Point Not Appearing in Surface

[rsebastiancoleman]

I'm creating a very simple surface defined by just four points in a PENZD file. No issues uploading, creating a point group, or adding that point group to the surface definition. However, the surface renders using only three of the four points (creating just one triangle) despite listing four points in its definition. Point 2022062101 is not included. Stranger still, when I add a dummy fifth point that is simply the average of all the others, the missing point shows up, but the surface still only creates three triangles when four would be expected. I have tried manually retyping the points into a new excel file, saving as both txt and csv to eliminate any strangling data from excel, neither has changed the rendering. 

The PENZD with both the four points and with a fifth point are attached. The attached drawing file has layers "2022-06-21" which includes just the four points and "2022-06-21 with dummy" which includes the fifth dummy point. Any assistance would be great. Thank you!

[jroot]

When I open the dwg I see a surface with 5 points:

[rsebastiancoleman]

Yes, the current drawing is using the additional fifth "dummy" point, which causes the M1 point to show up. However, when trying to define with only four points, M1 disappears. I'll clarify in the original post. 

[jroot]

Okay I see.  I removed the 2100 point and it made the surface with only the three points like you said.  I was able to add a line from the existing triangles to the point 2101 that wasn't included and then it worked. I don't know why it's throwing that one out.

[LucasHicks]

Have you seen this thread?

https://forums.autodesk.com/t5/civil-3d-forum/civil-excludes-csv-points-when-creating-surface/m-p/11...

[jroot]

I guess that maybe explains it, sortof.

In all my years of creating surfaces with point groups/cogo points I have never had this happen. It has never omitted outlier points no matter how small of a sliver it creates.

[rl_jackson · Accepted solution]

So, I took a look at this, and I'll agree with @jroot it's strange that it doesn't include all of your dataset similar to the linked post which I had responded to but the points in that case were at an inconceivable distance from the rest of the dataset. My take is this, its algorithmic deficiency in the TIN model. I tested it in 2020 for giggles and received the same results as 2023, and no longer have older versions on my system. My recommendation for this type of dataset which is very limited IMO would be to create a featureline through your data and added it to the surface as a breakline with addition points every 100'. I've attached my 2020 file as a sample which should give you the information you're looking for. As a side not stretching a surface at the distance your attempting has never been recommended, as it will always create undesirable results.

HTH

[rsebastiancoleman]

Thank you for your help, and apologies for taking a few days to respond. Did you have any trouble creating a line to 2101 within the surface? I attempted to draw one from both 2102 and 2013 but it stops giving me the option about a hundred feet from the point. 

[rsebastiancoleman]

Ok, thank you for the guidance on this. We'll reevaluate how we construction our surfaces.