The Sony A7 thread [Merged]

[JimKasson]

Advertisement (gone after registration)

Jim

I notice in one your well written blogs you recommend against compression of any sort. Still feel that way?

 

In a word, yes. I'm from Missouri. If I don't understand the algorithm, I turn compression off. It's really easy for the marketing folks at some camera company to call their algorithm lossless. I've seen that happen when the algorithm turned out to be lossy, as defined by computer science.

 

Jim

[chris_livsey]

Chris, that's not such a simple question.

Jim

 

Jim, Thanks, I do try Hampered by my programming skills being stuck at the Sinclair ZX81 ( Timex Sinclair 1000)

 

ZX81 - Wikipedia, the free encyclopedia

[Guest polygamer]

Hi, Oldies Radio stations replay old songs, such as Chubby Checker's Limbo Rock with How low can you go?

 

I am not sure about UWA performance on the A7. So, I took my ZEISS Distagon 4/18 and the A7 on a trip to the Niederwald Monument.

 

Its picture and the lower left corner (made a little brighter) I am satisfied with

(maybe I should have the novoflex adapter, not the metabones):

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!

[Guest polygamer]

However, a little vignetting and Sony-typical corner softness is not satisfactory in the landscape picture of the Rhine:

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!

[Guest polygamer]

And now 3 crops: centre, centre right, lower right (brightened), all, as all pictures f8:

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!

[Guest polygamer]

For dealing with corner softness I have several options:

 

- Use the Voigtländer Heliar 5,6/12 on the A7 in APS-C crop mode, image size 10MP

 

- Use the Voigtländer Heliar 5,6/12 on the NEX-6 (APS-C), image size 16MP

 

- Use the ZEISS Distagon 4/18 on the M9.

 

Presently, I would say, the A7 and the ZEISS Biogon 4/18 are a good match for architecture and travel pictures in cities,

but not for landscape...

[CaptZoom]

Advertisement (gone after registration)

Polygamer[/mention]

If you can, try multiple copies if the same adapter to find the best fit for your lenses and camera.

[Guest polygamer]

Polygamer[/mention]

If you can, try multiple copies if the same adapter to find the best fit for your lenses and camera.

Hi, up to now, I was quite happy with the metabones, however it fits not tightly.

I also have novoflex and quenox adapters and they all performed equally well

- till totoday ...

[CaptZoom]

Hi, up to now, I was quite happy with the metabones, however it fits not tightly.

I also have novoflex and quenox adapters and they all performed equally well

- till totoday ...

 

 

Bummer.

[Rick]

In a word, yes. I'm from Missouri. If I don't understand the algorithm, I turn compression off. It's really easy for the marketing folks at some camera company to call their algorithm lossless. I've seen that happen when the algorithm turned out to be lossy, as defined by computer science.

 

Jim

 

And that would be whom? Leica states the M uses lossless compression. Are you indicting Leica when giving advise to Lou knowing he is talking about his M240? And, how did you determine the M240 lossless is a shame... I'm all ears.

[TheGodParticle/Hari]

Hi all,

 

Any experience if the Elmar-M 50/2.8 (last version) and Macro Elmar 90/4 can be collapsed on the A7/A7r?

 

Thank you

[JimKasson]

And that would be whom? Leica states the M uses lossless compression. Are you indicting Leica when giving advise to Lou knowing he is talking about his M240? And, how did you determine the M240 lossless is a shame... I'm all ears.

 

I'm not pointing fingers at any camera company in particular. I didn't read the question as being about the M240.

 

Jim

[Rick]

I'm not pointing fingers at any camera company in particular. I didn't read the question as being about the M240.

 

Jim

 

Thanks for the reply. Do you suspect the the M240's compression is lossy? There has been a lot of discussion about this and I haven't heard any source that believes the compression is anything but lossless.

 

The reason I ask is that there seems to be some confusion that lossless compression is somehow not as good as no compression. Statements that question lossless compression as being somehow suspect are perpetuating a sort of pseudoscience or HiFi-like legend that nothing is better than no compression even if it is lossless.

[JimKasson]

Thanks for the reply. Do you suspect the the M240's compression is lossy? There has been a lot of discussion about this and I haven't heard any source that believes the compression is anything but lossless.

 

I don't have an opinion on this, as I haven't looked into it. I don't even know if the M240 compression algorithm is by Leica, by Adobe, or by some third party. I have not seen a description of the algorithm, in either words or code. I suppose I could do some research, but, for my own personal purposes, using the uncompressed mode doesn't seem burdensome.

 

The reason I ask is that there seems to be some confusion that lossless compression is somehow not as good as no compression. Statements that question lossless compression as being somehow suspect are perpetuating a sort of pseudoscience or HiFi-like legend that nothing is better than no compression even if it is lossless.

 

I am a believer in lossless compression for many purposes. I use it for some cameras' raw files. I use it sometimes when creating TIFFs. I use it all the time (through no action on my part) when saving .psd files. I used to use it when sending and receiving faxes, again through no action on my part. With faxes of documents, lossless compression usually produced some impressive reductions in data, especially when combined with lossy algorithms to remove paper texture. I use it when creating Acrobat documents. I regularly create and consume .zip files, and consume their self-extracting .exe siblings. I have found, however, that the amount of lossless compression available for photographic images is not large, certainly when measured against, say, text files.

 

In many contexts, it's easy for a user to prove to herself that an algorithm is indeed lossless, at least for a particular input file. Take the file, compress it, expand it, and compare it to the original. When a camera does the compression, you can't do that, because you can't access the original version of the compressed file. Taking one picture with compression on and another with it off and comparing the two doesn't do what you'd like, because the read and photon noise are different in each original.

 

I'd feel a lot better about the whole situation if the camera manufacturers would just say, here's our algorithm. As it is, probably to best place to find out is to look at the DCRAW source code, if you (and I'm not) are facile with tightly-written C++.

 

Jim

[JimKasson]

Do you suspect the the M240's compression is lossy?

 

You got me thinking. Always dangerous. I looked at Dave Coffin's DCRAW code, and saw this:

 

<snip>/*

Not a full implementation of Lossless JPEG, just

enough to decode Canon, Kodak and Adobe DNG images.

*/

<snip>

 

int CLASS ljpeg_start (struct jhead *jh, int info_only)

{ <snip>

 

Lossless JPEG as defined in the first JPEG standard is, if memory serves, differential pulse code modulation followed by Huffman coding. That should be truly lossless. There is another JPEG coding that is lossess/near lossless. If you look at Barry Pearson's DNG specification web page,

 

DNG specification

 

it looks like DNG uses the DPCM (truly lossless) version.

 

Barry also has an explanation for the choice of Lossless JPEG.

 

DNG specification

 

which contains a quote from Thomas Knoll that indicates that DNG truly uses the lossless JPEG variant:

 

"The DNG spec (and Canon CR2 format, plus several Kodak raw formats) uses the old original lossless JPEG standard. I suggest you download the DNG SDK and read the source code yourself if you want all the details".

 

So, I now believe that an accurate implementation of the lossless DNG compression is indeed lossless. I wonder why Coffin didn't say anything about Leica images.

 

As an aside, using DNG is not sufficient to make the raw file lossless. Here' a quote from Barry about the M8:

 

"The Leica M8 camera uses DNG as its raw file format. It stores 8-bits (1 byte) per pixel in its raw files. It also stores a LinearizationTable of 256 16-bit entries in the DNG file to convert those value into 16-bit values. It can have useful dynamic range, while still having only 256 discrete levels. (The way the Leica M8 derives its 8-bit values from its original larger values is not constrained by the DNG specification. In fact, it converts its original values to 16-bit values, then takes the square-root, making it an 8-bit value. The LinearizationTable happens to square the 8-bit value, but that has nothing to do with the DNG specification)."

 

Jim

[JimKasson]

I've done a comparison of the Leica 50mm f/1.4 Summilux-M ASPH on the M240 to the Carl Zeiss 55mm f/1.8 Sonnar FE on the a7. Conclusions:

 

 

• Center resolution is at the sensor level at f/5.6 and f/8 with both setups, but the Sonnar has greater micro-contrast at the wider apertures
  
• Corners are smeared with the Summilux at f/2.8 and f/4, but not with the Sonnar
  
• Corner performance is similar at f/8 and narrower
  
• The a7 anti-aliasing filter doesn't appear to hurt it in this comparison
  

 

Details here:

 

Testing the Sony a7, part 9 | The Last Word

 

The a7/Zony is a great combination for those seeking Leica-level IQ at a fraction of the cost, if they're willing to give up the RF experience.

 

Possibilities for further comparisons for semi-cheapskates: 90mm 'cron/m240 vs Zeiss 100mm f/2 Makro/a7, and 135mm APO-Telyt/M240 vs 135mm f/2 Zeiss APO Sonnar/a7.

 

Jim

[JimKasson]

The comparison of the 90mm APO Summicron-M/m240 to the Zeiss 100mm f/2 Makro Sonnar ZF/a7 is up. Short conclusion is that both of these lenses are capable of delivering performance so far beyond what a 24 megapixel sensor can resolve that preference is a matter of taste.

 

Jim

[viramati]

I've done a comparison of the Leica 50mm f/1.4 Summilux-M ASPH on the M240 to the Carl Zeiss 55mm f/1.8 Sonnar FE on the a7. Conclusions:

 

 

• Center resolution is at the sensor level at f/5.6 and f/8 with both setups, but the Sonnar has greater micro-contrast at the wider apertures
  
• Corners are smeared with the Summilux at f/2.8 and f/4, but not with the Sonnar
  
• Corner performance is similar at f/8 and narrower
  
• The a7 anti-aliasing filter doesn't appear to hurt it in this comparison
  

 

Details here:

 

Testing the Sony a7, part 9 | The Last Word

 

The a7/Zony is a great combination for those seeking Leica-level IQ at a fraction of the cost, if they're willing to give up the RF experience.

 

Possibilities for further comparisons for semi-cheapskates: 90mm 'cron/m240 vs Zeiss 100mm f/2 Makro/a7, and 135mm APO-Telyt/M240 vs 135mm f/2 Zeiss APO Sonnar/a7.

 

Jim

 

I agree the sony/zeiss 55 is just so good on the A7 that I really don't use the leica lux 50 asph anymore on my M. Apart from the IQ I just can hit the focus so much more wight the 55 on the A7

[chris_livsey]

The comparison of the 90mm APO Summicron-M/m240 to the Zeiss 100mm f/2 Makro Sonnar ZF/a7 is up. Short conclusion is that both of these lenses are capable of delivering performance so far beyond what a 24 megapixel sensor can resolve that preference is a matter of taste.

 

Jim

 

Let's see how the Sony 54MP 35mm chip does then

 

Not rumour, it is on the Sony site with price comment

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[pierovitch]

Sony has been going through some difficult times recently, but surely they are not that desperate that they saw going after the Leica M market as vastly (!) broadening their customer base. I would think they are after bigger fish.

 

Hasselblad