Tag Archives: Topaz Denoise 4 Review

1798 – Don’t Kick it Around

If you’re interested in a continuation of Juha’s and my explorations into high-ISO quality and processing of images made with the LX5, Juha has made another post with an image that is an excellent candidate for noise reduction. Colorful, artificial surfaces without texture, straight lines with hard contrasts. I have chimed in and demonstrated one more time the quality of Topaz DeNoise. This time it stuck, because Juha has bought it and now uses it in Apple Aperture.

What else has happened? Of course Nikon has revealed their EVIL strategy, the so-called CX format, two cameras with a 2.7 crop factor, right in the middle between Micro Four Thirds and quality compacts like the LX5, one camera with a high-resolution electronic viewfinder, one without. It’s called the Nikon 1 system.

Due to the relatively small sensor and the high price there was a lot of ridicule around on the net today, but if you just look at the features and the image quality, there is a lot to like. First of all, these cameras seem to be blindingly fast. Not only can you shoot at 10 fps, you can also shoot short “video” bursts, and then choose the best image of it. It’s not just 1080p video, no, it’s a burst with the whole image resolution of the 10.1 megapixel sensor. You may call this an amateur feature, but believe me, in unpredictable action scenarios it is a heaven-sent.

There is also a new hybrid-use sensor, doubling as AF phase-detect sensor. Cool, and the camera even automatically switches to contrast AF when that gives better results. Thus there is obviously a lot of horse power in its “Expeed III” processor, and we will certainly see more of that in the upcoming DSLRs.

As I said, high-ISO image quality is pretty good, I’d say it is certainly better than that of the D200, and what I can see in the official samples and in those from imaging-resource.com, it can’t be much worse than that of the D300. ISO 800 is beautiful, ISO 1600 seems to be usable, and even ISO 3200 seems to be acceptable for web presentation. Not bad for such a small sensor.


There’s always a “but”, huh? Well, in this case at least for me there are some good reasons why I completely disregard these camera.

  • No square aspect ratio. Really, that is a killer for me. They could easily have done it in software. No problem, but they didn’t. Too bad, I love composing with the square and I will never again buy a camera with electronic viewfinder that doesn’t do squares. Sorry Nikon, everybody else has it, this is inexcusable.
  • No sensor stabilization. There is simply no reason for that, especially with an EVIL camera, and of course the available lenses are either long, stabilized and slow, or short and non-stabilized. Basically that’s Olympus vs Panasonic (Micro Four Thirds) over again. In a way I can understand and excuse it, because Nikon has always stabilized lenses, that’s where their know-how is. If I remember correctly, this makes Olympus the only EVIL brand with stabilized sensors. Almost, that is, because the new Sony Alpha cameras with the semi-transparent mirror are EVIL as well. But then, from their size and feature set they certainly play in the DSLR league.
  • No tilt/swivel LCD. Nikon seems to see this as an expensive consumer feature. The only current DSLR that has it is the consumer top-model, the D5100. I doubt that the D400 will get it. On the other hand, Canon and Sony seem to have realized how useful this feature can be.

Did I ever tell you that I contemplate selling my Nikon D300 and its lenses, and buying a Sony A77? Well, it doesn’t shoot squares either, but it has sensor stabilization and the sensor is, well, sweet 🙂

Well, let’s see what the D400 will be, and then, if I’m honest to myself, I don’t really need a new camera at all 😀

The Song of the Day is “Don’t Kick it Around” by Anita O’Day. It’s on CD 4 of the collection “Young Anita”. Hear it on YouTube.

1796 – Compared To What

Some days ago Juha posted his findings about the new “High ISO Noise Reduction” setting, introduced with version 2.0 of the Panasonic DMC-LX5’s firmware, that was released on September 13.

Juha compared two images at ISO 1600, one developed from RAW using Aperture, and the other a JPEG from the camera, with “HIGH ISO NR” set to “on”. The image was a “twigscape” with slightly out-of-focus background.

Well, I thought I try it myself, but I use my own RAW workflow, trying my best to wrestle the most out of the RAW files, adapting the level of noise reduction to the particular image and trying to sharpen in a way to retain as much detail as possible.

The first comparison is with an image pretty ideal for noise reduction. There are plenty of hard contrasts, and in between we see lots of artificial, uniform areas. ISO is 1600 and you see two exposures: At the left is a JPEG out of the camera with “HIGH ISO NR” set to “off”, in the middle a second JPEG with “HIGH ISO NR” set to “on”, and on the right you see the second image developed from RAW.

The RAW file was converted using the Adobe DNG Converter, then developed in Adobe Camera RAW 4.6. In ACR I’ve completely turned off noise reduction, color as well as luminance. Then, in Photoshop CS3 I have used Topaz DeNoise 5 with a “RAW strong” preset, duplicated the layer, set it to “Luminosity” mode and sharpened that with “Unsharp Mask” (radius 0.3, strength 500, threshold 0). Basically that’s it and the result is pretty obvious.

There is a slight improvement from “HIGH ISO NR = off” to “HIGH ISO NR = on”, but the difference between that and the version from RAW is striking.

OK, that was easy, now let’s look at something notoriously hard, let’s look at foliage. Well, grass or fur would have been even better, but this detail from my living room window in Vienna, seen though the telephoto end of the LX5’s zoom range, is representative enough of a typical “landscape” situation. It was a gloomy, cloudy day (and cold!), so there are contrasts, but much less accentuated, there is not much differentiation in terms of color, and there is a lot of fine detail.

To the left is the JPEG with “HIGH ISO NR” set to “on”. This is a situation that is much harder for noise reduction algorithms – and it shows. On the RAW side I have used much less aggressive noise reduction, and the sharpening layer was tempered with an edge mask. You may have noticed in the last example that the colors from RAW were more vivid than from the JPEG, and so it was here, but while that looked attractive on the artificial surfaces, it looked gaudy and wrong here. Thus I have desaturated the yellows, mixed a tiny amount of magenta into the yellows and finally darkened the image slightly. In the end I got something sufficiently similar in color to the JPEG from the camera, that the eye can fully concentrate on noise and detail.

I think the result is again a win for RAW, although with this material it is less dramatic. There is more detail, yes, but there is also more noise. I’ve tried a few things, different settings of Topaz DeNoise, different sharpening, fiddled with the edge mask, but whatever I did, when I got less noise, the shadows became invariably mushy, just like in the JPEG from the camera. I don’t say it’s the optimum, but I think it’s as far as I would go for a normal Image of the Day. Every tiny bit of additional progress would have to be bought with obscene amounts of time.

So, here we are. I see an improvement possible by setting “HIGH ISO NR” to “on”, but it is not comparable to what you can reach by going through RAW and using a high-quality noise reduction program that you can parametrize according to the image’s needs.

We can also see that ISO 1600 is possible on this camera, albeit only for certain subjects. Artificial surfaces with hard edges and not much texture work best, for landscapes though I would not go higher than ISO 400.

The Song of the Day is “Compared To What” from Roberta Flack’s 1969 debut album “First Take”. Hear it on YouTube.

1305 – Pink Noise Waltz III

There are some last words that I want to say about noise reduction. First, almost all noise reduction plugins on the market work globally, reducing noise in the whole image. I believe that is what Neat Image does, Noise Ninja does so and Topaz Denoise does so as well. Only Nik Dfine is an exception.

Of course you can always use masks to either paint the noise reduction onto the image, just like you do it with Noise Ninja’s “noise brush”, or use an inverted luminance mask to tie the intensity of the noise reduction to tonal value, or use a mask derived from a color range selection. In that way Nik Dfine’s control points (called U-Points) are just another, admittedly very clever way to define a mask.

I have never used Nik Dfine, but I know Nik’s U-Point technology from Nikon Capture NX2. It is a very easily understandable way of selection and of direct control. I have no doubt that it works for noise reduction just as well as it does for contrast, brightness and saturation.

Topaz Denoise, on the other hand, works globally, but it also offers plenty of control, and it does so in very meaningful ways. You can control the overall strength of the effect, adjust separately for the red and the blue channel, for the shadows and the highlights, you have separate sliders for recovering details and for removing blur (which is a kind of sharpening, but not exactly so), and you can even add in a variable amount of fine monochrome noise. But although this is plenty of control, even my first efforts came out better than what Noise Ninja would offer me. Since then I have played around many times, often with the images that I have already shown you, and in almost all cases I can do even better now.

To conclude this, Topaz Denoise works reasonably well with one of the presets, and with experience and some effort you can turn the results to excellent. While the advantage over other plugins, when using the presets, is not that big, it becomes dramatic as soon as you know what you’re doing and as soon as you invest a minute tweaking an image. Thus if you are looking for a good noise reduction plugin, Topaz Denoise should be among those that you try. I’m sure you’ll like it.

This is hail in the Image of the Day. We had quite some thunderstorm on Monday, and I was lucky to not get caught in the open 🙂

One last time the Song of the Day is “Pink Noise Waltz” from the Diablo Swing Orchestra’s 2006 album “The Butcher’s Ballroom”. Hear it on on YouTube 🙂

1304 – Pink Noise Waltz II

As promised, here’s more about Topaz Denoise 4, the new version of the noise reduction plugin, that Topaz claims gives you four additional stops. Again I compare to Noise Ninja, but now I not only use auto-profiling, now I use camera noise profiles as well.

Paul Butzi’s comment to yesterday’s post caused me to download and install Noise Ninja noise profiles for the D200 and the D300. Of course I have tried them before, at least for the D200, and all I could remember, was that I had not been overly impressed. Now I know why.

This is a detail that I have also shown in yesterday’s post. I have omitted RAW and instead added the profiled Noise Ninja. Yes, there is more detail, but in order to deliver detail, Noise Ninja leaves much coarse noise in the image. In fact, the result does not look much better than the RAW file without noise reduction.

The next detail is from yesterday’s train image. This is from the bright part with some kind of word or number scribbled on the side of the wagon. Topaz Denoise almost completely smudges it away, and the profiled Noise Ninja actually does a marvelous job.

Unfortunately this is over as soon as we get into the dark parts. The profiled Noise Ninja is really bad at removing color noise.

The next image is quite old and it was taken with the D200, a Nikon 50/1.8 at f4, 1/40s, -0.33EV and ISO 1600. The effective ISO is somewhere around 2000. The image is interesting, because it has a lot of architectural detail, combined with a still blue sky.

The first detail from this image is of one of the most difficult areas. The dark, low-contrast upper part of the remote building is easily smudged away into kind of a dark cloud. In fact that’s exactly what happens when you use any of Topaz Denoise’s presets.

In the second detail Noise Ninja shows mottled sky again.

And here is a final look at the sky. Again the low-frequency color noise is Noise Ninja’s major weakness.

I will conclude today’s post (that actually counts not for today, but for yesterday, Sunday) with two details from the Image of the Day. This was taken from the escalator of a shopping center, but actually I have no idea what it really is that we’re seeing here.

I have taken this image with the Nikon D300, my Sigma 20/1.8 at f8, 1/200s and ISO 6400. This is a correctly exposed, well-lit image, more or less a best case scenario.

It’s not that Noise Ninja were bad, not at all, but again Topaz Denoise 4 delivers the best results. Just look at the texture of the white fur on the right side of the first detail, and then look at the two lamps in the next detail.

The conclusion so far is clear: Topaz Denoise 4 wins over Noise Ninja, regardless of camera profiles.

The Song of the Day is still “Pink Noise Waltz” from the Diablo Swing Orchestra’s 2006 album “The Butcher’s Ballroom”. Just as yesterday, it can be heard on on YouTube 🙂

1303 – Pink Noise Waltz I

It’s Sunday night, I’m on the train to Vienna and I have not taken a single image all through the weekend. I’ve processed quite a few though.

In the last post I promised to tell you more about my experiences with Topaz Denoise 4, the latest contender to the throne of “Best Noise Reduction Utility”. Personally I own Noise Ninja (latest version), and for a comprehensive comparison I also wanted to use the test versions of Neat Image and Nik Dfine 2.0. Unfortunately the test version of Neat Image does not work on 16 bit images and the test version of Nik Dfine 2.0 immediately claimed that the 15 day demo period was over. I suppose the latter is connected to the fact that I installed the plugin as an admin user and then used it under my normal account. Whatever, the comparison is now limited to Noise Ninja, a plugin that I have used for years, and the new Topaz Denoise 4.

I have used Noise Ninja in the same way that I use it always. Basically I let it auto-profile the image, and then I use the defaults. If I feel that too many details are smudged, I lower the degree of luminance noise reduction from the default of 10 down to a minimum of 7. That’s it. I never use their “Noise Brush”, because I find it inconvenient, but in rare cases I may apply a mask to the denoise layer, probably based on luminance, probably based on color.

Thus what we compare here is Noise Ninja, used in that particular way, with the best I can get out of Topaz Denoise 4. Thus I arguably put more effort into Denoise than into Noise Ninja. That’s right, but that is because I feel the controls in Denoise are more useful. Noise Ninja is much faster than Topaz Denoise, but when I begin to tune the output of Noise Ninja or optimize by blending two different versions, e.g. one for lighter tonal values and one for the darker values, things begin to be pretty awkward. Topaz has just that concept built in: you can vary the effect depending on tonal value.

Whatever. That’s what I do. I use Noise Ninja in a pretty generic way because it couldn’t inspire me in years to put more effort into optimization. I do more optimization with Topaz Denoise, because I can and because it is easy and intuitive.

Before we look at the first bunch of results (another post with more of that will follow), let’s think about what we can realistically expect:

When we used film, “noise” was called “grain”, and it was a physical property of the particular type of film used. High sensitivity meant big crystals and that meant coarse grain. If you wanted to take photographs at another sensitivity, you needed to use another film, or to a certain degree you could underexpose the film and vary chemical process, temperature and time. This was called “pushing” the film.

In digital photography there is no such thing as a film that you could change. You have a sensor, when the sensor is exposed to light, it produces a certain electric signal, and that’s it. The different ISO settings are there for our convenience and because photographers expect them. In reality the digital ISO value specifies an amplification.

Digital noise is exactly the same as the noise you can hear on analog tape recordings. Analog? Yes, analog. The signal from the senor is analog, and only then it gets quantified and digitized.

Remember a very noisy tape recording. Even with the best of ears you couldn’t hear any fine details. The melody was recognizable, but dynamics and clarity were gone.

And exactly the same happens in digital images. The “real” pattern of tonal values is buried in noise, but because we don’t know what part of the signal is “real” signal and what part is noise, they become indistinguishable. The dynamic range of a noisy image is diminished, the effective resolution goes down.

That’s what we see in the Image of the Day. I took this image sometime last winter, using the Nikon D300, the Sigma 20/1.8, f3.2 at 1/80s and ISO 6400. The focus was on the white billows in the middle of the street. Like almost always when I have set the camera to ISO 6400, it was also set to B&W, but of course that affects the JPEGs only.

The full image was made with Topaz Denoise 4. In Adobe Camera RAW I have completely disabled noise reduction and I have set the exposure parameters so that the histogram fills the whole range. Then I have prepared four different details for you:

This first detail shows an area where we have some sharpness and some soft transitions in the billows. The RAW image is completely unusable, Noise Ninja has eliminated much noise but also most of the texture, and while Denoise comes out smoothest and a pretty soft, it has more detail than Noise Ninja and none of that mottled look that comes from low frequency color noise. In my book Denoise wins easily.

Noise Ninja always produces its own kind of fine noise. At print size, much of this noise will, if visible at all, be mistaken for detail. Denoise is cleaner here as well, but the difference is not that big. At print sizes or downsized for web, this is negligible. Actually this detail is sort of a best case: well-defined lines, good contrast, no delicate details. Both do fine, Denoise probably a little better.

This is the downfall for Noise Ninja. Again we see the mottled look, awkwardly obvious because we know the street is more or less monochrome. Noise Ninja does away with fine, but preserves coarse detail, and here it gets visible in an unpleasant way.

The last detail from this image shows more of the same. Low frequency color noise, the windows in the background are hardly recognizable in the Noise Ninja version, while they are clearly defined in the Topaz version. For this image, D300, ISO 6400, night scene, Topaz Denoise is definitely the better choice.

The next image, an old one, was taken with the Nikon D200, a Nikon 50/1.8 at f1.8, 1/30s, ISO 900 and -1EV, thus we’re really looking at an equivalent of ISO 1800 here. The D200 has a much noisier sensor, thus we are in desperate need of noise reduction.

The first detail is a spectacular win for Topaz Denoise. Just look at the shadow detail. Incredible.

Now we finally get to an advantage of Noise Ninja. Both do quite well with the blue spot, but look at the digit “3” below the orange lever, look at the three letters right of it and look at the lines around them. They are clearly visible in RAW, Noise Ninja preserves some of it, but in Denoise almost all is gone.

We see the same thing in the final detail. Again the fine letters are gone in Topaz Denoise, while Noise Ninja preserves at least part of them.

That’s it for today. I am one day behind and may be so for the rest of the week. I hope to catch up next weekend. In the next post, counting for Sunday, I will show you a few more interesting details.

The Song of the Day is “Pink Noise Waltz” from the Diablo Swing Orchestra’s 2006 album “The Butcher’s Ballroom”. Fantastic music, hear it on YouTube.