Knowledge Base

System Requirements

For optimal performance and functionality of EthoVision, an up-to-date professional workstation is strongly recommended. EthoVision 19 is tested on Windows 11 Pro 64 bit, version 25H2 on a Desktop Dell Pro Max Tower T2. Laptop compatibility testing is pending. For this system we guarantee compatibility.

Recommended Computer Requirements

• Processor (CPU) – 14th Gen Intel Core i7-14700, 20 cores, 2.1–5.4 GHz
• Internal Memory (RAM) – 32 GB for desktop PCs and 16 GB for laptop PCs
• Graphics Card (GPU) – NVIDIA RTX 5070, 12 GB GDDR7
• Connectivity – USB 3.0 ports or an additional Ethernet network interface adapter, depending on the cameras
• Hard Disk – 1 TB free space (to store experiments including video files)

Performance will depend on the functionality used. Some processes, such as live acquisition or tracking multiple animals within one arena, will require more from the processor and may influence the working speed.

Requirements Deep Learning

Deep learning relies on the CUDA computing platform. For this reason, the graphics processing unit (GPU) driver must support CUDA version 13.1. For tracking live or from video files, we tested and guarantee compatibility with:

• Desktop Computer (Dell Pro Max Tower T2) – NVIDIA GeForce RTX 5070, 12 GB GDDR7, driver version 581

Laptop compatibility testing is pending.

When a driver version is not sufficient for supporting CUDA, EthoVision gives the message in the Experiment Settings: "The graphics card driver is not installed or the graphics card does not support CUDA". We recommend downloading and installing the latest driver from the card's manufacturer.

Deep Learning Limitations

• One Subject Tracking – Up to four arenas, live tracking or from video file. No minimal trial duration required. Maximal trial duration tested: 72 hours.
• Two-Subject Tracking – Up to four arenas, only tracking from video file. Tracking of hooded rats not supported. It is recommended to perform trials of at least five minutes. Maximal trial duration tested: 1 hour.
• Subject Size – The apparent length of the subjects should be at least 10% of the size of the arena. We recommend that the apparent length of the subject's body is at least 120 pixels for rats and 50 pixels for mice (nose to tail-base).

Apparatuses

EthoVision's neural networks have been trained with videos of:

• One Subject Per Arena, Subjects of Uniform Color – Open field (regular or with round objects in it), PhenoTyper (with or without bedding material), Elevated plus maze, Three-chamber social approach cage, Barnes maze, Fear conditioning cage with floor grid, Y-maze (with no objects), and a maze with multiple chambers and openings.
• One Subject Per Arena, Hooded Rats – Open field (regular or with round, rectangular or triangular objects in it), PhenoTyper (with or without bedding material), and Elevated plus maze.
• Two Subjects Per Arena – Open field (with no objects), PhenoTyper (with or without bedding material), and home cage.

Requirements Behavior Recognition

When using Rat or Mouse Behavior Recognition, take the following requirements into account:

• Behavior Recognition requires the detection of the nose point, center of gravity and tail base. The subject's apparent length (nose to tail base in walking position) must be at least 60 pixels for rats, and 55 pixels for mice. In all cases its apparent length must not exceed half the arena size.
• Rats must be older than approximately three weeks. In all cases subjects must be able to walk.
• The animal needs to be uniform in color. (Hooded rats have not been tested.)
• You can track one animal per arena. When tracking live, you can use only one arena.
• Behavior Recognition is tested in open field and home cage situations without a shelter. Using a shelter or other objects in your arena can influence the results.
• It is preferred that the animal is visible in the arena at all times.
• Use a top-view camera position and lighting from the top (instead of backlight).
• Slow changes in the light during the course of a test are not a problem. Moving spotlights can reduce the reliability of the detection.
• Video frame size must be greater than 352 × 288 pixels per arena.
• The frame rate of the recorded videos can only be a multiple of the sample rate. The sample rate must be set between 25 and 31 samples per second.
• The video image must not be overexposed. Details of the subject's fur must be visible.
• Video compression should be as low as possible.

Source: EthoVision XT 19 Technical Specifications - Rodent, Noldus Information Technology