What were the best hardware configurations for a RealityCapture photogrammetry workstation as 2020 drew to a close?
In the past we have found that RealityCapture can perform better on high core count processors when SMT is turned off. With the release of AMD's Ryzen 5000 Series processors in late 2020, those became some of the fastest chips available for RealityCapture - and the top-end models in that family sit right on the edge of where we found disabling SMT to be helpful in our older tests. So does turning SMT off help or hinder these new Ryzen CPUs?
In past testing, we found that RealityCapture benefited somewhat from adding a second video card. Is that still the case with the latest NVIDIA GeForce RTX 30 Series cards, and the recent update to RealityCapture 1.1 Blaze?
With the launch of the Ryzen 5000 Series, AMD has brought their updated Zen 3 microarchitecture to mainstream desktop computers. They claim these CPUs have substantial performance per clock improvements over the previous generation. Does that impact photogrammetry processing in RealityCapture? And if so, how do these new chips stack up against other AMD and Intel models?
NVIDIA's first GeForce RTX 30 Series cards launched in September, and now the RTX 3070 has joined its bigger siblings. How does it stack up to the RTX 3080 and 3090? And how do they all compare against the previous generation of cards? Here we look at how they all perform in RealityCapture.
NVIDIA's GeForce RTX 30 Series cards are here, with NVIDIA boasting significant performance gains over the previous generation. The RTX 3080 launched last week, and now with the RTX 3090 released today we can compare these models to each other as well as the older 20 Series find out how large those gains are in photogrammetry applications like RealityCapture.
The RTX 3000 series cards are here, with NVIDIA boasting significant performance gains over the previous generation. With the RTX 3080 now launched, we can find out how large those gains are in photogrammetry applications like RealityCapture.
AMD and Intel have both released small revisions to their latest-gen consumer CPU lines, in the form of the Ryzen XT models and Core i9 10850K. On paper these appear to be identical to current products except for very slight adjustments in clock speed, but how do they stack up in real-world photogrammetry applications?
A short article covering some of the best computer system configurations for photogrammetry in RealityCapture.
We've observed that some photogrammetry applications seem to perform better with lower core count processors, so we wanted to look at whether Hyperthreading and Simultaneous Multithreading could be negatively impacting performance in these programs. Next up is RealityCapture.
Intel's new 10th Gen Core processors are out now, with an increased number of cores and very high clock speeds. How do they stack up against AMD's Ryzen chips and other current models for photogrammetry workloads in RealityCapture?
When AMD launched the 64-core Threadripper 3990X, some reviewers reported that performance of this 128-thread beast was hindered by running a normal version of Windows 10 Pro - and that instead using Windows 10 Pro for Workstations or Windows 10 Enterprise gave better results. We have investigated that claim using our in-house Pix4D and RealityCapture benchmarks to see if the choice of operating system could impact our customers.
AMD has launched a new top-end processor, equipped with a whopping 64 cores: the Threadripper 3990X. After extensive testing we are ready to report on how this new CPU performs in RealityCapture, compared to several of AMD and Intel's other current desktop chips.
Picking the processor, or CPU, is one of the most important decisions when building or buying a workstation - but there are dozens of options to pick from at any given point, with varying specifications and price points, so making the right choice isn't always clear-cut. We've tested all of the latest processors from Intel and AMD across three popular photogrammetry applications to help you know what the best choice is for your next system.
Intel and AMD have both launched new lines of high-end desktop processors, with different approaches to increasing value. AMD's 3rd Gen Threadripper CPUs are based on a newer microarchitecture, bringing better performance for around the same price as previous models. Intel, on the other hand, focused on reducing price while still using the same underlying technology as their last series. We tested both to see which approach gives the best results RealityCapture.
AMD's new Ryzen 3rd generation processors feature both an increase in core count and per-core performance, allowing them to rival Intel's mainstream Core processors in many professional applications. The first chips in this series did well with RealityCapture in our previous testing, but didn't quite beat Intel's Core i9 9900K, so now we are going to see how the top-end Ryzen 9 3950X fares.
AMD's new Ryzen 3rd generation processors feature both an increase in core count and per-core performance, allowing them to rival and sometimes beat Intel's mainstream Core processors in professional applications. In this article we are going to take a look at how these new CPUs handle RealityCapture, which benefits from both core count and clock speed at various points throughout its workflow.
RealityCapture, like other photogrammetry applications, is built to take a batch of photographs and turn them into digital, 3D models. The algorithms used during that process are designed to be "out of core", meaning that not all of the data has to be loaded into system memory (RAM) at the same time - allowing for full processing without requiring a ton of available memory. Having more RAM can still be beneficial, though, so we decided to test how much impact it has on performance.
RealityCapture, like other photogrammetry applications, is built to take a batch of photographs and turn them into digital, 3D models. The many steps involved in that process can take a lot of time, and utilize both the CPU and GPU at different points. We recently put together a benchmark tool for RealityCapture, and after looking at processor performance last week we are now diving into a comparison of the current NVIDIA GeForce and Titan video cards.
RealityCapture, like other photogrammetry applications, is built to take a batch of photographs and turn them into digital, 3D models. The many steps involved in that process can take a lot of time, and utilize both the CPU and GPU at different points. We recently put together a benchmark tool for RealityCapture, and in our first article focusing on this software we are going to take a look at how Intel's Core i7 and i9 processors stack up against AMD's Threadripper chips.
Over the last couple of years, several new vulnerabilities have been found in modern microprocessor architectures. Many of these have been related to speculative execution techniques used across most of Intel's product line, and a patch was released for Windows 10 last week which was intended to address some of those weaknesses. Unfortunately, it has also reduced performance in some applications - which this article will demonstrate by comparing processing speed in three photogrammetry programs before and after the update.
Here at Puget Systems, we have put together a benchmark utility for RealityCapture which measures system performance for photogrammetry by running two small projects - a model and a map - and tracking the time each step takes to process. This benchmark is freely available to download, though running it requires a valid installation of RealityCapture.