DLP3010EVM-LC: unexpected patterns appears in internal patterns mode

Hello User,

The team will look into this behavior. To see if this is a GUI or FW related can you change the Pre/Exposure/Post times to 300/1500/200? Does this behavior occur when the times are lowered to 10s of ms?

Regards,

John

Hello,John

It seems to have nothing improved, only the period becomes shorter.I'm wondering if this is a common situation. if so, I would like to switch to a more stable product.

Hello user,

Can you please try to set the pre/exposure/post time as 300/2000/200 in us? 

Please refer to the sequence timings table for DLPC347x Light control platform from the [FAQ] here. It's also shown in the DLPC347x firmware release notes when you downloaded from TI.com.

The internal pattern mode only supports 1D patterns, you may want to consider an external source for 2D patterns use.

Regards,

Lori 

Hello TI,

Thank you for your advice. I find it more effective to display the sequence step by step when it runs continuously，since the patterns are difficult to recognize within such a short exposure time if it is played once. Is there a way to adjust the time using the Table in the link to make the exposure time longer without unexpected patterns appeared? (By the way, sometimes, when I click "Run Once," the GUI mistakenly recognizes it as "Continuously Run". It's not a major issue, but I recommend that you fix it to avoid confusion in the future).And another problem has appeared: the trigger in J11 of the main board does not work properly when I click "Step." It seems to be triggering continuously. How can I synchronize with camera？

Regarding the 2D patterns, I am unsure how to add the patterns to the GUI or Flash. Could you please provide more details or a demo?

Thank you for your patience.

Best regards,

Yu

Hello user,

The table shows the supported sequence timing, you may not see the internal pattern work properly beyond the supported exposure time, but you can experience on the EVM to increase the exposure time. What's the purpose to have a longer exposure time of 10+ seconds for your application? 

We have not noticed that there is issue of the GUI for "Run Once". Can you please describe what you see when you set "Run once"? Do you ever see the patterns stop displaying? 

How do you use the trigger in signal and capture it as continuously? Please make sure you have the jumper set up correctly for trigger in mode and enable Trigger in mode. Please refer to the diagram for the trigger in mode. It's in DLPC3478 datasheet.

For displaying 2D patterns in external pattern mode, we don't add the patterns to the GUI or write data into flash. The patterns are sent to the DLPC3478  controller over parallel interface and display them at runtime. You can use the HDMI port for external inputs. 

Regards,

Lori 

Hello TI engineers,

I am currently working on projecting 2D images. In order to achieve my desired projection results, I have decided to abandon the internal patterns  mode. However, I have encountered difficulties in solving phase information when using the external patterns  mode. This issue is likely caused by nonlinearity.

Hence, I would like to inquire about the preprocessing performed by the DLP system in the external patterns mode. Specifically, I am curious to know if the DLP system recognizes the external mode as an extended monitor and applies any preprocessing techniques such as gamma correction to the input images. If this preprocessing is indeed happening, I would like to know how I can disable or bypass it.

Additionally, I would like to discuss the exposure time. While a 10-second exposure is not mandatory, I prefer longer exposure times to enhance the contrast captured by the camera.

I greatly appreciate any insights or guidance you can provide regarding these matters. Thank you in advance for your assistance.

Best regards,

Yu

Hello User,

There is gamma correction. The team will have to look into available options.

The team will have to look into what exposure times are achievable for this platform.

Regards,

John

Hello TI Engineer,

I would like to inquire if it is possible to obtain a firmware version for the external mode that does not include gamma correction or any other preprocessing functionalities. This would be highly beneficial for my project's requirements. If this firmware version could also extend the exposure time in the internal mode, it would greatly enhance the capabilities of my project. Given the tight timeline of my project, I would appreciate it if you could provide an estimated timeframe for when I can expect a response with the updated firmware.

Additionally, I have experimented with a method of “Run Continuously” in the “Internal Patterns” by “pause” and projecting step by step to achieve sufficient exposure time. However,  the obtained phase still exhibits the aforementioned issues and is not usable. I hope that my findings can contribute to improving the performance of your product.

Thank you for your attention to this matter, and I look forward to your prompt response.

Regards,

Yu

Hello User,

Unfortunately, we are not able to support the custom firmware without gamma correction and extend the exposure time neither in internal pattern mode nor external pattern mode. I would recommend you to follow the supported exposure time that I mentioned before. 

We have not noticed that there is issue when displaying the patterns with "Run continuously". You need to follow the supported pre-exposure/exposure/post-exposure timing as well. 

Regards,

Lori 

Hi TI Engineer,

Thank you for your clarification. So, does this mean that I cannot project 2D images without gamma correction? However, Engineer John mentioned yesterday that the team will have to look into available options about gamma correction.

regards,
Yu

Hello User,

I apologize for the confusion. Please allow me to double check if there is gemma correction in external pattern mode.

Regards,

Lori 

Dear TI Engineer,

Thank you for your suggestion. It appears that the light control mode indeed does not have gamma correction. I have conducted tests on the R, G, B, and RGB options within the light control mode, as well as the "display"-"video and color"-"switch to external video" (referred to as "colorful" in my analysis) for the 256 grayscale responses.

During my tests, I have observed non-monotonic and discontinuous behaviors（ I have  marked them with yellow ellipses and boxes） in the external mode, which affects the linear relationship between input and output. Furthermore, these non-monotonic and discontinuous phenomena seem to occur at specific grayscale levels. I have documented my test results and the exposure time settings for your reference.

I would appreciate your guidance on whether this issue is related to the settings or if there are any specific methods or recommendations from your testing procedures that I can consider. In the attached video, there are also visible fluctuations in brightness .

Thank you once again for your assistance.

Best regards,

Yu

input grayscale.zip

Hello Yu,

Thank you for this information. The team will have to look into this behavior. Thank you for your patience.

Regards,

John

Hello TI Engineer,

I hope this message finds you well. I wanted to follow up on the investigation progress regarding the issues mentioned earlier. I'm particularly interested in knowing if there are any suggestions or recommendations regarding the configuration settings that could potentially address the observed non-monotonic and discontinuous behavior in the external mode.

Furthermore, if this phenomenon is specific to my device, I would appreciate exploring the possibility of a replacement.

Thank you for your attention to this matter, and I look forward to your response.

Best regards,

Yu

Hello Yu,

Our team is actively investigating this issue, though we do not yet have a conclusive solution. The current task is to recreate the behavior that you are observing.

In your most recent video with the red image, what values are being input when seeing the jump in brightness?

Regards,

Austin

Hello TI Engineer,

Thank you for your response. In the provided video, the red image sequence corresponds to the initial section of the red line in the figure（from 0 to around 20）. It represents a grayscale test starting from 0, with a resolution of 1280*720. Each frame in the video corresponds to a grayscale level increment of 1.I have uploaded the grayscale images in a compressed format in my previous post where I initially reported the phenomenon.

Considering this setup, I expected the brightness variation to be monotonically increasing. However, the observed fluctuations in brightness seem to deviate from this expected behavior.

Best regards,

Yu

Hello Yu,

Thank you for the details. This will help with further debug.

Regards,

John

Dear TI Engineer,

I hope this message finds you well. I wanted to kindly remind you that it has been a month since I purchased the product. Given the pressing nature of my project, I would greatly appreciate receiving timely debugging results.

To facilitate the troubleshooting process, would it be possible for you to provide me with some test images and the corresponding settings that demonstrate the expected linear response?

Additionally, if it proves difficult for the engineers to replicate the described abnormal behavior on their end, would it be feasible for me to request a replacement device that functions properly?

Regards,

Yu

Hello Yu,

We are actively working on your question within our team. Can you please allow us more time to analyze this question? We may noticed that there could be a brightness difference between two steps, but we will need to investigate more and check if this is normal operation. We will get back to you by the end of next week. Thank you for your patience. 

Regards,

Lori 

Hello TI Engineer,

I am writing to inquire about the progress regarding the analysis of the issue I raised earlier regarding the non-monotonic and discontinuous behavior in the projected images.

I would greatly appreciate any updates or information you can provide regarding the investigation. Additionally, if there are any specific testing procedures, patterns, or configuration settings that could help me achieve linear response, please share them with me.

Thank you for your attention to this matter, and I look forward to your response.

Regards,

Yu

Hello Yu,

I'm sorry for the late response. We are still investigating this behavior internally with our team since we have not aware this issue before. I will update you when we have more information. 

Thank you for your patience!

Regards,

Lori 

Hello TI,

I wanted to inquire whether your team has encountered a similar issue with the device you are using. If you have experienced the same problem, I would like to request a return of the product and explore the possibility of purchasing a different model that does not exhibit this behavior.

On the other hand, if your team has not encountered this issue with the equipment, I kindly request a replacement of my current device instead of waiting for a resolution. Given the tight project timeline, it would be beneficial to have a functioning device as soon as possible.

Thank you for your attention to this matter, and I appreciate your prompt response.

Best regards,
Yu

Hello TI,

Thank you for your suggestion regarding warranty. I would like to inquire if you have encountered the same issue depicted in the video. This information would greatly assist me in determining whether I should consider alternative product options or simply request a replacement.

Additionally, if your equipment does not exhibit such anomalies, would it be possible for you to provide me with detailed configuration settings that you have used? This information would help me ensure that I am using the device optimally.

Thank you once again for your support.

Regards,

Yu

Dear TI engineer,

Thank you for providing the detailed information. Could you please offer some recommendations for EVM products with optical path that can project 2D images with a square pixel arrangement and linear response?  These characteristics are really necessary for me. 

Best regards,

Yu