This is what HDR is... I think :S

[Alasdair]

Ok guys,

So, I have had a go at summarizing/explaining HDR as I understand it to be (in a very basic manner).
Here is a link to that post

http://macleod.arknet.co.uk/photoblo...01/what-is-hdr

I know this is a pretty complicated subject but I have given it my best bash lol.
Please feel free to comment and correct me in any areas I am not 100% right.

Thanks for all your help and advice on here.

Regards
Alasdair MacLeod
http://www.macleod.arknet.co.uk

[Colin Southern]

Good effort - I suspect that in writing the article you clarified a lot in your own mind. Just a few thoughts came to mind whilst I was reading it ...

- An EV of 0 equates to a 1 second exposure at F1, not a "correctly exposed scene" ... what you (and most others) is referring to is "delta EV" or the change in exposure from a reference ("normal") exposure.

- In a normal HDR bracket it's a waste of time taking shots 1 stop apart - 2 stops will get you the same result, and you thus need less of them.

- "even though the display device would normally not be able to reproduce the full range of luminance values in an un Tone Mapped image." ... 99.99999% of monitors can only reproduce about 1/2 the range of a SINGLE RAW exposure, let alone an HDR composite.

- "but others love the super real/fantasy/art/painting effect." - I think you mean "surreal" - nothing "super real" about over-processed images!

- For me, the most common way of producing an HDR image is to simply use a GND filter at the time of the capture, but this technique doesn't appear to get a mention. Another technique is to transition between two (or more) images using a layer mask (or equivalent), but no mention of that either.

- Probably the only other thing that springs to mind is that "there's really no such thing as a HDR final image" - any image that can be displayed on a regular monitor will have quite a low dynamic range.

Hope these comments help

[Alasdair]

Wow Colin,

Thanks for the super fast reply.

Where do I start lol.

With regards to delta EV, how would you suggest that paragraph be written to explain that section correctly. From what you have said I need to go back to the drawing boards on that one lol. If im honest, I haven't heard the term delta EV yet (I am very new to photography at this point).

"99.99999% of monitors can only reproduce about 1/2 the range of a SINGLE RAW exposure, let alone an HDR composite."

Isnt that one of the reasons we tone map the image? To allow the details we want to see to be brought out.

"I think you mean "surreal""

Yeah, Ill go change that now lol :P

"For me, the most common way of producing an HDR image is to simply use a GND filter at the time of the capture, but this technique doesn't appear to get a mention. Another technique is to transition between two (or more) images using a layer mask (or equivalent), but no mention of that either."

Maybe its because im new to photography (I started Jan this year), but I don't know what a GND filter is and I don't know how to transition between two or more images using a layer mask yet. This is why I didn't mention them. I did say it was a brief overview of what I understand HDR to be lol :P (yes, again I am sticking my toung out at you).

"Probably the only other thing that springs to mind is that "there's really no such thing as a HDR final image" - any image that can be displayed on a regular monitor will have quite a low dynamic range."

you know what i was saying lol.

Regards
Alasdair MacLeod
http://www.macleod.arknet.co.uk/photoblog

[Colin Southern]

Wow Colin,

Thanks for the super fast reply.

You're welcome

Where do I start lol.

At the beginning

With regards to delta EV, how would you suggest that paragraph be written to explain that section correctly. From what you have said I need to go back to the drawing boards on that one lol. If im honest, I haven't heard the term delta EV yet (I am very new to photography at this point).

I don't know if "delta EV" is a correct term ... I just use the term "delta" as a generic way of saying "Relative" or "a change of"

Isnt that one of the reasons we tone map the image? To allow the details we want to see to be brought out.

Yes, although in essence all we're really doing is just compressing the total range (potentially in a non-linear way).

Maybe its because im new to photography (I started Jan this year), but I don't know what a GND filter is and I don't know how to transition between two or more images using a layer mask yet. This is why I didn't mention them.

GND is Graduated Neutral Density filter - it's a bit like those sunglasses that had the darker bits at the top. Singh-Ray make the best ones. You might find some of the articles I've written for them on the use of their filters interesting as well (or then again, you might not!). To use a layer mask technique you timply take (typically) 2 shots; one exposed for the sky - one exposed for something else - chuck them both onto the same image as layers - and then paint a mask that hides the bad bits of both, revealing the good bits.

[proseak]

"Another technique is to transition between two (or more) images using a layer mask (or equivalent), but no mention of that either." Well, I've been experimenting with just this technique - not my own idea (found it on luminous landscapes, I think)...Let's see!

1) Open both images in PS. Go to the darker of the two, Ctrl+A, Ctrl+C, Ctrl+tab(you should now be on the lighter image), and Ctrl+V. Click on the "add layer mask" icon at the bottom of the Layers palette. Now, make the background layer active, and Ctrl+A, Ctrl+C. Alt+click on the mask on layer 1 - it should go white. Ctrl+v, and you should see a greyscale version of the background. Blur "to taste", say 15-50 Gaussian. At this point, you can just click on the background layer in the Layers palette; with any luck, that will be it.

There is another point of control that you may find useful. However, at this point it goes a bit version specific - i use CS3, and for me the way to go is Window> Arrange >New Window for <image name>...I do know that it was different in CS2. This new window allows you to preview the effect on the final image of tweaking the mask - I often steepen the curve to achieve a cleaner knockout, for instance.

HTH

Peter

[Colin Southern]

1) Open both images in PS. Go to the darker of the two, Ctrl+A, Ctrl+C, Ctrl+tab(you should now be on the lighter image), and Ctrl+V. Click on the "add layer mask" icon at the bottom of the Layers palette. Now, make the background layer active, and Ctrl+A, Ctrl+C. Alt+click on the mask on layer 1 - it should go white. Ctrl+v, and you should see a greyscale version of the background. Blur "to taste", say 15-50 Gaussian. At this point, you can just click on the background layer in the Layers palette; with any luck, that will be it.

There is another point of control that you may find useful. However, at this point it goes a bit version specific - i use CS3, and for me the way to go is Window> Arrange >New Window for <image name>...I do know that it was different in CS2. This new window allows you to preview the effect on the final image of tweaking the mask - I often steepen the curve to achieve a cleaner knockout, for instance.

HTH

Peter

Sounds like a lot of work. I just open both images - use the move tool to flick one onto the other - add a layer mask and then paint the mask black / white / gray etc as required.

[proseak]

Sounds like a lot of work. I just open both images - use the move tool to flick one onto the other - add a layer mask and then paint the mask black / white / gray etc as required.

It's actually very easy, and very fast(both appeal to me!) - after the first time, I could do it as fast as my fingers could move, and certainly faster than I'm typing this. The real time-saver is that almost all of the mask-painting is done for you by the machine, something else that appeals to me

Regards

Peter