[pics] SD10 vs. 20d sample

Started Feb 21, 2005 | Discussions
Jafalt Regular Member • Posts: 161
Thank You

Thank You for doing this test
It is well done

I would really like to see

  • a huge print

  • shot with the same lens (ex or canon)

  • Upsamling to Sigma Dobbelt size from 20D and virce versa

(why does have to be on Canons conditon/requirements all the time)

jac

dr.noise wrote:

20D + 50/1.4:

SD10 + 50/2.8 EX upsampled to 8mp:

Both shots made at F11, 1/200, ISO100.
20D shot saved from C1Pro, low noise reduction, no sharpening.
I saved SD10 shot double-size from SPP, then resized to 8mp in
ACDSee. I also tried saving same size then bicubic upsample in PS,
but got absolutely identical result.

Originals saved as 100% JPG:

; (20D; 8.7 mb)

; (SD10; 8.17 mb)

RAW files:
http://zdg.ru/tmp/200502/IMG_0802.CR2 (20D; 8.4 mb)
http://zdg.ru/tmp/200502/IMG07695.X3F (SD10; 8.04 mb)

Probably the SD10 shot could be upsampled better with less
aliasing. I was thinking about pulling sharpness setting back to
-1.0 before upsampling. You can try raw files yourselves.

About color: I didn't do any precise color management as my goal
was to compare details. 20D was shot on Auto WB and it got white
sheet right. So I assumed there is no need for further processing.
SD10 on Auto WB produced bluish white sheet, so I switched to
Custom WB.

And for what we have: 20D shot is mostly wrong except white sheet.
Striped cloth on the left is pinkish, should be more yellow like on
SD10 shot (spot on!), the toy dog is pinkish too (SD10 spot on),
blues are lighter than needed. Yellow stripe on the magazine cover
is too pale and wrong hue. SD10's one is right, but slightly
oversaturated. The red pepper of SD10 is right but a bit pale,
20D's is well-saturated but too orange. Yellow pepper is good on
both shots. I must note that reference white sheet is white on both
shots.

Pepe-Lepue Senior Member • Posts: 1,942
Re: Yo!!! Hold the horses!!! [IMG]

docmaas wrote:

thanks Pepe. I would give up on that sigma lens too if that is the
best it can do. It's bad. I can hardly believe Sigma is saying
its ok. I could see the "chip" was a reflection in the sigma image
but it sure looks like a "chip" in your image but overall you make
the canon look quite good.

The reason I bought the Sigma was to substitute my 24-70L while Canon was working on it. I've seen these Sigma 24-70 lenses produce better results than this, but now I am learning it's rare to get a stellar copy. Maybe that's why the lens cost less than $400?

I'm going to give Sigma another shot at making the lens right, but I have a feeling that they will again say that the lens is performing "optimal".

-- hide signature --

'I am ze locksmith of love, no?'
Stephen Reed

http://www.pbase.com/domotang

 Pepe-Lepue's gear list:Pepe-Lepue's gear list
Sony RX100 III Sony a6000 Sony E 30mm F3.5 Macro Sony Carl Zeiss Sonnar T* E 24mm F1.8 ZA Sony E 50mm F1.8 OSS +5 more
SigmaSD9 Senior Member • Posts: 2,013
Lenses

The 20D has a big lens advantage in this test. The 50/1.4 has much better color balance, no yellow cast (both evident in your pics) and is significantly sharper than the 50 EX at all apertures.

An f/11 comparision was a good choice, but if you do nothing but compare X3F file sizes at the aperture, the 50/1.4 might produce about a half a MB larger X3F than the 50 EX file, all things equal. That is about the closest they get.

I can offer you a 50/1.4 loaner for Sigma SA, but I don't know how well, quickly, or even if it would make the trip.

dr.noise wrote:

20D + 50/1.4:

SD10 + 50/2.8 EX upsampled to 8mp:

Both shots made at F11, 1/200, ISO100.
20D shot saved from C1Pro, low noise reduction, no sharpening.
I saved SD10 shot double-size from SPP, then resized to 8mp in
ACDSee. I also tried saving same size then bicubic upsample in PS,
but got absolutely identical result.

Neither is adaptive like the 20D's upscaling routine.

ghoerdt Senior Member • Posts: 1,250
Re: Yo!!! Hold the horses!!! [IMG]

Hi Pepe,

Pepe-Lepue wrote:

One thing I noticed just now is that the 20D has much better shadow
detail than the Sigma.

When I stated "Colour was no issue" that included the shadows also, I simply didnt pay any attention to it in that case.

Especially this can be optimized with Sigmas FillLight feature very well.

Greetings

Günter

Pepe-Lepue Senior Member • Posts: 1,942
Re: Yo!!! Hold the horses!!! [IMG]

dr.noise wrote:

For me both cameras seem to produce very similar results.

I tend to agree. I've always liked the Foveon images. In fact I considered switxghing to the Sigma system not long ago. I just couldn't make the Sigma system comfortable for me so I stuck with Canon (and bought the Nikon D70).

I didn't notice it. Where?

I was noticing the Canon's shadow details (the face of the statue and the shadowy top of the heart on the dog) are more clear with more detail.

-- hide signature --

'I am ze locksmith of love, no?'
Stephen Reed

http://www.pbase.com/domotang

 Pepe-Lepue's gear list:Pepe-Lepue's gear list
Sony RX100 III Sony a6000 Sony E 30mm F3.5 Macro Sony Carl Zeiss Sonnar T* E 24mm F1.8 ZA Sony E 50mm F1.8 OSS +5 more
Pepe-Lepue Senior Member • Posts: 1,942
Re: Yo!!! Hold the horses!!! [IMG]

ghoerdt wrote:

When I stated "Colour was no issue" that included the shadows also,
I simply didnt pay any attention to it in that case.

I figured that. That's why I was asking. More than likely the shadows got clipped (or truncated) during initial processing.

-- hide signature --

'I am ze locksmith of love, no?'
Stephen Reed

http://www.pbase.com/domotang

 Pepe-Lepue's gear list:Pepe-Lepue's gear list
Sony RX100 III Sony a6000 Sony E 30mm F3.5 Macro Sony Carl Zeiss Sonnar T* E 24mm F1.8 ZA Sony E 50mm F1.8 OSS +5 more
Pepe-Lepue Senior Member • Posts: 1,942
Re: Lenses

Agreed.
The 50/1.4 is going to produce an advantage over the 50EX

SigmaSD9 wrote:
The 20D has a big lens advantage in this test. The 50/1.4 has much
better color balance, no yellow cast (both evident in your pics)
and is significantly sharper than the 50 EX at all apertures.

An f/11 comparision was a good choice, but if you do nothing but
compare X3F file sizes at the aperture, the 50/1.4 might produce
about a half a MB larger X3F than the 50 EX file, all things equal.
That is about the closest they get.

I can offer you a 50/1.4 loaner for Sigma SA, but I don't know how
well, quickly, or even if it would make the trip.

dr.noise wrote:

20D + 50/1.4:

SD10 + 50/2.8 EX upsampled to 8mp:

Both shots made at F11, 1/200, ISO100.
20D shot saved from C1Pro, low noise reduction, no sharpening.
I saved SD10 shot double-size from SPP, then resized to 8mp in
ACDSee. I also tried saving same size then bicubic upsample in PS,
but got absolutely identical result.

Neither is adaptive like the 20D's upscaling routine.

-- hide signature --

'I am ze locksmith of love, no?'
Stephen Reed

http://www.pbase.com/domotang

 Pepe-Lepue's gear list:Pepe-Lepue's gear list
Sony RX100 III Sony a6000 Sony E 30mm F3.5 Macro Sony Carl Zeiss Sonnar T* E 24mm F1.8 ZA Sony E 50mm F1.8 OSS +5 more
SigmaSD9 Senior Member • Posts: 2,013
Re: [pics] SD10 vs. 20d sample

dr.noise wrote:

ghoerdt wrote:

So much for 8.2MP being more than 3.34x3MP,
and for the 50/1.4 killing the EX 50 Macro...

Actually, there is some aliasing in SD10 upsampled shot. I think it
can be reduced by decreasing sharpness before upsampling. I am
still not good at upsampling tasks.

And 50/1.4 is really excellent lens and pleasant to work with.
Might outperform 50EX at 2.8, but I would not say kill, no.

It kills it. The difference is largest at or around f/2.8, where the 50 EX is very soft and the 50/1.4 is already approaching optimum sharpness, which probably occurs around f/3.5 to f/4.

With both at f/2.8 using a well controlled scene, if the 50/1.4 produces a 9MB X3F, the 50EX might produce a 6.5MB file, just to give you an idea.

I have plenty SD10 shots with 50EX delivering single-pixel and
maybe sup-pixel resolution. That means, even if 50/1.4 is sharper,
it will not gain any extra resolution on SD10.

I recommend you use both before reaching a conclusion.

SigmaSD9 Senior Member • Posts: 2,013
Re: Lenses

Pepe-Lepue wrote:

Agreed.
The 50/1.4 is going to produce an advantage over the 50EX

I know it will too. If you do nothing but look at the thumbnails, the yellow cast of the SD10 shot is purely a glass issue.

dr.noise
OP dr.noise Veteran Member • Posts: 3,767
Re: Lenses

SigmaSD9 wrote:

The 20D has a big lens advantage in this test. The 50/1.4 has much
better color balance, no yellow cast (both evident in your pics)

But funny thing is that the yellow cast is real. Colors were THAT yellow in real, rather than Canon's pinkish shades. I double checked them. And the white sheet is white - no cast. Looking at Canon shot you'll never believe its colors are off - until you see the real scene. Interesting thing that ACR fixed the reds - the red pepper is much better in ACR than is SPP.

and is significantly sharper than the 50 EX at all apertures.

Looking at the signs at the bottom of blue Tesla box, I cannot imagine any lens can be any sharper since they are already beyond the sensor's resolution limit. There are plenty of single-pixel details in other areas.

I can offer you a 50/1.4 loaner for Sigma SA, but I don't know how
well, quickly, or even if it would make the trip.

Thanks alot, but no. It would be very long and dangerous trip for a lens. I'll better find someone with 50/1.4 in Moscow (And I probably know who

 dr.noise's gear list:dr.noise's gear list
Sigma DP2s Sony DSC-RX0 Sigma SD10 Sigma fp
dr.noise
OP dr.noise Veteran Member • Posts: 3,767
Re: Lenses

SigmaSD9 wrote:

Pepe-Lepue wrote:

Agreed.
The 50/1.4 is going to produce an advantage over the 50EX

I know it will too. If you do nothing but look at the thumbnails,
the yellow cast of the SD10 shot is purely a glass issue.

Again: there is no yellow cast in Sigma shot. There is pink cast in Canon shot.
Both backgrounds are pure white. No cast.
Canon renders striped cloth and toy dog pinkier.
Sigma yellower.
The truth is that Sigma renders them just right.
The only Sigma flaw is pale red pepper which gets completely fixed in ACR.

 dr.noise's gear list:dr.noise's gear list
Sigma DP2s Sony DSC-RX0 Sigma SD10 Sigma fp
dr.noise
OP dr.noise Veteran Member • Posts: 3,767
Re: Yo!!! Hold the horses!!! [IMG]

Pepe-Lepue wrote:

You had to process the images . . . right? Using any RAW converter
is a type of process applied to the data to get the image.

Right. But by "processing" I mean extra steps like sharpening.

There is no doubt that the Bayer needs to have some USM applied to
optimize the image, but when done at (300, .3. 2) the amount of
halo is hardly visible.

But I see it perfectly in your example. You've done really good job improving 20D shot, I cannot argue. But the halos are visible. And SD10 looks soft beside it, but it wasn't sharpened as much and there are no halos. That means SD10 shot can be sharpened further, and you'll see it will no longer be that soft.

BTW, I tried SD10 shot upsampling with minimum initial RAW sharpening setting. That produced much less aliasing, and then I applied sharpening just like you mentioned (300 .3 .2) and got very similar results. Halos, too. And both cameras may have resolution advantage in certain areas.

Most pros don’t pixel-peep like I do (and most here in these forums)

 dr.noise's gear list:dr.noise's gear list
Sigma DP2s Sony DSC-RX0 Sigma SD10 Sigma fp
SigmaSD9 Senior Member • Posts: 2,013
Re: Lenses

dr.noise wrote:

SigmaSD9 wrote:

Pepe-Lepue wrote:

Agreed.
The 50/1.4 is going to produce an advantage over the 50EX

I know it will too. If you do nothing but look at the thumbnails,
the yellow cast of the SD10 shot is purely a glass issue.

Again: there is no yellow cast in Sigma shot.

I find the bear in particular a bit yellow in the SD10 shot, with an overall yellow cast to the scene as well. That is almost purely due to lens color balance, not sensors.

There is pink cast in
Canon shot.

Well then you'd lean toward the same pink cast in the SD10 shot with a 50/1.4 attached. However you want to characterize the difference, the difference that you are noting is not atypical of the lens change alone.

dr.noise
OP dr.noise Veteran Member • Posts: 3,767
Re: Lenses

SigmaSD9 wrote:

I find the bear in particular a bit yellow in the SD10 shot

...because it is!
I checked this specially. Twice.

Looking at 20d shot I was completely sure it is right and SD10 is wrong. But after examination of the real scene SD10 happened to be right.

an overall yellow cast to the scene as well.

can't be overall cast as all whites are whites.
if there is some cast then it exists in real.

That is almost purely due to lens color balance, not sensors.

I used Custom balance on SD10. Shouldn't it deal with lens color balance?

Well then you'd lean toward the same pink cast in the SD10 shot
with a 50/1.4 attached. However you want to characterize the
difference, the difference that you are noting is not atypical of
the lens change alone.

I am just noting it by the way, this is not the subject of my test right now. I wanted to note it specially for eliminating any "yellow cast" suspicions for SD10.

 dr.noise's gear list:dr.noise's gear list
Sigma DP2s Sony DSC-RX0 Sigma SD10 Sigma fp
pyatak Forum Member • Posts: 70
Re: Lenses

SigmaSD9 wrote:

the yellow cast of the SD10 shot is purely a glass issue.

Again: there is no yellow cast in Sigma shot.

I find the bear in particular a bit yellow in the SD10 shot, with
an overall yellow cast to the scene as well. That is almost purely
due to lens color balance, not sensors.

hahaha, you are trying to argue with person who can actually see this dog (it's not a bear) with his own eyes in exact light conditions. if you preffer this dog to be pink not yellow it doesn't mean that sigma lens wrong

dr.noise
OP dr.noise Veteran Member • Posts: 3,767
BTW...

the thumbnails are deceiving.

For example, I cannot tell that 20d's blues are too light. They seem completely similar to SD10's blues. But in original images there is obvious difference and 20d's blues are lighter indeed.

SigmaSD9 wrote:

dr.noise wrote:

SigmaSD9 wrote:

Pepe-Lepue wrote:

Agreed.
The 50/1.4 is going to produce an advantage over the 50EX

I know it will too. If you do nothing but look at the thumbnails,
the yellow cast of the SD10 shot is purely a glass issue.

Again: there is no yellow cast in Sigma shot.

I find the bear in particular a bit yellow in the SD10 shot, with
an overall yellow cast to the scene as well. That is almost purely
due to lens color balance, not sensors.

There is pink cast in
Canon shot.

Well then you'd lean toward the same pink cast in the SD10 shot
with a 50/1.4 attached. However you want to characterize the
difference, the difference that you are noting is not atypical of
the lens change alone.

 dr.noise's gear list:dr.noise's gear list
Sigma DP2s Sony DSC-RX0 Sigma SD10 Sigma fp
J. Michael Veteran Member • Posts: 4,673
An obvious difference to me...........

is the different way the cameras render the Baileys lable.
the curved red area with the Baileys lettering in it seems harsher
in the SD10 picture than in the 20D image.

Which one is more accurate is not for me to say, but I do find the tonal gradation in the 20D more pleasing.

Mike

Pepe-Lepue wrote:

I don't know how you processed the 20D image, but frankly I think
you didn't optimally do it. Since the 20D is part of my workflow on
a daily basis, I will show how I might have processed this shot.
I really appreciate you posting the RAW images.

Here is my comparison of the 20D processed using my workflow (Adobe
ACR) verses the image you posted.
If you can't see the image below due to the incompetence of PBase
lately, then please click the link below;
http://www.pbase.com/domotang/image/40007480/original

Jafalt Regular Member • Posts: 161
Re: Lenses

What about the monitor?

pyatak wrote:

SigmaSD9 wrote:

the yellow cast of the SD10 shot is purely a glass issue.

Again: there is no yellow cast in Sigma shot.

I find the bear in particular a bit yellow in the SD10 shot, with
an overall yellow cast to the scene as well. That is almost purely
due to lens color balance, not sensors.

hahaha, you are trying to argue with person who can actually see
this dog (it's not a bear) with his own eyes in exact light
conditions. if you preffer this dog to be pink not yellow it
doesn't mean that sigma lens wrong

dr.noise
OP dr.noise Veteran Member • Posts: 3,767
Canon's phenomena

pyatak wrote:

hahaha, you are trying to argue with person who can actually see
this dog (it's not a bear) with his own eyes in exact light
conditions. if you preffer this dog to be pink not yellow it
doesn't mean that sigma lens wrong

You know, even I who have seen the dog, even I could not believe that it is wrong on 20d shot! I had to force myself to look at the real dog several times to be sure that it is not a trick. 20d has some ability to convince you that colors are right. They look OK while you look at the picture. So OK that you couldn't think about them being wrong. But look at the real scene, and the difference can be perfectly seen.

Looking at Sigma shot you are thinking "now that's yellow cast!" but looking at real scene you see the same cast.

 dr.noise's gear list:dr.noise's gear list
Sigma DP2s Sony DSC-RX0 Sigma SD10 Sigma fp
pyatak Forum Member • Posts: 70
pleasant colors vs real colors

dr.noise wrote:

real dog several times to be sure that it is not a trick. 20d has
some ability to convince you that colors are right. They look OK
while you look at the picture. So OK that you couldn't think about
them being wrong. But look at the real scene, and the difference
can be perfectly seen.

isn't that the point of auto wb? give people what they expect to see. i think you should have used gray card for both shots...

Looking at Sigma shot you are thinking "now that's yellow cast!"
but looking at real scene you see the same cast.

hehe, reality check...

Keyboard shortcuts:
FForum MMy threads