With hundreds of updates, so much has changed since version 5 was first released in June 2019. But 6.5 years later, it’s time to take Blue Iris to the next level with this major upgrade to version 6. Here’s what you can find today:
The UI takes on a decidedly more “dark mode” aesthetic. Attention was paid to contrast, legibility and consistency. Initial changes are in place to offer user-defined color palettes.
The ability to keep track of video assets, search them efficiently, and export them for sharing when required is a critical aspect of high-performance video security software. The DB code has been completely rewritten to more effectively and reliably meet this key requirement. Simplicity is the word as the DB now consists of only two files, a smaller file with critical data for a backup or rebuild, and a much larger (and replaceable, less valuable) file consisting primarily of video preview images.
You may never need to add 128 cameras, but many systems have felt the need to go beyond the previous limit of 64 for quite some time, and this is now possible.
Gone are the days when you had to use an external tool like STunnel for this basic modern-day security feature. Traffic in and out of your server via LAN or Internet can be fully encrypted using a self-generated certificate, or for a more seamless experience, a real CA-signed certificate (globally trusted, not self-signed).
The software is now being built on the latest development platform, using the latest Microsoft Visual Studio and MFC code libraries, along with the latest in video processing libraries.
As H.265 has gained in popularity and maturity, it is now increasingly leveraged for functionality such as direct-to-wire streaming (saving decoding and re-encoding processing time as video may be sent directly from camera to browser or phone app) and exports (with possible conversion from H.264 to H.265).
Encoding H.264 or H.265 video using either the Intel QVS/OneVPL platform or the Nvidia NVENC platform is now officially out of its version 5 “beta” designation and has been fully integrated and tested with compatible hardware. Video encoding is perhaps the most “heavy lifting” the software can be asked to perform, and now it may be assisted by your GPU. Decoding has also received updates to allow the use of DXVA2 and Direct3D12 technologies along with improved Intel and Nvidia integration.
The software can today control basic Sonoff (eWeLink) devices that have been configured for “DIY mode”. Plans are in place to continue this momentum by adding support for Zigbee devices and a basic “home layout” view.
As with version 5 before it, development will be continuous and often rapid, largely based on user requests and feedback—please keep them coming! However, there are already some major plans:
As ARM processors become more compelling and prevalent, we have chosen to shelf support for legacy 32-bit CPUs and instead focus effort on the current AMD/Intel 64-bit architecture as well as ARM64. This will be a requirement for running this software on new Apple silicon such as the M5 chip via Parallels or potentially an Apple native variant.
As remote access remains a struggle for many due to networking security and architecture, the use of reverse-proxy services like NGROK were an interesting, although technical, shortcut. NGROK has recently announced the end to any free services they may still provide. However, a roadmap exists for us to develop and offer a similar service using a public network (likely Amazon AWS), allowing users to very simply engage remote access. The technical details and internal costs are still being worked out, but this will be an exciting and welcome addition.
Long overdue, the phone apps will receive much love in early 6.x development.
Many users today rely on third-party services for AI processing of alerts for the presence of people, vehicles, wildlife, and associated facial and license plate recognition. These work great, until they do not, as they rely on image encoding, windows sockets interfaces, and python and other external libraries, generating many points of failure. It is a short-term goal (6.0.x) to offer a “built in” option, simplifying the interface and generally making AI a more dependable and internally-manageable service.
Blue Iris development has not only been a “job” but it has been a passion, something we use ourselves daily and we remain excited to see where it all goes!