Fix 23.976 fps Subtitles to 25 fps
Or paste subtitle text
Options
Which frame rate the timings are counted in. On the format pages this is what MicroDVD needs, because it stores timing as frame numbers. On the frame rate pages it is the rate the file was timed against — the version the subtitles were made for — and the page's second box takes the rate of the video you are going to use them with. 23.976, 24, 25, 29.97, 30, 50, 59.94 and 60 are the common ones. Getting it wrong shifts the whole timeline, and on the frame rate pages it shifts it by a ratio rather than by a fixed amount.
The frame rate of the version you are going to use the subtitles with — the video you are actually going to play. The factor applied is the first box divided by this one, so 24 into 25 gives 0.96: a film runs 4% faster on PAL, and a 100-minute feature fits a 96-minute tape. Getting the two boxes the wrong way round shifts the whole timeline in the wrong direction by the same amount.
This page takes a subtitle file timed against a 23.976 fps version and puts it on a 25 fps PAL version. Every timestamp is multiplied by 23.976 divided by 25, which is 0.95904, so the whole timeline comes out 4.096 percent earlier and each cue is 4.096 percent shorter. The two boxes are already filled in; change them if your file is timed against something else.
The pair comes up whenever a film made for the American market is published in Europe. A film shot at 24 fps is carried in an NTSC signal at 23.976 fps — that is 24000/1001, the odd divisor that every NTSC rate carries — and the PAL version replays the same pictures at 25 fps. The pictures are identical and the frame count is identical; what changes is how fast they are shown, and the subtitle timings have to be replayed at the same speed as the picture.
4.096 percent is not the same as the 4 percent of the clean 24 to 25 pair, and the difference is not academic: it is 3.5 seconds an hour, and 5.5 seconds over a 95-minute film. Which of the two pairs is yours depends on what the file was timed against, not on what the video looks like — a 23.976 disc and a 24 fps master are different animals, and each has its own page in this group.
The ratio is applied to the distance from zero, so a cue one second in moves 41 ms and a cue forty-five minutes in moves 1 minute 51 seconds. Both edges of every cue move, so a two-second subtitle is given 1.92 seconds. Nothing is uploaded, nothing is sent anywhere, and a whole season goes through the queue in one pass — SRT, WebVTT, ASS, SSA, SBV, SAMI, MicroDVD, TTML, JSON and CSV files are read as what they actually are rather than as the extension suggests.
Examples
A short cue and a long one
Input
1
00:00:01,000 --> 00:00:03,000
Hello there.
2
00:01:00,000 --> 00:01:02,000
One minute in.
Output
1
00:00:00,959 --> 00:00:02,877
Hello there.
2
00:00:57,542 --> 00:00:59,460
One minute in.
0.95904 is 23.976 divided by 25. The cue at one second loses 41 ms and the one at a minute loses 2,458 ms — sixty times the distance from zero, sixty times the movement. Both edges of each cue are multiplied, so the first cue's 2 seconds become 1.918 and the second cue's 2 seconds become the same 1.918.
The error is proportional to the clock
Input
1
00:45:00,000 --> 00:45:02,000
Forty-five minutes in.
2
01:00:00,000 --> 01:00:02,000
An hour in.
Output
1
00:43:09,408 --> 00:43:11,326
Forty-five minutes in.
2
00:57:32,544 --> 00:57:34,462
An hour in.
A minute and 51 seconds early at the forty-five minute mark, and two minutes and 27 seconds early at the hour. This is what makes the pair recognisable before you convert anything: the subtitles are close to right at the start and minutes out by the end, and the error keeps growing at the same rate. A fixed offset can never fix that, and a proportion can.
A MicroDVD file changes only its declaration
Input
{1}{1}23.976
{120}{144}Hello
Output
{1}{1}25
{120}{144}Hello
MicroDVD stores its timing as frame numbers, and the same pictures replayed at 25 fps are still frames 120 and 144. Rewriting those numbers would double the error instead of fixing it. What the conversion edits is the frame rate declared on the first line — 23.976 becomes 25 — and that is what tells the player how to turn the frame numbers back into seconds.
Frequently asked questions
What is 23.976 fps, and why is it not exactly 24?
It is 24000 divided by 1001, which is very slightly slower than 24. The reason is NTSC: a 29.97 frame rate was needed for colour television in the United States, and 29.97 is 30000/1001, so everything that has to sit inside an NTSC signal inherits the same 1001 divisor. Film carried in NTSC therefore runs at 24000/1001 rather than 24, and that is why this pair is not the same as the clean 24 to 25 one.
What factor does 23.976 to 25 use?
0.95904, which is 23.976 divided by 25: every timestamp keeps 95.904 percent of its value, so the timeline runs 4.096 percent earlier. A 100-minute feature comes out at 95 minutes 54 seconds rather than the round 96 minutes you get from 24 to 25, and every cue is given 4.096 percent less time on screen. The two rates go in the boxes in that order — the one the file was timed against first, the one you are going to play it with second.
My file was timed for the American release and I have the European disc — is this the pair?
That is what this page is for, as long as the American release you have is film carried at 23.976. The European disc is 25 fps PAL, and the subtitles have to be replayed at its speed: 0.95904. If your file was timed against a film master running at a true 24, the pair you want is 24 to 25 and the factor is 0.96; the two boxes here accept both, but the pages explain different situations.
Will the words change, or only the timings?
Only the timings. Text, line breaks and whatever styling the format carries are left exactly as they are, and the loss report tells you if the file itself lost anything on the way in — a frame rate conversion on its own discards nothing. What changes is that every cue is given 4.096 percent less time, so the reading speed of each line goes up by that much and a line that was already close to the limit may go over it.
How far out would the subtitles be if I used 0.96 instead of 0.95904?
About 3.5 seconds an hour, and 5.5 seconds over a 95-minute film — always early, and always growing towards the end. That is small enough to look like a rounding error and large enough to be visible on screen, which is why the two pages exist separately. If you are not sure which rate your file was timed against, the test is where the error sits after converting: the wrong pair leaves a residual that grows, the right pair leaves none.
Does the PAL version really run four percent faster?
The rate goes up by 4.167 percent, from 24 to 25, and the duration therefore comes down by 4 percent — which is the number people quote when they say PAL speed-up. At 23.976 the rate rises by 4.271 percent and the duration falls by 4.096 percent, a shade more than the round figure. The duration is the one that matters here, because a subtitle timeline is measured in time, not in frames.