So, who's buying the 16-35mm?

[AZach]

Advertisement (gone after registration)

Congratulations, my birthday was same but somehow was missed by "the TNT guy"  . The lens cap in the bottom picture looks a bit small...  Looking forward to your review!

[Advertisement]

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

[MarkinVan]

Congratulations and Happy Belated Birthday.

 

What a gift.

 

Look forward to hearing all about it and seeing your results.

 

Waiting patiently to hear on the arrival of my own version.

 

Cheers!

 

 

Yesterday was my birthday, and thanks to the great guys at NewOldCamera in Milan (http://www.newoldcamera.com/novita.aspx) look what the TNT guy brought me today: the new Leica Super-Vario-Elmar-SL 16-35mm ASPH is finally here!

 

Now I just have to wait for the new Firmware, and then I'll start working on my review... looking forward to trying it out!

 

 

 

Best regards,

 

Vieri

[Fedro]

my birthday's tomorrow, will keep an eye out for the TNT man

 

Happy birthday, Vieri!

[Vieri]

Congratulations, my birthday was same but somehow was missed by "the TNT guy"  . The lens cap in the bottom picture looks a bit small...  Looking forward to your review!

 

Thank you very much! The lens cap is just fine in the real world... this is courtesy of iPhone!

 

Congratulations and Happy Belated Birthday.

 

What a gift.

 

Look forward to hearing all about it and seeing your results.

 

Waiting patiently to hear on the arrival of my own version.

 

Cheers!

 

Thank you very much! A self-made present sometimes is good

 

my birthday's tomorrow, will keep an eye out for the TNT man

 

Happy birthday, Vieri!

 

 

Thank you Fedro, and happy Birthday to you too for tomorrow - one never knows what the TNT man can bring!

 

Best regards,

 

Vieri

[Vieri]

I got that information from the most trusted of sources... The internet....

 

**********************************************

 

All in all, this means that image quality is a known quantity: it matches that of the H5D-50c and H6D-50c, which is to say – short of the 54x40mm 100MP sensor, is about the best you can get today. Like all Hasselblad cameras, sensors used in the X1D are individuallyprofiled to a fixed reference color standard at all ISOs – I shoot the X1D comfortably to ISO 12,800. In fact, each camera has nearly half a gigabyte of calibration data in it. I believe Hasselblad is the only company to do this – it’s one of the reasons output is so spectrally neutral and tonally natural across the sensitivity range, and no dark frame subtraction is required even on exposures up to one hour. Color accuracy is one of the main reasons I switched; those of you who have Workflow III will see that the Hasselblad profiles have almost no adjustments, and by far the least HSL adjustments of any camera included. Individual sensor calibration also means consistency is excellent – my H5D-50c, H6D-100c, CFV-50c and X1D all produce identical tonal response (with the exception of course of the H6D-100c, which has a bit more dynamic range extension at either end).

 

****************************************

 

From here and he should know.....

 

https://blog.mingthein.com/2018/01/31/long-term-review-the-hasselblad-x1d/

 

In the comments he puts the start up time to being able to load the calibration data.

 

Gordon

 

 

Gordon,

 

a question for you. While I can accept that the X1D is loading lens calibration data to correct problems such as distortion, vignette and so on, and that might take forever, and therefore that this is the reason for the very slow startup time of the X1D (why they don't load these data once and for all in the camera's memory is a different question - Ming said that it's because "ASICs were too expensive at the time of development", but the SL, the Sony, et.a. can do that...), I have a very hard time believing that that data is responsible for not needing LENR's dark frame.

 

Reason is very simple: noise patterns are random. The reason for shooting a dark frame right after the main frame is to try and recreate the closest possible conditions to the photograph you want to subtract the frame from. Meaning: if, due to whatever reason, you get noise on pixel x - y, shooting a second, dark frame right after the main image has a high chance to get noise in the same pixel, and therefore being able to subtract it from the main image. Loading calibration data makes no sense to me for that purpose, since there is no way the files you load could ever know which pixels will make noise on a given image - again, noise is random.

 

So, besides believing Ming, why do you think that this is the case with the X1D? To me, both logically and based on all I have ever read about LENR, this explanation makes no sense. I think that certain sensors have better noise characteristics than others, and that these are hardware-dependent. The X1D sensor, which I used in the Pentax 645z, is one of such sensors, hence it doesn't need LENR. Other sensors, such as the SL's, have less good hardware noise characteristics and therefore need LENR. This is my explanation for the absence of LENR in the X1D: it simply makes more logical sense.

 

Looking forward to hearing your thoughts on this. Best regards,

 

Vieri

Edited April 24, 2018 by Vieri

[tpenate2]

Vieri, happy belated birthday...  may you enjoy the lens in good health.  Sorry if it's a dumb question - why are you waiting for a firmware update before you use the lens?  We need to wait for that before it works correctly with the SL?  I placed my order when it was announced a couple of weeks ago - fingers crossed!  Thx / Tony

[Vieri]

Advertisement (gone after registration)

Vieri, happy belated birthday...  may you enjoy the lens in good health.  Sorry if it's a dumb question - why are you waiting for a firmware update before you use the lens?  We need to wait for that before it works correctly with the SL?  I placed my order when it was announced a couple of weeks ago - fingers crossed!  Thx / Tony

 

 

Thank you for your wishes! Well, while the lens might work even with an older FW (which is not a given, sometimes doing that can cause either camera or lens to freeze), the camera will not recognise it, EXIF will not be right, and if there are correction to be applied, the camera wouldn't have any chance to know what they are and therefore images wouldn't look as they are supposed to. Of course, I have no information on whether this is in fact the case with this particular lens, but I just prefer not to take the risk.

 

The new FW should be out soon anyway, waiting a couple of days more is not a big deal

Best regards,

 

Vieri

[FlashGordonPhotography]

Gordon,

 

a question for you. While I can accept that the X1D is loading lens calibration data to correct problems such as distortion, vignette and so on, and that might take forever, and therefore that this is the reason for the very slow startup time of the X1D (why they don't load these data once and for all in the camera's memory is a different question - Ming said that it's because "ASICs were too expensive at the time of development", but the SL, the Sony, et.a. can do that...), I have a very hard time believing that that data is responsible for not needing LENR's dark frame.

 

Reason is very simple: noise patterns are random. The reason for shooting a dark frame right after the main frame is to try and recreate the closest possible conditions to the photograph you want to subtract the frame from. Meaning: if, due to whatever reason, you get noise on pixel x - y, shooting a second, dark frame right after the main image has a high chance to get noise in the same pixel, and therefore being able to subtract it from the main image. Loading calibration data makes no sense to me for that purpose, since there is no way the files you load could ever know which pixels will make noise on a given image - again, noise is random.

 

So, besides believing Ming, why do you think that this is the case with the X1D? To me, both logically and based on all I have ever read about LENR, this explanation makes no sense. I think that certain sensors have better noise characteristics than others, and that these are hardware-dependent. The X1D sensor, which I used in the Pentax 645z, is one of such sensors, hence it doesn't need LENR. Other sensors, such as the SL's, have less good hardware noise characteristics and therefore need LENR. This is my explanation for the absence of LENR in the X1D: it simply makes more logical sense.

 

Looking forward to hearing your thoughts on this. Best regards,

 

Vieri

 

Probably a question for Ming Thein, not me.

 

Gordon

[scott kirkpatrick]

Vieri and others, noise that can be corrected by dark frame subtraction is not random.  There are several sources of error in a digital picture, only one of which ("shot noise" -- the variation in the number of actual photons that arrive in a pixel's area) is random.  The rest, like sensitivity variations, leakage during the exposure, etc, are quite predictable from the history of a pixel, but they do have some variation based on the camera's temperature, the exposure time...  LENR is the simplest, and probably the most accurate way to capture these variations for a specific image.  But it is tedious.  LENR is not an attempt to correct shot noise.

 

detail: what is shot noise?  If an exposure should lead to, on average, 10,000 photons hitting a particular pixel, and we repeat the exposure many times, the pixel will record a number of photons that varies by about +- 100 photons.  The variation is proportional to the square root of the average number.  Not a big problem for 10,000 photons, but if a bit of shadow detail requires 64 photons, then the variation is of order of +- 8 photons, and you can see that.  Smoothing over nearby pixels to get a more accurate shadow level with some loss of resolution, is the usual remedy, or we learn to like a little graininess in our shadows.

[Vieri]

Vieri and others, noise that can be corrected by dark frame subtraction is not random.  There are several sources of error in a digital picture, only one of which ("shot noise" -- the variation in the number of actual photons that arrive in a pixel's area) is random.  The rest, like sensitivity variations, leakage during the exposure, etc, are quite predictable from the history of a pixel, but they do have some variation based on the camera's temperature, the exposure time...  LENR is the simplest, and probably the most accurate way to capture these variations for a specific image.  But it is tedious.  LENR is not an attempt to correct shot noise.

 

detail: what is shot noise?  If an exposure should lead to, on average, 10,000 photons hitting a particular pixel, and we repeat the exposure many times, the pixel will record a number of photons that varies by about +- 100 photons.  The variation is proportional to the square root of the average number.  Not a big problem for 10,000 photons, but if a bit of shadow detail requires 64 photons, then the variation is of order of +- 8 photons, and you can see that.  Smoothing over nearby pixels to get a more accurate shadow level with some loss of resolution, is the usual remedy, or we learn to like a little graininess in our shadows.

 

 

Scott,

 

thank you for that. So, basically we can agree that:

 

1. LENR, while tedious, is the most effective way to fix noise;

2. If the majority of noise is non-shot noise, then it is hardware-dependent, since also camera temperature is partially hardware dependent; exposure time is the only true variable here, but one that can be easily taken into account when designing in-camera RAW image creation software (most cameras that do not have LENR apply a little RAW noise reduction in-camera, and I wouldn't be surprised if cameras with LENR did as well);

3. So, if noise is hardware-dependent, it makes sense that some sensors have better noise characteristics than others, and therefore that some cameras do not need LENR while others do (which would explain how all MF cameras using the same 50 Mp sensor in the X1D, 635Z, Fuji, etc do not have LENR);

4. If the above is true, either Gordon's reported "loading calibration data" has nothing to do with noise, or if it had it could just be kept loaded like every other camera, making startup times less atrocious;

 

Finally,

5. Neither LENR nor these alleged "noise data tables" can correct for shot noise, since is random, and the only thing that can fix that is a little noise reduction in software (either in-camera, or in post).

 

Am I correct so far?

 

Best regards.

 

Vieri

Edited April 25, 2018 by Vieri

[LocalHero1953]

At least Hasselblad (via Ming, assuming he is representing them correctly) gives a believable explanation for the absence of LENR exposures. Leica just ignores its customers here, which might be acceptable while no other brand has a system that works. As has always been the case for Leica and photographic challenges, there's a workaround, but one that is too tedious for most people to want to use. I felt the same way when I first saw Leica's solution to TTL viewing - the Visoflex - an engineering marvel, but.......... The difference was that I admired Leica for offering a solution to a problem that I hadn't expected or wanted them to solve. In this case, long exposure is a fairly mainstream use scenario that other brands offer without hassle.

[tpenate2]

Thank you for your wishes! Well, while the lens might work even with an older FW (which is not a given, sometimes doing that can cause either camera or lens to freeze), the camera will not recognise it, EXIF will not be right, and if there are correction to be applied, the camera wouldn't have any chance to know what they are and therefore images wouldn't look as they are supposed to. Of course, I have no information on whether this is in fact the case with this particular lens, but I just prefer not to take the risk.

 

The new FW should be out soon anyway, waiting a couple of days more is not a big deal

Best regards,

 

Vieri

 

 

Makes sense...  thank you!

[RoySmith]

Thank you for your wishes! Well, while the lens might work even with an older FW (which is not a given, sometimes doing that can cause either camera or lens to freeze), the camera will not recognise it, EXIF will not be right, and if there are correction to be applied, the camera wouldn't have any chance to know what they are and therefore images wouldn't look as they are supposed to. Of course, I have no information on whether this is in fact the case with this particular lens, but I just prefer not to take the risk.

 

The new FW should be out soon anyway, waiting a couple of days more is not a big deal

Best regards,

 

Vieri

 

 

Vieri,

 

Did the 16-35 SL  come with a warning not to use the lens unless the camera had FW 3.2 or newer ?

I recall the 50lux SL having a sticker on the lens cap with a warning like that.

 

Roy

[Vieri]

Vieri,

 

Did the 16-35 SL  come with a warning not to use the lens unless the camera had FW 3.2 or newer ?

I recall the 50lux SL having a sticker on the lens cap with a warning like that.

 

Roy

 

 

Hi Roy,

 

yes it does - a sticker telling you to use "the most recent Firmware" Best regards,

 

Vieri

[Vieri]

Makes sense...  thank you!

 

You are welcome! Best regards,

 

Vieri

[sillbeers15]

You are welcome! Best regards,

 

Vieri

Hi Vieri,

So you’re the first owner of the 16-35mm lens that I know. Without the firmware 3.2 becoming available, you are not recommended by Leica not to use the lens.

Humm,...I don feel so bad waiting for my lens then.

[lx1713]

I doubt if the firmware is going to stop Vieri from enjoying his birthday present 

[PBart]

Just got my 16-35mm today from the Leica Store New York SOHO. Will be running tests over the coming days and weeks before using it professionally but it's a beautiful lens as expected. Much like the 24-90mm and super smooth with zoom and focus. I'm really loving this SL system. If they made the SL camera higher resolution I'd be tempted to sell my S system. However, the S System has such incredible depth to the DNG files. 

[Advertisement]

Hallo,

Look here: Leica L-Mount

[Bill W]

Just got my 16-35mm today from the Leica Store New York SOHO. Will be running tests over the coming days and weeks before using it professionally but it's a beautiful lens as expected. Much like the 24-90mm and super smooth with zoom and focus. I'm really loving this SL system. If they made the SL camera higher resolution I'd be tempted to sell my S system. However, the S System has such incredible depth to the DNG files. 

 

Wow congrats. I assume your are going to wait for the software update? FWIW, I am selling my S system and going all in on the SL. Of course I am not a professional, just an old guy with a very bad back. The SL is at least a little lighter than the S and a couple of zooms that are superb...!

[PBart]

Wow congrats. I assume your are going to wait for the software update? FWIW, I am selling my S system and going all in on the SL. Of course I am not a professional, just an old guy with a very bad back. The SL is at least a little lighter than the S and a couple of zooms that are superb...!

 

I checked for a firmware update and found one today after the lens arrived. Downloaded, installed and all is good. Yes, the S System is heavy. The SL is so much better for versatility. Overall I I like the SL the best but still need the file quality from the S for pro work. Between the 16-35mm and the 24-90mm I have all I really need in the SL System other than a tilt shift lens at 24mm. I think it's very possible in the future to see a 24mm come about.