5/23/2019

Gpu For Video Rendering

Rendering speed is more dependent on the power of the CPU not the GPU. Essentially what you're doing is offline rendering after you have finished editing the video with title and effect tracks. Another possibility is that they still treat GPU rendering as experimental feature that's not stable enough to set it as a default yet. You don't want to waste hours rendering your video just to realize something is screwed up due to GPU rendering bug. If you decide to use it anyway, then you can't blame the software publisher - it was your.

Best Video Rendering Software

This guide will help you build a PC that's perfect for rendering scenes like the one above, which was made in Maya.

Last updated: December 2018

Rendering is the process of turning 2D and 3D models into images and animations. If you want to put together a PC that will give you the best performance for rendering out designs and/or animations at the lowest price, then you’ve come to the right article!

By building a PC for yourself, you can ensure that it is tailor-made for your exact needs with as little waste as possible. Below you’ll find four sections to take you through everything you need to know.

First, we’ll explain which components are most important for creative work, and how to get the most out of your new workstation.

Section 1: Example Rendering PC Builds

These builds are designed to provide the most power for rendering images and videos in programs such as Maya, Cinema 4D, Blender, Modo, and all the rest. General info on how these parts were selected is available in Section 2 below.

Note: When building a new PC, you’ll need to install an operating system. For maximum compatibility with 3D design and rendering, we recommend using Windows. You can purchase Windows here. Some people are eligible to get Windows free through school or work, so you may also look into that.

The Cost-Effective Rendering Build ($700)

This budget build takes advantage of the great performance of Intel’s budget processors, and goes for an inexpensive case. The build still packs dual-channel DDR4 RAM, a solid mid-tier 4GB graphics card, and a decent amount of storage.

This build even fits in a small SSD to act as a boot drive, and hold a few of your most-used software choices. Running programs that read and write as often as most modern modeling and animation tools can really take advantage of an SSD.

CPU: Intel G4620
Graphics Card: GTX 1050 Ti
Motherboard: Gigabyte GA-B250M
RAM: 8GB DDR4
Storage 1: 120GB PNY SSD
Storage 2: 1TB HDD
Power Supply: EVGA 450 B1
CPU Cooler: Stock
Case: Vivo Case-V00
Operating System: Windows 10

Balanced Price-to-Performance Build ($1250)

As usual, building at the higher mid-range yields an incredible machine for the price.

At this level, you get the latest architecture Intel 6-core CPU, a bigger SSD with enough room to fit your active projects in addition to your Windows and application installs.

The rest of the build is significantly upgraded too. A higher quality case, a more reliable, more efficient power supply, and more effective, quieter cooling.

CPU: Intel i5-9600K
Graphics Card: GTX 1060 6GB
Motherboard: MSI Z390-A Pro
RAM: 16GB DDR4
Storage 1: 250GB Crucial MX500 SSD
Storage 2: 2TB HDD
Power Supply: EVGA GQ 650W
CPU Cooler: CryOrig H7
Case: NZXT S340
Operating System: Windows 10

High-Performance Build ($2000)

Professional level build. Very high performance, although not twice as much as the higher value Price-to-Performance build.

If you’ve got 3D design work to do in a professional capacity and want to minimize the amount of time you're wasting with tools, then this is an excellent build for you.

This build sports a new Ryzen 7 CPU, with the some of the highest single-threaded and multi-threaded performance of any CPU. This build will have exceptional performance for creating, modifying, and animating 3D models. It will have very good rendering performance, although not as high as the more expensive builds.

This build also has a very efficient and quiet CPU cooler to keep the powerful CPU running cool and reliable, a graphics card with enormous performance for CUDA rendering, one of the fastest SSDs around, and a very high quality power supply.

If you want to do any gaming, this PC will handle all games smoothly at up to 4K.

CPU: AMD R7 2700X
Graphics Card: RTX 2070
Motherboard: MSI X470 Gaming Plus
RAM: 32GB DDR4
Storage 1: 500GB Samsung 970 EVO M.2 SSD
Storage 2: 3TB HDD
Power Supply: EVGA G3 750W
CPU Cooler: Scythe Fuma
Case: Fractal Design Define R5
Operating System: Windows 10

Professional Build ($3000)

The most performance you can get, before going off the deep end of the performance / price curve.

If you're not too concerned about budget, but balk at the price tag of the build below, this build is a good option.

The high-clock-speed, 8-core, 16-thread CPU has a great combination of single- and multi-threaded performance. Very balanced.

1 TB of extremely fast SSD storage will make booting up, shutting down, accessing files, and opening programs snappy.

CPU: Intel i9-9900K
Graphics Card: RTX 2080
Motherboard: Gigabyte Z390 Aorus Pro
RAM: 32GB Quad Channel DDR4-3200
Storage 1: 1TB Samsung 970 EVO M.2 SSD
Storage 2: 2x 4TB HGST HDD
Power Supply: EVGA G3 850W
CPU Cooler: Fractal Design Celsius S24
Case: Enthoo Evolv
Operating System: Windows 10

Forge of the Gods Build ($7500)

If money is no concern and you want the best tool available, this is it.

The AMD Threadripper 2990WX provides an unbeatable 32 cores and 64 threads. This build also sports a monstrous two RTX 2080 Ti cards, 64GB of RAM, 2TB of the fastest SSD storage available, and an extremely efficient and stable power supply. Quiet and sleek, with colossal computational capability.

If you want advice on how to customize this build to save money or be even more powerful, email us: contact@logicalincrements.com

CPU: TR 2990WX
Graphics Card: 2x RTX 2080 Tis
Motherboard: ASUS Prime X399-A
RAM: 64GB Quad Channel DDR4-3200
Storage 1: 2TB Samsung 970 Pro M.2 SSD
Storage 2: 2x 10TB HDD (for RAID-1 redundancy)
Power Supply: Seasonic Prime Platinum 1200W
CPU Cooler: Corsair H115i (with TR4 mounting bracket)
Case: Enthoo Primo
Operating System: Windows 10

Section 2: Choosing PC Hardware for Rendering

Best Video Rendering Software

CPU

Once upon a time, this would have been the most important part of a rendering PC—so much so that I would have recommended putting all possible resources toward obtaining and supporting the most powerful CPU that your budget would allow.

But CPU rendering is no longer the only game in town, and the focus can be shifted somewhat in favor of the graphics card. That said, the CPU is still vitally important. The CPU is used for everything you do on a computer, so a faster CPU will always be at least a little bit better.

CPU rendering can take advantage of many CPU cores. Intel CPUs typically offer the best single-threaded performance, while AMD CPUs typically offer the best multi-threaded performance. Both companies currently have some great chip options for rendering builds.

If you use a different CPU than we recommend in our builds, make sure you get a compatible motherboard, RAM, and CPU cooler.

Graphics Card

For rendering, unlike most other graphic design work, the strength of your graphics card is important for the speed at which you are able to complete your work. GPU rendering (when available) is much faster—in some cases, over ten times as fast.

You’re likely to want an NVIDIA graphics card. Here’s why:

Between the two most developed kinds of GPU rendering available now (AMD-backed OpenCL and NVIDIA’s proprietary CUDA), CUDA currently provides better performance. And while OpenCL is compatible with any graphics card (at least to some extent), CUDA is only available with NVIDIA cards.

To quote GPU Rendering Magazine’s article on OpenCL and CUDA: “As an easy rule of thumb, if your app supports CUDA, grab an NVIDIA card, even if it also supports OpenCL,” and “If the applications you use split their support between CUDA and OpenCL we recommend using a recent NVIDIA card.”

Hopefully AMD's OpenCL will have wider support and better performance in the future, but for now NVIDIA is the better option.

RAM

RAM is not going to be one of your biggest concerns. Modern consumer systems all use DDR4 RAM. The lower end systems here use dual-channel RAM, while the higher end systems use quad-channel. A Dual-channel configuration provides slightly better performance than a single stick, and quad channel is slightly better than dual channel.

Just make sure you’ve got at least 8GB, as the previewing process that makes up the view window of some 3D design programs like Cinema 4D can be a bit RAM-hungry. If you find that you need more RAM, adding more is easy as long as you still have open RAM slots on your motherboard.

Storage

Unlike computers intended for filmmaking, traditional animation, gaming, or general YouTube video-making, a computer for 3D design doesn’t necessarily need a mountain of storage space. Images, assets, and modeling files are significantly smaller than videos or games.

A terabyte or two of hard disk drive space, like a 1TB Seagate Barracuda, might be plenty.

If you intend to do loads of animating. Animation videos—especially high-resolution, high-frame-rate animation videos—take up just as much space as any other space-hogging high-quality video files. Fortunately, upgrading your storage capacity is one of the easiest changes you can make to an existing workstation PC. If you're on a budget, adding more storage is easy any time, and it always gets cheaper over time.

SSDs

Definitely get an SSD if you can. Everything will feel faster. Installing your OS, applications, and your project caches on an SSD will speed up your workflow in a very noticeable way. Programs and parts of programs will load faster. Cached information will be provided quicker.

SATA vs M.2 PCIe

Upgrading from a hard drive to a SATA SSD is a huge speed boost, and very noticeable.

Upgrading from a SATA SSD to a more expensive PCIe SSD is a big boost in peak performance. However, it is less noticeable in real world use, unless you're often transferring huge files. PCIe is the best and fastest, but it's often worth sticking with a cheaper SATA SSD and spending the money on a faster CPU or graphics card instead.

Power Supply (PSU)

Getting a PSU that doesn’t support a sufficient wattage for your build, or (even worse) getting a budget PSU that may not have been well-made, increases the risk of catastrophic failure for your build. The specifications or numbers on the packaging don't tell the whole story, so unless you're an electrical engineer, the best way to find a good power supply is to ask an expert.

Don’t risk damaging your components. Get a high quality PSU. Unlike many other components, the quality of power supplies does not increase as much over time. A high quality power supply bought today, will still be a high quality power supply in 5 or 10 years.

Section 3: Frequently Asked Questions About Rendering

FAQ 1: Why does rendering take so long?

It depends on which software you’re using, what options you have set, and what hardware you’re running.

You render a simple scene, e.g. a house sitting in a field next to a tree, with the sun shining on it. Your computer is calculating (or repeatedly estimating) the light bouncing from your one point light source (the “sun”) off of each individual polygon’s relative angle toward the camera. (There might be hundreds of polygons in the leaves of even a simple, not-particularly-realistic tree.)

At the same time, it is reflecting that light uniquely depending on the material, color, and texture of each simulated object. And then it is (again, depending on the software and settings) doing the same for the light reflected uniquely off of each polygon, such as from the house onto each leaf of the tree and vice versa.

It is attempting to map or reasonably approximate that set of calculations for each and every polygon visible in the scene in order to produce one cohesive image. And finally, it is doing any post-processing that you instructed your program to accomplish. And if you’re rendering out an animation, then it’s doing all of that for each frame.

The computer is simulating not just the eye that views the image, but also the world-like conditions that the eye is viewing. Some hardware is better than other hardware at accomplishing these kinds of tasks, but it is nothing short of astounding what kind of images can result from this herculean computational effort.

FAQ 2: How can I speed up my rendering?

If you’re still worried that your renders are taking too long after reading the answer to FAQ 1, I recommend going back to basics and checking your options:

Can you reduce computed passes, computed samples, or lighting bounces without harming your render’s quality?

Are you sure you’ve switched to GPU rendering?

If you are using a GPU rendering option, does your application require you to manually adjust a tile size so you can take full advantage of it?

Are you doing any post-processing that you can do without?

Test out all of your options in each of your applications. Different software will respond differently to different choices.

FAQ 3: Why would you choose GPU rendering over CPU rendering?

While CPU rendering is both accurate and reliable, it is very slow compared to what is possible these days with GPU rendering.

To give you a practical example, a couple of months ago I used Blender to make the animation that’s being worked on in the screenshot below this section (not exactly a complex scene). It’s an 11-second, 1080p, 60-FPS animated intro for my YouTube videos. Rendering the 660 frames of that animation via CPU rendering took over 20 hours (with an i5-6600K). Rendering those same 660 frames via GPU rendering took under 4 hours (with an i5-6600K and a GTX 1060 6GB).

Sequential operations are the CPU’s easy victory, but in the realm of outputting rendered images via parallelized rendering, the GPU is in its element. The CPU is still an important player regardless, but prioritizing the GPU when you’re budgeting can make a lot of sense.

FAQ 4: Which graphics card should I buy for GPU rendering? Why not an AMD card?

As for which kind of GPU I recommend, the answer for 3D modelers and animators (for the time being) is an NVIDIA card.

Secret games 3 full movie. Why? Well, as explained at length in the graphics card portion of section two above, NVIDIA got a head start on expanding into GPU computation and they provide top-notch support for integrating their CUDA systems into applications. So, while strides are made every year to make up the difference, NVIDIA GPUs are the more broadly supported and the higher-performance option for GPU rendering right now.

If you already have an AMD card, check your software. The AMD card may work just fine.

FAQ 5: Do I need a powerful CPU for rendering?

Yes. If you’re going to use your computer to render 3D scenes, then it’s got to have a powerful CPU. And if you’ll be using CPU rendering, I wouldn’t personally recommend going for anything below an Intel Core i3.

The longer—but more accurate—answer (which is reflected in the ‘cost-effective rendering’ example build in section 2) is that the CPU will always be an important part of the rendering process and can not be fully skimped on, but you can get significantly faster results with having a weaker CPU if (1) you get a more powerful GPU that is capable of GPU rendering and (2) your chosen rendering software supports GPU rendering.

If you are using a program that supports GPU computation, you might actually be better off with a GTX 1060 6GB and an i3 processor than a GTX 1050 and an i7. (Consider: my unlocked Skylake i5 averages just about 40% CPU usage when I’m doing GPU rendering.)

FAQ 6: Is the software for 3D design work expensive? Are there alternatives?

Yes, it is expensive; but it’s a little tricky to just leave it at that. It ultimately comes down to what kind of work you’ll be doing with the software, as that will dictate which software you need. Hobbyists can get away with free software like Blender, but professionals and aspiring professionals will likely need to shell out the funds for professional-grade software like Cinema 4D and Maya.

That said, most software companies offer a free trial to ensure the program is right for you. And if you’re a student right now, you might be able to get the relevant software (or at least a version of it) either completely free or discounted from the company or from your school. It never hurts to check. And there are also a few free, open-source software options.

At any rate, you can find a more thorough discussion of 3D design and rendering software in section four, directly below.

Section 4: 3D Design, Modeling, and Rendering Software

Alright, so you’ve got a PC now, or at least you have one in mind. Next, you’ll need one or more programs to do your 3D design work. Here’s an overview of the most prominent pieces of software for 3D design, modeling, animation, and rendering:

Maya

Developed by Autodesk, Maya is a professional-grade 3D modeling and graphics software. It sees a lot of use in video game projects and film projects, including animated features and VFX for live-action movies.

Maya features its own renderer, mental ray, which can actually be acquired separately from the full software package if you so desire.

Maya is available as a subscription for $185 per month or $1470 per year, with further bulk pricing available.

Cinema 4D

Developed by Maxon, Cinema 4D is a heavily lauded program for the artistic creation of 3D models and graphics. It is used very widely in numerous strata of the entertainment industry. Cinema 4D boasts a lot of the same features as its competitors, but also has a large number of proprietary tools, as well as premium-quality ease of use via intuitive interface design and a robust in-app help system.

Like Maya, it has a serious price tag. But Cinema 4D is one of the pieces of software that is most closely associated with professional-quality 3D design work.

A single-user commercial license of Cinema 4D Studio can be purchased for $3510.

Blender (Free)

Developed by The Blender Foundation, Blender is a free and open-source 3D design program. Like other developer-focused open-source software (e.g. the image editing software GIMP), Blender has a steep learning curve but is nonetheless very powerful in spite of its nonexistent price tag.

Although it’s not exactly beginner-friendly, Blender’s lack of precise specialization makes it well worth dabbling in if you aren’t yet ready to jump into one of the commercial programs.

You can download Blender for free from Blender’s website.

Modo

Developed by The Foundry, Modo is an all-in-one creative suite for modeling, sculpting, painting, texturing, animating, and rendering 3D designs. This one also boasts robust in-application physics simulation. In effect, Modo aims to be an entire modular creative pipeline in one package.

But it being comprehensive does not mean it’s a lightweight; for instance, if you watched the science series Cosmos hosted by Neil deGrasse Tyson a few years back, then you’ve seen a lot of 3D work done in Modo.

A single-user commercial license of Modo can be purchased for $1799.

ZBrush

Developed by Pixologic, ZBrush is a specialized 3D modeling and sculpting program that sees most of its use in commercial art and product images. Most of its marketing is focused on creative professionals who need a means of modeling what amounts to 3D art. Because it is well-suited to complex modeling, however, it can also be of great utility in games and professional CGI.

A single-user commercial license of ZBrush 4R7 can be purchased for $795.

Sculptris (Free)

Pixologic also offers a free, much more basic piece of 3D sculpting software called Sculptris, for beginners.

AutoCAD

1960's tv shows uk. Developed by Autodesk (the developers behind Maya), AutoCAD is a 2D and 3D design program that sees a lot of use in industry applications, for instance in the drafting of blueprints for architectural projects and computer components. It is one of the best-known CAD (Computer-Aided Design) programs. This is a professional-grade piece of design software, and it has the broad support systems and high price to prove it.

AutoCAD is available as a subscription for $1400 per year, with further bulk pricing available.

Rhino

Developed by Robert McNeel & Associates, Rhinoceros 3D is a highly customizable 3D design and CAD program. Like other prominent CAD programs, Rhino sees use in many product design, engineering, and architectural projects. Its specialty is mathematical modeling of 3-dimensional surfaces and curves.

A single-user commercial license of Rhinoceros 3D can be purchased for $995.

Conclusion

Balance your build between money put toward a CPU and money put toward a graphics card, with the prioritized weight of that pair leaning slightly toward the graphics card. Make sure to pick up an NVIDIA graphics card so you can take maximum advantage of GPU rendering options.

Go for dual-channel RAM if your budget is limited, but you don’t need buckets of it (8-16 GB should suffice). And invest in an SSD, but don’t go overboard on sheer storage capacity.

Consider choosing a case that sports at least one 5.25” external drive bay, as (1) many professional software licenses still come bundled with CDs, and (2) if you do animations or put your renders into videos you may want to burn your work onto physical media.

Get all of that right, and you’ll be ready to render!

About Us

Daniel is the Content Editor for Logical Increments.

He also does all of the graphic design, writing, and web development for The Gemsbok blog website and The Gemsbok YouTube channel, where you can find articles and videos on books, games, movies, and philosophy.

Logical Increments helps more than a million PC builders each year with hardware recommendations for any budget.

If you want to see our build recommendations for general purpose gaming PCs, check them out.

What is GPU rendering?

GPU rendering means, that there is used the power of GPU (Graphics card) for rendering, instead of CPU (processor).

Conventional CPUs (Intel, AMD) didn’t make such dramatic progress, like GPUs in last 10 years. What is GPU rendering – NVIDIA article.

GPU rendering is definitely the future of rendering. One single customer GPU has the same power as the huge and expensive Features selection in reflections computer clusters with many CPUs.

GPU rendering is incredibly fast and less power hungry. Since you will need fewer computers for the same performance, it is also much CHEAPER.

Which is faster GPU or CPU Rendering?

As GPU render engines become more popular and feature rich, you may be thinking that it’s time to jump in and integrate GPUs into your workflow. The driving force behind a migration to GPU rendering has always been speed. SketchUp users frequently ask, “How much faster is GPU rendering as compared to CPU rendering?” This article by BOXX may provide some insight: GPU Rendering vs. CPU Rendering

GPU Rendering for SketchUp

These Rendering Engines for SketchUp say that they use the GPU for Rendering:

(Source: SketchUp Extension Warehouse)

Allura GPU

Allura GPU – Powered by NVIDIA Iray® – is a new GPU based renderer which utilizes the power of you GPU to create faster renderings. – is the latest, GPU based rendering engine for SketchUp.

Using your video board GPU lets us achieve amazing rendering speeds, a high quality rendering image and a satisfying rendering experience.

NVIDIA Iray is a physically based 3D renderer which uses the power of your GPU – Graphics Processor to perform the rendering.

Raylectron

Raylectron utilize all the CPU cores and threads (user selectable) to render as fast as possible and new GPU version available.

Bloom Unit

Bloom Unit performs all processing in the Cloud using powerful NVIDIA GPU technology.

Thea

Thea for SketchUp is a combination of powerful rendering engines of Thea with the simplicity of SketchUp. Having biased, unbiased and interactive render modes (including GPU support)

WHAT IS GPU-ACCELERATED COMPUTING?

GPU-accelerated computing is the use of a graphics processing unit (GPU) together with a CPU to accelerate deep learning, analytics, and engineering applications. Pioneered in 2007 by NVIDIA, GPU accelerators now power energy-efficient data centers in government labs, universities, enterprises, and small-and-medium businesses around the world. They play a huge role in accelerating applications in platforms ranging from artificial intelligence to cars, drones, and robots.

HOW GPUs ACCELERATE SOFTWARE APPLICATIONS

GPU-accelerated computing offloads compute-intensive portions of the application to the GPU, while the remainder of the code still runs on the CPU. From a user’s perspective, applications simply run much faster.

GPU vs CPU Performance

A simple way to understand the difference between a GPU and a CPU is to compare how they process tasks. A CPU consists of a few cores optimized for sequential serial processing while a GPU has a massively parallel architecture consisting of thousands of smaller, more efficient cores designed for handling multiple tasks simultaneously.

GPUs have thousands of cores to process parallel workloads efficiently

– See more at: http://www.nvidia.com/object/what-is-gpu-computing.html#sthash.BBhjS0dG.dpuf

On this page: