928 Pasche pattern formula?
#16
#18
Rennlist Member
Join Date: Apr 2012
Location: Denver Foothills, CO
Posts: 128
Received 0 Likes
on
0 Posts
I remember seeing Pasha for the first time as a tween; a friend had pasha wallpaper! I was so jealous.
Pasha looks a bit like a sequency-ordered Walsh-Hadamard matrix. Of course, not exactly the same.
https://en.wikipedia.org/wiki/Walsh_matrix
Pasha-like patterns and some other very wild stuff drop out from Pascal's triangle.
#19
Rennlist Member
Join Date: Apr 2012
Location: Denver Foothills, CO
Posts: 128
Received 0 Likes
on
0 Posts
Pasha looks a bit like a sequency-ordered Walsh-Hadamard matrix. Of course, not exactly the same.
https://en.wikipedia.org/wiki/Walsh_matrix
Pasha-like patterns and some other very wild stuff drop out from Pascal's triangle.
https://en.wikipedia.org/wiki/Walsh_matrix
Pasha-like patterns and some other very wild stuff drop out from Pascal's triangle.
https://en.wikipedia.org/wiki/Walsh_...d_Walsh_16.svg
Here's a 8096x8096 sequency Walsh matrix image which shows more of the Pascha if you zoom in 400+% and pan around the image. You have to zoom because each pixel of the original image is one entry in the Walsh matrix.
https://www.flickr.com/photos/423912...37618/sizes/o/
#20
These guys might know the "formula".
http://www.lakewell.com/en/general/f...rboard-fabrics
Wish they had a Beige/Black combo.
http://www.lakewell.com/en/general/f...rboard-fabrics
Wish they had a Beige/Black combo.
#23
Rennlist Member
These guys might know the "formula".
http://www.lakewell.com/en/general/f...rboard-fabrics
Wish they had a Beige/Black combo.
http://www.lakewell.com/en/general/f...rboard-fabrics
Wish they had a Beige/Black combo.
If I read it right.
Last edited by soontobered84; 04-15-2015 at 01:53 AM.
#27
Rennlist Member
#28
Rennlist Member
I found myself studying this pattern for one of my projects, and I too had some difficulty figuring it out. First, there are numerous variations of the pattern that can be found online. However, when you study the one that Porsche appears to have used, as shown in the posts above, you will find the following:
The group of what I will call small blocks in the center of the small block area consists of 7 rows each way. In that group four of the rows are the same width and are a bit wider than the other three which are also the same width, and the seven rows are arranged alternately. In the center of that grouping there is a white square. (I am using black and white for this explanation.)
At the other part of the pattern, the larger blocks, the center of those is not a square, as one might expect, but rather the corner of four large blocks which are each square. Therefore, in order to accomplish that, rather than also having a white square in the middle, as with the group of small blocks. the number of rows between the small grouping and the center of the larger blocks must differ. The distance from center to center is apparently the same but the number of rows is not.
In looking at the pattern above, you will be able to count, left to right, from the center of the large blocks to the edge of the small block grouping, 7 rows of blocks on one side and 6 on the other. In order to accomplish that in the same distance the rows must differ in width on one side from the other.
Also, in looking at the pattern shown then counting up from the center of the large grouping to the edge of the small grouping there are 5 rows on one side and 6 on the other.
The pattern is not symmetrical, and if it was the center of the large block grouping would also be a white square rather than the four corners of four squares.
When you study the application of this pattern as Porsche applied it to the front seats of the 928 although the center of the small grouping is usually centered, the ends of each row, left to right, usually ends in white on one side and black on the other, which makes the pattern look like it is not centered correctly.
Edit: On closer inspection I find that of the 7 by 7 group of small blocks the row on the left is actually a bit narrower than all the rest. That must be where they started their "adjustment" to get the extra row on the left in order to end up with a four corner middle rather than a square in the middle. Similar might also be true of the center group of small blocks. There might only be 6 rows of them from top to bottom in that grouping rather than 7.
EDit: another look at the pattern shows that the grouping of small blocks actually consists of 7 vertical rows and 6 horizontal rows. The center of that cluster is on the line between the center most black and white squares.
The group of what I will call small blocks in the center of the small block area consists of 7 rows each way. In that group four of the rows are the same width and are a bit wider than the other three which are also the same width, and the seven rows are arranged alternately. In the center of that grouping there is a white square. (I am using black and white for this explanation.)
At the other part of the pattern, the larger blocks, the center of those is not a square, as one might expect, but rather the corner of four large blocks which are each square. Therefore, in order to accomplish that, rather than also having a white square in the middle, as with the group of small blocks. the number of rows between the small grouping and the center of the larger blocks must differ. The distance from center to center is apparently the same but the number of rows is not.
In looking at the pattern above, you will be able to count, left to right, from the center of the large blocks to the edge of the small block grouping, 7 rows of blocks on one side and 6 on the other. In order to accomplish that in the same distance the rows must differ in width on one side from the other.
Also, in looking at the pattern shown then counting up from the center of the large grouping to the edge of the small grouping there are 5 rows on one side and 6 on the other.
The pattern is not symmetrical, and if it was the center of the large block grouping would also be a white square rather than the four corners of four squares.
When you study the application of this pattern as Porsche applied it to the front seats of the 928 although the center of the small grouping is usually centered, the ends of each row, left to right, usually ends in white on one side and black on the other, which makes the pattern look like it is not centered correctly.
Edit: On closer inspection I find that of the 7 by 7 group of small blocks the row on the left is actually a bit narrower than all the rest. That must be where they started their "adjustment" to get the extra row on the left in order to end up with a four corner middle rather than a square in the middle. Similar might also be true of the center group of small blocks. There might only be 6 rows of them from top to bottom in that grouping rather than 7.
EDit: another look at the pattern shows that the grouping of small blocks actually consists of 7 vertical rows and 6 horizontal rows. The center of that cluster is on the line between the center most black and white squares.
Last edited by Jerry Feather; 10-20-2017 at 12:35 PM.
#29
Rennlist Member
Now that I kind of understand the pattern a bit I think I'll redraw it symmetrical, both directions, and then turn it 45 degrees for my seat project, if I can find a way to have it printed in Red on some light tan leather.
#30
Chronic Tool Dropper
Lifetime Rennlist
Member
Lifetime Rennlist
Member
Jerry:
I dinked with a Pascha "formula" somewhat on a lark, and came up with what seemed at the time to be a starting point. I plugged a sine curve into Excel and made a linked AutoCad image with the numbers. Look at post 20 in this related thread. Obviously we can plug in almost anything into Excel drive the output. If you happen to have a parametric CAD program I'm sure you can do something similar. Surely there are plenty out there in the freeware realm; I happen to know and use AutoCad. The best program is the one you know.
Next Option: If someone has an actual sample of the fabric from which I can get dimensions, .or. can take a picture with a scale on the pattern, I can generate PDF or EPS (or Gerber or whatever...) to support printing new fabric. I just need one panel for the scale or a matrix of dim's.
** the formula I came up with looks at 0-180º of the sine curve, 9º for each frame (180º/20 frames repeat), with a 1/2" fixed offset (minimum dimension of any rectangle). I'm open to any suggestions...
I dinked with a Pascha "formula" somewhat on a lark, and came up with what seemed at the time to be a starting point. I plugged a sine curve into Excel and made a linked AutoCad image with the numbers. Look at post 20 in this related thread. Obviously we can plug in almost anything into Excel drive the output. If you happen to have a parametric CAD program I'm sure you can do something similar. Surely there are plenty out there in the freeware realm; I happen to know and use AutoCad. The best program is the one you know.
Next Option: If someone has an actual sample of the fabric from which I can get dimensions, .or. can take a picture with a scale on the pattern, I can generate PDF or EPS (or Gerber or whatever...) to support printing new fabric. I just need one panel for the scale or a matrix of dim's.
** the formula I came up with looks at 0-180º of the sine curve, 9º for each frame (180º/20 frames repeat), with a 1/2" fixed offset (minimum dimension of any rectangle). I'm open to any suggestions...