Shots from new camera, should they be this bad?

Lakes_Puma

Suspended / Banned
Messages
47
Name
Richard
Edit My Images
No
Just purchased an Olympus TG-4 and took some shots with it this weekend and I'm shocked at the quality.

Even at fairly low ISO the images are oil painting like at 100%

P6130047.JPG


Am I expecting too much
 
I cant really see at this size, you might be better posting a 100% crop?
 
It's a compact, albeit an expensive one, with a tiny sensor. That looks like a JPEG straight out of the camera, with excessive sharpening applied (quite common). That's how a lot of people like them. Basically what sharpening does is add contrast to edges, but when it's overdone it pretty much draws an outline around small details and it looks artificial and naff.
 
Try using RAW if it does it.
If not there may be a 'natural' profile setting - or make a custom one with sharpness contrast and saturation turned down one or two clicks
 
Definitely fuzzy. 9.00 yesterday morning, facing SE :)

I'm surprised at the apparent pic quality from a 16mp camera.
My D200, at 10mp, looks better than that at 100%.
On the other hand, you are using f2, ISO400 at 1/500. Coupled with a small sensor and a 4.5mm lens- there is a limit.
Camera does take pics in raw, though I fail to see how that can help.
 
Thank you for the suggestions, I'll do some more experimenting with the settings, here is the above photo cropped

P6130047crop.jpg
 
Tbh, you have unrealistic expectations. There are very few compact cameras whose files will stand up to that much cropping and all of them will have larger sensors.
 
The sensor in that camera is known as 1/2.3" though that is very deceiving - it's actually only 6.17 x 4.55mm, which gives it roughly 1/30th the image area of a full-frame DSLR. That's really where the problem lies.

Edit: it might have 16mp, but that's really got very little to do with it. Most of them will be 'binned' anyway.
 
Last edited:
if you want a 100% viewable experience then u need some foveon in your life

if you do stick with bayer sensors you could see if dxo supports your camera, they do good corrections
 
if you want a 100% viewable experience then u need some foveon in your life

if you do stick with bayer sensors you could see if dxo supports your camera, they do good corrections

Dunno about needing a forveon.

Just out of interest I've just had a quick look at shots from my Canon Ixus 82 IS and Panasonic LX5 and both stand up to viewing at 100% better than the OP's shots. Deffo something 'orrible going on in those but if it's settings or limitations of the camera I just don't know.

Ixus 82 IS.


LX5


This one is a 100% crop from a Medion 85173 which was so cheap and crappy it can hardly be called a camera at all.


I don't know what the answer to the OP's problem is but the old to ancient compacts I have look better to me.

PS. These are via Photobucket and look a bit to a lot better on my screen than here.
 
Last edited:
foveon gives double the resolving detail, or 80% most of the time
 
Pretty much every Panasonic LUMIX compact file I've seen (which is a lot, as both my folks and my in-laws have them) look far more acceptable than that at 100%.
 
Dunno about needing a forveon.

Just out of interest I've just had a quick look at shots from my Canon Ixus 82 IS and Panasonic LX5 and both stand up to viewing at 100% better than the OP's shots. Deffo something 'orrible going on in those but if it's settings or limitations of the camera I just don't know.

Ixus 82 IS.


LX5


This one is a 100% crop from a Medion 85173 which was so cheap and crappy it can hardly be called a camera at all.


I don't know what the answer to the OP's problem is but the old to ancient compacts I have look better to me.

PS. These are via Photobucket and look a bit to a lot better on my screen than here.
And there lies the problem, imo, and very neatly demonstrated. More pixels really doesn't make for a better quality image. Then there is the small size with more pixels and any shake will be much less forgiving.

I'd start with the in camera settings and switch off sharpening all together. Then watch technique and perhaps use a tripod.

Hopefully that will improve it. Oh and finally stop looking at it at 1:1. ;)
 
MP count has nothing to do with image quality.....................!!!
 
foveon gives double the resolving detail, or 80% most of the time
This isnt about resolution or resolving detail, the OP's camera has plenty of resolution, on paper, and is clearly being out resolved by cameras with, on paper, less resolution.

Stop it with the foveon fanboy rubbish :) The OP never said he wants to look at all his images at 100%. The issue is the overall IQ whuich should be a lot better then the posted shot.
 
Last edited:
Back again to the OP, I find these 'tough' cameras (such as this and the Canon D20) although look good on paper, seem to suffer on the IQ side. I wonder how much of it is down to the waterproof filter / plastic covering the lens aperture and housing?
 
Thanks for the comments and advise everyone, I'm not expecting quality similar to my Lumix G6, but was hoping for better quality than my 6 year old Canon IXUS 860IS, which this camera is to replace.

I know the tough cameras compromise on IQ but all the reviews led me to believe the TG-4 was one of the better/best ones in class.

Looking at some sample shots on the web don't show the same kind of quality I am seeing e.g.

http://www.imaging-resource.com/PRODS/olympus-tg-4/TG4hSLI0400.HTM

I'm just trying to establish whether I have purchased a 'duffer'?
 
Last edited:
Have you tried a full reset and taken test shots?

And don't think you need to go down the lines of a tripod, you should only need a tripod for really low light stuff with a camera such as this.
 
Last edited:
Hi Richard

Have not read all of the above and I do not have your camera but your image looks to me like you have a severe noise reduction algorithm or such thing going on. Can you turn NR off at all to see if this is being applied in-body ?

The 100% image looks a bit off even for a tiny sensor camera !
 
The sensor in that camera is known as 1/2.3" though that is very deceiving - it's actually only 6.17 x 4.55mm, which gives it roughly 1/30th the image area of a full-frame DSLR. That's really where the problem lies.

Edit: it might have 16mp, but that's really got very little to do with it. Most of them will be 'binned' anyway.

How did they ever come up with that fractional decimal (?) It makes comparing sensor sizes really hard.

This looks pretty bad though.

Here's a 1:1 image from an iPhone:

View attachment 39631
 
Do you have it set to "vibrant" or any other mode that will up the saturation?
 
I think it came from tv tubes or some other unrelated thing, and yep its stupid..
 
I recently bought a Panasonic TZ Compact. It has a 16m pixel 1/2.3 sensor, and the images are absolute crap. They are worse than yours.

I recently found an old Canon Ixus 82IS Compact, in the loft that belonged to the missus. It has 8m pixels, but I think it may be 1/1.7 sensor. As the images from the very old Ixus are so much better, than my newer Panasonic.

Poor images are simply down to a poor sensor.
 
Didn't Panasonic go down in the MP count from the tz60 to the tz70. Unique these days, but I think it makes perfect sense for these kind of devices.
 
Oh I forgot to mention, I also have an even older Canon G2 that only has 4m pixels. The images from that are even better. No longer use it, as it's simply too old clumsy and bulky. Plus I hate Compact Flash. So again simply down to a reasonable sized sensor.
 
I recall making the change from my D70 to the D7000. The step was huge and it showed my terrible technique. I had to shoot so much faster to compensate for my shake.
 
Didn't Panasonic go down in the MP count from the tz60 to the tz70. Unique these days, but I think it makes perfect sense for these kind of devices.

Yes they did, no idea why they cram so many pixels into a diddy little sensor. Oh I remember now, to fool people into thinking more pixels is better !
 
Didn't Panasonic go down in the MP count from the tz60 to the tz70. Unique these days, but I think it makes perfect sense for these kind of devices.
Canon dropped from the 14.7Mp of the G10 down to 10Mp in the G11 and G12...a welcome step IMO.
 
Thanks for the comments and advise everyone, I'm not expecting quality similar to my Lumix G6, but was hoping for better quality than my 6 year old Canon IXUS 860IS, which this camera is to replace.

I know the tough cameras compromise on IQ but all the reviews led me to believe the TG-4 was one of the better/best ones in class.

Looking at some sample shots on the web don't show the same kind of quality I am seeing e.g.

http://www.imaging-resource.com/PRODS/olympus-tg-4/TG4hSLI0400.HTM

I'm just trying to establish whether I have purchased a 'duffer'?

Even the test shot at full size looks bad, perhaps you are expecting too much.
 
The tree looks in better focus than the rider! Did you have it on single focus or continuos focus?
 
Dunno about needing a forveon.

Just out of interest I've just had a quick look at shots from my Canon Ixus 82 IS and Panasonic LX5 and both stand up to viewing at 100% better than the OP's shots. Deffo something 'orrible going on in those but if it's settings or limitations of the camera I just don't know.





I don't know what the answer to the OP's problem is but the old to ancient compacts I have look better to me.

.

Ixus 82 IS and LX5 probably have the slightly bigger sensor 1/1.7 or whatever size, as opposed to the tiny tiny 1/2.5 sensor.
 
For that price I would have bought a Sony Cybershot RX100 it has a 1 inch sensor, plus it's cheaper than the tiny sensor Olympus :thinking: Can you not change it? Get a better spec camera, and get some money back.
 
I've done some more experimenting and set in manual mode the results are much better

P6150063Full.jpg


100% crop

P6150063.jpg
 
Not bad. So you switched everything off and used a tripod?
 
<snip>

I'm just trying to establish whether I have purchased a 'duffer'?

Unlikely IMHO, it's just a very small sensor with some aggressive post processing algorithms applied with the intention of making things look better. Or not, in this case.

With a little PP skill and care, you should be able to moderate the result more to your liking, but you obviously can't make the sensor any bigger, which is the source of the problem, and what you can do in PP may also be limited.

If most of the problem is caused by adjustments to the JPEG output, you can probably do quite a lot by either adjusting the JPEG parameters in-camera, or in PP. But it's quite possible that some of them may also be applied to the Raw data file at sensor level (pre analogue-to-digital conversion) and in that case you're stuck with it. It's a commonly held misapprehension that the Raw file is the untouched data straight off the sensor, but that has never been strictly true and increasingly manufacturers are adding more tweaks to the Raws that are permanent, eg lens aberrations corrections. The files are still Raw in the technical sense, ie the data is fluid, but far from unaltered.

Sensor sizes for compacts, and there are many of them, are (as mentioned above) derived from a historical measure of TV camera tubes. But they actually refer to the external diameter of the tube rather than the sensor inside. It's a nonsense, but the marketing departments like it ;)
 
Last edited:
Back
Top