Jump to content


Photo
- - - - -

SRTM data Comparison


  • Please log in to reply
30 replies to this topic

#1 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 22 May 2010 - 04:56

Nasa SRTM Data files

This post is for Landscape creators , if you can understand this you will create a much more accurate and historical mapping which in turn helps texturing and any other process you do afterwards with the smrt data.


Sometime ago i found a website that had a different outlook on a problem thats been caused by missing data from the NASA SRTM hgt files.
You may all ready know this information ,but no harm in posting some results so we can all see the differences.

SRTM data is basically a Radar bouncing from a satellite towards earth and bounces back to the satellite , in the form of a height value called SRTM or hdr files or hgt files , height files.

The data is collected in squares called "tiles" these tiles are then loaded into a program that joins all the tiles you have collected and presents you with a untextured height map or mesh or layer that once textured can be made into a flight sim map.

So what is Missing SRTM DATA ?


Issue 1
This issue starts with the satellite , when it transmits the radar signal the radio waves respond totally different when they come into contact with ICE and water and basically anything thats reflective like snow or a landscape thats wet from rainfall etc


Issue 2
The other issue with the satellite radar is the angle of which it transmitted the radar , the satellite is travelling very fast the data covered either a 3 second arc or a 1 second arc , which means if the radar was transmitted towards a steep walled valley in mountains the radar bounce did not always get bounced back to the satellite , or the steep valley hill side ment the radar did not bounce into the valley to recover the height and shape of the valley.
Or the radar passes over one steep walled side of the valley bounces of the opposite valley wall and the side of the valley wall it passed over was not hit by the radar , so the valley and one side are how missing data or Hole is created.

Combine both these issues and the problem becomes rather upsetting to map makers , especially if you are going to be making a HISTORICALLY Correct Map.
Flat areas can have the same results , because of rainfall ,snow,and weather system storms at the exact time the satellite passed over recording information, and of course the angel the radar was pointing at earth when it transmitted.

Ok , by now some mapmakers will be saying 3DEM and microdem and other programs fix this issue and you can load a map that has not got any missing data file areas because the program fixes it ….
Sorry but no.. you are just fooling your self and below it an example that proves it.



3DEM is a program that you can load SRTM radar data into and it will convert the data into a visual terrain contour , the program DOES have the ability to fix the missing data with a unusual set of rules that it follows , basically a guess fix but with a generally good intention.
The results are never going to be high quality, its a fix not a correction.


The way 3DEM fixes this problem of missing data is to look at the height values of the edges of the missing data area some call these areas "holes" and divides all the heights values of the rim with the distance circumference of the hole.


This means if you have a mountain and the nasa data is missing for the PEAK of this mountain , then 3DEM fills the missing patch with a make do patch a mathemathical equation that is not 100% perfect.

The mountain top will have a undulating bumpy, but basically flat top with no PEAK! , sometimes it actually can make a VALLEY instead of a mountain :)

Below is a screen shot of SRTM NASA DATA fixed using 3dem .
screenshot is from Concordia glaciar Valley looking towards K2 mountain on a Map i made just to show this example of fixing missing data srtm files.
Image
The flight sim is Condor flight sim ( for Sailplanes / Gliders )

Now some would say , Whats wrong with that !
Ok , would you believe that NASA data files have a huge missing piece of data that is actually the whole K2 mountain , otherwords the K2 Mountain does not exist if you fix the missing data using 3Dem or microdem or any smrt data correction tool.

Below is the same photo snapshot with the K2 location visible
Image

So 3Dem fixed the missing data area but it is not intelligent its just a mathemathical equation and a rough guess…

For those of you who fly IL-2 1946 flight game , this above photo is the quality you have been flying within for along time now , about 10 years :D , why ! , because the mapmakers did not know or thought it was fixed correctly by 3DEM / Microdem when it was a greyscale height map.

Not very good quality !

I hope Rise of Flight mapmakers , will produce far greater than this so i make this post to guide a correct direction.

Below is a snap shot from GOOGLE EARTH ! , the GOOGLE maping team realise that the programs that fix missing data files is incorrect so they have taken many many years to slowly rebuild the contours and shapes of the terrian to a fairly close to accurate effect.

Image

Just to backup GOOGLE Earths accuracy below is a actual PHOTO from roughly the same location as all FOUR photos location within the Concordia glaciar Valley , sorry they are not exact tripod location but you can see the huge improvement Google earth created.
Image

You can see that 3Dem fixed the missing data patch by deleting what should be visable and that google earth did a great job of rebuilding a famous landmark.

OK :)

Now the mapmakers from IL-2 and several other Flight sims MAKE the maps/scennery and use 3Dem or microdem to fix the patches in the missing data from the satelite smrt.files
Unfortunately the result is the FIRST TOP photo quality and the landscape is incorrect infact its totally wrong , for 10 years IL-2 mapmakers have been building maps inaccurate , and i DO NOT want it to happen in RISE OF FLIGHT !
Its historically correct and i think the map creation should follow the same path , so i hope the next photo you do not look at the COLOURS or texturing it is only a example ,,,,

BUT !!

Look at the contours and shapes of the mountains and the peaks ! this is the example i am showing ,so map makers do not make incorrect landscape in the future of rise of flight.
Image

compared with GOOGLE EARTH , i may be wrong but i think my terrain is better shaped, of course the real photo is much more better but the shapes and contours of the terrain are as good a quality that can be made with present day technology.

This last photo did not use 3dem patch fix to mend the missing data areas , it used ordinance survey data and topographic mapping and alot of skill by a very clever GROUP of people.

Anyone who in the future makes new maps either 3rd party of offical, please understand smrt data even if its NASA DATA does not mean its totally accurate , and 3DEM or microdem or any other data missing fix programs do not fix the terrain correctly , only a human touch can let the mountain be seen.

I hope this is helpful to mapmakers ,
The url link below is to Corrected SRTM Data is not ALL of the world corrected only some parts but europe has some 1 arc second smrt data fixes as well as 3 arc second data fixes so please use it , and your results will be far greater quality , using corrected ordinance survey map values and topoghaphical mapping to fix the missing smrt data areas , so 3DEM does not have to FIX with a lower quality patchwork result.

http://www.viewfinde...s.org/dem3.html" onclick="window.open(this.href);return false;">http://www.viewfinde...s.org/dem3.html

If you need help understanding this topic of smrt please post here and i will help all i can , because its for a good cause , and means we all fly in terrain thats the best quality.

Rise of Flight Developers
Thank you for creating a LANDSCAPE forum , i will make it my home :)

Best Regards
O_Wolfpac
Tony
  • 0

#2 J2_squid

J2_squid
  • Posts: 3815

Posted 29 May 2010 - 15:52

Wise words, but the terrain is also affected by the resolution of the scan. FSX shows this clearly when you adjust the Mesh detail. IIRC the maximum resolution is 15cm? I belive the SRMT is 14m? Thats one of the reasons the peaks look different as the gaps in the data are smoothed out.

Hmm, wonder if the Mesh's from some of the terrains for FSX can be converted? Also whats the resolition of the height map in ROF?
  • 0

#3 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 30 May 2010 - 01:43

thats what im trying to get my head around ive lots of maps i made for Condor sitting here in folders that i made 2-4 years ago some with photo realistic scenery textures

FSX map format, how many metres per pixel are the mapping res.
  • 0

#4 1PL-Lucas-1Esk

1PL-Lucas-1Esk
  • Posts: 1038
  • LocationPoland/Warsaw

Posted 30 May 2010 - 21:08

The FSX terrain mesh is compiled to BGL format, as well as the other scenery files (for example 3d model + xml file telling the game its coordinates)
The stock FSX comes with 38 meter mesh for the USA, 76 meter mesh for the Canada, Europe, Australia and New Zealand and Japan. I remember that the Oahu has 10meter mesh (or 19) and the cliffs of Dover have 10 meter resolution.
Some payware/freeware meshes come from the data provided by mr. J. de Ferrante if iam not wrong. I messed with the FSX scenery a bit (done couple of aerodrome sceneries with phototextures), but I have never added additional mesh to the game, so i cannot help :(

Lucas
  • 0

1PL-Lucas
CO of 1.Pułk Lotniczy
http://1pl.boo.pl


#5 taleks2

taleks2
  • Posts: 2040

Posted 31 May 2010 - 00:16

SRTM is correct name for Shuttle Radar Topography Mission.

As for defects removal, automated processing based on e.g. inverse distance weighting algorithm may remove area deffects in some cases, but have limitations if too much contextual information was lost.

There are several projects, which remove defects in SRTM, e.g. CGIAR-CSI.

I belive the SRMT is 14m?
it varies, depending on latitude.

1" data was released for USA territory only, which gives about 30 meters per pixel.
3" data released for rest of the scanned world, which gives about 90 m/pixel.

BTW: AFAIK, SRTM was superseded by GDEM in 2009.
  • 0

#6 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 02 June 2010 - 04:43

Been doing some reading on GDEM

It appears its good quality for low -to- mid range landscape , with higher landscape area being low quality with many errors, but even using this GDEM data some area's can be lower quality than other comparisons of smrt data even in low to mid range landscape and smrt data fixed via the technics used on http://www.viewfinde...s.org/dem3.html">http://www.viewfinde...s.org/dem3.html can be greater quality.

I believe there will be a report released very soon on a new comparison from the japan GDEM data , fingers crossed it has greater results
  • 0

#7 WW1EAF_Ming

WW1EAF_Ming
  • Posts: 2565

Posted 02 June 2010 - 09:41

the GOOGLE maping team realise that the programs that fix missing data files is incorrect so they have taken many many years to slowly rebuild the contours and shapes of the terrian to a fairly close to accurate effect

That's interesting Wolf

Terrain isn't going anywhere or if so at a glacial pace :) so it can't be that hard to fix missing data, from contour intervals on standard maps

How is the data arranged do you know, so you could interpolate/insert missing points. Is a terrain-net map simply a big list of XYZ points, or does it get more complicated by the time they process and order the data?

I'm bonkers about 3D modelling and also about maps, don't mind me poking my oar in please I'm after Afghanistan :)

Ming
  • 0

#8 J2_squid

J2_squid
  • Posts: 3815

Posted 02 June 2010 - 09:52

Some different theaters will be great. What we desperately need right now though is smaller sized maps of the frontline. Using the full map is causing problems for people with ROF.exe errors and long load times.

How hard is it to "cut" parts of the map out? Somme, Verdun etc would be nice.
  • 0

#9 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 02 June 2010 - 13:21

Squid

I think that will happen automatically when the map editor is released in my guess.

Question is , does that mean that no one is getting the same problems on a dogfight map which is around 50km x 50km comparison to original coop large map ?
  • 0

#10 J2_squid

J2_squid
  • Posts: 3815

Posted 02 June 2010 - 14:06

AFAIK the situation is much better. There are several causes to the errors, as you can see mission design is one. But surely loading up a smaller map has to help.

Cant wait for the editor. :)
  • 0

#11 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 02 June 2010 - 15:16

the editor = my guess :D

Maybe its not the actual size of the map , maybe it what its trying to render is the key.

The Dogfight map has less trees , less roads , less railway lines , more water !

Just a thought , if the radius distance of rendering the terrain was set to large distance then in theory that slightly slower cpu pc with its ghaphics set slightly to high a level would cause a hiccup error

again just a thought , its late and cant sleep so this may all be gobbledy gook
  • 0

#12 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 02 June 2010 - 15:39

Sorry EAF19_Ming i missed your post !

How is the data arranged do you know, so you could interpolate/insert missing points. Is a terrain-net map simply a big list of XYZ points, or does it get more complicated by the time they process and order the data?

Well im learning , and no i havent got it down correct yet, but i have figured out a work-around way of fixing the data :)
Yes , interpolate/insert missing points , or fill the holes with better contour shapes than what 3DEM tries to fix.
http://www.viewfinde...s.org/dem3.html">http://www.viewfinde...s.org/dem3.html
The website above they have ,,,,well,,, genius minds and wizkid fingers and have filled the mssing data holes with map data and survey readings aerial photos etc

I just cant get my head around how hehehehe so i sat down and worked out a way that suits flight sim mapping.

In flight sims 9 times out of ten we convert the nasa data smrt file into 3dem program then save into a greyscale map , it came to me one day to not fix the missing data and convert to greyscale to see just how bad the scenery would look without fixing the holes , at the time i was making maps for Condor Flight sim so i loaded the map and ouch it was like flying not in the ALPs but on the MOON !
Massive cliffs and trenches and well not nice , great lift but not good on the eye :D

then i loaded the map in Gimp and looked closly at the map in greyscale and could see the errors reasonably clearly , i figured i would have a go at mending them by hand to eye and alot of commonsense but when i tried it out ingame it was not even close to correct.
Then i got the lightbulb momment and loaded the map area into 3D Max Studios and it loaded ! never thought it would but then after reading alot of websites i learned how to MAKE a greyscale map in 3d max so i edited the vertices and added afew plains in max studios and converted it to greyscale and then made the map upagain so i could fly in condor to see results and its not great but when i compared with a different map i fixed the holes in 3DEM the obvious comes clear , 3d max made a better contour :P

not exactly a correct fix but it ment i could make slightly better , but it only works if the map /flight sim runs off a greyscale map as the foundation map.

Rule of thumb learned by mistakes ,being never try to create a vertical cliff :) i could explain why not,,, but it just does not happen hehehe seems the pixel to meters also has an odd effect aswell so becareful of vertices inbewteen pixel grid , if they get to close together you create a sort of ripple effect when it converts to greyscale like it cant workout the correct data so it gets a headcold heheehehe


Is a terrain-net map simply a big list of XYZ points well yes it is xyz , but its also the shape of the globe aswell as utm projection as WGS84 , no idea what it means but hey im no wizkid , im just bumpin around in the dark :D
  • 0

#13 WW1EAF_Ming

WW1EAF_Ming
  • Posts: 2565

Posted 02 June 2010 - 16:44

No worries Wolf

i got the lightbulb momment and loaded the map area into 3D Max Studios and it loaded !

So cool mate. I've got Softimage Autodesk 7.01, can't wait to get a terrain model in there :)

the shape of the globe

Ah of course yes, thanks!

Small areas are of course as near to flat as makes no difference. No great circles on the battlefield :)

Ming
  • 0

#14 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 02 June 2010 - 17:36

actually you will be supprised :) its present
  • 0

#15 WW1EAF_Ming

WW1EAF_Ming
  • Posts: 2565

Posted 03 June 2010 - 09:27

Double-precision, gnat's whisker :)

Ming
  • 0

#16 WW1EAF_Ming

WW1EAF_Ming
  • Posts: 2565

Posted 03 June 2010 - 09:28

I cleaned that up :)

Ming
  • 0

#17 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 03 June 2010 - 09:42

HAHAHA
  • 0

#18 Rama

Rama
  • Posts: 338

Posted 27 October 2010 - 16:53

BTW: AFAIK, SRTM was superseded by GDEM in 2009.

Just to let you know, GDEM quality is very poor. In most of the areas, SRTM DTED1 (i.e. 3" resolution) is better.
Something you should also know with SRTM is that 3 different releases has been made, the last one including patching of SRTM voids with other data (Ref3D mainly) for a lot of areas.
  • 0

#19 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 27 October 2010 - 21:41

Rama , Are the new patch void fixes good quality ?
  • 0

#20 taleks2

taleks2
  • Posts: 2040

Posted 28 October 2010 - 01:27

Just to let you know, GDEM quality is very poor. In most of the areas, SRTM DTED1 (i.e. 3" resolution) is better.
Something you should also know with SRTM is that 3 different releases has been made, the last one including patching of SRTM voids with other data (Ref3D mainly) for a lot of areas.
I know about different releases, including water shape areas in vector formats.
Anyway, I'm not working in GIS area since 2008 and was not working with GDEM myself.
Thus, I may be wrong.

PS:
As for SRTM, my work in GIS department of ITiG (part of FEB RAS) included development of software which fixes void area and additional modules for self-shading of 2d image on base of heights (well, and database/GIS interaction, but that's other story). SRTM have it's limitations in resolution per height and in resolution per pixel, also it uses WGC84 with datum, which is out of Europe AFAIR, thus have error accumulation in area of Europe, especially if change of projection is needed. Anyway, this is excellent freely available source of data.
  • 0

#21 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 28 October 2010 - 01:55

My word Taleks2 :) you really are a multitasking talent of all :)

Water shape in vector format would be handy , especially for lakes and coastal locations.

Did you know you can load SRTM Hgt files into 3D Max studios 2009 as Vertex co-ordinates :)
not sure if max 8 or 2010 will open them but 2009 does :)
  • 0

#22 Rama

Rama
  • Posts: 338

Posted 28 October 2010 - 12:02

Rama , Are the new patch void fixes good quality ?

Yes (I'll tell you more in a private message).
But all the voids are not covered with patches. (it stays some little voids everywhere and some big voids in some areas, like in Himalaya mountains or inter-tropical area).

Water shape in vector format would be handy , especially for lakes and coastal locations.
These water shapes were mostly digitized from old low-resolution Landsat data… and the quality isn't very good… not really suitable for RoF 25m resolution maps.

SRTM have it's limitations in resolution per height, and in resolution per pixel
Between -50° and 50° lattitude, pixel size is 3" for SRTM DTED1 (so around 90m at equator), above 50°, pixel resolution is 6".
Altitude résolution is 1m everywhere (I'm not talking about precision, but about resolution).

also it uses WGC84 with datum, which is out of Europe AFAIR, thus have error accumulation in area of Europe, especially if change of projection is needed.
That's not totally correct. SRTM altitudes are given above EGM96, which is a spheroïd (sphérical harmonic model of the Earth), itself given on WGS84.
There's no problems for projection change, everywhere on the world.
  • 0

#23 taleks2

taleks2
  • Posts: 2040

Posted 28 October 2010 - 13:45

These water shapes were mostly digitized from old low-resolution Landsat data… and the quality isn't very good… not really suitable for RoF 25m resolution maps.
SRTM data is also not good, if we talk about 25m/pixel :)

Water shapes helps to remove height artifacts near river/lakes/seas bounds or in it's areas. it's not panacea, but it really helps to save operators time for digitizing maps in practice.

BTW: some lakes and seas may be detected automatically (by chaotic or wave-structured pattern) and "flooded" with needed water-height level.

Between -50° and 50° lattitude, pixel size is 3" for SRTM DTED1 (so around 90m at equator), above 50°, pixel resolution is 6".
Don't you think it's limitation?

Altitude résolution is 1m everywhere (I'm not talking about precision, but about resolution).
well, precision is more correct word to describe limitation, I've meant.
You're correct, that height is integer value, which represents height in meters, thus resolution of height representation is 1m.

BTW: in Russia was problematic to use data with precision 1m/pixel about 5 years ago. We had even limitations on GPS (GLONASS was not usable at that time) usage for mapping and other purposes. Restrictions from military were pain in the ass, forcing keep in secret even those map, that were needed for civil usage. Now situation is a bit better.

There's no problems for projection change, everywhere on the world.
It's not problem to change projection, but it's problem to keep error in accepted interval during projection.
Problem exists, not only for SRTM, of course, Earth is not cube after all. That's why maps are created with some datum (in addition to reference ellipsoid), ortho-projection is not the best one for many applications (errors in route length calculations, areas of territories). It may be not significant in some applications, but is important in others.

PS:
not spheroid, but reference ellipsoid, in general. However, neither of them is geoid, in fact.
Also, EGM96 is not reference ellipsoid, it's gravitation model, that is used, for example, to calculate errors for reference ellipsoids. It's more correct Earth surface model, but it's not used on applications due performance difficulties. Usage of reference ellipsoids is better (and simpler to work with, from algorithms point of view) for applications.

PPS:
May be I've put it in wrong way. For simulation purposes, SRTM it mostly acceptable. But not for all real-world applications.
  • 0

#24 Rama

Rama
  • Posts: 338

Posted 28 October 2010 - 14:11

SRTM data is also not good, if we talk about 25m/pixel.
Of course SRTM data isn't good enough for 25m/pixel… never said the contrary.
RoF elevation map is 50m pixel resolution (and yes, SRTM isn't good enough in mountaneous areas). Water Map is of better resolution (25m/pixel). To digitize water layer for RoF map, the SRTM water bodies shapes are helpless… that's all I was saying.

However, neither of them is geoid, in fact.
Also, EGM96 is not reference ellipsoid, it's gravitation model
EGM96 is a geoid (I wrongly used the word "spheroïd", sorry for my bad english…) issued from gravitionnal measurement (a geoïd is a gravitationnal model in any case… or we don't have the same definition of the word…). Altitudes on maps are almost allways given "above mean sea level", which means that they are given above the iso-gravitationnal surface represented by the "mean sea level" (sea surface depends on gravitation).

It's more correct Earth surface model, but it's not used on applications due performance difficulties. Usage of reference ellipsoids is better (and simpler to work with, from algorithms point of view) for applications.
I'm a mapping ingeneer, and works daily on mapping application (in real-world), for civilian and army usage. All these applications use DEM (not only SRTM of course…) with elevation given on various Geoïd (and EGM96 ie one of them).
So it's wrong to say that "it's not used on applications".
  • 0

#25 taleks2

taleks2
  • Posts: 2040

Posted 28 October 2010 - 15:40

Water Map is of better resolution (25m/pixel). To digitize water layer for RoF map, the SRTM water bodies shapes are helpless… that's all I was saying.
Yes, in RoF it can't help directly.

a geoïd is a gravitationnal model in any case… or we don't have the same definition of the word…
In Russian traditional geo-related terms, geoid is surface of gravitational field, which is close to mean ocean level surface. Well, it's something like this if translate to English. So, it seems equal to your term.

"not geoid" was related to spheroid and reference ellipsoid, not EGM96.

So it's wrong to say that "it's not used on applications".
Yes, I'm wrong here. Wanted to say, that reference ellipsoids are used more frequently, than geoids, at least in my experience.

It's just matter of interest. Rama, you're from France, are you? I'm not sure which types of topography maps do you use, but from your post I conclude, that SRTM conforms to it with low error after projection (elevation level isolines are close). Am I correct? Thanks ahead.
  • 0

#26 Rama

Rama
  • Posts: 338

Posted 28 October 2010 - 16:49

It's just matter of interest. Rama, you're from France, are you? I'm not sure which types of topography maps do you use, but from your post I conclude, that SRTM conforms to it with low error after projection (elevation level isolines are close). Am I correct? Thanks ahead.
Yes I am from France… but work mostly on areas outside France. ;)
To answer your question, I can take it from 2 different angles:
- If I want to compare a DEM and level isolines from topomaps, I will not extract isolines from the DEM, but instead compute a DEM from the isolines and compare it with the other DEM; from the DEM differences you can then directly compute altitude error statistics. Of course, the result will also depends on the quality of the topomaps. I did participate to different global DEMs studies (SRTM, GDEM, REF3D, ICE3, DEM extracted from Topomaps, etc…) on different places of the world, with using absolute references (GCPs, ICESAT, Lidar acquisition, ect…). If you're interested, send me a PM.
- If I want to use SRTM to extract isolines for inclusion in a topomap, then I will restrict them to scales equal or smaller than 1/200K, and in not-to-mountainous areas. In this case, I can use any projection and the errors will be acceptable (low enough for the map specification).

If you want to talk more about this, then it's maybe better to use PM, I'm not sure many here are interested by this discussion… ;)
  • 0

#27 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 29 October 2010 - 04:50

I AM !!!! hey not fair !!!

Its a forum :) remember :D

Granted to the pilot just reading a post this MAY look like a argument :) but its not ,,,,,,,
good points made and opinions from different usage standpoints

Question is …..have i made a bad choice downloaded my corrected digital topograph hgt and blending these files over the srtm to correct the void missing areas and remaining within a a slightly low quality data ,,,,
Im not far into making a map so i can back-track very quickly to a correct route of data i just know that srtm data with voids fix/patched by 3DEM is so wrong its laughable ,,,,
Granted i was not aware of the new version of recent release , maybe this solves some issues , but quality counts especially within a terrain.

BUT gentlemen you know your mapmaking enought to see what i ment so have i still gone the wrong path.

Have any of you opened hgt files inside 3d max studios to create a greyscale map thats saved as hgt and manually corrected a void

remember its a forum
  • 0

#28 WW1EAF_Ming

WW1EAF_Ming
  • Posts: 2565

Posted 29 October 2010 - 09:44

Checking this thread daily Rama

It's a forum </echo> :)

Otherwise I would not see (on googling WGC84) fascinating stuff like this-

The coordinate origin of World Geodetic System 1984 is meant to be located at the Earth's center of mass; the error is believed to be less than 2 cm

http://en.wikipedia....Geodetic_System" onclick="window.open(this.href);return false;">http://en.wikipedia....Geodetic_System

When you think of those guys back then cutting through jungles to map continents… we are in another world :)

I see in the newspaper today, plans for one-way trips to Mars. What is it we are looking for, is there no end to our desire to explore? It could be bananas :)

Ming
  • 0

#29 taleks2

taleks2
  • Posts: 2040

Posted 29 October 2010 - 11:08

If you want to talk more about this, then it's maybe better to use PM, I'm not sure many here are interested by this discussion… ;)
Thanks, I just needed this sentence "and the errors will be acceptable" for 1:200000.
As I said earlier - I'm not working in GIS area last 2 years, thus it is not important anymore :)

have i made a bad choice downloaded my corrected digital topograph hgt and blending these files over the srtm to correct the void missing areas and remaining within a a slightly low quality data
mmm, if you're sure that your data is accurate and conforms to real one, then you've made correct choice.
Some data is always better than void data.

Have any of you opened hgt files inside 3d max studios to create a greyscale map thats saved as hgt and manually corrected a void
Actually, I've never used 3D Max for such purposes. Specialized software is much more problem specific and flexible.

BTW, it's possible to correct .hgt in Photoshop or any other graphics editor, which supports 16bit depth for channel. Just open it as raw format with correct dimensions and byte-order.

I see in the newspaper today, plans for one-way trips to Mars.
One-way? You know, it's not best planet to live (just check Ray Bradbury's novels).
  • 0

#30 WW1EAF_Ming

WW1EAF_Ming
  • Posts: 2565

Posted 29 October 2010 - 12:21

check Ray Bradbury's novels

I read them 40 years ago, when The Silver Locusts was the title of The Martian Chronicles :)

And still remember them, wonderful.

it's not best planet to live

Or to die and that's the prospect. But the views… :)

Ming
  • 0

#31 O_WolfPac

O_WolfPac
  • Posts: 1294

Posted 30 October 2010 - 23:30

Good point that needs more thought.

In the first post i show a way to correct missing voids in Mountain terrian ,,
Would be a very very nice addition to create landscape that changes direction of wind

Slope Soaring , and downdraft areas on downwind side of a hill / ridge
Thermals and the effects
  • 0


0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users