You've hit on a really important point about CRI and the evolution of color rendering metrics. You're absolutely right: CRI has limitations, and some manufacturers can "game" the system. Let's break down the issues and address your questions

The Problem with CRI (Ra)
Limited Color Samples

CRI only uses 8 specific color samples (R1-R8). It gives a decent indication of rendering these specific colors, but it doesn't necessarily translate to accurate rendering of all colors, particularly saturated colors. R9 (red), which is a supplemental metric, is often omitted or poorly rendered, even in "high CRI" LEDs. This is especially important for things like skin tones, artwork, or any application where reds need to be accurate.

Averaging Problems

CRI averages the rendering of the 8 colors. Even if one or two colors are rendered poorly, the average can still be high enough to achieve a seemingly respectable CRI score.

Doesn't Account for Preference

CRI focuses solely on accuracy compared to a reference illuminant (typically incandescent or daylight). It doesn't account for preferred color rendering. Sometimes, a slight boost in certain colors can be perceived as more pleasing, even if it's not strictly "accurate."

Outdated Reference Illuminants

The reference illuminants used in CRI are not ideal for modern LED technology.

Why TM-30 is Better (in theory)

TM-30 addresses the shortcomings of CRI with two key metrics: Rf (Fidelity Index)

Similar to CRI, Rf measures the average color fidelity. However, it uses a much larger set of color samples (99 Color Evaluation Samples - CES), providing a more comprehensive assessment. Rf of 80 or higher is generally considered good.

Rg (Gamut Index)

This is where TM-30 really shines. Rg indicates whether the light source tends to saturate (Rg > 100) or desaturate (Rg < 100) colors. This is valuable because it tells you how the light source modifies the colors, not just how accurately it renders them. An Rg of 100 means no change in saturation.

The Reality of TM-30 Usage

While TM-30 is a superior metric, you're right that it's not widely used by manufacturers for several reasons: Complexity

TM-30 is more complex to understand and communicate than CRI. It involves two scores (Rf and Rg) and a color vector graphic visualizing the changes in hue and saturation. This can be a barrier to adoption.

Cost

TM-30 testing is more costly than CRI, as you need a more advanced spectrometer and software.

Marketing Inertia

CRI is established. Customers think they understand it. Changing to a new metric requires education and effort, which some manufacturers are hesitant to invest in.

Lack of Regulation

There's no regulatory requirement (in most places) to use TM-30. Manufacturers can continue to rely on CRI, even if it's misleading.

Potential for Negative Results

A TM-30 test might reveal flaws in an LED's rendering that a CRI test would mask. Some manufacturers simply avoid the risk.

Answering Your Questions (Regarding Testing) Yes, people are testing LEDs with TM-30 and comparing to published CRI. Many independent labs, lighting consultants, and even some serious hobbyists with good spectrometers are doing this. However, the results are not always widely publicized. "Construction Grade" LEDs

These are often where the problems are most pronounced. Manufacturers prioritize cost and efficacy (lumens per watt) over color rendering accuracy. A cheap LED might have a high CRI score based on R1-R8, but fail miserably when rendering saturated reds, blues, or greens, resulting in low TM-30 values.

Availability of Data

Unfortunately, you're unlikely to find a readily available database of TM-30 test results for a wide range of "construction grade" LEDs. This is because it's costly and time-consuming to test and publish such data. You'll have to do some digging or conduct your own tests if you need specific data for a particular product.

Recommendations for Choosing LEDs with Good Color Rendering1. Beware of CRI Alone

Don't rely solely on the CRI number. It's a starting point, but not the whole story.2.

Look for R9 Value

If possible, get the R9 value. A good R9 value (above 50, ideally above 70) is a positive sign. Many manufacturers will hide this value because their LEDs perform poorly in this metric.3.

Consider TM-30 (If Available)

If you can find TM-30 data, use it! An Rf of 80+ and an Rg close to 100 is a good target.4.

Look at Spectrums

Pay attention to the SPD (Spectral Power Distribution) chart if available. A smooth, even spectrum is generally better than one with sharp peaks and valleys. Look for a good amount of energy in the red part of the spectrum.5.

Visual Inspection

Whenever possible, visually inspect the light output. Look at a range of colored objects (especially reds, blues, and greens) under the light. Do the colors look natural and vibrant? Are there any noticeable color casts?6.

Test Yourself (If Possible)

If you have access to a spectrometer (even a relatively inexpensive one), test the LEDs yourself. This is the most reliable way to know how they perform. There are even smartphone apps that can give you a rough estimate of CRI, but these are not as accurate as a dedicated spectrometer.7.

Brand Reputation

Stick to reputable brands known for quality. They are more likely to prioritize accurate color rendering.8.

Application Matters

What are you using the LEDs for? The acceptable level of color rendering accuracy varies depending on the application. For example, art galleries and retail spaces require much higher color rendering than a warehouse.

In summary, you're right to be skeptical of CRI alone. TM-30 is a more advanced metric, but its adoption is still limited. Use a combination of available data (including CRI, R9, and TM-30, if possible), visual inspection, and your own testing to choose LEDs with good color rendering. Flag for review