I was thinking the other day about sharpness in a photograph. There is nothing that compares to pure lens quality. Lets talk about hard core stuff here and not these Photoshop moves that any novice can get. So I want to ask the resources of the forum if you can post your

SHARPEST IMAGE WITHOUT ANY PHOTOSHOP MANIPULATIONS

Dont give me the Unsharp Mask or High Pass Filter images. No Parc Cruz unsharp mask photos please (sorry Parc:Grin: ). Just show us what kind of sharpness you can get from the quality of your lens. I believe that lens quality is the best for sharpness and anything else looks fake.

So let see how sharp you guys can get.

And if someone posts an Unsharp Mask or High Pass Fllter, I ask the moderators permission if it would be okay to rip the post to pieces.

:) Just curious Gabriel... How can you possibly accurately judge an image if it was sharpened or not Online and infront of your monitor; if most posts here are already cropped, resized, edited and saved for web-use? And even if, someone posted a full image, with full exif info, it can easily be fabricated--Not unless you own Image authenticity verification software of all major camera manufacturer available... :D

Would you care to start Gabriel? :)

I was thinking the other day about sharpness in a photograph. There is nothing that compares to pure lens quality. Lets talk about hard core stuff here and not these Photoshop moves that any novice can get. So I want to ask the resources of the forum if you can post your

SHARPEST IMAGE WITHOUT ANY PHOTOSHOP MANIPULATIONS

Dont give me the Unsharp Mask or High Pass Filter images. No Parc Cruz unsharp mask photos please (sorry Parc:Grin: ). Just show us what kind of sharpness you can get from the quality of your lens. I believe that lens quality is the best for sharpness and anything else looks fake.

So let see how sharp you guys can get.


Hi Gabriel,

On current Canon DSLRs, the AA filter somewhat softens the image that passed through the lens to mitigate moire and lessen false colors. As a result, what the sensor records is less sharp than what passed through the glass. So if absolutely no sharpening is applied to the photo during RAW conversion, we'll not see the full sharpness potential of the lens + sensor combo.

In-camera JPEGs of course always get some degree of sharpening during on-board conversion.

Let me quote a part of Canon's EOS white paper as regards AA filters and the need for sharpening:



"If little or no in-camera sharpening has been applied prior to viewing an EOS-1 class Digital SLR image at 100% magnification, especially with the EOS-1D Mark II, most photographers will agree that the image looks soft, even if it has been properly focused and exposed. This is mostly due to the detail-softening effect of the camera’s anti-aliasing (AA) filter, which is permanently mounted in front of the image sensor. The AA filter plays an important role in reducing or eliminating false colors and moiré, but the trade-off is that images must be sharpened either in the camera or in the computer prior to final usage."


If we really want to see the detail gathering performance of a certain lens, we have to do some sharpening to negate the effects of the AA filter. A processed 100% crop of a detailed pic is sufficient to give us an idea of the system resolution (glass + sensor). To the trained eye, it's actually easy to differentiate captured detail from sharpening artifacts/haloes. After all, no amount of PS manipulation can create detail where none is captured.

Just my humble opinion, of course.

Romy

Earl,

It can be done. The question is whether we are comparing apples to apples in the end. :)

But what's the point of the whole exercise? There are MTF graphs available on the net to indicate the len's resolving capabilities if sharpness is all you're looking for. Comparing shots of the same lens taken with poor vs bad technique would immediately introduce variability in the comparison. And there is evidence that there are sample variances - even with L lenses.

For me, this is like doing a half decent job of reinventing the wheel.

I was thinking the other day about sharpness in a photograph. There is nothing that compares to pure lens quality. Lets talk about hard core stuff here and not these Photoshop moves that any novice can get. So I want to ask the resources of the forum if you can post your

SHARPEST IMAGE WITHOUT ANY PHOTOSHOP MANIPULATIONS

Dont give me the Unsharp Mask or High Pass Filter images. No Parc Cruz unsharp mask photos please (sorry Parc:Grin: ). Just show us what kind of sharpness you can get from the quality of your lens. I believe that lens quality is the best for sharpness and anything else looks fake.

So let see how sharp you guys can get.

Let's see what you got...hehehehe...

No more talk. Here's my sharpest image (please don't rip this post to pieces). :D

http://static.flickr.com/94/230504931_12fce31ade_o.jpg

This was taken last night at OWG - Canon Users EB.

I was looking through some old photos taken with film and for some reason they look much sharper than my digital photos. I can try scanning some of these but my cheapo scanner automatically sharpens. Maybe I'll take them to a professional lab and tell them to turn off the sharpening and post them here.

Now, going back to sharp photos, I am not interested in seeing charts or tests that Arnel said can be found. I was hoping to see real world photos here. I personally feel that my pictures taken digitally look soft compared to film.

This thread is not meant to be a contest. I was hoping it would educate me and others on how well you guys can get sharp photos, and if you did, share with us your technique. I see that my second post here looked kinda intimidating so I'm going to delete it now. I was just getting tired of seeing all these fake sharp pictures that everyone puts here using the Unsharp Mask filter at maximum settings.

So can anyone share us their sharpest digital photo? Hope we can learn something here.

Thanks for posting Dys! Is it the reflection or did you run a High Pass filter on this image?

Thanks for posting Dys! Is it the reflection or did you run a High Pass filter on this image?
1. I took the picture using maximum contrast, sharpness, saturation, color tone parameters (Canon 350D) in large setting.
2. I resized it to 5 x 7.5 inches using Adobe Photoshop CS.
3. I applied Unsharp Mask and High Pass filter.
4. I saved it using Save For Web (JPEG, Very High - Quality: 80, Optimized).
5. I uploaded it to Flickr.com.

I think it is better if you put some parameters or qualifications for a digital photograph to be judged as sharp. As the others have said, there are many things that affect the sharpness of a digital photograph.

In this way, people will more likely participate in your little experiment. :D

besides....

how can the sharpness of a photograph be judged online, with all our screens varying in calibration and resolution? what might be sharp on my monitor might be soft in another's.

i was at OWG last night looking at the larrios prints and i was looking for sharpening artifacts (which i didnt find, wow!). it was then that i truly appreciated all the processes involved in coming out with those kinds of photographs (from the shot to the post processing to the archival printing, meticulous every step of the way).

gabriel, i think you have a nice idea. but this inquiry will be inconclusive at best. everyone aspires for sharp photographs. heck, even the national geographic photographers DVD i watched at OWG last night mentioned that sharpness is one of the things that editors look for in a photograph.

but with all the tools present, to novice and pros alike, havent we reached a point where critical sharpness good enough for any application is achievable from any of the cameras or lenses at our disposal?

if you really want to nitpick resolution prowess...then i guess you better do so with prints. thats the way to go.

I'm sorry but even the pros will agree that sharpening is part of the digital workflow. So for purists who ask me why I bother to sharpen my images, my only answer is this: Because I can.

In my workflow, sharpening is always one step before final output. Prior to sending the images to Fuji or Koday, I apply some degree of sharpening. Prior to uploading them to the web, they go through certain degrees of sharpening as well. Even before printing them on my inkjet printer at home, they go through some semblance of sharpening. Why do they go through sharpening in my workflow? Because they can.

So in response to Gabriel's sentiment of "sharing with us your technique," my suggestions are:

1) Find the "sweet spot" of your lens. Usually this is starts at F/5.6 or F/8 and ends at F/11 or F/16
2) Shoot with the best amount of light. You have more chances of getting sharp images when there is adequate light.
3) Expensive lenses have been proven, so to speak, to take "sharper" pictures than cheaper lenses. But then again, this is dependent on one's definition of "sharp."

However I do agree with Jay that posting such online will give you inconclusive results.

Looking at it in an abstract way, the very sharp photos are those images that pierced your hearts and senses, even if it's blurry images. Images on the monitor will just fool our eyes, just like the contrast tool at PS, you may find your images going sharper whenever you slide the density, but it only fool our eyes. Only the printed images will reveal a true sharp images, not on monitors:)

Okidokee likewise... Talk is cheap... So here you go...

D2X, 105 Micro Nikkor VR at F/4.0, ISO 100, 1/500 sec. No sharpenning. Slight wind blow...

First image... Full view but scaled down to conserve bandwidth...
http://www.earlgonzalez.com/Spider1.jpg


Second, 100 percent crop of a 4288 x 2848 image fitted to 550 x 365 pixel dimension,
compressed and saved for the web.
http://www.earlgonzalez.com/Spider2.jpg

Considering the factors that this spider is almost just as big as a letter in your keyboard plus the
slight breeze erratically coming... To me this is sharp, if not one of the sharpest image I took.

Best Regards. :) Now let's see yours...

Earl,

It can be done. The question is whether we are comparing apples to apples in the end. :)

But what's the point of the whole exercise? There are MTF graphs available on the net to indicate the len's resolving capabilities if sharpness is all you're looking for. Comparing shots of the same lens taken with poor vs bad technique would immediately introduce variability in the comparison. And there is evidence that there are sample variances - even with L lenses.

For me, this is like doing a half decent job of reinventing the wheel.

Hi Arnell :)

I wonder how do you propose to judge critical sharpness, as the author of this thread implied, infront of your monitor? It's a subjective thing Bro...

As for the MTF curves or graphs you mentioned... These things are not an accurate guage to a lens' critical sharpness or quality for that matter... One has also to take into consideration not only resolving power but also the lens' inherent contrast qualities or attributes, and colour rendition... :D "Good golly miss molly"... to state the truth, most photographers even have trouble interpreting these charts; or neither do most have the time and inclination to do the deciphering... :Grin:

So can anyone share us their sharpest digital photo? Hope we can learn something here.

Ok, I'll join the party...here's my most detailed digital capture ever - conventional SLR lens (not a telescope) + DSLR. Shot through Paranaque's polluted skies from a distance of about 384,000 km....:)

This is processed (conversion to BW, sharpening, levels, contrast, etc.), of course.

Full frame resized to 600x400
http://www.pbase.com/liquidstone/image/66109922.jpg


100% crop
http://www.pbase.com/liquidstone/image/66109709.jpg

Romy

Wow Romy, that's very sharp, even with the cropping. I'm just curious, what's your shutter speed?

Romy,

Pretty amazing!

If I read the thread correctly, I think the OP was looking for sharp images without the use of Unsharp Mask. How about posting one of your wild bird photos that didn't need any Photoshop filters because it was so clean, clear, and naturally sharp.

Nice shot Romy... yeah I almost forgot the moonshot thread... :)

Never mind the question. I found your exif which is 1/80 shutter speed at f18, at 1600mm. Wow, I want a lens like that...

Never mind the question. I found your exif which is 1/80 shutter speed at f18, at 1600mm. Wow, I want a lens like that...

Hi Nap,

This was actually at 2263 mm (800 mm + stacked 2x and 1.4x TCs). Exif didn't report the second TC's effect.

Romy

D2X HSC, 70-200 VR. HDR, tone, curves... CS2
http://www.earlgonzalez.com/Moonshotko.jpg

Not as good as Romy's one but it'll do for me... :)

Romy,

Pretty amazing!

If I read the thread correctly, I think the OP was looking for sharp images without the use of Unsharp Mask. How about posting one of your wild bird photos that didn't need any Photoshop filters because it was so clean, clear, and naturally sharp.

Ok, Nick.... here's one from the 20D + 500 f4 IS + Canon 1.4x TC.

Full frame resized to 600x400 (with USM):
http://www.pbase.com/liquidstone/image/66110865.jpg


And here's the 100% crop straight from the RAW converter (ZoomBrowser EX), the only things done in PS was cropping and saving as jpeg quality 9 (no USM or any enhancement):
http://www.pbase.com/liquidstone/image/66110876.jpg


Romy

Wow Romy... :) Talim...

Romy,

Post more! Excellent! Now that is sharp.

Romy,

Post more! Excellent! Now that is sharp.


Thanks, Earl and Nick.... ok, I'll oblige our editor with a one kilometer test shot...:)

Resized full frame:
http://www.pbase.com/liquidstone/image/52021274.jpg



100% crops:
http://www.pbase.com/liquidstone/image/52021338/original.jpg


Romy

These shots are all very great, nanliit ako. Of course you guys used a tripod in all these shots, right?

That shot was 1 kilometer away? That's simply amazing, Romy. That's one fine glass. Oh, and I just googled sigmonster and found out your name came out on top with your pbase gallery! :)

That shot was 1 kilometer away? That's simply amazing, Romy. That's one fine glass. Oh, and I just googled sigmonster and found out your name came out on top with your pbase gallery! :)


Marco, you're right... a tripod was used in my posted shots.


Lester, it was about a km away - that's the Y-intersection at Subic and I shot from the viewdeck of Jest Camp....:)

Romy

Cool!

Romy? Can you still make out the license plate on the bus in the original image?

Cool!

Romy? Can you still make out the license plate on the bus in the original image?


Sure, Dave...the bus was running (a bit of motion blur at SS of 1/100 sec) and focus was on the stoplight (the plate was a tad outside the DOF), but I can still read the plate number - CVX 325...:)

http://www.pbase.com/liquidstone/image/52045420.jpg

Romy

Sure, Dave...the bus was running (a bit of motion blur at SS of 1/100 sec) and focus was on the stoplight (the plate was a tad outside the DOF), but I can still read the plate number - CVX 325...:)

http://www.pbase.com/liquidstone/image/52045420.jpg

Romy

ha ha ha :D whooooaaaa... COOL! Now you're really tempting me to get some more range! :)

Hi Arnell

I wonder how do you propose to judge critical sharpness, as the author of this thread implied, infront of your monitor? It's a subjective thing Bro...

As for the MTF curves or graphs you mentioned... These things are not an accurate guage to a lens' critical sharpness or quality for that matter... One has also to take into consideration not only resolving power but also the lens' inherent contrast qualities or attributes, and colour rendition... :D "Good golly miss molly"... to state the truth, most photographers even have trouble interpreting these charts; or neither do most have the time and inclination to do the deciphering... :Grin:

That's the point, dude. I have put in time - my time. And the knowledge is not gonna come in free - it's gonna cost you. :Grin:

Simply put, sharpness, in practical terms, refers to the lens's ability to render fine detail distinguishably. Several factors affect this which describe the way the lens draws the image. Factors such as contrast, diffraction, etc.

So how did the engineers developed tests to evaluate the lens's capabilities and compare with those of other lenses. They must have figured that if the lens can render closer lines clearer, then it must be sharper. This is where lines per millimeter (LPM) comes in - similar to the printer's dpi.

Now, there is a lot of literature, especially on the net, that can explain this. This is the part where its gonna cost you - you gotta put in time to read and understand what is said. By doing so will enable you to separate chaff from grain in the future. I believe Reichmann has an article on MTF curves which is pretty accessible. :)

That's the point, dude. I have put in time - my time. And the knowledge is not gonna come in free - it's gonna cost you. :Grin:

Simply put, sharpness, in practical terms, refers to the lens's ability to render fine detail distinguishably. Several factors affect this which describe the way the lens draws the image. Factors such as contrast, diffraction, etc.

So how did the engineers developed tests to evaluate the lens's capabilities and compare with those of other lenses. They must have figured that if the lens can render closer lines clearer, then it must be sharper. This is where lines per millimeter (LPM) comes in - similar to the printer's dpi.

Now, there is a lot of literature, especially on the net, that can explain this. This is the part where its gonna cost you - you gotta put in time to read and understand what is said. By doing so will enable you to separate chaff from grain in the future. I believe Reichmann has an article on MTF curves which is pretty accessible. :)

he he he :D The info here goes to those who need it Arnell, not to me... But thanks for the reminder though... so you can freely charge someone else for this to your hearts content... You crack me up brother... Well, anyway since you mentioned Michael R. You should better read the article again... Here, an excerpt for you:

Footnote
All MTF charts on this page are courtesy of Canon Inc. You can view complete lens specs and MTF charts of all Canon lenses on the Canon US web site (http://www.usa.canon.com/eflenses/lineup/index.html).
Canon's MTF charts are based on theoretical calculations used in the design of the lens, while some other manufacturers use actually measurements. There are valid arguments in favour of both methods. Be aware though that different manufacturers have different measurement procedures, and therefore while comparing MTF charts between lenses in the same line is possible, and is in fact very useful in making a purchasing decision, doing so between different manufacturer's MTF charts isn't.
Why have I used Canon's charts? Simply because to my knowledge they are one of the few companies that regularly publishes such charts for all of their lenses. Leica also makes them available, but only as part of a large German language PDF (http://www.leica-camera.com/imperia/md/content/pdf/msystem/48.pdf). The only other that I know of is Contax / Zeiss (http://www.contax.ch/Site/Links.html#techfiles). Update: Apparently Nikon also publish MTF charts for some of their lenses, but only via their Japanese language web pages. These can be found here (http://www.nikon-image.com/jpn/products/af_nikkor/index.htm).

Like I said, MTF curves are guidelines and should be taken with a grain of salt... You should know better... Try reading Nikon's MTF charts, if you cannot read Japanese, I'll teach you for free... :)

Earl,

Didn't I write separating chaff and grain? Require's judgement dude, including on the things we read. Agree published MTF curves are guidelines and not the bible truth. But didn't we learn in school to be wary of statistics? ;-)

Absent some other readily available system, what do you recommend? The other option is testing the lens (which I have done in the past) to determine the characteristics of the specific objective you have. This is not rocket science but requires a lot of sweat. IMO, very few people are willing to go through this if we assume their reluctance to read the published results an indication.

Gotta see the forest first dude...then the trees. ;-)

Looking forward to reading your ideas on this. May take you up on the Japanese language lessons someday- I'll do my darndest she's pretty and worth our time.:D

Thumb Tacks, this sharp enough :)

D70 reverse mounted 20mm F4 AI at F16

http://www.photosbykengo.com/misc/DSC_1021_sm.jpg

I was looking through some old photos taken with film and for some reason they look much sharper than my digital photos.

I personally feel that my pictures taken digitally look soft compared to film.



Hi Gabriel,

I actually had a similar observation when I was looking at prints of my wedding (c.1998), shot on color neg and printed on color neg paper. They were not done in a custom lab, but through a 1 hour Kodak Express shop, using an analog minilab with an optical exposure system.

The negs showed that the pics were properly exposed, and the prints were very sharp (photog was using a Nikon F90x with some major Nikkor glass). There was no post processing done at all, and the only operator intervention was to correct for the "color failure" - or the color cast inherent to certain color combinations.

I then compared it to a collection of shots from various Digital Cameras I've borrowed, shot & tested (from the early 1.3Mp P&S units to current 8Mp D-SLRS). These shots were then brought to a Digital Minilab -- with only density and color correction changes performed by the operator. And printed the same type of minilab color paper.

And without post processing, the print outs of digital shots were noticeably less defined and "softer" than the film shots. Now, before I get crucified for this "blasphemy" - I believe this to be a function of the analog minilab having better optics to project the image onto the photopaper vs. the laser or LED of the digital lab.

Maybe that's why nyarp yung pics printed from film...

Cheers!

Edwin

Here's a shot of a squirrel I took just about 15 minutes ago. Nothing was done except raw conversion to jpeg
EF 100-400 @ 400mm + EF 1.4XTC handheld

Full frame
http://i.pbase.com/o4/20/655120/1/66129857.buOdS4Om.TEST04.jpg

Cropped
http://i.pbase.com/o4/20/655120/1/66129858.MoMdgjZN.TEST05.jpg

Nice one Reyno! :) This photo got me... So that's what a Squirrel's ding dong looks like 100% cropped! :D

Reyno,

Tack sharp and even handheld! You probably sleep early everynight and stay away from the double espressos and Marlboro lights.

Earl,

Didn't I write separating chaff and grain? Require's judgement dude, including on the things we read. Agree published MTF curves are guidelines and not the bible truth. But didn't we learn in school to be wary of statistics? ;-)

Absent some other readily available system, what do you recommend? The other option is testing the lens (which I have done in the past) to determine the characteristics of the specific objective you have. This is not rocket science but requires a lot of sweat. IMO, very few people are willing to go through this if we assume their reluctance to read the published results an indication.

Gotta see the forest first dude...then the trees. ;-)

Looking forward to reading your ideas on this. May take you up on the Japanese language lessons someday- I'll do my darndest she's pretty and worth our time.:D

:D Okidokee Arnell... Been there... done that as well... Assuming all else is equal, we're just splitting hair filaments here... :D Peace.

sir reyno, what camera you using? assuming a canon, what picture style are you using? did you tweak the settings?

and that picture is wicked sharp....

Thank you gentlemen.
Earl: I noticed that ding-dong too but while I was processing the image my son passed by and asked about that thing sticking out, and I told him that he is a he but then then replied if that's a ding... how come he's got boobs?

Nick: Coffee in any form and Marlboro lights keep's me going everyday. What I do is, I tuck my elbows against my ribcage then stop breathing when I take a shot. It's not confortable in the beginning but once you get used to it it becomes second nature. And of course, the IS helps a lot in these situations.

Christopher: The lens was mounted on a 30D and here's the settings:
ISO: 400 Av:f/8 Tv: 1/400 Exposure Compensation: +1/3
Contrast: 2 Sharpness:4 Color Saturation: 2 Color Tone: 2
I don't use the in-camera picture styles rather, I use the user define parameters and the above are what I normally use including birding and landscapes.

Here's a ready to print copy of the shot. Nothing much done except a bit of sharpening then lightened the shadow area to bring out more detail.
http://i.pbase.com/o4/20/655120/1/66130721.6E4FJaLm.TEST06.jpg

Here's a shot of a squirrel I took just about 15 minutes ago. Nothing was done except raw conversion to jpeg
EF 100-400 @ 400mm + EF 1.4XTC handheld

Cropped
http://i.pbase.com/o4/20/655120/1/66129858.MoMdgjZN.TEST05.jpg


Nick Tuason
Reyno,

...You probably sleep early everynight and stay away from the double espressos and Marlboro lights.


From the looks of it, I would say Mr. Love here has had either a busy night/morning. Lucky guy! :D

When you take a naturally sharp capture as the original, and then just put some simple moves to prep the image for print, you get quality that just literally pops off the page. This is the biggest difference between trying to fix soft images with the unsharp mask.
Excellent example again Reyno.

Now, when I first saw your picture and you said that you had taken it 15 minutes ago, I was asking myself 'How come I have never seen any squirrels like that in the Philippines?" Then Nino Carandang just told me that we had an inquiry for a subscription from Orlando, Florida and it was you!

Thanks for participating in our forum Reyno and for helping make DPP an international community. Coffee and Marlboro Lights on me when you visit. :)

Hey, check my signature... :D

Hey, check my signature... :D

Now that's what I call a witty-wise answer. :Grin:

Hey, check my signature... :D


Rolando,

Funny you mention this because I really was thinking about your signature when I wrote my post! :D