ThreeJS 61 - as small as it gets (near-perfect GZIP compression)

Choose a version:

The original file has 813502 bytes (794.4k) and is available from the project website.
There you can find the official minified version, too, which brings down the size to 410858 bytes (401.2k, 51%).

After GZIP compression these minified files vary in size:

Boot	119523 bytes (116.7k)	CDN
cdnjs	99716 bytes (97.4k)	CDN
gzip -6 (default)	98753 bytes (96.4k)	local copy
gzip -9	98346 bytes (96.0k)	local copy
libdeflate -12	95084 bytes (92.9k)	local copy
zultra	95025 bytes (92.8k)	local copy
7zip -mx=9 -tgzip	94973 bytes (92.7k)	local copy
pigz -11 -n	94684 bytes (92.5k)	local copy
kzip -s0 -rn -b0	94602 bytes (92.4k)	local copy
Zopfli	94539 bytes (92.3k)	local copy
Zopfli (defluff)	94537 bytes (92.3k)	local copy

perma-link to the smallest file on my server:
http://minime.stephan-brumme.com/files/threejs/three-r61.min.js (or via HTTPS)

You will automatically get the smallest ThreeJS 61 file, ETag caching is available and
if your browser doesn't support GZIP decompression then the uncompressed version will be sent.

Currently best Zopfli settings

Save 5177 bytes by using my ThreeJS 61 Zopfli version instead of the best available CDN (5.48% smaller than cdnjs, 94539 vs. 99716 bytes):

Download

You can use my super-compressed files for whatever purpose you like as long as you respect the library's original license agreement.
There are no restrictions from my side - but please avoid hot-linking if you run a high-traffic website.

These command-line settings yielded the best compression ratio so far (Linux version of zopfli-krzymod):
zopfli --i1000000 --mb8 --mls512 --bsr22 --lazy --ohh

(found February 7, 2017)

Description	Value	Parameter
iterations	1000000	--i1000000
maximum blocks	8	--mb8
maximum length score	512	--mls512
block splitting recursion	22	--bsr22
lazy matching in LZ77	yes	--lazy
optimized Huffman headers	yes	--ohh
initial random W for iterations	1	--rw1
initial random Z for iterations	2	--rz2

Even Smaller Files Thanks To Defluff

Zopfli's output can be further optimized by the defluff tool.
In this particular case, defluff saves 2 more bytes (94537 bytes).

Download

Verify file integrity

After decompression, my uncompressed files are identical to the original ones:

MD5:
curl --silent --compressed https://raw.githubusercontent.com/mrdoob/three.js/r61/build/three.min.js --location | md5sum
1a3257293bd9785fa7905dfa7f2ecf33 -
curl --silent --compressed https://minime.stephan-brumme.com/files/threejs/three-r61.min.zopfli.js.gz | md5sum
1a3257293bd9785fa7905dfa7f2ecf33 -

SHA1:
curl --silent --compressed https://raw.githubusercontent.com/mrdoob/three.js/r61/build/three.min.js --location | sha1sum
cb3c8723db7d000b60f840558f894ce8ece47895 -
curl --silent --compressed https://minime.stephan-brumme.com/files/threejs/three-r61.min.zopfli.js.gz | sha1sum
cb3c8723db7d000b60f840558f894ce8ece47895 -

All listed CDNs deliver identical contents:

CDN	Size (compressed)	MD5 (uncompressed)	Timestamp
Boot	119523 bytes	1a3257293bd9785fa7905dfa7f2ecf33	(invalid)
cdnjs	99716 bytes	1a3257293bd9785fa7905dfa7f2ecf33	(invalid)

Note: only the MD5 hashes are shown to keep things simple.

Other Versions

Available ThreeJS versions at minime.stephan-brumme.com:

148, 147, 146, 145, 144, 143, 142, 141, 140, 139, 138, 137, 136, 135, 134, 133, 132, 131, 130, 129, 128, 127, 126, 125, 124, 123, 122, 121, 120, 119, 118, 117, 116, 115, 114, 113, 112, 111, 110, 109, 108, 107, 106, 105, 104, 103, 102, 101, 100, 99, 98, 97, 96, 95, 94, 93, 92, 91, 90, 89, 88, 87, 86, 85, 84, 83, 82, 81, 80, 79, 78, 77, 76, 75, 74, 73, 72, 71, 70, 69, 68, 67, 66, 65, 64, 63, 62, 61, 60, 59, 58, 57, 56, 55, 54, 53, 52, 51, 50

The project site contains an overview how well these versions were compressed.
Other interesting projects are AngularJS, BackboneJS, Bootstrap, D3, Dojo, Ember, jQuery, Knockout, lodash, React, Socket.IO, UnderscoreJS and Vue.

Changelog

Best Zopfli parameters so far:

Size	Improvement	Parameters	Found
94539 bytes	-6 bytes	zopfli --i1000000 --mls512 --bsr22 --lazy --ohh	February 7, 2017 @ 03:44
94545 bytes	-7 bytes	zopfli --i100000 --mls512 --bsr22 --lazy --ohh	December 28, 2015 @ 20:30
94552 bytes	-10 bytes	zopfli --i10000 --mls512 --bsr22 --lazy --ohh	November 17, 2015 @ 13:10
94562 bytes	-5 bytes	zopfli --i10000 --mls4096 --bsr12 --lazy --ohh	November 17, 2015 @ 11:08
94567 bytes	-27 bytes	zopfli --i1000 --mls512 --bsr22 --lazy --ohh	November 16, 2015 @ 17:18
94594 bytes	–	zopfli --i100 --mls512 --bsr22 --lazy --ohh	November 15, 2015 @ 10:49

If there are multiple parameter sets yielding the same compressed size, only the first one found is shown.

Most recent activity on July 20, 2020 @ 12:49.

Heatmaps

This Zopfli heatmap visualizes how compression changes when modifying the --bsr and --mls parameter.
Cell's contents is the best filesize achieved (in bytes, hover with mouse over cells to see number of iterations).

Good parameters are green, bad are red. The best and worst are bold as well.
The brightness of the blue background color indicates how many iterations were processed:
10,000, 100,000 or 1,000,000.

bsr \ mls	2	4	8	16	32	64	128	256	512	1024	2048	4096	8192	16384	32768
bsr \ mls	2	4	8	16	32	64	128	256	512	1024	2048	4096	8192	16384	32768
2	94711	94713	94711	94695	94689	94694	94679	94702	94640	94709	94651	94628	94672	94669	94664
4	94649	94648	94636	94645	94634	94641	94650	94611	94606	94596	94630	94616	94674	94660	94659
6	94723	94634	94744	94640	94652	94751	94655	94652	94573	94638	94621	94660	94599	94666	94759
7	94736	94719	94720	94628	94689	94630	94637	94618	94779	94623	94579	94637	94616	94631	94657
8	94719	94738	94723	94780	94625	94714	94629	94709	94623	94685	94658	94658	94634	94668	94653
9	94744	94697	94692	94702	94631	94742	94632	94679	94629	94633	94686	94638	94576	94681	94708
10	94724	94722	94728	94771	94741	94619	94722	94622	94622	94703	94600	94661	94657	94656	94639
11	94724	94728	94752	94636	94632	94716	94690	94705	94644	94696	94641	94660	94632	94662	94736
12	94716	94715	94715	94709	94674	94628	94637	94712	94637	94622	94588	94556	94572	94651	94659
13	94727	94715	94725	94711	94693	94629	94689	94620	94696	94686	94608	94640	94653	94642	94657
14	94730	94722	94718	94687	94628	94690	94632	94680	94631	94638	94663	94641	94641	94715	94640
15	94646	94696	94695	94627	94621	94629	94678	94616	94601	94632	94602	94655	94599	94632	94637
16	94661	94659	94617	94636	94694	94691	94687	94614	94621	94641	94638	94641	94657	94675	94644
17	94721	94721	94724	94696	94623	94682	94630	94692	94685	94624	94628	94647	94600	94627	94640
18	94717	94720	94706	94707	94623	94630	94632	94693	94631	94623	94595	94664	94657	94653	94662
19	94644	94718	94654	94635	94681	94690	94705	94683	94625	94635	94624	94631	94678	94658	94650
20	94721	94718	94746	94655	94616	94630	94690	94614	94640	94628	94592	94637	94604	94658	94658
21	94736	94719	94803	94691	94737	94737	94736	94683	94625	94697	94608	94658	94656	94714	94644
22	94825	94725	94722	94750	94788	94729	94787	94683	94539	94656	94607	94654	94583	94713	94646
23	94657	94635	94714	94705	94637	94698	94697	94708	94644	94626	94649	94648	94629	94656	94639
25	94729	94723	94628	94630	94686	94628	94629	94683	94644	94627	94593	94647	94639	94689	94656
30	94727	94719	94631	94692	94619	94626	94630	94619	94597	94684	94623	94665	94643	94687	94645
40	94727	94717	94716	94636	94624	94624	94720	94708	94650	94689	94592	94641	94629	94719	94637

Due to the Monte Carlo design of my search algorithm, not all parameters have reached the same number of iterations yet:

Iterations	Min. Bytes	Reduction	Coverage
100	94594 bytes	–	100%
1,000	94567 bytes	-27 bytes	100%
10,000	94552 bytes	-15 bytes	100%
100,000	94545 bytes	-7 bytes	0.58%
1,000,000	94539 bytes	-6 bytes	0.29%
10,000,000	–	–	–

KZIP has far less options available for tuning/optimization. I only played around with the number of blocks (parameter -n):

Blocks	Min. Bytes	Compared To Best Zopfli	Compared To Best KZIP
(any)	94602 bytes	+63 bytes (+0.07%)	–
1	95067 bytes	+528 bytes (+0.56%)	+465 bytes
2	94942 bytes	+403 bytes (+0.43%)	+340 bytes
3	94942 bytes	+403 bytes (+0.43%)	+340 bytes
4	94829 bytes	+290 bytes (+0.31%)	+227 bytes
5	94785 bytes	+246 bytes (+0.26%)	+183 bytes
6	94805 bytes	+266 bytes (+0.28%)	+203 bytes
7	94699 bytes	+160 bytes (+0.17%)	+97 bytes
8	94665 bytes	+126 bytes (+0.13%)	+63 bytes

Non-DEFLATE Algorithms

Archivers based on completely different compression algorithms often produce superior results.
Unfortunately, browsers only support gzip compression at the moment.
However, support for Brotli is constantly growing - for example, your browser actually supports it !

Algorithm	Program	Parameters	Size	Compared To Best Zopfli
ZPAQ (Wikipedia)	zpaq	zpaq -method 69	64738 bytes	-29801 bytes (-31.52%)
RAR (proprietary)	RAR	rar a -m5 -md64m -mc63:128t -mt1	74424 bytes	-20115 bytes (-21.28%)
PPMd (Wikipedia)	7zip	7za a -mx=9 -m0=ppmd	79571 bytes	-14968 bytes (-15.83%)
Brotli (Wikipedia)	brotli	brotli -q 11	80758 bytes	-13781 bytes (-14.58%)
LZMA2 (Wikipedia)	xz	xz -9	81284 bytes	-13255 bytes (-14.02%)
Zstandard (Wikipedia)	zstd	zstd -19	86389 bytes	-8150 bytes (-8.62%)
Burrows-Wheeler transform (Wikipedia)	bzip2	bzip2 -9	86873 bytes	-7666 bytes (-8.11%)

Detailled Analysis

I wrote a DEFLATE decoder in Javascript. Click the button below to start a client-side analysis of the smallest gzipped files (may take a second):

Load and Parse Data

Notes: pigz is a fast open source multi-threaded implementation of gzip written by one of the original authors of gzip.
However, when using compression level 11, pigz actually switches to the slower Zopfli algorithm and isn't multi-threaded anymore.
KrzyMOD's extensions to Zopfli offer the highest level of configuration and is therefore used for my brute-force search.
Ken Silverman wrote the closed-source KZIP compression program and Jonathon Fowler ported it to Linux.
Defluff was created by Joachim Henke; DeflOpt is a tool by Ben Jos Walbeehm.

website made by Stephan Brumme in 2015 and still improving in 2024.
all timestamps are displayed in central european time. see my changelog.
no flash, not even images or external css files - and everything squeezed into a single html file.
which was handsomely compressed before releasing it into the wild internet - obviously.

please visit my homepage and my blog, too.
email: minime (at) stephan-brumme.com

All trademarks are property of their respective owners. You know, the boring legal stuff.