S9000 Image quality

[Mattias77]

You should always use the highest quality setting, unless you have no other option, like out of memory. That's what I think at least. Maybe you won't be able to see any difference in a high compressed image with a low compressed image. But if you look closely you will see the difference. For example, if you look closely at that picture (679 kB) you will see some "squares" that are a result of compression. By using a lower compression setting these squares will not be so easy to spot, which means better quality. But then again, if you don't see the difference then you can of course use whatever setting you like. But in my case I will always try to save my images as a raw file (uncompressed) (haven't bought a digital camera yet, but I have my eyes on the S9500 (S9000)).

Sorry for my English, but it's not my main language.

Regards, Mattias

Does that mean that aving the images in-camera at the lowest
compression (largest file size) is a waste of memory?

 

[Gearóid Ó Laoi Garry Lee]

Mattias77 wrote:

Your English is superb, so good that I'm going to correct it, as one only does with people who are almost perfect.

You should always use the highest quality setting, unless you have
no other option, like out of memory.

like you are out of memory.

That's what I think at least.

Maybe you won't be able to see any difference in a high compressed
image

high-compression image or highly compressed image. Low-compression image.

with a low compressed image. But if you look closely you will

see the difference. For example, if you look closely at that
picture (679 kB) you will see some "squares" that are a result of
compression. By using a lower compression setting these squares
will not be so easy to spot, which means better quality. But then
again, if you don't see the difference then you can of course use
whatever setting you like. But in my case I will always try to save
my images as a raw file (uncompressed) (haven't bought a digital
camera yet, but I have my eyes on the S9500 (S9000)).

Sorry for my English, but it's not my main language.

Well done sir.
Sehr gut. Ich spreche kein Deutch.

 

[jbabbage]

I thought I would share my latest update. I made the testing url available to the camera dealership. They too agreed that my results should be better than I have gotten, and I will be bringing the camera back in tomorrow morning for an exchange. I'll do the same test again with this camera and see if the results are any different. I will keep you all informed.

--
Jim Babbage

 

[Fleming]

Thanks for keeping us informed. Hope you'll be happier with the new one.

 

[jbabbage]

Me too!
--
Jim Babbage

 

[Martin_Kay]

Well it seems that most raw images are going to need a degree of sharpening in Photoshop. All digital capture is subject to sharpening somewhere. I personally think its a tall order to expect perfect images all then time from an 'automatic' exposure system, there being plenty of occaisons when the system trips up. Most pro photographers shooting E6 always bracket even if they are using a good spot meter or polaroids.

From what I've seen from the S9000, it compares favourably with other digital cameras of a like nature, the advantage being the 9 megapixel capture.

 

[83bj60]

Only post-processing was simple sharpening with ACDsee.

If the detail is disappearing, how do you explain the fanstastic
detail in the fine-print of the bottles??

It's in your minds, lads.

Well I have not seen this particular picture but from what I have seen in many other samples, the s9000 processing system is indeed very good at getting that sort of stuff very sharply resolved. What I don't like, however, is that low contrast detail (such as what can be found in landscapes where subtle variations can be found on very small scale detail) appears to me as if it was treated as noise by the camera noise suppression engine with consequential loss of fine detail, resulting in a something that appears to me to have been "painted" over, giving it an artificial, overly processed appearance.

It IS difficult to quantify this phenomenon and I agree with you, for lack of a better explanation, that it is indeed really just in our minds ;¬).

Just my 2 cents...

 

[Dash29]

Can someone let me know if i'm doing somthing wrong here... ive just looked at the jar pictures, ive then clicked on them to enlarge it..now is there anything else i should do before viewing them????

 

[jbabbage]

Never expected "perfect." Just better than wht I was getting.
--
Jim Babbage

 

[Dash29]

Jim is the text blurred to you on those jars?

 

[jbabbage]

Frankly Dash, I have not studied them too closely
--
Jim Babbage

 

[Dash29]

well ive downloaded the 100iso file and looked at it, the text is totally rubbish, am i doing something wrong?

 

[jbabbage]

Ok just took another look. The text looks WAY better than what I was suffering from with my camera. There's a little softness, but nothing unacceptable or unexpected for a prosumer camera, IMO.

400 seems OK - more noise obviously, but still better than what I was gettig at 80 or 100 ISO

1600 ISO is pretty blech but hey, it's 1600 and I doubt I'd be shooting things with text at that speed anyway.

I am hoping that I get better results when I exchange cameras tomorrow. I'll let you all know.

--
Jim Babbage

 

[Dash29]

well i dont know how to make the file that good i cant read the text on the green part of the label and the barcode is blurred too

 

[Fleming]

Jim, Someone in another thread noticed that some 9000 images show a version 1.01 and some a version 1.02 in the Exif data. You might want to see which yours is and get the other version. Just an idea.

 

[jbabbage]

Thanks for that. I've no idea how to get that updated - Fuji doens't have anything on their site for the S9000, other than marketing material.
--
Jim Babbage

 

[Gearóid Ó Laoi Garry Lee]

I have found that the lens in the S9500 is very sharp at most focal lengths, except at extreme edges where there is a bit of a fall-off, especially at wideangle setting, not surprisingly. You would only notice this if you went looking for it.

 

[jbabbage]

Thank you. As you could see from the tests I posted, it's moe than the edges that are soft, which is why I am hoping the exchange I make will solve my problem.
--
Jim Babbage

 

[jbabbage]

Good news, I think! I exchanged my S9000 for a new one and preliminary tests - handheld - are better than what I was getting with the original camera on a tripod. I'll do more structured testing tonight as right now, I have to pack up for a location shoot.

I am - at this point - very relieved!

I will keep everyone posted.
--
Jim Babbage

 

[Dash29]

I sent fuji 3 test shots they said if i was being hyper critical although they were a little soft, also they said that the ver 1.1/2 isnt in the Efif data.