Calculator Inputs
The page stays in a single main content flow, while the input grid adapts to large, small, and mobile screens.
Example Data Table
These sample scenarios help you compare how latency, loss, and overhead change network quality across common operating conditions.
| Scenario | Capacity | Observed Throughput | RTT | Jitter | Loss | Overhead | Estimated Goodput |
|---|---|---|---|---|---|---|---|
| Branch office WAN | 500 Mbps | 410 Mbps | 22 ms | 2.1 ms | 0.3% | 7% | 379 Mbps |
| Cloud backup window | 1 Gbps | 720 Mbps | 28 ms | 3.5 ms | 0.8% | 8% | 652 Mbps |
| Congested remote link | 200 Mbps | 118 Mbps | 80 ms | 14 ms | 2.7% | 11% | 101 Mbps |
Formula Used
The calculator combines raw link rate, overhead, delivery quality, and time-based transfer data. When packet counters exist, loss is derived from those counters first.
How to Use This Calculator
- Enter the provisioned line capacity and choose the correct rate unit.
- Add the observed throughput, or leave it at zero to derive throughput from transfer size and duration.
- Supply RTT, one-way latency, and jitter to reflect the path quality you measured.
- Enter packet counters when available. They override manual loss because counter-based loss is usually more reliable.
- Set retransmission and protocol overhead to model headers, encapsulation, encryption, and repeated sends.
- Use payload size and concurrent users to estimate packets in flight and per-user throughput share.
- Press Calculate Performance to display the result block above the form, review the chart, and export the summary to CSV or PDF.
Frequently Asked Questions
1. What is the difference between throughput and goodput?
Throughput is total delivered rate, including headers and recovery traffic. Goodput is the useful application data rate after subtracting overhead, loss, and retransmission penalties. Goodput is usually the more practical metric for user experience.
2. Why does packet loss matter even when bandwidth looks high?
Packet loss forces retransmissions and breaks traffic smoothness. That means file transfers slow down, real-time calls become unstable, and application response feels inconsistent even if the raw link capacity still seems large.
3. When should I enter packet counters instead of manual loss?
Use packet counters whenever you have them from routers, switches, agents, or test tools. Counters create a more defensible loss percentage than guesses, so this calculator automatically prioritizes sent and lost packet values.
4. What does protocol overhead include?
Protocol overhead can include Ethernet framing, VLAN tags, IP and transport headers, tunneling, VPN encapsulation, and encryption-related bytes. Combine them into one estimated percentage when modeling the real payload efficiency of a link.
5. Why is bandwidth-delay product useful?
Bandwidth-delay product estimates how much data should stay in flight to fully use a path. It helps size TCP windows and buffers, especially on long-distance or high-speed links where undersized windows limit throughput.
6. Can I use this calculator for Wi-Fi or internet links?
Yes. The model is general enough for LAN, WAN, Wi-Fi, VPN, cloud, and internet paths. Just remember that bursty wireless conditions may cause bigger jitter swings than a simple static snapshot can fully represent.
7. Is the MOS value exact for voice quality?
No. The MOS result here is an approximation meant for planning and quick screening. It can guide comparisons, but codec behavior, concealment, routing, and burst loss patterns still influence real call quality.
8. What score should count as healthy?
A strong network often scores above 75, while excellent conditions push above 90. Scores below 60 usually suggest noticeable efficiency, latency, jitter, or loss problems that deserve deeper troubleshooting.