c2035eb0f0
similarity uses reduce more often
2016-03-21 14:18:01 +00:00
385f8a167f
notes
2016-03-21 11:36:15 +00:00
063706f300
notes
2016-03-21 11:11:27 +00:00
ba3ae9e671
finished!
...
fix up preciscion and signedness issues, simpler inner loop
2016-03-21 11:08:38 +00:00
57cba39607
switch to 2.6 bits
...
almost working
2016-03-20 22:29:25 +00:00
74984f1049
off by one error
...
didn't always write the output
2016-03-19 10:37:02 +00:00
05156edd15
add orc path for reducev
...
works, but still not quite right, see some banding
2016-03-18 17:41:51 +00:00
c20eb9dca0
remove old unrolled case for reduce
...
not used any more
2016-03-15 17:34:31 +00:00
46a18118c6
works!
2016-03-15 16:46:18 +00:00
ff88087a28
try making the reduce mask larger with residual
2016-03-15 14:59:59 +00:00
45c3fea6ac
more resize tuning
2016-03-15 09:49:30 +00:00
ac4e5ce059
return of shrink-less-affine-more
...
though a bit less extreme
2016-03-14 14:58:02 +00:00
773b474116
Merge branch 'master' of github.com:jcupitt/libvips
2016-03-14 13:35:47 +00:00
82cecf3d9f
bilinear interp. supports complex types
...
in line with cubic etc.
2016-03-14 10:06:45 +00:00
fba2ac2f85
fix sizing issues with new reduce code
...
a couple of bugs and some changed conventions
2016-03-13 17:44:24 +00:00
f12fef4aa9
fix compile
2016-03-13 11:35:35 +00:00
dc7902b12f
Merge branch 'master' into add-lanczos3
2016-03-13 11:23:03 +00:00
6368ab0649
add switches to disable rad, analyze and ppm
...
vips has built-in support for rad, analyze and ppm ... add configure
switches to disable these readers
useful to reduce the attack surface in some applications
2016-03-12 16:48:27 +00:00
42224b6b0e
Merge branch '8.2'
2016-03-12 15:12:56 +00:00
abe4e70d02
make radload slightly more robust
...
stops some valgrind warnings
2016-03-12 15:10:52 +00:00
70db9aa77b
better rounding in resize
2016-03-12 13:14:34 +00:00
493482c8a7
hey ho, line cache goes back too
2016-03-11 14:56:06 +00:00
a832a0c254
notes
2016-03-11 14:42:12 +00:00
4f41617d89
TODO notes
2016-03-11 13:47:19 +00:00
08e8c27e38
moore vips_resize() tuning
...
put back the mild post-shrink sharpen, a little more anti-alias
see https://github.com/jcupitt/libvips/issues/404#issuecomment-195297607
2016-03-11 10:17:41 +00:00
5d31d8e824
turn up anti-alias for small shrinks
...
a little more anti-alias for small reductions
see https://github.com/jcupitt/libvips/issues/404#issuecomment-195276154
2016-03-11 09:25:33 +00:00
2ca7b29214
better gauss mat generation
2016-03-10 22:01:55 +00:00
4974a1ed9c
better rounding for vips_resize()
...
we were getting off by one size errors
2016-03-10 19:53:05 +00:00
9209fb25c5
update notes
2016-03-10 17:40:19 +00:00
d351e07a6b
finish up
2016-03-10 17:33:34 +00:00
e9bf936377
finish vl3
...
but it's still not producing identical results to reducev, seems to be
offset by one?
2016-03-10 12:16:21 +00:00
4d10bd12f9
still trying to get reducevl3 to vectorise
2016-03-10 10:24:44 +00:00
f58d941d99
Merge branch 'master' into add-lanczos3
2016-03-10 08:54:10 +00:00
61d2aa6586
formatting
2016-03-10 08:53:54 +00:00
013102de01
finish hl3 version
...
vl3 nest
2016-03-10 08:43:50 +00:00
f9d7a8df48
Merge pull request #402 from lovell/defer-LabQ2sRGB-table-building
...
Performance: only build LAB to sRGB look-up tables if needed
2016-03-10 08:41:08 +00:00
a26291a426
reducehl3 is done
...
sort of anyway, tinkering with optimization
2016-03-09 22:47:02 +00:00
a40397c438
Build LAB to sRGB tables only when first needed
...
Reduces libvips start-up time by ~20ms
2016-03-09 19:20:57 +00:00
dcd921d7bf
silly travis
2016-03-09 13:06:03 +00:00
e9728e389b
first commit
2016-03-09 12:56:30 +00:00
efa12029a6
add notes on new image format deps
2016-03-08 09:00:57 +00:00
9201248623
Merge branch '8.2'
2016-03-06 18:05:12 +00:00
acf5f5169a
fix hist_entropy
...
there was a copy-paste error in the call to vips_log(), thanks Lovell
see https://github.com/lovell/sharp/issues/295
2016-03-06 18:02:59 +00:00
2ea5c5f7e5
make python bandrank() work like bandjoin()
...
we had a custom wrapper for bandjoin(), but bandrank(), a function with
an almost identical interface, did not ... this was confusing
bandrank() now has a custom wrapper too ... this breaks API
unfortunately, but hopefully very few people were using this thing and
it's better to make this change as soon as possible
2016-03-01 12:12:51 +00:00
25c3c49d1c
better handling of cast+shift for non-int formats
...
before, im.cast(uchar, shift = true) where im was float and tagged as
rgb16 would not shift the image, since it's unclear how much to shift a float
type by
now we do two casts: first, we guess the numeric range from the
interpretation, so rgb16 would be ushort, so we cast float->ushort;
second, we cast to the target type and do the shift on the way
see https://github.com/jcupitt/libvips/issues/397
thanks apacheark
2016-03-01 11:31:54 +00:00
ad7ddc4f24
more vips_resize() tuning
...
now we have better tuning for the anti-alias blur, we can increase the
amount we shrink by and reduce the amount we affine by
this gives a noticable speedup
2016-03-01 11:19:30 +00:00
19a838470b
add webp shrink-on-load
...
and vipsthumbnail knows about it too
see https://github.com/jcupitt/libvips/issues/398
2016-02-28 14:36:21 +00:00
e582f13d13
tiny polish to hist_find
...
slightly faster again
2016-02-28 12:08:06 +00:00
e46e2b7b28
Merge pull request #392 from lovell/hist-find-perf
...
vips_hist_find: ~30% performance improvement for 1-4 band uchar
2016-02-28 11:29:14 +00:00
b197773095
oop, overflow in sharpen
...
plus another tweak of the defaults
2016-02-28 11:24:35 +00:00
John Cupitt
Lovell Fuller