Compare the same elapsed duration.

Enter a count, sample rate and picture rate. This browser calculator works with numbers; it does not read, convert or resample media files.

Samples are PCM samples per channel.
Whole picture frames; 0–1,000,000,000,000. No signed offsets.
Whole samples per second per channel; 8,000–384,000 Hz.
Fractional rates use their exact /1001 relationships.
Optional rounding applies to the converted count.

1,601.600000 samplesConverted count (decimal view)
8008/5 samplesExact reduced fraction
Not roundedWhole-unit result
No whole-unit roundingRounded minus original timing
0.033366667 sOriginal elapsed duration
1,601.600000 samples/frameSamples per picture interval

1 picture frame · 29.97 fps · 48,000 Hz. Exact relationship: 8008/5 samples per channel. Decimal views are rounded; the reduced fraction preserves the relationship.

Count samples per channel.

At 48,000 Hz, each channel contains 48,000 PCM samples for one second. Adding channels increases the amount of data, but does not change that duration. Enter the per-channel count, rather than adding together every sample value in an interleaved multichannel buffer.

Apple’s Core Audio terminology calls a simultaneous set containing one sample from each channel an audio frame. That audio frame is distinct from a picture frame. A compressed audio packet is another unit; an AAC packet count is not a suitable input here. Use the decoded PCM sample count and sample rate.

Divide by the rate to get duration.

seconds = picture frames / picture fps = audio samples / sample rate
samples = picture frames × sample rate / picture fps
picture frames = samples × picture fps / sample rate
rounding error = rounded output duration − original duration

The labels 23.976, 29.97 and 59.94 mean 24000/1001, 30000/1001 and 60000/1001 fps here. The reduced fraction is exact for these entered relationships. Decimal counts show six places and elapsed seconds show nine; displayed precision does not establish recording accuracy.

Exact samples per channel for picture intervals
Picture count and rateSample rateSample count
1 frame at 29.97 fps48,000 Hz8008/5 = 1,601.6
5 frames at 29.97 fps48,000 Hz8,008
1 frame at 23.976 fps48,000 Hz2,002
1 frame at 24 fps44,100 Hz3675/2 = 1,837.5

Round the final range once.

A fractional sample count describes a duration relationship, not a partially stored PCM sample. If an edit requires whole units, select a rounding rule. Nearest rounds a half-unit upward: 1,837.5 becomes 1,838. Positive timing error means the rounded duration is longer; negative means shorter.

One picture frame at 29.97 fps, 48,000 Hz
RuleWhole samplesTiming error
Round down1,601−12.500000 µs
Nearest, ties up1,602+8.333333 µs
Round up1,602+8.333333 µs

Rounding each of five separate one-frame intervals to 1,602 samples gives 8,010 samples. The exact five-frame range is 8,008 samples. Keep the exact relationship through intermediate calculations, then round the final range once where possible.

In reverse, 1,602 samples at 48,000 Hz last 0.033375 seconds. At 29.97 fps this is 4005/4004 picture frames. Rounding to one frame shortens the duration by 8.333333 µs. Zero is a valid count; negative offsets are outside this tool.

Check origin and clock assumptions.

These duration comparisons assume a common zero origin and compatible sample/picture boundary phase. They do not measure clock drift or guarantee sample-phase alignment, pitch preservation or resampling quality. Confirm offsets and actual playback rates in your editing system.

DF and NDF are timecode label-counting choices, not different picture rates. Convert labels with the timecode calculator first. Continue with the audio pull-up and pull-down guide, conform calculator, production tools and timecode mathematics reference.

Technical reference

The worked examples apply the equations above to the stated rates and zero-origin assumptions.