I probably forgot because I hadn't seen so many overdone HDRs popping up all over the place at the time. Now everyone and their brother are doing them, and waayyy over doing it. IIRC I just used the curves to keep certain parts from going completely black (trees, road) and completely white (areas around lights, building, sky above building and reflection in road).





I still say bah to the HDR mania.

I love HDR images... I don't see what's spacial about this one though....

How many exposures is that, and are you saying you applied curves to each individual image before merging?

I love HDRs when implemented correctly. I also don't mind the HDRs that are way over the top, I just don't see them as photographs anymore at that point. More like digital art, but still interesting.

Originally Posted by Wyn

I love HDR images... I don't see what's spacial about this one though....

Spacial?

Originally Posted by tardypizza

How many exposures is that, and are you saying you applied curves to each individual image before merging?

No, iirc I applied curves to the merged image to compress the midtones and bring highlights and shadows back from oblivion.

It was either 3 or 5 images with a total range of I think 5 stops. One image would never have been able to keep all the detail from the dark to light areas.

It's interesting that I also don't like overdone HDR's but a common and accepted practice in astrophotography is to stack many, if not hundreds of different exposures to bring out the most detail and properly expose both the lows and highs of bright stars and faint nebulosity. For most celestial objects, stacking is the only way to present the captured data in a meaningful way.

It's not exactly the same methodology as HDR in practice, but it's still the same theory. Trying to display way more information at the same time than the monitor is capable of showing.

Originally Posted by tardypizza

It's not exactly the same methodology as HDR in practice, but it's still the same theory. Trying to display way more information at the same time than the monitor is capable of showing.

It isn't the same methodology? I would figure it might be, except perhaps they apply a linear compression to the dynamic range to make it visible, which for a photograph isn't very attractive.

It's definitely not a new technique but it sure has become popular as of late.

I meant it's not the same method of capture. As you know, in an HDR, you expose 3 or 5 or 7 exposures at the same ISO at different compensation stops. What I do a lot in astro work is expose for different lengths of time at varying ISOs. Like for the Orion Nebula, it's common to do several shortish exposures at low ISO to properly expose the bright core of stars. Then several medium shots at ISO 400 for the green/blue hues in the nebula. Then several long exposures at ISO800 to get the very faint red hues. Exposure compensation alone could never get all the range, and that's what I meant by not exactly the same.

Sounds like the same thing to me. You change what you can to effectively create multiple different exposures. The details of how you do it don't matter much.

We're splitting hairs here, that's why I keep saying it's the same theory, different methodology.

Originally Posted by tardypizza

I meant it's not the same method of capture. As you know, in an HDR, you expose 3 or 5 or 7 exposures at the same ISO at different compensation stops. What I do a lot in astro work is expose for different lengths of time at varying ISOs. Like for the Orion Nebula, it's common to do several shortish exposures at low ISO to properly expose the bright core of stars. Then several medium shots at ISO 400 for the green/blue hues in the nebula. Then several long exposures at ISO800 to get the very faint red hues. Exposure compensation alone could never get all the range, and that's what I meant by not exactly the same.

Wow, that's interesting to see that you're going to that length to get a single image... Why do you have to change ISOs though? Seems like you'd be able to get it with just changing the shutterspeed....

The main thing you're battling is the telescope's tracking. The differences in brightness is huge between stars and faint red nebulosity (that the IR cut filter tends to block anyway). Imaging stars is easy, hence low ISO and short exposures. To get any red with an unmodified DSLR you're looking at very long exposures. And since tracking at long focal lengths is anything but easy, the best compromise is to limit your exposures to < 1 minute. You can compensate for the reduction in exposure time by bumping up the ISO to get more nebulosity. And the added noise at higher ISOs is easily dealt with by taking a series of "darks" (same thing your camera does with long exposure noise reduction), averaging them and canceling out noisy pixels that way.