What's the reasoning behind artillery attack values?

PC : Turn based WW2 goodness in the mold of Panzer General. This promises to be a true classic!

Moderators: Slitherine Core, Panzer Corps Moderators, Panzer Corps Design

Post Reply
rezaf
Colonel - Fallschirmjäger
Colonel - Fallschirmjäger
Posts: 1487
Joined: Mon Aug 15, 2011 3:27 pm

What's the reasoning behind artillery attack values?

Post by rezaf » Fri Dec 06, 2013 10:34 am

When I was browsing through all the artiller pieces for my Grand Armory project, I noticed that the attack values for these guns are entirely based on their caliber.
All guns of a size, no matter when manufactured or from whom, get exactly the same attack values. Sometimes, ROF is adjusted slightly, but usually not in a consistent manner.
Available ammo types etc. seem to play no role whatsoever.
Naval attack is basically always 1, as is Initiative.

What's the reasoning behind this? Is artillery of one caliber really always just as effective as any other piece of that same caliber? Are the old WW1 guns just as good as the ones fielded in '44, disregarding the carriage and ease of production?

I also noticed ranges are inconsistent and illogical, but since there's no established "scope" for PzC, this is of course not surprising. A piece with range 1 better be able to take a beating (i.e. be an assault gun), so even Nebelwerfers who couldn't fire very far get range 2, whilst some guns that could fire > 10km sometimes ALSO get range 2 only.
Railroad Artillery could fire 60 kilometers is capped at range 5 or something.
Since circumstances on how large a hex is change constantly, I wonder why there's no "range factor" scenario property that adjusts ART ranges accordingly. But ah well.

My real question is the attack value one. I want to provide some stats for additional artillery pieces and am struggling to differentiate them from their brethren...
_____
rezaf

BiteNibbleChomp
Lieutenant-General - Do 217E
Lieutenant-General - Do 217E
Posts: 3227
Joined: Mon Jul 01, 2013 6:35 am

Re: What's the reasoning behind artillery attack values?

Post by BiteNibbleChomp » Fri Dec 06, 2013 10:45 am

I think that this idea is a bit silly - a gun manufactured in 1897 with 75mm gun should not be the same as a 1940s 75mm gun. I am sure they would have improved on the ammunition etc. at least.

- BNC
Creator of American Civil War mod for Strategic Command WWI!
Discuss here: http://www.matrixgames.com/forums/tm.asp?m=4504986
Download: http://www.mediafire.com/file/tj4t11z3ttl142w/SCACW.zip/file

shawkhan
Staff Sergeant - StuG IIIF
Staff Sergeant - StuG IIIF
Posts: 282
Joined: Mon Dec 24, 2007 7:36 pm

Re: What's the reasoning behind artillery attack values?

Post by shawkhan » Fri Dec 06, 2013 11:51 am

Well, this is a game rather than a simulation, but it is true that artillery is also subject to rates of fire already which adjusts artillery effectiveness somewhat.
One of my pet peeves is that of the dreaded '88'. It was incredibly accurate, which isn't taken into account in the game. It was capable of hitting armored targets at 5k meters, and was able to act as artillery as well, an ability not even depicted in the game. Yet, the '88' is merely equal to the Russian 85mm gun which was about ten times less accurate. '88' attack values should be raised drastically. Accuracy and rate of fire could both be used to adjust artillery values. The other problem with artillery is that of range. In the game range is set by gun caliber. All 105s have a range of three, regardless of the antiquity of the artillery, whereas all 75s have a range of just two. More modern artillery simply had greater range in most cases. There were special cases as well, such as the 170mm gun, which had an incredible range for the times. This piece at least should have a range of four.
Ranges in the game have to be somewhat abstracted, but it could easily have a bit more detail.

Razz1
Panzer Corps Moderator
Panzer Corps Moderator
Posts: 3307
Joined: Tue Jan 18, 2011 4:49 am
Location: USA

Re: What's the reasoning behind artillery attack values?

Post by Razz1 » Sat Dec 07, 2013 1:46 am

I have been researching artillery and found out the data base needs to be updated more to use rate of fire.
Some initiative changes may be in order too.
It is fixedin the AT and AA mod.
Ammo is pretty much a supply issue unless it's mobile, then you could possibly make some adjustment.
As far as I can tell there is no difference in older artillery and more modern artillery.
In fact there were no new artillery pieces made during the war. They just made the same guns from the past and continued manufacturing of some units.
The artillery pieces that were new, were put into tanks or used as AT guns. Basically an artillery piece is just a tube.

There were improvements:
Traverse from left to right
Angle of Fire
loading Ammo
transport and break down of artillery.
Most of the artillery in the war was transported by horses for the Germans and Russians. It's not even modeled in the game! PG had horses.

It's very hard to model these changes in PC. The best thing we have that can be updated to reflect facts is rate of fire.
Unfortunately some artillery pieces should be considered as Corps artillery units. Range 4 and allot more expensive.
The developers considered range 4 to unbalance the game.
Railroad artillery was capped at 5 as they considered it to be unbalanced with a longer range.


The only way to model the 88 better is to have two versions. I have already re-balanced it but they are still too weak.
With two versions, we can have an early version that is not over powered. Then later you can upgrade for a little more cost to a powerful 88 which can take on the big tanks.

shawkhan
Staff Sergeant - StuG IIIF
Staff Sergeant - StuG IIIF
Posts: 282
Joined: Mon Dec 24, 2007 7:36 pm

Re: What's the reasoning behind artillery attack values?

Post by shawkhan » Sat Dec 07, 2013 4:57 am

I have to admit to being a little surprised at the statement that old artillery was no different from the newer artillery and that no improvements were made in artillery during the war. This is manifestly untrue in many instances.
A few examples:
Comparing the '88' to the Russian 85mm gun, the '88' had 50% more range and fired about twice as fast, and was incredibly accurate. In tests, it could repeatedly hit a breadbox-sized target at 1200 yards with every shot. The 88s range of three was accurate but the Russian 85 should have a range of two. The mount of the 88 allowed fast accurate fire from a hull-down position as it was one of the few AA guns that could depress its barrel. Its initiative should be increased as well as its close defense value.
The German 10.5cm howitzer should have a range of two, as its range was only 10km, compared to 9km for the 75mm and 13km for the 15 cm.
The 17 cm K18 was a super cannon, with twice the range (30km) of almost any other gun. Should have at least a range of four if not more.
I could go on and on, but suffice it to say, the increased accuracy of the more modern guns also made them more effective. Some of the German mounts made their guns faster to set up, traverse and load. The 88 for instance could get off its first round within 2-3 minutes of stopping. This was because of its superior mounting system.
The Russian 76mm gun was an excellent weapon. It had a range of 13km so should have a range of three.
Generally speaking, the more modern guns had better breech loading systems and thus had greater rates of fire. For example, the German 75 had a rof of 8-10 rpm, while the better more modern Russian 76 was twice as fast.
I could continue on a per case basis, but suffice it to say the artillery values could use some tweaking.

You shouldn't even get me started on tanks. The Russian 85 could in no way compare to the 88 in hard attack for instance. It is totally illogical for it to have the same attack value in the game. And what is the story with all the KV-85s and IS1s in some of the 1943 scenarios? These were prototype tanks, only 143 KV-85s and 207 IS1s were ever produced. I fought about 500 of them alone in Kremenchug or Kiev43. Where is the T34/85? There were actually more of them built than the T34/76. The initiatives of most Russian tanks until the T34/85 should be drastically lowered. Not only were radios almost unknown in these previous models, but they only had two man turrets, meaning that the tank commander was also the gunner. Combine that with no cupola and only one vision block in the turret and you have a tank that is almost blind, certainly affecting initiative.

The game is already fantastic, but I think making it more historically accurate may make it better.

McGuba
Lieutenant Colonel - Fw 190A
Lieutenant Colonel - Fw 190A
Posts: 1167
Joined: Thu Mar 15, 2012 7:34 pm

Re: What's the reasoning behind artillery attack values?

Post by McGuba » Sat Dec 07, 2013 12:34 pm

It had been discussed earlier. Sadly, Panzer Corps e-file does contain inconsistencies. Examples ranging from the Sturmpanzer I / sIG 38 / sIG 33 II self-propelled guns, all equipped with the 15 cm sIG 33 infantry gun with a range of only 4.700 m, and an in-game range of 3 hexes, to the afformentioned excellent Soviet 76mm guns which had a range of 13.000 m and an in-game range of only 2 hexes.

What happened here, is that they simply took over the stats from Panzer General which had the same inconsistencies. With all the DLCs and stuff released it may be a bit too late to change these stats as it would change the scenarios considerably and they were tested with the original, "erroneous" stats. Deducter released a mod a while ago to fix most of these, but it did change the gameplay of the DLCs considerably as I heard.

So, to answere your original question, there is no reasoning here apart from the silly old Panzer General range calculation i.e. 75mm = 2 hexes, 100mm or bigger = 3 hexes, attack value = caliber, full stop.

Now you have to either live with these stats or change them yourself and see what happens as it is very unlikely that the developers would fix them. Alternatively can make a mod yourself, or use Deducter's, or play custom mods that dare to make these changes.
My real question is the attack value one. I want to provide some stats for additional artillery pieces and am struggling to differentiate them from their brethren...
I would say reduce their attack value slightly if they are older types within the same calibre and vica versa. I would not play with the ROF value as it is not visible so the casual player has no clue about it. You can also reduce / increase its max ammo to represent the nation's general supply / wealth. E.g. Soviets never run out of ammo, minor nations has to count every single shot.
ImageImage
slitherine.com/forum/viewtopic.php?f=147&t=47985
slitherine.com/forum/viewtopic.php?f=147&t=36969

shawkhan
Staff Sergeant - StuG IIIF
Staff Sergeant - StuG IIIF
Posts: 282
Joined: Mon Dec 24, 2007 7:36 pm

Re: What's the reasoning behind artillery attack values?

Post by shawkhan » Sat Dec 07, 2013 2:03 pm

Well, the Russians in the latter part of 1942 on were not hurting for ammunition but they faced serious shortages earlier. For instance, one of the problems with the T34/76 in 1941 was that they had little AP ammo available. Some nations in the war such as Russia and the US had a different philosophy on tank doctrine, not expecting their armor to fight the enemy armor directly. The Russians expected their AT guns to do the job while the US thought Tank Destroyers a better idea. The Germans as is well known differentiated their tanks into tank killers(the PzIII) and infantry support (PzIV).

ThvN
Panzer Corps Moderator
Panzer Corps Moderator
Posts: 1408
Joined: Fri Jul 06, 2012 8:55 pm

Re: What's the reasoning behind artillery attack values?

Post by ThvN » Sat Dec 07, 2013 11:52 pm

rezaf wrote:When I was browsing through all the artiller pieces for my Grand Armory project, I noticed that the attack values for these guns are entirely based on their caliber.
Strange, isn't it? I was even more surprised about the ground/air defense values, actually.
Available ammo types etc. seem to play no role whatsoever.
This is a very good point, especially as late in the war much better shells became available for the Western Allies.
Is artillery of one caliber really always just as effective as any other piece of that same caliber?
If the technology is similar and the gun/shell dimensions as well, than they are very comparable. The Soviet 152mm has practically identical effects as the US 155mm, if they use similar shells. But the older French 155mm (also used by USA when WW2 started) was not in the same league, as that was a leftover from WW1, and even when modernized it wasn't up to par with the more modern guns.
Are the old WW1 guns just as good as the ones fielded in '44, disregarding the carriage and ease of production?
Between WW1 and WW2 lies a very big gap, which is caused by several factors. Like Razz1 said, artillery can be compared a tube, but there are certain technological leaps that make one tube more like a 1965 Ford Mustang GT and another more like a 1995 Ford Mustang GT. They are both Fords that have a V8 and are rear wheel drive, both have four tires and run on gasoline, but the difference is not just being 'newer and a little bit faster'.

Metallurgy: the steel they use and the manufacturing processes can produce much stronger tubes which will allow greater pressures, warp less (accuracy!) and can better withstand prolonged firing. Also, ammunition production benefitted from the same advantages, so they were able to increase the effectiveness of the ammo within the same dimensions.

Chemical: Gunpowder had evolved into 'smokeless powder' before WW1, but it was still being improved: more stable and powerful, so if you were to put in in old shell and fire it from an old cannon it could burst apart. Explosives that were put into the shells and fuses became better as well.

Ballistics: streamlining projectiles and better ballistic calculations and sights improved performance.

Mechanical: the WW1 guns started to use recoil systems that allowed 'rapid fire', this was first used for the famous 'French 75' Model 1897. This meant that the crews did not have to re-aim the gun after each shot (or roll it back into place if it had moved too much). These systems were much improved between the wars and guns that were designed in the interbellum generally could fire faster for longer, and could fire more powerful ammunition for the same caliber and gun weight. These systems also allowed the guns to be designed with features to speed up loading. One example is being able to load a gun at high elevations of the barrel: on some guns, the barrel had to be put horizontal before reloading, which lowered the effective rate of fire a lot.

For a good comparison, the gains were best observed at sea. WW1 battleship guns compared to WW2 models show a big performance gap.
Since circumstances on how large a hex is change constantly, I wonder why there's no "range factor" scenario property that adjusts ART ranges accordingly. But ah well.
No range scaling I'm afraid. With the latest changes, you can now assign a range penalty (for example -2 attack) for each hex over one for every ranged weapon. But that's about it.
My real question is the attack value one. I want to provide some stats for additional artillery pieces and am struggling to differentiate them from their brethren...
What sort of info do you need? I have collected quite a few bookmarks over the past few years, and modelling this sort of thing depends on many factors. Years ago, I've done some very rough calculations for effects of different shell types, and things like shell weight/payload weight, blast effects (pressure) and some comparisons to close matches of existing data can give some figures, but that was for a game with a much smaller scale, with individual guns.

In PzC, the artillery weapons need to be taken into account as well, together with any peculiarities (fusing, ammo quality, deployment, organization, training, etc.) The differences will be less pronounced on such a large scale, but if you want to create more diversity there is a lot of room to tweak some stats.

Bonus reading material: for fun, check out page 8 (in case the browser doesn't take you there already) and further of the pdf in this link:

http://sill-www.army.mil/firesbulletin/ ... df#page=10

It's about modern artillery tests versus armoured vehicles, real tests with live fire, no simulations. And it turns out they are a lot more vulnerable than people usually assume, even without direct hits.

rezaf
Colonel - Fallschirmjäger
Colonel - Fallschirmjäger
Posts: 1487
Joined: Mon Aug 15, 2011 3:27 pm

Re: What's the reasoning behind artillery attack values?

Post by rezaf » Sun Dec 08, 2013 2:48 pm

Thanks for the input everyone.
I'm going to try to address a couple of points... so WARNING! Wall of text incoming!
Razz1 wrote:I have been researching artillery and found out the data base needs to be updated more to use rate of fire.(...)
Yeah, I realized by now that this is a can-of-worms situation. My first (and last) impulse was to leave as much as possible unchanged for now.
I'm all for rebalancing the lot at some point, but for now, my goal was to make a equipment database that was as complete as possible using as much of the already available artwork as possible and letting interested modders know what icons are still missing or lacking.
This does mean, however, that quite a few artillery pieces are going to get added which hardly differ from others already in the game - because I'm going to stick to the copycat approach to determining attack values of same-caliber artillery pieces - I'm not nearly competent enough to make conclusions about what values artillery should use instead. I'm currently leaning towards adjusting ranges though, which would give me SOME influence. The idea is <=10000m range is 2 hexes, 10-15000m is 3 hexes, 15-20000 is 4 hexes and so on. Railroad artillery is the only real problem here, because it'd have to get a range of >10 hexes.
Also, this approach would lump a ton of artillery pieces with greatly different range in the 2-hex group, but due to how the game works, you NEED to be able to put your artillery behind the frontlines, which makes range 2 neccessary, even for things like the nebelwerfers, some of which had only a range of one kilometer.
But no better idea has come to mind yet.
Razz1 wrote:Most of the artillery in the war was transported by horses for the Germans and Russians. It's not even modeled in the game! PG had horses.
Yeah, I also realized one limitation of early-design artillery was the carriage, which often (usually) was not suitable for being pulled by a truck, only by much slower horses.
Maybe this can be incorporated in the game at some point by using a different transport category, like heavypull, but I think i'll neglect it for the time being.
Razz1 wrote:It's very hard to model these changes in PC. The best thing we have that can be updated to reflect facts is rate of fire.
Yeah, but as noted, RoF is a hidden effect, and I'm not sure it's actual effect is really explained anywhere. Got a link handy?
Basically, I think it should not be used at all, as it makes relatively little sense in a game where a turn is usually an entire day.

However, I have to add that I came to the conclusion that all PzC stats and calculations are completely irregular and often downright schizophrenic. I understand the base for everything was copying Panzer General, but maybe Rudankort should have invested some more time in this area. Maybe he'll do so in a sequel, as I think it's now too late for PzC and we'll have to live with the basic structure that's there. I can blame it on all the beta testers who weren't vocal enough in 2011. :twisted:
McGuba wrote:It had been discussed earlier. Sadly, Panzer Corps e-file does contain inconsistencies.
Heh, words more true haven't been spoken I guess.
McGuba wrote:Deducter released a mod a while ago to fix most of these, but it did change the gameplay of the DLCs considerably as I heard.
My big problem with Deducters approach is that his idea of realistic essentially means "more difficult for the player". In my eyes, PzC, like PG before it, is first and foremost a beer & prezels wargame, where realism oftentimes takes the back seat. Trying to shoehorn absolute realism, also taking into consideration stuff like the availability of equipment etc., into the e-file is a recipe for disaster, imo, and thus I think deducters modifications are only suitable for the most expert of players.
McGuba wrote:Now you have to either live with these stats or change them yourself and see what happens as it is very unlikely that the developers would fix them. Alternatively can make a mod yourself, or use Deducter's, or play custom mods that dare to make these changes.
Oh, of course, I wasn't trying to lobby for official changes, as I think it's too late for those now. My main reason for posting the thread, like I wrote, is that I want to make an achive of sorts of all available unit icons and offer "suggested stats" wherever possible.
The end goal would be to have an ultimate e-file with everything that was used anywhere by anyone in WW2 in any significant quantity, but I'm not sure this is remotely achievable. Maybe a decent percentage of stuff used by the major powers will do.
ThvN wrote:Strange, isn't it? I was even more surprised about the ground/air defense values, actually..
I actually have no idea about the reasoning here. I guess low GD was chosen to make unprotected artillery easy pickings, and AA is mostly 10 ... whatever?
ThvN wrote:(Stuff about older vs. newer artillery)
I guess there were advancements after all, but I'm still unsure about how significant they actually were. A big problem in the context of the game is that a lot of variables are irrelevant to the player in most circumstances. That an artillery piece can be easier mass-produced, for example. Could be extremely important in real life, but if I want to buy a piece of artillery in 1943 in the game, does it really matter if it costs 200 prestige more or less?
And most variables play no role whatsoever - for example, almost all artillery have INI 1. What gives? Which pieces should get higher ini, and why?
INI is another of those things that have mostly hidden effects that can have immense impact, are communicated badly to the player (at least it's in the log...) and even are hard to judge for the game itself (which knows all variables) - hence the often borked predictions.
ThvN wrote:No range scaling I'm afraid. With the latest changes, you can now assign a range penalty (for example -2 attack) for each hex over one for every ranged weapon. But that's about it.
Yes, I know, but why? Is anyone aware? It seems like such an obvious idea, not only affecting range, but also movement range and fuel consumption.
Should be quite easy to program as well, so I'm wondering why something like it was never introduced.
ThvN wrote:What sort of info do you need?
I'm not sure, actually. If I ever get around to rebalancing the stats of all artillery pieces, maybe you can lend a hand, but like I wrote, I think I'll leave most things intact for now, lest the house of cards come down upon me.
_____
rezaf

ThvN
Panzer Corps Moderator
Panzer Corps Moderator
Posts: 1408
Joined: Fri Jul 06, 2012 8:55 pm

Re: What's the reasoning behind artillery attack values?

Post by ThvN » Mon Dec 09, 2013 11:47 pm

rezaf wrote:Thanks for the input everyone.
I'm going to try to address a couple of points... so WARNING! Wall of text incoming!
Ah, my favourite response. Thanks for taking the time.

About your range troubles, to add to the confusion some guns had different ranges depending on ammo or type of application. An example, the 10,5cm leFH 18 (105mm light field howitzer):

The towed piece had a range of 10.600 meters, the later 'M' model 12.300, because that could handle higher charges due to a muzzle brake. The rate of fire (RoF) is usually given as 4-6 rounds/minute (unknown if that is sustained or max rate, and I've seen a source list 6-8 rounds/min, but that might be for the uprated 18/40 model).

The gun was also mounted on the Wespe, where it had the same range but reduced RoF; it has less crew to operate it and conditions were relatively cramped. I could find no figures, unfortunately.

The StuH 42 also used the same gun, and apparently had a range of just 5.400 meters, because elevation was severely restricted (less then half that of the Wespe). The rate of fire was lower as well, quoted as 3-5 rounds/min, due to even more cramped conditions than in the Wespe.

Yeah, but as noted, RoF is a hidden effect, and I'm not sure it's actual effect is really explained anywhere. Got a link handy?
It's a very easy system, and it is partly visible in the combat log. The numbers are percentages (8=80%, 11=110%). These are mulitplied with the strength of the unit to get the adjusted number of shots. viewtopic.php?f=121&t=26663

A big problem in the context of the game is that a lot of variables are irrelevant to the player in most circumstances. That an artillery piece can be easier mass-produced, for example. Could be extremely important in real life, but if I want to buy a piece of artillery in 1943 in the game, does it really matter if it costs 200 prestige more or less?
Cost might start to matter more with the prestige 'soft cap' ? I've started to look at cost-effective units due to this, and there are some interesting oddities: 15cm sFH + SdKfz 251 halftrack = 371 prestige, but 17cm K18 + SdKfz 7 halftrack = 380 pr. They have the same RoF (8), so the difference is the gun stats and the extra attack of the 251 halftrack... Which I won't ever want to use. Protection is equal, both halftracks are 'hard' targets (?) and have the same defensive values (??) except close defense. This makes no sense to me at all.
And most variables play no role whatsoever - for example, almost all artillery have INI 1. What gives? Which pieces should get higher ini, and why? INI is another of those things that have mostly hidden effects that can have immense impact, are communicated badly to the player (at least it's in the log...) and even are hard to judge for the game itself (which knows all variables) - hence the often borked predictions.
I've noticed the 'borked' predictions are often due to the INI dice roll at the start of the combat. This can literally turn things around if the INI's are already close, very annoying. I'm hoping it will be made selectable in the future, it would be interesting to see if the combat results will start to fall better within players' expectations.

Some say INI is supposed to resemble mainly range (I'm assuming they mean effective range). But if that were true, than artillery would have the highest INI's of all. So I like to think of it more as the ability to quickly engage (multiple) targets before they can shoot back, were not just range plays a role, but also accuracy, stealth, sighting equipment, training, etc.

So depending on wether a gun isn't too big (cumbersome) and slow to reload/re-aim at attackers, they certainly should be some adjustments to INI in my opinion. So, I'm trying to raise the INI for lighter pieces which could fire and be re-aimed quickly and/or fire directly at targets (some guns had direct-fire telescopic 'anti-tank' sights).

To prove your point, check the self-propelled pieces and their INI, example: The INI for the Wespe=4, StuH=2 (leFH 18=1). They all have the same gun, doesn't make sense. The StuH 42 was designed to be fired directly (and has INI=4 in AT mode...) and has a very low profile.
Another one, the 15cm sFH 18 (INI=1) was mounted in the Hummel (INI=5), yes, that makes perfect sense as well. :roll:
I actually have no idea about the reasoning here. I guess low GD was chosen to make unprotected artillery easy pickings, and AA is mostly 10 ... whatever?
Probably due to some 'rock, paper, scissors' design, but it's very inconsistent sometimes. About the defensive values, it makes no sense to me that a tiny 3.7cm AT gun has the same AD as a massive artillery piece. They are not exactly as easy to hit (let alone detect from the air), and their GD is quite different. Even with the same guns sometimes: the QF 25pdr has GD=2, but switch it to AT and it's suddenly GD=6? Okay, they might dig it in a little better, but it's still a fairly big difference? The 88mm FlaK has GD=2, but its AT switch has GD=5. But both guns will always have AD=10 regardless of switch state.
I'm not sure, actually. If I ever get around to rebalancing the stats of all artillery pieces, maybe you can lend a hand, but like I wrote, I think I'll leave most things intact for now, lest the house of cards come down upon me.
I agree with being cautious, it's easy to mess up a working system (I already did that several times). Just let me know when you need anything.

rezaf
Colonel - Fallschirmjäger
Colonel - Fallschirmjäger
Posts: 1487
Joined: Mon Aug 15, 2011 3:27 pm

Re: What's the reasoning behind artillery attack values?

Post by rezaf » Wed Dec 11, 2013 1:57 pm

ThvN wrote:Ah, my favourite response. Thanks for taking the time.
Heh, no problem at all - I have to thank you (and everyone else) for participating.
ThvN wrote:It's a very easy system, and it is partly visible in the combat log. The numbers are percentages (8=80%, 11=110%). These are mulitplied with the strength of the unit to get the adjusted number of shots. viewtopic.php?f=121&t=26663
Well ... let me say I think it's a horrible system, as rolls tend to be very important in combat and this change to rolls isn't communicated at all. When it comes to artillery, it probably doesn't change things too much, but anyway.
Furthermore, in the context of a PzC turn - usually an entire day - RoF makes no sense whatsoever.
ThvN wrote:I've noticed the 'borked' predictions are often due to the INI dice roll at the start of the combat. This can literally turn things around if the INI's are already close, very annoying. I'm hoping it will be made selectable in the future, it would be interesting to see if the combat results will start to fall better within players' expectations.
What do you mean, selectable?
I imagine one could run tests just now by giving units different INI values (for example give two units identical ones).
Since you appear to have such things on hand, is there a in-depth analysis of how the different combat values work and how much impact changes have somewhere?
My amateurish understanding was that essentially, who wins the initiative roll shoots first, high noon style.
ThvN wrote:Some say INI is supposed to resemble mainly range (I'm assuming they mean effective range). But if that were true, than artillery would have the highest INI's of all. So I like to think of it more as the ability to quickly engage (multiple) targets before they can shoot back, were not just range plays a role, but also accuracy, stealth, sighting equipment, training, etc.
I think it often IS about range, though. Mainly, if you have two tanks, one of which has a gun that can shoot much farther, that one ought to have a much higher INI value.
If you have two identical tanks, the one with better maneuverability ought to have higher INI, though.
But what if you compare a tank with double gun-range to another which can drive (exaggeration) 300mph and thus the first one can usually not hope to hit him, despite greater gun range?
Even worse when it comes to aircraft. A fast one ... should it have high INI? High movement? Great air and ground defense?

It's one of Rudankorts greatest failings that he chose not to overhaul these inconsistencies dating back to PG when developing Panzer Corps. INI is horrible, RoF is completely intransparent, artillery stats make little sense, prestige is too limited a mechanic, hard-attack applies to two completely different things (armored vehicles and immobile fortrifications) ... the list goes on and on.
ThvN wrote:So depending on wether a gun isn't too big (cumbersome) and slow to reload/re-aim at attackers, they certainly should be some adjustments to INI in my opinion. So, I'm trying to raise the INI for lighter pieces which could fire and be re-aimed quickly and/or fire directly at targets (some guns had direct-fire telescopic 'anti-tank' sights).
Artillery (or ranged units in general) are an exception to the INI rule outlined above, though, because the enemy cannot shoot back. Giving those units high INI would mean when assaulted directly, they'd be able to lay down heavy defensive fire at whomever tries to assault them, which would probably be very painful to soft attackers (infantry) especially. Unless I have even less understanding of the mechanics than I already admitted, which is entirely possible.
ThvN wrote:To prove your point, check the self-propelled pieces and their INI, example: The INI for the Wespe=4, StuH=2 (leFH 18=1). They all have the same gun, doesn't make sense. The StuH 42 was designed to be fired directly (and has INI=4 in AT mode...) and has a very low profile.
Another one, the 15cm sFH 18 (INI=1) was mounted in the Hummel (INI=5), yes, that makes perfect sense as well. :roll:
Yeah, I obviously have no idea of the reasoning behind SP-Art either. They are more likely to get into ground combat and should have a bit more survivability, maybe that's the reasoning? I dunno.
ThvN wrote:About the defensive values, it makes no sense to me that a tiny 3.7cm AT gun has the same AD as a massive artillery piece. They are not exactly as easy to hit (let alone detect from the air), and their GD is quite different. Even with the same guns sometimes: the QF 25pdr has GD=2, but switch it to AT and it's suddenly GD=6? Okay, they might dig it in a little better, but it's still a fairly big difference? The 88mm FlaK has GD=2, but its AT switch has GD=5. But both guns will always have AD=10 regardless of switch state.
Same as above, tweaks to GD are neccessary because the AT unit will engage in direct (range 0) combat and having low stats will be a suicide-recipe in that context. Again, a turn measuring an entire day is a big problem, because you can do quite some digging in in that timeframe, but it also should be no problem (in real life) to elevate the gun and do some artillery bombarding at any time AFTER having dug in. Digging in can just be a handwaving gesture when it comes to these stats anyway, since there is an existing, entirely different mechanic for digging in.
ThvN wrote:I agree with being cautious, it's easy to mess up a working system (I already did that several times). Just let me know when you need anything.
For now, I stuck with the vanilla PzC approach, which has resulted in a ton of mostly identical artillery pieces. Ah well.
It's very hard to come up with plausible stats, especially for switchable stuff. For example, guille has made AT graphics for the german field howitzers. I found proof that they were actually used in this role, so ... fine, I'll include the switched state.
But with which stats? How good should a 7.5 cm gun be against tanks? It has to be worse than the dedicated AT guns available at the time, right? We already have that problem with the .88 which is better at the AT job than the AT guns.
Diminishing prestige aside (personally, I think punishing good play is a terrible idea, which is why I think this is a bad mechanic), really cheap prices are no encouragement. So the guns need to be much more expensive OR much worse than the AT guns available. But how good should 7.5 cm ART with AT rounds REALLY be?

More often than not, I did some comparisons and threw some semi-random number in the pzeqp file, sometimes realizing later that, compared with some other unit's stats, having it there can't be a good thing.
Some of these things remain unfixed, once I'm done with the germans, I'll post in the TGA thread and maybe you can take a look. I still got quite some work ahead of me, though.
_____
rezaf

shawkhan
Staff Sergeant - StuG IIIF
Staff Sergeant - StuG IIIF
Posts: 282
Joined: Mon Dec 24, 2007 7:36 pm

Re: What's the reasoning behind artillery attack values?

Post by shawkhan » Wed Dec 11, 2013 5:03 pm

Personally, I would like to see a quick interim fix which could take care of the most egregious inaccuracies. Perfection takes much too long, usually forever.
Having read Deducter's new unit values I think he got about 90% of them correct already.
Nerfing the fuel allotment to Tigers I don't agree with but his HA of 19 shows the 88 in a more accurate fashion, especially when compared to the inferior Russian 85mm. The inferiority of the 85 to the German 88 was the reason the Russians went to the100mm and the 122mm ATs after all.
The range for both the German 75 and 105 should be only 2 hexes.
The Russian 76.2mm artillery had a range of three. There was a reason it was called the 'crash boom'. With its high muzzle velocity and flat trajectory, you were given little warning before the round arrived.
The 17 cm should have a range of 4 as per Deducter's mod.
German tanks should have much higher initiative than any allied tank until the T34/85. The units in the game are at least battalion size. CCC, command, control and communication on the German size dictate this. Their tanks were also much more user-friendly, due to the excellent optics, crew layout, vision blocks and radios in all their vehicles.
AT guns should almost always get the first shot against vehicles. They were small, hard to see, especially when dug in. There is a story of a German 37mm AT banging away at a Russian T34/76, getting off some 23 mostly harmless shots w/o being spotted by the poor nearly blind Russian tank.
Artillery already has a HA factor. Unless suppressed, artillery should almost always get the first shot when attacked by vehicles. After all, it has the longer range.
Fix just these things and I would be a happy camper.

ThvN
Panzer Corps Moderator
Panzer Corps Moderator
Posts: 1408
Joined: Fri Jul 06, 2012 8:55 pm

Re: What's the reasoning behind artillery attack values?

Post by ThvN » Fri Dec 13, 2013 2:54 am

rezaf wrote: (about RoF) Well ... let me say I think it's a horrible system, as rolls tend to be very important in combat and this change to rolls isn't communicated at all. When it comes to artillery, it probably doesn't change things too much, but anyway.
Furthermore, in the context of a PzC turn - usually an entire day - RoF makes no sense whatsoever.
It depends, I think, but you are right that in the context of the game RoF sometimes doesn't make much sense, and it should be shown pre-purchase. The RoF partly 'decouples' the unit strength from the amount of rolls-to-hit, which can be beneficial sometimes. And because I have been modding values for a while now I have noticed some effects that cannot be compensated by simply tweaking the attack values, so I think the RoF has its merits.

The bottleneck design constraint seems to be the strength of the unit, which is usually fixed at 10. So if every unit has 10 chances to hit, this might unbalance certain things, because than you can only vary the kill chance per roll, not the number of times this roll gets made. The only way to replicate this mechanism would be to adjust the basic strength of the unit, and buying a 7-strength 21cm Mrs or upgrading a 11-str 7.5cm FK to an 8-str 15cm sFH might not be desired. These units would be proportionally more vulnerable or might even be 'Conscript'-like, treating them unfairly and mess up the combat system. But than your question would probably, and rightfully, be: why not just adjust the attack values to replicate the different effectiveness? I've been tweaking numbers for a while now, so I'll just show some examples of what can happen.

The RoF king, the Flammpanzer (RoF= 14), would be a very different unit if I would remove the RoF bonus. When stock (SA=6, HA=1), it has the same attack values as a Fallschirmjäger. So the chances of each roll to supress/kill are equal, but the Flammpanzer would do a lot more damage (on average 40% more kills than the Fj unit). If I were to remove its RoF, and apply some percentages, theoretically I'd have to make his SA=7.7 and HA=1.3 to compensate, but I found out it isn't that easy.

Let's say I made it SA=8/HA=1. This compensation will mean it keeps it advantage, percentage-wise, over the Fj, right? Well, the chances of hitting increase a little, but it will not compensate for the number of kills it will achieve. You will have to increase the attack values a lot to compensate, but it still can't replace the ability to make more kills than the unit strength (10). In its stock form, its specialty is attacking soft units with very low GD and making a lot of hits (very handy against Conscripts and overstrength soft units with low GD). For example, a soft attack tank like the PzIIIN can't do that without overstrength, and that will require lots of experience and be relatively expensive.

So the Flammpanzer occupies it's own little niche because of its peculiar stats. I think the idea is that the Flammpanzer quickly loses effectiveness rapidly as soon as the defense values of the enemy increase. If I increase its SA to make it kill a lot more than the Fj can, it will become very effective against soft units with higher defensive values as well. It becomes another soft attack tank like the PzIIIN. It will still have its trait but it will be less specialized, and will be even more devastating against entrenched soft units instead of just the ones with low GD.

Similarly, the reduced RoF of heavy artillery is to preserve their ability to damage units with high defenses, but make sure they do not become too powerful against units which have low defenses. This focuses the effectiveness of these units within a certain range of targets. So in order to get the most out of them, the player will have to use the large artillery against targets with high defenses. If he were to use them against some weak unit in an open field (see below), their ability would be wasted. This effect cannot be replicated by changing attack values, that would make them simply better or worse over the whole range of targets (global increase of ability), but thanks to RoF tweaks they must be used against a specific type of target in order to benefit from their increased abilities.

Think of it as a trade-off. Let me show you some numbers.

Artillery units attack a T-34/43 (GD=15) in an open field. All units have zero experience and normal strength:

Artillery Unit Miss Supp Kill [Shots] (Predicted result)
7.5cm FK nA= 84% 15% 1% [11] (0 killed, 2 suppressed)
10.5cm leFH= 80% 18% 2% [10] (0 killed, 2 suppressed)
21cm Mrs 18= 50% 40% 10% [7] (1 killed, 3 suppressed)

So, the 21cm is much better versus targets with high defenses. Who would have thought? :wink:


The same artillery vs. a poor Soviet Regular (GD=6) (entrenchment 0) in an open field.

Artillery Unit Miss Supp Kill [Shots] (Predicted result)
7.5cm FK nA= 45% 44% 11% [11] (1 killed, 5 suppressed)
10.5cm leFH= 35% 51% 14% [10] (1 killed, 5 suppressed)
21cm Mrs 18= 14% 59% 27% [7] (2 killed, 5 suppressed)

Mmmh... Still better, but not exactly what I expect from a unit costing twice as much as the 10.5cm. And if it wasn't for the lack of range, the 7.5cm would be quite competitive.


But now, the same artillery vs. a poor Soviet Conscript (GD=2) (entrenchment 0) in an open field.

Artillery Unit Miss Supp Kill [Shots] (Predicted result)
7.5cm FK nA= 28% 54% 18% [11] (2 killed, 7 suppressed)
10.5cm leFH= 24% 56% 20% [10] (2 killed, 7 suppressed)
21cm Mrs 18= 10% 58% 32% [7] (2 killed, 5 suppressed)

Yes, the conscripts would rather be shot at by the 21cm than the lighter guns. This is were the trade-off shows. The low RoF keeps it from being better in every discipline, but if the defense values increase it quickly becomes the better choice. There are situations where a smaller gun will be better, like attacking transport trucks and other units in the open with very low defenses, although entrenchment will quickly raise defensive values to where the bigger guns are needed. But still, a player must use the 21cm correctly to avoid wasting its capabilities, you can't just upgrade your guns and expect a general increase in lethality, despite the bigger numbers. That the player isn't told this is a big letdown, just like most traits are hidden.

I don't like the RoF being next to invisible. Almost all stats are clearly labeled when purchasing, so this very important one should be as well. And I wouldn't mind having some kind of pre-purchase way of determining if a unit is switchable...
ThvN wrote:I've noticed the 'borked' predictions are often due to the INI dice roll at the start of the combat. This can literally turn things around if the INI's are already close, very annoying. I'm hoping it will be made selectable in the future, it would be interesting to see if the combat results will start to fall better within players' expectations.
What do you mean, selectable?
I imagine one could run tests just now by giving units different INI values (for example give two units identical ones).
Since you appear to have such things on hand, is there a in-depth analysis of how the different combat values work and how much impact changes have somewhere?
'Selectable' means that I want it as an option, like the 'dice chess' option. I think the random combat results are not that bad, it's mostly the +1 or +2 INI dice roll at the start of the combat that can mess things up.

The chances to miss, suppress or kill with each roll depend on the relative attack/defense values of the units, which I read somewhere are taken from tables (hardcoded) in the game. The artillery has a much bigger chance of getting a 'suppress' result, so there are definately differences between unit classes. Since I have no hard data, I mostly rely on small tests like the one above.
My amateurish understanding was that essentially, who wins the initiative roll shoots first, high noon style.
Essentially, yes, but it is not all-or-nothing... For each point of advantage in initiative, 20% of shots are made before the opponent. So you have ini=8, the other has ini=6, your unit can use 40% of its shots before the other one can return fire. The strength points you kill and/or suppress with your initiave advantage can't shoot back, and the higher the difference, the more 'free' shots you get. But if the difference in ini is more than 5, this has no effect, because than you have a 100% advantage. Let's say you have ini=10, the enemy ini=3. You have 7 more, which is 5 or more, so your unit can use all its attacks before the other may return fire. If you kill 3 and supress 5, that leaves the enemy with 2 shots to return fire with.

So higher ini is not an 'all-or-nothing' advantage, but the higher the difference the better your chance to inflict damage before taking return fire, up until a certain point.

Imagine you are going to have this high noon shootout, and both you and your opponent get 10 shots each (no hi-cap magazines for you, and no six-shooters :P ). But there is this crazy rule: the judges hold a fiscal audition where the richer party can 'win' free shots against your opponent before he may return fire. For each 1,000 dollars more you own than your opponent you will get a free shot, but no more than five shots can be free, no matter how rich you are. So you show them all your savings and your opponent does likewise. The judges count, and let's say you win the audition by having 8,000 dollar more. That's a pity, you get five free shots but your last 3,000 dollars don't count, they are 'wasted'.

OK, so now you can open fire, but after you have shot five times your opponent is allowed to start shooting as well. So the first five are made without return fire, but your remaining shots will be done the classical way: after the free shots, both opponents will shoot simultaneously at each other as long as both still have ammo. If one party runs out the other may continue shooting, if he still can, until he's empty too.

Of course, if you manage to wound (suppress) your opponent, he will not be able to return with all his remaining shots. But if you miss al your free shots or they don't incapacitate your opponent, you will still be shot at a lot. So if you are a very lousy shot or your opponent is wearing a bulletproof vest these free shots won't help a lot. And being very rich is not much help either, because you only need to be 5,000 dollar richer than your opponent to get the maximum benefit.

ThvN wrote:Some say INI is supposed to resemble mainly range (I'm assuming they mean effective range). But if that were true, than artillery would have the highest INI's of all. So I like to think of it more as the ability to quickly engage (multiple) targets before they can shoot back, were not just range plays a role, but also accuracy, stealth, sighting equipment, training, etc.
I think it often IS about range, though. Mainly, if you have two tanks, one of which has a gun that can shoot much farther, that one ought to have a much higher INI value.
If you have two identical tanks, the one with better maneuverability ought to have higher INI, though.
But what if you compare a tank with double gun-range to another which can drive (exaggeration) 300mph and thus the first one can usually not hope to hit him, despite greater gun range?
Range still plays a big part, of course, but it depends on what you consider 'range'. It doesn't have a singular meaning, unfortunately, just like the rate of fire for a weapon can vary depending on what definition you apply. I try to use 'effective range' or 'practical range', which is something completely different than maximum range or theoretical range.

About the 300mph tank: the speed advantage could influence INI but I would also increase the GD, because it is harder to hit. So the tank with the longer ranged gun might still be able to shoot first at the 300mph tank before the other side can shoot (= higher ini), but the 300mph tank would probably not be hit by it, because its speed makes it very hard to hit (so its GD would be very high). The speed needn't automatically increase its INI, what if it only has a flame-thrower? It's all relative, unfortunately.

If a target is harder to hit, I try to reflect that by tweaking INI and/or GD, which is a separate mechanism from the whole INI system. But realistically, yes, a more mobile tank can compensate INI despite having a shorter ranged gun, because it might be able to negate the range advantage of the opponent by manoeuvering well.
Even worse when it comes to aircraft. A fast one ... should it have high INI? High movement? Great air and ground defense?
Don't get me started :shock: . I'm trying to test out an idea I had a long time ago, which will radically change some airplanes. Basically, a lot of airplanes will become switchable into 'high' and 'low' versions. The idea is that this will force the player to choose each turn between keeping your Focke-Wulf switched as a 'pure' fighter to cover units, or to switch to a 'low' flying version with the ability to engage ground targets at the cost of reduced air fighting capabilities. The 'high' versions will generally have better ground/air defense, higher INI, etc.

This will also mean that fighters can be made more different and I can make planes that were historically less suited to higher altitudes relatively better in their 'low' versions. But I've discovered that tweaking plane defense/attack values is very difficult, and I've been forced to re-think my approach to include the AAA and air defense/attack stats of ground units. It's a mess and because my job has been taking more and more time I have not been able to continue modding for a while now.

Regarding your question: a fast airplane isn't fast everywhere. The P-51D has the famous top speed of 437mph (705km/h) at 25,000 ft., but at sea level it was barely faster than a P-40, which was generally considered to not be a very fast plane. In airplanes, high INI for me would be the ability to dictate the terms of engagement, which generally means having altitude/speed available quickly when needed. This is the concept of the 'energy fighter', which can strike and disengage almost at will against lower and slower opponents. But typical 'dogfighters' could have a high air defense, being hard to hit and having the advantage when the 'energy fighters' have wasted their potential initiative advantage. These are all just ideas which I have been trying to test, but it is difficult as the gamerules have changed in between and the new custom experience table opens up new possibilities.
ThvN wrote:So depending on wether a gun isn't too big (cumbersome) and slow to reload/re-aim at attackers, they certainly should be some adjustments to INI in my opinion. So, I'm trying to raise the INI for lighter pieces which could fire and be re-aimed quickly and/or fire directly at targets (some guns had direct-fire telescopic 'anti-tank' sights).
Artillery (or ranged units in general) are an exception to the INI rule outlined above, though, because the enemy cannot shoot back. Giving those units high INI would mean when assaulted directly, they'd be able to lay down heavy defensive fire at whomever tries to assault them, which would probably be very painful to soft attackers (infantry) especially. Unless I have even less understanding of the mechanics than I already admitted, which is entirely possible.
ThvN wrote:To prove your point, check the self-propelled pieces and their INI, example: The INI for the Wespe=4, StuH=2 (leFH 18=1). They all have the same gun, doesn't make sense. The StuH 42 was designed to be fired directly (and has INI=4 in AT mode...) and has a very low profile.
Another one, the 15cm sFH 18 (INI=1) was mounted in the Hummel (INI=5), yes, that makes perfect sense as well. :roll:
Yeah, I obviously have no idea of the reasoning behind SP-Art either. They are more likely to get into ground combat and should have a bit more survivability, maybe that's the reasoning? I dunno.
I think you understand the game quite well. Like you said, INI isn't used in ranged combat. So the INI value should only reflect the ability to engage those attacking units. So why are very comparable units so very different? I think it's a left-over from a rock-paper-scissors scheme, but it has so many exceptions now that some of the logic seems rather randomly applied. If an SP Arty unit has a higher INI than the towed model it could mean that the mobility causes the difference, but this logic is not applied consistently. But like you, I don't know, the stats mostly have pattern where I think I know what the logic is, but with the artillery I'm a bit lost.



ThvN wrote:About the defensive values, it makes no sense to me that a tiny 3.7cm AT gun has the same AD as a massive artillery piece. They are not exactly as easy to hit (let alone detect from the air), and their GD is quite different. Even with the same guns sometimes: the QF 25pdr has GD=2, but switch it to AT and it's suddenly GD=6? Okay, they might dig it in a little better, but it's still a fairly big difference? The 88mm FlaK has GD=2, but its AT switch has GD=5. But both guns will always have AD=10 regardless of switch state.
Same as above, tweaks to GD are neccessary because the AT unit will engage in direct (range 0) combat and having low stats will be a suicide-recipe in that context. Again, a turn measuring an entire day is a big problem, because you can do quite some digging in in that timeframe, but it also should be no problem (in real life) to elevate the gun and do some artillery bombarding at any time AFTER having dug in. Digging in can just be a handwaving gesture when it comes to these stats anyway, since there is an existing, entirely different mechanic for digging in.
That is exactly my problem with this sort of inconsistent values, it is not clear why there should be a difference except for giving it a quick 'fix' to make sure it won't be useless in its intended role. I don't mind handwaving and bending things to fit the intended result, that's fine, but apply too much of and it will start to have unintended effects, like ending up with overpowered or useless units.
For now, I stuck with the vanilla PzC approach, which has resulted in a ton of mostly identical artillery pieces. Ah well.
It's very hard to come up with plausible stats, especially for switchable stuff. For example, guille has made AT graphics for the german field howitzers. I found proof that they were actually used in this role, so ... fine, I'll include the switched state.
But with which stats? How good should a 7.5 cm gun be against tanks? It has to be worse than the dedicated AT guns available at the time, right? We already have that problem with the .88 which is better at the AT job than the AT guns.
Diminishing prestige aside (personally, I think punishing good play is a terrible idea, which is why I think this is a bad mechanic), really cheap prices are no encouragement. So the guns need to be much more expensive OR much worse than the AT guns available. But how good should 7.5 cm ART with AT rounds REALLY be?
Same here. I'm still busy figuring out good stats for switchable AAA units. At first I made them too powerful, after I fixed that I realized it might be a better idea to not switch them all into 'AT' units. But if I change one thing I have to make it more expensive, and start redoing other things as well.

BTW, not all artillery pieces were automatically worse than 'real' AT guns, if you look at the power of the gun and ammo. Some guns were supplied with fairly good AT rounds, but they only had a few. The Germans used captured Soviet 76.2mm artillery and rebuilt them into antitank guns, they even made their own ammo for them. But most artillery was far more useful in its intended role, and anti-tank use was usually only during emergencies. I think their initiative should be a bit lower than 'real' AT guns, and the attack values would depend on what gun it was and how common AT ammo was. HE could also be very effective against armored targets, but it would have to be a fairly big round.

I have been changing the soft cap into different values for each GC, so far it seems it needs to be a bit higher for the later years, instead of 800pr I'm now working with a 1000 prestige ceiling, and I'm in '44 West now.
More often than not, I did some comparisons and threw some semi-random number in the pzeqp file, sometimes realizing later that, compared with some other unit's stats, having it there can't be a good thing.
Some of these things remain unfixed, once I'm done with the germans, I'll post in the TGA thread and maybe you can take a look. I still got quite some work ahead of me, though.
Take your time, you will probably work a lot quicker than me :oops: . In case you encounter problems, just let me know, I might think of some suggestions or my ramblings might help get you unstuck.

Tarrak
Panzer Corps Moderator
Panzer Corps Moderator
Posts: 1183
Joined: Mon Jul 11, 2011 11:01 pm

Re: What's the reasoning behind artillery attack values?

Post by Tarrak » Fri Dec 13, 2013 9:40 am

A very nice analysis of the RoF ThvN.

captainjack
Colonel - Ju 88A
Colonel - Ju 88A
Posts: 1568
Joined: Thu Sep 13, 2012 7:42 am

Re: What's the reasoning behind artillery attack values?

Post by captainjack » Sun Dec 15, 2013 7:16 am

That's a very detailed and interesting analysis ThvN. I have to say that the artillery stats seem to work quite well to provide balance for the German units as looking at my roster in 42 East I have a mix of 15cm and 17cm because I want some hard hitting but I also want the greater ammo capacity and improved chance of suppression on less well protected targets. I have a 21cm Nebelwerfer because the ROF in combination with high soft attack is very good against heavily entrenched infantry, and I have a Stug 3 for tank support and for the high ROF and fortkiller ability, and a couple of Sig33, though the low ammo and GD and AD are an issue. So if I'm using a wide mix of units for different purposes, the overall picture is probably about right, even if the details are open to debate.

I'm not convinced the mix is quite right for AC artillery units (the US mobile unit with unlimited ammo seems a bit too good for the price), although there are some interesting decisions to be made with the high defence and short range of Bishop rather than longer range and weak defence of M7 Priest.

I would like to see ROF and switchability on the purchase screens, so it's easier to understand the trade offs, although forums and Control + L help get the message through.

While I like the 1.21 tweaks for AT I still think a general 3 or 4 INI upgrade to reflect the relative ease of concealment and the low profiles of the guns and probably 2 for the mounted ones would lead me to use them more - I usually have one but rarely more. However, this should probably apply only for true (non-switching) AT whether to reflect the better training of crew or the deliberately low profile or more abundant AT ammo.

rezaf
Colonel - Fallschirmjäger
Colonel - Fallschirmjäger
Posts: 1487
Joined: Mon Aug 15, 2011 3:27 pm

Re: What's the reasoning behind artillery attack values?

Post by rezaf » Mon Dec 16, 2013 7:23 am

ThvN wrote:(Amazing analysis of RoF)
ThvN wrote:That the player isn't told this is a big letdown, just like most traits are hidden.
Ok, you convinced me that RoF has it's uses, but we're appearently in agreement that it's an intransparent mechanic and should be better communicated to the player.
ThvN wrote:'Selectable' means that I want it as an option, like the 'dice chess' option. I think the random combat results are not that bad, it's mostly the +1 or +2 INI dice roll at the start of the combat that can mess things up.
So you want an option that basically makes each unit have 0 INI (or some other, equal INI value), so there are no advantages stemming from that stat.
Did you run any tests with giving all units equal INI?
ThvN wrote:(Equally amazing analysis of INI)
The analogy is SLIGHTLY flawed, since the opponent having less shots each time you "hurt" him isn't reflected well, but other than that, bravo. 8)
ThvN wrote:(Stuff about aircraft)
Hmm. I'm not sure the low/high flying aircraft thingy is really necessary. I hardly ever strafe with my fighters, but the AI LOVES to do it, sometimes even neglecting juicy unescorted TacBombers in the process.
AND the AI is bad at deciding when (and when not) to switch - it's a massive improvement to the early AI, which was unable to switch AT ALL, but still.

Aircraft having different speed at different height is all good and well, but a turn takes a day, so I'm not convinced raw speed shouldn't play an important role. However, you are absolutely correct when it comes to this (and also the effective range of artillery pieces etc. you mentioned earlier), it SHOULD play a role how fast a plane is able to "field" it's advantages. A problem, though, is that this is sometimes arguable (some say x is better than y, some say vice versa) and in a lot of cases, you're not going to be able to find any info on this AT ALL. I had a hard time figuring out the production numbers of some of the Bf109 variants, for example (and I didn't get numbers in all cases in the end), trying to find reliable info on their raw performance in combat conditions (compared to some obscure russian plane and some polish double decker) would probably be a nightmare, if not an excercise in futility.

Another issue I ran into when trying to come up with values for new planes is the game is handling their movement range in a highly inconsistent fashion. Sometimes, a plane you'd think of as fast has high movement and another one you'd think of as slow has less movement - but it isn't consistent. For example, from what I understand the Ju 52 was a rather slow moving plane, but it has relatively high movement - after I had added some seaplanes and game them pretty low MOV, I added the Ju 52 Behelfsbomber and began with copying the stats of the transport plane, thereupon realizing it's high MOV value (I think I gave the slowest seaplanes a value of 8 or 9).
Often enough, I ended up just leaving such inconsistencies alone for the time being. If I dive into them all, I'll never get a base version finished people like you can help improve, after all. :wink:
ThvN wrote:Same here. I'm still busy figuring out good stats for switchable AAA units. At first I made them too powerful, after I fixed that I realized it might be a better idea to not switch them all into 'AT' units. But if I change one thing I have to make it more expensive, and start redoing other things as well.
Generally, mine are usually NOT worth switching - but should they? From what I understand, with VERY few exceptions (8.8 being the most famous and obvious one), switching was an emergency measure to provide SOME means of (more or less) heavy firepower when no other was available. In the context of the game (broken record: a turn is a day), switching makes no sense anyhow. For an AA unit operating independendly, I think the only situation where it'd consider "switching" was in defensive role, which is nicely represented by it's negative "attack" values. But hey, people (guille, for the most part) bothered to make the artwork, so I'll include the switched state. Whether it's useful or not, I'll ignore for now.

Phew, I hope I addressed most of your major points, that was quite a lengthy reply of yours - thanks for bothering to write it.
captainjack wrote:While I like the 1.21 tweaks for AT I still think a general 3 or 4 INI upgrade to reflect the relative ease of concealment and the low profiles of the guns and probably 2 for the mounted ones would lead me to use them more - I usually have one but rarely more. However, this should probably apply only for true (non-switching) AT whether to reflect the better training of crew or the deliberately low profile or more abundant AT ammo.
But how much does it help that you are able to conceal (or dig in) better when what you want to do is attack? This is the main issue I have with the towed AT guns, in a vast majority of situations, their role should be defensive, in a game that's all about offense. Even the defensive scenarios usually involve losing objectives and then turning the tide and recapturing them. When placed defensively, even the BEST AT guns usually disappoint, plus they are one-trick-ponies that will have a hard time dealing with INF. Pre-placed AT guns are usually literally overrun in no time at all. Defensive artillery vs. tanks is often worthless. Most INF, at least the '43 variants, have enough HA to make them able to at least put a dent into an attacking tank, especially in close-combat, and of course they'll fare much better against INF. So why get a towed AT piece?
Plus, in a twist to the classic saying of "the best defense is a good defense", the best defenders in this game are overstrength tanks out in the open, even if they just sit there. The AI is afraid to attack them because of the bad odds, and thus they can literally act as dams to any offensive.
To sum it up: in a world with a hard limit on how much units I can field, this equipment (towed AT) has no place, imo. Even if there was a "floating point unit cap" where most AT pieces only took half a slot or less, I'd personally not want to field them (with the possible exception of the 8.8, in some - but rare - situations).
_____
rezaf

Aloo
Sergeant - 7.5 cm FK 16 nA
Sergeant - 7.5 cm FK 16 nA
Posts: 217
Joined: Mon Oct 31, 2011 12:38 pm
Location: Poland
Contact:

Re: What's the reasoning behind artillery attack values?

Post by Aloo » Mon Dec 16, 2013 2:46 pm

captainjack wrote: While I like the 1.21 tweaks for AT I still think a general 3 or 4 INI upgrade to reflect the relative ease of concealment and the low profiles of the guns and probably 2 for the mounted ones would lead me to use them more - I usually have one but rarely more. However, this should probably apply only for true (non-switching) AT whether to reflect the better training of crew or the deliberately low profile or more abundant AT ammo.
Wouldn't this mess up the balance in most single player campaigns, where there is a lot of AT and most of the computer spam around the last town is AT? With higher INI it would be harder for inf to dig them up without casualties, while it still wouldn't make them a valid choice for the player.

rezaf wrote:
ThvN wrote:'Selectable' means that I want it as an option, like the 'dice chess' option. I think the random combat results are not that bad, it's mostly the +1 or +2 INI dice roll at the start of the combat that can mess things up.
So you want an option that basically makes each unit have 0 INI (or some other, equal INI value), so there are no advantages stemming from that stat.
Did you run any tests with giving all units equal INI?

rezaf
I think he wants the option to turn off the random roll at the start of combat that gives each unit 0,1 or 2 INI. This can change the whole outcome of combat.

Molve
Master Sergeant - U-boat
Master Sergeant - U-boat
Posts: 524
Joined: Wed Jul 27, 2011 10:06 am

Re: What's the reasoning behind artillery attack values?

Post by Molve » Mon Dec 16, 2013 3:51 pm

Aloo wrote:
rezaf wrote:
ThvN wrote:'Selectable' means that I want it as an option, like the 'dice chess' option. I think the random combat results are not that bad, it's mostly the +1 or +2 INI dice roll at the start of the combat that can mess things up.
So you want an option that basically makes each unit have 0 INI (or some other, equal INI value), so there are no advantages stemming from that stat.
Did you run any tests with giving all units equal INI?

rezaf
I think he wants the option to turn off the random roll at the start of combat that gives each unit 0,1 or 2 INI. This can change the whole outcome of combat.
Yes, that's my reading too.

He doesn't want the INI to be zero or some other static number, equal for all units.

He wants the INI to always be the number listed for any given unit, with no variability. Furthermore; in my reading, he asks this being under the assumption that
a) it is this random 0-2 INI bonus that sometimes shifts a combat outcome drastically from the predictions
a) and, that said predictions, does not properly take into account an unlucky INI roll.

Another way of saying this, I believe, is that he wants the combat predictions to have a high accuracy, and that each time the actual outcome differs drastically from the predictions this should be because of some clearly communicated circumstance.

For instance, compare how clearly "rugged defense" is displayed with how "bad INI luck" is displayed. Both cases can have a great impact on the actual outcome; however, while one is exceedingly clearly communicated, the other is an obscure detail of a combat log most players don't even know exists.

(My personal interpretation only.)

ThvN
Panzer Corps Moderator
Panzer Corps Moderator
Posts: 1408
Joined: Fri Jul 06, 2012 8:55 pm

Re: What's the reasoning behind artillery attack values?

Post by ThvN » Tue Dec 17, 2013 12:17 am

To everybody, my thanks for the replies and compliments... and yes, I was trying to say the initial +0/+1/+2 INI roll should be made optional (or reduced to +1 max), the rest of the INI system is fine.

It's just my impression that often, when I get a 'reversed' outcome and I open the combat log, a +2 INI roll caused a big shift in combat outcome. Especially when you consider that this is added after the INI cap for terrain etc. takes place. So you attack, your experience/mass attack bonus gets nullified because of the terrain cap, and even though you have a +1 INI hero, which should guarantee the first shot, you roll a +0 and the opponent +2. This will hurt a lot.

It can also swing aerial combat, which is already notorious for knife-edge outcomes. And of course it will also work in your favour, but the attacker can't count on such rolls, so usually when you get a beneficial roll it will just increase the hits, and it will be taken for granted. It's when it turns against you that you notice it, just like people tend to remember their 'bad' luck better than their 'good' luck.

And like Molve said, it's a bit like rugged defense, which I don't mind, but somehow I have come to dislike the pre-combat ini rolls (at least the +2 seems excessive), as they make combat appear too random in results sometimes. Rugged defense can be frustrating, but it is tied to entrenchment and can be ignored by traits, so as a game mechanism it is acceptable to me and adds a risk that appears manageable. The random ini roll doesn't, it just represents the 'fortunes of war' that are already present in the hit dice rolls I think. So maybe it is the lack of communication, but also the inability to stack things in my favour often enough, so I can at least reduce the risk. On a scale of frustration, in ranks between 'rugged defense' and 'surrounding a submarine and getting 3 EVADES! in a row'

Limiting the roll to a +1 max, like the INI heroes are now, would make sense, but simply adding the option (in-game or in the gamerules file) to tweak this roll would help a lot.


Rezaf, to help explain, the 'high' and 'low' flying planes are just an experiment, to try and break the rigid barriers between the airplane classes (notably the tactical bombers). So don't worry about them, it was just to try and make the point that modding the stats for them can provide interesting dilemmas, and it depends on what you view as important.

And as for the movement rates, during a multiplayer game I managed to get into a train and behind enemy lines. My opponent sent a Bf110 after it to try and bomb it. He wasn't yet aware of the movement rate of the trains... let's just say he was too late to catch the train. These locomotives are some sort of secret weapon, easily twice as fast as any airplane, and faster than a V2 even...

About switching (smaller) AA units: it's a matter of personal preference of course, but although they were initially not often used to combat ground forces except in emergencies, later on it became more common, and especially the SPAAG's could be used offensively. The main reason for me to make them switchable is that they become more useful to have in a players core. So it's not because I think it's more 'realistic', I'm more concerned with gameplay and balance. If the weather turns bad, at least they can lend a hand in ground combat, and they can be very handy in mopping up 1-strength survivors. Right now they offer very little benefit and are very inflexible, like towed AT guns.

To me it makes no sense that a fighter plane can attack ground targets and provide escort for a B-24 at the same time, but an AAA unit has to keep pointing the barrel skywards all the time, especially as some switchable units seem almost randomly chosen (StuGIV). So it's mostly a matter of personal preference, there are good arguments for both 'switchers' and 'non-switchers'.

rezaf
Colonel - Fallschirmjäger
Colonel - Fallschirmjäger
Posts: 1487
Joined: Mon Aug 15, 2011 3:27 pm

Re: What's the reasoning behind artillery attack values?

Post by rezaf » Tue Dec 17, 2013 6:52 am

Ah, thanks for clearing up what you want as optional ThvN.

I agree that this roll is the root cause that makes combat highly erratic. I ran a series of tests long ago which kinda proved the predictions are technically "correct", however there are situations where the RNG just likes to get back at you.
Some things happen WAY too often, especially in these situations. My "favorite" is the "Whip of the Balrog" (alternatively Rudankorts Revenge), where a supposed clean wipe of some unit (0:8 losses prediction on a unit which has 8 Str left) ends up scoring a single counter-kill.
Anyway, like I said earlier, I feel like not revamping these mechanics was one of the few true failings when making PzC.

About your point regarding remembering only bad luck - a major factor is that you usually will only attack in the first place when you have good odds. So noticing a reversal-of-fates victory after a disastrous prediction will almost never happen to you, since you wouldn't attack at such bad odds anyway (funny enough, neither will the AI). So you get to see all the worse-than-expected's and almost none of the better-than-expected's.

Regarding train movement ... funny, isn't it. Once again, 1 turn = 1 day comes to the rescue - one can "explain it" by saying the air range represents the operational range in which the airplane can fly multiple missions over the course of that day, whilst train movement involves bringing in a big train and moving the unit, done. This rationale, of course, is severly lacking, because it doesn't matter where the next airfield is in this context ...
We can also say it is how it is because it is a mechanic brought in by Rudankort and not present in the original PG. I always considered it completely useless, but read in some changelog that it now wastes less turns, so I might consider using it more often now ... maybe.

Your reasoning concerning unit switching is sound, though in the end, you'd need to make them a killer unit for me to consider using them. Like towed AT, I think they make little sense in an offensive core.

Thanks for your reply.
_____
rezaf

Post Reply

Return to “Panzer Corps”