Hi Ruochun (accidentally sent it to you and not the forum here is the one
to the forum),

Thanks for clarifying. I think I understand the approach you’re suggesting
now. Instead of loading all the CSV files at once—which would consume an
enormous amount of memory—I should:

   1.

   Set up a preset Blender scene that contains the environment, lighting,
   and any globally-applied materials or textures right from the start.
   2.

   Use a Python script to handle the data one CSV at a time. For each CSV:
   - Load the data into the scene.
      - Apply the appropriate modifications, such as adjusting object
      positions, textures, or particle systems based on that CSV’s contents.
      - Render the resulting frame.
   3.

   Once the frame is rendered, clear the scene of that CSV’s data, then
   move on to the next CSV and repeat the process.

In the end, I’ll have a series of rendered frames, each generated from a
single CSV to keep memory usage down. Then I can use a tool like FFmpeg to
compile all the individual frames into a final animation.

Does that workflow align with what you had in mind?

thanks for your continued support

On Thu, Dec 12, 2024 at 9:45 PM Ruochun Zhang <[email protected]> wrote:

> Hi Liam,
>
> My question is, if you have around 250MB per CSV, why is 40GB of CSV data
> loaded into Blender? What we usually do is load the data for one scene and
> render a picture at a time. In the end, we combine all pictures as frames
> into a movie using FFmpeg instead of trying to load all data at once.
>
> Thank you,
> Ruochun
>
> On Thursday, December 12, 2024 at 10:31:57 AM UTC+8 [email protected]
> wrote:
>
>> Good afternoon, Community!
>>
>> I hope you're all doing well. I have a question regarding post-processing
>> with the DEM Engine in Blender, inspired by how some folks handle fluid
>> simulations using CSV files. I'm trying to create a high-quality animation
>> by importing a large volume of CSVs and VTKs, around 250 MB per csv.
>>
>> Would this require a co-simulation add-on from regular Chrono, or is
>> there a more efficient approach? For example, could I use ParaView to
>> export the data as .ply or .usd files and then import them into Blender?
>>
>> I also experimented with a Python script in Blender to import the CSVs by
>> reading their columns (x, y, z, r, absv), but Blender crashed under the
>> load of 40 GB of CSV data—unsurprisingly, that's quite the chunk of data to
>> handle.
>>
>> If anyone has insights, advice, or experience with this kind of workflow,
>> I'd really appreciate your input!
>>
>> Thanks,
>> Liam
>>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "ProjectChrono" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/projectchrono/eJNw4ZcF6f4/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> [email protected].
> To view this discussion visit
> https://groups.google.com/d/msgid/projectchrono/6d8b3d4e-99af-46d1-9a14-60e477697547n%40googlegroups.com
> <https://groups.google.com/d/msgid/projectchrono/6d8b3d4e-99af-46d1-9a14-60e477697547n%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>


-- 
Best,

Liam Murray
*Emerging Spaceman and swimmer*

-- 
You received this message because you are subscribed to the Google Groups 
"ProjectChrono" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/d/msgid/projectchrono/CADSLfQzfWEd4pRD3R0p2WNv-ehfHO871CANVWY1Oi%2BrFQO-2kw%40mail.gmail.com.

Reply via email to