Matrix Games Forums

 Forums  Register  Login  Photo Gallery  Member List  Search  Calendars  FAQ 

My Profile  Inbox  Address Book  My Subscription  My Forums  Log Out

Airfield attack from B17s

 
View related threads: (in this forum | in all forums)

Logged in as: Guest
Users viewing this topic: none
  Printable Version
All Forums >> [Current Games From Matrix.] >> [World War II] >> War In The Pacific - Struggle Against Japan 1941 - 1945 >> Airfield attack from B17s Page: [1]
Login
Message << Older Topic   Newer Topic >>
Airfield attack from B17s - 10/31/2004 8:07:26 PM   
Knaust

 

Posts: 177
Joined: 3/5/2001
From: Rivoli ITALY
Status: offline 		My clever opponent is harassing my Jap airfields with B17s raids at 35000 feets...every time he scores heavy tolls from my planes...on 140 based planes, some 40 got lost from a 24 plane raid!
Those B17 cannot be catched by no Jap plane...however, is it realistic scoring such a heavy toll from that height?
Post #: 1
RE: Airfield attack from B17s - 10/31/2004 8:13:18 PM   
pasternakski


Posts: 6565
Joined: 6/29/2002
Status: offline 		This is a known bug that is ostensibly fixed in the 1.30 poatch ... which is rumored to exist somewhere in cyberspace and may or may not find its way into the hands of all the impatient CUSTOMERS who have suspended play pending its release. There are some other bugs that some of us see as showstoppers (the leader leak, the perpetual US old battleships without radar, and the randomly squirting ground combat retreat effect, to name three of them).

(in reply to Knaust)
Post #: 2
RE: Airfield attack from B17s - 10/31/2004 8:17:23 PM   
KPAX


Posts: 735
Joined: 6/3/2004
From: Where the heart is; Home of the Fighting Irish
Status: offline
quote:

ORIGINAL: Knaust

My clever opponent is harassing my Jap airfields with B17s raids at 35000 feets...every time he scores heavy tolls from my planes...on 140 based planes, some 40 got lost from a 24 plane raid!
Those B17 cannot be catched by no Jap plane...however, is it realistic scoring such a heavy toll from that height?


As Basternakski so well put it.....

This is a known bug.

You need to contact your opponent and tell him to stop flying at 30k or above.

He may not be aware of this, so you may need to refer him to the forum.

_____________________________

"War makes Heros on both sides." Hero (the movie)



Thanks !!

KPAX

(in reply to Knaust)
Post #: 3
RE: Airfield attack from B17s - 10/31/2004 8:20:00 PM   
Knaust

 

Posts: 177
Joined: 3/5/2001
From: Rivoli ITALY
Status: offline 		thx....does new patch need restarting the game?...and in the meanwhile, which house rule must the Allied player use?

(in reply to pasternakski)
Post #: 4
RE: Airfield attack from B17s - 10/31/2004 8:21:25 PM   
Knaust

 

Posts: 177
Joined: 3/5/2001
From: Rivoli ITALY
Status: offline 		thx KPAX...ouch...I had to contact the forum a while before...I lost 300 planes on the ground till now

(in reply to KPAX)
Post #: 5
RE: Airfield attack from B17s - 10/31/2004 9:11:22 PM   
Platoonist


Posts: 1342
Joined: 5/11/2003
From: Kila Hana
Status: offline 		Patch shouldn't require a re-start although if you want all the OOB changes you should. You and your opponent should also beware of the over-effectiveness of night air attacks as well. Somes like there is always one more bug to kill.

_____________________________


(in reply to Knaust)
Post #: 6
RE: Airfield attack from B17s - 10/31/2004 9:35:59 PM   
freeboy

 

Posts: 9088
Joined: 5/16/2004
From: Colorado
Status: offline 		I think the actual flight level to error out is 33200 or there abouts

(in reply to Platoonist)
Post #: 7
RE: Airfield attack from B17s - 10/31/2004 9:58:24 PM   
Mr.Frag


Posts: 13410
Joined: 12/18/2002
From: Purgatory
Status: offline 		It is a math problem with signed integers.

Max is 32767 feet which is 7fff in hex, adding one more makes it 8000 in hex which is actually a negative number so the accuracy numbers go through the roof.

(in reply to freeboy)
Post #: 8
RE: Airfield attack from B17s - 11/1/2004 9:34:28 AM   
Knaust

 

Posts: 177
Joined: 3/5/2001
From: Rivoli ITALY
Status: offline 		uhmm...I don't think it is a matter of altitude...it seems to me that 40 planes lost on the ground out of 140 based to an attack of 24 B17s flying at 35000 feet is far too excessive.
Otherwise this is the secret Allied weapon in the starting days

(in reply to Mr.Frag)
Post #: 9
RE: Airfield attack from B17s - 11/1/2004 10:22:38 AM   
michaelm75au


Posts: 13500
Joined: 5/5/2001
From: Melbourne, Australia
Status: offline 		The problem is that the B-17 attack is currently TOO accurate at the excessive (>32K) altitude due to this bug.
When the patch is released, the accurracy will be corrected and you probably wont get the same result of 40 out of 140 destroyed.
Michael

< Message edited by michaelm -- 11/1/2004 6:05:02 PM >

(in reply to Knaust)
Post #: 10
RE: Airfield attack from B17s - 11/1/2004 12:25:01 PM   
Mr.Frag


Posts: 13410
Joined: 12/18/2002
From: Purgatory
Status: offline 		A little math lesson for those interested in why bugs happen.

a signed short integer can hold decimal values from -32,768 to +32,767

a unsigned short interger can hold decimal values from 0 to 65535

The difference between a signed and unsigned value is the leftmost bit is reserved for the sign.

This is known as 2's complement (-2 to the power of n-1 through +2 to the power of n-1 -1) where n is the number of bits used.

A 16 bit number in our case becomes a 15 bit number + the sign bit.

in decimal, 32767

in binary, 0111 1111 1111 1111

when you look at the binary it becomes very obvious what happens when you add 1

1000 0000 0000 0000 is the new value.

Looks fine but in reality due to the sign bit, this is -32768, not 32768. As you continue to add 1, 32769 is actually -32767, 32770 is -32766 etc all the way until 65535 is reached which is -1.

Since the routine happened to be checking alt vs effectiveness, attacking from a large negative altitude results in the effectiveness going into the 100%+ range.

(in reply to michaelm75au)
Post #: 11
RE: Airfield attack from B17s - 11/1/2004 12:52:20 PM   
Knaust

 

Posts: 177
Joined: 3/5/2001
From: Rivoli ITALY
Status: offline 		damn...who will pay me for over 300 planes lost on the ground due to these B17 attacks?

(in reply to Mr.Frag)
Post #: 12
RE: Airfield attack from B17s - 11/1/2004 1:10:38 PM   
Oliver Heindorf


Posts: 1911
Joined: 5/1/2002
From: Hamburg/Deutschland
Status: offline 		why ou guys used a signed bit ? since a plane cant fly below the ground height....why using the signed bit ? just courious

_____________________________


(in reply to Knaust)
Post #: 13
RE: Airfield attack from B17s - 11/1/2004 2:00:32 PM   
Mr.Frag


Posts: 13410
Joined: 12/18/2002
From: Purgatory
Status: offline
quote:

why using the signed bit


I would assume that it was not intentional ... compilers are funny things at times, one finds out that in one version the defaults are one way and in another, they are another way. Probably just a missing type declare statement somewhere and it happened to default the wrong way.

(in reply to Oliver Heindorf)
Post #: 14
RE: Airfield attack from B17s - 11/1/2004 3:04:23 PM   
Barlock

 

Posts: 12
Joined: 2/26/2004
From: Sydney, Australia
Status: offline
quote:

ORIGINAL: Knaust

damn...who will pay me for over 300 planes lost on the ground due to these B17 attacks?


Win the war - then you can sue the Allies for reparations!

(in reply to Knaust)
Post #: 15
RE: Airfield attack from B17s - 11/1/2004 3:10:49 PM   
Oliver Heindorf


Posts: 1911
Joined: 5/1/2002
From: Hamburg/Deutschland
Status: offline 		thanks Ray , this explains it ( probably )

btw : at 35000 feet you ait see a s**t and no, usually, all bombs should land into the nirvana as it is just too height esp in the pac war area because of the foggy air ( too much water in the air becasue of the temperature - dont know the word Luftfeuchtigkeit in english )

_____________________________


(in reply to Barlock)
Post #: 16
RE: Airfield attack from B17s - 11/1/2004 3:22:50 PM   
Raverdave


Posts: 6520
Joined: 2/8/2002
From: Melb. Australia
Status: offline
quote:

ORIGINAL: Mr.Frag

A little math lesson for those interested in why bugs happen.

a signed short integer can hold decimal values from -32,768 to +32,767

a unsigned short interger can hold decimal values from 0 to 65535

The difference between a signed and unsigned value is the leftmost bit is reserved for the sign.

This is known as 2's complement (-2 to the power of n-1 through +2 to the power of n-1 -1) where n is the number of bits used.

A 16 bit number in our case becomes a 15 bit number + the sign bit.

in decimal, 32767

in binary, 0111 1111 1111 1111

when you look at the binary it becomes very obvious what happens when you add 1

1000 0000 0000 0000 is the new value.

Looks fine but in reality due to the sign bit, this is -32768, not 32768. As you continue to add 1, 32769 is actually -32767, 32770 is -32766 etc all the way until 65535 is reached which is -1.

Since the routine happened to be checking alt vs effectiveness, attacking from a large negative altitude results in the effectiveness going into the 100%+ range.


I read this four times now and I am buggered if I can understand what he is saying

_____________________________




Never argue with an idiot, he will only drag you down to his level and beat you with experience.

(in reply to Mr.Frag)
Post #: 17
RE: Airfield attack from B17s - 11/1/2004 3:32:33 PM   
Mr.Frag


Posts: 13410
Joined: 12/18/2002
From: Purgatory
Status: offline 		You sure you want more math? It only hurts more from here on.

Unsigned integers: (0-15)

0 = 0000
1 = 0001
2 = 0010
3 = 0011
4 = 0100
5 = 0101
6 = 0110
7 = 0111
8 = 1000
9 = 1001
10 = 1010
11 = 1011
12 = 1100
13 = 1101
14 = 1110
15 = 1111

In twos complement, the leftmost bit becomes the sign of the number (0 for positives, 1 for negatives)

Signed Integers (-8 to +7)

0 = 0000
1 = 0001
2 = 0010
3 = 0011
4 = 0100
5 = 0101
6 = 0110
7 = 0111
-8 = 1000
-7 = 1001
-6= 1010
-5 = 1011
-4 = 1100
-3 = 1101
-2 = 1110
-1 = 1111

To convert a negative, one applies the twos complement and adds 1:

looking at lets say -5, we see 1011. We ignore the sign bit (1)011 and reverse the other bits:

011 becomes 100 ... now look up 100 and we see that = 4. To this we add the 1. and we get 5. To this we add the minus bit and end up with -5 ... simple stuff

Anytime you do math and the number of bits exceed the number of bits available to store the number, you get an overflow condition.

With unsigned math (using the above 4 bit values) you can see that adding 1 to 15 or 0001 + 1111 is going to result in 16 or 10000 (hmm, wheres that extra bit go? we overflowed into the value sitting beside us in memory. whoops!)

With signed math (using the above 4 bit values) you can see that adding 1 to 7 or 0001 + 0111 is going to result in -8 or 1000 (hmm, whoops, we just broke something)

Detecting unsigned overflow is a lot easier then signed overflow as it corrupts the value in the next location. Signed overflow does not corrupt anything, it just converts the number into something unexpected.

Now, in our particular bug ...

we are using much larger numbers, but the rules are the same ...

0111 1111 1111 1111 (32767) + 0000 0000 0000 0001 (1) = 1000 0000 0000 0000 (-32768)

< Message edited by Mr.Frag -- 11/1/2004 8:49:59 AM >

(in reply to Raverdave)
Post #: 18
RE: Airfield attack from B17s - 11/1/2004 3:37:28 PM   
Oliver Heindorf


Posts: 1911
Joined: 5/1/2002
From: Hamburg/Deutschland
Status: offline
quote:

ORIGINAL: Mr.Frag

You sure you want more math? It only hurts more from here on.



you cant annoy me with that , no, you cant. But think about the poor Raver hehe, have mercy with him

_____________________________


(in reply to Mr.Frag)
Post #: 19
RE: Airfield attack from B17s - 11/1/2004 3:53:45 PM   
Mr.Frag


Posts: 13410
Joined: 12/18/2002
From: Purgatory
Status: offline 		BTW: This is exactly the same problem people ran into when putting devices in slots past 255. The device location for ship weapons is an unsigned 8 bit value or 255 = 1111 1111.

Adding 1 to it does not get you to slot 256, it gets you to slot 0 with the 1 0000 0000 (extra bit) being ignored.

(in reply to Oliver Heindorf)
Post #: 20
RE: Airfield attack from B17s - 11/1/2004 3:58:10 PM   
tsimmonds


Posts: 5498
Joined: 2/6/2004
From: astride Mason and Dixon's Line
Status: offline
quote:

Luftfeuchtigkeit
= humidity (what you feel), or haze (what you can't see through)

_____________________________

Fear the kitten!

(in reply to Oliver Heindorf)
Post #: 21
Page:   [1]
All Forums >> [Current Games From Matrix.] >> [World War II] >> War In The Pacific - Struggle Against Japan 1941 - 1945 >> Airfield attack from B17s Page: [1]
Jump to:





New Messages No New Messages
Hot Topic w/ New Messages Hot Topic w/o New Messages
Locked w/ New Messages Locked w/o New Messages
 Post New Thread
 Reply to Message
 Post New Poll
 Submit Vote
 Delete My Own Post
 Delete My Own Thread
 Rate Posts


Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI

1.221