Digital Zoooooom!!! Another experiment!;-)

[nahau]

A while back, I did an experiment with digital zoom. The main discussion involved myself and Nadim. A lot of you don't know Nadim, but he is very knowledgable in the use of Photoshop, Genuine fractals, etc. In any case I thought it would be interesting to see just how many people think of digital zoom as a "marketing ploy" with no added value.

Below are two captures split left and right. I won't say which is the digitally zoomed image and which is computer enlarged...it would be more fun to see how many people can guess which is which...correctly!;-)

I purposely did not put the digitally zoomed photos on any specified side of the either capture...which means, that the digitally zoomed photos could either be on the left side or the right side. Hmmmm...which ones are they??





Any Guesses???
nahau

 

[Blur]

fractals, etc. In any case I thought it would be interesting to
see just how many people think of digital zoom as a "marketing
ploy" with no added value.

I actually do use the digital zoom on my Oly 3030 on the rare occassion that I'm just too far away and I need the shot. Yes it does look visibly "worse" than pure optical zoom, but sometimes it just can't be helped. I don't find it as absolutely horrible as people make it out to be. No one in my family seemed to complain when I gave them prints of the photos I used digital zoom on, they were all really impressed that I could even get a decent shot at that distance.

I found the digital zoom on the Fuji FinePix 4700 was actually pretty good. The camera already used interpolation on it's default resolution, a little more camera processing and the use of SuperCCD arranged pixels seemed to not have as much of an adverse effect on it's digital zoom photos as regular CCD based cameras. Just my experiences.

I purposely did not put the digitally zoomed photos on any
specified side of the either capture...which means, that the
digitally zoomed photos could either be on the left side or the
right side. Hmmmm...which ones are they??

Any Guesses???

Now I'm hoping you didn't deliberatly dirtied-up the non-digital zoom photo here... =)

For the first image, I'd say it was the right image. There's more pixelation which is particularly noticable in the reflected areas, and the logo or whatever it is at the top of the photo to the left of the metal circle.

For the second, I'd say it was the left image. This one was a little harder since it is predominantly a solid colour in the back. However, there is noticable pixelation on the Rayovac tag (you can't even read it in the left image), the edges of the pens, and very noticably in the brushes shadow.

It's hard to tell because your screen shots are so small compared to what the actual image would be. The resized screen shots are causing a lot of lost pixels (just look at the title bar), so it isn't really indicative of what the images really look like. What camera is this from?

 

[nahau]

Hi Blur,

These photos were taken with my G1. Don't worry, the pixelization is not a problem, its just the screen capture that makes the title bar look weird. The photos are actually as clean as I shot them less a little web compression which affected each photo equally...I made sure to check that before posting. All the photos are unedited and actually the digital zoomed photo is at 100%. They were all shot using a B300 teleconverter placed on the G1 at full zoom (with macro enabled) so the actual size is not far from what is represented here...and, they are not "dirtied-up" LOL!

So your guess is that the digitally zoomed photos are "right" and "left" respectfully. Very interesting. We will see what others say before I reveal the truth. Make sure to come back and check this thread out later. Hopefully more will participate in it.
Thanks for your input.
nahau

fractals, etc. In any case I thought it would be interesting to
see just how many people think of digital zoom as a "marketing
ploy" with no added value.

I actually do use the digital zoom on my Oly 3030 on the rare
occassion that I'm just too far away and I need the shot. Yes it
does look visibly "worse" than pure optical zoom, but sometimes it
just can't be helped. I don't find it as absolutely horrible as
people make it out to be. No one in my family seemed to complain
when I gave them prints of the photos I used digital zoom on, they
were all really impressed that I could even get a decent shot at
that distance.

I found the digital zoom on the Fuji FinePix 4700 was actually
pretty good. The camera already used interpolation on it's default
resolution, a little more camera processing and the use of SuperCCD
arranged pixels seemed to not have as much of an adverse effect on
it's digital zoom photos as regular CCD based cameras. Just my
experiences.

I purposely did not put the digitally zoomed photos on any
specified side of the either capture...which means, that the
digitally zoomed photos could either be on the left side or the
right side. Hmmmm...which ones are they??

Any Guesses???

Now I'm hoping you didn't deliberatly dirtied-up the non-digital
zoom photo here... =)

For the first image, I'd say it was the right image. There's more
pixelation which is particularly noticable in the reflected areas,
and the logo or whatever it is at the top of the photo to the left
of the metal circle.

For the second, I'd say it was the left image. This one was a
little harder since it is predominantly a solid colour in the back.
However, there is noticable pixelation on the Rayovac tag (you
can't even read it in the left image), the edges of the pens, and
very noticably in the brushes shadow.

It's hard to tell because your screen shots are so small compared
to what the actual image would be. The resized screen shots are
causing a lot of lost pixels (just look at the title bar), so it
isn't really indicative of what the images really look like. What
camera is this from?

 

[bob g]

For the first image, I'd say it was the right image. There's more
pixelation which is particularly noticable in the reflected areas,
and the logo or whatever it is at the top of the photo to the left
of the metal circle.

For the second, I'd say it was the left image. This one was a
little harder since it is predominantly a solid colour in the back.
However, there is noticable pixelation on the Rayovac tag (you
can't even read it in the left image), the edges of the pens, and
very noticably in the brushes shadow.

My guess is same as Blur's!

Hi Nahau,

How ya doing?

Guess you better get that Penn Reel in the first shot ready. Lotsa Albacore, Bluefin and yellotail caught about 100 mi south of SD last few days.

Best regards,

Bob G (HB, Ca)

 

[nahau]

Hi bob,

Funny you mention Albacore. I just went fishing 2 weeks ago, just as the Albacore run began deminishing. I only caught one and my friend caught one and also one yellowtail. The entire boat didn't do too well. We went out overnight on the Legend out of Seaforth landing (SD). The bummer part of the whole trip, was that while we were putting our tackle in the truck, someone stole our fish! We were so PO'ed...to say the least!!! I am waiting for the count to go back up then probably take another trip.
Thanks for the input...for the thread...and the fishing!!;-)
nahau

For the first image, I'd say it was the right image. There's more
pixelation which is particularly noticable in the reflected areas,
and the logo or whatever it is at the top of the photo to the left
of the metal circle.

For the second, I'd say it was the left image. This one was a
little harder since it is predominantly a solid colour in the back.
However, there is noticable pixelation on the Rayovac tag (you
can't even read it in the left image), the edges of the pens, and
very noticably in the brushes shadow.

My guess is same as Blur's!

Hi Nahau,

How ya doing?

Guess you better get that Penn Reel in the first shot ready. Lotsa
Albacore, Bluefin and yellotail caught about 100 mi south of SD
last few days.

Best regards,

Bob G (HB, Ca)

 

[Tom Davenport]

Nahaua, are you really comparing apples to apples? When a camera digitally zooms an image, is it really "enlarging" it, or simply cropping it and producing output file with smaller size dimensions? If the latter, then I would expect digital enlargement to look worse than digital zoom. However, even if digital zoom were actually "enlarging" the image, it still may not be an apples to apples comparison if Microsoft Photo Editor (which you seem to be using in this example) uses a different enlargement technique than the camera. You could make the same argument for in camera sharpening, noise reduction, etc. Whether or not the camera does a "better job" depends on the sophistication of the software you are using for comparison.

In any case, the images on the top right and bottom left do not look as good as their counterparts. My guess is that these are digitally enlarged pictures (as opposed to the digitally zoomed pictures) due to Microsoft Photo Editor being used for the enlargement...Tom

I won't say which is
the digitally zoomed image and which is computer enlarged

 

[sarbos]

Hello!

Digital zoomed:
1.- Right (quite obvious pixelation in the curve)
2.- Left (quite obvious pixelation in the paintbrush)

did we win?

Sarbos

 

[sarbos]

Hello!

I would expect digital enlargement to look
worse than digital zoom.

???

Can you define both, please? they are the same to me...

 

[nahau]

Hi Tom,

A valid concern. Regardless, the comparison is a fair one. The MSPhotoed is only being used for viewing. I always use it because it is easy to position these types of comparison photos without the larger interface of PS or PSP7, of which, I own both. That said, comparison photos in both PS, and PSP7 show the same results for pixelization and such. These captures were done with PSP7 by the way.

The gist of the experiment is to show whether or not digital zoom is a viable alternative to "enlarging". A lot of people contend that digital zoom is useless, and say that taking a photo then enlarging the part that you need is better than using digital zoom. Far be it from me to make that determination without some sort of data.

It is curious that you state you feel digital enlargement would look "worse" than digital zoom. I guess you feel that way because MSPhoto editor is not as "expensive" as PS or PSP7 and cannot do as good a job in a test of this sort...so I do understand your reasoning. As stated above though, the results are the same in all three applications.
Thanks for the input,
nahau

In any case, the images on the top right and bottom left do not
look as good as their counterparts. My guess is that these are
digitally enlarged pictures (as opposed to the digitally zoomed
pictures) due to Microsoft Photo Editor being used for the
enlargement...Tom

I won't say which is
the digitally zoomed image and which is computer enlarged

 

[nahau]

Hi sarbos,

I always look forward to your posts. They are always light and cheerful for me to read!;-)
I will reveal the answers later...and then you can see if you won!
nahau

Hello!

Digital zoomed:
1.- Right (quite obvious pixelation in the curve)
2.- Left (quite obvious pixelation in the paintbrush)

did we win?

Sarbos

 

[nahau]

Here are the same photos as processed with PS5.5. They were screen captured with PSP7...the same program used with the previous MSPhotoed captures. If there is any doubt whether or not MSPhotoed can be used for "viewing" experimentals, these PS5.5 images should remove that. I tried to capture them exactly the same, but it was difficult...they are close though and the order and placement for each photo is the same.





regards...
nahau

 

[Tom Davenport]

I would expect digital enlargement to look
worse than digital zoom.

???

Can you define both, please? they are the same to me...

digital zoom: camera takes raw image, crops it, [presumably] enlarges it, and writes compressed jpg.

digital (outside of camera) enlargement: camera takes raw image, writes compressed jpg, Nahau uncompresses jpg, crops it, enlarges it (technique dependent on software used by Nahau), and writes [re]compressesed jpg.

Ignoring differences in the enlargement technique between the camera and Nahau's software, I would expect digital (outside of camera) enlargement to look worse due to the extra compression of the jpg.

Basically, I don't think folks have enough information to make an educated choice. For example, is Nahau's digitally enlarged imaged done straight from a RAW image file? If so, there is not an extra jpg compression. But then comparison is not apples vs. apples due to essentailly comparing RAW vs. jpg output from the camera...Tom

 

[Tom Davenport]

Hello!

Digital zoomed:
1.- Right (quite obvious pixelation in the curve)
2.- Left (quite obvious pixelation in the paintbrush)

did we win?

Sarbos

I have to chuckle why folks seem to think digital zoom should look worse.
Ignoring any...

1) differences in the enlargement "technique" used by the camera vs. external software
2) distortions created by additional or differing quality jpeg compression

there should be absolutely NO difference in quality between the two pictures, or am I missing something?

 

[nahau]

Tom,

Yes, I believe you are...if you think there should be "NO" difference in quality with the two photos. The results are obvious! All photos by the way were shot in jpeg mode...which means they all have the same camera compression. You cannot rule out camera vs computer processing as sooner or later, you will have to load your files into your computer for processing. Agreed that different programs use different algos for compression, but if you look at the updated PS5.5 photos, do you think we are talking "apples to apples" now?

Let me restate the purpose of the experiment in other words in case you are reading more into it than there really is. The experiment is meant to find out just how many people think digital zooming is useless. Nothing more! The results of the experiment may be surprising to some and maybe will make people realize some things too. We will see!

I am not trying to convince you one way or the other, it is just an experiment...and so laugh at people if you want, but I personally don't think it's very polite...maybe we are not as intelligent as you! ;-)
nahau

Hello!

Digital zoomed:
1.- Right (quite obvious pixelation in the curve)
2.- Left (quite obvious pixelation in the paintbrush)

did we win?

Sarbos

I have to chuckle why folks seem to think digital zoom should look
worse.
Ignoring any...
1) differences in the enlargement "technique" used by the camera
vs. external software
2) distortions created by additional or differing quality jpeg
compression

there should be absolutely NO difference in quality between the two
pictures, or am I missing something?

 

[Yang]

I remember only one or twice used digital zoom and compared the results with detail enlargement of the corresponding "normal picture". I've got the impression the former looks better (smoother). So, I quess is
Pic 1 digital Zoom left
Pic 2 digital Zoom right

The same discussion can be applied to ISO settings: My experiments showed it's better to shot at a higher ISO value than to have a underexposed picture and try to make it brighter with PC software afterwards while shooting in a low lighting condition without a tripod.

Yang

A while back, I did an experiment with digital zoom. The main
discussion involved myself and Nadim. A lot of you don't know
Nadim, but he is very knowledgable in the use of Photoshop, Genuine
fractals, etc. In any case I thought it would be interesting to
see just how many people think of digital zoom as a "marketing
ploy" with no added value.
Below are two captures split left and right. I won't say which is
the digitally zoomed image and which is computer enlarged...it
would be more fun to see how many people can guess which is
which...correctly!;-)
I purposely did not put the digitally zoomed photos on any
specified side of the either capture...which means, that the
digitally zoomed photos could either be on the left side or the
right side. Hmmmm...which ones are they??





Any Guesses???
nahau

 

[nahau]

Hello Yang,

Thank you for your input. I do remember some very nice fruit photos that you shot on a rainy day with higher iso...so I believe there is much credit to your statement on doing it right the first time then trying to fix it later! ;-)
nahau

The same discussion can be applied to ISO settings: My experiments
showed it's better to shot at a higher ISO value than to have a
underexposed picture and try to make it brighter with PC software
afterwards while shooting in a low lighting condition without a
tripod.

Yang

A while back, I did an experiment with digital zoom. The main
discussion involved myself and Nadim. A lot of you don't know
Nadim, but he is very knowledgable in the use of Photoshop, Genuine
fractals, etc. In any case I thought it would be interesting to
see just how many people think of digital zoom as a "marketing
ploy" with no added value.
Below are two captures split left and right. I won't say which is
the digitally zoomed image and which is computer enlarged...it
would be more fun to see how many people can guess which is
which...correctly!;-)
I purposely did not put the digitally zoomed photos on any
specified side of the either capture...which means, that the
digitally zoomed photos could either be on the left side or the
right side. Hmmmm...which ones are they??





Any Guesses???
nahau

 

[Tom Davenport]

Nahau, I wasn't trying to be impolite or come accross as big headed (a certain someone in this forum does it much better than I . I just wanted to better understand why folks think digital zoom should look worse? I mean, what you are doing is essentially "digital zooming" outside of the camera. What rationale is there for thinking external zooming should look "better"? Other than the various hypotheses I have already made, perhaps the quality of the two pictures BEFORE the internal or external digital zooming is different. Either that or I am simply clueless as to how digital zooming works, which is entirely likely...Tom

Hello!

Digital zoomed:
1.- Right (quite obvious pixelation in the curve)
2.- Left (quite obvious pixelation in the paintbrush)

did we win?

Sarbos

I have to chuckle why folks seem to think digital zoom should look
worse.
Ignoring any...
1) differences in the enlargement "technique" used by the camera
vs. external software
2) distortions created by additional or differing quality jpeg
compression

there should be absolutely NO difference in quality between the two
pictures, or am I missing something?

 

[Bob Williams]

Actually, i would have thought the same, and that the tip off
is the display artifacts caused by not resizing the other, but instead
showing at a percentage on the screen when using the capture.
I would say left/right.

The same discussion can be applied to ISO settings: My experiments
showed it's better to shot at a higher ISO value than to have a
underexposed picture and try to make it brighter with PC software
afterwards while shooting in a low lighting condition without a
tripod.

Yang

A while back, I did an experiment with digital zoom. The main
discussion involved myself and Nadim. A lot of you don't know
Nadim, but he is very knowledgable in the use of Photoshop, Genuine
fractals, etc. In any case I thought it would be interesting to
see just how many people think of digital zoom as a "marketing
ploy" with no added value.
Below are two captures split left and right. I won't say which is
the digitally zoomed image and which is computer enlarged...it
would be more fun to see how many people can guess which is
which...correctly!;-)
I purposely did not put the digitally zoomed photos on any
specified side of the either capture...which means, that the
digitally zoomed photos could either be on the left side or the
right side. Hmmmm...which ones are they??





Any Guesses???
nahau

 

[sarbos]

Hello!

Hi sarbos,
I always look forward to your posts. They are always light and
cheerful for me to read!;-)

Glad to know!

 

[nahau]

Hi Tom,

I really felt like retaliating with another insult, but I see that exchanging insults will not solve anything. It will only aggravate the matter. So..lets start again!

I am an engineer by profession and in my work, experimentation is part of my daily routine. I have been doing engineering for 20 years and so naturally, I carry it into my "hobby" as well.

I have read Phils comments on digital zooming...and while I never take his word for gospel, it did make me wonder just how many people did. Phil goes on to state that digital zooming is a "marketing ploy". While he may have been referring to other cameras...and not the G1, it makes me wonder if he even tested the G1's capabilities before posting that blanket statement.

I wanted to let this thread run for a while and find out thoughts about using digital zooming. Whether or not people think it is useful...or not. From you posting, you must think it is not. My thoughts really don't matter one way or the other as I am not you and won't be using your camera to take pictures with.

There are some people here that would benefit from the knowledge gained by experiments done with these cameras...just read the Raw vs jpeg thread, and you will see comments that lead to this conclusion. It is important that this forum give out as much information possible because of instances like the magenta cast issue which lead to many months of verbal abuse between forum members...all of which could have been avoided with better understanding of how to use the camera. MichaelW who has also been looking at the magenta casting issue has also stated that it is more prevalent in superfine as compared to fine compression. This is another finding that will help people...providing of course they read it.

It is not my appointed venue to help everyone here, but I like to think that I can contribute some knowledge...no matter how miniscule.

Anyways, I probably gave away the answer, with this reply. I had hoped to get more responses and make this "fun" rather than what it turned into. There are many who need this kind of exercise in thinking. Those with more experience will probably get the answers correct, those with less, will probably not. But in the end, at least something may be learned.
nahau

Hello!

Digital zoomed:
1.- Right (quite obvious pixelation in the curve)
2.- Left (quite obvious pixelation in the paintbrush)

did we win?

Sarbos

I have to chuckle why folks seem to think digital zoom should look
worse.
Ignoring any...
1) differences in the enlargement "technique" used by the camera
vs. external software
2) distortions created by additional or differing quality jpeg
compression

there should be absolutely NO difference in quality between the two
pictures, or am I missing something?