My first test / 828

I may be wrong
Good, you are starting to learn the facts of life. Next step is admitting that you actually are wrong :-)

I'm really tired and don't have enough energy to produce scientific comparison, but i uploaded a PSD file that i saved earlier today for my own reference when i was trying to decide between NN and NI. I think the differences are pretty dramatic.

http://hkarapuu.tietoverkot.net/NN-NI.psd

It's about 1mb and contains screen captures from both programs @300% magnification. Profiles were created using just the sample images, and settings were adjusted so that noise reduction amounts were equal. One sample is Oly E-1 sample from phil's gallery, other is Sony's "hidden" F828 sample.

I'll try tomorrow some ready made profiles, hopefully they will help noise ninjas detail eating.. If not, my choice is clear.

--
Henkka Karapuu
http://www.iki.fi/hkarapuu
 
Why not compare quality of pic of Canon D300 and Sony 828? More than comparision on computer monitor I wonder how it is on paper? Which lenses for Canon D300 do you recommend for low budget?

Lep pozdrav iz Nove Gorice,
Boyan
I am tests 828 still under bad light conditions. From 64 to 800 ISO.
New pictures are here.

http://www.e-fotografija.com/artman/publish/article_156.shtml

Matjaz

Picture quality is good for me (for which is camera intended). OK
noise, but if you do not enlage picture to much is very good. I
have some 20 x 30 cm pictures from lab.
But handling. minolta A1 have better.

And don't compare 828 with 300D about quality of picture!

Competitor Minolta A1 article
http://www.e-fotografija.com/artman/publish/article_75.shtml
http://www.e-fotografija.com/artman/publish/article_84.shtml
 
Why not compare quality of pic of Canon D300 and Sony 828? More
than comparision on computer monitor I wonder how it is on paper?
Which lenses for Canon D300 do you recommend for low budget?

Lep pozdrav iz Nove Gorice,
Boyan
Bouth this camera are for different main group. Don't look pictures on monitor. If you want to buy 300D take EF-S lens with body. If you have only body take Sigma 18 - 50 mm.

Look, poglej na stran
http://www.e-fotografija.com/artman/publish/article_123.shtml
http://www.e-fotografija.com/artman/publish/article_150.shtml
http://www.e-fotografija.com/artman/publish/article_117.shtml

Zgoljo
Slovenia
 
Did you see a film in this conditions? HAHA
Very true. After NN processing this reminds me a lot ISO400 film scans. Two differences: the 828 is better because its 'grain' is finer; the 828 is worse because there still is some amount of very low frequency yellowish spots.

Dmitry
828 is not so bad on 800 ISO. But like Matjaz said don't compare
828 with 300D. They are from different main group.

But sharpen?

Zgoljo,
Slovenia
 
Hi Mato,

I have noticed that all but ISO 64 image of your latest series have strong yellow cast. This is weird since you say WB settings were the same (tungsten) during all shots. Could you please find out what made the difference?

BTW this has indirect influence to the noise issue too; albeit this is easy to correct colors, the change ends up with nasty bluish spots all around that make image look absolutely bad and are hard to get rid of.

May you try the same test with custom WB set?

TIA,
Dmitry
Today I have tested 828. More about test and about camera in next
days. If will your interest bigger I will make translation also to
English.

Matjaz

http://www.e-fotografija.com/artman/publish/article_155.shtml
 
Good, you are starting to learn the facts of life. Next step is
admitting that you actually are wrong :-)
Oh, I'm certain that I'm learning a lot about your suppositions. Is that close enough? heheheh... ;-)

The problem with judging between NI and NN is that so much depends upon your familiarity with one as compared to the other.

However, most users seem to prefer the interface of NN. Which in itself may not mean much --- maybe most users are lazy. :-)

--

Ulysses
 
The problem may simply be that the Tungsten preset is simply not doing as well as it ought to do. Oddly, MANY of Sony's cameras (and many other manufacturers as well) do poorly with the Incandescent/Tungsten setting. Others do VERY well.
May you try the same test with custom WB set?

TIA,
Dmitry
Today I have tested 828. More about test and about camera in next
days. If will your interest bigger I will make translation also to
English.

Matjaz

http://www.e-fotografija.com/artman/publish/article_155.shtml
--

Ulysses
 
You've misunderstood me. I didn't ask why there is color cast but why the difference between ISO64 shot and all the rest.

Dmitry
May you try the same test with custom WB set?

TIA,
Dmitry
Today I have tested 828. More about test and about camera in next
days. If will your interest bigger I will make translation also to
English.

Matjaz

http://www.e-fotografija.com/artman/publish/article_155.shtml
--

Ulysses
 
You talking about the one at the bottom of the page? Didn't he edit that one with Photoshop levels? His note mentioned so, if I remember correctly.
You've misunderstood me. I didn't ask why there is color cast but
why the difference between ISO64 shot and all the rest.
--

Ulysses
 
However, most users seem to prefer the interface of NN. Which in
itself may not mean much --- maybe most users are lazy. :-)
Yes, it's easy to get more favourable first impression for NN because of user interface is simplified, the software is more responsive, and default reduction level is lower and thus less plastic. But, at least for me it's the final result that counts.

As i mentioned NN and NI have exactly same basic controls available, but NI offers individual tuning of high,low,med frequency bands and color channels, while NN has just one setting for all.

--
Henkka Karapuu
http://www.iki.fi/hkarapuu
 
Hi Henkka & everyone else,

I've been contemplating adding an "advanced" control set in Noise Ninja that will allow more flexible control over the different frequency bands.

So far, I've intentionally tried to err on the side of simplicity of use, since most users want a simple interface, and they are very happy with the results they can obtain using Noise Ninja even without the extra controls. However, I'd also like to add more control for folks like you who want it. Mostly, I just need to take a little time to work out an appropriate user interface, since internally all of this is already implemented.

Also, the best advice for anyone deciding between the two products is simply to try them both and see which one is more to your personal liking.

I've seen a whole range of feedback regarding effectiveness, ease of use, and productivity for both products, so it's hard to make a blanket statement that, for all users, one tool is superior. Some people even end up using both products; here's an example:
http://forums.dpreview.com/forums/read.asp?forum=1004&message=6950976

The good thing is that customers now have a choice between two effective but differentiated tools, and as long as the market allows us to both stay in business, we'll continue to compete to improve our products, to everyone's benefit.

As always, feel free to contact me by email with any questions, problems, or suggestions concerning Noise Ninja.

Regards,
Jim Christian, Noise Ninja developer
PictureCode
http://www.picturecode.com
 
Hi there, Jim. Thanks for weighing in on this one.

Question for you:
take a little time to work out an appropriate user interface, since
internally all of this is already implemented.
Since NN is already doing these things internally, making interface adjustments won't at all affect program speed and performance, will it?

Is it a matter of adding a tabbed interface, or perhaps an Advanced button to take those users who want it over to a new dialog?
Also, the best advice for anyone deciding between the two products
is simply to try them both and see which one is more to your
personal liking.
Hmmm... just like with cameras. :-)

--

Ulysses
 
Jim, thanks for your reply. It's definitely one plus for noise ninja that you are very responsive for user feedback.

However, i don't feel that Neat Image's more advanced controls are that significant advantage, although they can be handy sometimes.

The reason why I feel neat image is superior is simply because of the results. When equalizing amount of noise reduction or loss of detail between these two programs using same sample, it seems that neat image image has always better results in the other parameter (eg. if removing same amount of noise, neat image will preserve more detail).

Here is one non-scientific comparison for reducing same amount of noise (and mistake in layer names, correct order is NI, NN, NI, NN), which shows rather dramatic difference in favor of neat image:

http://hkarapuu.tietoverkot.net/NN-NI.psd

Naturally i need to do more testing, and use properly created profiles. One possible explanation for the results is that neat image does just better job creating profile with "quick and dirty" method.

--
Henkka Karapuu
http://www.iki.fi/hkarapuu
 

Keyboard shortcuts

Back
Top