Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[cameni]

35.05'09.55"N
85.20'57.49"W
500 meters
facing 165 degrees.

[ZeosPantera]

For the record, here's how ATI cards currently render the same location

That's the best looking problem I have ever seen. I got all excited thinking you had added desert textures and cacti.. Turns out ATi is just trying to help.

[cameni]

ATI is definitely getting better with each fixed bug we are submitting No flying trees and shiny blinking colors anymore.

That desert color is the result of fallback test texture kicking in, as there is some bug preventing our job processor loading compressed textures. Apart from that the scene would look almost identical.
There are further performance problems and stalls that will have to be addressed once everything principal works.

[Airhogg]

Airhogg wrote:
35.05'09.55"N
85.20'57.49"W
500 meters
facing 165 degrees.

Thanks Camini!  The elevations are spot on!  We're just missing the Tennessee River.  Thanks for posting the pic!  It's VERY promising and is the best rendition I've seen of this area as far as elevations are concerned.  

[|ZUTI|]

Thank you cameni. All pictures are almost spot one. With exception of west/northwest one. It seems to be missing the highest mountain chain in my country. But it might be a bit to far, can't say for sure

Thanks again, much appreciated.

[Mikey999]

Hi Cameni,

Your engine looks fantastic and very promising for the future...

I have some coordinates, if you have some time.

45.971361, 7.722294
+45° 58' 16.90", +7° 43' 20.26"
3000 meters
facing west and east

The location is a cable car station above Zermatt in Switzerland. The mountain on the west is the Matterhorn (4478m), on the east the Monte Rosa range with the Breithorn in the front (4164m). Very curious about that :-)

Keep up the excellent work!

Cheers,
Mike

[cameni]

+45° 58' 16.90", +7° 43' 20.26"
3000 meters
facing west and east

The location is a cable car station above Zermatt in Switzerland. The mountain on the west is the Matterhorn (4478m), on the east the Monte Rosa range with the Breithorn in the front (4164m). Very curious about that :-)

These mountainous regions beg for a better resolution base data, I'm interested to see what 30m data will bring here once we prepare the enhanced dataset.



Ouch, what an ugly snowline there

[razgriz 53992]

I think it looks great

But there should probably be bits where the snowline blends and there are patches of snow below the main snow line, so it looks more natural.

Plus grass should be a a bit darker, sensitive to region/ altitude? I don't know, disregard my noknowledgeofthisstuff comments

[cameni]

I think it looks great

But there should probably be bits where the snowline blends and there are patches of snow below the main snow line, so it looks more natural.

Plus grass should be a a bit darker, sensitive to region/ altitude? I don't know, disregard my noknowledgeofthisstuff comments

Thanks
Snow is done by way of hack, and although there is a fractal-like blending between snow and not-snow, it's done in a finer detail level thati is not visible in that screen. Usually it's not a problem because the mountainous terrain tends to be quite ragged, but in this case that stretch of terrain has a constant slope. Might be an error in the data as well.

The color of grass (or generally of the vegetation) will be determined from real land color data, this will give it the variation and appropriate colors. The variation in color is currently completely missing in this coarser range of detail.

[Mikey999]

Thanks for the screenshots - it looks fantastic    Might be great to have better resolution base data, but already now you can easily recognize the peaks with their characteristics.

There is no error in the data - the constant slope is due to a huge glacier with a more or less constant slope. Do you plan to implement glaciers at a later time? I think glaciers are an important part of the looks in such regions.

The tree line is approx 1000 meters too high for the alps, but as i understand this will change as soon as you implement different climate and land classes. Snow line is perfect for summer and this region  

[phlerp]

What parameters will effect the snow? For instance the snow line  (at least during spring) is different depending what side of a mountain you are. On the sunny side it melts faster. This also applies for trees and vegetation I guess. Or is that too complicated and time-consuming calculations?

[Omkalfatre]

Any idea when we will have a first look with final "land class data"?

It would be interesting also, if you find the time to see;

59 09 39.06 N, 10 12 12.26 E (Looking to the south - Close to ENTO airport)

Spent countless hours in FSX in this area, would be fun to compare

As I understand, the data for this area is not that high resolution, but would be nice to see it in your engine anyway

[cameni]

Snow line depends on many factors that affect the temperature and moisture at given place. Temperature alone is simpler, depending mainly on the latitude and elevation, and that's what we want to use first. It could be also used for vegetation; both depend on average minimum temperatures over a period of time, though the periods are different for snow and ability of particular vegetation to survive. By the way, tree line altitude is roughly set for southern Himalayas at the moment.

Later we would use real world temperature & precipitation data to compute a displacement map that will correct the computed values, encompassing other factors that affect the climate. This map will be quite coarse and it should not be taxing the resources too much.
Other, more localized factors like the sunny sides could be accounted for as well, at least as by computing a value that shifts the computed snow line elevation.

[cameni]

Any idea when we will have a first look with final "land class data"?

It would be interesting also, if you find the time to see;
59 09 39.06 N, 10 12 12.26 E (Looking to the south - Close to ENTO airport)

Spent countless hours in FSX in this area, would be fun to compare

As I understand, the data for this area is not that high resolution, but would be nice to see it in your engine anyway

Um, I don't know what to consider to be the final land class data, especially when the land class data are divided in two parts - raster based fractal refined "natural" data that should get the planet the look from some ten thousand years back, and the civilization layer vector data that should bring in all the unnatural (in the sense of - to be reverted without human maintenance) modifications. I guess the latter solely depends on the game/sim project type. The former should be brought in somewhere during this year (I hope).

59° is still within the 90m region of source data, here's how it looks now:

[Omkalfatre]

Thank you very much! It looks MUCH better than FSX, with less detail, no buldings and "landclass"    

The detail that is there, the terrain is so much smoother and the lightning/atmospheric model is just perfect! Cant wait to see what you make out of this with higher resolution data ++!

Looking forward to later this year then!