A colorspace is the basis for color management.  By itself, the concept of a colorspace isn’t too difficult to grasp: it’s just a defined range of colors and brightness. But looking further, it can help to understand why we have different colorspaces and where they came from.

Although it’s common to see colorspaces represented as triangles on the CIE 1931 chromaticity diagram, it’s actually quite significant that a colorspace is really a 3D volume – a space.  In the video above, I show some 3D renderings from Bruce Lindblooms website, which can be found here.  More 3D renderings can easily be found on YouTube.

A bit more on sRGB

I noted in the introduction that the sRGB colorspace, the defacto standard for desktop computers, is “a bit crap”.  There are numerous sources on the internet that describe how sRGB was designed around the cheapest, lowest quality monitors that were on the market in the early 1990s.  Although this isn’t exactly wrong, it’s not exactly correct either. There’s an implication that the creators of sRGB – Hewlett Packard and Microsoft – actively measured consumer monitors.  But they did not.  In actual fact, as the original paper details, they proposed using the same color primaries as the previously defined rec 709 standard.  In suggesting that sRGB use the same color primaries as rec 709, the paper does not mention cost or quality at all, but rather the “unanimous worldwide agreement” of rec 709 as the key to the acceptance and adoption of sRGB.

But the notion that sRGB was based on the cheapest, lowest quality monitors isn’t exactly wrong, either.  But while it’s wrong to imply that the creators of sRGB actively surveyed monitors in the 1990s, there’s some truth that cost and quality were behind the choice of the color primaries.  It’s a good example of the difference between theory, and real-world applications.

As mentioned in the video above (as well as in the introduction to this series) NTSC analogue television was developed in 1953.  The red, green and blue primary colors that were defined by the original NTSC standard were based on what could be manufactured in 1953.  Although it was first to market, the NTSC standard had a few problems, which prompted other countries to develop the alternative PAL and SECAM formats.  SECAM was developed in 1956, and PAL in 1962.

Originally, these newer analogue formats were going to use the same primary colors as NTSC, however it had become apparent that the color TVs being manufactured weren’t following the official NTSC guidelines. Early televisions were relatively dim, and there was public demand for brighter screens.  TV manufacturers quickly worked out that they could sell many more TVs by using brighter CRT tubes, even if they didn’t match the official NTSC colors. Mario Orazio, writing for TVTechnology, suggests that there was only ever one model of TV that actually used a CRT matching those in the NTSC standard.

In response to the difference between the official standard and the real-world products being sold, the red, green and blue primaries for the PAL and SECAM standards were changed to match the consumer TVs that were already being manufactured. The color gamut for PAL and SECAM was noticeably smaller than the original NTSC definition, but by matching the TVs in common usage, it meant they could reproduce colors more accurately without additional color correction circuitry – which kept the cost down.

By 1968, it was clear that the difference between the official NTSC standard and the actual NTSC televisions being sold to the public was a problem for broadcasters.  One particular CRT manufacturer, the Conrac Corporation, produced all of the broadcast monitors used in TV studios but they had very different primary colors to the original NTSC standard.  This prompted the development of the SMPTE-C color standard – with primary colors based on the Conrac CRTs already used in studio monitors.  Like PAL and SECAM, the SMPTE-C color primaries were based on real-world red, green and blues already used in commercial products, and reflected the advances made in CRT manufacturing since 1953. As might be expected, the color values defined for SMPTE-C are very similar, but not identical, to PAL and SECAM.  As all of the existing NTSC broadcast equipment was effectively using the SMPTE-C colorspace anyway, this was a case where the official standard was updated to match existing equipment- I don’t think the studios had to actually change anything.

So the notion that the cheapest, most common products on the market were used as the basis for a new colorspace applies more accurately to the PAL and SECAM standards, dating back to the 1950s and early 1960s, and then the SMPTE-C standard, in 1969.  But the world had four different analogue television colorspaces.

When the work began on a new format for digital, high-definition television, it presented the opportunity to agree on a single, global color standard.  The very first draft of the rec 709 standard, in 1990, proposed the same red and blue primary colors as those used by the PAL and SECAM formats (dating back to 1956), and a green that was half-way between the PAL/SECAM formats and SMPTE-C.  As noted in the sRGB paper, these recommendations were adopted with unanimous approval.

That’s why, following the global acceptance of the rec 709 color primaries by the broadcast industry, Hewlett-Packard and Microsoft proposed using the same colors for the sRGB standard.

So yes, if you trace the lineage of sRGB back a few more steps then the color primaries were based on existing, low-cost consumer products. But it wasn’t Hewlett-Packard and Microsoft that sampled cheap monitors, it was the people who developed the SECAM, PAL and SMPTE-C colorspaces long before either company existed.  The real motivation for Hewlett-Packard and Microsoft was acceptance and adoption of their proposed sRGB standard, and this was more likely by using a set of color primaries that already had global support.

Coming up…

There’s a wide range of colorspaces in common usage today, and once we understand what they are – and why they exist – the next step is to look at an actual color managed workflow.

This is part 8 in a long series on color management.  If you’ve missed the other parts, you can catch up here:

Part 1: The honeymoon is over

Part 2: Newton’s Prisms

Part 3: It’s all in the brain

Part 4: Maxwell’s spinning discs

Part 5: Introducing CIE 1931

Part 6: Understanding the CIE 1931 chromaticity diagram

Part 7: Introducing gamma

AND – I’ve been writing After Effects articles and tutorials for over 20 years. Please check out some of my other ProVideo Coalition articles.