Ad
Ad
Ad
Pages: « 1 [2] 3 4 5 »   Bottom of Page
Print
Author Topic: I'm finding DxO better than Lightroom (dives for cover...)  (Read 29260 times)
stewarthemley
Guest
« Reply #20 on: October 02, 2012, 03:48:37 PM »
ReplyReply

I think it would be MORE helpful to SHOW us what convinced you.k
Dear Jeremy

It would be more useful to YOU if YOU played with it, like I did, and found out if it was convincing for YOU.

Honestly, I have no agenda here. I don't give a flying f*** if other people agree with me or not. I found it improved my images and so wanting to help others I posted.

And if people like Eric can suggest ways to help me get more out of LR then that's a bonus for all of us.

Why does it have to be a pain trying to help people?
Logged
robgo2
Sr. Member
****
Offline Offline

Posts: 344


WWW
« Reply #21 on: October 02, 2012, 07:33:11 PM »
ReplyReply

Totally agree about the interface. Just did about thirty images in DxO and by number ten I was crying out for LR's simplicity. I guess they want the thing to be as automatic as possible, which it almost is, but then when you want that extra bit of tuning, which most of us do, it's a right royal pain. I'm not sure how I'm going to react with my next batch of several hundred images, which I do about every ten days. Maybe I'll regret my original posting!

Back in the day when I was a steady user of DxO Optics Pro, I was unconvinced by the developer's claim that optimal results could be achieved by using preset adjustments.  Individualized adjustments are almost always going to give superior results, regardless of the program.

Rob
Logged
JimAscher
Sr. Member
****
Offline Offline

Posts: 336



WWW
« Reply #22 on: October 03, 2012, 07:31:25 AM »
ReplyReply

As I work entirely in black-and-white, I begin with a mass conversion of my RAW takes with DxO Pro (which I invariably find in comparison with the initial RAW provides an improvement).  I pass these images to Lightroom, principally for reviewing and cropping, then conclude by processing selected images in Silver Efex Pro 2 for eventual printing.  For me, this is the simplest procedure.   
Logged

Jim Ascher

See my new SmugMug site:
http://jimascherphotos.smugmug.com/
ario
Full Member
***
Offline Offline

Posts: 121


« Reply #23 on: October 03, 2012, 11:45:46 AM »
ReplyReply

I am giving a try to DXO but I have not found the way to move my work between different computers as I do with LR, Aperture and Capture One.
I usually have both originals and catalogs/libraries/settings on portable HD's which I can easily move from one station to an other working always on the most updated version.
Apparently I cannot do the same with DXO but may be I am missing something.
Any suggestion will be highly appreciated.
Ario
Logged
Slobodan Blagojevic
Sr. Member
****
Offline Offline

Posts: 5951


When everybody thinks the same... nobody thinks.


WWW
« Reply #24 on: October 03, 2012, 12:36:49 PM »
ReplyReply

... Why does it have to be a pain trying to help people?

Stewart,

I think you might have a problem with your spell-checker, i.e., its Auto-Correct part. I think you meant to write:

"Why do I have to be a pain trying to "help" people?

Damn you Auto-Correct!
Logged

Slobodan

Flickr
500px
Keith Reeder
Sr. Member
****
Offline Offline

Posts: 252


WWW
« Reply #25 on: October 03, 2012, 12:50:48 PM »
ReplyReply

Was that really necessary, Slobodan?

Sometimes when an uncalled-for smartarse comment comes to mind, it's best just to leave it there.
Logged

Keith Reeder
Blyth, NE England
Slobodan Blagojevic
Sr. Member
****
Offline Offline

Posts: 5951


When everybody thinks the same... nobody thinks.


WWW
« Reply #26 on: October 03, 2012, 01:29:35 PM »
ReplyReply

Was that really necessary, Slobodan?

Sometimes when an uncalled-for smartarse comment comes to mind, it's best just to leave it there.

Just as necessary as some dude on the Internet posting what he had for breakfast, or another dude posting "I prefer product A to product B." Just as I do not give a damn what some dude on the Internet had for breakfast, I just as well do not give a damn that another dude on the Internet prefers A to B.

Now, I am sure that the "dude" in question is a smart guy and a fine gentlemen in person, but so far I only know his forum persona, and, as I far as I am concerned, this is the first post I remember him for. So, why on Earth would I care what he thinks about A vs. B? Or what he had for breakfast?

Unless, of course, he would post examples so that I can see for myself. His repeated whining that examples do not make sense as they reflect his subjective criteria is just as annoying. His main claim is "...DxO extracts significantly more detail from these cameras without planting artifacts..." i.e., that the demosaic algorithms and detail recovery are superior. That statement leaves little room for subjectivity. While color and tone preferences are subjective (although it is also possibly to quantify them in terms of color accuracy and dynamic range), detail is much less so. It is there or it is not, nothing subjective about it, and easy for everyone to see. What is subjective about detail is, of course, how significant the difference is (i.e., the debate between 'incomparable' and 'barely noticeable under ideal circumstances').

To illustrate demosaicing superiority, all one has to do is to post the first thing that comes out of both converters, without any other processing (i.e., his subjective preferences for contrast, highlights, color, etc.). If one wants to be thorough, there are three options to present the initial rendering: sharpening zeroed, default sharpening, and the optimum sharpening for each converter (the latter is, of course, subjective, but as long as it is labeled as such, I will be free to form my own opinion).

That would be helpful for some guy from the Internet.

If, however, Jeff Schewe would post something like : "Hey guys, I just tried converter X and it is significantly better than L...," I would be downloading it before even finishing that sentence. Not because I am a sucker for authority (I am anything but), not because someone else proclaimed him as such, but because I know him for years, on this forum or else, and I formed my own opinion of him as authority.

If, however, some dude on the Internet said the same, I really, really couldn't care less, let alone expose myself to the trouble of downloading it, installing it, learning the basics, trying it, de-installing it, then chasing and cleaning all the bits and pieces an installation leaves across the system...

Logged

Slobodan

Flickr
500px
stewarthemley
Guest
« Reply #27 on: October 03, 2012, 03:58:06 PM »
ReplyReply

Wow. What did I ever do to you, Slobodan? Must have been pretty bad.

So I think I'll just bow out here. I'm sorry if I offended anyone else. Just trying to be helpful. Good luck guys.
Logged
kers
Sr. Member
****
Offline Offline

Posts: 756


WWW
« Reply #28 on: October 03, 2012, 04:06:10 PM »
ReplyReply

Ok lets put some meat in the dinner !
I have here an image developed in DXO 7.51 and adobe Raw CS6.

what you need to know :
I am working  with adobe raw so DXO would need some fine tuning - the colors for instance...

d800e  24mm1,4G  d8 1./320sec on tripod liveview-sharp
so about as good as it gets

development in Adobe  camera colors ( not adobe )
everything set to zero in the first workspace called Basic( color : as shot)
lenscorrection profile used + removal of chromatic aberration
sharpened 60 %  amount    0,6 radius   60 detail
-- photoshop smart sharpened  50%, 0,1,0

development in DXO ( i am never sure if it does what i wanted)

DXO default Neutral  color (RAW only)( color : as shot)
it uses its dxo lensprofile
after that in photoshop:    25% smart sharpen 0,1

download HQ full-size Jpegs: 

http://www.beeld.nu/LL/dxo-adobe.zip

cheers !

Pieter Kers
« Last Edit: October 03, 2012, 04:11:50 PM by kers » Logged

Pieter Kers
www.beeld.nu
kencameron
Sr. Member
****
Offline Offline

Posts: 669



WWW
« Reply #29 on: October 03, 2012, 05:10:22 PM »
ReplyReply

Interesting, thanks for these. The DXO colors certainly need work. I am left seeing a possible case for DXO, but not enough of one to incline me to stop using Lightroom. I am also left with some sympathy for the OP's reservations about the limited value of comparing individual posted shots. It would be nice, for example, to have a comparison between the best work of experts in both programs - your post gives the impression that you are more expert in Adobe/Lightoom. Still, there is no doubt that we now have something to discuss rather than just vent about.
Logged

kers
Sr. Member
****
Offline Offline

Posts: 756


WWW
« Reply #30 on: October 03, 2012, 05:52:49 PM »
ReplyReply

to Ken Cameron: et al

The adobe colours give a very strange unsaturated blue- that is why i use ' camera colours'
so also adobe has a problem i think...
My idea is that the good lensprofile of DXO is what it makes it better.
I see benefits of both but like the workflow and colours of adobe-
I like the differentiation of the treeleaves in DXO but do not like the colours (but probably they can be addressed) nor some other details that are clearly less good.
I had done this test before with a 14-24mm lens and there DXO had less detail. - here it is about even.
To do good testing you have to put a lot of time in it- i do not have...or want.
From one image you cannot make a judgement. But it gives an idea.

to Stewart Hemley: is this in line with your findings?
Logged

Pieter Kers
www.beeld.nu
kencameron
Sr. Member
****
Offline Offline

Posts: 669



WWW
« Reply #31 on: October 03, 2012, 06:45:16 PM »
ReplyReply

Yes, all this makes sense when I look closely at the posted images. I mean no offence in what I said about your expertise in DXO. Just that you said you are never sure it does what you wanted. Maybe nobody does.
Logged

allegretto
Sr. Member
****
Offline Offline

Posts: 466


« Reply #32 on: October 03, 2012, 07:52:10 PM »
ReplyReply

for one-button an then just a bit of fine tuning to suit your needs, it's pretty darn good
Logged
stewarthemley
Guest
« Reply #33 on: October 04, 2012, 03:06:47 AM »
ReplyReply

Back again. I was going to bow out but you can't give in to the people who want to bring their own problems to the discussion...

Kers,  first, thanks for doing the comparison. As you mention, there are so many variables, and both LR and DxO are quite complicated to really get the best out of, more so DxO! So I stand by my view that it's more instructive for individuals to do their own experimenting. And checking around this forum, there are many posters who don't include examples, in fact most don't, but lets not go there again!

My feelings are roughly this: the developers of DxO make a wrong decision by making it so difficult to go manual. Yes, the auto/default settings are close, to my tastes at least, but you (I) ALWAYS have to fine tune. One reason I spend time checking other converters is that I want the best image I can get, and that is very unlikely to be someone else's (default/auto) settings.

Also agree that it is lens dependent. Some lenses respond better to LR, but I have found (checking about 10 lenses from various makers) that with most, the lens profile in DxO has the edge. Check for things like barrel distortion.

Re colors: looking at my screen, I like Adobe colors but then when I do a side-by-side, I definitely prefer DxO. Especially leaves, grass, the natural colors. Of course, you can get them similar but that's quite a bit of work, including presets/profiles, etc which is just too much for me. And that is always going to be personal taste.

Re workflow: no brainer!!! I would like DxO to have a further option to the simple and advanced they have now. Manual. Then I think we could make easier comparisons. I have a built-in dislike of programs that make decisions for me, so I suppose I must be looking for trouble even trying DxO.

To those who don't want to do the testing, which I agree is time-consuming, that's fine, obviously. And LR (and C1) are good programs so if you stick with them, you're hardly losing out.  It's all in the quest of getting the result you (and your clients if you're a pro) like best. Competition is so tough these days that when I'm not being paid to get images I feel it's an investment to see how I can improve the ones I've got. YMMV.
« Last Edit: October 04, 2012, 03:10:28 AM by stewarthemley » Logged
kencameron
Sr. Member
****
Offline Offline

Posts: 669



WWW
« Reply #34 on: October 04, 2012, 05:35:48 AM »
ReplyReply

My feelings are roughly this: the developers of DxO make a wrong decision by making it so difficult to go manual.

My experience so far, as a Lightroom user trialling DXO, is that DXO often gives me a starting point which I like better, but further tweaking is incomparably easier with Lightroom, so the final version which I like better almost always comes from Lightroom. As an amateur who can play as long as he likes with single images, this leaves me thinking I probably won't persist with DXO.
Logged

Ligament
Full Member
***
Offline Offline

Posts: 178


« Reply #35 on: October 04, 2012, 09:32:48 PM »
ReplyReply

I did not find this true with past versions of DxO, but with the latest version and D800e files I am quite impressed. I agree the UI is poor, and ability to fine tune to taste is difficult and non-intuitive.

With some of my files, DxO presets get it right the first time with a click. I can't really understand why or how, and I'm almost ashamed to admit it.
Logged
stamper
Sr. Member
****
Offline Offline

Posts: 2723


« Reply #36 on: October 05, 2012, 03:39:42 AM »
ReplyReply

If you were using presets that DXO manufactured then you are relying on the program to make decisions as to how your image looks and not using your vision as to how it looks. I don't see the point of having an advanced DSLR and not taking the time to process the output in a way that suits you. It is your image and you should be responsible for the final outcome?
Logged

BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3685


« Reply #37 on: October 05, 2012, 04:35:08 AM »
ReplyReply

Can anyone explain, in plain english, what deconvolution does?  I have googled it, of course, but explanations like this don't work for me. And I have tried looking at what it does, by playing with the slider, but can't see any consistent impact.

Hi Ken,

Since your question was left unanswered, I'll have a go at it.

Try visualizing blur (e.g. from our optics) as the signal intended for one pixel, spilling over a bit to neighboring pixels. There is more spill over to the immediate neighbors, a little less to their neighbors, and so on. The distribution of the spill over amounts can be characterized in a mathematical 'formula' (a convolution kernel, AKA Point Spread Function (PSF) ), and that formula allows to mathematically reverse the spill, back to the original pixel, for for each and every pixel. Hence, the restoration of the original detail is called deconvolution (because it reverses the convolution).

That deconvolution operation works best when the exact PSF (shape of the blur pattern) is known. The theoretical benefit of DxO is that it has a more accurate knowledge of how that PSF looks, and differs across the image, and at different focus distances, at each aperture setting. Lightroom/ACR and other Raw converters typically only allow to dial in one shape for that PSF, dictated by the Radius setting and Amount, for the entire image. A Raw converter like Capture One Pro allows to vary the Amount towards the corners to compensate for corner sharpness fall-off.
 
Cheers,
Bart
Logged
kers
Sr. Member
****
Offline Offline

Posts: 756


WWW
« Reply #38 on: October 05, 2012, 05:34:42 AM »
ReplyReply

Now i also have tried two other Raw converters on the same photo above.  NX2, Nikons own and the new Photo Ninja.
The outcome differs from type of photo and lens used so it seems. NX2 does a remarkable job on this photo where i found Adobe raw much better in other cases.
It seems that a good lensprofile is key to the success and off course Nikon has the best one in this case. I am a user of Nikons TS lenses and there lensprofiles do not work- so are not used- too bad.
I got very different outcome with a photo shot with a 45PCE lens where photoninja was king and adobe raw turned flat and 2d.
So the more i see the more i get puzzled.  Good thing is that there are no real bad converters.
If you really are serious about one photo you could develop it in all 4 and decide.. But also then you probably find parts where one is better than the other.
edit: Photo Ninja is the new kid on the block and seems to be a very good Raw converter- at least in my d800e case. I am new to the software but already get some great results.




« Last Edit: October 05, 2012, 07:25:41 AM by kers » Logged

Pieter Kers
www.beeld.nu
jeremypayne
Guest
« Reply #39 on: October 05, 2012, 06:29:01 AM »
ReplyReply

Back again. I was going to bow out but you can't give in to the people who want to bring their own problems to the discussion...

Get real, buddy.  You aren't any kind of victim.

You made a claim.  You should reasonably expect that peope would like see the evidence.

Your refusal to share your results is a bit strange and your over-reaction to being asked is even weirder.

Oh well ... Moving on ... Nothing to see here, unfortunately ...
Logged
Pages: « 1 [2] 3 4 5 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad