Does Rhino 5 benefit from double or triple video cards?

Justin, I think you need to start listening and start learning from what is shown to you.

Let me break it down for you:
Rhino USES the hardware. It is BUILT to run on OpenGL.

1- STABILITY is always more important than raw speed.

2- ALL software have limitations and are bottlenecks in one way or another.

3- ALL hardware have bugs and issues that needs to be solved as workarounds by the software

4- Rhino has chosen an approach and are changing all the time, respect and understand that.

5- There are no huge corporations that are “controlling” things, only people doing their best. Understanding and respecting that is important.

6- EVERYTHING takes time, and something ALLWAYS comes up, that needs attention. Bugs, family, health, money, new hardware, new OS’s, internal conflicts etc. etc. If you want to work with development then get used to this ASAP! You do NOT have to like it, but you have to deal with it none the less. It’s how things work.

7- Everybody hates bottlenecks, but it is part of life. So you have to research what you need, and how to deal with it. It goes with everything in life. Even relations with people.

8- When summing up be accurate. " I learn Rhino is not built to use the hardware available to" is not true. “I learn Rhino is not currently built to push the hardware to it’s limits in all situations, and that they are working on it” is more accurate.

2 Likes

From a high end user view Justins subsumption isn’t wrong. Only I think users need to understand that Rhino isn’t created for high end, it’s for medium usage. It’s not wrong, it’s like the price - medium range.
Only for interest my question is - where will Rhino be in the future? I suppose so it will be what it is and if some one like to use it in the high end range he needs to know tricks and workarounds.

I frustrated about the limits often too, most since I use hardware from the high end range. But Rhino is a wonderful allrounder and I don’t see there an alternative. :wink:

Yep, like I said:

–Mitch

Yes, I heard all that. Their may some confusion about what I’m chiefly talking about, and that is FPS(frames per second). That Rhino doesn’t use the latest Bells and whistles isn’t chiefly important to me.

And if I can’t achieve Reasonable FPS, then having some work arounds is important for me, and for the direction Rhino takes. This keeps the product versatile, and keeps me stuck to it (In the house vs mouse way I mean, as the product states itself, you can build anything from a ‘house to a mouse’).

Rhino needs to stick with its’ ‘you can build anything’ motto.

Customers all understand software is limited by hardware, by the available technology, It just sounds very bad when you say its the softwares fault not he hardware. And Im talking about FPS, not all the bells and whistles on newer tech. We’ve all played video games before, and know that older games do not use the fancy features on new cards, but they run much much faster, because the newer hardware allows them to run at higher FPS.

I am not a software engineer. A detailed Wiki on this subject would most helpful.

I am wondering if some of your problem is the Quadro 6000 you are using. The builders of the computer I am looking at getting say the Quadro work horribly in some applications, and Geforce cards are much better and cheaper.

But I am not a software engineer. A detailed wiki on this would be most helpful.

Hi John- IF Rhino 5 supports 2-3 video cards using SLI Nvidia setup, does Rhino itself parse threads to the GPU’s for processing floating point calculations?? In other words, if I had a quad-CPU core and three SLI Nvidia cards connected, are all the threads processed on the CPUs or are some passed over to the GPUs?? If not, it would make a lot of sense for rendering! Thanks, Stuart

Just my two cents… I am a college student in Brazil, at FAU-USP. Since I am moving in a couple of years to my end thesis and graduation, I have spent some time and money researching and then savings to buy what would be a decent to good 3D notebook to last for the next 2 to 3 years and become my rendering workhorse.

So, with some research, I decided on a Clevo 650SG. The current configuration is a Core i7 CPU, with nVidia 980M GPU with 4GB of DDR5 memory, and have installed 16Gb of HyperX DD3 Memory out of 32 Maximum possible. And two SSD HDs ( a Samsung EVO 850 and a Samsung 840 Pro ). And still have a PCI-e slot for an extra “HD” to go. That said, it’s not at all a flimsy machine.

This semester I spent in frustration over SketchUp’s incapacity to handle large models. And it was a semester spent looking for alternatives that could be integrated in my future professional life, as SketchUp felt more and more as a toy, and not a tool, and that search led me to Blender, which proved to be somewhat not adequate for Architectural design. And talking to a friend, who had a great experience with Rhino and Grasshopper, I evaluated and finally got an educational license of both Rhino and Thea Render to be used as a render engine. And I am both impressed with my learning curve, productivity, and quality of the end products I presented just last week as the semester ended.

And for the end of the semester I ended up working on my grand plan for a neighborhood as 3D model. The idea was to be able to work the whole project as one big file, as one “Digital Model” as I would a phisical model to develop the concept, including tree placement, object placement such as benches and trashcans, which are part of the finished project for not only the Grand Plan of the new neighborhood, but also the character of the streets created and quality of the spaces. And in the end, be able to create both plans, section cuts and 3D renders out of the same “design object” if it is correct to use that term.

I just wanted to add my two cents as a student, on a budget, who is making the effort to move on 3D modeling, on a somewhat poor country and a public university ( USP ). Many students, when buying new notebooks, seek good to decent videoboards and SSD disks to be able to better use softwares such as Rhino. Most times, CPU and Video boards are the main item, and then the struggle to get the best combination to fit within that constraint. Usually that results in an i7 with 2Gb of VRam DDR3, And I sat in somewhat frustration to see my system was mostly idling as I had problems handling the model, with very low FPS even when I was on wireframe view. Yes, I am still on my learning curve, and I fully understand I will get better at this. I will also study more on how to work such models on Rhino and possibly the next time I will have a much smoother experience. But one thing is to have your model behaving heavily when you are using all your resources and you think “Oh, yeah… either this is really too complex for this machine or I am doing it all wrong”, another is when you have less than 50% of your CPU being used, 0% of the actual graphics card, and less than half your available memory. There is the feeling my work could be running smoother, but isn’t.

So, far from questioning your business model, as I am grateful for the license I got and it is a price I can go for, but I just wanted to leave my two cents to say that the paradigm of students with poor machines is changing. We cannot afford SLI desktops dedicated to the task, but 16Gb notebooks are becoming the norm, as well as nVidia cards with 2Gb of Ram, even if most still cannot invest as I did in CUDA cards with DDR5 ram. Many students tend to get the best they can for their money in order to be able to move into digital design fully.

I am saying this, because maybe the way to compatibilize this would be a graphics configuration tab in which says as in Adobe products how much of the available system memory I want Rhino to take a hold of, how many CPU cores I would like to have it take as dedicated to itself ( as for example QGIS does ), and I daydream of a checkbox saying “Use Old 3D Hardware Rendering Model” in which users with limited hardware can check and go for an older OpenGL implementation, but the rest of the users can opt to use at least a middle ground implementation that takes greater advantage of newer hardware.

I am fully aware of the complexity of what I am saying, this is sort of a wishlist of a newbie user. Comparing to Autodesk products, I already think Rhino feels faster on same sized models. It is already very close to a dream tool that brings together the pieces I need from Autocad for example, with the easy of use and 3D thinking on modeling I brought from SketchUp and brings that to a new level I am loving. But most of my work has always been complex and models have been big, and I think us students would greatly benefit from Rhino allowing itself to have a little more elbow room within the system to use more resources and allow for bigger models to be deal within it.