Ad
Ad
Ad
Pages: « 1 ... 6 7 [8] 9 »   Bottom of Page
Print
Author Topic: Sony RX100...  (Read 50822 times)
mac_paolo
Sr. Member
****
Offline Offline

Posts: 424


« Reply #140 on: September 26, 2012, 11:32:05 AM »
ReplyReply

Was wondering if anyone has been running LR 4.2 Beta to process their RX100 RAW images?  Do you convert the images in 4.2 and go back to 4.1 to edit?  Or is 4.2 stable enough to complete the editing, and final printing? 
[...]
It's a RC (Release Candidate). It means that it's stable enough for everyday use. It's not a beta.
It overwritten my 4.1 and works flawlessly for me since then. Smiley
Logged
thierrylegros396
Sr. Member
****
Offline Offline

Posts: 691


« Reply #141 on: September 26, 2012, 11:35:23 AM »
ReplyReply

LR4.2 RC is as stable as a final version Smiley Wink

Magnetic fixed filter will be the best solution Wink

Thierry
Logged
One Frame at a Time
Full Member
***
Offline Offline

Posts: 169


« Reply #142 on: September 26, 2012, 12:28:06 PM »
ReplyReply


Magnetic fixed filter will be the best solution Wink


It seems like a good alternative but I am concerned about placing a strong magnet onto a device that depends on moving electrons about on a nano scale Huh

Glad to hear about 4.2! Thanks!!
Logged
mac_paolo
Sr. Member
****
Offline Offline

Posts: 424


« Reply #143 on: September 26, 2012, 12:59:45 PM »
ReplyReply

LR4.2 RC is as stable as a final version Smiley Wink
Not exactly. A GM would be quite as stable as a Release Candidate. Beside that I didn't find any serious bug Smiley
Logged
One Frame at a Time
Full Member
***
Offline Offline

Posts: 169


« Reply #144 on: September 26, 2012, 01:04:23 PM »
ReplyReply

Whats a GM?
Logged
kikashi
Sr. Member
****
Offline Offline

Posts: 4088



« Reply #145 on: September 27, 2012, 01:41:35 PM »
ReplyReply

Whats a GM?

A genetically modified version  Wink

Alternatively, a golden master.

Jeremy
Logged
AFairley
Sr. Member
****
Offline Offline

Posts: 1199



« Reply #146 on: September 27, 2012, 02:10:04 PM »
ReplyReply

Just saw this filter adapter (threaded ring is attached to front of lens with double-sided tape).  Filters I don't care so much about, but this would let me use a lens hood, which I do care about.  Of course, and attached lens hood destroys the compact form factor of the camera (somewhat).  http://www.youtube.com/watch?v=pNv6IXJZ3xI
Logged

thierrylegros396
Sr. Member
****
Offline Offline

Posts: 691


« Reply #147 on: November 24, 2012, 10:45:03 AM »
ReplyReply

To wake-up the post and give more information about lowest ISO Wink Wink


I've made a lot of tests this afternoon, and now I've understood almost everything.

So, I've made manual shots (@f3.5) of a colorchecker, 1st 1/4s @ISO80, 2nd 1/5s @ISO100, 3rd 1/6s @ISO125.
All shots are properly exposed (ETTR) and need only a slight correction (max 0.1Ev) to obtain exactly same exposure in LR4.2.

Results: RAW with same exposure exhibit more noise @ISO100 and even more @ISO125, so DXOMark DR results are correct !

JPG doesn't need any exposure correction, exposure really identical, but it's clear that ISO125 is progressively compressed in the highlights, but neither ISO80, nor ISO100, so DPReview results are correct too !

So, if you shoot RAW, use ISO80 if it's possible.

Also noticed that the 2 last patches of my colorcheker were blinking (@ISO80 and 100), but RAW data were not clipped !!!


Have a Nice Day.

Thierry
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7930


WWW
« Reply #148 on: November 25, 2012, 12:15:47 AM »
ReplyReply

Hi!

What happens is that exposure is actually reduced. So if a certain patch would see 1000 photons at 1/4s it would only see 666 photons at 1/6s. From photon statistics, SNR would be 31.6 at 1/4s and 25.8 at 1/6s. That is physics, nothing software can do anything about. The JPEG processing pipeline probably does some noise reduction and sharpening (among many other things) and will take ISO setting into account.

Best regards
Erik


To wake-up the post and give more information about lowest ISO Wink Wink


I've made a lot of tests this afternoon, and now I've understood almost everything.

So, I've made manual shots (@f3.5) of a colorchecker, 1st 1/4s @ISO80, 2nd 1/5s @ISO100, 3rd 1/6s @ISO125.
All shots are properly exposed (ETTR) and need only a slight correction (max 0.1Ev) to obtain exactly same exposure in LR4.2.

Results: RAW with same exposure exhibit more noise @ISO100 and even more @ISO125, so DXOMark DR results are correct !

JPG doesn't need any exposure correction, exposure really identical, but it's clear that ISO125 is progressively compressed in the highlights, but neither ISO80, nor ISO100, so DPReview results are correct too !

So, if you shoot RAW, use ISO80 if it's possible.

Also noticed that the 2 last patches of my colorcheker were blinking (@ISO80 and 100), but RAW data were not clipped !!!


Have a Nice Day.

Thierry
Logged

thierrylegros396
Sr. Member
****
Offline Offline

Posts: 691


« Reply #149 on: November 25, 2012, 06:08:16 AM »
ReplyReply

More tests this morning that confirm results of DXO Wink Wink

I've made 9 shots of the colorchecker in manual mode @ISO100, the first to be just below onset of clipping, others with speed doubled each time (-1EV steps).

Developped in LR with exposure compensation.

Up to +5EV, no problems, but after I've to tweak with curves to obtain the good correction.

318 is -6EV, 316 is -4EV, 313 is -2EV, 311 is correct exposure.

As you can see, results are very good up to -6EV.
Logged
thierrylegros396
Sr. Member
****
Offline Offline

Posts: 691


« Reply #150 on: November 25, 2012, 06:17:17 AM »
ReplyReply

The last one -8EV had to be tweaked with massive NR (100 lum. and 50 chrom.), some contrast and some vibrance to restore colors.

Of course, it's far from a perfect result !!!

So, if the DR of a colorchecker is 6EV, fairly good results of the -6EV exposure confirms the 12EV DR found by DXO.

And NR of LR4.2 allows me to obtain 1 more EV.

Have a Nice Day.

Thierry


Logged
thierrylegros396
Sr. Member
****
Offline Offline

Posts: 691


« Reply #151 on: November 25, 2012, 12:54:21 PM »
ReplyReply

To compare, I've tested my old G10, very disapointed Embarrassed

1392 is properly exposed, 1394 is -2EV, 1396 is -4 EV, 1396 is -6EV.

But the biggest difference is that even @ISO80 the properly exposed shot exhibit noise in the blue channel !!!

Hopefully LR4.2 has a very good noise reduction system that works well up to -4EV with no drawbaks at all.

But below -4EV it's another matter.
« Last Edit: November 25, 2012, 01:06:48 PM by thierrylegros396 » Logged
thierrylegros396
Sr. Member
****
Offline Offline

Posts: 691


« Reply #152 on: November 25, 2012, 01:04:02 PM »
ReplyReply

So, my G10 DR is just under 10EV @ISO80, with really worse SNR18% than the RX100.

The last G10 shot (-8EV), just to show you that it's really unusable.

But in real world, with good light, G10 is not crappy, and sometimes I prefer G10 result over RX100 !!!

Why, it's very difficult to explain, just like you may prefer valve audio amp Wink Wink Wink

Thierry

« Last Edit: November 25, 2012, 01:11:10 PM by thierrylegros396 » Logged
hasselbladfan
Sr. Member
****
Offline Offline

Posts: 407


« Reply #153 on: November 25, 2012, 05:40:57 PM »
ReplyReply

Michael,

How does it compare to the Sigma DP2M?

Pro's / con's of both? Independent of price difference.
Logged
hasselbladfan
Sr. Member
****
Offline Offline

Posts: 407


« Reply #154 on: November 27, 2012, 09:36:15 AM »
ReplyReply

I guess Michael is on vacation ...... Sad

....... or he is still figuring out which one is best. Smiley
Logged
K.C.
Sr. Member
****
Offline Offline

Posts: 662


« Reply #155 on: November 27, 2012, 10:12:46 PM »
ReplyReply

Or he doesn't want to indulge the question of what the differences are between apples and oranges.
Logged
AFairley
Sr. Member
****
Offline Offline

Posts: 1199



« Reply #156 on: November 28, 2012, 10:46:42 AM »
ReplyReply

Thierry,

Thanks for the research.  DR is more important to me that noise on the RX100, so I always have shot at 125 assuming that's what will give me the most DR.  What could you tell about where the DR "sweet spot" is from the tests you ran?
Logged

thierrylegros396
Sr. Member
****
Offline Offline

Posts: 691


« Reply #157 on: November 28, 2012, 10:58:07 AM »
ReplyReply

From all my tests, it appears that you'd better choose ISO80 for RAW.

But if you use Jpg ISO125 will give you the best results.

Just pay attention that "Blinking Clipping Indicator" is extremely conservative (about 0.7EV), so you don't need to add a safety margin when using ETTR in RAW  Wink Wink Wink

Have a Nice Day.

Thierry
Logged
jeremypayne
Guest
« Reply #158 on: November 28, 2012, 11:26:44 AM »
ReplyReply

DR is more important to me that noise on the RX100, so I always have shot at 125 assuming that's what will give me the most DR. 

Bill Claff's data shows that the photographically useful dynamic range is as follows:

 - ISO 81 ... 8.73 stops
 - ISO 100 ... 8.41 stops
 - ISO 126 ... 8.71 stops

http://home.comcast.net/~nikond70/Charts/PDR.htm
Logged
jeremypayne
Guest
« Reply #159 on: November 28, 2012, 11:30:56 AM »
ReplyReply

Or he doesn't want to indulge the question of what the differences are between apples and oranges.

I believe Michael may have said he was going to compare the Sony RX1 with the Sigma and this could be the source of confusion here ...
Logged
Pages: « 1 ... 6 7 [8] 9 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad