Thanks a lot Lagom! amazing resource!! nice little manual to remember and learn new things. BTW that corner blended is what I wanted to archive, looking forward to try out with your way .
I find Rhinoās surfacing tools a bit overwhelming, there are so many ways to model surfaces, and several tools seem to do similar things but with different results. For example, when creating a surface between two curves, you could use EdgeSrf, Loft, or Surface from 2, 3, or 4 edges. But which one should you actually use to achieve the best quality? Everyone seems to do it differently, with varying results.
You understood what I meant about that corner blend and its surface quality. It would be really helpful to have a proper resource showing how to model and solve common surfacing cases, like corner blends, Y-fillets, and multi-blends, with a high-quality approach, similar to what you did here. Iāve seen examples online, but many donāt meet the level of quality Iām aiming for. I can see itās possible to achieve, but as Skysurfer mentioned, it requires multiple steps and specific tools, which makes the process trickier, and also there are not enough quality resources to learn this online like exits with other softwares.
From what I understand, this new Patch works like a Swiss knife, it handles most cases for most users, and can do the work of many surfacing tools already in rhino, and is amazing. But what about the other tools? I feel Iām lacking a bit of structure on the approach, or more info regarding the tools or maybe itās just me IDK
Hence my proposal for a limited preview consisting no more than 20 control points in either direction. Or, alternatively, set a minimum distance between control points (or a percentage of the entire patch surface, such like 5%). That will eliminate situations where the āPatchā tool tries to add a huge amount of control points in a small area.
Thanks . now I see that Patch isnāt meant for Class-A surfacing, but itās still exciting to see how itās evolving.
I will definitely play around more and post some examples when I see something interesting. Really looking forward to seeing whatās coming with the new surfacing tools!
Make sure to add an optional limit of the minimum distance between adjacent control points. Or minimum percentage (2-5%) of the entire length of the surface). That will prevent the tool generating a surface where 5 rows of control points are located super close to each other.
If you remember, several months ago we had a similar discussion about the āBlend surfaceā and āMatch surfaceā, as well. They both tend to overly-build the surface, too.
I was not able to get to an appropriate result like the old patch could for this surface, and the new Patch just made a mess, but maybe I didnāt provided the right input to it:
What I really was looking for was the result I got with NetworkSrf, BlendSrf getting close too, and no patch variant got me this result. The issue is that with this technique I have to split the edge to make the NetworkSrf or the BlendEdge work, losing history:
Worked well! Faster than NetworkSrf. The new patch seems so powerful, it really needs to be deeply documented in the help file, or at least a video tutorial explaining what its possible when R9 launches. Thanks Gijs!
(Other than that, I could have used SubD like Iām used to, but this specific file is a shared workflow with a few dependencies that cannot support SubD at the moment.)
Imho, as a tool in the development state, I find it easy to find all info about the new patch in a single topic, rather than to hunt around for several threads.
I have an expansive Grasshopper sketch that started with the legacy Patch command and extensive surface point rebuilding. Shifted to XNurbs, quirky but productive. After a few false starts (the paid XNurbs $500, it aināt broke, donāt fix it factor) with the ContinuousPatch, commited this month to integrating as test and quickly realized fully capable of switching over.
I am blown away. Incredible for my application. It gives consistent solutions where XNurbs failed and the granular control is very useful, loving the exploration. Has opened a lot more conditions of control.
Observation, in my application, the flexibility input has smaller variations through 0.1 to 0.7 to 0.9 and large effects from 0.994 > 0.995 > 0.998, natural maximum at 1.0
I can deal with this given a high resolution input, it just seems unusual.
Question: Can internal points be given weights? Either a single value 0.0-1.0 or a list according to the points.
I am using 100% the GH ContinuousPatch component, all day, everyday.
Just wanting to know if it has all of the capabilities incorporated and the same algorithms that are (seem to be) being tuned with the Rhino command. Feels like more than the dialog is being worked on since my incorporation of the GH component.
To make it more apparent the depth of my interest, here is my workspace in progress.
The first screen below is my front end for the ContinuousPatch in GH, elements to be toggled in and out of the solution on the left side. Elements to be displayed lower left. Upper panel for the current Patch settings quick toggles.
Middle section for control of the input vectors, with the high resolution tuning legend for a physical dial box on the lower left.
Hi, is this just my problem or are others seeing it too?
In the latest Rhino WIP version I no longer see the visible icons for G0, G1, G2 and I during the command activation.
I canāt select the edges or set the values anymore.
.
