Samsung S20 ultra comparison

Started 9 months ago | Discussions
Lee Jay Forum Pro • Posts: 56,297
Re: Samsung messed up the implementation

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor.  It doesn't have pixel binning.  So, yes.

if you look at the crops you'll see the processing is the same.

Nope.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM Canon Extender EF 2x III +10 more
Jostian
Jostian Veteran Member • Posts: 4,322
Re: Samsung messed up the implementation

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor. It doesn't have pixel binning. So, yes.

but the Samsung processing algorithms will be the same.

if you look at the crops you'll see the processing is the same.

Nope.

-- hide signature --
 Jostian's gear list:Jostian's gear list
Sony a9 Sony a6600 Sony FE 100-400mm F4.5-5.6 Tamron 28-75mm F2.8 III Tamron 20mm F2.8 Di III OSD +3 more
RDM5546
RDM5546 Senior Member • Posts: 1,722
Re: Samsung messed up the implementation

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor. It doesn't have pixel binning. So, yes.

but the Samsung processing algorithms will be the same.

if you look at the crops you'll see the processing is the same.

Nope.

You must be speaking loosely.  Two processings from different source code are at best similiar in many cases but definitely are not exactly the same.  Comparing two sets of different source code will not be exactly the same unless are identical which more than likely they not the identical in this case.    Similar to your eye Jostian the cases may be.  If you speak precisely how can they be the same?

 RDM5546's gear list:RDM5546's gear list
Canon G5 X II Canon EOS 70D Canon EOS 7D Mark II Canon EOS 5D Mark IV Canon EOS R +37 more
RDM5546
RDM5546 Senior Member • Posts: 1,722
Re: Samsung S20 ultra comparison

Lee Jay wrote:

Jostian wrote:

mmm, dont understand why the crops of the Huawei and iPhone are at 200% (or more) to match size of crops from Samsung which are at 100%,

108MP for the Samsung, 40MP for the Mate 30 Pro and 12MP for the iPhone 11 Pro. That's why.

anything above 100% will be a mess, so thats already a big problem.

Hence why having more pixels matters for cropping.

All comparison reviews will compare 100% crops relative to megapixels

Then they are all stupid. You should compare at constant final size, and that's what they did here.

, cropping a 40MP shots to match the crop size of a 108MP sensor is ludicrous,

It shows the advantage of having more pixels and a larger sensor.

hence the horrid crop comparisons...though I guess ubergizmo are not renown for their camera reviews.

They did it right. Showing 100% crops makes no sense and never did when doing cross-camera comparisons at constant field-of-view. Doing that led many to falsely believe that smaller pixel counts were "better" for a long time.

Lee Jay,

Have you tried the GCAM 7.3 download for the S20 Ultra. The current versions are available from three different developers seeking to build their own preferred camera apps.

https://www.gizmochina.com/2020/03/18/download-samsung-galaxy-s20-gcam-mod/

I have tried the first of these three downloads but it is a little buggy but still provides interesting alternative images from the 108MP sensor binned to 12MP.  I am interest in the quality of JPEG color and smoothing this Pixel camera developed source produces in comparison with Samsung App.  My investigation is still a work in progress.  I have no conclusions other than it can be used to make taste images in JPEG from alternative camera processing code.

 RDM5546's gear list:RDM5546's gear list
Canon G5 X II Canon EOS 70D Canon EOS 7D Mark II Canon EOS 5D Mark IV Canon EOS R +37 more
Jostian
Jostian Veteran Member • Posts: 4,322
Re: Samsung messed up the implementation

RDM5546 wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor. It doesn't have pixel binning. So, yes.

but the Samsung processing algorithms will be the same.

if you look at the crops you'll see the processing is the same.

Nope.

You must be speaking loosely. Two processings from different source code are at best similiar in many cases but definitely are not exactly the same. Comparing two sets of different source code will not be exactly the same unless are identical which more than likely they not the identical in this case. Similar to your eye Jostian the cases may be. If you speak precisely how can they be the same?

as per any phone that has more than one camera (wide, UW and tele) the jpeg processing is the same in terms of jpeg IQ produced, you dont get different processing across different lenses in terms of jpeg IQ produced.

Obviously the code will be different as the different sensors have different specs etc.

-- hide signature --
 Jostian's gear list:Jostian's gear list
Sony a9 Sony a6600 Sony FE 100-400mm F4.5-5.6 Tamron 28-75mm F2.8 III Tamron 20mm F2.8 Di III OSD +3 more
RDM5546
RDM5546 Senior Member • Posts: 1,722
Re: Samsung messed up the implementation

Jostian wrote:

RDM5546 wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor. It doesn't have pixel binning. So, yes.

but the Samsung processing algorithms will be the same.

if you look at the crops you'll see the processing is the same.

Nope.

You must be speaking loosely. Two processings from different source code are at best similiar in many cases but definitely are not exactly the same. Comparing two sets of different source code will not be exactly the same unless are identical which more than likely they not the identical in this case. Similar to your eye Jostian the cases may be. If you speak precisely how can they be the same?

as per any phone that has more than one camera (wide, UW and tele) the jpeg processing is the same in terms of jpeg IQ produced, you dont get different processing across different lenses in terms of jpeg IQ produced.

Obviously the code will be different as the different sensors have different specs etc.

I agree with part of what you say. The JPEG  Image Compression Codec algorithm  is a standard that the same for all cameras in a smartphone however the JPEG IQ being discussed the quality of the JPEG file generated by the phoone which is the result of an image processing pipeline including the CODEC as one of the final steps before saving the file.   This output differes from a raw file output in the image processing pipeline incorported before the JPEG CODEC step on the sensed image.  Some of this pipeline of "corrections" may be internal to the sensor chip  included in the raw file data but most of the severe corrections are done in software and only included in the JPEG file.  These severe corrections are dependent in part on the sensor chip and lens in use.   The JPEG file aberations IQ will vary to some degree for the different cameras in the S20 despite the same software codeset being used for all camera the parameters and performance for each camera will change the IQ for each camera.  The JPEG IQ will vary with the cameras in the S20 but there will be common IQ issues across all cameras as well since they do share the same  software code set just not the same sensor and lens parameters..

 RDM5546's gear list:RDM5546's gear list
Canon G5 X II Canon EOS 70D Canon EOS 7D Mark II Canon EOS 5D Mark IV Canon EOS R +37 more
RDM5546
RDM5546 Senior Member • Posts: 1,722
Re: Samsung messed up the implementation

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Panino Manino wrote:

Can we agree that this thing is a disaster (the camera)?

No.

It probably has the best main camera and certainly the best telephoto camera of any phone.

rubbish, the p30 pro beats it at 5X optical zoom

No way, no how.

The P30 pro has a 14.5mm lens with an 8MP sensor behind it with 1 micron pixels working out to 253 pixels per degree.

The S20 Ultra has a 19mm lens with a 48MP sensor behind it with 0.8 micron pixels working out to 415 pixels per degree.

your obsession that more pixels equals better IQ is just not fact!

Your pictures aren't the same size and the top one is out-of-focus.

I tested the Ultra's tele at 1:1 (10x), and it was pixel-sharp.

My results in testing the zoom are that the 10X the zoom is pretttey darn sharp.  I have to agree with Lee Jay that is no problem with the 10X image sharpness with other taken at the same distance.

It is easy to make an error testing the S20 Ultra if you do not back away from the subject at higher zoom factors.   While at 1X you can shoot as close as 1 foot or less but more is always better for reduced geometricall distoration.

At 4X the minimum focus distance is around 2feet and at 10X it is 3 feet.  This is the smallest zoom factor that uses the 19mm telephoto as indicated in the EXIF data.

 RDM5546's gear list:RDM5546's gear list
Canon G5 X II Canon EOS 70D Canon EOS 7D Mark II Canon EOS 5D Mark IV Canon EOS R +37 more
Jostian
Jostian Veteran Member • Posts: 4,322
Re: Samsung messed up the implementation

RDM5546 wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Panino Manino wrote:

Can we agree that this thing is a disaster (the camera)?

No.

It probably has the best main camera and certainly the best telephoto camera of any phone.

rubbish, the p30 pro beats it at 5X optical zoom

No way, no how.

The P30 pro has a 14.5mm lens with an 8MP sensor behind it with 1 micron pixels working out to 253 pixels per degree.

The S20 Ultra has a 19mm lens with a 48MP sensor behind it with 0.8 micron pixels working out to 415 pixels per degree.

your obsession that more pixels equals better IQ is just not fact!

Your pictures aren't the same size and the top one is out-of-focus.

I tested the Ultra's tele at 1:1 (10x), and it was pixel-sharp.

My results in testing the zoom are that the 10X the zoom is pretttey darn sharp. I have to agree with Lee Jay that is no problem with the 10X image sharpness with other taken at the same distance.

It is easy to make an error testing the S20 Ultra if you do not back away from the subject at higher zoom factors. While at 1X you can shoot as close as 1 foot or less but more is always better for reduced geometricall distoration.

At 4X the minimum focus distance is around 2feet and at 10X it is 3 feet. This is the smallest zoom factor that uses the 19mm telephoto as indicated in the EXIF data.

Post some 10x SOOC jpegs I'd love to see them...

-- hide signature --
 Jostian's gear list:Jostian's gear list
Sony a9 Sony a6600 Sony FE 100-400mm F4.5-5.6 Tamron 28-75mm F2.8 III Tamron 20mm F2.8 Di III OSD +3 more
Lee Jay Forum Pro • Posts: 56,297
Re: Samsung messed up the implementation

Jostian wrote:

RDM5546 wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor. It doesn't have pixel binning. So, yes.

but the Samsung processing algorithms will be the same.

if you look at the crops you'll see the processing is the same.

Nope.

You must be speaking loosely. Two processings from different source code are at best similiar in many cases but definitely are not exactly the same. Comparing two sets of different source code will not be exactly the same unless are identical which more than likely they not the identical in this case. Similar to your eye Jostian the cases may be. If you speak precisely how can they be the same?

as per any phone that has more than one camera (wide, UW and tele) the jpeg processing is the same in terms of jpeg IQ produced

That statement is entirely false.

, you dont get different processing across different lenses in terms of jpeg IQ produced.

Yes, you do.

Obviously the code will be different as the different sensors have different specs etc.

And hence, the results are different.

Differences include:

  • IR filters
  • Dye transmission curves
  • Well capacities
  • Read noise
  • Quantum efficiency
  • Microlens array
  • Lens aberrations
  • Black point measurement

All these things mean that processing is tuned to each sensor.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM Canon Extender EF 2x III +10 more
Lee Jay Forum Pro • Posts: 56,297
Re: Samsung S20 ultra comparison

RDM5546 wrote:

Lee Jay wrote:

Jostian wrote:

mmm, dont understand why the crops of the Huawei and iPhone are at 200% (or more) to match size of crops from Samsung which are at 100%,

108MP for the Samsung, 40MP for the Mate 30 Pro and 12MP for the iPhone 11 Pro. That's why.

anything above 100% will be a mess, so thats already a big problem.

Hence why having more pixels matters for cropping.

All comparison reviews will compare 100% crops relative to megapixels

Then they are all stupid. You should compare at constant final size, and that's what they did here.

, cropping a 40MP shots to match the crop size of a 108MP sensor is ludicrous,

It shows the advantage of having more pixels and a larger sensor.

hence the horrid crop comparisons...though I guess ubergizmo are not renown for their camera reviews.

They did it right. Showing 100% crops makes no sense and never did when doing cross-camera comparisons at constant field-of-view. Doing that led many to falsely believe that smaller pixel counts were "better" for a long time.

Lee Jay,

Have you tried the GCAM 7.3 download for the S20 Ultra.

I ended up buying an S20, for several reasons.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM Canon Extender EF 2x III +10 more
Jostian
Jostian Veteran Member • Posts: 4,322
Re: Samsung messed up the implementation

Lee Jay wrote:

Jostian wrote:

RDM5546 wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor. It doesn't have pixel binning. So, yes.

but the Samsung processing algorithms will be the same.

if you look at the crops you'll see the processing is the same.

Nope.

You must be speaking loosely. Two processings from different source code are at best similiar in many cases but definitely are not exactly the same. Comparing two sets of different source code will not be exactly the same unless are identical which more than likely they not the identical in this case. Similar to your eye Jostian the cases may be. If you speak precisely how can they be the same?

as per any phone that has more than one camera (wide, UW and tele) the jpeg processing is the same in terms of jpeg IQ produced

That statement is entirely false.

no, the processing algorithms are the same, you don't get different looking outputs from different lens, the IQ SOOC has the same look as per the numerous reviews we've see for the S20 Ultra and other phones, the jpegs SOOC have the same look for the different lenses.

, you dont get different processing across different lenses in terms of jpeg IQ produced.

Yes, you do.

not in terms of SOOC jpegs

Obviously the code will be different as the different sensors have different specs etc.

And hence, the results are different.

Differences include:

  • IR filters
  • Dye transmission curves
  • Well capacities
  • Read noise
  • Quantum efficiency
  • Microlens array
  • Lens aberrations
  • Black point measurement

All these things mean that processing is tuned to each sensor.

OMG we know this, but the SOOC jpegs have the same look across lenses, look at the crops from the different lenses that have been posted they have the same look to them.

top crop is S20U wide lens, bottom crop is the ultrawide, see the same smeary look to the branch detail and leaves on both...same jpeg processing algorithms!

-- hide signature --
 Jostian's gear list:Jostian's gear list
Sony a9 Sony a6600 Sony FE 100-400mm F4.5-5.6 Tamron 28-75mm F2.8 III Tamron 20mm F2.8 Di III OSD +3 more
Lee Jay Forum Pro • Posts: 56,297
Re: Samsung messed up the implementation

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

RDM5546 wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor. It doesn't have pixel binning. So, yes.

but the Samsung processing algorithms will be the same.

if you look at the crops you'll see the processing is the same.

Nope.

You must be speaking loosely. Two processings from different source code are at best similiar in many cases but definitely are not exactly the same. Comparing two sets of different source code will not be exactly the same unless are identical which more than likely they not the identical in this case. Similar to your eye Jostian the cases may be. If you speak precisely how can they be the same?

as per any phone that has more than one camera (wide, UW and tele) the jpeg processing is the same in terms of jpeg IQ produced

That statement is entirely false.

no, the processing algorithms are the same, you don't get different looking outputs from different lens,

It's not "different lens", it's "different cameras" - sensors and lens combinations.

the IQ SOOC has the same look as per the numerous reviews we've see for the S20 Ultra and other phones,

Then why don't they in any of the images I took?  It's actually not even close.

, you dont get different processing across different lenses in terms of jpeg IQ produced.

Yes, you do.

not in terms of SOOC jpegs

The SOOC jpegs from both of the S20 Ultras I tested were entirely different between the three cameras.

Obviously the code will be different as the different sensors have different specs etc.

And hence, the results are different.

Differences include:

  • IR filters
  • Dye transmission curves
  • Well capacities
  • Read noise
  • Quantum efficiency
  • Microlens array
  • Lens aberrations
  • Black point measurement

All these things mean that processing is tuned to each sensor.

OMG we know this, but the SOOC jpegs have the same look across lenses,

No, they do not.  The main camera is much more heavily noise reduced and smeared compared to the telephoto camera.  It's not even close.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM Canon Extender EF 2x III +10 more
RDM5546
RDM5546 Senior Member • Posts: 1,722
Re: Samsung messed up the implementation

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

RDM5546 wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor. It doesn't have pixel binning. So, yes.

but the Samsung processing algorithms will be the same.

if you look at the crops you'll see the processing is the same.

Nope.

You must be speaking loosely. Two processings from different source code are at best similiar in many cases but definitely are not exactly the same. Comparing two sets of different source code will not be exactly the same unless are identical which more than likely they not the identical in this case. Similar to your eye Jostian the cases may be. If you speak precisely how can they be the same?

as per any phone that has more than one camera (wide, UW and tele) the jpeg processing is the same in terms of jpeg IQ produced

That statement is entirely false.

no, the processing algorithms are the same, you don't get different looking outputs from different lens, the IQ SOOC has the same look as per the numerous reviews we've see for the S20 Ultra and other phones, the jpegs SOOC have the same look for the different lenses.

, you dont get different processing across different lenses in terms of jpeg IQ produced.

Yes, you do.

not in terms of SOOC jpegs

Obviously the code will be different as the different sensors have different specs etc.

And hence, the results are different.

Differences include:

  • IR filters
  • Dye transmission curves
  • Well capacities
  • Read noise
  • Quantum efficiency
  • Microlens array
  • Lens aberrations
  • Black point measurement

All these things mean that processing is tuned to each sensor.

OMG we know this, but the SOOC jpegs have the same look across lenses, look at the crops from the different lenses that have been posted they have the same look to them.

top crop is S20U wide lens, bottom crop is the ultrawide, see the same smeary look to the branch detail and leaves on both...same jpeg processing algorithms!

I agree with you that both images have similiar looking smear but to be precise I for one can nto realy say it is the same.   I do not know what Zoom factors you used for each image.  Is it 0.5X  and 1X?    They appear to be cropped as you state. Are the both cropped the same amount.  From the angle of view and not knowing the zoom factor judging the smear factor that is present in both trees to some degree is hard to quantitatively judge.   I do not believe from my picturee taking the smearing that I can see at times appears in all images at all light levels.   I do disagree with your belief that smearing exists and that I do wish Samsung would work on this a  lot and I also wish that Samsung would provide  means for taking 108MP raw image for exporting  to computer processing in the procamera mode.

I would like Verizon to hurrary up and distribute to US users the new version of the camera that Samsung provided them with to so that I see what is still there.  AT&T has distributed it to their customers according to a Dpreview report yesterday so hopefull Verison version is not far away.

 RDM5546's gear list:RDM5546's gear list
Canon G5 X II Canon EOS 70D Canon EOS 7D Mark II Canon EOS 5D Mark IV Canon EOS R +37 more
Jostian
Jostian Veteran Member • Posts: 4,322
Re: Samsung messed up the implementation

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

RDM5546 wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor. It doesn't have pixel binning. So, yes.

but the Samsung processing algorithms will be the same.

if you look at the crops you'll see the processing is the same.

Nope.

You must be speaking loosely. Two processings from different source code are at best similiar in many cases but definitely are not exactly the same. Comparing two sets of different source code will not be exactly the same unless are identical which more than likely they not the identical in this case. Similar to your eye Jostian the cases may be. If you speak precisely how can they be the same?

as per any phone that has more than one camera (wide, UW and tele) the jpeg processing is the same in terms of jpeg IQ produced

That statement is entirely false.

no, the processing algorithms are the same, you don't get different looking outputs from different lens,

It's not "different lens", it's "different cameras" - sensors and lens combinations.

doesnt change the SOOC IQ processing, same algorithms for jpeg output!

the IQ SOOC has the same look as per the numerous reviews we've see for the S20 Ultra and other phones,

Then why don't they in any of the images I took? It's actually not even close.

, you dont get different processing across different lenses in terms of jpeg IQ produced.

Yes, you do.

not in terms of SOOC jpegs

The SOOC jpegs from both of the S20 Ultras I tested were entirely different between the three cameras.

SHOW ME!!!!!!!!!!!!!!!!!!!!!!

Obviously the code will be different as the different sensors have different specs etc.

And hence, the results are different.

Differences include:

  • IR filters
  • Dye transmission curves
  • Well capacities
  • Read noise
  • Quantum efficiency
  • Microlens array
  • Lens aberrations
  • Black point measurement

All these things mean that processing is tuned to each sensor.

OMG we know this, but the SOOC jpegs have the same look across lenses,

No, they do not. The main camera is much more heavily noise reduced and smeared compared to the telephoto camera. It's not even close.

SHOW me your examples then...

ok, enough is enough, some people you just can't reach!

-- hide signature --
 Jostian's gear list:Jostian's gear list
Sony a9 Sony a6600 Sony FE 100-400mm F4.5-5.6 Tamron 28-75mm F2.8 III Tamron 20mm F2.8 Di III OSD +3 more
Jostian
Jostian Veteran Member • Posts: 4,322
Re: Samsung messed up the implementation

RDM5546 wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

RDM5546 wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

its hybrid, and the IQ is not great at all, as shown, or is it also out of focus. Your own photos of the tree showed very poor jpeg processing (smeary and waxy) as mentioned by numerous others.

From the main camera,

yes. Not from the telephoto camera

mmm, here's a 100% crop from the 108 mp jpeg file from another thread on the S20 Ultra, if you are happy with this, great for you, I know I wont be...

I've trimmed and highlighted this so your amazingly short attention span and inherent illiteracy can (hopefully) follow it.

oooh, such harshness... your 'inherent illiteracy' chirp is quite funny, coming from you! And you think the telephoto lens will miraculously use different jpeg processing to the other lenses...

It uses a Sony sensor instead of a Samsung sensor. It doesn't have pixel binning. So, yes.

but the Samsung processing algorithms will be the same.

if you look at the crops you'll see the processing is the same.

Nope.

You must be speaking loosely. Two processings from different source code are at best similiar in many cases but definitely are not exactly the same. Comparing two sets of different source code will not be exactly the same unless are identical which more than likely they not the identical in this case. Similar to your eye Jostian the cases may be. If you speak precisely how can they be the same?

as per any phone that has more than one camera (wide, UW and tele) the jpeg processing is the same in terms of jpeg IQ produced

That statement is entirely false.

no, the processing algorithms are the same, you don't get different looking outputs from different lens, the IQ SOOC has the same look as per the numerous reviews we've see for the S20 Ultra and other phones, the jpegs SOOC have the same look for the different lenses.

, you dont get different processing across different lenses in terms of jpeg IQ produced.

Yes, you do.

not in terms of SOOC jpegs

Obviously the code will be different as the different sensors have different specs etc.

And hence, the results are different.

Differences include:

  • IR filters
  • Dye transmission curves
  • Well capacities
  • Read noise
  • Quantum efficiency
  • Microlens array
  • Lens aberrations
  • Black point measurement

All these things mean that processing is tuned to each sensor.

OMG we know this, but the SOOC jpegs have the same look across lenses, look at the crops from the different lenses that have been posted they have the same look to them.

top crop is S20U wide lens, bottom crop is the ultrawide, see the same smeary look to the branch detail and leaves on both...same jpeg processing algorithms!

I agree with you that both images have similiar looking smear but to be precise I for one can nto realy say it is the same. I do not know what Zoom factors you used for each image. Is it 0.5X and 1X? They appear to be cropped as you state.

yes just to show the similar IQ output (similar processing for SOOC Jpegs).

Are the both cropped the same amount. From the angle of view and not knowing the zoom factor judging the smear factor that is present in both trees to some degree is hard to quantitatively judge. I do not believe from my picturee taking the smearing that I can see at times appears in all images at all light levels. I do disagree with your belief that smearing exists and that I do wish Samsung would work on this a lot and I also wish that Samsung would provide means for taking 108MP raw image for exporting to computer processing in the procamera mode.

I would like Verizon to hurrary up and distribute to US users the new version of the camera that Samsung provided them with to so that I see what is still there. AT&T has distributed it to their customers according to a Dpreview report yesterday so hopefull Verison version is not far away.

-- hide signature --
 Jostian's gear list:Jostian's gear list
Sony a9 Sony a6600 Sony FE 100-400mm F4.5-5.6 Tamron 28-75mm F2.8 III Tamron 20mm F2.8 Di III OSD +3 more
Lee Jay Forum Pro • Posts: 56,297
Re: Samsung messed up the implementation

Jostian wrote:

Lee Jay wrote:

The SOOC jpegs from both of the S20 Ultras I tested were entirely different between the three cameras.

SHOW ME!!!!!!!!!!!!!!!!!!!!!!

Same phone, taken at the same time, both 100% crops, both taken in full sunlight with fast shutter speeds, both with tap-to-focus on the subject.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM Canon Extender EF 2x III +10 more
Jostian
Jostian Veteran Member • Posts: 4,322
Re: Samsung messed up the implementation

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

The SOOC jpegs from both of the S20 Ultras I tested were entirely different between the three cameras.

SHOW ME!!!!!!!!!!!!!!!!!!!!!!

Same phone, taken at the same time, both 100% crops, both taken in full sunlight with fast shutter speeds, both with tap-to-focus on the subject.

Please post the originals.

-- hide signature --
 Jostian's gear list:Jostian's gear list
Sony a9 Sony a6600 Sony FE 100-400mm F4.5-5.6 Tamron 28-75mm F2.8 III Tamron 20mm F2.8 Di III OSD +3 more
Jostian
Jostian Veteran Member • Posts: 4,322
Re: Samsung messed up the implementation

See here from another S20U thread, just to show the poor jpeg processing, top crop is the SOOC jpeg and the bottom is from the RAW file. Why can't Samsung get their SOOC jpegs looking more like what you can get from RAW?? not this over-processed waxy rubbish.

-- hide signature --
 Jostian's gear list:Jostian's gear list
Sony a9 Sony a6600 Sony FE 100-400mm F4.5-5.6 Tamron 28-75mm F2.8 III Tamron 20mm F2.8 Di III OSD +3 more
Lee Jay Forum Pro • Posts: 56,297
Re: Samsung messed up the implementation

Jostian wrote:

Lee Jay wrote:

Jostian wrote:

Lee Jay wrote:

The SOOC jpegs from both of the S20 Ultras I tested were entirely different between the three cameras.

SHOW ME!!!!!!!!!!!!!!!!!!!!!!

Same phone, taken at the same time, both 100% crops, both taken in full sunlight with fast shutter speeds, both with tap-to-focus on the subject.

Please post the originals.

Those are crops from the originals.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM Canon Extender EF 2x III +10 more
Lee Jay Forum Pro • Posts: 56,297
Re: Samsung messed up the implementation

Jostian wrote:

See here from another S20U thread, just to show the poor jpeg processing, top crop is the SOOC jpeg and the bottom is from the RAW file. Why can't Samsung get their SOOC jpegs looking more like what you can get from RAW?? not this over-processed waxy rubbish.

And those are at what ratio (mine were 1:1) and from what cameras?

Without useful information, these mean nothing.

And if you can't answer the question you answered, you don't understand what "raw" means.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM Canon Extender EF 2x III +10 more
Keyboard shortcuts:
FForum MMy threads