NetDiag+

NetDiag+ 1.4.10, now on Android too: a report for your provider, an evening re-check and a widget

Health Score tells you who is at fault in one run. But one run is rarely enough: a line that is fine at lunch can fall apart at nine in the evening, and your provider needs to see that pattern, not a single number. This release is about the second run and what you do with the results. It ships as 1.4.10 on iOS and as 1.0.3 on Android.

A report for your provider

Once you have run Health Score at least twice, one button turns your checks into a report, as text or PDF. It opens with a summary:

Below the summary is a table of up to 30 recent checks: score, idle latency, jitter, loss, bufferbloat grade, speeds and who was flagged. A short note at the end explains how each check measures and which time zone the times are in.

The report is in English, so it reads the same at any support desk. It is the document the evidence guide describes, built from checks you have already run.

Check again tonight

Home lines are at their most crowded in the evening. If you tap “Remind me” after a daytime check, you get one reminder at 21:00; after an evening check, one at 13:00 the next day. It comes once and does not repeat.

The Health Score screen now also lists your last five checks and shows the daytime and evening averages side by side, which is the comparison that matters.

A home-screen widget

A small widget shows your last score, who was at fault and how long ago the check ran. Tap it to open Health Score.

No more false A+ where the test server is throttled

To measure bufferbloat, the app fills your line with a download and then an upload from Cloudflare’s test server, and watches whether latency rises while it does. That only works if the data actually arrives.

Since June 2025, connections from Russia to Cloudflare have been throttled so that only about the first 16 KB of each response gets through. The test still looked successful, but almost nothing was downloaded, so latency stayed flat and the line was graded A+.

Now, if the download stays below 1 Mbps, the upload below 0.5 Mbps, and latency does not rise, the run is shown as ungraded: the test server could not be loaded from this network. Health Score then leaves its load and speed points neutral instead of counting them as perfect. And if the server refuses part of the download, that is no longer mistaken for a slow line. The bufferbloat guide now explains how to read an ungraded run and what each grade means.

In Russia and mainland China, the global reference servers behind the idle check can give misleading numbers. There the check now also asks public DNS servers inside the country: Yandex DNS in Russia, AliDNS and DNSPod in China. Latency is taken from whichever server answers best, so a local server can only correct the number, never make it worse.

Ping, Traceroute and MTR

Now on Android

NetDiag+ has been on Google Play since 1 October. Health Score arrived there on 2 October with 1.0.1, and Android 1.0.3, released alongside iOS 1.4.10, has the report, the evening re-check, the widget and the throttling fix. It needs Android 10 or later and speaks the same 13 languages. Ping uses Android’s unprivileged ICMP socket, so there is no root and nothing to configure.

Free · 31 tools · one-time purchase · no subscription

← All release notes