I have a few questions:

#1 (parts 3 + 4):

If you have 16-bit color, doesn't that mean that each pixel is represented by 16 binary digits? What would the number of bytes required to represent a certain # of pixels depend on whether the 16 bit color works like 8 bit color or 24 bit color?

#2 I can't figure out how the Manchester encoding works if the increments are not specified (as they are on the lecture slide). When you have a long flat line there is no up/down signifying an electric current. I am so confused!

#3 Are we suppose to be plugging in actual variables? If so I have no clue what these are or where to find out. Is there a text I could use to get a more detailed explanation of the compilation of sine waves because I have gone over the lecture notes but I am still totally confused. I don't understand what "r" represents.

Jen Dolloff

I just posted an answer to a previous question that I'm hoping may help with the 16-bit color problem. If not, let me know.

As for Manchester encoding, the idea behind it is that you can't have long flat lines where "long" is defined to be longer than the time to transmit two bits. This is because each bit's transmission involves a voltage change. So, if you look at where the voltage changes occur you should be able to tell where the bits are without the extra lines I included in lecture to guide you.

Finally, you are actually supposed to plug in 1 and 2 for "r" and see what happens. I'm hoping you can then generalize to see what impact increasing the value plugged in for r would have. You don't really have to worry about exactly how the sine wave behaves. The only relevant property of the sine wave is that its values move smoothly from -1 up to +1 and then back down again.

Professor Murtagh:
In problem 2, part #1, I still don't get the idea behind Manchester encoding. If I only note the changes in voltage, then the answer would ALWAYS be 1010101010101, etc. I read your last response, and I still don't understand how one handles voltages lasting for longer lengths of time. Also, could you explain the difference to me to question #1, parts 3 & 4? I think my head is going to explode. -Matt

In Manchester encoding, every bit's representation includes a change in voltage. The bit represented, however, is determined by the direction of the change. A rise in voltage represents a 1. A drop in voltage represents a 0. To send two 1's in a row, the voltage musts rise twice. It can't do this without dropping in the middle. The drop in the middle, however, does not represent a 0. For example, if you look at the diagram of manchester encoding I showed in class, you will notice there is a drop in voltage after every "1" bit in the long sequence of 1's.


To tell which changes represent digits and which don't you have to figure out the length of time (width on paper) of each bit. You can do this because each change in voltage must be preceded and followed by a period of fixed voltage whose duration equals half the time to send a bit.

In the image questions, the big difference I'm asking you to look at is that if 16 bit color functioned like 8-bit color it would depend on a palette for each image. If 16 bit color worked like 24 bit color, it would not use a palette.

Tom

"I think my head is going to explode. -Matt"

I think mine already did.

I'm not sure how to interpret the longer lines on
the Manchester coding, either. It does seem like
the answer should always be 101010etc. What does
the longer line significy in Manchester coding
signify?

And I don't even want to look at #3 now...my body
will probably explode, too, if i did that.

-andy

Look at the slide I linked to in answer to Matt's question. Think about what the Manchester encoding for "10" looks like. Because the voltage rises for the 1 and then doesn't fall until the middle of the "0" you get a "long" line.

#2 is much clearer after studying the diagram on
your lecture notes. Thanks for the help