This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
DLPC3438 Write to Output delay
Part Number: DLP3010
we see that there is one frame latency for the controller IC DLPC3433, that is 16.6 ms for 60Hz frame rate.
what about the DLP3010 ? what's the latency for the time:
between DLPC3433 starts sending the first pixel data of one frame
and DLP3010 completes displaying the last pixel ?
Allow me to look into this a bit. However, I believe there is no fixed latency and the DMD essentially displays the information as soon as it's received.
We are glad that we were able to resolve this issue, and will now proceed to close this thread.
If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.
In reply to Kyle Rakos:
After looking into this I have the following information. As soon as the controller (with its one frame latency) begins sending video data to the DMD it will be displayed within a few hundred microseconds. However, since the DMD displays data time sequentially (i.e. different biplanes are shown during the duration of the frame being displayed), the DMD will continue receiving and displaying data from the frame throughout the frame duration.
Therefore, while the data begins being displayed with nearly no delay, taking into account the time sequential display, the latency could be considered about half a frame (for our standard display sequences).
thanks for your explanation, I want to ask about different conditions.
if the frame rate is 60hz and the colour rate is 60hz, the bitplanes are shown within 5.55ms for each colour. it takes 16.6 ms to display 3 colours.
if the frame rate is 60hz and the colour rate is 120hz, the biplanes are shown within 2.77ms for each colour. it takes 8.3 ms to display 3 colours.
So it is thought this latency might be decreased by boosting the colour rate.
is that right ?
In reply to kai:
The duty cycle of each color is generally not uniform (usually green is ~45%, red ~40%, and blue ~15%). Additionally, you will not generally have multiple color cycles per frame. Therefore, your first description of the color rate matching the frame rate generally doesn't occur. Your second description is more accurate. However, the color rate (in other words number of color cycles per frame) is optimized by our sequences to be as high as possible.
Based on the second description, the color rate is 120Hz with 2 color cycles per frame. One color cycle is 8.33 ms (green 3.75ms, red 3.33 ms, blue 1.25 ms). All the frame data ( bit planes planes) have been shown on the DMD within this 8.33 ms.
Is it correct to say that the latency is 8.33 ms?
if the color rate is 240Hz, one color cycle could be 4.17 ms. so that this latency decreases. right ?
The above numbers were just given as a typical example. The exact duty cycles and number of color cycles will vary. Additionally, less color cycles may be used as frame rate increases. We try to optimize these sequences depending on the application and use case. Unfortunately I am unable to give an exact latency answer due to all of these variables.
If you are able to describe the concern and use case perhaps I could provide some suggestions.
we are using 240Hz (4 color cycles per frame) . That is |RGB|RGB|RGB|RGB| in 16.6 ms ( one color cycle for 4.15ms).
The image data is 24-bit RGB. So each color time is divided into 8 bit-planes(8 Red bit-planes, 8 Green bit-planes, 8 Blue bit planes) . right?
The time of 8 bit-planes of one color is shown in one color cycle |RGB| . right? That means 4 times of 8 bit-planes are shown in one frame.
what is the format of the frame buffer ?
If the image data is R(1000 0000) G (1100 0000) B (1110 0000) for all pixels.
The bit data transmitted by sub-LVDS is 1280*720 of 1->0->0->0->0->0->0->0 -> 1 -> 1-> 0-> 0 -> 0 ->0 ->0 ->0 ->1 ->1 ->1 ->0 ->0 ->0 ->0 ->0.
All these bit data will be used in one color cycle. right ?
1. I'm surprised you are able to get 4 color cycles per frame for each color at 240 Hz. In either case the 24-bit input bitplanes are not linearly mapped to 24-bit output colors on the DMD. Additional processing may be done to use even more bitplanes.
2. The controller will handle the processing of the video data and format it into a specific format for the DMD. We do not share the specific operation of the subLVDS bus.
what's the highest color cycles per frame for DLPC3433+DLP3010 ?
It would depend on the frame rate and minimum duty cycle desired. Please feel free to reach out to me via email for more detailed information.
All content and materials on this site are provided "as is". TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with regard to these materials, including but not limited to all implied warranties and conditions of merchantability, fitness for a particular purpose, title and non-infringement of any third party intellectual property right. No license, either express or implied, by estoppel or otherwise, is granted by TI. Use of the information on this site may require a license from a third party, or a license from TI.
TI is a global semiconductor design and manufacturing company. Innovate with 100,000+ analog ICs andembedded processors, along with software, tools and the industry’s largest sales/support staff.