RGB Spaces Colorimetric Data

I present the following defined and precomputed data for each RGB color space.

This page is by Agatha Mallett (homepage). The theory of how to derive RGB matrices can be found here. Additional notes on sources and data are below. The important thing is that, barring any mistakes, all these numbers are exact to double-precision.

TL;DR Converting XYZ to sRGB

The most useful thing you can do with this data is convert from CIE XYZ to RGB. If you don't know what you're doing, that's probably sRGB.

See the main article for many more details, but basically, you:

Convert a spectral power distribution (in units of spectral radiance) into absolute XYZ space by taking the dot-products (not integrals) with the three (1 nm versions of) the CIE XYZ 1931 2° standard observer functions, which are just tabulated data you can just download.
Convert absolute XYZ to relative XYZ by multiplying by η.
Left-multiply relative XYZ by the space's M^-1 matrix to convert to linear RGB.
Finally, do any encoding transform (e.g. gamma encoding) for the space.

For sRGB:

Compute the absolute XYZ value.
  ⎡X⎤             830 nm       ⎡̄x(λ)⎤         (Note: CIE 015 specifies that this is a sum, not an
  ⎢Y⎥       =      Σ      s(λ) ⎢ȳ(λ)⎥ Δλ      integral, and that the dataset to use is the Δλ = 1 nm
  ⎣Z⎦abs        λ=360 nm       ⎣̄z(λ)⎦         one.  Data s(λ) should be in spectral radiance.)

Convert to relative XYZ.
  ⎡X⎤                             ⎡X⎤
  ⎢Y⎥       =   8.537524999999999 ⎢Y⎥         (Multiply by η value for sRGB)
  ⎣Z⎦rel                          ⎣Z⎦abs

Convert to linear RGB.  Note this matrix is M⁻¹; it's already inverted; you're welcome.
  ⎡R⎤           ⎡ 3.2404462546477406   -1.5371347618200821 -0.49853019302272933⎤ ⎡X⎤
  ⎢G⎥       =   ⎢-0.9692666062446794    1.8760119597883693  0.04155604221443006⎥ ⎢Y⎥
  ⎣B⎦ℓRGB       ⎣ 0.055643503564352756 -0.2040261797359601  1.0572265677227024 ⎦ ⎣Z⎦rel

Gamma-encode to get sRGB.  To be clear, you apply the `enc` function to each channel separately.
  ⎡R⎤          enc( ⎡R⎤     )                 ⎧ 255                                 if v ≥ 1.0
  ⎢G⎥       =  enc( ⎢G⎥     )        enc(v) = ⎨ 255*12.92*v                         if v ≤ 0.0031308
  ⎣B⎦sRGB      enc( ⎣B⎦ℓRGB )                 ⎩ 255*(1.055 val^(1.0/2.4) - 0.055)   otherwise

Note: if you're using the data spectrum for the white-point itself (e.g. D65), you need to be aware that that is not in units of spectral radiance. It was normalized to some arbitrary value. Therefore, you cannot use it as s(λ) in the above! If you want to do this, you must first multiply the values by σ (for sRGB, σ=11084.41934789392), which will return an SPD in SI units at the space's white luminance.

Colorimetric Data Table

Columns:

Color-Space: The name, links, and notes about the color space.
White-Point - Reference: The reference SPD for the white-point. The white-point is usually the chromaticity of a CIE standard illuminant, but sometimes may be only of a CIE illuminant. Rarely, it is a pre-defined chromaticity directly, in which case it is given in the next column.
White-Point - Chromaticity: The CIE 1931 xy chromaticity for the white-point. The z-component is also provided..
White-Point - Scaling: (Lum, σ, η): Respectively, the RGB space's target luminance (in nits), the scaling of the reference SPD to convert it to real units, and the scaling to convert from absolute XYZ to relative XYZ.
White-Point - Absolute XYZ: The CIE 1931 XYZ value of the reference spectrum scaled to have real units.
White-Point - Normalized XYZ: The CIE 1931 XYZ value of the reference spectrum, normalized to Y=1.
Primaries' Chromaticities: The CIE 1931 xy chromaticity (xyz provided) of the red, green, and blue primaries.
M^-1 (Normalized XYZ → RGB): The transformation matrix that converts from normalized XYZ space to the RGB space.
M (RGB → Normalized XYZ): The transformation matrix that converts from the RGB space to normalized XYZ space.

Accuracy not guaranteed, page subject to change at any time. Please contact me (see links on homepage), for corrections or additions.

Color-Space	White-Point					Primaries' Chromaticities			Matrices
Color-Space	Reference	Chromaticity	Scaling: (Lum, σ, η)	Absolute XYZ	Normalized XYZ	R	G	B	M^-1 (Normalized XYZ → RGB)	M (RGB → Normalized XYZ)
Adobe RGB (1998) ^[1]	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	160.0 22168.83869578784 4.268762499999999	0.22265716554956688 0.23425992896067657 0.2550689424384432	0.950470558654283 1.0 1.0888287363958846	0.64 0.33 0.03	0.21 0.71 0.08	0.15 0.06 0.79	2.0413657621150683 -0.5649454968342187 -0.34469384114836343 -0.9692666062446794 1.8760119597883693 0.04155604221443006 0.013447404369185286 -0.11838989336792537 1.0154108671712263	0.576731796112626 0.1855538404930999 0.18818492204855708 0.2973773323705728 0.6273486988100044 0.07527396881942283 0.027034302942779344 0.07068717733070472 0.9911072561224006
Apple RGB	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	??? ??? ???	N/A	0.950470558654283 1.0 1.0888287363958846	0.625 0.34 0.035	0.28 0.595 0.125	0.155 0.07 0.775	2.9515306569680746 -1.2894086677719054 -0.473843413014217 -1.085110047028224 1.9908579085392373 0.037202585408552404 0.08549346829862978 -0.26949671148806825 1.0912989776942743	0.4497298473899065 0.31624840281282596 0.1844923084515505 0.24465303698010912 0.6720278559772551 0.08331910704263572 0.025184871453834762 0.1411823226842973 0.9224615422577526
BestRGB ^[2]^[3]	D50	0.34568422232262525 0.35850402590425184 0.2958117517731229	??? ??? ???	N/A	0.964240837883783 1.0 0.8251281168377378	0.734710465507832 0.265289534492168 0.0	0.21496142445710442 0.7750320586936871 0.01000651684920852	0.12999487967229903 0.03500256016385049 0.8350025601638504	1.7549551966024226 -0.4834852176621251 -0.2529476905455372 -0.5441018600094364 1.5068718470800138 0.021540153519099044 0.0063507589001491425 -0.017588213710685727 1.2258272769857792	0.6327599424300101 0.20450534805234277 0.12697554740143013 0.2284772008202175 0.737333227648501 0.03418957153128157 0.0 0.009519783450483122 0.8156083333872547
Beta RGB ^[4]	D50	0.34568422232262525 0.35850402590425184 0.2958117517731229	??? ??? ???	N/A	0.964240837883783 1.0 0.8251281168377378	0.6888 0.3112 0.0	0.1986 0.7551 0.0463	0.1265 0.0352 0.8383	1.6831353903038802 -0.42821298806711844 -0.23600564200581922 -0.7710381038030288 1.706590764458632 0.044690832902468454 0.04000542406593118 -0.08854670982202248 1.2724955370867224	0.6712902170491151 0.1745799459975276 0.11837067483714034 0.30328907599547705 0.6637729970933187 0.032937926911204264 0.0 0.04070015860868846 0.7844279582290493
Bruce RGB ^[5]	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	??? ??? ???	N/A	0.950470558654283 1.0 1.0888287363958846	0.64 0.33 0.03	0.28 0.65 0.07	0.15 0.06 0.79	2.7454611464778576 -1.135811238097162 -0.4350259464377835 -0.9692666062446794 1.8760119597883693 0.04155604221443006 0.011272309552389816 -0.11397557436305258 1.013255380875929	0.46741713762640236 0.29445105501870916 0.18860236600917146 0.2410119615886137 0.6835470920077177 0.0754409464036686 0.02191017832623761 0.07361276375467729 0.9933057943149698
CIE RGB ^[6]^[7]	E	0.3333143807773521 0.3332877057993163 0.3333979134233316	??? ??? ???	N/A	1.000080035889628 1.0 1.0003306681347606	0.7346900232582807 0.26530997674171924 0.0	0.2736826753487811 0.7174214263192364 0.008895898331982469	0.1665346669372639 0.008883964460463396 0.8245813686022727	2.3634619417619147 -0.8958167862405376 -0.4676796705361663 -0.5151238487223274 1.4264686049372775 0.08866715277852716 0.005239624275601002 -0.014509441854323716 1.0089357753098176	0.490204212440407 0.3098806236482379 0.19999519980098315 0.17702168817329358 0.8123093605508466 0.010668951275859812 0.0 0.010072491872249317 0.9902581762625111
ColorMatch RGB	D50	0.34568422232262525 0.35850402590425184 0.2958117517731229	??? ??? ???	N/A	0.964240837883783 1.0 0.8251281168377378	0.63 0.34 0.03	0.295 0.605 0.1	0.15 0.075 0.775	2.6420569141623558 -1.2233203105671813 -0.3929800178475029 -1.1120034969749846 2.0590686762168375 0.015961772683851454 0.08217951526977244 -0.28075840705729826 1.4561587926508257	0.5093882890297654 0.3208992332658685 0.13395331558814907 0.27490796550812735 0.6581153766977981 0.06697665779407454 0.024256585191893592 0.10877940110707407 0.6920921305387702
DCI-P3 (Min Gamut) ^[8]^[9]^[10]	0.3139814397505545 0.35096063804161015 0.33505792220783537		48.0 N/A 14.229208333333332	0.0628730932745358 0.07027797868820297 0.06709354544039793	0.8946343427644687 1.0 0.9546880358933889	0.68 0.32 0.0	0.265 0.69 0.045	0.15 0.06 0.79	2.725367990343361 -1.0179932795788438 -0.44015898959085253 -0.7951931279824355 1.6897853969626755 0.022647905543803543 0.04122946074031344 -0.08761260407316605 1.1005975492067004	0.4451740690408129 0.27712566079607664 0.17233461292757915 0.20949367954861783 0.7215724752803505 0.06893384517103165 0.0 0.047059074474805465 0.9076289614185835
Display P3	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	??? ??? ???	N/A	0.950470558654283 1.0 1.0888287363958846	0.68 0.32 0.0	0.265 0.69 0.045	0.15 0.06 0.79	2.4931760142341752 -0.9312637545793583 -0.4026589580510947 -0.8295035249805175 1.7626949905835996 0.023625100395014736 0.03585366842286135 -0.07618904539858036 0.9570937598436042	0.48663357539456065 0.2656630314594593 0.19817395180026304 0.22900403547979326 0.6917263838001015 0.07926958072010522 0.0 0.04511259024783271 1.043716146148052
DonRGB4 ^[11]	D50	0.34568422232262525 0.35850402590425184 0.2958117517731229	??? ??? ???	N/A	0.964240837883783 1.0 0.8251281168377378	0.696 0.3 0.004	0.215 0.765 0.02	0.13 0.035 0.835	1.7602891057319137 -0.48809176042255037 -0.25359805045552036 -0.7126417986283907 1.6527733422095734 0.04167229562198291 0.007821534523828094 -0.03474463975550847 1.244900974517047	0.645808237607865 0.19334757262406196 0.12508502765185603 0.2783656196585625 0.6879576421274762 0.033676738213961235 0.0037115415954475 0.017985820709215065 0.8034307545330752
ECI RGB v2	D50	0.34568422232262525 0.35850402590425184 0.2958117517731229	??? ??? ???	N/A	0.964240837883783 1.0 0.8251281168377378	0.67 0.33 0.0	0.21 0.71 0.08	0.14 0.08 0.78	1.782656278583064 -0.4969553308581943 -0.2689941699140685 -0.9593793410018586 1.9478307832461978 -0.02758122425568669 0.08594085012568876 -0.17448596843700448 1.3229682382984815	0.6502427338133577 0.17807419929034182 0.13592390478008343 0.3202688091916538 0.6020603880768699 0.07767080273147624 0.0 0.0678377902058445 0.7572903266318933
Ekta Space PS5	D50	0.34568422232262525 0.35850402590425184 0.2958117517731229	??? ??? ???	N/A	0.964240837883783 1.0 0.8251281168377378	0.695 0.305 0.0	0.26 0.7 0.04	0.11 0.005 0.885	2.0042580210400773 -0.7304392626332076 -0.24499004067937002 -0.7110436960734704 1.6202471107247933 0.07922437402763591 0.038130253944037205 -0.08688697210198641 1.2726753611485204	0.593928081193343 0.27297430736779327 0.09733844932264672 0.26064469750211455 0.7349308275286742 0.004424474969211215 0.0 0.04199604728735282 0.783132069550385
HDTV (HD-CIF)	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	??? ??? ???	N/A	0.950470558654283 1.0 1.0888287363958846	0.64 0.33 0.03	0.3 0.6 0.1	0.15 0.06 0.79	3.2404462546477406 -1.5371347618200821 -0.49853019302272933 -0.9692666062446794 1.8760119597883693 0.04155604221443006 0.055643503564352756 -0.2040261797359601 1.0572265677227024	0.41245744558236713 0.357575865245516 0.18043724782639986 0.21267337037840806 0.715151730491032 0.07217489913055994 0.019333942761673457 0.11919195508183866 0.9503028385523725
M.A.C.	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	??? ??? ???	N/A	0.950470558654283 1.0 1.0888287363958846	0.67 0.33 0.0	0.21 0.71 0.08	0.14 0.08 0.78	1.9708347408378295 -0.549414288366556 -0.29738938544611754 -0.9537164251386219 1.936333348008717 -0.027418420924731162 0.06376928331735335 -0.12947096915947498 0.9816604825822879	0.588156514605899 0.1791315567830031 0.1831824872653808 0.28968902958201 0.605635263409201 0.10467570700878903 0.0 0.06824059306019166 1.020588143335693
NTSC-FCC	C	0.310089784498196 0.3162190653343711 0.3736911501674329	??? ??? ???	N/A	0.9806169788349289 1.0 1.181747690552088	0.67 0.33 0.0	0.21 0.71 0.08	0.14 0.08 0.78	1.9101298560238589 -0.5324914432393878 -0.2882293133130631 -0.9845310935942275 1.9988964627519166 -0.028304312714053162 0.05833602460227607 -0.11843980752583323 0.8980212272107952	0.6068484236188597 0.1735249491622592 0.20024360605381003 0.29889549223018463 0.5866795900247811 0.11442491774503431 0.0 0.0661047425380035 1.1156429480140846
NTSC-J ^[12]^[13]	D93	0.283140310338486 0.29705130864562385 0.41980838101589013	??? ??? ???	N/A	0.9531697120926225 1.0 1.4132520840590304	0.63 0.34 0.03	0.31 0.595 0.095	0.155 0.07 0.775	4.208210888348164 -2.0882485518339178 -0.6530261794394725 -1.0600359719078445 1.961107100285515 0.03487494016223207 0.0420028741931487 -0.14693739034350636 0.7832300658369493	0.32785561766182303 0.3683632121738962 0.2569508822569032 0.17693795238892035 0.7070197136886073 0.11604233392247242 0.015612172269610619 0.11288550050490367 1.284754411284516
NTSC, MUSE	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	??? ??? ???	N/A	0.950470558654283 1.0 1.0888287363958846	0.63 0.34 0.03	0.31 0.595 0.095	0.155 0.07 0.775	3.505386700609879 -1.7394847585786744 -0.5439625877342246 -1.0690728932885512 1.9778257505521932 0.035172252801383086 0.05632008773415409 -0.19702286747138334 1.0502039889251087	0.39359012225685513 0.36524942132870863 0.1916310150687192 0.21241371677354084 0.7010432441631665 0.08654303906329254 0.018742386774135956 0.11193127427815265 0.958155075343596
PAL / SECAM	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	??? ??? ???	N/A	0.950470558654283 1.0 1.0888287363958846	0.64 0.33 0.03	0.29 0.6 0.11	0.15 0.06 0.79	3.0628900410173245 -1.3931759151054413 -0.47575057119781294 -0.9692666062446794 1.8760119597883693 0.04155604221443006 0.0678775990963322 -0.2288550745488359 1.0693503726487745	0.43062002921388537 0.3415417093833163 0.1783088200570813 0.22203845256340965 0.7066380194137578 0.07132352802283252 0.020185313869400878 0.12955030355918892 0.9390931189672949
PAL-M	C	0.310089784498196 0.3162190653343711 0.3736911501674329	??? ??? ???	N/A	0.9806169788349289 1.0 1.181747690552088	0.67 0.33 0.0	0.21 0.71 0.08	0.14 0.08 0.78	1.9101298560238589 -0.5324914432393878 -0.2882293133130631 -0.9845310935942275 1.9988964627519166 -0.028304312714053162 0.05833602460227607 -0.11843980752583323 0.8980212272107952	0.6068484236188597 0.1735249491622592 0.20024360605381003 0.29889549223018463 0.5866795900247811 0.11442491774503431 0.0 0.0661047425380035 1.1156429480140846
ProPhoto / ROMM ^[14]	D50	0.34568422232262525 0.35850402590425184 0.2958117517731229	??? ??? ???	N/A	0.964240837883783 1.0 0.8251281168377378	0.734699 0.265301 0.0	0.159597 0.840403 0.0	0.036598 0.000105 0.963297	1.3458878267624148 -0.25559125739413247 -0.051105698036871774 -0.5446141763869226 1.5082019697524534 0.020526824458691955 0.0 0.0 1.2119330072431054	0.797706810310927 0.13518540007788032 0.03134862749497562 0.2880532224520508 0.7118568380461466 8.993950180262419e-05 0.0 0.0 0.8251281168377378
Russell RGB	D55	0.33243977703264577 0.34743862275763054 0.3201216002097237	??? ??? ???	N/A	0.95683022916123 1.0 0.9213759761908712	0.69 0.31 0.0	0.18 0.77 0.05	0.1 0.02 0.88	1.5869131539993353 -0.35978787736254514 -0.1721540430144121 -0.7535404101584284 1.6772351064816633 0.0475106123706927 0.037041892292602116 -0.08244808284482406 1.136349663556923	0.7016217794292249 0.15541232637623156 0.09979612335577351 0.3152213791638547 0.6648193961649906 0.019959224671154702 0.0 0.043170090660064324 0.8782058855308068
SGI ^[15]	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	??? ??? ???	N/A	0.950470558654283 1.0 1.0888287363958846	0.625 0.34 0.035	0.28 0.595 0.125	0.155 0.07 0.775	2.9515306569680746 -1.2894086677719054 -0.473843413014217 -1.085110047028224 1.9908579085392373 0.037202585408552404 0.08549346829862978 -0.26949671148806825 1.0912989776942743	0.4497298473899065 0.31624840281282596 0.1844923084515505 0.24465303698010912 0.6720278559772551 0.08331910704263572 0.025184871453834762 0.1411823226842973 0.9224615422577526
scRGB	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	80.0 11084.41934789392 8.537524999999999	0.11132858277478344 0.11712996448033829 0.1275344712192216	0.950470558654283 1.0 1.0888287363958846	0.64 0.33 0.03	0.3 0.6 0.1	0.15 0.06 0.79	3.2404462546477406 -1.5371347618200821 -0.49853019302272933 -0.9692666062446794 1.8760119597883693 0.04155604221443006 0.055643503564352756 -0.2040261797359601 1.0572265677227024	0.41245744558236713 0.357575865245516 0.18043724782639986 0.21267337037840806 0.715151730491032 0.07217489913055994 0.019333942761673457 0.11919195508183866 0.9503028385523725
sRGB	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	80.0 11084.41934789392 8.537524999999999	0.11132858277478344 0.11712996448033829 0.1275344712192216	0.950470558654283 1.0 1.0888287363958846	0.64 0.33 0.03	0.3 0.6 0.1	0.15 0.06 0.79	3.2404462546477406 -1.5371347618200821 -0.49853019302272933 -0.9692666062446794 1.8760119597883693 0.04155604221443006 0.055643503564352756 -0.2040261797359601 1.0572265677227024	0.41245744558236713 0.357575865245516 0.18043724782639986 0.21267337037840806 0.715151730491032 0.07217489913055994 0.019333942761673457 0.11919195508183866 0.9503028385523725
SMPTE-240M, SMPTE-C	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	??? ??? ???	N/A	0.950470558654283 1.0 1.0888287363958846	0.63 0.34 0.03	0.31 0.595 0.095	0.155 0.07 0.775	3.505386700609879 -1.7394847585786744 -0.5439625877342246 -1.0690728932885512 1.9778257505521932 0.035172252801383086 0.05632008773415409 -0.19702286747138334 1.0502039889251087	0.39359012225685513 0.36524942132870863 0.1916310150687192 0.21241371677354084 0.7010432441631665 0.08654303906329254 0.018742386774135956 0.11193127427815265 0.958155075343596
UHDTV	D65	0.31272687102656477 0.329023206641284 0.35824992233215125	??? ??? ???	N/A	0.950470558654283 1.0 1.0888287363958846	0.708 0.292 0.0	0.17 0.797 0.033	0.131 0.046 0.823	1.7165084742369088 -0.3556412150986606 -0.2533452177770312 -0.6666932627772987 1.6165028426244092 0.01576875657728227 0.01764365960186648 -0.04277983218534749 0.9423061854365046	0.637011006188227 0.1446149706534515 0.1688445818126044 0.26272205339966426 0.677989009475299 0.059288937125036655 0.0 0.02807231783272882 1.060756418563156
Wide Gamut ^[16]^[17]	D50	0.34568422232262525 0.35850402590425184 0.2958117517731229	??? ??? ???	N/A	0.964240837883783 1.0 0.8251281168377378	0.7346900232582807 0.26530997674171924 0.0	0.11416071960667966 0.8262069597811889 0.059632320612131445	0.1566409325773071 0.017704804990891342 0.8256542624318015	1.4593613408175363 -0.18194500055000798 -0.2729646425797244 -0.5228813372696496 1.447950456133572 0.06815072893724455 0.035369294559978476 -0.09794380204630061 1.2893019424686913	0.7175363176525349 0.1000814406190751 0.14662307961217302 0.2591154605631665 0.7243120319256421 0.016572507511191478 0.0 0.05227795142569759 0.7728501654120402

Data Notes

This data is based on various online resources, but especially Wikipedia's list and Bruce Lindbloom's list. Sources are gratefully acknowledged.
The entire end-to-end calculation of derived results is computed using arbitrary-precision arithmetic and rounded to the nearest double-precision value^[18]^[19]. Every number is written with the minimum number of digits to prevent any precision loss for that particular number.
As per CIE 015:2004 §7.1, for the standard observer I use the official 1 nm dataset.
For standard illuminants D50 and D65, I use the official 1 nm datasets (1 and 2)^[20].
For illuminants C and D55, I use the official 5 nm datasets (1, 2). CIE does not maintain these as standards, so 5 nm is the highest official source provided. I interpolate these linearly to 1 nm.
In keeping with the recommendations of CIE 015, I use the modern 1 nm versions of CIE 1924 V(λ) and CIE 1931 ȳ(λ), which are numerically identical (unlike the 5 nm originals). Newer datasets have been defined, but using them in the context of RGB spaces defined for the older standards would make a bogus result.
For a fair few spaces, I didn't find the reference luminance of the white-point. This means some information cannot be calculated (represented as "???").

Additional Notes

[1]	This supposedly began as an incorrect implementation of SMPTE 240M.
[2]	This is apparently one of two print-optimized spaces by Don Hutcheson (the other being DonRGB4). According to the author, "This is the best all-round color space for maximum gamut without exceeding the legal CIEYxy diagram.".
[3]	I have pulled the primaries out of the ICC profile; they appear here in higher precision than anywhere else.
[4]	Bruce Lindbloom invented this as the smallest color space still enclosing important colors, and having a few other handy features. It no longer appears on his site, though; perhaps he has decided against it somehow.
[5]	This space was invented by Bruce Fraser as a compromise between ColorMatch RGB and another space. The documentation says the other was Adobe RGB, though other sources say SMPTE 240M. It should be noted that at the time, Photoshop's definitions were screwed up, complicating the issue. Not to be confused with Bruce Lindbloom, who made Beta RGB.
[6]	The R, G, and B primaries are the chromaticities of 700, 546.1, and 435.8 nm light, respectively.
[7]	The normalized XYZ coordinates don't work out to be exactly ⟨1,1,1⟩. I believe this is because the standard observer was normalized before CIE interpolated it to 1 nm. I am not sure what, if anything, to do about this.
[8]	See "Digital Cinema System Specification" (e.g. v1.0 pg. 67; is removed by v1.3).
[9]	The white point comes from a certain gamma transform (see v1.0 §8.3.4.11) of 12-bit gamma-encoded XYZ values ⟨3794,3960,3890⟩.
[10]	DCI-P3 specifies a minimum gamut as an RGB space, but DCI-P3 itself is encoded in (gamma-transformed) XYZ. See "SMPTE Digital Cinema White Gamut Practices Study Group Report" section "DCI Primaries".
[11]	This is apparently one of two print-optimized spaces by Don Hutcheson (the other being BestRGB). The "4" is because it was the 4th version; the name itself is "by Damon Rando in 1996". "Captures the Ektachrome color gamut with virtually no clipping."
[12]	An obscure illuminant D93 is used for this mainly-Japanese color-space. It's not clear where the data originates. CIE doesn't have it, even though people call it a CIE illuminant. See also.
[13]	The data table I have starts at 380 nm. This is something of a problem since the standard observer starts at 360 nm. The calculations might be somewhat bogus, but it's best thing I can think of to do.
[14]	Bruce's numbers are consistent with being rounded to 4 digits.
[15]	See "A review of RGB color spaces".
[16]	Nominally, the space's R, G, and B primaries are the chromaticities of 700, 525, and 450 nm light, respectively. In practice, the green primary is slightly wrong, and everyone copied the wrong value from everyone else, so everyone got it wrong. What's presented here is the correct value—though by the same token, in a compatibility sense, this is quite useless.
[17]	It's not known with certainty where this mysterious color space comes from, and there are many similar and even more obscure contemporaneous RGB spaces kicking around. Allegedly, it was inspired by "the maximum additive triangle of Hardy and Wurzburg", this in turn likely defined in "The Theory of Three-Color Reproduction" [Hardy and Wurzburg 1937]. It has been said Adobe was responsible, but I know of no evidence for that. The oldest source I can find is "A Review of RGB Color Spaces", whose author, Danny Pascale, kindly replied to me, corroborating the inspiration from Hardy and Wurzburg. Slight edits for context and clarity: I remember doing quite some reading at the time, so I looked at an old book I had consulted, "Color Theory and Its Application in Art and Design" [Agoston 1979], in which I had a post-it on page 64 where it is mentioned that a large gamut space called the Hardy-Wurzburg triangle is defined by the 700-535-400 nm wavelengths. [...] this article, "Dilemma of the primary colors" [Edge and Howard 1979], proposes a 700-520-400 triplet to get a "triangle of the largest area". Edge & Howard mention that Hardy & Wurzburg show that best results are obtained with 535 nm for the primary green "because there are so many saturated yellows in everyday use". The triplet shown in my document (700, 525, 450) is perhaps from yet another reference or from a wrap-up of my readings at the time. For instance, why 525 nm? Perhaps because I may have felt that this would be yet another compromise between 520 nm, the wavelength where the green cones are stimulated with minimal contributions by the red and blue cones, while not too penalizing the saturated yellows. This is a chain of thought I have faint memories of. So, I may have defined this wide-gamut space as a sort of consensus on the issue or I got it from a reference I can't find . . .
[18]	This was surprisingly hard to confirm since it isn't documented explicitly. See `__float__(⋯)` for `mpmath.mpf`, which calls `to_float(⋯)`, which leans on correctly rounded mantissa and Python's `math.ldexp(⋯)`. That doesn't document precision either, but it leans on the platform's `ldexp(⋯)`, which is exact in this context.
[19]	I was surprised how much difference this can make. Many other sources are simply wrong after 3, or even 2, decimals!
[20]	These used linear interpolation instead of the higher-quality interpolation seen elsewhere, so practically speaking they are no better (and are actually slightly worse) than the coarser data. Nevertheless, they are recommended, so I used them here.