Software for a Student

Looking at internship postings, it seems to be expected that as a junior-level undergrad I should have experience with some CAD and rendering softwares. Since my university starts teaching these fairly late in the game, I am looking to buy some student-editions to get a little experience on my own. The problem is, every internship posting I see wants different software, and I’m a little lost as to the use and pros/cons of each.

Which software options are most generally used in the industry, and what is their purpose? So far I’ve seen SOLIDWORKS, Rhino, Keyshot, Alias, Modo, Autodesk, Maxwell, Vray, Luxrender, Mental Ray, Softimage, Maya, Blender, etc, etc. My head is positively spinning with all of these, and I have no idea where to start.

Any advice or opinions would be welcome!

So it depends on the industry

Most product design:
SOLIDWORKS, (sometime only)
Rhino - ID firms but usually are rebuilt in SW
Keyshot - has become a simple go to tool for most ID

If you focus on these to be honest all other packages require the same base foundation knowledge - it then becomes more about finding the same tools with in or becoming accustomed to the different interface.

I think Rhino, Keyshot, and Solidworks are really good starting points as they are the most widely accepted. I have noticed quite a few firms starting to use CREO (Mostly in engineering) as well as Fusion 360. It really is dependent on the firm, but from what I have seen Rhino, Keyshot, and Solidworks are the most used. Also, you may want to check with your school for things like Keyshot. My school had a floating license of Keyshot for the students to use.

I think a good way to think of these software packages is to break them down into categories:

Parametric CAD: Solidworks, Inventor, CREO and the enterprise grade Catia and Siemens NX.

Explicit Surface Modelers: Rhino and Alias would be the two biggest players.

Renderers: Keyshot, Maxwell, Octane, Vray maybe to a certain extent Blender

The popularity of the different packages varies by location and industry. The good thing is that learning one of the tools in each category will allow you to learn the others quickly. Most companies looking for a designer rather than a CAD monkey won’t care a whole lot which tool you master.

For most product design needs in consumer products, Solidworks is very popular for the CAD work. Detailed surfacing work is a bit of a specialty but Rhino is quite popular for this. But honestly, as someone who’s been using Rhino full time for the past 2 years, it’s about as fun as removing glass shards from your eyes. Also, the future of detailed surfacing work may be changing in the very near future. I wouldn’t be surprised to see the need for dedicated surfacing tools to be greatly diminished as better surfacing tools get integrated into CAD packages. For example, Autodesk’s Fusion 360 which is a hybrid between a CAD and a surfacing tool and makes for nice workflow.

Lastly, for rendering, it’s hard to go wrong with Keyshot. It’s simple and intuitive which makes it great to learn the basic concepts of rendering. While it may be lacking some of the very advanced nitty gritty details, it’s so fast to get great results from Keyshot that it’s a bit of an industry staple. My biggest suggestion when learning rendering would be to dabble in product photography. It’s way more intuitive to learn how to light an object by moving lights and modifiers around a space than moving around little sliders and waiting for the results to show up.

Lastly, I’d also recommend learning the crap out of Photoshop. Not only for sketching but also doing the post production of pictures and renders. It’s a vast and somewhat intimidating program but if you can wield it, you can save a lot of time and bring your results to a higher level.

TL;DR I’d start by learning Solidworks and rendering in Keyshot and doing your post production in Photoshop. If you eventually see a need for it, have a look at the advanced surfacing options that suit your needs and industry.

I recommend Rhino just because it is the most practical choice for a student.
The fully functional trial gets you up and running with a professional CAD package in no time and if you follow some tutorials, in 2 weeks you can get yourself already to a decent level. The full software is not very expensive compared to the other ones you mention and Rhino has great plug-ins and scripting capabilities.

The step from Rhino to Solidworks is often made when you have to develop a product towards manufacturing which requires many revisions.
Fortunately this is not a huge step as surfacing approaches are quite similar (mostly Sweep/Lofting - Trimming/Blending - Finishing operations)

If you post your surfacing work here we can provide you with feedback.

As a student Rhino is $128 to buy, and you own it. They even let you upgrade it to a commercial license after graduation (at a cost).

Solidworks is $86 (as a student), but that is for a one year license. They are both really powerful, and industry standards.

Fusion 360 is free to students. It isn’t an industry standard, although it is getting adopted fairly quickly. Fusion will let you model in the same ways that Rhino, and Solidworks do, plus it offers T-splines sculpting.

For rendering. Go with Keyshot. They have a student version available too, and Keyshot 7 (getting released soon) looks like another big jump in its capabilities.

You can’t go wrong with any of the above. With the internet, it is so easy to over analyse every decision you need to make. The most important thing is to just pick one, and learn it.

Autodesk Fusion 360 is a good option to look into, the step to/from SW or Rhino does not seem big and Autodesk is the strongest company, linking also to the 3D printing and movie visual effects space which will have great advantages in the future if they can couple these programs. I am coming across more and more studios requiring employees to master Fusion 360.

I may be biased because I am a long time SolidWorks user but I am now learning to use Rhino because I am going to be teaching it come September to university students.
I can honestly say that I see absolutely no point in learning Rhino whatsoever. The notion that it is easier or faster than SolidWorks is entirely false in my experience thus far. With a good SolidWorks teacher you will pick it up as fast as Rhino and it is infinitely easier to model to scale and account for manufacturing, etc, etc.

I think if you are just trying to replicate the way you work in Solidworks then Rhino won’t be any fun for you. Are you sure you want to teach it next month if you are just learning it now? You say with a good Solidworks tutor you can learn Solidworks fast, I say with a good Rhino teacher, that knows his craft longer than two weeks, you learn Rhino much much faster :wink:
Rhino is a fantastic tool for design ideation which lets you sketch in 3D without any restrictions. Rhino’s biggest strength is that it allows you to be really really dirty and let’s you do things that Solidworks doesn’t offer. Rhino is not a good tool to bring things into manufacturing, but that’s not what it is made for. Do your students a favor and don’t tell them there is no point in learning Rhino because there are many many designers that would strongly disagree with that (including me).

Totally disagree. I am very proficient in both packages,and really enjoy using both. However, if I need to build something fast, or sketch out ideas, I go to Rhino every time. It is so much faster to build in than SW, and lets my ideas flow more.
To me Rhino is more like working in a 3D version of Illustrator. The layers, work the same (very handy for bringing parts into Keyshot), and you can draw in a similar way.
To fully define a sketch in SW so that it is editable later (which is really the benefit of SW) takes way longer than creating that same shape in Rhino.

Well this was the point I was going to make also. If you are using Rhino you have to know EXACTLY what dimensions your design is going to be before you model it up. Properly modelling in SolidWorks you can change the dimensions of a part with one change and everything updates? Rhino? Make one change and almost start again.
I don’t believe it takes more time to properly define sketches and features than the value it adds. If you know EXACTLY every single dimension of your design before you enter CAD then you are a wizard, but I sincerely doubt that is common and the ability to explore and iterate in SolidWorks with design intent makes it invaluable over Rhino.
There are some complex surfacing forms that are easier and faster in Rhino for sure, but again I would still rather take the effort to model it properly in SolidWorks personally and have it editable.
Ultimately I still think that Fusion 360 is the best package going now in terms of versatility and I wish I had the time and motivation to learn it but I do not at the moment.

I can see your point, but in my studio we are typically showing 2-3 final concepts rendered out in 3D. To fully define three models is very time consuming, and ultimately a waste of time for 2 of the design directions. Having said that, you are correct that any big changes to your design modeled in Rhino will take more time as it will require a rebuild.
I agree that Fusion 360 has a lot going for it, and we did buy it for our studio to evaluate. The big problem is, we are always under a time crunch, so we almost never have the extra time it would take to model a design in a package we aren’t familiar with. We are chipping away at it slowly, and Autodesk have been amazing in offering support and free onsite training. I think they will definitely be growing their base in the next few years.

In the end it really just comes down to personal preference. NURBS vs SubD, 7th degree curvature continuity, making changes to adjust for client request. “Speed” is relative and workflow/project specific. There are times when making a certain model will need whatever tool is needed to get to the end goal. You’d no sooner use a drill press to try and cut a plank of wood…right tool for the right job. This is the area where I feel like most designers fall short (or maybe the finger could be pointed at the company) for not knowing the full range of tools that should be available in a designers tool set. It’s defacto that a company will have the Adobe suite and at least Sketch Up, Rhino, SW…(insert CAD software of choice). What is rare to find is company adding in say Houdini, C4D, zBrush or Modo in to the mix which, at first glance having some of these other software tools may seem like too much, but if viewed like a traditional shop then this just makes sense. But this does an another layer of complications on a multitude of levels, I get it, not saying it doesn’t complicate matters, but the options possible brings more to the table. Add to the fact that there is never enough time to actually properly evaluate and/or implement any of these other tools which does make it that much more challenging.

There’s another thread here about creating a displaced surface onto a piece of geometry and the idea of Rhino and Grasshopper as the solution, if the question is about speed, is the absolute wrong way to go in this specific instance. zBrush would knock that out in like 1/10th the time and Rhino/GH but most designers won’t know or have the knowledge or skills set with which to even know either the possibilities and/or capabilities to execute that workflow.

…bah I’m off on a tangent here but hope this at least begins to make sense to some.

This is why I never got round to learning it, as much as I’d like to and as much support and tutorials etc that are out there it is time that blocks it. I don’t know how much leeway they will make in industry until more universities adopt it and that is happening in the UK slowly.

I have done a few smaller projects with Fusion 360 because everybody keeps saying it is the future and consider myself pretty proficient in it now.
I don’t know, I think most people who claim that it is the future actually never really used it and just really WANT it to be the next chapter in CAD because they are tired of their old packages.
But let me tell you, you would be surprised about what kind of super basic things you simply can’t do with Fusion. It is just not there yet. You start a project and think “this is great”… then after some time you can’t do this or that and start to frenetically googling how to do it until you figure out that there really is no way of doing that yet. It’s quite frustrating and probably because the software didn’t have time to ogranically grow over the course of a decade. I honestly wouldn’t really want to use it as a production tool since I constantly hit various feature walls. One of my most favourite examples is that you simply cannot draw an orthogonal line to a non-straight spline. It is mindblowing to me how something so basic can’t be done.
So be careful with praising Fusion of being the silver bullet of the industry, it really is far away from it. They just have a big marketing budget right now, but the software will still need several years until it can compete.

And even if you ignore the missing features, it is still not as fast and as versatile as Rhino and it also lacks the useful bells and whistles of Solidworks. And the SubDs simply suck compared to e.g. Blender. Fusion is just not really “the best” at anything which will be the hardest hurdle to overcome for Autodesk.

The sub-D in F360 can convert to perfect nurbs surfaces which can be thickened etc though? If you use sub-D in Modo then you can’t?

I am mostly annoyed that DS didn’t buy T-Splines when they had the chance and rather than buy Power Surfacing and implement it they bought out an entirely new CAD package which noone is switching to. DS have shown they really don’t care about their user base despite its size.

Well, yes, you can convert subDs to Nurbs in Fusion… but so can T-Spline for Rhino, or Inventor, any other incarnation of T-spline and even Maya can convert entire polygon objects to nurbs. Doesn’t even Modo have some plug in to do that?

The advantage of Fusion is of course that you can have the conversion within the modelling process in the timeline and can always go back to your subD and make changes after the fact.
But what you also need to be aware of is that converted SubDs are not “clean” or “perfect” at all and quite the opposite of single span Class A modelling. The math of these surfaces is REALLY heavy. Yes, you can thicken them, etc. But it sounds much more useful than it is in reality cause they are so friggin overdefined and complicated that they break at every chance they get when you start building upon them and do things that go beyond a simple thicken of the surface. The SubDs can be a quick solution and sometimes are the easiest way out of a problem, but they are usually always the least elegant solution.

I really like the idea of what Fusion is trying to do with SubDs/T-Splines… but it is just not really finished and useful yet and almost creates as many problems as it solves, from a technical standpoint.

There is a thicken command in Modo or you can pick geometry and offset the surface and then bridge or loft between them. Just as an aside you don’t “have to” convert to NURBS to take something to be manufactured you can totally go direct to CNC, 3DP…etc.

Whom are you speaking of when you say DS bought another CAD package?

The plug in is called Power SubD and honestly there’s nothing else quite like it in industry in terms of pre-defining edges on a SubD model so that when converted to NURBS you can pick said edges and continue and CAD operations required.

[/quote]But what you also need to be aware of is that converted SubDs are not “clean” or “perfect” at all and quite the opposite of single span Class A modelling. The math of these surfaces is REALLY heavy. Yes, you can thicken them, etc. But it sounds much more useful than it is in reality cause they are so friggin overdefined and complicated that they break at every chance they get when you start building upon them and do things that go beyond a simple thicken of the surface. The SubDs can be a quick solution and sometimes are the easiest way out of a problem, but they are usually always the least elegant solution…[/quote]

If at all possible to show some examples of SubD that you’d consider not clean or perfect. I’m not necessarily disputing what you’ve experience just saying that there are possible ways to have bad NURBS geometry for a whole host of reasons. There are a fair amount of car companies using zBrush, Modo…etc as part of their design process. Not saying it’s the end geometry used but this notion of “class A” being this be all end all is also kind of urban myth that people throw around with really thinking about what’s really behind it. This could mean it was done in Rhino, SW, IV…etc. And when you say that SubD’s are heavy, one that that you can do with mesh geometry that NURBS can’t is actually decimate and reduce the heaviness. With NURBS you really can’t do this… you can try to rebuild it, yes, but that’s about it. It’d be great if there were a shrink wrap feature for NURBS surfaces but alas…:frowning:

I think you got me wrong, the SubDs themselves are mathetacially as clean, simple and smooth as it gets. It is really hard to mess up SubDs. It is the CONVERSION to NURBS that really f**ks things up. Converted SubDs are simply NASTY. Not only are they really heavy NURBS, also they are not able to maintain the perfect curvature of the poly mesh. Converted SubDs tend to “pinch” on starpoints. Even if the polysurface is absolutely pristine in Blender or Modo or whatever, once you convert it the curvature will do funny things. T-Splines are good to get you to where you want to be by 90% to visually ok data that is good for some prototyping. When you get into the nitty gritty surface developement it simply doesn’t offer the necessary control over surface curvature and produces way way overdefined NURBS. And when I say Class A I mean single span. And converted meshes usually have a thousand spans. Think of it of converting a pixel image (that’s the mesh) to a vector graphic (that’s NURBS). The output might look ok, but it is usually overdefined beyond usability.

And yes, car companies use it for concept cars and prototyping because poly modelling is friggin fast. But first they usually stay exclusively in polys and don’t convert, so they avoid all the curvature problems and second for an actual production car the data is useless and they will rebuild it single span… because the final precise control over curvature isn’t there.