Ugh… I don’t know if McNeel knows how big of a problem this whole toolbar/workspace crisis is. Let me explain why I think so:
I’m not even touching V8 until I can have my tools in there.
there’s no reason for me to use default V8, or any default Rhino because it would slow me down sooooo much. I think I would model faster in my custom V4 than on default V8. So I launch V8 WIP for 5-10 minutes at a time to check a new tool, but that’s about it. SO I’m not seeing the rough edges, which based on the trajectory of McNeel development, I’m sure there are tons of them.
So if they plannd to strategically leave power/advanced users out of the WIP feedback cycle, McNeel is doing an excellent job because that’s what’s happening right now IMO.
But if this is not what they want, I think their prioritization is all wrong, and all development should be reprioritized to make something usable for customization ASAP. Not because of the 1% of users who want to customize their tools is a relevant market, but rather because they are shunning most of us, and they will only get feedback by newbies, or advanced users trying things thing out in a playground-like setting: basic geometry for a quick test, not the real/complex/working/production files we all use.
I haven’t used Icem myself, so I don’t really have a qualified opinion on it…it seems to be used a lot amongst the European OEM’s and most people i talk to seem to have a positive experience.
I’m generally not a fan of Alias especially over the last several years, mostly because of stability issues, which in my opinion is the most important factor for this kind of work. I’m willing to deal with shortcomings in workflows as long as i can produce the same results at the end of the day. But Autodesk has ben shipping their yearly updates with tools that worked in previous version, but no longer do, or tools at are numerically inaccurate like absolute crown delta in the skin tool in 2023. Besides that Autodesk has done very little to update the core of the Alias base code since they acquired it and that present a whole other host of issues.
Most of our studio math gets sent to sculpting teams dedicated to modeling for production and that is pretty much done in NX for us…it seem you may have a similar situation where you work?
I agree with you on subd but I think it would improve to eventfully be more useful closer to production surface given time. Its great for quickly hammering out concept development though, where the emphasis is on quick surface dev and change management
Yeah, but BlendSrf history replay has a 75% fail rate. If the input srf edges get any kind of re-parameterization, the blend becomes an unfixable nightmare.
So far I’ve had very few issues with the UI. What Bobi is experiencing could he related to some variable specific to his system.
Just out of curiosity ,do you have some example of your custom tools? So far I really like the new macro editor and more importantly the Macro Library in version 8.
I think the strategy of moving some of the command options into dialog boxes and out of the command prompt has been an improvement like with the new filleting tools. It’s still in WIP so there are some kinks to be ironed out but I think its an overall good strategy, especially for highly interactive commands.
I think added support for ETO forms will go a long way in improving customization as well.
I use a 4K resolution (3840x2160 pixels) with 200% scaling from the Windows 10 settings. Rhino 7 works wonderfully well that way. Not sure why Rhino 8 WIP’s UI has so many weird bugs…
Yeah I’ve seen some of the bugs you’ve posted. I’ve had a few issue, but not as many as you have. My biggest interface issues with version 8 at the moment are related to camera rotation and navigation.
Sometimes what rotating around the model the “depth” picks up something in the distance behind the area of interest or just a random point out in space that throws off my camera target and center of rotation.
I don’t use a spaceball much at all. I usually use CTRL+SHIFT while rotating to sort of set my rotation point to whatever is closest to the mouse…this works almost perfectly in v7 but is buggy in v8. I also use CTRL+SHIFT with a quick press to snap into a plan view in the perspective viewport and then use CTRL+SHIFT again to break out of it…(I do this A LOT, lol) and in version 8 when I break out of plan view the camera suddenly gets zoomed out EXTREMELY far away from the origin. This behavior was much more stable and accurate in version 7
This also happens in Rhino 7 sometimes. I’m lucky enough to use a 3d mouse which has an almost perfect automatic camera rotation center based on the closest point to the center of the viewport, so I don’t deal with the inconsistent behaviour of the camera when it’s rotated by the right mouse button. On rare occasions the camera target goes somewhere in the distance, but that usually happens when I look at some object with a hole and there is a distant object visible through the hole of the nearby object.
They fixed a specific corner case coming from _ExtendSrf. The failure rate improved from 80 → 75%. History replay has been utterly useless since it was released in V6. It’s actually a regression from V5, because the history replay often leaves a srf in an unfixable state.
Same. I had to skip V6, and it’s looking like I’ll want to skip V8. Still, I think I’ll upgrade all of my licenses. Rhino is the cheapest software I own per hour of usage, and I want McNeel to have the resources to build better tools.
On the other hand, I could care less about SudD or whether it works on a Mac.
My job is to follow the stylistic development of each component in my company (automotive) and Icem and Alias are the main part for class-a modeling.
A large company has a very granular division of labor and there are few factotums or those who design, model, follow up on feasibility with suppliers and engineers.
Tools such as subdivision surfaces are used to make drafts for the research phase and never exceed 10% of the modeling time and all the designer use Blender because is absolutely complete, the remaining 90% must be done with classic surfaces in order to keep under control the thousands of constraints that need to be satisfied.
With subdivision surfaces it is mathematically impossible to bring into production a model that needs to interface with other aesthetic and functional pieces with our tolerance (0,001mm positional and 0,01 degree continuity, etc…) the only exception for the subdivision surfaces are the seat foams where, all in all, the soft behavior of the material allows for more permissive gap and flush management.
For example, a steering wheel is the classic object that some non-professionals would think of making with subdivision surfaces since it seems to have an organic shape… nothing more wrong: every single cent of the steering wheel must be under absolute control since it must be satisfied the visibility of the cluster to the tenth of a millimeter, to the pixel, the visibility of the stalks pictograms, the ergonomic sections, the safety requirements for the bag, the feasibility of molding, the wrapping of the leather panels, the capacitive and heated foils, the assembly of the bezels, the switches with molding axes not consistent with that of the steering wheel, switches gaps and flushes vs the bezels, rocking study of the bag, and any additional bezels with multiple mold axes!
And all must satisfy style surfaces that are perfectly clean and coherent with the intention of the style design.
Let’s not even try to start with subdivision surfaces, we don’t have time to waste, even if it takes more than a year of work to make a steering wheel just for the convergence between engineering and style!
Every single ergonomic, technical, functional and stylistic aspect is checked every day. And there is not a single step where subdivision surfaces can be used!
Those steering wheels you see online made with Maya or Blender or Modo or whatever don’t even have 1% feasibility.
If you want to get serious and check every trend of the surfaces you can only use traditional CAD and I must admit that the analysis and control tools of Icem and Alias are incredible.
Rhino could very well have the potential to do more and I use it effectively as it is unbeatable at setting up jobs and analyzing complex assemblies where Alias and Icem simply crash.
But I can’t believe the Rhino developers don’t know what an Automotive license of Alias or Icem does and they probably know every single tool and maybe they envy it!
But there are the commercial aspects: how much does it cost to develop competitive software with these tools? and how much profit can you take home? and how much profit do tools that promise to do magic like subdivision surfaces or Grasshopper make instead?
Architects also don’t work with Grasshopper as it would seem to emerge from the percentage of topics in the Rhino forum, there are not many houses or buildings generally produced by generative patterns, or at least in the percentage corresponding to the topics, but it is clear that this is the trend that McNeel has undertaken: a tool that was competitive at release 4 and that has abandoned the fighting spirit by bending to the commercial logic of tools with a great public appeal.
@Massy3D … what a rollercoaster of opinions!
I agree and disagree with you. But more the second one.
What you accused must be discussed at least to defend a bit McNeel… partially.
From what you wrote is seems like McNeel did a bad thing developing SubD or Grasshopper.
It’s NOT.
SubD are only 10% part of your workflow? Ok.
I design mountainbikes in the last 2+ years. For me is much more than 10%, but even if it were that, is the fact that we have them!
Yes, I spend much more time aligning other parts and working with surfaces after, but everything is possible because the initial model can exist thanks to SubD. Without SubD? We probably would do much uglier designs or in much much more time overall. Any edit? Restart completely. Instead with SubD we can correct the whole structure in an hour or less!
We correct the project following the manufacturer requests by working on the SubD, not only with surfaces.
You can’t compare 1:1 Rhino with other software that costs 20x as much. Really.
Let’s be honest: yes, Rhino users are probably poorer, but it lets a small team achieve virtually the same results as a big multinational company.
And Grasshopper. Let’s just not discuss it. You clearly didn’t understood what Grasshopper is and how can it be useful. I’ll pretend I didn’t read …
Try to go to Alias official forum: in the first page you see 2 weeks old posts.
Here you have to scroll for some seconds to reach “yesterday”, and a big portion of the posts are Grasshopper related.
This mean something.
Maybe the “public appeal” you mentioned … or maybe something else.
I agree.
It depends on the work you do and on the needs of each one.
I know bike designers who use solidworks exclusively simply because they design bikes that are probably very different from yours.
But this post was born out of the need to have better surfacing tools than the current one and more in line with those software that cost much more than 20 times!
But I am a Rhino user and I was also an Alias modeler and I work with people who use Alias and Icem for 10 hours every day and I assure you that in order to make certain products with the required industrial precision and with the quality and complexity of automotive design the subdivisions surfaces are totally useless except in a few situations.
Sure, we have an initial creative phase where, if 3D, Blender Is widely used and we have a later phase where we start from scratch using the Blender files just as a guide but we could very well just use 2D sketches which we do on a regular basis.
Rhino has enormous potential, perhaps unknown to too many users, but it is not enough for certain phases of work and for a certain type of industrial need.
I remain of the idea that Grasshopper and subdivision surfaces are just claims, a part that I could also appreciate if it were about innovation and research, however, after the surfacing tools have been solved and strengthened.
I hate Alias for the fact that they don’t fix their bugs, that they are always busy inventing new ones, and that the workflow is very limiting in the analysis of complex assemblies.
But I must also admit that they often introduce really useful features and that they know the automotive world very well and in fact have a version dedicated to it.
But nobody beats Rhino in the selection of objects because it often happens to analyze 3D with hundreds of pieces on a single layer and with Rhino there are different ways to distinguish them… it seems secondary but it improves understanding and this is essential for a designer.
It allows me in record time to set up and represent solutions. And he is so quick in building curves and surfaces that I don’t need any history but I simply do and trace and erase what is no longer needed. And Rhino’s layer management is something Icem and Alias can’t even dream of: think that in Icem the management of layers is so complex (and has just another name and another paradigm) that to keep more wips of the same object in the same stage they have to duplicate it and move it by 500mm in order to have it easily available!
In Rhino I use layers as history and it’s very easy and perfectly ordered.
But unfortunately when I have to translate a steering wheel grip into -A class surfaces, to speed up the work of the modelers, I’m sorry but it’s not the best.
And support that Rhino is not far away from achieving better results because the development team knows very well what we are talking about and because others have developed interesting tools like VSR and XNurbs with very small teams.
But Rhino must go back to being the competitor of Icem and Alias and not run away by proposing trendy tools.