Troubleshooting
Why Is My Download Speed Slow?
Slow downloads can involve units, Wi-Fi, competing traffic, a device or a remote service. Use a controlled comparison before changing your router or plan.
First establish whether the file is actively downloading, waiting, or being processed locally. A progress bar can include waiting and local processing as well as network transfer. One slow application is not the same as a slow connection across several services. Compare equivalent measurements before blaming Wi-Fi, the provider or the server.
Compare the right units and reference values
- Mbps or Mbit/s means megabits per second; MB/s means megabytes per second. Eight bits make one byte. Check the application's label, including whether it uses MB/s or MiB/s.
- 100 Mbps is theoretically 12.5 MB/s before overhead. A displayed 10 MB/s equals 80 Mbps, not 10 Mbps. These are unit conversions, not promised download rates.
- For a 1 GB file expressed in decimal units at a constant 100 Mbps, the ideal transfer time is 80 seconds; at 10 Mbps it is 800 seconds, or 13 minutes 20 seconds. Overhead and processing can extend this.
- Read the download values for your actual plan. An advertised maximum is not a prediction for every file. The adapter's link speed is also different from measured Internet throughput.
Make a repeatable baseline
Use the same device, browser and test destination, preferably with Ethernet. Pause your own large transfers and backups and note the time and server. Run a speed test between file-transfer attempts, not alongside the download being compared, since they compete for capacity. Observe the app's transfer state and displayed unit separately. Testing uses traffic deliberately, so choose a time that does not disrupt important calls.
Use the pattern to choose the next check
- Ethernet is faster than Wi-Fi on the same device: compare wireless coverage at the usual location and nearer the router. This is a clue about that path, not proof of a particular defective part.
- Only one service is slow: compare another trusted, permitted download source and check the affected app's rate limit, queue and service status.
- Only one device is slow: compare another device under similar conditions and inspect the affected device's resource usage and transfer processing. A browser change helps narrow the issue, not exclude every software cause.
- Several services and devices are slow over Ethernet: repeat with recorded targets at another time and keep the results. Router, access connection and wider network paths remain possible causes; evening timing alone proves none of them.
VMAXCHECK uses M-Lab NDT, a single-stream test to a particular measurement server. It reflects the device and path used at that moment, not the speed of every file host. Different test methods or destinations can differ. Record whether a VPN or proxy is involved because it changes the path; for a managed connection, ask the administrator about comparison conditions rather than bypassing required protection.
Change what the comparison supports
- Check the app's documented transfer limits and scheduling. In OneDrive, rate settings or pause options may be available; work settings can be managed by IT. Follow the relevant app's instructions.
- For a local bottleneck, inspect coverage or the actual capacities of the port, adapter and cable. Compare a known working compatible cable where appropriate. Router age alone is not a reason to replace it.
- Save router events before a manufacturer-guided restart. It interrupts other users and calls and is a test, not proof of the cause. Change one setting at a time and preserve the previous configuration.
- Repeat the same speed and real-transfer comparison after a change. Contact the service for a persistent isolated problem, or the network provider for problems across several devices and services.
Give support times, connection type, router and device models, actual plan, units, test targets and servers, background load, affected apps and before/after results. Remove private account details and credentials from logs. An unusable connection need not wait several days for help; repeated comparable results improve the evidence without guaranteeing a particular support procedure.