A bit of Mars fun, crazy images anyone can see.

Mr Plough

Monty Burnz wrote: »

What the hell are these supposed to be??
Congratulations, you've found conclusive evidence of smudges on the surface of Mars!

Probably a result of the smudge tool in photoshop.

King Mob

quasar2010 wrote: »

OK, the pictures have been manipulated to appear at different angles, according to you, the pixel distortion which appears in one particular image is the result of processing and software issues - we have established this. When the next viewing angle is required for a different picture the processor reverts back to the original picture - the one from above not the one it just created with pixel distortion, jesus man its hard to get through to you.
Why would any system create a modified image from an original image then create another image without reverting back to the same original image.
It doesn't take pictures of pictures it created, it always reverts back to the original raw image for its data for the next.
Your hard work

This is getting silly.
There is one initial image that is overlaid a three dimensional wire mesh model of the elevation of the area that was photographed.
This image would have compression artefacts on it for various reasons, thus when the image is overlaid onto the mesh the artefacts are still there.
So the artefacts are on the model.
Next you open your 3D model in a program that lets you view it, turn it, zoom in etc. The artefacts form the original image that you used to make the model are still there at the same size and orientation as in the original image.Then you move the model about in your viewer program to simulate what the area would look like, then take a screenshot.
This is what the images you posted are.
This isn't a difficult concept to grasp.

And for the fourth time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?

Edit: here's a video showing a 3D model of the face on Mars made using the exact same process.

https://www.youtube.com/watch?v=gXWQdqHLmkk

Here's another showing a crater not unlike the one we're discussing.

https://www.youtube.com/watch?v=RjlJsw0GIMs

And here's a good one of a canyon

https://www.youtube.com/watch?v=hQLd-kbs4Fw

Monty Burnz

King Mob wrote: »

And for the fourth time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?

I think you'll find that's the corner of Mars.

Your explanation sounds pretty convincing to me. I'd love to find a few castles and pyramids on Mars, it would be the greatest discovery in history - by sadly I don't think we have any evidence of that sort of thing here.

quasar2010

King Mob wrote: »

This is getting silly.
There is one initial image that is overlaid a three dimensional wire mesh model of the elevation of the area that was photographed.
This image would have compression artefacts on it for various reasons, thus when the image is overlaid onto the mesh the artefacts are still there.
So the artefacts are on the model.
Next you open your 3D model in a program that lets you view it, turn it, zoom in etc. The artefacts form the original image that you used to make the model are still there at the same size and orientation as in the original image.Then you move the model about in your viewer program to simulate what the area would look like, then take a screenshot.
This is what the images you posted are.
This isn't a difficult concept to grasp.

And for the fourth time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?

No, completely and utterly wrong.
Imagine you have an aerial picture and you want to create a simulated picture from a particlar angle - the west. The processor takes the available data and re-creates the image for you. Very simple.
Lets say the newly created picture has distortions because of this process.
Next, you want to create another image from the northwest, what does the processor do to create this image?
It reverts back to the original image to create the angle the best it can. It doesn't go back to the previous picture it created because the previous picture is of the wrong prospective ie - from the west.
The processor creates a complete new simulation without the same distortions it made in the first simulation.
If the processor wanted to create simulations from the north, the west, the east or the south then it will revert back to the original overhead picture not to the one it created previously and replicate the same image distortions.
Or it may combine the original and the west picture together, in which case it would create a different distortion pattern.

Mr Plough

Pretty good site to see th effects of compression to jpeg.

Quasar has a point, considering the clarity of the images on the esa site, there is very little compression.

http://www.ammara.com/support/technologies/jpeg.html#JPEG_COMPRESSION

King Mob

quasar2010 wrote: »

No, completely and utterly wrong.
Imagine you have an aerial picture and you want to create a simulated picture from a particlar angle - the west. The processor takes the available data and re-creates the image for you. Very simple.
Lets say the newly created picture has distortions because of this process.
Next, you want to create another image from the northwest, what does the processor do to create this image?
It reverts back to the original image to create the angle the best it can. It doesn't go back to the previous picture it created because the previous picture is of the wrong prospective ie - from the west.
The processor creates a complete new simulation without the same distortions it made in the first simulation.
If the processor wanted to create simulations from the north, the west, the east or the south then it will revert back to the original overhead picture not to the one it created previously and replicate the same image distortions.
Or it may combine the original and the west picture together, in which case it would create a different distortion pattern.

Except it only has one skin to go on. The actual surface of the model is a photo laid onto a wire mesh.
The actual original overhead photo used to create the skin has these artefacts.
The compression artefacts are inherent to the skin of the model, regardless of perspective or rendering.

And for the fifth time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?

King Mob

Mr Plough wrote: »

Pretty good site to see th effects of compression to jpeg.

Quasar has a point, considering the clarity of the images on the esa site, there is very little compression.

http://www.ammara.com/support/technologies/jpeg.html#JPEG_COMPRESSION

Unless they used lower res, compressed versions of the overhead images when making the models. Then you'd agree there'd be plenty of artefacts.

Monty Burnz

quasar2010 wrote: »

It reverts back to the original image to create the angle the best it can.

This doesn't make any sense to be honest. If it's only working from one image, and has one model of elevation to work from, it's going to come up with the exact same thing every time. It's just that whatever perspective the 'viewpoint' is placed at, you're going to see different things (e.g. you'll see one side of the hill the first time, and the opposite side of it from the opposite viewpoint). There's no new information in the original picture to make an improved model.

Again, I'm sure there are people on Boards who have intimate knowledge of how such models are created - GIS folks etc. I would suggest though that the idea of one model created from one photo image overlaying one terrain model is correct. You then move your viewpoint around that one model to create the different perspectives (like spectator mode in Counter Strike or whatever).

Mr Plough

King Mob wrote: »

Unless they used lower res, compressed versions of the overhead images when making the models. Then you'd agree there'd be plenty of artefacts.

No... I'm pretty sure I wouldn't actually. On the site I linked to, you can click the lower rez links... as low as 10% quality.
The whole image is affected.
Whereas the images on the esa are crystal clear.

There is no comparison.

King Mob

Mr Plough wrote: »

No... I'm pretty sure I wouldn't actually. On the site I linked to, you can click the lower rez links... as low as 10% quality.
The whole image is affected.
Whereas the images on the esa are crystal clear.

I don't understand your point.
Some of the images on the ESA site are of every high resolution, particularly the raw overhead ones from the orbiter.
Even the images used for the model have a very high resolution, but they likely use a lower resolution to the other overhead ones on the site.

I would imagine this is because it makes rendering the models a lot easier.

quasar2010

King Mob wrote: »

Except it only has one skin to go on. The actual surface of the model is a photo laid onto a wire mesh.
The actual original overhead photo used to create the skin has these artefacts.
The compression artefacts are inherent to the skin of the model, regardless of perspective or rendering.

And for the fifth time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?

Something you are forgetting here, there are two pictures not one. The creator of the first simulated picture asked a processing system to create an image using available data from an aerial shot and it did just that.
The same creator then asked the same system to create another simulated picture at a different angle from the same aerial shot and it did just that.
King mob is telling us that the the creator of these pictures instead of using the same aerial shot to create each individual picture actually used the newly created west picture to form the newly created northwest picture when that makes absolutely no sense at all.
The big original overlay which king mob explains does not exist in a two stage operation which he imagines. The processor doesn't create an overlay then swivel around it taking pictures. It creates each picture from a completely new original prospective.

King Mob

quasar2010 wrote: »

Something you are forgetting here, there are two pictures not one. The creator of the first simulated picture asked a processing system to create an image using available data from an aerial shot and it did just that.
The same creator then asked the same system to create another simulated picture at a different angle from the same aerial shot and it did just that.
King mob is telling us that the the creator of these pictures instead of using the same aerial shot to create each individual picture actually used the newly created west picture to form the newly created northwest picture when that makes absolutely no sense at all.
The big original overlay which king mob explains does not exist in a two stage operation which he imagines. The processor doesn't create an overlay then swivel around it taking pictures. It creates each picture from a completely new original prospective.

That's not what I'm saying at all.
I'm saying that they used one and only one overhead image to create the overlay on top of a model then they simply moved the model about in a viewer as demonstrated in the videos I posted, then took two separate screenshots of the same model at different angles.

But you know that. You're just trying hard to make a strawman.

So again for now the sixth time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?

namloc1980

quasar2010 wrote: »

Something you are forgetting here, there are two pictures not one. The creator of the first simulated picture asked a processing system to create an image using available data from an aerial shot and it did just that.
The same creator then asked the same system to create another simulated picture at a different angle from the same aerial shot and it did just that.
King mob is telling us that the the creator of these pictures instead of using the same aerial shot to create each individual picture actually used the newly created west picture to form the newly created northwest picture when that makes absolutely no sense at all.
The big original overlay which king mob explains does not exist in a two stage operation which he imagines. The processor doesn't create an overlay then swivel around it taking pictures. It creates each picture from a completely new original prospective.

You haven't told us why almost all the perspective images on the ESA website show artifats. You also haven't told us why the much higher resolution images of the floor of the Hale Crater show no sign of these supposed strutures?

Mr Plough

King Mob wrote: »

That's not what I'm saying at all.
I'm saying that they used one and only one overhead image to create the overlay on top of a model then they simply moved the model about in a viewer as demonstrated in the videos I posted, then took two separate screenshots of the same model at different angles.

But you know that. You're just trying hard to make a strawman.

So again for now the sixth time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?

Blackness = photoshop.

King Mob

Mr Plough wrote: »

Blackness = photoshop.

So why did they photoshop away everything outside the square even is it cuts through a mountain, yet leave the incriminating stuff in the middle?

Does it not look like the borders of the rotating 3D images in the video I posted?

quasar2010

King Mob wrote: »

That's not what I'm saying at all.
I'm saying that they used one and only one overhead image to create the overlay on top of a model then they simply moved the model about in a viewer as demonstrated in the videos I posted, then took two separate screenshots of the same model at different angles.

But you know that. You're just trying hard to make a strawman.

So again for now the sixth time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?

When those images are rotated in the viewer, there is absolutely no way on this planet that those image distortions can be replicated in a still shot twice.
The same compression distortion that appears on one shot cannot appear on another because of all the data the processor is using at any given time.
Like I have said, the processor is using every available date, not the data contained in just one particular image that for some reason you believe the processor is clung onto for life.
You are completely wrong and you know it.
processors create beautiful things for us to see, the use ALL available data in which to create those images, they DONT swivel around a central overlay remembering the last image, THEY RE-PROCESS EVERY SINGLE TIME.

Mr Plough

King Mob wrote: »

So why did they photoshop away everything outside the square even is it cuts through a mountain, yet leave the incriminating stuff in the middle?

Does it not look like the borders of the rotating 3D images in the video I posted?

They like to wave it in your face but always leave speculation. They know we'll argue like idiots over it. But you can never say they didn't tell you.

Georgia guidestones for example.

Monty Burnz

quasar2010 wrote: »

processors create beautiful things for us to see, the use ALL available data in which to create those images, they DONT swivel around a central overlay remembering the last image, THEY RE-PROCESS EVERY SINGLE TIME.

What is this 'processor' you keep referring to?

How do you know for certain that they use the cumbersome and odd system you describe, rather than the common-sense system that I know for a fact is used in other applications?

And there is no 'last image' - there is a model. The model is composed of a visual image of the surface (the photo, with whatever artifacts are in there) stretched over a wireframe representation of the surface. This is a commonly used system. This model is then rotated at will to show a '3D' image from whatever perspective. This is very simple to understand.

The black bits in the picture are presumably the edges where the single photo ends.

Here's how they do it on Google Earth - surely you can follow this video. I'm referring to the bit at the start where the satellite photo is stretched over the terrain detail that they have. http://www.youtube.com/watch?v=6DndO_Myk20

THEY RE-PROCESS EVERY SINGLE TIME.

No they don't. I think the burden of proof is now on you to show that they do it the odd way that you describe. Good luck with that

King Mob

quasar2010 wrote: »

When those images are rotated in the viewer, there is absolutely no way on this planet that those image distortions can be replicated in a still shot twice.
The same compression distortion that appears on one shot cannot appear on another because of all the data the processor is using at any given time.
Like I have said, the processor is using every available date, not the data contained in just one particular image that for some reason you believe the processor is clung onto for life.
You are completely wrong and you know it.
processors create beautiful things for us to see, the use ALL available data in which to create those images, they DONT swivel around a central overlay remembering the last image, THEY RE-PROCESS EVERY SINGLE TIME.

This is getting ridiculous now.
I have stated repeatedly that the model uses one image as it's skin.
I am not saying that the compression artefacts are on the images of the model which are on the sight I am saying (have have been saying very clearly in each and every post) that the compression artefacts are on the model itself.

Processors mightn't swivel around an overlay if they are rendering a totally simulated object (but even this seems like a stupid way to do 3d models). However the model is question is not totally simulated because it uses a real photograph as a skin. And this photograph has compression artefacts.

And again for the seventh time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?

the fact you've not even acknowledgeing this question for which there is only one honest answer means you know well you're talking nonsense.

King Mob

Mr Plough wrote: »

They like to wave it in your face but always leave speculation. They know we'll argue like idiots over it. But you can never say they didn't tell you.

Georgia guidestones for example.

Or, and bare with me cause this might be crazy, it's an image of a 3-D model of a set area of Mars, and some one is mistaking compression artefacts for something else.

Since you ignored the other question I'll assume you don't want to give the honest answer: Yes, it looks exactly like what happens at the edges of the 3-D model is the videos posted earlier.

quasar2010

And this photograph has compression artefacts

So when a new image is taken from the overlay, are you saying that the new image doesn't suffer from compression artifacts?
You don't make sense, you are saying on one hand that you can stretch an image digitally over an area, take further images from this overlay and consequently not suffer from further distortion.

Monty Burnz

King Mob wrote: »

Since you ignored the other question I'll assume you don't want to give the honest answer: Yes, it looks exactly like what happens at the edges of the 3-D model is the videos posted earlier.

Game, set and match. :cool:

Monty Burnz

quasar2010 wrote: »

So when a new image is taken from the overlay, are you saying that the new image doesn't suffer from compression artifacts?
You don't make sense, you are saying on one hand that you can stretch an image digitally over an area, take further images from this overlay and consequently not suffer from further distortion.

It doesn't matter. If you are saying the opposite, that just means there are two layers of artifacts introduced, not just the one layer from the original photo skin! What do you think that does for the likelihood that your 'structures' are artifacts?

namloc1980

So quasar2010 any chance you'll now tell us why the very high resolution images of the floor of Hale Crater shows no signs of artificial strutures etc?

King Mob

quasar2010 wrote: »

So when a new image is taken from the overlay, are you saying that the new image doesn't suffer from compression artifacts?

It probably does, but the artefacts you're talking about are form the compression on the original overlay, not any subsequent compression.

quasar2010 wrote: »

You don't make sense, you are saying on one hand that you can stretch an image digitally over an area, take further images from this overlay and consequently not suffer from further distortion.

That's not what I said at all.
No where did I say anything that can be possibly taken to mean this.

The artefacts you are talking about are on relatively flat ground, so any stretching is not going to be an important factor.

For the eighth time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?
Can you at least acknowledge the fact you're ignoring this question?
If not, why not?

quasar2010

King Mob wrote: »

It probably does, but the artefacts you're talking about are form the compression on the original overlay, not any subsequent compression.


That's not what I said at all.
No where did I say anything that can be possibly taken to mean this.

The artefacts you are talking about are on relatively flat ground, so any stretching is not going to be an important factor.

For the eighth time: how do you explain the blackness beyond the square area?
Why are you ignoring this question?
Can you at least acknowledge the fact you're ignoring this question?
If not, why not?

I can argue this lemon forever but I have to concede that I'm riding a lemon and no matter what I say, the fact remains that the overlay was done with the compression error which i cannot escape from, therefore I concede the argument and apologise for my stubborness.

Monty Burnz

quasar2010 wrote: »

I can argue this lemon forever but I have to concede that I'm riding a lemon and no matter what I say, the fact remains that the overlay was done with the compression error which i cannot escape from, therefore I concede the argument and apologise for my stubborness.

Fair play

RoboClam

quasar2010 wrote: »

I can argue this lemon forever but I have to concede that I'm riding a lemon and no matter what I say, the fact remains that the overlay was done with the compression error which i cannot escape from, therefore I concede the argument and apologise for my stubborness.

No one likes to be wrong; it is especially hard to admit fault after having defended a position for so long. However, you've come out and said it and now I'm sure that I and a lot of other people will respect you more for that.

Mr Plough

quasar2010 wrote: »

I can argue this lemon forever but I have to concede that I'm riding a lemon and no matter what I say, the fact remains that the overlay was done with the compression error which i cannot escape from, therefore I concede the argument and apologise for my stubborness.

He's not a lemon.

squod

RoboClam wrote: »

No one likes to be wrong; it is especially hard to admit fault after having defended a position for so long. However, you've come out and said it and now I'm sure that I and a lot of other people will respect you more for that.

I can only wish other posters would do the same..........