to 50. Every hit appeared solid and there were a few Afganit intercepts, some of which still resulted in vehicle kills. of 50 shots, 12 resulted in no damage. That's a ratio not wildly different from the last test and, I believe, statistically significant.

I'll swap the Armatas out for T-90s and do the test again.

Armata vs Leo2A6 side on 2km below turret aim.sce_9_10-27-19_12_57_21.htm Armata_vs_Leo2A6_side_on_9420_102719HP-Z4401257.aar

Do medical vehicles carry red cross insignia that are visible as such through thermals?

Hi. This is almost certainly not a bug, but it's a problem I'm having.

When I'm on a gentle slide slope, tracking targets passing up a valley at 90 degrees to me, and I track and fire, I miss low and the commander tells me to aim up 200. I can make the manual adjustment required which is presumably for the reason that the real life FCS can't cope with trunnion tilt or targets going up slopes relative to the Leo. That I could understand. However, when I switch to an AI gunner and observe from F8 view, the Commander keeps telling the gunner to add 200 over and over and the gunner keeps missing low.

13 minutes ago, Ssnake said:

I'll go even further, and go on the record (not for the first time) that I categorically reject the idea of "game balance" for a combat simulation.

No way. Some combinations of equipment and tactics are unwinnable, and should be, period. If that forces some of you of of their comfort zone - good. That means that you're learning something.

I learn something every time I play. I can't fault SB as a learning tool.

OK, I sat down and had a think about how to exclude random point of impact, so I re did the test, once (1 x 30 targets).

New methodology: Range 1000 metres, gunner set to hold fire so I could take the shots manually. I moved along after each six shots to try and keep the angle side on. I fired one round per Armata and aimed about half way down the side skirt approximately in the middle of the turret, so all shots would hopefully be good solid hits in approximately the same place. I had five penetrations that, according to the report, caused no damage (apart from possibly mobility as I had already disabled both tracks on all target Armatas).  I then reviewed the AAR against the timestamps on the report. The hits were all good and solid. You can check them as follows:

01:08 vs HQ/036

02:19 vs HQ/031

08:02 vs HQ/046

08:09 vs HQ/045

08:32 vs HQ/042

Caveat: it could be that in the design of the T-14 (as far as ESim Games' simulation thereof goes), there is nothing worthwhile at that point to hit that would cause damage that registered in the report.

Armata vs Leo2A6 side on 1000 m one shot per target.sce_11_10-26-19_23_15_21.htm Armata_vs_Leo2A6_side_on_11832_102619HP-Z4402315.aar

16 minutes ago, Ssnake said:

WRT the question why the M829A4 hasn't been added earlier - well, to model it accurately more needs to be done than to simply add another set of ballistic property data and call it M829A4. Some programming will have to be involved to accurately reflect capabilities and limitations. When that'll fit into the workplan, we'll have to see.

Totally off topic, but I hope fixing the target fixation problem with Gill/Spike is working its way up the list as that system works very well vs Armata in game when it functions correctly (though often only mission killing, in contrast to the catastrophic kills it almost invariably inflicts on other Russian/Soviet tanks, as you might expect).