D3/L1 Vario-Elmarit photo quality

[spylaw4]

Advertisement (gone after registration)

I think it is about time that we had a D3/L1-fest thread like the D2 one, and I see that the topic has already been broached there. Also it would be nice if we could have more identifiable D3/L1 photos posted in the Photo forum so that we can get a better idea of the Vario-Elmarit's way of 'painting' images.

 

In this thread please try to stick to the standard Vario-Elmarit and NOT go off into other lenses (Oly, Leica R, Rokkor etc.).

 

Personally, I am still getting to grips with the L1 and its multitude of permutations, but so far I have found the following (using firmware v1.1) and using Lightroom V1 for processing:

 

+ Set the WB manually - Auto tends IMO to be inconsistent as evidenced when one shoots a series of the same subject with the same lighting. The color temperature/deg.K setting procedure is a bit tedious, but if one has plenty of time (not always possible) to set up then could be a good idea.

 

+ Autofocus is ok, but using the centre zone only is more reliable and quicker.

 

+ Use OIS Mode 2 if only because it avoids the noise!

 

+ I primarily use ISO 200

 

+ The Film Mode (FM) setting appears to affect the RAW image (try shooting the same scene using the various standard FM settings and compare the unadjusted RAW files). Apart from some initial quick looks at the jpegs accompanying the RAW files (not nearly as good as the D2 so far) I have not seriously investigated jpeg quality yet.

 

+ I have tried Dugby's suggestions for FM settings but found them to be to my eyes a little harsh. I will however be re-visiting this aspect (see my jpeg comment above).

 

+ So far the Vario-Elmarit tends to 'paint' softer/silkier than the D2's Vario-Summicron even with the in-camera sharpening turned right up to max, and even using stopped down manual aperture settings.

 

I don't think I've missed out anything vital that would affect the image quality, but I'm sure that the question will be asked if I have.

 

So maybe we might get a D3/L1 thread started rather like the D2-fest but with an eye to finding a consensus on the best settings? Comparing it with the D2 might be more appropriate here, than there?

 

Over to you!

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[spylaw4]

With apologies to the original posters but, for the sake of completeness and to avoid having to dot back and forward, here are some relevant posts from the D2 thread ......

 

@chkphoto:

Do any D3 owners feel that that camera has the image quality/look of the D2?

 

@audiaudi:

Yes and No ... personally, I prefer the look of the LC1/D2 images but it's hard to compare them directly to the L1/D3 images because the advantages offered by one camera relative to the other change depending upon the circumstances in which they're being used. Also, the L1/D3's image files are 50% larger than the LC1/D2 image files, which means one or the other has to be resampled in order to compare them directly to each other.

 

But your question has started me thinking about an experiment I should (and will!) perform when I have a little time later this week ... stay tuned.

 

@dugby:

Yes and Yes.......

 

1. The Jpegs out of the D2 are superb and usable immediately. The jpegs outs out of the L1 are similar to the D2's. However the JPEGs created by Silkypix from the L1's RAW are much much better.

 

I need to go back and compare these to JPEGs from the D2 RAW files....now

 

 

2. The camera's themselves are interesting. Now that I have my D2 back from the Solms hospital........ it feels light, thin and tin-can like. The L1 feels heavier, solid and like a precision Metallic Precision tool.

 

Don't get me wrong...... my emotions ..."well up" positively and nostaligically every time I hold the D2 and I think yes, I can use the D2 at the 'right times'...... Whereas the L1 makes me feel ready to take the world on.....anytime

 

My family have asked me if/when I might sell off one of these superb cameras..... and deep down I keep thinking.....No, they are both excellent image capturers with vastly different 'modus operandi'...... that I need to keep both.

[audidudi]

Here's one I shot earlier this week and just processed a few minutes ago (I'm using a new monitor profile and am not sure it's quite right, so if the color balance/contrast of this image seems out of whack to you, let me know!)... As for the "softness" referred to above, I agree ... sort of. I actually started a thread about this recently but have since discovered that L1 files require a lot more sharpening than I've been accustomed to using in the past and respond very well to a multi-pass technique, where you sharpen several times at different settings. I would pass along my suggested settings but I'm still polishing my approach, so I'll hold off for now.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[designdog]

While you are at it, best to have the latest firmware, just updated today at:

 

http://panasonic.jp/support/global/cs/dsc/download/L1K/

 

-ddog

[spylaw4]

designdog - the link is still showing Ver 1.1 so no update as yet - I've been monitoring this regularly today.

 

audiaudi - I look forward to your posting of the sharpening technique (I hope it's Mac friendly and doesn't involve the dreaded Photoshop)

[audidudi]

I look forward to your posting of the sharpening technique (I hope it's Mac friendly and doesn't involve the dreaded Photoshop)

Alas, while I use an Apple Cinema Display via an adapter, I'm running a Windows box and Yes, my sharpening technique does involve Photoshop as well as Irfanview ... sorry.

 

Oh, and version 2.0 of the firmware is now available at the link above...

[audidudi]

Advertisement (gone after registration)

While commuting to work this morning, I was thinking about how L1/D3 and LC1/D2 images compare with each other when it suddenly dawned on me: Why not shoot identical images with each camera and -- gasp! -- compare them, rather than addressing this in a more theoretical/philosophical manner as seems to have been the case thus far?

 

As it happens, I have both cameras with me at work (which is where I seem to do most of my photography these days <sigh>) and although my schedule is busy today, I plan to do some shooting with them tomorrow and over the weekend. Since I always use a tripod for this sort of work, it should be possible to frame images with each camera that are (for all practical purposes) identical, process them using the same software, then pull 100% crops out of various sections of each image and compare them. I'm not sure whether it's best to skip the issue about file size differences or to resample up/down as appropriate, but I can sort that out once I have the RAW captures.

 

Anyone else want to take a crack at this? Collectively, we may be able to get a better handle on how these two cameras perform and the extents to which they're similar and different...

[spylaw4]

V 2.0 is indeed available and has the following :

 

Added a multiple exposure function.

 

Added "panning mode"(MODE3) to the optical image stabilizer.

 

Added a "low-angle" viewing mode that makes it easier to view the LCD display when the camera is held below eye level.

 

Added feature that simplifies setting the aperture when using a lens without an aperture ring.

 

Added two manual white balance settings to GUI. GUI design uses navigational cursor to easily select, and set, white balance (WB) to change the color temperature, or set the ISO sensitivity or flash.

 

Time restriction for making changes was removed.

 

Added "HOLD" option to Auto Review time, enabling continuous viewing.

 

Added feature to allow zoomed playback of photos. It's now possible to move back to the preceding images or forward to the next images while the image is still zoomed.

 

Added [AUTO] for the ISO sensitivity in the aperture-priority AE mode [A] and shutter speed-priority AE mode .

 

Added GUI option to easily check the firmware version.

 

Improved the performance of AE(Auto Exposure) and AWB(Auto White Balance).

 

This last one is much needed.

Presumably the OIS panning mode is to be used when tracking moving subjects?

 

I'll be installing it later today.

[spylaw4]

I'm perfectly happy to do this - and indeed was thinking about doing so over the Easter weekend. Not in as quite a refined and scientific way as you have proposed - but close enough I think.

 

Looking forward to seeing the results. Sorry to hear about your use of my personal bete noirs!

 

Oh - and IMHO your image wile probably ok for colour could possibly do with a bit more contrast?

 

BTW I though Sean Reid was going to take a look at the D3. Maybe he might opine on the D2/D3 divide?

 

While commuting to work this morning, I was thinking about how L1/D3 and LC1/D2 images compare with each other when it suddenly dawned on me: Why not shoot identical images with each camera and -- gasp! -- compare them, rather than addressing this in a more theoretical/philosophical manner as seems to have been the case thus far?

 

As it happens, I have both cameras with me at work (which is where I seem to do most of my photography these days <sigh>) and although my schedule is busy today, I plan to do some shooting with them tomorrow and over the weekend. Since I always use a tripod for this sort of work, it should be possible to frame images with each camera that are (for all practical purposes) identical, process them using the same software, then pull 100% crops out of various sections of each image and compare them. I'm not sure whether it's best to skip the issue about file size differences or to resample up/down as appropriate, but I can sort that out once I have the RAW captures.

 

Anyone else want to take a crack at this? Collectively, we may be able to get a better handle on how these two cameras perform and the extents to which they're similar and different...

[audidudi]

Oh - and IMHO your image wile probably ok for colour could possibly do with a bit more contrast?

Thanks for the info ... it looks very flat on my notebook but, somewhat strangely, it actually looks pretty good on my uncalibrated NEC monitor here at work. In any event, I'll be looking into this further over the next few days...

[Stigbert]

Hi folks,

Just curious as to the D3 firmware update.

The L1 now had several but as far as I know, there is nothing from Leica?

 

Any clues anyone.

 

Cheers

[jdlaing]

I have the Digilux 3 and we get NO updates.

 

Nothing

Nada

Zip.

 

[audidudi]

The L1 now had several but as far as I know, there is nothing from Leica?

I can't speak to Version 2.0, but it was my understanding the previous L1 firmware update brought it into alignment with the D3's firmware, so no corresponding Leica update was necessary. Was this not the case?

[Stigbert]

Interesting, that may be the case but I have not seen that as any "official" standpoint.

 

And also, is there any chans for a black D3 in the near future?

 

StigBert

[spylaw4]

I can't speak to Version 2.0, but it was my understanding the previous L1 firmware update brought it into alignment with the D3's firmware, so no corresponding Leica update was necessary. Was this not the case?

 

That was my understanding also. It will be interesting to see what happens now re the D3.

[audidudi]

Well, this experiment I've proposed is certainly going to produce some interesting results! As I left work tonight, I thought why not? and unpacked my camera equipment in the parking garage. I basically reshot the photo I posted above and while the test was a bust from a scientific point of view (exposure and white balance differences mean the results cannot be directly compared and I didn't have time this evening to identify and address this on the spot, so I'm not going to bother posting the images here), the initial results were eye-opening to me in two respects: Noise and detail. As expected, the LC1 image was both noisier and slightly less detailed, yet strangely, it was also noticeably more three-dimensional and less "plasticy" looking than the otherwise less noisy and more detailed L1 image. Also, the auto white balance of the LC1 had it all over the auto white balance of my (v 1.1 firmware) L1 and even after correcting them while processing the RAW images, the colors were more attractive and I believe accurate as well. Lastly, the LC1 image was less contrasty and handled the wide range between the highlights and shadows with much more aplomb than did the L1 (although this could very well be affected by the exposure and white-balance mismatch noted above, so don't read too much into this yet.) Overall, while it's still too early to go out on a limb, if I had to pick one camera over the other based upon just this one image, then I'd pick the LC1. Hmmm... (and my apologies for the run-on paragraph, as it appears this new version of FireFox doesn't want to play well with the code behind this site.)

[Riley]

Here's one I shot earlier this week and just processed a few minutes ago (I'm using a new monitor profile and am not sure it's quite right, so if the color balance/contrast of this image seems out of whack to you, let me know!)... As for the "softness" referred to above, I agree ... sort of. I actually started a thread about this recently but have since discovered that L1 files require a lot more sharpening than I've been accustomed to using in the past and respond very well to a multi-pass technique, where you sharpen several times at different settings. I would pass along my suggested settings but I'm still polishing my approach, so I'll hold off for now.

 

In my imagination, there's a lot the same about L1 and typical Olympus 4/3 cameras, this seems to be with sharpening too. Mercifully the files are large, so they can handle quite a bit of PP work without falling apart. LC-1 was never like that, although thankfully, files straight from the camera with LC-1 usually require a whole lot less work.

 

So the durable files can take a multi pass sharpening technique, and like you I have no solid answer as to what that might be, but have several methods on the table that I commonly use.

 

the first variant I began with, i duplicated the layer, and applied a simple UM a tad on the high side, then i varied the opacity and watched the results. What I attempted to avoid was the blow out of white pixelation which is a dead give-away for excessive UM. Sliding the opacity achieves this, but then I came across this profile by Benjamin Kanarek, a fashion photogrpaher working in Paris.

 

Benjamin's profile is detailed below

 

Take an unsharpened neutral JPEG, you accentuate the image i.e.unsharp mask at 0.3 pixels at 240 to 330 percent. After doing so you unsharp mask at 40 to 60 pixel at 10 to 15 percent.

 

Once you have done that, you go to layers and click on duplicate layer. Here is where you have two very important choices. This will affect the look of the image. One will accentuate and highlight the blacks and the other will accentuate the colours and the shadows. For the later you will de-saturate the layer. Go to the layers menu and you will see an icon labelled "Normal". Click on that icon and you will open a file with several lighting options. Choose soft light. This in my opinion will have too much contrast. You have an opacity and fill bar. I generally reduce the fill and opacity any where from 50 to 80 percent.

 

Once you have done that and if pleased by the result, flatted the image in the Layers Menu and "Voila" a medium format film look in digital. Second option is DO NOT de-saturate the image. Keep it in colour and do exactly what was done in the previous method. You will have a highly saturated and profound looking image.

 

That profile can do lovely things for 'some' images. What Benjamin is chasing is the wider tonality of film, and a more saturated look. But it inadvertently releases some ideas on sharpening

 

My mix on Benjamin's profile, is to apply the phases of UM, duplicate the layer and add final strong UM which the files now seem more able to take. I leave out the Soft Light layering as it is less useful on some subjects. But reduce the opacity of the top most layer to around 50-80% before merging them flat.

 

I dont say this is the answer, but it is in some way a part of the answer.

[spylaw4]

All very interestng. How can a manufacturer get the in-camera process so right in one model (LC1/D2) and not follow that through to the next (L1/D3)? Well at least Panasonic are trying hard.

 

@Jeffrey: When I do my test I"ll bear in mnd the exposure differences, and take a shot at the same exposure on each camera, as well as letting the autoexp deal with it. My L1 is now on V2 firmware, so it will be interesting to see what auto WB+ autoexp are now like. There are certainly more settings for WB in the GUI - we need an updated pdf manual to confirm what they are I think.

 

@Riley: If I had to follow that process for every image I would have no time for anything else! Apart from it using PP which I dislike, don't have and currently have no intention of getting!

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[Riley]

All very interestng. How can a manufacturer get the in-camera process so right in one model (LC1/D2) and not follow that through to the next (L1/D3)? Well at least Panasonic are trying hard.

 

@Riley: If I had to follow that process for every image I would have no time for anything else! Apart from it using PP which I dislike, don't have and currently have no intention of getting!

 

I think Panasonic inherited some of the issues with Olympus 4/3rds, I suspect to an extent too strong an AA filter. I also think that the combined input of Olympus/Panasonic/Leica will have an impact on future developments. Pretty interesting when we see real competition within 4/3rds as a format.

 

As to Benjamin's profile, you can record the procedure as a macro in PS or PSP, and execute it with one click, then all one needs do is wait for the results ... and tweak the levels a tad. At least, that's what I did.

 

anyway, this is it in operation with the full Benjamin profile that i called 'wet plastic'

added a crop

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[audidudi]

So the durable files can take a multi pass sharpening technique, and like you I have no solid answer as to what that might be, but have several methods on the table that I commonly use.

My basic sharpening method was developed by yours truly years ago -- nearly a decade, in fact -- and I've been using it with some success ever since. I gather that others may have come up with similar, if not identical approaches, so I'm not claiming it's unique to me or that I'm anything special for having developed it, only that it works well for me. (As an aside, I've tried several of the commercially available sharpening programs and in every case, I've always gone back to my manual method as I preferred its results.) Basically, I start with an initial gaussian blur of a small amount, followed by three passes of unsharp masking of decreasing amounts and radii, using the fade tool in the luminosity mode after each pass to fine-tune things. Depending upon the image, I will sometimes use layers and masks to limit the amount of sharpening applied to various portions of the image, but usually I sharpen the whole image identically. I also sharpen less for images to be viewed online instead of printed, as I find images that print well usually look slightly oversharpened onscreen. Anyway, no magic here and also not something that can be automated very well, since I tweak each pass to fine-tune it for each image and it takes a bit of experience to determine the settings that will be optimal ... start with too much sharpening in the first pass and you'll know by the third pass you need to go back and start over; too little and you'll find that you can sharpen four and five passes without any noticeable change in how the image looks. I'll post more later, but as I said, the L1 images look superficially more detailed than the LC1 images, but the LC1 images look more three-dimensional and life-like to my eye...