F550 in good light.

PAUL TILL · Apr 18, 2011

dotbalm said:
So these are SOOC.

What happens if you apply your sharpening method?

It doesn't work on F550 images, I think the Jpeg engine is over sharpening to begin with, I first noticed it when I played with one of Kims Raw files and I mentioned it to him.

Apr 5, 2011

"It is probably the worst RAW file I have ever had to work with and it has very little leeway especially the sharpening."

If the Raw files have very little leeway what hope is there for the Jpegs?

I've always said no camera produces good Jpegs SOOC but I think Fuji is trying to reach that goal with their new cameras. The images in my first post look PP'd to me SOOC, a lot of people will like that but I don't.

Paul.

PAUL TILL · Apr 18, 2011

jk641 said:
I personally don't see anything wrong with these images.
The colors could be a little more saturated, but it's not a big issue.

By the way, I've noticed that the original full-size images here are all more than 3MB in size.

In your previous thread, the images were much smaller in size, about 1MB. Did you do something differently?

John

These are original SOOC shots uploaded straight to ImageShack, the night shots were saved at high quality 9 in CS5 (which look no different from the originals) I saw no point in saving crappy pictures at full resolution when they still looked crappy.

Paul.

PAUL TILL · Apr 18, 2011

Wayne N. said:
buckshot said:

Paul has always posted clear crisp images with a variety of cameras, if he is unable to do so with this camera I would think its a bad copy... or faulted

Click to expand...

I agree. My problem is that these also look clear and crisp to me. I'm sure that someone will show me the error of my eyes though.

Wayne, that's the problem they look like I have PP'd them, (all be it badly) but I haven't they are SOOC. The cameras Jpeg engine appears to be doing my PP for me and I like to do my own but it's leaving very little leeway for any adjustment.

Have a play with the originals, that's what they are there for.

Yes I could shoot Raw but SilkyPix is a pain and from playing with HS20 Raw files before I found that even the Raw's don't have the leeway that I'm used to for adjustment.

I think the camera needs some extra settings in the menu to:

1. Turn down the sharpening.
2. Turn down/off the NR.
3. Adjust the contrast.

I PP all my Jpegs in Adobe ACR and a typical F200EXR or S200FS Jpeg needs the blacks raised to 5 or 6 but these start to clip at 1, I found it the same with HS10 Jpegs. I think Fuji has pushed the level of in camera processing to it's max.

Paul.

Wellington100 · Apr 18, 2011

Barry Fitzgerald said:
This is hardly news common problem on smaller sensors and yes the F70 smears away too I just think Fuji ramped it up even more since the F80 and newer models bad move IMO

I wonder if its more complicated than that. The diffraction problem must be huge for this camera which the diffraction calculator shows is significant for a 16mp tiny sensor. Diffraction smears data intended for one pixel into the surrounding pixels which reduces resolution. The effect of diffraction gets significantly worse as aperture reduces.

PAUL TILL · Apr 18, 2011

Wellington100 said:
Barry Fitzgerald said:

This is hardly news common problem on smaller sensors and yes the F70 smears away too I just think Fuji ramped it up even more since the F80 and newer models bad move IMO

Click to expand...

I wonder if its more complicated than that. The diffraction problem must be huge for this camera which the diffraction calculator shows is significant for a 16mp tiny sensor. Diffraction smears data intended for one pixel into the surrounding pixels which reduces resolution. The effect of diffraction gets significantly worse as aperture reduces.

But at what point does it stop and an ND filter take over? There is obviously an ND filter in there as you can't get the star effect with lights like you can with a real iris that opens and closes.

Paul.

Wellington100 · Apr 18, 2011

PAUL TILL said:
Wellington100 said:

Barry Fitzgerald said:

This is hardly news common problem on smaller sensors and yes the F70 smears away too I just think Fuji ramped it up even more since the F80 and newer models bad move IMO

Click to expand...

I wonder if its more complicated than that. The diffraction problem must be huge for this camera which the diffraction calculator shows is significant for a 16mp tiny sensor. Diffraction smears data intended for one pixel into the surrounding pixels which reduces resolution. The effect of diffraction gets significantly worse as aperture reduces

Click to expand...

PAUL TILL said:
But at what point does it stop and an ND filter take over? There is obviously an ND filter in there as you can't get the star effect with lights like you can with a real iris that opens and closes.

Paul.

Thats a good point you make, the F550 always operates at maximum real aperture and the ND filter would not affect diffraction. According to the advanced diffraction calculator, it looks like diffraction limits resolution from F2.7 for a 16mp sensor with a 1.1/2 size sensor. The effect of diffraction increases as the aperture reduces so by zooming in, the aperture reduces and diffraction gets worse, therefore resolution goes down further.

I don't think the EXR pixel binning changes this basic equation because there are still 16m individual pixels being imaged to begin with.

PAUL TILL · Apr 18, 2011

I've just been looking at aig's SOOC shots with his HS20 here:

http://forums.dpreview.com/forums/readflat.asp?forum=1012&thread=38233622

They look quite good at 1200 pixels in size, see below:

But now take a look at a 100% crop of the original at ISO200.

I think this shows how the Jpeg engine in both the F550 and HS20 is over processing the images and he has the NR turned down on the HS20 something you can't do on the 550.

It looks like Fuji has upped the sharpening to give the impression of detail.

However looking at those HS20 images and my F550 images at 100% I think the little F550 is actually doing a better job. Or is that overheating problem on the HS20 degrading the images? He said the temp warning light came on but the camera never shut down..!

Your thoughts on this?

Paul.

Billx08 · Apr 18, 2011

dotbalm said:
. . .

I hate to admit it, but I have to RTFM because I don't know/recall how many real apertures the F550 has. Is it one with two NDs? Is it two with one ND? If at a given FL f8 were achieved using a ND...then is f8 as diffraction-inducing than if it were based on a true aperture based f8 using a smaller radius than what is used in conjunction with a ND? I could look this up, but I'm late for dinner as it is..REALLY late to the point my wife wants to clean up and if I miss the window...guess who "gets" to clean up.

Same here. I don't know if it's one or two ND filters. All the manual says is "uses Neutral Density (ND) filter" after showing that it provides 3 different apertures. So either there's really only one real aperture or two, which would imply two ND filters or only one. This calls for a test.

Ok, test completed. From wide open to the next smaller aperture is a 2 stop change. To go to the last and smallest aperture is an additional 1 stop change. So if you zoom wide open to an f/4.0 aperture, the other two available apertures are f/8 and f/11. Focusing on a close object and comparing a distant object shows that the two smaller f/stops have the same wider DoF, so in the above example, if f/4 is wide open, f/8 and f/11 use the same smaller aperture and f/11 adds a one stop ND filter. It's really easy to see, which surprised me. The difference in DoF between f/4 and f/8 is large, and I couldn't see a difference in DoF between f/8 and f/11. The only difference I noticed was a slight color shift, but that's because the shutter speed wasn't able to be slow enough for the smallest apertures so I had to boost the ISO to get equal exposures.

Wayne N. · Apr 18, 2011

PAUL TILL said:
Wayne N. said:

buckshot said:

Paul has always posted clear crisp images with a variety of cameras, if he is unable to do so with this camera I would think its a bad copy... or faulted

Click to expand...

I agree. My problem is that these also look clear and crisp to me. I'm sure that someone will show me the error of my eyes though.

Click to expand...

Wayne, that's the problem they look like I have PP'd them, (all be it badly) but I haven't they are SOOC. The cameras Jpeg engine appears to be doing my PP for me and I like to do my own but it's leaving very little leeway for any adjustment.

I take it you feel that the F550 over-processes more than the F70?

PAUL TILL said:
Have a play with the originals, that's what they are there for.

Good idea.

PAUL TILL said:
Yes I could shoot Raw but SilkyPix is a pain and from playing with HS20 Raw files before I found that even the Raw's don't have the leeway that I'm used to for adjustment.

That's a bummer about Sillypix being a pain.

PAUL TILL said:
I think the camera needs some extra settings in the menu to:

1. Turn down the sharpening.
2. Turn down/off the NR.
3. Adjust the contrast.

Having those options would indeed be nice - they helped a lot on my old Oly C770 superzoom. I didn't miss the lack of those adjustments as much on the F70 due to flatter tone curve OOC which I find bodes fairly well for PP (and I don't find that the F70 oversharpens too badly). The at-times-mush-factor still bugs me though when it occurs.

PAUL TILL said:
I PP all my Jpegs in Adobe ACR and a typical F200EXR or S200FS Jpeg needs the blacks raised to 5 or 6 but these start to clip at 1, I found it the same with HS10 Jpegs. I think Fuji has pushed the level of in camera processing to it's max.

Ok. Thanks for the explanations.

Cheers,

Wayne
http://www.pbase.com/wayne_n

Billx08 · Apr 18, 2011

Wellington100 said:
Wellington100 said:

Wellington100 said:

Wellington100 said:

This is hardly news common problem on smaller sensors and yes the F70 smears away too I just think Fuji ramped it up even more since the F80 and newer models bad move IMO

Click to expand...

I wonder if its more complicated than that. The diffraction problem must be huge for this camera which the diffraction calculator shows is significant for a 16mp tiny sensor. Diffraction smears data intended for one pixel into the surrounding pixels which reduces resolution. The effect of diffraction gets significantly worse as aperture reduces

Click to expand...

Click to expand...

Wellington100 said:

But at what point does it stop and an ND filter take over? There is obviously an ND filter in there as you can't get the star effect with lights like you can with a real iris that opens and closes.

Click to expand...

Thats a good point you make, the F550 always operates at maximum real aperture and the ND filter would not affect diffraction. According to the advanced diffraction calculator, it looks like diffraction limits resolution from F2.7 for a 16mp sensor with a 1.1/2 size sensor. The effect of diffraction increases as the aperture reduces so by zooming in, the aperture reduces and diffraction gets worse, therefore resolution goes down further.

I just posted the results my F550 test shots (see "F550 aperture" reply near the bottom of the thread). When you go from wide open to the next smaller aperture, the actual aperture decreases in size by two full stops. When you go to the next and smallest aperture, the real aperture doesn't change. The F550 just adds a one stop ND filter. The examples I gave are f/4.0 to f/8.0 to f/11. Zoom to a different focal length and the three apertures change, but they're still separated by the same 2 stops then 1 stop difference.

PAUL TILL · Apr 18, 2011

Billx08 said:
Ok, test completed. From wide open to the next smaller aperture is a 2 stop change. To go to the last and smallest aperture is an additional 1 stop change. So if you zoom wide open to an f/4.0 aperture, the other two available apertures are f/8 and f/11. Focusing on a close object and comparing a distant object shows that the two smaller f/stops have the same wider DoF, so in the above example, if f/4 is wide open, f/8 and f/11 use the same smaller aperture and f/11 adds a one stop ND filter. It's really easy to see, which surprised me. The difference in DoF between f/4 and f/8 is large, and I couldn't see a difference in DoF between f/8 and f/11. The only difference I noticed was a slight color shift, but that's because the shutter speed wasn't able to be slow enough for the smallest apertures so I had to boost the ISO to get equal exposures.

That's not how my F550 looks Bill, your test images might have helped but as you have not provided any I've had to do it myself.

You said "The difference in DoF between f/4 and f/8 is large , and I couldn't see a difference in DoF between f/8 and f/11." I started wide open at f4.4 so I had f9 and f13 as my other two apertures.

As you can see from the images below there is no difference in DOF between the two largest apertures f4.4 and f8 but you quite clearly see that f13 has a deeper DOF.

Either we have totally different cameras or you cocked that one up.

Looking at the shutter speeds in the pictures I come to the conclusion that the difference between f9 and f13 is the iris closing as they both have the same shutter speed yet the f13 picture has a greater DOF.

If it was an ND filter dropping in the DOF would be the same but the shutter speed would be different.

Paul.

PAUL TILL · Apr 18, 2011

So in conclusion to the above it looks like the F550 has 2 real apertures, wide open and closed down to get the middle f stop it's dropping in an ND filter.

The trouble with that conclusion is I don't see any difference in the exposure or DOF between f4.4 and f9.

Is my ND filter stuck and not coming into play?

Paul.

PAUL TILL · Apr 18, 2011

Scrap this bit Is my ND filter stuck and not coming into play ? I had the last 2 images shutter speed locked in my head and forgot it had changed in the first two.

Paul.

PAUL TILL · Apr 18, 2011

Bill... you've gone all quiet yet your still posting in other threads!

Paul.

dengx · Apr 18, 2011

PAUL TILL said:
I've always said no camera produces good Jpegs SOOC

Hello Paul,

S5 and S3 produced very good JPEGs SOOC

For sure it could be touched even more, white-balanced, sharpened etc but it still looked very good from the beginning.

Regards

Billx08 · Apr 19, 2011

PAUL TILL said:
So in conclusion to the above it looks like the F550 has 2 real apertures, wide open and closed down to get the middle f stop it's dropping in an ND filter.

Well, you're correct. I see that you ran into the same problem that I initially had where the longest shutter speed wasn't long enough to get equal exposures, so your f/13 photo is darker than the others, but that didn't affect the results. I re-shot the photos to use subjects that were clearer and got the same result that I did the first time, but the first time I must have been up too long and mis-stated the conclusion because the smallest aperture photos in both my initial set and the latest set were actually the only ones that had the greatest DoF and sharpest out-of-focus background object. So thanks for posting your photos and making me take a closer look. FWIW, here are the photos I shot. They're all SOOC, not cropped, not resized or sharpened, just uploaded directly from the SDHC card. They're 4MB images but I don't see an "Original" option. Maybe that will show up after this reply is posted. The F550 focused on the lens hood and the booklet in the background (perched on the red box) is the out-of-focus object.

Billx08 · Apr 19, 2011

PAUL TILL said:
Bill... you've gone all quiet yet your still posting in other threads!

I was busy re-shooting my test photos and I just posted the corrected results. Yep, I screwed it up. I didn't start the tests until after I saw the car/truck photos that you posted.

PAUL TILL · Apr 19, 2011

Billx08 said:
PAUL TILL said:

Bill... you've gone all quiet yet your still posting in other threads!

Click to expand...

I was busy re-shooting my test photos and I just posted the corrected results. Yep, I screwed it up. I didn't start the tests until after I saw the car/truck photos that you posted.

That's OK, the pictures are very much appreciated. You made it harder on yourself for this particular test by raising the ISO's the mush can be mistaken for blur.

Paul.

P.S. Thanks for keeping me of the bottom.

Paul De Bra · Apr 19, 2011

The F500/F550 have a mega-zoom lens which is bound to be less sharp than a more modest zoom lens like on the F200. On top of that you should never stop down the lens because wide open the sharpness is already diffraction-limited on a small sensor.

Yes I do see some mush, but mostly on images taken at f8 or so. Not so much on the wide-open images. And I see a hint of a problem with the upper right corner, but not conclusive from these images.

The mega-zoom lens requires more in-camera sharpening, but also means that sooner than with a sharp lens a part of an image will be soft beyond recognition so mush is unavoidable. The RAW images should reveal how soft the lens really is.

I have seen the same mush problems with images posted from the F70, F80, F300... it's all due to these mega-zoom lenses being soft.

This also means that there can be quite a bit of copy variation. If you stumble on the occasional pretty sharp copy of the lens you might get surprisingly good results. But on average I would not expect anything more than what you are getting (sad as this is).
--
Slowly learning to use the 450D, the Canon G6 and the Fuji F200.
Public pictures at http://debra.zenfolio.com/ .

Adam-T · Apr 19, 2011

Good shots Paul - the mush is there at full size but in all fairness, it's no more than the F70 is at 5Mp - it won't compete with the F200's 1:1.6" CCD but I'd not expect it to either anymore than I expected the F200 to match the detail from the older F100FD (Proper 12Mp SCCD-HR).. The F200 was a nice balance of detail and EXR capabilty, the problem is that a pocket megazoom made from it would have a honking great lens on, it'd look like a NEX with the 18-200 on at full zoom ..

It'd be interesting to see how it compares to the older CCD based pocket megazooms as well as the new Canon SX220HS

--
A Problem is only the pessimistic way of looking at a challenge

F550 in good light.

Veteran Member

Veteran Member

Veteran Member

Forum Pro

Veteran Member

Forum Pro

Veteran Member

Forum Pro

Forum Pro

Forum Pro

Veteran Member

Veteran Member

Veteran Member

Veteran Member

Senior Member

Forum Pro

Forum Pro

Veteran Member

Forum Pro

Forum Pro

Keyboard shortcuts