V-Ray CPU Comparison: New 14, 16, and 18-core Skylake-X Processors
Always look at the date when you read an article. Some of the content in this article is most likely out of date, as it was written on September 25, 2017. For newer information, see our more recent articles.
Table of Contents
V-Ray, from Chaos Group, is widely used for creating realistic 3D graphics. When rendering those graphics, both the central processor (CPU) and graphics processors (GPUs) can be utilized to increase performance. On the CPU side, rendering generally scales well with both clock speed and core count – but those specifications cannot be directly compared across different brand or generations of processor. Here at Puget Systems we do real-world testing to ensure we provide our customers with the right computer for their needs.
Intel has just released a trio of new Core X series processors, with higher core counts than anything they have offered in this line to date: the Core i9 7940X, 7960X, and 7980XE – with 14, 16, and 18 cores respectively. We ran these chips through a few tests in V-Ray Benchmark 1.0.6 to see how they perform with rendering. We had recently put out an article using this same benchmark that covered a wider range of processors, and at the time AMD’s Threadripper CPUs were found to be the fastest single-CPU option. This time we will focus just on the Core X series in comparison to Threadripper, to see if the new models can take the performance title back for Intel. For reference we included a dual Xeon system as well, to show how it compares to these single chip configurations.
To see how these different CPUs perform in V-Ray, we ran the free benchmark in CPU mode on the following configurations:
The main focus here is on the three new Core X (formerly code named Skylake X) processors, and specifically how they compare to AMD’s Threadripper models. In our last round of testing we found that the 16-core 1950X took the lead in single-CPU performance with Keyshot, as well as many other CPU-based rendering engines. Here are some details about how we conducted our testing, but if you just want to skip straight to the results then feel free to scroll past this section.
The results presented below are from V-Ray Benchmark 1.0.6, which is a free benchmark released by Chaos Group. It is designed to test CPU and GPU performance within V-Ray without requiring a full installation of that software. Since the focus of this article is on CPU performance we only ran the CPU portion of the benchmark, which gives a time in seconds for how long it took to render a single scene. The rendering can also be watched in real-time during the benchmark, and the benefit of additional cores can be seen visually that way.
It is also worth noting that there are some differences in the amount and speed of RAM across the various test platforms. We prefer to use the speed of memory that each CPU is rated for, according to its manufacturer. For the current crop of Core X and Threadripper processors that is DDR4-2666, while the older Xeon uses slightly slower 2400MHz memory. Some of these platforms could be run with even faster RAM modules, but that is pushing the memory controller built into the CPU past its rated speed – overclocking it, effectively. That may lead to slightly increased performance but we have also found it to lead to stability issues and higher rates of memory failure. For that reason we stick with the manufacturer specs when it comes to selecting RAM for our systems.
Here are the results for the various CPUs we tested in V-Ray Benchmark 1.0.6. The new Intel processors are shown in light blue:
Since rendering in V-Ray is a heavily threaded application, there is a clear spread between the different processors based on core count. The dual Xeon, with a total of 28 cores, definitely wins out – but that is a much more expensive system and is really just included here as a point of reference. Among the single CPU workstations, the new 14 to 18-core models from Intel take back the lead that AMD’s 16-core Threadripper had since its own launch last month. Intel has a 10-20% lead, depending on which Core i9 you look at, but it is worth noting that the 1950X is still less expensive… and outperforms the Intel processors which are in its price range.
It is also worth noting that AMD and Intel both have server-class processors with even more cores as well: up to 32 on AMD’s EPYC and up to 28 from Intel’s latest Xeon Platinum line. With their focus on the multi-CPU server segment those may not come into play for V-Ray, but if a manufacturer puts out a single-socket workstation motherboard that is compatible with either of those platforms then they may be worth a look.
Based on these results, it looks like the tables have turned since our last V-Ray CPU performance article. Intel took back the lead, but their new Core X processors are also substantially more expensive than AMD’s Threadripper. If you want the fastest single-CPU rendering speeds then go for the i9 7980XE – but if you want a great value, the 1950X has strong performance while costing $1000 less.
Puget Systems offers a range of powerful and reliable systems that are tailor-made for your unique workflow.
Configure a System!
Labs Consultation Service
Our Labs team is available to provide in-depth hardware recommendations based on your workflow.
Tags: AMD, Chaos, Core, CPU, Group, i9, Intel, Performance, Processor, Rendering, Threadripper, V-Ray
What is the difference between Mobile CPU and Desktop CPU?
For each new generation, Intel provides a Desktop version of the processor as well as a mobile version. We explain the difference between the two so you have all information in hand to make your choice!
Comparing Mobile CPU vs Desktop CPU
A desktop version will have a higher speed processing and better cache than a mobile CPU which will feature a reduced speed. Compare the two models below: the mobile core i7 CPU has the same clock rate as the Desktop i5 core. The Desktop has twice more cores than the mobile core i7 and a small improvement on the GPU Clock rated speed.
Intel Core i5
Intel Core i7
2. 7 GHz
It means that Desktop CPUs are more powerful. They can handle more devices: that’s why they are designed for full size PC or workstations like our Airtop.
Heat and Power Consumption
Due to their higher number of cores, computers with Desktop CPU are usually more powerful but therefore generate more heat and have a bigger power consumption than PCs powered by mobile processors.
As an example, our Airtop with desktop CPU draws from 65W whereas our fanless computers with mobile processors like Intense PC3 draws from 25W and Fitlet RM from 10W only.
Because computers with mobile CPUs generate little heating, it is easier to get a fanless design with smart passive cooling. Fanless computers using desktop CPU can also be cooled but they will require a different design as more heat is generated.
Mobile CPU for which applications?
We recommend mobile CPUs for applications requiring low processing performance such as firewall, CCTV client, thin clients, remote controlling, M2M communication, real-time monitoring, SOHO…
Of course, it depends on your overall requirements and needs: Ask our Team!
Articles related to this topic:
Operating System: the difference between 32-bit and 64-bit
Introducing Airtop! Desktop PC just got “cool” again
What is the Function of the BIOS on a mini PC?
Performance comparison of PCs and smartphones, including iPhone 11 / Sudo Null IT News
After the release of the iPhone 11 with SoC Bionic A13, once again there was a desire to compare its performance with a PC. A couple of years ago, Apple’s chips already outperformed the mid-range notebook segment. And since there is practically no progress in performance there, the new pocket gadget should now bypass the entire laptop fraternity and have a good “bite” on desktop systems.
Bypassed in many ways. Bitten. Details under the cut.
When looking at opinions about who is faster (smartphones or laptops), the most common option turned out to be: “how can a smartphone for 60 thousand be slower than a PC that is cheaper?” True, these opinions were not expressed on Habré. But technically savvy people, on the contrary, asked, they say, how can a baby with a TDP of 3-5 W get around monsters with a TDP of 65 W or more, despite the fact that they are produced according to similar technical processes?
Two different camps formed. I myself, being a systems engineer in the first VO, belong to the second. And I have an answer to the question about watts. But let’s get to the heart of the matter.
What ruler will measure
We will compare performance in the Geekbench 5 cross-platform test, which emulates the work of real user tasks such as archiving and encryption. How legitimate it is to compare different platforms in it is a good question. Let’s take it a little lower. And now I’ll just say that the creators of the test are pushing hard on this:
I use this test periodically. But the results for this post were taken from the official charts. In them, the creators put the average values from what gets into the database from users. Most often, such results turn out to be slightly underestimated, because users are not professional testers. During the test, some software may be running in the background, or power saving mode is enabled. However, we don’t care. The extreme lower values are already discarded there anyway. In addition, I do not have the goal of obtaining precision data. It is enough to outline a certain general picture.
First — Apple is cool, and over the past couple of years it has increased its lead over Qualcomm and Samsung with their licensed and finished armaments.
Second, the performance of top smartphones in office-consumer tasks has caught up with advanced laptops and good office PCs (see caveats below).
The third is single-core performance. It is she who is responsible for the responsiveness of the interface and the speed of applications, most of which are poorly adapted to parallelization.
Who’s who on the chart
Now let’s take a look at the interior of the test subjects. For convenience, I have collected everything in one plate.
If we combine these data with the performance chart, we can see that the limited thermal package does not allow all cores of mobile chips to thresh to their fullest. Additional restrictions are introduced by the big.LITTLE architecture, within which it is not always possible to simultaneously operate a high-performance on-chip cluster and an energy-efficient one.
Is it possible to compare different architectures?
A full-fledged comparison of processor architectures is extremely difficult, and I have no idea how to do it competently. ARM belongs to the RISC type, and x86 belongs to the CISC. With fewer instructions and fewer blocks, the ARM chip must execute individual instructions faster and more energy-efficiently. But as soon as it comes to performing complex functions, for which the x86 has prepared hardware blocks and instruction sets, ARM will smoke on the sidelines. But that’s in theory.
And there are different operating systems, different compilers. And it seems to me that the developers of Geekbench slightly scored on all this, simplifying everything to monitoring the performance of some typical tasks by the system, such as decoding jpg or assembling cached web pages. At the same time, they tried to optimize the code of these tasks for each system separately.
As a result, the happy owner of the latest iPhone can proudly say that his smart phone is able to open photos from the gallery as quickly as the top-end five-gig “stove” from Intel. But Geekbench no longer allows you to make more serious statements. However, for most everyday situations this is quite enough.
More details on their tests can be found in this pdf.
What are these comparisons for?
Three years ago, I was struck by the idea that with the growth in performance of mobile chips, they could encroach on the segment of inexpensive netbooks, a third of the price of which, at times, is Microsoft’s OS. With Google distributing its OS for free and only $1 per device to add Google Play services and more, the idea of capturing the lower segment looked quite realistic.
However, Qualcomm’s marketers have gone the other way, and for the past couple of years they have been trying to surprise the world with $1000 systems in which their top chips get along with Windows 10…
Links to charts
If you are interested in the average results of other systems, you can use the Gikbench online database. Here are direct links to automatically updated charts for Android, iOS and PC. In the same place, you can enter any keywords in the search (models of chips, smartphones) and see the results obtained by other users for these devices. The only thing is that you will have to filter inadequate options yourself.
and yes, the owners of top smartphones can be proud that they carry in their pocket a thing more powerful than most modern laptops. At least if we compare them in terms of work in everyday applications.
Server and desktop processors — what’s the difference for the user
Readers often ask which type of processor they should choose: server or desktop .
While choosing the option that provides the best bang for your buck and building a processor-based system may seem intuitive, it’s often worth taking a closer look.
«Server» and «Desktop» are not just categories of processors, but separate platforms of which the processor is a part.
We’ll quickly go over the computer platforms to better explain the difference between server and desktop processors.
What types of platforms exist
There are four main platforms:
- Mainstream / consumer sometimes referred to as desktop
- Workstation or HEDT (high end desktop)
- Professional workstation
There is no official definition of what each of these platforms consists of, but they will generally be divided into broad levels of form factors, hardware performance, expandability and specialization, and intended use case.
Comparison of different platforms
Desktop or consumer computer
Basic or consumer platforms are what you will use on a computer built for light workloads such as graphic design, word processing, browsing, and general day-to-day work and gaming.
Desktop platforms are also evolving to better handle multi-threaded workloads, making them a cheaper alternative to workstation platforms. One good example is CPU rendering, where this platform is starting to gain popularity due to the ever-increasing number of CPU cores.
Workstation platforms are suitable for more demanding workloads requiring high multi-threaded performance and better connectivity. By providing many compute cores and access to more, such as PCIe lanes, applications that are highly parallelized run great on this platform.
Workstation platforms fit into desktop form factors, making them ideal for desktop or home use where you actively work on the system.
The Professional Workstation platform offers many of the features you’d find on a high performance server platform, but the key difference is that it matches the desktop form factor.
This platform is ideal for applications such as rendering, simulation, or workloads that require access to more PCIe lanes, ECC memory, more memory capacity, or CPU security features than the Workstation/HEDT platform can provide.
The server platform is designed for reliability, flexibility and scalability. They are deployed primarily as rack-mounted units in data centers, allowing a large amount of processing power to be packed into a small space.
Server platforms are configured for a variety of purposes, from storing large amounts of data to resource-intensive applications with a large number of computing cores and memory.
They are also designed to operate 24 hours a day, 7 days a week for extended periods without instability or system failures, and are not intended for direct use.
Key differences between platforms
What does it have to do with platforms? You just wanted to know the difference between desktop and server CPU!
Bear with me a little longer — we’ll get there!
Processors are inevitably tied to their platform. You can’t insert CPU desktop to the server platform and cannot install server CPU to the desktop platform.
The difference between server and desktop processors lies in their platform! Let’s take a look at the key platform differences:
Platform Form Factor
One obvious difference between desktop and server platforms is their form factor. For both the server and desktop platforms, there are standards for measuring the space occupied by the system.
For desktop platforms, typical form factors are XL-ATX, E-ATX, ATX, M-ATX, and M-ITX, in order of size. ATX is the most popular option for desktop PC systems, followed by M-ATX and M-ITX.
The E-ATX form factor is often used in workstations and enthusiast systems. The larger size allows them to pack extra features, such as more PCIe and RAM slots, while still staying within the limitations of a desktop PC case.
Desktop cases also come in a variety of form factors. They are classified into Full Tower, Mid Tower, Mini Tower and Small Form Factor (SFF). Hardware compatibility for a particular case depends on its size: Full Tower cases support most sizes, while SFF cases only support M-ITX.
Most servers are available in either a tower or rack configuration. The tower configuration is similar to the desktop platform and is a good choice if you are running a small number of servers.
Rack servers are designed for use with standard 19-inch server racks, allowing servers to be placed vertically and save space. In addition to servers, you can also purchase rackmount storage arrays and network switches to add additional functionality to your installation.
Rack components are sized in height units from 1U to 6U for regular servers. Most server racks are 42U high, which allows you to fit quite a lot of equipment in a small space.
While some server motherboards fit common desktop PC form factors, many are custom made to fit more effectively into the server chassis you’ll be purchasing from the company.
Server platforms support memory with ECC or Error Correction Code for the entire set of processors. ECC support on desktop platforms is limited to professional workstation, workstation, and a select number of consumer platforms.
On consumer platforms such as AMD Ryzen on B550 and X570 motherboards, ECC support exists but has not been tested for use on servers or workstations and compatibility varies by motherboard manufacturers.
ECC RAM corrects memory corruption due to random bit switching, preventing system crashes and data corruption. This is important when you cannot afford system failures when using your computer 24/7 for an extended period of time.
Professional workstation servers and processors also support terabytes of RAM. For reference, a typical consumer platform supports no more than 128 GB of memory.
The number of available RAM channels also depends on the platform. Server and Pro workstations have eight channels, workstation platforms four, and consumer two. More memory channels improve the net bandwidth between RAM and CPU.
|Platform||ECC support||Memory channels||Max. memory|
|Worktable||Consumer/mainstream||No, some platforms have uncertified support||2||128 GB|
|Professional workstation||Yes||8||2 TB|
A unique feature of the server platform is multi-processor support . Multiple processors in one system not only increases the number of cores, but also gives access to more memory and PCIe lanes in one system.
With a single system with multiple processors, you save a lot of space and money that separate systems would take up. This is great, for example, for rendering farms that require many compute cores in a limited space.
Platform expansion and connection
Server and professional workstation platforms offer a large number of PCIe lanes. These PCIe lanes are required to add expansion cards such as GPUs, NVMe SSDs, SATA SSDs, hard drives, or network cards. To find out more, you can refer to our article on how many PCIe lanes you need.
|Platform||Maximum number of PCIe lanes|
Server platforms are very versatile in allocating PCIe lanes. Do you need a lot of NVMe drives? You have a server that does this by providing a full four lanes of PCIe connectivity per drive.
Need a large number of GPUs in one system? You also get this by supporting more GPUs than most desktop systems.
This just goes to show how versatile and customizable the server platform is. Servers are easily configured with multiple GPUs, NVMe drives, or hard drives while maintaining a compact footprint.
Of course, servers aren’t optimized for silent operation, so you don’t want to work directly on a server that’s under your desk. That is why they are usually hidden in a data center or a separate room.
Professional workstations are also customizable with an abundance of x16 PCIe slots. However, they are not as versatile as , mainly due to lack of space in the desktop form factor, which may force you to use expansion card adapters to get the most out of the platform.
Consumer systems have fewer PCIe lanes, which will limit your expansion to one or two GPUs and a pair of NVMe drives. Some specialized systems support multiple GPUs via a single 1x vertical channel for applications such as mining, although this will severely impact performance in non-mining workloads.
How processors relate to platforms
This was an overview of platforms. What do processors have to do with platforms?
Processors are platform specific and will only be compatible with that platform . Take AMD’s Ryzen and Epyc consumer/workstation series for example. There are no Ryzen processors for servers or Epyc processors for desktops.
The CPU of each series is made for its platform. Even processors that are identical on paper (for example, with the same number of cores and clock speed) will differ significantly depending on the platform they are running.
Which processor types are associated with which platform?
Here is a quick overview of which processor types are tied to which platform and their typical number of cores:
|Intel||Pentium / Celeron||2-4|
|Professional workstation||AMD||Threadripper Pro||12-64|
Desktop and server processors
Key differences between desktop and server CPUs
As discussed above, processors on their platforms offer several features that help distinguish the two platforms. We will now focus on the actual differences between the processors .
Processor clock speeds
Desktop and consumer PC processors have higher clock speeds, making them a great option for active and single-threaded workloads that cannot be easily parallelized, such as graphics design, many types of video editing, and computer games.
Intel and AMD also allow manual overclocking of many of their desktop processors, which provides additional performance at the expense of increased power consumption and reduced stability. The thermal power of a processor core increases exponentially with its clock speed.
Single processor desktop platforms in a well ventilated system with options for large air and liquid cooling solutions allow manufacturers to push higher core clock speeds without worrying about overheating.
Server platforms operate in confined environments where the only method of cooling is high velocity air through a small heatsink. Add to that multiple processors with more cores, and you’ll need lower clock speeds to keep temperatures stable and running for a long time.
Lower clock speeds also reduce power consumption, which may seem useless for a single processor. However, if you plan to use hundreds or even thousands of processors, this will have serious consequences, even with a difference of several watts per processor.
Server processors also need to run 24 hours a day, 7 days a week under heavy load, which greatly reduces their lifespan. This is why even server processors with low core counts run at lower clock speeds than comparable desktop processors.
Number of processor cores
While desktop platforms match servers by the maximum number of cores per processor, server processors have the unique advantage of being able to use multi-processor configurations .
The Intel Xeon Scalable is a great example of how easy it is to pack one server with multiple cores. Intel offers these processors as nodes that fit easily into a 2U chassis, offering up to 224 cores per server.
Processor price difference
Desktop processor prices are simple and usually proportional to processor clock speed and number of cores.
Server processors are usually made up of components with a large number of bins, which means that they run more stable and consume less power, making them more expensive even compared to desktop processors, which on paper have the same characteristics.
A CPU must have built-in logic to access all platform features, so it’s not surprising that server CPUs that have access to more memory channels or more PCIe lanes, for example, carry a higher price — all other factors are the same.
Correct CPU: server or desktop processor
When choosing between a server and a desktop processor, not only the processor is important, but also the platform .
Once you migrate to a certain platform, it will be costly to migrate to another due to the lack of compatibility between platforms.
Choose a desktop platform if you plan to actively work on your PC or workstation from your desk. Although the servers are in a tower configuration, they are not designed to operate as a standalone device and will offer much lower performance (low clock speeds, etc.).
Use the server platform if you plan to leave it unattended. Servers are easy to mount in equipment racks and allow you to expand with additional servers, storage arrays or network switches without taking up much space.
Be aware that rack servers are very loud, so they are not suitable for home or work use.
For applications such as render farms that require a lot of processing power, you can use the desktop platform as render nodes if you only need a few of them. For large-scale operations, the server platform’s space and energy savings make rack-mounted servers the best option.
Frequently Asked Questions
Rack or Chassis Server: Which is Best?
Tower Servers The is more of an entry level server for those who want to get started with a server or for environments where there are no server racks.