Matrix Games Forums

Forums  Register  Login  Photo Gallery  Member List  Search  Calendars  FAQ 

My Profile  Inbox  Address Book  My Subscription  My Forums  Log Out

Airplanes

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

Logged in as: Guest
Users viewing this topic: none
  Printable Version
All Forums >> [New Releases from Matrix Games] >> War in the Pacific: Admiral's Edition >> Scenario Design and Modding >> Airplanes Page: [1] 2 3 4 5   next >   >>
Login
Message << Older Topic   Newer Topic >>
Airplanes - 9/10/2013 9:11:45 PM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline
Wandering into the world of airplanes is a totally new experience for me. But I received a gift of the original calculation parametrics from a friendly Air Team soul. So off I go through airplane life, fat, drunk, and stupid, with my programs in one hand and a stack of TAIC reports in the other.

Just doing Fighters and FBs because they are the ones mainly impacted by the a2a routines. Data is collected at specific power ranges and at a plane’s several critical altitudes (speed, climb, power). All data is normalized, in order to be self-consistent and rationally related across nations and years.

Several IJ planes will show up much more nicely, several Allied planes will have their beards trimmed, some planes will perform well at higher altitudes, some less so, some planes can operate at bomber altitudes, most can’t.

Oh, yummy! Franks and Georges, flown by good pilots, against Hellcats and Corsairs in a later war scenario? Woof !!! Fun stuff.

[ed] but who knows who can effectively oppose the high altitude 4E bomber stream? The Shadow knows, muaha ha ha ha haaa!

Ciao, JWE

< Message edited by Symon -- 9/10/2013 9:17:41 PM >


_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.
Post #: 1
RE: Airplanes - 9/10/2013 11:54:42 PM   
MateDow


Posts: 218
Joined: 8/6/2002
Status: offline
I am looking forward to seeing the data.

Are the calculations something that will be shareable?

(in reply to Symon)
Post #: 2
RE: Airplanes - 9/11/2013 2:41:02 AM   
GaryChildress

 

Posts: 6830
Joined: 7/17/2005
From: The Divided Nations of Earth
Status: offline

quote:

ORIGINAL: MateDow

I am looking forward to seeing the data.

Are the calculations something that will be shareable?


I second that! Any chance more of the internal workings of the AE system might be made public so modders might have an easier time of it?

Thanks.

_____________________________


(in reply to MateDow)
Post #: 3
RE: Airplanes - 9/11/2013 9:21:40 AM   
HerzKaraya


Posts: 195
Joined: 1/19/2005
From: Madrid, Spain
Status: offline
I would like to see those formulas too! Or at least how much "weight" is given to each parameter (Wingloading, HP/Weight ratio, rate of turn, rate of roll, turn radius, speed, acceleration, dive speed,...) when assigning a MVR value.
And then how important is Max Speed in relation to MVR when pitching to fighters against each other? Or does the better pilot choose which attribute he is going to use against the other? (Slashing attack or tight dogficht)

Thanks

_____________________________

Vista, suerte y al toro!

(in reply to GaryChildress)
Post #: 4
RE: Airplanes - 9/11/2013 9:56:53 AM   
Terminus


Posts: 41459
Joined: 4/23/2005
From: Denmark
Status: offline
Pretty sure we're not allowed to make those calculations public.

_____________________________

We are all dreams of the Giant Space Butterfly.

(in reply to Symon)
Post #: 5
RE: Airplanes - 9/11/2013 1:29:36 PM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline
quote:

ORIGINAL: Terminus
Pretty sure we're not allowed to make those calculations public.

Absolutely right. If it belongs to Matrix or Henderson Field Designs the algorithms are under non-disclosure.

Besides, most of the new stuff is just ordinary, everyday data. Speed - self explanatory, Ceiling - we're using operational ceiling (the 1000 fpm limit) as opposed to max, Cd0 to get mo' bedda values for FB configurations, that sort of thing. Oh, and climb rates - we're using a weighted average of CL, max, and critical altitude.

< Message edited by Symon -- 9/11/2013 2:04:07 PM >


_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.

(in reply to Terminus)
Post #: 6
RE: Airplanes - 9/11/2013 1:55:45 PM   
fcharton

 

Posts: 1112
Joined: 10/4/2010
From: France
Status: offline
But could some agregated statistics per model be made public? Averages would be fine, and not disclose the algorithms (and a variance or two would soothe the more suspicious of us).

Francois

(in reply to Symon)
Post #: 7
RE: Airplanes - 9/11/2013 2:01:08 PM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline
quote:

ORIGINAL: fcharton
But could some agregated statistics per model be made public? Averages would be fine, and not disclose the algorithms (and a variance or two would soothe the more suspicious of us).

Francois

All the individual data things will be fully documented

_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.

(in reply to fcharton)
Post #: 8
RE: Airplanes - 9/11/2013 3:04:09 PM   
PaxMondo


Posts: 9750
Joined: 6/6/2008
Status: offline


_____________________________

Pax

(in reply to Symon)
Post #: 9
RE: Airplanes - 9/11/2013 5:58:01 PM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline

quote:

ORIGINAL: PaxMondo


LOL. But oh! no, no, no. Not tweaking anything that has to do with the calculations. We were always hesitant about approaching airplane data because we were not familiar with how things propagated. Having the programs just makes sure we don't do something with unintended consequences. Evaluating the new stuff shows that the Air Team was thinking in exactly the same conceptual space that we are, so no need to tweak their results. Verification only; Air Team was way smarter than us and it shows .

If the Babes Team comes up with a new calculation algorithm, we'll talk about it. But Matrix/HFD algorithms are private. We think the Air Team's algorithms are righteous. So no disclosure; instead a poop load of new data that works within the game concept

Ciao. JWE

_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.

(in reply to PaxMondo)
Post #: 10
RE: Airplanes - 9/12/2013 1:24:18 AM   
GaryChildress

 

Posts: 6830
Joined: 7/17/2005
From: The Divided Nations of Earth
Status: offline
quote:

ORIGINAL: fcharton

But could some agregated statistics per model be made public? Averages would be fine, and not disclose the algorithms (and a variance or two would soothe the more suspicious of us).

Francois


I don't think the intent should be to "soothe suspicions" or otherwise scrutinize or pick apart the methodology used by the AE team (if that's what you mean). That will only devolve into another "this game is broken" thread and would be immensely counter productive to the game and the modding community.

The point is to make modding even POSSIBLE. It is nigh impossible to mod something that we don't understand except by just guessing numbers or pulling them out of our posteriors. Gamers want mods that work. What modders need are solid methods to come up with new stats in the editor. Right now I don't think the majority of modders are getting what they need to work on their mods. Looking in the editor manual and reading that "device accuracy is the accuracy of the device" helps no one. Modders need formulas or means to producing stats that work in the game.

This is what I think would be a useful entry in the editor manual for something like accuracy for instance:

quote:

"Accuracy is a number between X and Y, the higher the number the more accurate the weapon. Factors that affect accuracy are P, Q, and R. In order to calculate accuracy of a new device you would multiply P by Q and divide by R (or whatever the formula is)."


This would really help tell us four things: 1. What the highest and lowest possible values are, 2. what higher and lower values mean in game terms, 3. what factors affect accuracy, and 4. how to arrive at numbers on our own that are compatible with the game. I don't see the crime or danger in Matrix or Henderson Field publishing the formulas or methods used to arrive at a number in the editor. It would help modders thrive and that in turn would help the game thrive.

EDIT: Snipped out the comments caused by frustration. Don't think it helps the conversation.

< Message edited by Gary Childress -- 9/12/2013 2:54:36 AM >


_____________________________


(in reply to fcharton)
Post #: 11
RE: Airplanes - 9/12/2013 1:24:58 AM   
Commander Stormwolf

 

Posts: 1623
Joined: 2/19/2008
Status: offline
on the a2a model:

1) the co-ordination penalties and abolition of the WITP uber-cap
are a BIG step to improving the a2a model

2) the altitude effects and MVR penalties are wrong and are the core
of why the a2a model is currently broken

3) aerial armament values need to be re-evaluated (particularly accuracy)
recommend my formula

accuracy = (muzzle velocity/3000) x (ammo/500)
______________ ______________


presently A6M2 and N1K2-J have almost the same accuracy
60 rounds vs 200 rds (1970 ft/s vs 2460 ft/s)

4) night defensive gunner accuracy reduced significantly

< Message edited by Commander Stormwolf -- 9/12/2013 1:26:49 AM >


_____________________________

"No Enemy Survives Contact with the Plan" - Commander Stormwolf

(in reply to Symon)
Post #: 12
RE: Airplanes - 9/12/2013 2:13:11 AM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline
You stinkimg piece of wet feces. Leave and do not ever come back.

_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.

(in reply to Commander Stormwolf)
Post #: 13
RE: Airplanes - 9/12/2013 3:58:39 AM   
MateDow


Posts: 218
Joined: 8/6/2002
Status: offline
I was more thinking that, like the values for accuracy and such with guns, we'd have access to something that would allow us to put our own planes in. Something like for wing load X it will give a maneuver rating of Y. I'm sure that there would be more variables. I'm really not interested in how the game uses that number to determine combat, just how to give it numbers that will be used properly.

Hopefully that makes sense.

(in reply to Symon)
Post #: 14
RE: Airplanes - 9/12/2013 4:46:53 AM   
GaryChildress

 

Posts: 6830
Joined: 7/17/2005
From: The Divided Nations of Earth
Status: offline
I suppose "internal workings" was a bad way of me putting it. Not really what I meant either. It sounds like we are looking for the same thing MateDow, just a way of introducing new devices, planes, etc into the editor and have it actually work correctly in game. Right now I'm having a devil of a time trying to figure out how to add the German Pocket BBs for my whacked out mod. They're a lot UNLIKE anything currently in the game. So I'm not sure what values to assign them. Wish I had some sort of simple formula for giving them compatible stats for maneuver.

< Message edited by Gary Childress -- 9/12/2013 4:47:40 AM >


_____________________________


(in reply to MateDow)
Post #: 15
RE: Airplanes - 9/12/2013 3:44:31 PM   
Dili

 

Posts: 4708
Joined: 9/10/2004
Status: offline
Energy fight should only work one level/altitude band down . High altitude fighters going to low level would always under perform.

Bombers should have also altitude bands, this specially to affect their range.

_____________________________


(in reply to GaryChildress)
Post #: 16
RE: Airplanes - 9/12/2013 4:30:52 PM   
Terminus


Posts: 41459
Joined: 4/23/2005
From: Denmark
Status: offline

quote:

ORIGINAL: Gary Childress

I suppose "internal workings" was a bad way of me putting it. Not really what I meant either. It sounds like we are looking for the same thing MateDow, just a way of introducing new devices, planes, etc into the editor and have it actually work correctly in game. Right now I'm having a devil of a time trying to figure out how to add the German Pocket BBs for my whacked out mod. They're a lot UNLIKE anything currently in the game. So I'm not sure what values to assign them. Wish I had some sort of simple formula for giving them compatible stats for maneuver.


They're not that unusual. Ship maneuverability is a function of tonnage, ship length and speed. Find a warship in the DB that's close enough and then fudge some stats from that. That's what I do.

_____________________________

We are all dreams of the Giant Space Butterfly.

(in reply to GaryChildress)
Post #: 17
RE: Airplanes - 9/12/2013 6:37:28 PM   
Dili

 

Posts: 4708
Joined: 9/10/2004
Status: offline
It is also a a function of steering gear including its area and propellers configuration and propulsion control, for example Bismarck was unable to maneuver with propellers only.

_____________________________


(in reply to Terminus)
Post #: 18
RE: Airplanes - 9/12/2013 6:42:05 PM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline

quote:

ORIGINAL: MateDow
I was more thinking that, like the values for accuracy and such with guns, we'd have access to something that would allow us to put our own planes in. Something like for wing load X it will give a maneuver rating of Y. I'm sure that there would be more variables. I'm really not interested in how the game uses that number to determine combat, just how to give it numbers that will be used properly.

Hopefully that makes sense.

It does make sense. Without giving anything away, it's pretty clear that maneuverability is a function of 'loopability' and acceleration, in turn, a function of wing loading and HP/wt ratio. That's because it was, to the first order, and the air guys weren't dumb. So come up with your own calculations of WL and HP/Wt and get a number. Then apply that number to the game number and find your normalization factor (what you have to multiply your number by in order to get the game number). Rinse and repeat a couple times to make sure you got a good normalization factor (there are outliers that will catch you up). Then do the same calcs for your new airplane, apply the normalization factor and you should be in the ball park.

It doesn't take much inspection to figure out the maneuver band cut-offs. Figured these out long before I got the programs. It's a simple cut-off at the HP-RoC critical altitude. If you want to get totally gnarly, find your data for each of the maneuver bands

Some things are very important. You must make sure that your weight numbers are consistent, in order to ensure consistency between and among nations. Makes no sense to calc planes with different empty, loaded, operational, max takeoff weights. Babes is using a "clean", loaded standard. Also your HP needs to be unified. Do you use SL, MilP Max, MilP at Crit, WEP Max, WEP at Crit? It must be consistent. Babes uses MilP at Crit because that plays into the maneuver band cut-offs and represents the most general condition. So use that as your HP for HP/Wt calcs.

Takes some work, but it's well worth it. Generates a better understanding of airplane performance characterists while doing the research and figuring out what's up with low and high altitude performance.

Ciao. John

_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.

(in reply to MateDow)
Post #: 19
RE: Airplanes - 9/12/2013 7:01:37 PM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline
quote:

ORIGINAL: Terminus
quote:

ORIGINAL: Gary Childress
I suppose "internal workings" was a bad way of me putting it. Not really what I meant either. It sounds like we are looking for the same thing MateDow, just a way of introducing new devices, planes, etc into the editor and have it actually work correctly in game. Right now I'm having a devil of a time trying to figure out how to add the German Pocket BBs for my whacked out mod. They're a lot UNLIKE anything currently in the game. So I'm not sure what values to assign them. Wish I had some sort of simple formula for giving them compatible stats for maneuver.

They're not that unusual. Ship maneuverability is a function of tonnage, ship length and speed. Find a warship in the DB that's close enough and then fudge some stats from that. That's what I do.

They are not unusal at all, in fact they are pretty mundane. Pocket BBs were pretty pathetic in most all respects. Make a list of Hp/Dspl and Sp/L for ship types (BB, BC, CA, etc..); plot them. Find yours (it's not all that good, you will find it among the CAs). Plot it and find your number. Enter the number. Different ship types have different Constants that multiply the H/D and S/L values. This is because of the maneuver tests being dependant on size and hull form factor. That's how it's done.
{ed] BtW that's a bit OT for an airplane thread.

Ciao. JWE

< Message edited by Symon -- 9/12/2013 9:33:20 PM >


_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.

(in reply to Terminus)
Post #: 20
RE: Airplanes - 9/14/2013 4:50:48 PM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline
Sort of at, not really a quandary, rather a speed bump. Climb numbers are all over the map. They are consistent with what’s in the mass-market stuff, but that’s what seems to be the source of the inconsistency. Publications often just report “Rate of Climb” without saying where it comes from. If it does say SL, it doesn’t give any hints as to the power regime.

So we looked at RoC, from the test reports, at several conditions. Initial RoC at SL (at both Military and WE Power), RoC at Critical Altitude (again at both Military and WE Power), and an Effective RoC to 20000’ (again at both Military and WE Power), for various aircraft. The results show up the inconsistencies very well.

What I propose is to use Eff RoC to 20000’ at Mil Power as the baseline for our F and FB tweaks. Some of these are better (for both sides), some are worse (for both sides). Some really interesting things crop up for those IJ planes that were later war bomber interceptors; their ERoC to 20k was … good!

These numbers don’t stand alone, they are folded into the new “operational” ceilings (combat ceilings in some references) which are the 1000 fpm points rather than the “gasping for air pressure, how high can I possibly go” 100 fpm points. RoC probably has more to do with available power than speed, so Climb critical altitudes determine the shape of the maneuver band cut-offs. Air Team knew this very well and I haven’t found but 2 planes that wanted a tweak in this regard.

So … Eff RoC to 20000’ at Mil Power as the baseline? That’s where we are going unless somebody says that dog don’t hunt … Ian, Tom, Brian?





Attachment (1)

_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.

(in reply to Symon)
Post #: 21
RE: Airplanes - 9/16/2013 10:46:26 AM   
LoBaron


Posts: 4776
Joined: 1/26/2003
From: Vienna, Austria
Status: offline
Chiming in to say I highly apprechiate that the DaBabes team finally got wings.

The approach looks very sound, I will keep an eye on this. Thank you for enriching the game with your knowledge and dedication.



There is one thing where I would propose a discussion, and this is the Eff RoC to 20k, which I assume is averaged out RoC SL to 20k. The stat is used for scrambling fighters on intercept, and for ACM. Personally I would regard performance in A2A combat the more important aspect, as small discrepancies in scrambling RoC can be rationalized away more easily.

Averaged out RoC SL - 20k might favour the non-turbocharged fighters a bit too much for ACM, which usually happens between fighters already airborne and at some altitude.

This is just a random numbers, you guys will understand where to put the optimal RoC window much better than me, but what about 5k (or even 8k) - 25k as a baseline? This could lessen the negative effect on fighters whose RoC only begins to excell at or over 15-20k, which was of significant advantage in air combat, even if most of the fights took place at lower altitudes.

Just a humble suggestion.

_____________________________


(in reply to Symon)
Post #: 22
RE: Airplanes - 9/16/2013 2:20:24 PM   
Dili

 

Posts: 4708
Joined: 9/10/2004
Status: offline
In my lost post i said that time to altitude which are also avalable is better than climb rates.

_____________________________


(in reply to LoBaron)
Post #: 23
RE: Airplanes - 9/16/2013 6:49:52 PM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline

quote:

ORIGINAL: LoBaron
There is one thing where I would propose a discussion, and this is the Eff RoC to 20k, which I assume is averaged out RoC SL to 20k. The stat is used for scrambling fighters on intercept, and for ACM. Personally I would regard performance in A2A combat the more important aspect, as small discrepancies in scrambling RoC can be rationalized away more easily.

Averaged out RoC SL - 20k might favour the non-turbocharged fighters a bit too much for ACM, which usually happens between fighters already airborne and at some altitude.

This is just a random numbers, you guys will understand where to put the optimal RoC window much better than me, but what about 5k (or even 8k) - 25k as a baseline? This could lessen the negative effect on fighters whose RoC only begins to excell at or over 15-20k, which was of significant advantage in air combat, even if most of the fights took place at lower altitudes.

Just a humble suggestion.

Makes sense.

I'm out of town. Lunch break. Wi-fi is a wonderful thing isn’t it? One of the guys in the group used to fly A4Ds. Been picking his brain unmercifully. Showing him what I’m trying to accomplish and he’s nodding and then says “where’s your blower altitude breaks and what’s your power up there?” Whoah !!! Okey Dokey, then !!! Light bulbs go on like a flash-bang !!!

Don’t know what I can come up with but, as you say, something that weights critical altitude performance. I’ll keep plugging along. You keep thinking.

Thanks. Ciao. John


_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.

(in reply to LoBaron)
Post #: 24
RE: Airplanes - 9/16/2013 7:19:03 PM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline

quote:

ORIGINAL: LoBaron
Chiming in to say I highly apprechiate that the DaBabes team finally got wings.

We've had wings for a while. Just a different kind




Attachment (1)

_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.

(in reply to LoBaron)
Post #: 25
RE: Airplanes - 9/17/2013 1:08:34 AM   
Big B

 

Posts: 4870
Joined: 6/1/2005
From: Old Los Angeles pre-1960
Status: offline
Dog gonnit', somebody told me those were pilot wings!
quote:

ORIGINAL: Symon


quote:

ORIGINAL: LoBaron
Chiming in to say I highly apprechiate that the DaBabes team finally got wings.

We've had wings for a while. Just a different kind





(in reply to Symon)
Post #: 26
RE: Airplanes - 9/17/2013 1:37:11 PM   
Jorm


Posts: 545
Joined: 6/25/2002
From: Melbourne
Status: offline
goodo,

but it seems there will be some WWII a/c that the data you seem to refer to is not available to even the well resourced MODer ?
i.e. all French A/c etc






< Message edited by Jorm -- 9/18/2013 7:35:15 AM >

(in reply to Symon)
Post #: 27
RE: Airplanes - 9/17/2013 4:08:47 PM   
Dili

 

Posts: 4708
Joined: 9/10/2004
Status: offline
I have seldom seen french aircraft manuals, but they certainly exist, i have found almost all Italian aircraft manuals and it is a country not usually followed about in WW2.
It probably needs a research with French terms.

Concerning manueverabiliy for example Macchi 200 an 1940 production low altitude fighter was not faster and had difficulty to be stable at over 8000m altitude than a 4 engine bomber at over that level.

_____________________________


(in reply to Jorm)
Post #: 28
RE: Airplanes - 9/18/2013 2:12:44 AM   
Symon


Posts: 1928
Joined: 11/24/2012
From: De Eye-lands, Mon
Status: offline
Hi LoBaron,

Think we have a decent way of figuring this out. Kinda painful, but does account for altitude performance. I'll put up a preview spreadsheet with climb at various altitudes along with the HP ratings at critical altitudes. Things are making a bit better sense now. Fun stuff. JWE

_____________________________

Nous n'avons pas peur! Vive la liberté! Moi aussi je suis Charlie!
Yippy Ki Yay.

(in reply to LoBaron)
Post #: 29
RE: Airplanes - 9/18/2013 9:50:21 AM   
LoBaron


Posts: 4776
Joined: 1/26/2003
From: Vienna, Austria
Status: offline
Looking forward to it JWE, I knew you´d find a good solution.

Thanks!

Edit: Although I doubt you need it with all the heavyweighters you can mine for knowledge, should you consider another brain helpful for a consistency check of the implementation method please PM, I am happy to help. I am equally happy to just profit from the results of your work though.

< Message edited by LoBaron -- 9/18/2013 10:03:05 AM >


_____________________________


(in reply to Symon)
Post #: 30
Page:   [1] 2 3 4 5   next >   >>
All Forums >> [New Releases from Matrix Games] >> War in the Pacific: Admiral's Edition >> Scenario Design and Modding >> Airplanes Page: [1] 2 3 4 5   next >   >>
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

0.907