My new >3W RGB projector

[White-Light]

That should be fine. A higher CCT white is still white. 2700K looks white by itself, but so does 6500k.

I know.

I think Chroma is more accurate that people give it credit for. The problem is most people (myself included until recently) focus on the very white star shaped area in the centre thats also very small in size and interpret that as the point of white balance.

In actual fact, when you take a stand back and compare actual real world results with chroma and look at the graph carefully, what you can see is that in actual fact, there is a large white triangle at the centre of the graph (highlighted very roughly in black below), that actually represents the white area. If I put the one up without my black triangle highlight, you can see the feint white triangle area for yourself:







Only when you realise that do you realise that you can achieve white anywhere within that large triangle which is why many previous plots were misintrepreted as proving chroma to be wrong.

Whats more I actually think you can use this to your advantage to try to obtain the type of white balance you want.

eg Steve's plot, was commented by him to be a little greenish and can be seen to be on the edge of green but its still within the white triangle area (slightly out on my rough black triangle drawing above but you can clearly see its in the white area when you look at the unannotated version).

To my mind this gives chroma a use in obtaining a white ablance. eg, want a warm reddy white, then plot for the red side of the triangle. Want a cool white, plot for the green side like Steve did or for the blue area for a more bluey white.

I'm pretty sure that with such a wide area within the white triangle pretty much every colour temperature of white is covered.

Its just a pity so many people wrote chroma off so quickly before anyone actually realised that the white balance extended far beyond the star area in the centre, and I'm as guilty as anyone at that. Its just that with further analysis against real world feedback from members on here, I've now realised that the quite large faint white triangle appears to actually be the white balance area not the small starred centre area and that the large area is large because it encompasses many different areas of colour temperature within white.

Now awaits the inevitable flaming.

[instinct]

I never noticed that white triangle and couldn't see it.. and then I tilted my laptop screen down.. WOW, it was so obvious.

[danielbriggs]

In actual fact, when you take a stand back and compare actual real world results with chroma and look at the graph carefully, what you can see is that in actual fact, there is a large white triangle at the centre of the graph.

I think Chroma is more accurate that people give it credit for.

Accurate? Like real-world-mirror-lens-dichro-losses-hell-even-temperature-as-well-as-un-quantifiable-losses accurate?
Nope.

Precise?
Yep.

To my mind this gives chroma a use in obtaining a white ablance. eg, want a warm reddy white, then plot for the red side of the triangle. Want a cool white, plot for the green side like Steve did or for the blue area for a more bluey white.

+

I'm pretty sure that with such a wide area within the white triangle pretty much every colour temperature of white is covered.

You mean like this?

Its just a pity so many people wrote chroma off so quickly before anyone actually realised that the white balance extended far beyond the star area in the centre

No. Hold on a second...
You appear to have somewhat muddled this "Chroma Banter"
It's not the program, governing equations or proved theory I've been shaking a stick at here... That's all well and good.
It's the USE of such a precision tool with no underpinning real world guidance to substantiate and interpret such claims.

The Chromatic Conjecture if you will...



There are plenty of other examples, not just Chroma where tools can very easily generate ill, come-borderline nonsensical data if used without sensible direction.
Take Computational Fluid Dynamics (CFD) as an example. Truly a fantastic and powerful tool. Genuinely.

However how do you ensure the data it spits out is accurate?
Validation!

And how is that achieved?
By using a wind tunnel to confirm the model is correct!


Now relate that example back to RGB lasers...

Now awaits the inevitable flaming.

Bah! You're no fun...