Sahaar › Why prayer times differ

Why prayer times differ between apps

Published 26 August 2026 · Sahaar requires iOS 26 or later

Why do two apps in the same city differ by ten minutes?

Because Fajr and Isha are defined by twilight, not by the clock. Each authority states how deep below the horizon the sun must be, and one degree moves the time by four to six minutes in the middle latitudes — more as you go north.

Take the two extremes: the Egyptian method and ISNA are 4.5° apart at Fajr, and in London in March that is over half an hour. Neither app is broken — they follow two different rulings, and neither will tell you which one your mosque uses. That is why Sahaar keeps the method name on screen next to the times.

Before you blame the method, though, rule out two smaller causes: times shift with longitude minute by minute, so a timetable printed for the provincial capital will not match a village forty kilometres west, and a device clock set by hand drifts quietly, taking every prayer with it.

Why is Asr an hour later in some apps?

Because of the shadow measure — the largest disagreement in the day. The standard measure (asr al-awwal, 1× shadow) begins Asr when an object’s shadow equals its own length, on top of its noon shadow; the Hanafi measure (asr al-thani, 2× shadow) waits until the shadow is twice the object’s length, which puts Asr later. How much later depends on where you are and the season, which is why an “Asr is wrong” report is nearly always a question about the measure.

The 1× measure is not one school’s alone: it is the view of the Shafi‘i, Maliki and Hanbali imams, and of Abu Yusuf and Muhammad al-Shaybani within the Hanafi school, which is why Sahaar labels it “Standard (1×)” rather than by a school name. Which measure applies where is not a guess either: in August 2026 we checked what each authority publishes against both measures, and the two were far enough apart that there was no doubt which one each timetable follows. That is what the table below records.

Two rows surprised us: Sarajevo publishes the standard measure although the Balkans are widely assumed Hanafi, and “the West is on the 1× measure” is wrong twice over, since the UK and South Africa both publish Hanafi. Sahaar suggests the measure for your country; you can change it under Settings › Prayer time calculation › Asr — except in Türkiye. Diyanet’s timetable uses the standard measure and that is when the adhan is called, so the Türkiye method follows the timetable instead of offering a second measure that would not match the adhan.

What happens at high latitude, when Fajr never arrives?

Above roughly 48°, for a few weeks around midsummer, the sun never sinks far enough below the horizon. There is no moment when it is 18° down, so Fajr cannot be calculated — a rule has to stand in for it. Three are in common use: the middle of the night, one seventh of the night, and an angle-based rule. Sahaar picks the rule from your method and location; there is no setting for it. The Türkiye method keeps the real angle for as long as there is one and switches to one seventh of the night when there isn’t; the other methods fall back to the middle of the night — except in the United Kingdom, which is covered below.

The rule makes a real difference. With the midnight rule, a June night in London gives Fajr 01:02 and Isha 01:03 — the arithmetic is right, but that is Isha and Fajr one minute apart, in the middle of the night. That is why the Türkiye method switches to one seventh of the night: 03:40 and 22:26.

Diyanet’s practice further north is often described as the “45-degree rule”: working out Fajr and Isha as though you were at 45° latitude. We tried it and checked it against Diyanet’s published timetables for cities abroad; it did not hold. What the timetables actually show is a real 18° Fajr wherever the angle still occurs, and an Isha angle near 16.1° abroad against 17° inside Türkiye. Over 32 days per city, Isha came out at +0.2 minutes in Cologne and −0.1 in London.

The United Kingdom is a separate case. London’s Unified timetable gives Fajr 04:25 and Isha 21:14; Muslim World League angles with the midnight rule produce 03:50 and 22:04 — three quarters of an hour out at Isha. The Moonsighting Committee method lands within about four minutes, so that is what Sahaar suggests in the UK, with the Hanafi (2×) Asr measure.

Which authority rules does Sahaar apply?

Angles alone are not enough. Several authorities publish a rule on top of their angles, and an app that skips it shows the wrong time in that country on those days. Sahaar applies them, and each one was checked against the authority’s own timetable:

• Umm al-Qura in Ramadan. Isha is 120 minutes after Maghrib instead of 90, with the month read from the Umm al-Qura calendar itself. Without it, Isha in Riyadh runs half an hour early for a month.
• Canada. 13° and 13°, following the Fiqh Council of North America’s 2017 position, not the ISNA angles used further south.
• Indonesia. Kemenag’s ihtiyat — a couple of minutes added to most prayers, on top of the Singapore angles.
• South Africa. Karachi angles, the Hanafi Asr measure, and three minutes added to Maghrib.
• France. The UOIF position of 12° and 12° — of nine Paris mosques checked, none published Muslim World League times.
• North Africa. Algeria 18°/17°, Tunisia 18°/18°, Morocco 19°/17°, not all three folded into the Egyptian angles.

Those rules, together with the regional angle sets, are why Sahaar comes with eighteen methods rather than the familiar six.

When does a published timetable replace the calculation?

Whenever an authority publishes one and its licence allows us to use it. A calculation matched to an authority’s own rules lands within a minute or so; the authority’s own table is the same to the second. Singapore is the first: MUIS publishes its timetable as open data, and Sahaar serves it as a yearly file from its own server. The request carries the provider, the city identifier and the year — never your coordinates — and if the file hasn’t arrived yet, the times are calculated in the meantime.

When a table is in use the label says so: “MUIS timetable”. If it disagrees with Sahaar’s own calculation by more than a minute, the table is set aside, the calculated time is shown, and the label goes back to the method name. The same rule is written under “Calculation only, not the timetable” in Settings. More authorities follow as licences are granted.

Can I match my own mosque?

Yes, and this is the setting most people actually need. Settings › Prayer time calculation › Minute adjustment shifts each prayer separately, from minus fifteen to plus fifteen minutes — enough to match the mosque down the road that has always called Fajr three minutes late.

The source label moves with it. Once any prayer is adjusted it reads “Diyanet-compatible · +5 min adjustment”, or “manual adjustment” when the prayers differ, on the Today screen, in the widget and in Settings — because after you have shifted the times, calling them the authority’s times would not be true. Adjustments made by an authority itself, such as Kemenag’s ihtiyat, are that authority’s own time and produce no label.

Check the clock first, though. A five-minute complaint in August 2026 came from a device whose time had been set by hand; the engine was within a minute throughout. If every prayer is off by the same amount, look at the clock before the settings.

How do we know the engine is right?

By measuring it on a schedule. On the first of every month, on the morning after every daylight-saving change, and on any change to the calculation code, an automated run starts. It makes two comparisons: six provinces across Türkiye and Northern Cyprus against a mirror of the published Turkish timetables, at a tolerance of one minute; and twenty-two cities worldwide against an independent calculator, in UTC, with a tolerance per prayer. A deviation fails the run and sends an email.

Every run is written to a log with its result and a link to the run itself. That log lives in the project repository today and will be published here alongside this page, so you can see the result for yourself.

There is one limit. Above about 45° latitude, differences from a locally published timetable get larger, because that is where authorities apply substitutions they rarely document. Those runs are recorded as information, not as failures. If your city differs consistently, send us the city, the date and your method at contact@sahaar.app.

Common questions