blob: 64269dfb57ada788c2dd64dfa1b8c2829514f77c (
plain) (
blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
|
Known BUGS:
1. <fixed>
2. rate control and specifics files interact badly
3. using REMOTE parallel execution:
`command` to find INPUT files does not work
SPECIFICS_FILE (CDL_FILE)
USER_DATA
dont work (should send the files over the socket, but do not
4. using any sort of parallel code:
REFERENCE_FRAME DECODED
does not work
5. Cannot use both STDIN and CDL_FILEs (ok, since it doesnt make sense...)
6. <fixed>
7. Patterns of BBI... seem to cause problems for parallel encoding.
I believe the bugs are fixed for sequential, but if you hit one, let us
know!
8. Sometimes parallel encoding results in a zero size being written into
the stream. To fix, specify SIZE of the actual size.
8.1 (related? minor, anyway) in parallel is also writes the Quant matrix
even when it is the default one
9. It is unclear (to us anyway) if we should generate Y's from 0..255 or 16..224
10. <fixed>
11. <fixed>
12. reading JPEG from stdin, the encoder will sometimes stop before the end
of the stream (if the images are smaller than half the buffer size).
13. Sometimes INPUT_CONVERT * is not good (requires "cat *"), in a parallel
case reported by a user it was needed.
14. Parallel encoding often generates improper temporal_ref's...
|
