Need help understanding GFDisplay Masking

[Guest jcardana]

I hate to bother anyone with this... I've been through the docs several times, I'm just not getting it.

I'm setting up 8 switches to control the lighting (offset 0D0C U8). The part I'm not understanding is using the Mask.

Mxxxx: Mask

This element provides a hexadecimal value to be used as a mask on the value selected by an Xxxxx or IXxxxx element. This is

most often used to test individual bits in an array of flag bits. The result of an offset value and a mask is the logical AND of the

two.

Yeah, I don't understand. This element provides or requires a hex value. I understand it to require one. I don't know HEX well enough to know what to use to look at certain bits.

I've been through the GFdisplay guide and there's a few references which I used to try to figure this out.

4=X3300 U16 M0100 ; FS LOC acquired  
8=X3300 U16 M0080 ; FS NAV CRS acquired
44=X3300 U16 M0002 ;Active Nav1
45=X3300 U16 M0004 ;Active Nav2

The examples in the GFDisplay.ini show M0002 either looking at bit 1 or 4 / M004 looking at bit 2 or 5 / M0080 looking at bit 7 and M0100 looking at bit 8

0 = reserved

1 = good NAV1

2 = good NAV2

3 = good ADF1

4 = NAV1 has DME

5 = NAV2 has DME

6 = NAV1 is ILS

7 = AP NAV1 radial acquired

8 = AP ILS LOC acquired (incl BC—see 10)

9 = AP ILS GS acquired

10=AP ILS LOC is BC

11=good ADF2 (FS2004)

12=NAV2 is ILS

13–15 reserved

(hex 8000, or bit 15, the top-most bit in the 16-bit word).

I don't understand how M8000 looks at bit 15. I just don't understand the hex sequence between bit 0 and bit 15.

[Pete Dowson]

I don't know HEX well enough to know what to use to look at certain bits.

Time to learn, then, as it is really very very simple.

Hex is like decimal, but instead of using 10 digits (0-9) it uses 16 (0-9 then A-F). So each position in a number can be any one of 16 values instead of 10 values.

Letters A-F represent values 10-15.

Now, in decimal, each digit in a number represents 10 times the next digit. i.e. 123 = 1 x 100 + 2 x 10 + 3 x 1.

In hex, because the digits go up to 16, each digit represents 16 times the next digit. i.e. HEX 123 = 1 x 256 (16x16) + 2 x 16 + 3 x 1.

Okay so far? That's Hex.

Now to bits. In binary there are only 2 digits, 0 and 1. But the principle still applies. Now each digit represents 2x the next one. Remembering this it is easy to see that

Decimal  Hex  Binary
 0            0      0000
 1            1      0001
 2            2      0010
 3            3      0011
 4            4      0100
 5            5      0101
 6            6      0110
 7            7      0111
 8            8      1000
 9            9      1001
 10          A      1010
 11          B      1011
 12          C      1100
 13          D      1101
 14          E      1110
 15          F      1111

See how one HEX digit represents exactly 4 binary digits, or bits? That is VERY useful, and why it makes HEX so useful and easy as you'll now see:

For a 16-bit value, like offset 0D0C (the lights), there are 4 x 4 bits (4 x 4 = 16, easy). So, these can be written as 4 HEX digits. A pattern of bits like this:

0 0 1 0  0 1 1 0  1 1 1 0  1 0 0 1

would be written in HEX as 26E9, because

0 0 1 0 = 2 (see table above)
0 1 1 0 = 6
1 1 1 0 = E
1 0 0 1 = 9

Now all you need to know is how the bits in a 16 bit value, like 0D0C, are numbered, and you then have the whole story.

By convention, bits are normally numbered according to their "power of 2". In effect this means that the LAST bit, the one worth a measly 1, is bit 0 (2^0 = 1). The next one up is bit 1 (2^1 = 2 ), the next 2 (2^2 = 2 x 2 = 4) and so on, up to the first, or "topmost" bit 15 (worth 2^15 = 2 x 2 x .... = 32768).

So that hex pattern above, 26E9, which as we saw represents these bits:

0 0 1 0  0 1 1 0  1 1 1 0  1 0 0 1

the numbers of the bits which are set are, from left to right:

13, 10, 9, 7, 6, 5, 3 and 0.

Finally, if you ever need the DECIMAL equivalent of such a HEX number, you'd probably do better with a calculator, but the method is:

e.g. 26E9

Step 1: 2 x 16 = 32

Step 2: Add the 6 = 38

Step 3: x 16 = 608

Step 4: Add the E (= 14): 622

Step 5: x 16 = 9952

Step 6: Add the 9 = 9961.

Answer = 9961.

I have to do a lot of such conversions as a programmer, so i use a special calculator that does Hex to Decimal and Decimal to Hex conversions. ;-)

I don't understand how M8000 looks at bit 15. I just don't understand the hex sequence between bit 0 and bit 15.

Well, I hope that's clear now. If not, then I'm sorry but that's the best I can do. You need a basic guide to computers.

Regards

Pete

[Guest jcardana]

Wonderful explaination and thanks for taking the time.

I'm good up until this part... this is where I get lost.

So that hex pattern above, 26E9, which as we saw represents these bits:

0 0 1 0  0 1 1 0  1 1 1 0  1 0 0 1

the numbers of the bits which are set are, from left to right:

13, 10, 9, 7, 6, 5, 3 and 0.

I would have thought this...

  2     6     E     9
0010  0110  1110  1001
  4    64   16384  512

Is the last bit on the right or left? Am I reading this backwards? It just hasn't clicked yet.

[Pete Dowson]

I'm good up until this part... this is where I get lost.

So that hex pattern above, 26E9, which as we saw represents these bits:

0 0 1 0  0 1 1 0  1 1 1 0  1 0 0 1

the numbers of the bits which are set are, from left to right:

13, 10, 9, 7, 6, 5, 3 and 0.

How can you get lost here when all we are doing in numbering the bits. the same as numbering houses in a road or students in a line.

Look, if I have a 4 bit number: 0 1 0 1 (which represents 5 in decimal), the LAST bit, the 1 at the end, is bit 0, the one to its left it bit 1, the next is bit 2 and the 4th from the right is bit 3.

With a 16 bit number, the 16th bit from the right is bit 15 (it is always one less because we count from 0, not 1).

I would have thought this...

  2     6     E     9
0010  0110  1110  1001
  4    64   16384  512

What system of bit numbering are you using which gets up to 16384 bits in a 16 bit vaslue? And goes up and down? Can you explain what those values, 4, 64, 16384 and 512 are supposed to represent and how you arrived at them. you've got me completely lost. i've no idea how you are thinking, even! :-(

Is the last bit on the right or left? Am I reading this backwards? It just hasn't clicked yet.

What do you mean by "last" bit, exactly? It isn't a very precise term except in an ordered list. The last bit reading left to right is the right-most bit, the least significant bit, or bit 0. I've no idea how you are using your "English" words I'm afraid.

Pete

[Guest jcardana]

I was looking at the values that the set bits represent, I was looking at the bytes, not the bits.

I now understand how the bits are numbered,

the numbers of the bits which are set are, from left to right:

13, 10, 9, 7, 6, 5, 3 and 0.

This just threw me off, I though you were saying we counted starting from the left.

I'll keep studying to figure out how the mask in GFdisplay is working. I understand it will allow me to look at specific bits within the offset.

[Guest jcardana]

IT CLICKED!!!

M8000 is looking at the 16th bit because...

8    4    2    1
8    0    0    0

8*8=16

and if I wanted to look at bit 5 I would use M0101???

[Pete Dowson]

I was looking at the values that the set bits represent, I was looking at the bytes, not the bits.

Well, that's not relevant at all to want you want to do in GFdisplay, but if you want that, then this:

      2     6     E     9
    0010  0110  1110  1001
      4    64   16384  512

still makes absolutely no sense. The highest bit, the "2" is "26E9", is bit 13 so, as I so carefully explained, it represents 2^13, or 2 x 213 times. ie. 8192.

There are then 2 bits in the next hex digit (the 6), so how do you get 64?

Ditto how do you get 16384 for the 3 bits in the 3rd hex digit, "E"?

And the last hex digit is 9, which, wonderfully simply, represents decimal 9 too. But you get 512 from somewhere? How?

Even if you are reading right to left for some reason, your numbers make no sense. Can you explain them?

Like decimal numbers, binary and hex read with the high values on the left, the low values on the right. why would you think everything was reversed in order just because we use a different base?

I really cannot help any further sorry. I suggest you get a simple book on computers.

Regards

Pete

[Guest jcardana]

I was doing the powers of 2, i just did it wrong...

Dec Hex    Bin    2^x
0	0000	0000	1
1	0001	0001	2
2	0002	0010	4
3	0003	0011	8
4	0004	0100	16
5	0005	0101	32
6	0006	0110	64
7	0007	0111	128
8	0008	1000	256
9	0009	1001	512
10	000A	1010	1024
11	000B	1011	2048
12	000C	1100	4096
13	000D	1101	8192
14	000E	1110	16384
15	000F	1111	32768

[Guest jcardana]

I hate to bother you so much, I don't know why I'm having such a hard time with this. You probably think I'm a moron. But, here's something else I'm confused about.

The Programmers Guide states...

Offset  Size             Use
0D0C    2       Lights (FS2k/CFS2), a switch for each one (bits from lo to hi):
                0 Navigation
                1 Beacon
                2 Landing
                3 Taxi
                4 Strobes
                5 Instruments
                6 Recognition
                7 Wing
                8 Logo
                9 Cabin

According to that, x0D0C should be a type 2 offset.

But, according to the GFDisplay manual...

S 8: Signed 8-bit byte                       1
U 8: Unsigned 8-bit byte                     2
S 16: Signed 16-bit word                     3
U 16: Unsigned 16-bit word                   4 (also 0)
S 32: Signed 32-bit double word              5
U 32: Unsigned 32-bit double word            6
F 32: Floating point 32-bit (“float”)        7
F 64: Floating point 64-bit (“double”)       8
STR: Character string (ASCII)*               9

a type 2 offset is a Unsigned 8-bit byte

But you wrote

For a 16-bit value, like offset 0D0C (the lights), there are 4 x 4 bits

Is this messing me up?

[Guest jcardana]

I doubt you'll read this far Mr. Dowson, but I wanted to say thanks. :wink:

Just a few minutes ago time I searched for "GFDISPLAY MASK" and found a post from last year. Enter the wrong keywords and get the wrong answers. GIGO. :roll:

It made more sense. You mentioned a calculator earlier... My calc on Win2k has a scientific mode which will convert Binary to Hex

To read the fifth bit... I entered 100000 into the calculator and hit the hex button, and got 20, so M0020 now shows me the status of my Instrument lights.

So thank you, thank you, thank you for your time and patience. I hope I didn't frustrate you too much.

I'll leave you alone for a while,

Joe

[Pete Dowson]

I was doing the powers of 2, i just did it wrong...

Powers of 2 apply to individual bits, according to their position in the complete number, the same as powers of 10 applying to decimal digits in a decimal number. For some reason your table lists powers of two for hexadecimal digits 0 - F. Why? One hexadecimal digit represents no powers of anything -- it's just one digit of any hex number, the same as 0-9 are digits in a decimal number!

You seem to be wanting to make something very complex and mystifying and completely wrong out of something basic and simple. Please take more time to consider how decimal numbers actually work. Then apply that to binary numbers, where you only have two digit values 0 and 1 instead of 10 (0-9). Once you've got that far, you can simply group 4 bits together to give you hexadecimal.

Pete

[Pete Dowson]

I entered 100000 into the calculator and hit the hex button, and got 20, so M0020 now shows me the status of my Instrument lights.

But why not just split the 100000 into groups of 4 bits:

0010 0000

which (as you can see from the first table I gave you) is 20.

That's the whole point! Each hex digit represents 4 binary bits, that's all!!

Pete

[Guest jcardana]

I think my expectations got in the way

This is all I was looking for...

0000 0000 0001 =0001

0000 0000 0010 =0002

0000 0000 0100 =0004

0000 0000 1000 =0008

0000 0001 0000 =0010

0000 0010 0000 =0020

0000 0100 0000 =0040

0000 1000 0000 =0080

0001 0000 0000 =0100

and I got my self all kinds of confused.

[Pete Dowson]

This is all I was looking for...

But that list is specific for 12 bits. The very first list I gave you is completely general for ANY size - 4, 8, 12, 16 .... 64 bit values, anything!

    Decimal  Hex  Binary
    0            0      0000
    1            1      0001
    2            2      0010
    3            3      0011
    4            4      0100
    5            5      0101
    6            6      0110
    7            7      0111
    8            8      1000
    9            9      1001
    10          A      1010
    11          B      1011

Just look up each group of 4 bits. It's so easy! I don't understand why you are still making it more complicated!

Anyway, I'm not answering any more here. I've run out of ways of trying to convince you to take the easiest ways.

Regards

Pete

[Guest jcardana]

I understand what your saying now. You're right, I made it far to difficult.

Thanks for your time.