Vibration in large scenery

Questions, comments, concerns, & fixes
User avatar
Richard Vo
Veteran
Veteran
Posts: 519
Joined: Mon Dec 14, 2015 8:19 pm
Favorite Aircraft: A380
Location: Sweden
OS: Windows 10
Has thanked: 315 times
Been thanked: 111 times
Contact:

Vibration in large scenery

Post by Richard Vo » Wed May 25, 2016 5:49 pm

When I'm flying in a large scenery (ex: UK, Guam to SoCal 2015), I see that the aircraft is vibrating. Is there any problem? :?:
Regard, Quan Vo
Image

"Airlines will need the re-engined A380!" - me, 2019

By any chance if you have Instagram, please drop a follow on @vhaq.aero . Appreciate that!

User avatar
Neocon
First Class Membership
First Class Membership
Posts: 3839
Joined: Wed Oct 19, 2011 5:01 am
Favorite Aircraft: Baron 58
Location: Tennessee
OS: Win 10
Has thanked: 1605 times
Been thanked: 1726 times
Contact:

Re: Vibration in large scenery

Post by Neocon » Wed May 25, 2016 6:08 pm

That's just how YSF works with big maps. The further from center, the worse the shaking.

User avatar
Ace Lord
Contributor
Contributor
Posts: 383
Joined: Thu Feb 10, 2011 8:45 pm
Favorite Aircraft: Too many to list
Location: Lost Hills Bunker Complex
OS: Windows 10
Has thanked: 52 times
Been thanked: 125 times

Re: Vibration in large scenery

Post by Ace Lord » Wed May 25, 2016 8:56 pm

Remember reading up on this back in the YSP 2.0 days...member by the name of h3csc was trying to work on a 1:10 scale world map but abandoned the project because of this issue.
Image

User avatar
Turbofan
Senior Veteran
Senior Veteran
Posts: 1891
Joined: Thu Feb 03, 2011 4:22 pm
Favorite Aircraft: Boeing 747, all variants
Location: Central OK
Has thanked: 1004 times
Been thanked: 677 times
Contact:

Re: Vibration in large scenery

Post by Turbofan » Wed May 25, 2016 11:18 pm

My educated guess after making maps and watching how the largest maps behave is that the magic numbers are in the range of 300-350 miles. There is still shaking but not to the point of "quaking". Anything farther away and the shaking (or quaking) gets worse quickly.

And yes, I remember h3csc, very talented guy. Wonder what happened to him?
Check out my YSFS videos on my Youtube channel!
-----------------------
ATRP: When used in my posts refers to "Air Traffic Replay Project"
-----------------------
Avatar: Pratt and Whitney JT8D-219
-----------------------
My general rant: I will call them blended winglets, not Sharklets; There is no CEO, it's just simply put, an Airbus 319/320/321, or the 320NEO, 321NEO etc. I didn't land IN DAL, I landed AT DAL. On the other hand, I did land in Dallas.
-----------------------
Website GRID updated for 2019. New stuff in the works!
Turbofan's Terminal

Hi! I make average, undesirable, boring grey maps for YS Flight!

User avatar
Gunny
Business Class Membership
Business Class Membership
Posts: 3224
Joined: Sun Feb 06, 2011 6:17 pm
Favorite Aircraft: Any thing with a prop.
Location: New Iberia.LA.
OS: Windows XP SP3
Has thanked: 575 times
Been thanked: 731 times

Re: Vibration in large scenery

Post by Gunny » Thu May 26, 2016 1:15 am

I have found that flying from the cockpit inside view F1 lessens percieved vibrations.
Image[/align=centre]

User avatar
Barr
Veteran
Veteran
Posts: 1011
Joined: Wed Mar 28, 2012 8:49 pm
Favorite Aircraft: The Boeing 757
Location: Providence, Rhode Island
OS: Windows 10/Gentoo
Has thanked: 272 times
Been thanked: 205 times

Re: Vibration in large scenery

Post by Barr » Thu May 26, 2016 1:52 am

Lies.

It's turbulence. You all need to contemplate and admire Sojis implementation of hidden features like turbulence some more.
Image
Beginner ATC for hire!
Current Projects:
-Pilatus PC12
-Mooney M20
-New England Scenery
Fine Print:
**Word of Caution**
My knowledge is limited. Anything I write in my posts may or may not be accurate. To be honest I think its all crap.

User avatar
AdityaB
Veteran
Veteran
Posts: 547
Joined: Sat Oct 17, 2015 12:40 pm
Favorite Aircraft: All
Location: India
OS: Windows 10
Has thanked: 163 times
Been thanked: 144 times
Contact:

Re: Vibration in large scenery

Post by AdityaB » Thu May 26, 2016 2:30 am

BARRACUZ wrote:Lies.

It's turbulence. You all need to contemplate and admire Sojis implementation of hidden features like turbulence some more.
So true :lol: :lol: :lol: :lol: :lol:
Gunny wrote:I have found that flying from the cockpit inside view F1 lessens percieved vibrations.
and some F7 and F8 views at specific angles :)

@forum, no specific ways to reduce this issue, other than decreasing the size of the map :?

Thanks a lot :)
Please note that I do not mean any kind of offence in any of my posts
Get full list of
my releases at:
AdityaB's Facility
:) :) :)
*temporarily unavailable further details

Deleted User 2704

Re: Vibration in large scenery

Post by Deleted User 2704 » Thu May 26, 2016 5:04 am

It is very simply reason came from YSFLIGHT data structure.
YSFLIGHT manage position with single-precision floating point structure.
This point structure is near 0.0 is very high precision but not so very low precision.

So if your plane in far from map center, vibration by these reason.

User avatar
Nodoka Hanamura
Veteran
Veteran
Posts: 923
Joined: Thu Apr 07, 2011 8:01 pm
Favorite Aircraft: Su-24, Mig-29, Su-27, F-18, F-15, ADF-01, Icon A5
Has thanked: 513 times
Been thanked: 266 times

Re: Vibration in large scenery

Post by Nodoka Hanamura » Thu May 26, 2016 6:25 am

mrmofumofu wrote:It is very simply reason came from YSFLIGHT data structure.
YSFLIGHT manage position with single-precision floating point structure.
This point structure is near 0.0 is very high precision but not so very low precision.

So if your plane in far from map center, vibration by these reason.
Would it be possible to resolve this by using a different data structure other than floating point, Mofumofu-san?
█████████████████████████████████████████████████████████████████████
Image

Deleted User 2704

Re: Vibration in large scenery

Post by Deleted User 2704 » Thu May 26, 2016 6:56 am

Nodoka Hanamura wrote:Would it be possible to resolve this by using a different data structure other than floating point, Mofumofu-san?
It is impossible. But vibration is controlable in creation.
Example in conbat-map, if you create fighting area, the area median will set to near (0, 0, 0).
Other ex in large map, farest airport center point will set to near (0, 0, 0).

User avatar
Turbofan
Senior Veteran
Senior Veteran
Posts: 1891
Joined: Thu Feb 03, 2011 4:22 pm
Favorite Aircraft: Boeing 747, all variants
Location: Central OK
Has thanked: 1004 times
Been thanked: 677 times
Contact:

Re: Vibration in large scenery

Post by Turbofan » Thu May 26, 2016 10:06 pm

^You learn something new everyday. But I did not understand mrmofumofu's last post, can you please elaborate?

One other thing that I have discovered is if you properly set the parent-child relationship in your scenery, the shaking is definitely mitigated, there is a noticeable reduction.
Check out my YSFS videos on my Youtube channel!
-----------------------
ATRP: When used in my posts refers to "Air Traffic Replay Project"
-----------------------
Avatar: Pratt and Whitney JT8D-219
-----------------------
My general rant: I will call them blended winglets, not Sharklets; There is no CEO, it's just simply put, an Airbus 319/320/321, or the 320NEO, 321NEO etc. I didn't land IN DAL, I landed AT DAL. On the other hand, I did land in Dallas.
-----------------------
Website GRID updated for 2019. New stuff in the works!
Turbofan's Terminal

Hi! I make average, undesirable, boring grey maps for YS Flight!

User avatar
waspe414
Staff
Staff
Posts: 1248
Joined: Thu Feb 10, 2011 9:35 pm
Favorite Aircraft: CH-136 Kiowa, Yak-130 Mitten
Location: Canada
Has thanked: 115 times
Been thanked: 214 times

Re: Vibration in large scenery

Post by waspe414 » Thu May 26, 2016 10:38 pm

It would make sense to be a floating point issue, though I've not looked into it.

The way floating point works is you have a certain number of digits that you can store, say 8. With a fixed point system, you would decide how many decimal places you want, and set the point there.

If we want four decimal places, we set the digits to 0000.0000. This can support numbers as high as 9999.9999, but as soon as we add 0.0001 to that, we have no more digits, the program throws an error.

We can avoid this arbitrary maximum by letting the decimal point move around, or float. Adding one to 9999.5555 would become 10000.556. Shaking occurs when you get few or no decimal places. If we're measuring in meters, with 8 digits, If you're less than a kilometer from the 0 point, we have five digits of decimals, and can accurately record positions to within micrometers. Any variation in those last few digits are entirely invisible.

When we fly away from center long enough for our position to become more than 100km, 100,000.00 meters, the program only has two decimal places of precision, or one centimeter. Flying 10,000 kilometers (10,000,000m) pushes the floating decimal all the way off the number, giving us a precision of one meter. Any variation in calculations that cause this last digit to fluctuate will mean the aircraft is rendered as jumping a meter or two every render cycle.

I don't know if floating point can go into the powers of ten, but if it can then the precision is reduced to tens, or hundreds of meters.

A great illustration of this is Kurt J Mac's walk to the Farlands in Minecraft. It might be fun to try and verify this by finding the floating point boundaries in YS.
Actual UltraViolet
Image Image
Image
I stole your turnip. Now you must come and defend your honour. ~ Hank Green

User avatar
Flake
Core Staff
Core Staff
Posts: 4350
Joined: Thu Feb 10, 2011 8:47 pm
Favorite Aircraft: Boeing F/A-18A
Location: Australia
OS: Windows 10 (x64)
Has thanked: 835 times
Been thanked: 1238 times
Contact:

Re: Vibration in large scenery

Post by Flake » Fri May 27, 2016 12:00 am

waspe414 wrote:A great illustration of this is Kurt J Mac's walk to the Farlands in Minecraft. It might be fun to try and verify this by finding the floating point boundaries in YS.
I ran an experiment on a custom fld I made. intervals were 1m, 10m, 100m...

I found that at 100,000m things got a little weird and at 1,000,000m things got very weird. At 10,000,000m things just plain didn't want to work at all.

This makes sense with floats having 8 digits. At 100,000 units are rounded to nearest 0.01m, and at 10,000,000 to nearest 1m.

If you were to do 100,000,000m the shaking would be to nearest 10m.
I am an accountant working full-time (and some). I'm not here as often as I would like to be. Send a message if you need me. There are a few people in the community who can get in contact with me urgently if you need - don't be afraid to ask. I just don't check here as frequently as I used to. Sorry!

User avatar
Nodoka Hanamura
Veteran
Veteran
Posts: 923
Joined: Thu Apr 07, 2011 8:01 pm
Favorite Aircraft: Su-24, Mig-29, Su-27, F-18, F-15, ADF-01, Icon A5
Has thanked: 513 times
Been thanked: 266 times

Re: Vibration in large scenery

Post by Nodoka Hanamura » Fri May 27, 2016 9:53 am

Okay then. Would it be possible to rectify this using something akin to how the Gamebryo Engine (Fallout 3/NV/4, TES:V Skyrim) Work and have YSFlight generate cells with independent Origins, and that when we get out of that cell we simply move to another with it's own origin?

I'm busy right now, but later I can draw a diagram of what I mean.

If this problem can be resolved, YSFlight could overcome one of it's biggest flaws - Map limitations.

Postscript:

On a Side note (also irrelevant, but correcting a error above) - Apparently Bethesda forked Gamebryo to make their own version of the engine for their own projects (Skyrim/FO4) called the Creation Engine. Interesting.
Last edited by Nodoka Hanamura on Thu Jun 02, 2016 12:35 pm, edited 2 times in total.
█████████████████████████████████████████████████████████████████████
Image

User avatar
u2fly
Senior Veteran
Senior Veteran
Posts: 1247
Joined: Fri Oct 02, 2015 9:27 am
Favorite Aircraft: Antonov An-225 Mriya
Location: Ukraine
OS: Linux
Has thanked: 744 times
Been thanked: 276 times
Contact:

Re: Vibration in large scenery

Post by u2fly » Fri May 27, 2016 11:16 am

Think its possible only after implementing geographical coordinate system for scenery logic in YSFlight, but this is a task for Soji Yamakawa.

Now we have coordinates system something like from 0 to ∞.

But in geographical coordinates there are something like for N and S we can set from -90DEG to +90DEG and for E and W we can use circle (from 0 to 360, where 360 is again come to 0 and repeat).
Also spherical coordinates will make better reality for YSFlight
/!\ READ YSFLIGHT HANDBOOK (online + PDF) | updated 2019/04/25

U.S. Senator John McCain wrote:
“Spending #NewYearsEve w/ brave #Ukrainian Marines at a forward combat outpost - we stand w/ them in their fight against #Putin's aggression”
http://pic.twitter.com/vkz0gdozVV
(31 DEC 2016)

User avatar
decaff_42
Staff
Staff
Posts: 3539
Joined: Sat Oct 29, 2011 7:23 pm
Favorite Aircraft: SR-71 Blackbird
Location: Eastern United States
OS: Mac OSX
Has thanked: 613 times
Been thanked: 1323 times
Contact:

Re: Vibration in large scenery

Post by decaff_42 » Fri May 27, 2016 11:25 am

That is a really interesting idea, to swap the coordinate systems, but I'm not sure it would solve the problem. You would still run into errors at the outside edge of the map, where the arc length of the 0-360 would get really big.
Check out my website YSDecaff for all my tutorials and addon releases.
Visit by Bitbucket page for Python codes: https://bitbucket.org/decaff42/

Civilian Aviation Server Flight Log

User avatar
Nodoka Hanamura
Veteran
Veteran
Posts: 923
Joined: Thu Apr 07, 2011 8:01 pm
Favorite Aircraft: Su-24, Mig-29, Su-27, F-18, F-15, ADF-01, Icon A5
Has thanked: 513 times
Been thanked: 266 times

Re: Vibration in large scenery

Post by Nodoka Hanamura » Tue May 31, 2016 7:23 pm

u2fly wrote:Think its possible only after implementing geographical coordinate system for scenery logic in YSFlight, but this is a task for Soji Yamakawa.

Now we have coordinates system something like from 0 to ∞.

But in geographical coordinates there are something like for N and S we can set from -90DEG to +90DEG and for E and W we can use circle (from 0 to 360, where 360 is again come to 0 and repeat).
Also spherical coordinates will make better reality for YSFlight
I was going to make a long winded post about how I might have had a solution, but now realize it was a waste of time.

Anywho - Long story short for the uninitiated - to paraphrase one of the VOs from Tats' The Theater section of his 10 Gaming Creepypastas video -

"Strange, Odd things can occur, Values can reach numbers they were never designed to reach, etc."

This is because with, what waspe said, Floating point values can only be 8 numbers, however, to get around it, we push back the decimal one space. However, this decreases the accuracy of geolocation on the map each time to a certain degree every time a frame is rendered. When it reaches - say - XXXXXXX.X, accuracy becomes so degraded that there is game instability.

Another example, again, like waspe said, is the farlands in Minecraft. Chunks of data in Minecraft become more and more corrupted as you go farther and farther away from the PO (Point of Origin), so far, in fact that it can even result in a crash, the world failing to load altogether after a certain point, and game instability.

However, games like FSX don't do this, same with another game that shares a somewhat similar scenery system to YSFlight - VSF, which I brought up a while back when it came out on Steam. VSF uses a (if I'm right) Cartesian Coordinate system. Scenery is placed on a globe, and has a origin and boundaries on that globe. The only way for this to work is to - somehow, implement a CCS or similar coordinate system into YS like u2fly said. But ultimately, that's all up to Soji.

Postscript:

@decaff

Wouldn't a globe to place scenery on be able to solve the solution? For instance, at Say I hit the end of the coordinate numbers - I'd just pop up on the other side of the globe, now with N/S or W/E Positions swapped.


Just a FYI - I have no degree in math or sciences, but I couldn't help but engage in the discussion, as this really interests me, and want to learn more about this kind of problem. So take what I say with a grain of salt unless someone more qualified says otherwise :P
Last edited by Nodoka Hanamura on Thu Jun 02, 2016 10:50 am, edited 1 time in total.
█████████████████████████████████████████████████████████████████████
Image

User avatar
Flake
Core Staff
Core Staff
Posts: 4350
Joined: Thu Feb 10, 2011 8:47 pm
Favorite Aircraft: Boeing F/A-18A
Location: Australia
OS: Windows 10 (x64)
Has thanked: 835 times
Been thanked: 1238 times
Contact:

Re: Vibration in large scenery

Post by Flake » Wed Jun 01, 2016 1:19 am

My solution has always been a "floating cell" co-ordinate system. there is cell 0,1 ... 0,2 ... 0,3... and in each of those a 0 based center floating point co-ordinate map. just load the cells you need at the time and translate accordingly.
I am an accountant working full-time (and some). I'm not here as often as I would like to be. Send a message if you need me. There are a few people in the community who can get in contact with me urgently if you need - don't be afraid to ask. I just don't check here as frequently as I used to. Sorry!

User avatar
Nodoka Hanamura
Veteran
Veteran
Posts: 923
Joined: Thu Apr 07, 2011 8:01 pm
Favorite Aircraft: Su-24, Mig-29, Su-27, F-18, F-15, ADF-01, Icon A5
Has thanked: 513 times
Been thanked: 266 times

Re: Vibration in large scenery

Post by Nodoka Hanamura » Wed Jun 01, 2016 12:58 pm

OfficerFlake wrote:My solution has always been a "floating cell" co-ordinate system. there is cell 0,1 ... 0,2 ... 0,3... and in each of those a 0 based center floating point co-ordinate map. just load the cells you need at the time and translate accordingly.
Actually, that's what I was planning to bring up, at least somewhat.

Here's the diagram I came up with:
Image
█████████████████████████████████████████████████████████████████████
Image

User avatar
Dragon_Mech
Veteran
Veteran
Posts: 571
Joined: Wed Jul 15, 2015 12:18 pm
Favorite Aircraft: F-14, F-15D&E, SR-71, King Air 200, B737
Location: Mansfield, Mo. USA
OS: Windows 7 Ultimate
Has thanked: 94 times
Been thanked: 127 times

Re: Vibration in large scenery

Post by Dragon_Mech » Thu Jun 02, 2016 7:06 am

i have noticed that from a very high altitude(25,000+ feet) , a very large square can be seen directly below the aircraft when looking straight down at the ground. perhaps this has something to do with aircraft tracking?
Owner, Dragonic Machine Industries
Image

Post Reply

Who is online

Users browsing this forum: Bing [Bot] and 1 guest