Ad
Ad
Ad
Pages: [1] 2 »   Bottom of Page
Print
Author Topic: 2.4 now available for download  (Read 7399 times)
Rhossydd
Sr. Member
****
Offline Offline

Posts: 2002


WWW
« on: June 24, 2009, 01:13:50 AM »
ReplyReply

Adds support for more cameras and some bug fixes.
Logged
barryfitzgerald
Sr. Member
****
Offline Offline

Posts: 609


« Reply #1 on: June 24, 2009, 05:48:38 AM »
ReplyReply

I guess we will have to wait till LR 3 for a fix to the mushy high ISO raw problem.

Logged
Morris Taub
Sr. Member
****
Offline Offline

Posts: 372



WWW
« Reply #2 on: June 24, 2009, 05:43:40 PM »
ReplyReply

Quote from: barryfitzgerald
I guess we will have to wait till LR 3 for a fix to the mushy high ISO raw problem.

Hi Barry...can you tell me what this mushy high iso problem is please?...I mean I've developed images at 1600 iso without problem...is it higher iso's?...
Logged

phila
Sr. Member
****
Offline Offline

Posts: 267



WWW
« Reply #3 on: June 25, 2009, 03:42:35 AM »
ReplyReply

Anyone else not been able to install it?

I've downloaded the dmg (twice!) but am unable to progress past the first 'page' of the installer as the "Continue" button is greyed out. One strange thing I've noticed is that, after double clicking on the dmg it takes about 30 seconds to open the installer, displaying a progress bar in the meanwhile that says "Attaching...".

OS 10.5.7 & LR v2.3
Logged

francois
Sr. Member
****
Offline Offline

Posts: 7002


« Reply #4 on: June 25, 2009, 04:43:11 AM »
ReplyReply

Quote from: phila
Anyone else not been able to install it?

I've downloaded the dmg (twice!) but am unable to progress past the first 'page' of the installer as the "Continue" button is greyed out. One strange thing I've noticed is that, after double clicking on the dmg it takes about 30 seconds to open the installer, displaying a progress bar in the meanwhile that says "Attaching...".

OS 10.5.7 & LR v2.3
No problem. I'm also on Mac OS X 10.5.7. After verifying the .dmg (about 5 seconds), it only take 2-3 sec to mount the image. Is there any error message in the Console (/Applications/Utlilities/Console.app).



Logged

Francois
phila
Sr. Member
****
Offline Offline

Posts: 267



WWW
« Reply #5 on: June 26, 2009, 02:53:48 AM »
ReplyReply

Quote from: francois
No problem. I'm also on Mac OS X 10.5.7. After verifying the .dmg (about 5 seconds), it only take 2-3 sec to mount the image. Is there any error message in the Console (/Applications/Utlilities/Console.app).

Nothing in Console.

Here are screen grabs:

www.philaphoto.com/images/LR-1.jpg
www.philaphoto.com/images/LR-2.jpg
Logged

francois
Sr. Member
****
Offline Offline

Posts: 7002


« Reply #6 on: June 26, 2009, 04:23:59 AM »
ReplyReply

Quote from: phila
I would repair disk permissions and re-download the file with a different browser.
Another thing you can do: try to install the file from a different user account.
Logged

Francois
phila
Sr. Member
****
Offline Offline

Posts: 267



WWW
« Reply #7 on: June 26, 2009, 06:08:21 AM »
ReplyReply

It finally worked after a Restart. Who knows why?!?
Logged

francois
Sr. Member
****
Offline Offline

Posts: 7002


« Reply #8 on: June 26, 2009, 06:27:38 AM »
ReplyReply

Quote from: phila
It finally worked after a Restart. Who knows why?!?
Glad to hear that your problem is solved. Hdiutil was probably having some kind of issues and got stuck (hdiutil is the tool that deals with hd images).
Logged

Francois
Rhossydd
Sr. Member
****
Offline Offline

Posts: 2002


WWW
« Reply #9 on: June 28, 2009, 05:01:28 AM »
ReplyReply

How are people finding this update ?

I'm finding it less responsive and generally slower than 2.3.

Paul

Running on Win XPP sp3
Logged
stevebri
Full Member
***
Offline Offline

Posts: 176


« Reply #10 on: June 28, 2009, 01:42:09 PM »
ReplyReply

Hi Paul,

I'm using it on Vista 64 bit with 8g of ram and it seems fine.

two things to consider;

Firstly go to edit>preferences>general>go to catalogue settings>general>relaunch and optimize
Secondly, do you often defrag your drives and registry and general 'puter clean ups...?

i ask because i do both my machines once a fortnight and the response and general speed up is very noticable.

What bits is 2.4 most sluggish for you?

S
Logged
James R
Sr. Member
****
Offline Offline

Posts: 260


« Reply #11 on: June 28, 2009, 03:38:44 PM »
ReplyReply

Quote from: stevebri
Hi Paul,

Firstly go to edit>preferences>general>go to catalogue settings>general>relaunch and optimize
Secondly, do you often defrag your drives and registry and general 'puter clean ups...?

Thanks for reminding me why I moved to Macs.  
Logged
DarkPenguin
Guest
« Reply #12 on: June 28, 2009, 04:36:25 PM »
ReplyReply

Quote from: Rhossydd
How are people finding this update ?

I'm finding it less responsive and generally slower than 2.3.

Paul

Running on Win XPP sp3

It crashes less than 2.3.  Vista 32bit.
Logged
paulbk
Sr. Member
****
Offline Offline

Posts: 469



« Reply #13 on: June 28, 2009, 07:36:05 PM »
ReplyReply

2.4, so far so good. No crashes yet. Much better performance than 2.3.

2.3 was a nighmare.

xp sp3
Logged

paul b. kramarchyk
Barkhamsted, Connecticut, USA
barryfitzgerald
Sr. Member
****
Offline Offline

Posts: 609


« Reply #14 on: July 01, 2009, 02:27:19 PM »
ReplyReply

Quote from: momo2
Hi Barry...can you tell me what this mushy high iso problem is please?...I mean I've developed images at 1600 iso without problem...is it higher iso's?...


The issue is Adobe are apply non avoidable noise reduction on raw. Obviously low ISO not much of a problem, but higher ISO levels and it's very obvious. Different cameras will have different results, Sony raw's are by far the worst, though LR is not doing a great job on my Minolta ones either.
Other problems from this include low kelvin temps and colour noise being smeared into images (and near impossible to remove in ACR)

It's the smearing effect that is the problem. The solution is amazingly simple, Adobe should NOT be applying base level NR to raw files.
Download a trial of capture one 4 or raw therapee..and you will see what is going on.
Logged
oldcsar
Full Member
***
Offline Offline

Posts: 126


« Reply #15 on: July 01, 2009, 07:37:45 PM »
ReplyReply

Quote from: barryfitzgerald
The issue is Adobe are apply non avoidable noise reduction on raw. Obviously low ISO not much of a problem, but higher ISO levels and it's very obvious. Different cameras will have different results, Sony raw's are by far the worst, though LR is not doing a great job on my Minolta ones either.
Other problems from this include low kelvin temps and colour noise being smeared into images (and near impossible to remove in ACR)

It's the smearing effect that is the problem. The solution is amazingly simple, Adobe should NOT be applying base level NR to raw files.
Download a trial of capture one 4 or raw therapee..and you will see what is going on.

Is it really as big an issue for most users as it is for you? As Jeff Schewe suggests, looking at 1:1 crops may be equivalent to magnifying an image 2x-4x in relation to typical print size. It is entirely true that ACR is using a small amount of NR in their demosaic process, but I believe this does not have a detrimental effect on micro-detail. I can only speak for its effects on Canon Raw (Canon 300D/30D), but what I see compared to other converters (C1 4, Rawtherapee) is that it is only removing a fine layer of noise which is either equal size to the real detail, or slightly smaller. C1 4 and Rawtherapee do preserve this fine layer across the image, but this is not real detail and becomes an obstacle when sharpening. When sharpening Canon files in C1, they can quickly become 'crunchy' when noise reduction is not used at all, but employing a tiny amount allows the details to be sharpened and helps avoid the sharpening of noise (and reduces the cross-hatching artifacts that can sometimes be seen while adjusting files in DPP, C1, etc).

I bring this up to express how a small amount of NR may be beneficial at the Raw conversion stage. I will also admit that this smearing effect, as you would like to call it, was much worse in earlier versions of ACR (or LR 1.1 or LR 1.2 for instance). At this stage in development, I don't think that Adobe will pull a developmental U-turn with their demosaicing technology and offer different demosaicing techniques (such as those seen in Rawtherapee). There is also the fact that Minolta cameras were never on par with Canon CMOS technology even at the time of the 20D, and so it is probably also the case that Minolta files look worse with Adobe's demosaicing because there is less detail to start with (especially if we're talking the Maxxum 7D, which was only 6MP and could not stand up to the clarity of high-ISO Canon files. This is not to say that Minolta cameras cannot still take great pictures- I believe they can- but the low-resolution, noisier Minolta sensors of around 2005 probably do not handle ACR processing as well as cleaner, higher resolution sensors. But, let's be realistic, who is going to make a huge enlargement of high-ISO 6MP shot (for instance) from a noisy sensor and expect great results? Because really, a big enlargement would be the only instance where this "smearing" might be apparent.

For the sake of clarity, I will repeat by saying that ACR should not actually smear any fine detail. What it does is remove a fine layer of noise, and this micro-noise will not be seen in reasonably sized prints (as long as one is not pushing huge sizes with files from older, noisier sensors). Additionally, RAW files can be more effectively sharpened with a small amount of noise reduction, C1 included IMHO. If you can actually show that ACR is removing fine detail in your files and that it is not simply fine noise, I will revise my statement. I do think that the grumblings about ACR NR has decreased significantly in the recent builds of ACR, and I think this is because it is actually working reasonably well for most people. For those who don't like the results, it may be the case that they do not understand what they are seeing with respect to 1:1 crops and mistaking noise for meaningful detail.
Logged

Schewe
Sr. Member
****
Offline Offline

Posts: 5542


WWW
« Reply #16 on: July 01, 2009, 10:58:43 PM »
ReplyReply

Quote from: oldcsar
I don't think that Adobe will pull a developmental U-turn with their demosaicing technology and offer different demosaicing techniques (such as those seen in Rawtherapee).


I wouldn't bet on that (which should tell ya something right there).

Since the Camera Raw engineering teams is rather small and headed up by Adobe's "sacred cow" (a fellow with the initials TK), if the team things something needs modification and knows how to do the modification, I'll guarantee that the boys will make a U-turn (on a dime too).

But that's the thing...they do need to be convinced that something needs to be done (or fixed). Given the motivation and the time frame (the time is the real hard part) the engineers are capable of making great strides between versions....and if that means totally revamping the demosiacing, redoing the noise reduction, the sharpening and even adding things (some already hinted at like lens corrections for the Panasonic) I think many people might be surprised at what the team may have in store for the future...can't really say anything mind you (NDAs don't ya know) but it's just my experience sorta burbling up...

:~)
Logged
Andrew Fee
Jr. Member
**
Offline Offline

Posts: 87


WWW
« Reply #17 on: July 02, 2009, 01:44:30 AM »
ReplyReply

Quote from: James R
Thanks for reminding me why I moved to Macs.  
Sadly, this is more of a myth than anything else. Yes, HFS+ generally does a good job of keeping files defragmented, but it does a terrible job of keeping free space defragmented.

What this means is that you might find you suddenly go over some threshold point and disk performance is absolutely terrible. That said, I just loaded up iDefrag to check the status of my drive after seeing this post.

Due to some problems I've been having lately, I just wiped my Mac and reinstalled a couple of days ago, and already the drive is pretty fragmented. (and this is files, not free space)

(the bar along the bottom is the status of the whole drive, red is fragmentation)


It would be great if there's a new demosaicing algorithm in Lightroom 3, as I'm constantly running into problems with images from my 1000D. (yeah, I know it's a low-end camera, but that shouldn't matter)

I don't have any example images to hand, but it's very similar to this example from the DxO site:

(DxO's conversion for comparison)


I also agree that it would be great if that base noise reduction could be removed for a finer grain structure. I realise that it sounds like I just want DxO… well, I like some of the technical aspects of what they can do with RAW files, but I much prefer Lightroom's interface and find it's much easier to get the look I want from images using it.



As for 2.4… it all seems to be running smoothly here. Can't say I've really noticed any performance loss. (or gains) Don't think I've ever had Lightroom crash on my Mac, and I use it all the time. (ever since the 10.5.2 update or so, it's had the ability to crash Spaces if I switch to Lightroom—which should take me to the space it's on but doesn't—and then manually go to that space)
Logged
barryfitzgerald
Sr. Member
****
Offline Offline

Posts: 609


« Reply #18 on: July 02, 2009, 05:44:15 AM »
ReplyReply

Quote from: oldcsar
Is it really as big an issue for most users as it is for you? As Jeff Schewe suggests, looking at 1:1 crops may be equivalent to magnifying an image 2x-4x in relation to typical print size.

My comments are not on crops, but on prints.

Yes it does make a difference.



I bring this up to express how a small amount of NR may be beneficial at the Raw conversion stage. I will also admit that this smearing effect, as you would like to call it, was much worse in earlier versions of ACR (or LR 1.1 or LR 1.2 for instance). At this stage in development, I don't think that Adobe will pull a developmental U-turn with their demosaicing technology and offer different demosaicing techniques (such as those seen in Rawtherapee). There is also the fact that Minolta cameras were never on par with Canon CMOS technology even at the time of the 20D, and so it is probably also the case that Minolta files look worse with Adobe's demosaicing because there is less detail to start with (especially if we're talking the Maxxum 7D, which was only 6MP and could not stand up to the clarity of high-ISO Canon files. This is not to say that Minolta cameras cannot still take great pictures- I believe they can- but the low-resolution, noisier Minolta sensors of around 2005 probably do not handle ACR processing as well as cleaner, higher resolution sensors. But, let's be realistic, who is going to make a huge enlargement of high-ISO 6MP shot (for instance) from a noisy sensor and expect great results? Because really, a big enlargement would be the only instance where this "smearing" might be apparent.


I was not really talking about CMOS v CCD, as far as the KM cameras go, other makers such as Nikon and Pentax were using them too. I have been entirely happy with the 6mp CCD for high ISO work.
If other software such as C1 4 can handle them very well, then I feel ACR can be improved in this respect
The CMOS sensor (8mp) is likely better for high ISO than the 6mp CCD, however I don't think the gap is that large. And we are wandering off the topic here!

BTW the 6mp CCD was not made by Minolta, but by Sony..who also provided it to Nikon, Pentax etc etc



For the sake of clarity, I will repeat by saying that ACR should not actually smear any fine detail. What it does is remove a fine layer of noise, and this micro-noise will not be seen in reasonably sized prints (as long as one is not pushing huge sizes with files from older, noisier sensors). Additionally, RAW files can be more effectively sharpened with a small amount of noise reduction, C1 included IMHO. If you can actually show that ACR is removing fine detail in your files and that it is not simply fine noise, I will revise my statement. I do think that the grumblings about ACR NR has decreased significantly in the recent builds of ACR, and I think this is because it is actually working reasonably well for most people. For those who don't like the results, it may be the case that they do not understand what they are seeing with respect to 1:1 crops and mistaking noise for meaningful detail.



What ACR does is evidently base level NR. I am no technical expert, but I can see the effects of this lead to large big grain areas, and it can also smear chroma noise. When you use ACR you have a harder time removing the noise with 3rd party programs such as noise ninja etc..simply because it is more smeared to start with.

I think Adobe should leave the NR to the user, and not attempt to remove any..the sliders for NR are basic, but I mostly use chroma NR in ACR. I don't want luminance NR, so for me 0 should be 0

I will point out LR is a fantastic program and of great use to photographers. Not to take away from their superior workflow..but I would like to see this addressed, I believe that this view is supported by many as well.
« Last Edit: July 02, 2009, 05:45:07 AM by barryfitzgerald » Logged
oldcsar
Full Member
***
Offline Offline

Posts: 126


« Reply #19 on: July 02, 2009, 06:03:36 PM »
ReplyReply

Quote from: barryfitzgerald
What ACR does is evidently base level NR. I am no technical expert, but I can see the effects of this lead to large big grain areas, and it can also smear chroma noise. When you use ACR you have a harder time removing the noise with 3rd party programs such as noise ninja etc..simply because it is more smeared to start with.

I think Adobe should leave the NR to the user, and not attempt to remove any..the sliders for NR are basic, but I mostly use chroma NR in ACR. I don't want luminance NR, so for me 0 should be 0

I will point out LR is a fantastic program and of great use to photographers. Not to take away from their superior workflow..but I would like to see this addressed, I believe that this view is supported by many as well.
Smearing chroma noise? Really? I remember hearing about problems with particular Alpha models and earlier versions of ACR, with respect to noise reduction being embedded by Sony into the RAW file. My understanding was that a firmware update improved this issue. I do remember seeing horrendously blotchy crops from Sony cameras converted by ACR in camera reviews online. Nonetheless, part of me wonders if part of the problem is due to the signal to noise ratio of these affected cameras, in relation to other cameras which are not expressing this chroma smearing (for instance, I have not seen this phenomenon with my own cameras but that doesn't mean it doesn't exist). If signal to noise is not so great to begin with at high-ISO, can we really expect great enlargements, regardless of the converter?

Nonetheless, your point is well-taken... we should be able to choose whether or not to have base NR. What I was interested in challenging you with is the premise that this base NR is a serious issue across camera makers, which I don't think it is according to my own experience making 20x30 enlargements from high-ISO shots (CR2). Additionally, that 1:1 crops may not be relevant to the noise characteristics in the actual print.

Thank you Jeff for your hint. It is good to know that Adobe will be addressing the problem for those who are concerned by the base NR effect. In principle, I think the idea of NR at the demosaicing stage offers some advantages with respect to capture sharpening. It should be improved upon, not necessarily thrown out. If it did not offer some sort of advantage, I dare say that it has been a moronic decision to put off 'fixing' it for so long. I personally would like to believe that it has not been a moronic decision, and that something can be gained from this approach, albeit with some refinement. Choice is good for everyone though, no question about that.
Logged

Pages: [1] 2 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad