DLPC900: Question in pattern display LUT definition

Hi Fizix,

I would like to check if I understood you correctly. Do you mean to say that the images will be packed into blocks of 24 bit frame and the index will depend on which block the pattern is? So if I have 25 1bit images, the first 24 images will have image pattern index of 0 while the 25th image have an index of 1? Also, is this packing determined when I add the images into the flash memory or based on how I add them before 'Pattern display LUT definition'? e.g. I added 25 1bit images to the flash memory and when I program the pattern display, I only use two images, 1st and 25th image. In this case, is the index of the second image 1 or 0? Similarly for image bit depth, can I say add a 8bit image into flash and when I call it I use 1bit? If this is possible, how do we calculate the index? Is it calculated based on 8bit or 1bit?

Sorry, my mistake. My question should be what is "Number of patterns in sequence" and why is it 0?

Jiayun,

Fizix is out-of-office this week. I am replying on his behalf.

"Do you mean to say that the images will be packed into blocks of 24 bit frame and the index will depend on which block the pattern is? So if I have 25 1bit images, the first 24 images will have image pattern index of 0 while the 25th image have an index of 1? "

Yes. This is correct. 

Say I am adding 26 1-bit patterns. Let's call them A to Z:

Image A will have Pattern Index = 0, Image Pattern Index=0, Bit position in image pattern=0

Image B will have Pattern Index = 1, Image Pattern Index=0, Bit position in image pattern=1

.

.

.

Image X will have Pattern Index = 23, Image Pattern Index=0, Bit position in image pattern=23

Image Y will have Pattern Index = 24, Image Pattern Index=1, Bit position in image pattern=0

Image Z will have Pattern Index = 25, Image Pattern Index=1, Bit position in image pattern=1

The remaining 22-bit planes are filled with zeroes(a black image).

Say I am adding 26 8-bit patterns. Again A-Z:

Image A will have Pattern Index = 0, Image Pattern Index=0, Bit position in image pattern=0

Image B will have Pattern Index = 1, Image Pattern Index=0, Bit position in image pattern=8

Image C will have Pattern Index = 2, Image Pattern Index=0, Bit position in image pattern=16

Image D will have Pattern Index = 3, Image Pattern Index=1, Bit position in image pattern=0

.

.

.

Image X will have Pattern Index = 23, Image Pattern Index=7, Bit position in image pattern=16

Image Y will have Pattern Index = 24, Image Pattern Index=8, Bit position in image pattern=0

Image Z will have Pattern Index = 25, Image Pattern Index=8, Bit position in image pattern=8

The remaining 8-bitplanes are filled with zeroes.

"Also, is this packing determined when I add the images into the flash memory or based on how I add them before 'Pattern display LUT definition'? "

This is determined when you add the images to FW. This is how it is stored in the external flash. Note that only the image data is stored in the flash. Not the 'pattern display LUT definition' like exposure time, etc. This is information is taken when you 'indicate' the C900 to display an image.

Say you have added 8-bit image to FW. Now, when you are indicating the images to display, you say 1-bit image. Then it displays 1st bit-plane of the first 8-bit image.

Number of LUT entries is the total number of patterns you are adding, can be a maximum of 400.

"Ok I believe you misunderstood what I am doing and what the problem is. I am trying to verify if I correctly added images into the DMD flash. To do that, I operate in "Video mode" under Operating mode and I choose "Images from Flash" under Source Select. I then choose Index 0 and click Set. I see a blurry image which is nothing like what I see when I operate in "Pre-stored Pattern" mode."

I am not able to reply on the other thread, answering it here.
You say you are adding 72 1-bit images with 120 ms exposure time.

So, when you select Image from flash, 0th index, it displays the first 24 1-bit images together. This is why you see the blurred image.

I hope this is helps.

Thank you,

Regards,
Aishwarya

Jiayun, 

"Also, what does 'Number of LUT entries'? Why is it 0 in this case?"
I assume you are asking about Command 0x1A31? Number of LUT entries is the total number of images that you want to display.

"My question should be what is "Number of patterns in sequence" and why is it 0?"
May I ask where are you referring this from?

Thank you,

Aishwarya

I am referring to the 1A31 command. The data is 02 00 00 00 00 00 which is 00 02 00 00 00 00 in MSB LSB form. According to Table 2-133, the first 2 bytes are the number of LUT entries which in this case is 2 hence 00 02. The next 3 bytes are "Number of patterns in the sequence" which we can see is 0. So my question is what does "Number of patterns in the sequence" refer to

Hello Jiayun,

This description may be incorrect.  I am sorry I misunderstood your question. Let us look at this and get back to you after the weekend.

Fizix

Hello again Jiayun,

We are still researching this.  Sorry for the delay.  Please allow another day or two.

Fizix

Jiayun,

You are very welcome!

Fizix