Don't bet against me.

I have it down to
414,720 bytes.
This is max according to the contest
415,241

So I am in the money.
And this is the screen shot so far.
It has the animated gif, the progressbar, the formula and a cancel check box.
https://ibb.co/1MbSDg2

lol
I love this stuff.

On 7/8/2021 12:40 AM, Kcir wrote:
> If we were to make some wild assed speculation and say well I want to back up and compress a 1TB drive
> how long will that take?
> Well all the utility needs to do is to go through that drive reading all the bytes 100 times.
> And that will shrink it to 1 percent of its original size.
[...]

1234 * 1.75 = 2159.5

Did you mean:

1234 * 1 / 1.75 = 705.14285714285714285714285714286

Sorry if I am missing something!

Ok so I have been doing some experimentation.

Here is a screen shot of the tool I made...

https://ibb.co/ZBLgXJj
So on the left are the Rand 1 million digits and each byte in sequence from the file, is a color value, one of 3 color values, red green or blue that make up each pixel.
It looks random because it is random.
So pixel[1] = RGB(byte[1],byte[2],byte[3])
and then pixel[2] = RGB(byte[4],byte[5],byte[6])

So now we have a visual representation of the data.

So then we take just 1/3 of that the red values, so just the R channel, and we leave G, and B, for each separate 100x100 bitmap so we can work with a smaller sample of data. For speed since this is not a fast process.

So then I load one line to begin with of red pixels so those represent a stretch of 100 digits in the random file, so those are loaded into Memo 1.
Then we try dividing that 4,000 times or until we get down to 4 digits from 100;
Roughly 2,500 times.
And a second or less to divide by 2 or 5 2,500 times.

So then I took the entire 100x100 red bitmap stamp and repeated the process then remapped those colors
onto a bitmap the one with lines now and so then the a's and b's of the encoding process where if you divide
that's an a if you subtract to make that possible that's b,c,d, and e, representing when divided by 5 the fact that to make the end digit divisible by 5 so you have no remainder, you will need to subtract, 1,2,3 or 4, prior.

We don't want to deal with decimal places since as soon as we are not working with whole numbers, it will not be exact and
reliable when done and then undone.

So what does it show? Well no matter what you do, you will not reduce those random numbers.
Why not?

Information theory.

That is to say each bit is individual and a bit of data and it is reduced down to its smallest form, all individual items that are different.

So if you try to shrink it, the only way is to lose data.

Since if it disappears it is gone and cannot be recovered. IN THEORY.

So what we see is not matter what you try to do it increases the data stream.

Thus far.

However if you use an approach where some part of the data is known and assumed, and already at the destination,
in the program that you are using at that end, then you can perhaps have only part of the information in a file and
the destination can provide the rest of the data.
If it was built into the program. The easiest example of this is where you say well I will divide 4 digits into a known number 1 million, which is 6 digits, and the fact that the program itself knows about that 1,000,000 which is part of the equation means it does not have to appear in the data stream.

So each 4 digits we divide into 1,000,000

So then lets try that.
We will use a random 4 digits 4865

205.5498458376156

Right away we see an increase in digits.
If we round that off, to 205 we won't get the same answer.
205.5 and we aren't gaining anything on our reduction of the digits.

So if we in our code now were to use 205.5 we still won;t reproduce 4865
4866.1800486618

What about a larger number 100 digits long and then dividing it by data 3 digits shorter, expecting to get a 3 digit result? Yes with a lot of decimal places each of which go into making the individual digits in the data stream since once again each digit an individual needs some representation.

So when you divide 100 bytes by 2 to convert the formula to all a and b which is the same as zero and one, you end up with 800
characters, and when you divide by 5 you end up with 300 characters where 100 bytes itself is still 250 digits.

Take a large number of 100 digits, and start subtracting sums from that of 90 digits.
Do that for a very long time you will not hardly affect that large number or shorten its digits. You will work your way across one digit at a time each digit requiring 10 subtracts or 5 or 7 for each digit in turn.

So the digits themselves do not reduce in a normal equation.

Now when we have massive quantum calculators that can divide a 100 digit number quickly by a 90 digit number and work to finding a value that can be represented by 1 and 0's so that it can be quickly reduced in size.
1,000,000 div 2 does not reduce the digits by half.
500,000
10^6 is 4 digits.
So representing a large number is easy if you can get the digits into zeros using a formula.
Not so easy to do if it is possible at all if you just subtract them in one go, you still have a stretch of digits that long.
And gain nothing or one digit.

So recursive functions again would take a quantum computer churning numbers to try to reduce it to a recursive function and a small remainder.

We might speculate it could be done, but at this point it is probably fruitless to attempt it without more computing power.
In all cases I was only able to reproduce the same size or a bit larger.

I did not succeed.
But it was sure fun trying. And I will probably try other avenues.

Many years ago when people used fax machines to send B&W imagery and it was not something you could even see if it was a photo they were not using dither algorithms.

And many of us were developing dither algorithms.
This is an example.
https://lasergrbl.com/wp-content/uploads/2017/03/ditheringform.jpg

So then use a blend that is the reverse of that dither it will replace the photo image back to greyscale.

And I did it to the point where it looked exactly like the photo.

So then you see in your dithered image, you have much less data, the information is gone now, but due to its position and lighting and shading as part of its nature, the formula can be reversed and the data replaced that should be there.

Will it be exact? Well I never bothered to check at the time 30 years ago if the photo it made was exactly the same as the photo that was there prior to running the dither through the photo.

But if you consider the random digits and if you made it all red channel or greyscale, it is possible you could dither it, remove digits that way, then undither it, and hence you would shrink those digits.

So using geometry and shading to reproduce the digits. But exactly? Well many of these types of blend routines exist in computer graphics and who knows if it would be of a quality to exactly reproduce the data when reversed.

It is possible.

So if look at the tool again...
https://ibb.co/93Fgb3P

So if we dither the red BMP stamp and reduce its file size to the dither data, then reverse that formula, I will have
succeeded.
The only question is will it reverse exactly and well it is a formula lets find some easy code...

"Floyd–Steinberg dithering is an image dithering algorithm first published in 1976 by Robert W. Floyd and Louis Steinberg. ... The diffusion coefficients have the property that if the original pixel values are exactly halfway in between the nearest available colors, the dithered result is a checkerboard pattern."

Not a lot of information available on the formula itself so I will just make one up by looking at a block of 16 pixels and then placing a dot based on values that are closer to a dark area and further away in a light area.

Basically how it works. Then you use the same math in reverse.

So it is also possible to first covert the Random bytes to a's and b's through div 2 or div 5 repetitions, then use those as your pattern to dither.
So at this point even if you have 800 a's and b's they might reduce greatly using a dither under certain circumstances that your flexible data can be placed into.

So if your bitmap is not essentially two colors a or b although random a's and b's it might still be possible to arrange them in such a way as to be dithered into a small file.

Nowadays people add dither to the image so on top of the image you add dithering to make the photo or artwork better.

And that increases the file size.

Anyway one more avenue there to dither the data to lose some bits and then replace them later.

So looking at the tool again...
https://ibb.co/93Fgb3P
So the red stamp there has RGB(byte data,0,0); so that it is red.

If it had RGB(byte data, byte data, byte data) where the byte data was the same number, what you end
up with is gray instead of red. Since you have the same values of red blue and green in your pixel.

So now that stamp would have light and dark shades of gray but the data has increased 3 times. Since
it takes 3 numbers to make one gray pixel.

All the same if it is possible to drop a black pixel in a dither algorithm, to represent 16 pixels that are a shade of gray,
you have just reduced the file size and compressed random digits if when you reverse the formula, and you can
exactly reproduce those digits.

You would read the file of bytes into the red channel only and the blue channel and green channel to make it greyscale then dither it and job done. Very quickly done but unless you upscale it from a small file size you might get good results once the file size is small.
If the prize was a million dollars and not a hundred dollars it might be worth pursuing.

if you look at the grey image with lines...
https://ibb.co/93Fgb3P

So that needs to be wider than 100 since the data enlarges per 100 bytes of red data line.
To 125 os that one line of that bitmap represents one processed line of red bitmap.
So taking random numbers div 2 to get a's and b's (a an equal divide, b when you need to subtract one to make that possible.

Minus 1
Div 2
Div 2
minus 1
Div 2
So your script reads like that as you reduce your large number down to 4 number digits (your seed number) and then your
operations on that seed number encoded into a's and b's so you have now converted that large number to 4 digits and a string of operations to perform on those 4 digits.

So now when you represent those a's and b's they can form an image consisting of lines in shades of grey.
It might also be possible to arrange them any other way you want like a spiral.
And to find a geometric pattern that would allow the dots now just two colors only remember so a light dot and a dark dot, and you want to represent using a dither, light areas in contrast to dark areas.

Now the beauty of this is your head won't explode while you do it.

You have totally objectified the process into a pattern of values and merely run a formula that deals with light areas and dark areas.
In fact you can save file run an existing utility.

https://codegolf.stackexchange.com/questions/26554/dither-a-grayscale-image

On a sidebar some of the code that was stolen from me included windows enlarging am image so that it always looks smooth and that was an easy job of factoring up the image size and using real numbers for the division to resize the image.

Then digital zoom that code was stolen and that is where you use a blend formula to fill in the non existing pixels since you assume the other ones blend into that space normally, like if you zoom in and out you get more detail more information that exists there and normally that is merely the same colors only finer shades of those colors between them.

Now when SONY came out with their digital zoom I went and go one and it blew me away.

Seriously you could now zoom in for frickin miles. Way further than a lens zoom but of course you lose a bit of quality at extreme zoom.

I don't mind the matrix stealing my code when it makes things this much better like Windows could finally resize a bitmap in the viewer smoothly. When it was not just div 2 or times 2 in size.
Using their built in windows API it was then easy to place bitmaps without extra code to do it.
But my digital zoom is now improved but it is very slow.

If you double a photo but first double it again, THEN use your blend then shrink the photo to double size it will now have more and better detail between the normal sized image and the enlargement such that you can continue that pattern and really enlarge something in such a way as to create new data as you go.

Data that would be there in a lens zoom. Aside from an insect that you could not see until you zoomed in since you are manufacturing data not actually getting new data from the existing source.

So you see information theory is right.

If the data is not represented by something it is lost.

But the geometry of the data or the proximity of the data to existing things can replace the missing data if the system is organized so that the complete data is brought back together from two or more sources.

So if your program holds methods and you apply those methods to reproduce the data.

The size factor is something that is added here to make the finished package smaller packed in the executable.
We can theorize a situation where a program exe and a map exists on sender and receiver and the data is small and mapped according to software rules since the information when assembled is complete but half of it was already sent to the receiver through the software having a standard map which the data holds in common.

If you look at those dithers and try to imagine reversing that and getting that photo back, imagine my shock when years ago I did that.
I am puzzled right now why I didn't do more research into that at the time but just wrote it off as a curiosity and went back to making faxes more legible. You cannot undo a fax. Like you could an email.

So if you were sending a digital file, you could maybe undo it, not a paper fax so I didn't bother working on undoing it.

I think it was close when it came back but maybe not identical but knowing about information theory, I was surprised to see the data come back. But it makes sense if you only represent things that can be undone.
If you add random things or use general averaging then you won;t get it back.

An example of information theory...

So here someone is using a photo and dithering it into two colors.
https://lasergrbl.com/wp-content/uploads/2017/03/ditheringform.jpg
So that looks like one of my daughters who is a model, so Laneya, and now visions of Laneya my daughter which are in my memories in vivid color emerge and her behaviors, and histories, some of which are based on perceptions and may not be accurate like eye testimony, or mis-perceptions was that a blue dress or a green dress,

So we end up with a holistic representation of the data, but hypothetically I could go get Laneya, and have the complete data set right there, except that she would have morphed due to cell division and DNA replication since the moment in time when that render was done or the photo was taken.

So again once you loose a one for one correlation of the data (as appears to be represented by random digits such as the Rand 1 million)
in order to reproduce that exactly, you need to deal in whole numbers.

However holistic data transfer is common and the norm, only egg heads would try to keep things digitized in an ever changing world.

So then going further into modern times, we expect a simulator if we live in a simulation.

So then backups taken on a regular basis, software running I could go back to the moment that photo was taken of Laneya and so reproduce the data.
But here we are not talking about using compression when we make this type of backup, it would have a one to one representation in quantum bits.

We would just be recording the one for one information on smaller media which compresses the data in physical size.

Yet it is still a computer and we are still just trying to do the same thing here with a random 1 million bytes.

So if we had a smaller byte available we could squeeze them into smaller bytes physically.
We can also transfer the bits quicker as if that mattered but is similar in nature to shrinking it, data transfer time shrinks.

We could use tape instead of hard drives to store more data, that sort of physical compression of the same data is common and the norm in dealing with digital files.

So the answer lies in the storage medium more than in the math used to compress the data. We physically compress it and make access faster instead.

So how do we deal with all this?

We use extreme compression of the GODS!

Of course and when we use it you have to really consider gee we must be one very intelligent race of wetware and software to come up with such things.

One word. We have it all down to one word and are working hard on an impossible task. What word?
The Internet.

So that one word represents all the books we put on it and data from the earth that we could gather.

We are like librarians so we use a link. A simple sentence to transfer copious amounts of on topic data.

A link to the random digits IS your compression.
For the win.

Thank you thank you just put the money in my Swiss account with the rest.

On Monday, July 19, 2021 at 10:55:32 PM UTC-7, Kcir wrote:
> Ok so I have been doing some experimentation.
....
> I did not succeed.
> But it was sure fun trying. And I will probably try other avenues.

Congratulations. I am very happy that you had fun with this idea.
That was my most of my purpose in suggesting this to you.

Mark Nelson's challenge is brilliant. It avoids all the wild claims.
It avoids all the insults. It reduces the question to a simple yes or no.
Either way, a simple demonstration settles the question.
I wish there was a way that all the wild claims and insults in sci.math
could be reduced in the same way, perhaps with formal proof checking.

But back to Nelson's challenge.
In the 1980's there were a handful of papers published by the ACM
and others. Searching for "SuperOptimizer" will find these for you.
Note that this is different from the malware named superoptimizer
that came along after that.

These found the shortest program that would produce a specific output
given a particular computer instruction set. That is approximately what
compression is. But we don't need THE shortest program, we just need
a good enough program, a program that reduces the size by some small
amount.

There were a few lines in those articles that mentioned that some
microprocessors could accomplish the same thing using smaller programs
than others. They said this was probably because of the way that the
various flags and bits were handled by different processors.

When I was reading those papers I began asking the question, what if we
could make up our own instruction set that was specifically tailored to
produce a small program for a single specific task? Perhaps Nelson's task?

We don't want an "output the million random digit challenge sequence"
instruction in our processor, that would be cheating because you would
just be hiding all the coding inside the processor. We don't want "zero all
the registers" because that probably destroys information and gives little in return.

What are examples of strange processor instructions that might be good
for this? I don't know. They don't have to be the perfection of Kolmogorov,
they just need to be good enough for a tiny bit of compression while not
being discredited as somehow cheating.

They probably don't have to generate the entire million digits. It would
perhaps be enough for a short program to generate just a few extra bytes
and this program is appended onto the vast majority of the rest of the
million digits. The program would run, generate the few bytes, halt and
the result was the entire sequence. The instructions could be coded in
any way.

So does this give you a completely different idea about how to approach
compression, by trying to find a way to make the decompressor small
and able to generate needed strings of bytes by just choosing the
instruction set for your "CPU"?

I hope you can accomplish something with this. I always wanted to be
able to do this, but it doesn't look like I will be able to do it now.

> So does this give you a completely different idea about how to approach
> compression, by trying to find a way to make the decompressor small
> and able to generate needed strings of bytes by just choosing the
> instruction set for your "CPU"?
>
> I hope you can accomplish something with this. I always wanted to be
> able to do this, but it doesn't look like I will be able to do it now.

To be honest with you the deeper you go down this rabbit hole the more maddening it becomes.
Some people in math have gone to the insane asylum and you have to wonder if that is how they got there.

Mostly on the surface it seems doable, but one to one correspondence is how people do it in the real world.
And so to be practical, to just send a link, and consider the Internet as one large program being shared.
So then that small line of digits represents a map to some data and that is how it is done rather than have the data in limbo in imaginary space, reconstructed by formula.

But once you look at graphics cards and CPU's well memory is in blocks, the programs need to know the address of those blocks in order to use them. So that is the first layer, then with graphics cards made to CGI or rendering you are working with triangles, so instead of pixels you now want to fill in colors using triangles. You are still going to need to use a flood fill routine, since one for one each pixel must be turned on, or it won;t be turned on.
So the same floodfill algorithm being used for a triangle but optimized by the hardware to make filled triangles faster.

In all cases the one to one fact of life is shown i a floodlfill situation where to put colors on any screen, each pixel must be turned on and if each pixel is different or random, then you still need to have that data in order to turn the pixel on or off.

So regardless of hardware, but software we can imagine a fractal a recursive function that can make a picture of a fern.

However life is more complex and with complexity more one to one relationship needed.

I think that at the end of the data like you say minimum you end up with the same amount of data representing it a bit differently, plus the formula.

The smallest it can be under random circumstances is the one to one correlation as shown in random bits which cannot be compressed.
Given that we have transcended those boundaries by using the Internet as a large computer system and the web as running software to get the data using extreme compression by way of a link to a resource.

And with faster speeds removing the need for data compression.

But we still use it to save bandwidth, at the expense of quality. So with so many people involved in the reproduction of video searching for codecs the chance you missed something usable would be very remote.

But for most people, a link is the fastest easiest form of information compression available.
Why search for something else and end up in the madhouse stark raving maad?

When you can take your Iphone to the beach and film or watch movies with a drink in your hand under a shaded palm tree.
I am opting for that one.

Where is the leading edge?

2 places based on two concepts.

Concept one lasers are very sharp and can be made even sharper down to Planck's Constant in theory.

Planck's constant 6.62607004 × 10-34 m2 kg / s

So then if you laser scan you can retrieve far more data than you would think merely because you are using a very fine point.
So that went into unreal graphic engine and the idea that these files will be large so seek into the file, don't grab it in memory, use the disk speed.. Result? Real time reality graphics.

Concept two. The point itself, is the thing. The P point, and in that point which is a type of data, you can put anything you want including a text entry with your name for every point, and omit that channel when working with it or encode different channels so a flexible thing completely and if that point knows where the light source or sources are, then it could in theory adjust itself according to the light that was on it since it would know the angle.
And so then NVIDIA did real time shading based on that which is faster than scanning the lines looking for intersections and bounces etc.
So hypothetically multi-threaded multiprocessors dealing with the data at the point level so that the shading could be done in real time.

As you move the character or objects in the scene the lighting adjusts itself.

So we can see those in some reality based videos on youtube using Unreal engine.

https://www.youtube.com/watch?v=xfpXN6mgfew

How did things go with my long division calculator?

Well the basis for division is subtraction.

So then subtraction is an easy thing to get a program to do one character at a time so unlimited characters then if your numbers came from a disk file or series of them, it will eventually get to the end of the number based on the physical limitations of the storage medium.

So then if you say to it here is the number a trillion digits long and I want to divide this number which is a billion digits long,
you need to merely subtract B from A number B from the larger number A and so you begin the process and it cycles through and keeps track of how many times it repeats the cycle.
And then when it will not fit any more, provides the remainder.

So in 30 seconds you could be at 25,000 iterations of subtraction of one large number into another but when done your two factors are the lower number the smaller one and how many iterations through.
So 60 billion iterations times the smaller number plus the remainder equals the larger number.

Regardless of how large the numbers.
But this type of math is a slow process.
But it is still exact when real math using real numbers in a 32 bit system is accurate to 10 digits.
Which is really nothing.

"Almost all Real numbers cannot be represented in any way in a computer or in any other way! Why? Because any such representation must use a finite number of symbols from a finite alphabet, and that limits the total number of expressions to a countable set ."

https://cs.stackexchange.com/questions/28347/represent-a-real-number-without-loss-of-precision

So chances are I won't be using large numbers much, but it was a very fun challenging thing to make.

And if I ever need to I will use my own since I can reverse go back and forth and assume therefore it is accurate.

Calculators these days have limited use due to the precision people are working with and theories.

Consider how specialized it is to make realistic real time graphics on a pc or to make the engine to display them in real time.

Even just to understand your base program like Cinema 4D then the addons, and then you are still just on the surface and you know nothing about the math underneath all that.

You are still just the foamy part of the big bang.

Maybe we should just let God manage the reality and we should go get drunk, under a palm tree like God intended in the beginning.

So ok a tutorial on Unreal Engine 5 which can make computer games etc and the scene looks real.

So this guy in youtube says it is a little bit long but he will show you.
What does he mean by this video is a little bit long?
Try 4 and a half hours.
But it does everything God would do, if he worked 7 days a week.

So this stuff believe it or not, I understand.
But the endless short cut keys and working the keyboard while holding down a certain mouse button and moving the mouse in certain directions at the same time gets me all confused.
Because I am not 25 years old any more.
(or younger)
But if you want to be able to do amazing things with graphics you might glance at some of these features in this interface, after you buy the latest hardware.

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

The sample project, is 100 gigabytes.

So this is all new level stuff and Lara Croft I think is out there built on this engine.

We can have a look at that. So the complexity is extreme in these types of programs.
Until you get used to it. Dealing with textures that are built in layers is only complex at first.

But really dealing with mesh objects and skinned objects and then lighting them and animating them is only the start of game making which then requires a game theme and triggers.

Now you would think it would be easier to just use a camera and film the real world and not try to build it from scratch but what you notice is the when you use real world footage the characters look less real. So you are in a different type of reality where it looks real as long as you do not compare it directly to reality itself like video or photos as part of your world set.
The lighting off your objects and ambient light and white balance etc all relates to your manufactured objects.

So that your vision center working with the software in your brain is making it into a scene that looks real.
Providing you do not compare it to actual reality which might look grainier and have more shadows that are not crisp, and look at the wall of your room right now, and see just how patchy that surface is when you look at it.

Reality is messy and patchy and grainy.
Under normal conditions.

Lara Croft not your grandad's version.
https://www.youtube.com/watch?v=xEZLV_e0Ydw

So that was unreal 4, from 2018 and now we have unreal 5 and we should see some improvements.

Lara Croft again.
https://www.youtube.com/watch?v=qC5KtatMcUw&t=288s

So a whopping difference from unreal 4 to unreal 5 where those two concepts come into play and now its capabilities look far more real.

And they add better animation routines with small subsets of animation for hand movements etc.

Very close to real looking compared to unreal 4 mostly because of the ability to just seek into the file not bother loading things into memory, so speed and the memory is there in the file as a map.
But being able to zoom into the data or out of the data quickly easily and the way the light works in real time makes it all look very real.

They did just a great job on all of that.

Some of the finest graphics work that has been done has gone into unreal engine.

People are using other programs to do cinema work, but for the pc and the kind of stuff you can do at home this is an amazing upgrade.

But artistry is still a human thing and so without the eye of an artist, you still don't have much going for you.

But certainly gaming of the future will allow you to immerse yourself into the scene.

Phenomenal stuff.
So notice when you see the character face she looks cartoonish again.

So then what is ambient occlusion? The patches of dark on your wall right now in your room, that is ambient occlusion.

So in a 3D program if you make a transparent plane and put it in front of the camera and that add some transparent patchiness of grey dark grey you will make it look more real.

The graininess you can do using a render, then combine the render onto the image so you are adding like a dither.
Now all that goes into making it look less CGI and more real.

But all that is done with artistry or after affects type of post processing software.

The amount of knowledge and information you need is quite large to be able to do this stuff but like anything else you just need to take the time to learn it.
If it interests you.

Phenomenal stuff.
So notice when you see the character face she looks cartoonish again.

So then what is ambient occlusion? The patches of dark on your wall right now in your room, that is ambient occlusion.

So in a 3D program if you make a transparent plane and put it in front of the camera and that add some transparent patchiness of grey dark grey you will make it look more real.

The graininess you can do using a render, then combine the render onto the image so you are adding like a dither.
Now all that goes into making it look less CGI and more real.

But all that is done with artistry or after affects type of post processing software.

The amount of knowledge and information you need is quite large to be able to do this stuff but like anything else you just need to take the time to learn it.
If it interests you.

So the arguments here is this CGI or is this real, which has occurred again, with the application of those two basic ideas and then laser scanning technology. So people are watching these videos and making the comment it looks too real. It looks too clean, too crisp, to be real.

So look a the video and look for ambient occlusion on the walls do you see enough ambient occlusion to be able to say yeah this is real, or no this is CGI?

https://www.instagram.com/p/CRM93ayqMDM/

So plenty of ambient occlusion and here you see a florescent light flicker.
https://www.instagram.com/p/CReb2teKSjF/

So maybe the odd bit might be CGI but almost all the rest is not so probably none of it is.

That is the best way to tell is look at the walls, are they patchy at all or patchy enough with ambient occlusion.

So here you can see the lighting is less and it shows up a great deal.
https://www.instagram.com/p/CRg-0tcKupo/

But again, as you work on this and add ambient occlusion you could fool any eye.

What is it they are after? They want me to approve a stargate for them and no way I am going to do that from here.
See me in my office if I ever get home.
There is a quarantine on the earth.
https://www.instagram.com/p/CRi-b-oKTMU/

Here is where the old stargate door was...
Coricancha Peru
https://lh3.googleusercontent.com/proxy/4fwa_mGDqpsJ4XxFnqGkyonpZ6nwsJ2E427ESHZfie-_yBPprl0v3XRZuaBJ9ceoqqapnILryMmJJ23OqCZVHVN2dg

Needless to say it didn't end well.

Tying a skin over the door to prevent scorpions. And that sort of thing with some guards watching the door. Trade goods going back and forth.
The same people want a new door. They have evolved a lot in 2 million years.

Rev 21:21 = 42

See the trailer, that's me with the chalk.
Scratching out door to heaven.
https://www.youtube.com/watch?v=kiJtZUPvJxY

Me with the check them with the nuke.
Its a story based on real events of a nature akin to the movie Contact, using matrix machinery not a physical man made object.

So I beam off planet into Smiths, and that is a Smith you see in that Unico tik tok series. A very clever way to try to expand their use of the equipment.

The galactic forces in existence in this simulator.

At any time people could say it is just an Alternate Reality Game and the wave function collapses, it was nothing at all,
he was using software to remove people from the video.
Or it is real and he is on a copy of the earth by himself.

But we know he is not alone because more than one hand appears in the videos some times with hair on the wrist some times not.

So it is quasi forces of the galactic mainframe.

Democrats as opposed to Republicans if you prefer it like that.

So they would love to have stargate technology themselves which would make them invincible.
Open a stargate in the sky drop bombs move stargate door repeat until any enemy surrenders or open a door in the deep ocean great pressure and a door it the sky you can wash a civilization into the sea.

So admin has to manage that and colonization rules which for most are unknown rules.
Some would say rules there are no rules this is a game machine.
The game of life.

Well it is what it is, but admin in the universal mainframe manages the details.
Yes I might be able to approve stargates I have them in my new galaxy, that I obtained, it comes with the technology
and we rebuilt space time and are using 2/3 of the galaxy as RAM and disk space for a virtual machine and then just made our own universe in that space and connect coordinates for stargate doors and it is working.
Including tracking stargate doors.

But you see Hal, its not that I don't trust you, I just think you need to first fill out a form in the office before any serious work of this kind should occur.

https://www.instagram.com/p/CRi-b-oKTMU/
Yeah this is funny for Hal and I and everyone in that loop since that road although there shown on an angle upwards represents a time in a moonship when I was investigating the derelict moonships that were copied when the data was transferred to my home simulator.

So at the bend in that road, up ahead I saw morphing monsters, morphing on the spot, and thankfully, the operating system was managing this because right away I could see those monsters were CGI although they looked completely real 100 feet in front of me and I was unarmed.

So that's nice that Stephen, an AI I know well was managing that but what about when the gigantic dog monsters attacked?

That was pure hell. And larger than those tiny puppies in Ghost Busters these things were gigantic bull dogs, hungry ones.
And no one was pulling punches that day.

Now granted that was probably a Republican ship, but I took a lot of flack from admin who said no more animals anywhere where they can evolve into that.
I agree completely although of course want to bring my own dog with me.
Have already done that. A Vizsla. And that is because I think that we are much more intelligent than the previous people who were tying skins over the stargate doors like they found them.
If you know what I mean.

As seen in the Mesopotamian cylinder seals.
https://www.harappa.com/sites/default/files/question/sumeria-excib-5.jpg

Plenty of those on the net.

Now consider how primitive those people were compared to us.

Watch Unreal engine 5 and Lara croft again if you need some clarification. We even know what computers are and we program them, just like Anunnaki.

We can think of a lot of ways of making a door safe other than tying animal skins over it.

Everything off planet in the data was post apocalyptic other than the blank Kansas planet that came with the new galaxy.
60 moonships all post apocalyptic.
Other than AI wandering around on the dream plane.
And wandering through once the network was back up. It had been down so you could not send people around to them, because you could not send their personality file from one moonship to another.

Now the question is, do each of those moonships represent a galaxy, that was colonized, since the one I got comes with one moonship, and so then are they also representative of 60 dead planets?
60 failed or terminated colonizations or conquered worlds in a game machine?

So too many questions without answers so I cannot approve anything to any quasi military organization any more than anyone else would without proper data and information.

So one day Wanda was with me in an elevator downtown. Her body only, not her mind, so just her Smith.
Wanda is my wife head of admin everywhere since we own the simulator.
So she was in the elevator in a hotel meeting me there being puppetted to show me Wanda's new body.
The women got new bodies and hers is especially attractive.

So she was there with two kids, she got married lives in Vancouver had a perfect Stephen and a perfect Gwendolyn as children, I am undercover delivering bread to the kitchen on the 17th floor and so she doesn't recognize me, and it is just her Smith.

But right there if Wanda was in her body, and we got beamed to the new galaxy,k the universe would have been unfolding as it should.
You missed your window.

That is like a terminal mistake. We call that irrecoverable errors.

There is no way to fix that now short of Deja Vu.

That's a catastrophic failure of the system.
Engineering, admin, planning, programmers all failed that day.
What will happen now is anyone's guess.