- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report
Hello,
Since I've never had formal training with Inventor, I'm not always certain that I'm executing tasks in the best way to help out the software that has inherent performance limitations from time-to-time, perhaps from my hardware (touched on later).
Our company operates in the automotive performance parts world; we receive OEM automaker files so that we can better design our own performance parts. So far, I've seen the files provided in X_T, STEP, and igs formats primarily. Surface parts can still emanate from the X_T and STEP formats of course, and they typically have errors no matter what because they are complex parts like engine blocks, cylinder heads, or molded plastic parts for example.
I've been conducting the repairs on these parts, but I've wondered if some things could be done better/differently to make the software more stable/work faster. Side note: I also keep running into a memory truncation and request to expand issue, which I researched using other forum posts, but still cannot wrap my head around why it does not use RAM or the graphics card more appropriately, in general.
Getting on with my repair process, I enter the repair environment and then execute command, "Find Errors." That goes through the motions and then I'm presented the results in the part tree.
Now, I presume the category folder that it opens first and expands automatically is the correct place to start, yes? There was another category before Modeling Uncertainty that I already took care of in this image.
So as the next image shows, this is usually how I try to execute my repair process. I basically always try the repair option with a tolerance of .003 first, and then go to .004 (where a warning is triggered that I say OK to) but it's usually far too slow and doesn't work anyways.
The boundary patch isn't a bad option, but this is where I'd like some guidance.
Is it better to try and use the boundary patch repair option, OR simply delete the faulty surface and then add-in a new Boundary Patch Quilt like I've done in another location? Sometimes it seems like it doesn't necessarily turn out the same as before, but it does seem like it would be more stable since it's usually executed faster and thus requires less processing I'm thinking.
Also, what if I decided to tackle the Intersecting Faces Folder first? Is that not recommended? I just didn't find the exact answers I needed on the help pages, and once again, I'm trying to make sure I use the software right so that I can help it out for this large, fragile stuff. How significant is the certified/approved graphics card for surface modeling over standard modeling?
How in the world would I handle a file with multiple (50+) errors reliably? I haven't had immense luck with this yet, but I don't want to leave these errors in a file so that I can reduce the possibility of future workflow errors. My typical problem is software hang-up/timeout.
One cannot save in this environment either which is balls-to-the-wall insanity. I know this is an operation, but with this memory allocation issue it's just a recipe for disaster. I don't want to keep entering a repair environment because I need to save my file to keep it somewhat stable (maybe not even). Aside from the issue of that being convoluted, each repair operation adds more data to the part tree which doesn't seem practical for streamlined performance either since it's just more data.
Feel free to copy past my text and input your responses in the appropriate fields. Thanks!
- Dell XPS Tower
- Windows 11 Pro - 64-bit 22H2
- Intel Core i7-8700 @ 3.20 GHz
- 32 GB RAM
- NVIDIA GeForce GTX 1060 6GB
- Inventor 2022 Professional & Vault Basic installed on SSD
- Window Defender
Remote Machine Specs:
- HP Z2 Mini G4
- Windows 10 Pro - 64-bit
- Intel Core i9-9900 @ 3.10 GHz base
- 32 GB RAM
- NVIDIA Quadro P1000 4GB
- Inventor 2022 Professional & Vault Basic installed on M.2 SSD
- Kaspersky Internet Security Product Suite
Solved! Go to Solution.