There have been a lot of pictures taken of the laser spot test taken and passed around, but as it is it really doesn’t show much of value. The reason it isn’t a very good test the way it’s being handled currently is neither the naked eye nor any camera can capture the full range of detail in the laser spot, because of the huge range of intensity it contains. So in order to give a more accurate picture of the range of intensity represented I made a series of shots with different exposures.
The method I used was pretty simple I took my DSLR and set it to manual mode. I set my ISO to 100, which is as low as it will go. For those that don’t know this just means the camera will be less sensitive to light, so it will take more light coming in to the camera for something to show up in a picture. The reason for this is I wanted to catch a picture of the brightest part of the laser spot test. I then set my camera to use a shutter speed of 1/8000 of a second, this is my shortest shutter speed. The reason for this is it minimizes the amount of light coming in to the camera when I take a picture. Next I set my aperture wide enough to take a picture in the regular lighting of the room. For those that don’t know the aperture is how wide the hole letting light into the camera is. I set my camera up on a tripod and pointed it at and focused it on the laser spot test template print out designed by @Steve_Johnstone. Next I turned off Auto-Focus so my focus would remain the same for all of my shots. I set up a remote trigger for the camera to eliminate shake. Now I iteratively narrowed my aperture by whole stops and took pictures of the laser spot until I found an aperture where the intensity of the brightest spot dropped noticeably, in my case this was f-16. Then I set the aperture back to one stop wider, so in this case f-11. The reason for all of this was so that I could start out showing only the very brightest part of the laser spot.
Now comes the fun part I then took a series of shots, each successive shot doubling the time the shutter stayed open, until I could see most of the laser spot as it looks to the naked eye. The interesting part about this is that the part of each shot that is white in one shot but was not white in the last shot is about 1/2 a bright as the part of the last shot that was white. What I found is the brightest spot in the laser is about 8000 times brighter, than some of what is so bright it just looks white in a normal laser spot test picture.
My laser spot when I did this test was bad enough so I was having artifacts from the flare. I would be really interested to see how other peoples laser spot tests look at these same settings, they should be vaguely comparable.
All shots taken at ISO 100 / f-11
<img src="/uploads/formlabs/2151/753bc22076fc3779.JPG" width=“345” height=“230"”>
Coincidentally, this is exactly what the flashaholic community (flashlight and LED lighting enthusiasts) does when reviewing flashlight beam profiles
(Although usually with way less steps - it’s usually just to show the hotspot shape vs. the flood.)
I like having it incremented in single stops, the fact that each successive image increases the sensitivity two fold helps to illustrate the magnitude of the range. However it does also take it and flatten it like a log graph of an exponential function which is unfortunate.
Log scale might actually be a good thing.
I took your photos and made a false color composite for fun.
Hope you don’t mind.
Normalized to just the laser intensities and cropped.
Thanks you saved me the effort I’ve been meaning to but had no time to work out how and do it. Would you mind explaining how you did that so others can do the same should the want to do this same process to get a clearer representation of the details of their laser spot.
Also finding a way to weight the smallest center point areas to red so you can tell the highest intensity spot is actually close to round would be good. Wow I wish I had time to research this, and devise a tool for doing this so others could manufacture a better laser spot test image.
- Create a new document in Photoshop, fill the background layer with black
- Stack all the photos in Photoshop, each on its own individual layer.
- Desaturate all layers (do NOT use Black & White, use Desaturate, as that is a surefire way to get just the luminance channel)
- Set all the layers to Screen blending mode.
- Select all the layers (shift+click) and adjust the opacity, while checking the Info panel and hovering over the hotspot of the image - rinse and repeat until the brightest point you can find is at 254,254,254 (leaving one value of headroom just in case)
- When that’s done, switch to Grayscale mode (Image>Mode>Grayscale) and immediately into Indexed Color mode
- Change the palette (Image>Mode>Color Table) to Spectrum
That will yield an image covering (linearly) the whole luminosity scale of all your photos combined (with the false color palette). You can then optionally clamp it, then stretch it so it only covers the laser intensities (without the paper, for example) using Levels (as i did in the second image). Bear in mind that this will result in a loss of scale resolution and make the image NOT comparable to other people’s images.
Oh, that part is easy.
In step 6, you just need to customize the palette to your wishes - e.g. use a blackbody palette as a starting point and make the brightest value green would make it stand out well, or using grayscale and picking a few colors for the brightest few values.
It should be possible to write a simple python script that does this in an automated way. I might give that a try, but not until weekend.
Yeah the current method suffers from exactly what I said earlier it’s a log scale representation of the data making it look like all that red stuff in the middle is all about the same intensity, when really that is the area with the most variability in brightness, meanwhile the outside halo is covered by multiple colors and represents the least amount of variability in brightness.
Ah, so you want an inverse log.
That’s relatively easy to do too, you’d just need to tweak the opacities of the layers.
Yeah it cold likely be done as a function of the shutter speed.
Wow this is really cleaver stuff guys.
I will have to get my wife to have a go at it with her DSLR using the same settings etc. and do the photoshop work as per @Ante_Vukorepa.
Just one thing - do you need to take the photos at night?
No they were taken in normal lighting.
@Steve_Johnstone and @Annino you guys still gonna give this a try?
Yip, hopefully over the weekend
@RocusHalbasch, yes I will try to get around to it this weekend… no tripod tho so I’ll have to get creative.
Only 6 days left. This new limit is ridiculous. This thread is probably going to Close before other people have a chance to take the shots and post them. @Sam_Jacoby is there anything you can do about this ridiculous limit?
We are all a bit short of time at the moment but will see if my wife, Sharon, could have a go at taking the photos today.