Welcome to the Onshape forum! Ask questions and join in the discussions about everything Onshape.
First time visiting? Here are some places to start:- Looking for a certain topic? Check out the categories filter or use Search (upper right).
- Need support? Ask a question to our Community Support category.
- Please submit support tickets for bugs but you can request improvements in the Product Feedback category.
- Be respectful, on topic and if you see a problem, Flag it.
If you would like to contact our Community Manager personally, feel free to send a private message or an email.
InStep STL import alternative
apearce
Member Posts: 5 ✭
InStep is a common required step for a lot of people looking to work with OnShape.
Without conversion I have to redraw my models. This is something I'm looking into - to generate views from different angles and use this to generate a true nurbs model. But this is time consuming so I'm still interested in 'conversion' if it can speed things up.
Some observations.
There's a number of problems I've had with InStep. The first is that it's not a cloud app in any way. It's a desktop app which just has a connector to OnShape for upload and download, reducing just a few clicks. Since conversion is quite processor intensive even a model of just 10,000 triangles takes a long time on my system. I've had the process crash at all 3 stages - import, error correction and export - usually with the 'failed to generate new window'. Contacting solveering is only possible through a contact form designed for commercial consultancy jobs - nothing to do with software. The company doesn't seem to really be a software company and I get the impression they are not really interested in the software they have created - that they simply made it for their own need and decided to sell it 'on the side'. The forum http://instep.solveering.com/forum/ as of now is down.
When I convert from .obj to .stp the resulting model still comes up blank in onshape. Same for conversion to .x3d. It also won't load in a .stp I've just converted and corrected... as 'file does not appear to be valid'... so I go through the import and correction again to try different formats (this seems clearer to me than direct export to onshape as I can see what the software is doing and inspect the converted model).
The pieced facet correction sounds interesting and useful because it picked up a lot of errors that meshmixer inspector had not. I have 37k features and 57,000 edges. Possibly the reason why I get these errors more than other people have is because the model I'm trying to import to OnShape is a surfboard which is nothing but curves.
I appreciate that going from a mesh model to full 3D is a bit like text OCR in scanning documents but somehow I have to get a model in. Assisted conversion with tracing might be a better approach.
A new app seems important to see. Probably the reason why it's not a very popular area for developers is that it's a very difficult thing to do but users just expect it to work because of their ignorance as to what is involved. A better approach can be an assisted tracing method whereby the process has an element of automation but the user is making choices as to how the solid body is created.
Without conversion I have to redraw my models. This is something I'm looking into - to generate views from different angles and use this to generate a true nurbs model. But this is time consuming so I'm still interested in 'conversion' if it can speed things up.
Some observations.
There's a number of problems I've had with InStep. The first is that it's not a cloud app in any way. It's a desktop app which just has a connector to OnShape for upload and download, reducing just a few clicks. Since conversion is quite processor intensive even a model of just 10,000 triangles takes a long time on my system. I've had the process crash at all 3 stages - import, error correction and export - usually with the 'failed to generate new window'. Contacting solveering is only possible through a contact form designed for commercial consultancy jobs - nothing to do with software. The company doesn't seem to really be a software company and I get the impression they are not really interested in the software they have created - that they simply made it for their own need and decided to sell it 'on the side'. The forum http://instep.solveering.com/forum/ as of now is down.
When I convert from .obj to .stp the resulting model still comes up blank in onshape. Same for conversion to .x3d. It also won't load in a .stp I've just converted and corrected... as 'file does not appear to be valid'... so I go through the import and correction again to try different formats (this seems clearer to me than direct export to onshape as I can see what the software is doing and inspect the converted model).
The pieced facet correction sounds interesting and useful because it picked up a lot of errors that meshmixer inspector had not. I have 37k features and 57,000 edges. Possibly the reason why I get these errors more than other people have is because the model I'm trying to import to OnShape is a surfboard which is nothing but curves.
I appreciate that going from a mesh model to full 3D is a bit like text OCR in scanning documents but somehow I have to get a model in. Assisted conversion with tracing might be a better approach.
A new app seems important to see. Probably the reason why it's not a very popular area for developers is that it's a very difficult thing to do but users just expect it to work because of their ignorance as to what is involved. A better approach can be an assisted tracing method whereby the process has an element of automation but the user is making choices as to how the solid body is created.
0
Comments
I'm going to go ahead and comment on a few of the items you raised, even if a bit late.
The application is not a cloud app - simply because that is difficult to do and forces the cost to be higher. The desktop model allows all the CPU resources to be your own, the cloud app requires this to be done on something like AWS which would be an additional cost to the user.
Conversion of a 10k facet model should be very fast, on the order of a minute start to finish. The part that will take long is the check for 'pierced facets' basically checking that there are no overlaps, intersections or similar. If your data is good, then you can bypass this step.
The application really should not crash unless you are forcing it to do something unnatural -loading files that do not follow the format's specs, loading files that exceed your computer's capabilities, etc. The same goes for cases where users think that they can use a Mac to do a PC's work... if it says Windows compatible, it should be run on a Windows machine (XP SP3 or later with Instep 2.x). "Failed to generate Window" does not sound like an application error...
Contacting Solveering is done -yes- through the contact form which is for any form of contact, not just consulting work. Alternatively, you can report an issue directly from within the application's help menu or you can use the support email (support@solveering.com). Not sure what prompted the comment about a lack of interest in the product...
The forum is down, yes, and will likely stay that way. There has been a flood of spam and a trickle of user interest.
The comment about the file being 'nothing but curves' is an indicator as to what might be going wrong. Data needs to be defined in a facet format... otherwise how does the software know where surfaces are? Garbage in, garbage out very much applies here and it should not come as a surprise if there is no data available if you didn't load anything that contains data. Though the conversion of point and line data is something that is being actively pursued, the current version is simply not set up to handle that.
I appreciate your detailed and candid feedback. Please contact me directly if you would like to further discuss this.
Ben