[-Pv-]

You can go too far improving the DD AI in SH4 in the Pacific compared to what you saw in the Atlantic. One major problem the Imperial forces had was bad intel on US sub performance. One major point is despite the US subs not diving as deep as German subs, the Japanese DDs set their max DC depth even shallower believing the US subs didn't go as deep as they really could. The US subs used welded boats sooner than the Germans and could take a pounding. Although the US fleat boats were slower than hoped, they were faster and generally larger, longer-ranged than German boats. Because of this it was possible to evade escorts on the surface at night using speed and low profile. The patrol area was much larger, but the US boats could cross the distance a little faster, and patrol farther.

One difficult feature in DD AI is the logic behind DD tactics. Some convoys would only have one escort. More valuable convoys would have several, but the AI needs to decide how much attention to give the sub (keeping it submerged until the remaining convoy gets away) and when to abandon the sub and run at full speed to get back to the convoy. As much as we all like a good knife fight, if the programmers get it right, there will be times when a more persistant and skillful DD will pound you relentlessly, and times when they will completely ignore you once they have you slow and low so they can stay with the running convoy. Also, if you've played Destroyer Command (I assume most here have) you also realize on the DD side you can't sit there and drop your entire DC load on one target (esp if it appears to be particularly elusive) because you have to save something for the next fight or your value as an escort for the remaining route is null.

A hunter-killer group is expected to persue until all possibility of regaining contact is lost or ammo expended. Escorts however will favor keeping the convoy ships in view even if it means abandoning a contact. Sub hunter-killers were rare for Japan in the Pacific because they had a shortage of escort ships over an area much larger than the Atlantic theater. Convoys were generally smaller than we saw with the Brittish and US in SHIII, but a valuable convoy with large target ships will be heavily and aggressively escorted. The Japanese gunnery on smaller ships was considered above average making surface actions in less than ideal conditions (in favor of the sub) risky.

"Know when to hold them, know when to fold them, know when to walk away, know when to run."

I realize we are all just shooting the brease here for our own amusement. The programmers will do what interests them with the time and budget they have appropriated. I guess asking the community to maintain realistic expectations is a waste of bandwidth. Since the SHIII dev team demonstrated above-average attention to detail, provided an affordable experience, and patched the most game-stopping issues promptly, I expect any forthcomming product to meet the same standards, so I'll buy it if the sub gammer sites review it possitively, especially since this is the sub battle theater I enjoy the most.

Special ops were very important aspects of the Pacific Island hopping campaign. Gathering port intel, picking up and dropping off special forces near islands and picking up downed flyers near the major sea battles would be important features to me to give the sim a period and location feel.
-Pv-

[DeepSix]

Very well stated. I especially like the skillful incorporation of a Kenny Rogers "Gambler" quote.

[oche]

It appears that some people really miss the laser guided depth charges from SHII since they wish to see even more precise depth charge attacks in SH4...that's not realistic in any way and they know it, japanese sonar and ASW capabilities was crap compared to US and British technology.

[Beery]

[The Noob]

Okay, make it Like in Silent Service 1!

Make an Realism Option!

Expert Destroyers On = AI Gets Improved

Expert Destroyers Off = Destroyers Get as Inneffecient as it Would be Realistic!

That Would be an Solution for that Fu*kin' Problem! Both Sides Would be Pleased................

[Safe-Keeper]

[Beery]

[-Pv-]

Good job Beery. You expressed what I have experienced in SHIII and what I expect for SHIV. The Rock was feared as a heavily patrolled chokepoint, but not impassible. I think you will find the many very shallow harbors in the Pacific very challenging also. The inland Japanese convoys would thread very complex jump and dodge courses through the islands taking advantage of shallow water, night, moon out to sea and the complications islands and reefs made in the attempt to get a good firing solution.
-Pv-

[Godalmighty83]

in sh3 destroyers were too effective, from the right viewing angle you could se the DD's rudder moving exactly as you moved your meaning a great deal of AI cheating was going on.

that and the infinte supply of rapid fire depthcharges which were always set to the perfect depth.

[The Noob]

What about my Solution?'

Wouldn't it Be The Best of all?

[Sulikate]

[Subnuts]

I've been playing the original Silent Hunter recently, running patrol encounters at 60% realism (limited depth data, visability, and realistic charts off) and laughing at the schizophrenia of the escort AI. If you dive below a thermal layer (which don't seem to be implemented until after you get your bathythermograph) and jettison debris, the escorts, and every aircraft in the area, will gang up on the debris and depth charge it over and over again.

After a while I got bored and started shooting torpedoes at random. When the torpedoes exploded, the DDs would go chasing after the spot where it exploded and depth charge it. I fired three torpedoes, each in a different direction, and the escorts went nuts and chased after each explosion. Then I came up to periscope depth and released more debris. A passing Zero sighted it and crashed into it!

The escorts came back and I stayed at PD until they detected me. Then I dived back below the thermal layer, accelerated to ahead full, releasing debris everywhere. Again, the escorts ran around, DCing the debris. A second Zero roared in, and I heard a loud BAM! I came up to PD, looked around, and noticed that one of the destroyers was on fire! I dove back below the layer, and kept releasing debris until the escorts ran out of DCs and went away.

[Trout]

I dont think the problem is about the skill level of the hunters or their technolgy, it is:

1) Hunters give up WAY too easily as compared to RL

2) There is no strategic AI so that it gets hotter for you the longer you stay in an area

3) OUr subs are not difficult enough to handle while being depth charged. See below:


We all know that the damage modeling is not complete enough in the game. THere are all kinds of minor damage that we simply dont take (even very minor leaks) that are survivable, but make evasion more difficult. How would you like to maintain your depth with broken gauges, bent planes, jammed valves or a sudden change in dive angle because a charge went off over you? Or how about damage that makes your boat louder? (there are many kinds!)

And fixing ANY kind of damage basically means you can be heard better underwater.

Perhaps if the focus was put on fuller modeling of our boats, the hunters would pick up more sounds which would "encourage" them to stick around longer?

Any way you slice it, SOMTHING needs to be done to make evasion more of a challenge and more fun (but perhaps more of a "patrol ender" than a "life ender" when you do get seriously tagged).

By the way, have you EVER run out of battery or air in this game? GEtting held under for many hours and having a very scary experience was a common thing for submariners. It hardly ever happens in this game, although when it does you typically die in the end.

Trout

[Beery]

I've said it before, and I'm sure I'll say it again: surviving in a SH3 U-boat is far MORE difficult than it was in real life. Nothing needs to be made harder for the sub except perhaps for aircraft attacks. Certainly, depth charge attacks are MORE deadly than they were in real life, and U-boats get killed by DDs FAR more easily in this game than in real life. I've posted stats on this elsewhere on this site (I'll repost them below if I can find them). I don't know what else I can do to prove that U-boats weren't as easy to kill as some players want them to be.

Here's the post I was referring to. What I did was take real life stats and figure out what the game should give the player if it was accurate in terms of survivability.

[-Pv-]

In response to Trout,

I assume you are talking about your experiences in SHIII compared to what you expect in SHIV.

Don't expect DD AI in SHIV to ALWAYS be more aggressive or lethal than SHIII. That would not be historically accurate. However due to the fact that 52 subs were lost in the Pac war and 5,200 submariners died, there should be times when you unexpectedly meet a lethal or lucky opponent.

Concerning the lack of hot zones due to sub activity, I have definitely seen evidence in SHIII that the AI does respond to attacks and sinkings where if I stuck around I would see aircraft and hunter patrols show up eventually. The ruling factor in SHIII appears to be spotting. If I was SEEN by any enemy (even if I shot them down or sunk) there would be follow-up patrols by the enemy. When this happens, I quietly leave the area or suffer unending attacks.

In the Pac the best hunting should be near ports, but they will also be heavily patrolled making them very dangerous.

Historically, 1.5 percent of the nation's naval manpower sinking one-half of the Japanese tonnage would indicate there will be a target-rich environment and the fleet boats will be tough to sink compared to the Germans. In addition, the Pac boats had the latest technology the Allies could produce (with the exception of the torps.)

How do you know that in SHIII there were no damage or noise effects that affected your chances of detection and being sunk?
Just because you didn't get a text message telling you "...the compressor is making too much noise sir..." doesn't mean the game logic did not account for some of these things. Remember when you selected Silent Running you elected to forsake repairs and the noise that made. That is indication SHIII likely had some logic in place for the effect of repair noise.

I think it's important in our speculations to avoid the these assumptions:
1) The game logic in SHIV will be unaltered ports of SHI or III.
2) SHIV will be like SHIII with different visuals and map.

I've seen no evidence these two are likely based on past releases.

What controls game development more than anything are:
Skill, size, and enthusiasm of the development team.
Time/Budget.
Current devel technology available.
Target consumer technology available.

Market pressures as in potential sales and purchaser requests are a relatively small part of the picture (except for the facets that contributed to the investment decisions) although SHIII worked hard at the last moment to include a requested feature. The dynamic campaign came close to not being there.

When programmers can make AI as smart as humans, it's time for them to stop creating games and switch to creating millennium man robots. They'll make a lot more money. Games that cannot be won by humans playing at the highest reality level will not be well-received by the majority of gamers. Each person has their own challenge assessment. Some think the SHIII at highest is too hard, some too easy. For me, I can no longer spend 50 hours a week gaming.

What I CAN expect from past releases:
1) Game play will be more or less historically pertinent to the theater and time period.
2) Graphics will exceed expectations and game logic will take a measurable back seat.
3) There will be notable bugs at release that will be patched after sales have begun and the sales are positive enough to warrant the added investment.

Point two is unfortunately due to human behavior. Many people will buy and play a game once or twice even if they're uninterested if the graphic representation wowed them. Some people will play a flawed game for the graphics only. I call these people TV Gamers.
Lots of eye movement, little cognitive activity.
-Pv-