Like most software developers, Bitplane maintains a list of system requirements for Imaris that can be used to help ensure the hardware in your system will work with Imaris. However, most "system requirements" lists tend to cover the required hardware, not what hardware would actually give the best performance. Although Bitplane is better than most in that they list separate requirements for a minimum, recommended, and high-end system, we've taken the time to perform testing to determine what hardware runs Imaris the best. Based on this testing, we have come up with our own list of recommended hardware for Imaris.
After extensive testing and analysis of this software, we found that the most important information necessary to make an educated decision on hardware for Imaris will come from the type of analysis you will be performing and the size of the data sets with which you'll will be working.
- If your tasks will be primarily single threaded, you will get the best performance from the "standard" system because of the lower core count, but higher frequency, CPU (unless you are performing analysis on larger and/or more complex data sets, in which case, you will want more RAM so you'll need to move to the "high-end" system).
- If your tasks will be primarily multi-threaded ones, then you will get the best performance from the "high-end" system as it has more cores and also allows for much more RAM (up to 512GB) for any large and/or complex data sets you may need to analyze.
Due to the mixture of single and multiple core CPU tasks within Imaris, a high frequency CPU with a medium number of cores (4-8 cores) will provide the most balanced performance. [More Information]
Video Card (GPU)
NVIDIA Quadro cards give the best overall performance in Imaris. The Quadro M2000 4GB is a great option for the majority of users, although in some instances the greater power and larger VRAM capacity of the Quadro M4000 8GB makes it a better choice. [More Information]
Bitplane recommends between 4 and 64GB of RAM for Imaris, although we generally recommend a minimum of 16GB if possible. Depending on the size complexity of the data sets you will be using however, you may need even more RAM than the 64GB maximum Bitplane recommends. [More Information]
Storage (Hard Drives)
We recommend having a large, fast SSD as your primary drive paired with traditional hard drives for any necessary extended storage. [More Information]
Detailed Hardware Information:
When it comes to CPUs there are two main specifications that define the capability of a CPU:
- The frequency directly affects how many operations a single CPU core can complete in a second (how fast it is).
- The number of cores is how many physical cores there are within a CPU (how many operations it can run simultaneously).
Whether a higher frequency or higher core count is faster depends on how well a program is designed to take advantage of multiple CPU cores. In the case of Bitplane's Imaris, there are some tasks that can only use a single CPU core while some can take advantage of multiple CPU cores.
Examples of single core CPU tasks:
- Rendering or viewing the 3D model
- Surface analysis
- Filaments analysis
Examples of multiple core CPU tasks:
- Spots analysis
- Tracking analysis
Video Card (GPU)
The GPU (video card) plays an important role in rendering the 3D model, although often the CPU will actually be the limiting factor. The speed of your video card won't have any impact on how long it takes to perform an analysis, but it does make a impact on how smoothly you are able to rotate and view 3D models. We've tested a range of video cards using data sets provided by Bitplane and have found that NVIDIA Quadro video cards give the best overall performance.
If you would like to see the results of our testing, feel free to view the graph to the right. On interesting aspect we found during our testing is that in five of the six renders, the lower-end Quadro K2200 performed almost identically to the much more expensive Quadro K5200. The only exception was a model of blood vessels with a surface analysis applied in which the Quadro K5200 showed much better performance.
Most models do not require more than 2-4GB of VRAM (dedicated memory on the video card), but extremely large models may require 4GB or more.
The amount of RAM you need is going to depend on the size and complexity of the data sets with which you will be working:
- Smaller data sets and/or data sets with low complexity will need 16-32GB of RAM.
- Medium data sets and/or ones with some complexity will need between 32-64GB of RAM.
- Larger data sets and/or highly complex ones will need between 64-256GB of RAM.
Storage (Hard Drives)
For the hard drive, the most critical factor is simply that you have enough. Simply loading the data sets can take quite a while and faster storage (SSD's) will help reduce loading times. However, faster storage is often more expensive per GB so we suggest supplementing the SSD with larger capacity, more traditional, storage drives.