End-to-end encryption
AES-256-GCM with ECDH key exchange. Traffic is never decrypted at a relay.
A remote desktop built on a native C++ core. The encryption, video encoding and network layers are our own, designed to hold up on a poor connection.
AES-256-GCM with ECDH key exchange. Traffic is never decrypted at a relay.
H.264 encoding with GPU-based screen capture via DXGI. No frames are produced while the screen is unchanged.
Hybrid UDP/TCP transport, XOR FEC and adaptive bitrate. Packet loss does not drop the session.
A Windows desktop client, a browser client that needs no installation, and mobile.
Remote desktop is a solved category with entrenched products in it. AnyDesk and TeamViewer carry years of deployment, support organisations and compatibility work behind them, and we are not going to claim trviewer displaces them for a company already running one at scale.
What we can be specific about is what is inside ours. The encryption, the video pipeline and the network transport are code we wrote rather than a thin wrapper around somebody else’s sample. The rest of this page is about those three layers, because they are the only honest reason to look at a fourth remote desktop.
A session is encrypted with AES-256-GCM and the key is agreed directly between the two endpoints by ECDH. The relay in the middle forwards bytes it cannot read: traffic is never decrypted there.
That is a constraint rather than a slogan, and it takes options away. A relay that cannot read the stream cannot transcode it for a weak client, cannot cache anything, and cannot inspect a session when you want to debug one. We took the trade because a support tool with reach into every desktop in an organisation is precisely the wrong place to put a machine that can see the pixels.
Screen capture runs through DXGI Desktop Duplication, which hands frames over from the GPU instead of reading the desktop back through the CPU. Encoding is H.264.
The part that matters more than either is what happens when nothing moves. A desktop under a support session is static most of the time, because somebody is reading, or talking, or thinking. No frames are produced while the screen is unchanged, so an idle session costs almost nothing in bandwidth or CPU and the budget is still there when the screen finally does change.
Transport is hybrid UDP and TCP with XOR forward error correction and an adaptive bitrate. UDP carries the video, because a late frame is worthless and asking for it again only makes the next one later; TCP carries what must not be lost.
Forward error correction means a certain amount of loss is repaired from data that has already been sent rather than by requesting a retransmission. That is the difference between a session that degrades on a bad line and one that stalls on it. Packet loss changes the picture; it does not drop the session.
This is the layer that decides, more than any other, whether a remote session over a mobile connection is usable or infuriating.
The desktop client is Windows and Qt6, the browser client runs over WebRTC with nothing to install, and there is a mobile client. They speak the same protocol rather than being three products that happen to share a name.
The browser client is the one that changes how support actually goes. The person being helped does not have to be talked through an installer and a permission dialog before anything can be fixed, which is the step where a lot of remote support calls quietly fail.
A session carries multi-monitor support, file transfer and a shared clipboard, and machines can be configured for unattended access so a server or a workstation is reachable with nobody sitting in front of it.
These are table stakes rather than differentiators. They are listed because their absence makes a remote desktop unusable in practice, not because they are the interesting part.
trviewer is our own product and we run it, but the honest position is that it is young against incumbents with a decade of edge cases behind them. If what you need is a remote desktop already deployed across ten thousand endpoints in your industry, that is not an argument we can win today.
The case for looking at it is narrower. You want to know exactly what the encryption does, where the keys live and what happens to the stream in the middle, and you want the people who wrote those layers to be reachable when you ask. That is a real requirement for some buyers and irrelevant to most, and we would rather say which one you are than pretend everybody needs it.
Mainly in what we can tell you about the internals. Sessions use AES-256-GCM with an ECDH key exchange, the relay never decrypts traffic, and the capture, encoding and transport layers are ours. We are not claiming a feature-for-feature match with products that have been deployed for years.
No. The key is agreed between the two endpoints, so the relay forwards bytes it cannot read. The same decision also means the relay cannot transcode or cache the stream.
Not necessarily. There is a browser client over WebRTC with nothing to install, alongside the Windows desktop client and mobile.
Transport is hybrid UDP and TCP with XOR forward error correction and adaptive bitrate. Loss is repaired from data already sent where it can be, and the bitrate adapts, so the session degrades instead of dropping.
Yes. Unattended access is supported, along with multi-monitor sessions, file transfer and clipboard sharing.
It keeps frames off the CPU path, and the capture produces nothing at all while the screen is unchanged. A support session spends most of its time watching a static desktop, so an idle session ends up costing very little.
A short call is enough. We run demos for our products and a free discovery call for services.
If you would rather not fill in a form, call us or write on WhatsApp.
bilgi@mnethic.comPalandöken, Erzurum