Independently Validated

Performance That
Speaks for Itself

Every claim backed by independent testing from Europe's foremost network testing facility. No marketing spin — just data.

Rigorous, Independent, Repeatable

Every performance figure on this page was produced by Broadband-Testing, Europe's foremost independent network testing facility, led by Steve Broadhead. These are not internal benchmarks or cherry-picked scenarios — they are structured, methodical evaluations conducted under controlled conditions using industry-standard equipment.

The test infrastructure employed Spirent Axon traffic generators to produce realistic, mixed-protocol traffic loads — hundreds of simultaneous FTP transfers, HTTP sessions, and VoIP calls running concurrently. WAN conditions were precisely emulated using iTrinegy WAN emulators, introducing calibrated latency, packet loss, and bandwidth constraints that replicate real-world scenarios from transatlantic fibre to satellite links. The result: data you can trust, from a source the industry respects.

TCP Acceleration Performance

Tested on a 100Mbps symmetric link with 100ms round-trip time — typical of a UK-to-NYC connection. Standard TCP settings unless otherwise noted.

Without Window Scaling

TCP's default window size caps throughput at approximately 5Mbps on a 100ms link — regardless of available bandwidth. This isn't a configuration error; it's a fundamental protocol limitation. ViBE eliminates this ceiling entirely, utilising the full 100Mbps from the first packet.

Standard TCP ~5 Mbps
5 Mbps
ViBE ~100 Mbps
100 Mbps

0.1% Packet Loss

Even a seemingly trivial 0.1% packet loss rate — well within normal internet conditions — triggers TCP's congestion avoidance algorithms, causing dramatic throughput reductions and extended transfer times. ViBE's protocol handles loss transparently, completing the same transfer in less than half the time.

Standard TCP — transfer time Baseline
Full duration
ViBE — transfer time < 50%
Less than half
25×
Faster Under 1% Packet Loss
ViBE vs standard TCP on the same 100Mbps / 100ms link

1% Packet Loss

At 1% loss, TCP enters a catastrophic feedback loop: every retransmission risks further loss, which triggers further backoff, which increases latency, which causes timeouts. Throughput can collapse to single-digit percentages of available bandwidth. ViBE sidesteps this entirely — its protocol is designed to maintain throughput irrespective of packet loss, delivering a measured 25× performance advantage over standard TCP.

Standard TCP ~4% utilisation
~4%
ViBE ~100% utilisation
~100%

Satellite Conditions: 500ms RTT + 0.1% Loss

Satellite links compound TCP's weaknesses: high latency magnifies the impact of every lost packet, and TCP's slow-start algorithm takes even longer to ramp up. This combination makes standard TCP virtually unusable for bulk transfers. ViBE maintains consistent, high throughput regardless of latency, completing transfers 7× faster than standard TCP under these conditions.

Standard TCP Baseline
Full duration
ViBE 7× faster
7× faster

Bidirectional Transfers

When TCP streams run in both directions simultaneously, they compete for the same congestion window, causing severe mutual interference. Upload traffic degrades download performance and vice versa — a constant reality in any office environment with cloud sync, backups, and real-time applications running concurrently. ViBE manages bidirectional traffic independently, with both streams completing cleanly in approximately 28 seconds without interference.

Standard TCP — bidirectional streams interfere Degraded
Mutual interference
ViBE — both directions clean ~28s each
Clean, ~28s completion

Latency Stability Under Load

Network latency isn't just about averages — it's about consistency. TCP under load causes latency spikes to 250ms or more as buffers fill and congestion algorithms react. These spikes destroy VoIP quality, cause POS timeouts, and make interactive applications unusable. ViBE maintains latency at a rock-steady 100ms (the baseline link RTT) regardless of traffic volume, because its byte-level QoS prevents buffer bloat from ever occurring.

Standard TCP — latency under load 250ms+ spikes
Spikes to 250ms+
ViBE — latency under load Stable 100ms
Stable at 100ms

Voice & Mixed Traffic Performance

Tested on an 8Mbps WAN link with Spirent Axon generating 500 FTP sessions, 500 HTTP sessions, and increasing VoIP call volumes simultaneously.

6 Concurrent VoIP Calls

Even at just six simultaneous calls — a trivial load for most businesses — standard networking already shows significant degradation. MOS scores drop to 3.4 (audible distortion), latency balloons to 225ms, and 4.4% of packets are lost. With ViBE, the same six calls achieve a perfect MOS of 4.0, latency drops to just 5ms, and packet loss falls to zero.

Without ViBE — MOS 3.4 / Latency 225ms / Loss 4.4%
MOS 3.4
With ViBE — MOS 4.0 / Latency 5ms / Loss 0%
MOS 4.0

40 Concurrent VoIP Calls

At 40 concurrent calls alongside 1,000 data sessions, standard networking collapses: MOS degrades to 3.1, 1,933 packets are dropped, and latency exceeds 211ms. ViBE restores order completely — MOS returns to 4.0, dropped packets fall from 1,933 to just 5, and latency is reduced to 7ms. The voice traffic is effectively invisible to the data sessions, and vice versa.

Without ViBE — MOS 3.1 / 1,933 dropped / 211ms
MOS 3.1
With ViBE — MOS 4.0 / 5 dropped / 7ms
MOS 4.0

80 Concurrent VoIP Calls

At 80 calls (160 channels), the standard network manages only 127 successful channels — nearly 20% of calls fail entirely, with those that connect suffering degraded quality. ViBE handles all 160 out of 160 channels flawlessly, with every single call maintaining toll-quality audio alongside the full data workload.

Without ViBE — 127 of 160 channels successful
127 / 160
With ViBE — 160 of 160 channels successful
160 / 160 — flawless

180 Concurrent VoIP Calls

Pushing to 180 calls (360 channels) makes the disparity even starker. Standard networking manages only 200 successful channels with a degraded MOS of 3.0. ViBE handles all 360 channels with a perfect MOS of 4.0 — nearly double the capacity at dramatically higher quality.

Without ViBE — 200 of 360 channels / MOS 3.0
200 / 360
With ViBE — 360 of 360 channels / MOS 4.0
360 / 360

180 Calls at 50% Bandwidth (4Mbps)

In perhaps the most remarkable test, bandwidth was halved to just 4Mbps while maintaining 180 concurrent calls and the full HTTP session workload. ViBE still delivered every single call — all 360 channels — at MOS 4.0, while simultaneously completing all HTTP sessions successfully. This demonstrates ViBE's byte-level QoS at its most potent: absolute priority management that ensures voice quality is maintained even when bandwidth is severely constrained.

ViBE at 4Mbps — 360/360 calls + all HTTP sessions
MOS 4.0 — all traffic served
45:1
Latency Reduction Factor
Measured across mixed voice and data traffic under load — from 225ms down to 5ms

Extreme Condition Performance

400
Simultaneous Calls
On a link that couldn't handle 30 without ViBE

ViBE carried 400 simultaneous voice calls on an 8Mbps WAN link where standard networking failed at just 30. This isn't a theoretical claim — it's a measured result from independent testing that demonstrates ViBE's ability to fundamentally transform what's possible on constrained infrastructure.

📡

448kbps ADSL

Over a 448kbps ADSL connection — a link most would consider barely adequate for browsing — ViBE carried 40+ simultaneous voice calls alongside data traffic with zero packet loss. Every call maintained toll quality.

📞

256kbps Extreme

Even at just 256kbps — a link that most SD-WAN vendors would consider unusable — ViBE sustained 28 concurrent voice calls. This capability makes ViBE uniquely suited for satellite deployments, rural connectivity, and emergency communications where bandwidth is at an absolute premium.

🛡️

RAIN Mode

ViBE's Redundant Array of Inexpensive Networks mode was tested against severe packet loss conditions. At 10% network loss, RAIN reduced effective loss to just 1%. Even at a catastrophic 50% packet loss rate — a link most would consider completely broken — RAIN maintained effective loss at just 8%, keeping applications functional.

Any company from SMB to ISP who wants to optimise bandwidth consumed by a mix of VoIP and data should invest in Voipex's ViBE technology. It really is that much of a no-brainer.

Steve Broadhead
Broadband-Testing — Europe's foremost independent network testing facility

Experience It Yourself

These results were produced on standard hardware using real traffic profiles. We're confident ViBE will deliver the same transformative performance on your network — and we're happy to prove it.