I had quite a busy Saturday. Starting at 9:00 am, I attended a workshop produced by my friends John & Susan of Catchlight Studios. I’ll post my own images later, but you can get a sample of what it was like on this post on their blog. One problem I started at that workshop dogged me for the rest of the day – heat exhaustion. It was hot & muggy and, though I was drinking fluids (both water and my ever-present Diet Dr. Pepper), I found myself feeling wiped out.
After that workshop, I stopped by my gym to shower and change clothes, had lunch, and then went on to the Star Wars Celebration V at the Orange County Convention Center. Lots of people dressed up, lots of interesting sets, and lots of walking. The lighting inside convention centers isn’t that great. I treated it like a low-light event and shot at high ISO. However, take a look at this set of portraits by Rafael Concepcion. This is one of the reasons I really admire the guy’s take on things. He treated the show as an opportunity for HDR portraits and created some stunning results.
After the show, I attended the Last Tour to Endor party at Disney’s Hollywood Studios. The revamped the Indiana Jones live-action show there for the special event – Indiana Jones and the Jedi Temple of Doom. Presenting it as a Star Wars fan film, they did a great job of making a cross-over mashup scenes from both trilogies that blended very well. It also gave me another opportunity for some high ISO shooting from the stands.
Modern cameras do an outstanding job of reducing noise at high ISO, but they still need some help – particularly when you create 100% crops like the ones I’m about to present. I have two products in my arsenal for reducing noise, so I thought it could be interesting to put them against each other in a head to head test.
Let’s start with the original image from the Indiana Jones show. I selected this image crop because it contains a selection of skin tones, different fabrics and other materials which display high ISO noise in different ways. Let’s be honest. I also selected it because it shows Adrianne Curry dressed as “Slave Leia” in the show. How often do I get to post a picture of America’s Top Model?
This is a crop from the RAW image with no processing at all. Take a look at how the noise affects her skin, the flowing material on her…I suppose it’s a skirt. Then look at the striped fabric of the dude she’s punching. Look at the green “skin” of the Gamorrean Guard in the background, or the black pants in the background. There’s noise on all of it, but it affects each area differently. When you consider that I shot this photo at ISO 4000, it’s not bad. However, it can get better.
The image above is from NIK Software’s DFine 2.0. I selected a pretty strong Contrast Noise setting of 175 and a Color Noise setting of 110. As you can see, the noise is almost non-existant in this photo. The only adjustment to this image is from DFine 2.0 – nothing else. I like the result, but I can see some noise on the floor matting. I also notice a slight loss of detail on furry clothing covering the Gamorrean Guard’s crotch area (I can’t believe I just wrote that).
This image uses Topaz DeNoise 5. I used the RAW Image Strong setting. I think it cleans up much more noise, but at the loss of some details. Once again (and I can’t believe I’m going to write this), look at the furry cloth covering the Gamorrean Guard’s crotch area. Compared to the original, and even the NIK version, it looks blurrier. Still, I like the overall results we’re seeing here. The detail on other parts of clothing in the image seems quite acceptable to me.
I’m not going to pick a winner here. Partly because I’m not sure if the settings I chose were really comparable between the products – I just moved some sliders or pressed a preset button to get a result I liked. The other reason is because each image has some strengths over the other, so it’s a matter of preference. The real reason is to show that neither is a loser. Both NIK and Topaz make products that do an outstanding job of reducing noise quickly and easily.

.jpg)
.jpg)
.jpg)




