help please from rx100 iii users

[george]

Hi

 

Maybe its me doing something stupidly wrong, maybe I have the wrong settings incamera, maybe I have been 'away' too long. I am hoping its the second of the two! 
I am finding with my RX100 iii , which is a new camera to me, that I just cannot get acceptable landscape images, maybe only 5% that I am happy to upload. I have tried every different mode available on the camera, I don't think any of the images require the ISO to go above 125. I seem to be ending up with images with a horrendous amount of luminence noise and struggling to get a good clearly focused image with many too, unless I manual focus on something close in the foreground, but then the background is shocking. The noise is the main issue, as obviously once I try to correct it in Lightroom the image becomes soft. I try to ensure that I do not go above f5 /f5.6 as I have read with this particular camera that is the best fstop for landscape. I live by the beach, the weather is fabulous right now and really would like to get a good bank of images to work on once the weather decides summer should really be over.  Many of the images I am struggling with contain almost entirely sky, sea and beach, and I am having to bin them, the luminence noise is shocking. I have just read that you should have the NR turned off in camera, is that what most of you do? What settings are you using to produce good landscapes? I am seriously considering getting my 5D back out of storage just to be able to get myself a stock to work on. Hoping some of you will be willing to share your settings. Just to confirm, this problem is really only when I am taking landscapes, anything remotely close is fine. Many thanks, and feel free to be a blunt as you like  Jenny

[wiskerke]

You do shoot in RAW?

Is it a new camera or a used one?

Could you do a simple focus test?

Do you have a file available somewhere?

 

wim

[george]

Hi Wim

yes always shot in Raw then convert via dng converter as I use Lightroom 4 which doesn't take the raw files. The camera was purchased new by me a couple of months ago. I played around with some product work indoors the other day and focus was fine for that and don't recall seeing luminence noise (although only looked at them quickly as was testing lighting). I will have a look at the instructions for uploading via Dropbox that was out here the other day and will out one on. Thanks

 

[george]

Can I upload raw to dropbox? Or does it need to be the dng?

 

[spacecadet]

19 minutes ago, george said:

Can I upload raw to dropbox? Or does it need to be the dng?

 

Just export a jpeg at 90.

[george]

https://www.dropbox.com/s/4s7bmu3mnfba0hm/peniche Nov 2017 16th 070815.jpg?dl=0

https://www.dropbox.com/s/51mcy0o45p2w1zy/peniche Nov 2017 16th 069814.jpg?dl=0

https://www.dropbox.com/s/cp6rhz7zt52gmjv/sunday 19 nov 17 003824.jpg?dl=0

https://www.dropbox.com/s/ji4n9y6rs8e8isl/sunday 19 nov 17 007828.jpg?dl=0

 

please feel complete free to give brutal opinions. These are none edited obviously

 

hope,I have done these links correctly!

 

 

[John Mitchell]

I don't have an RX100. However, I find that DxO OpticsPro does a good job of fixing "sky noise" -- without blurring the image -- when I use the "prime" selection for NR. I believe that DxO will handle RAW files from your RX100 iii. It might be worth downloading a trial version and giving it a try.

[george]

9 minutes ago, John Mitchell said:

I don't have an RX100. However, I find that DxO OpticsPro does a good job of fixing "sky noise" -- without blurring the image -- when I use the "prime" selection for NR. I believe that DxO will handle RAW files from your RX100 iii. It might be worth downloading a trial version and giving it a try.

Thanks John, I will have a look at that

 

[John Mitchell]

1 hour ago, george said:

Thanks John, I will have a look at that

 

 

The DxO lens modules (you have to download the one for your camera) also do a good job of correcting lens weaknesses, at least that is what I find with the Sony lenses that I use. That said, Alamy QC seems OK with a bit of sky noise. However, the images that you uploaded do have a lot more noise than I've seen in the ones from my Sony NEX-6. The somewhat harsh lighting probably doesn't help matters.

 

Here's a download link for the trial version, now called PhotoLab by the looks of it,

[Betty LaRue]

I guess I’m blind. I don’t see the noise.  But if you have PS, it is blindingly easy to use the selection tool on the sky and go to filters>noise>reduce noise at whatever % you prefer. I usually do 10.

It can be done with the adjustment brush in LR but I’ve done it so long in PS, that’s the fastest for me.

 

if you don’t mind my saying, your horizons are crooked.  Easily fixed in LR or PS. But you will lose some image. Best to get it right while taking the shot. I naturally tilt the same way you do, so I have to pay close attention or I look like I’m shooting while inebriated! 

Betty

[george]

2 hours ago, John Mitchell said:

 

The DxO lens modules (you have to download the one for your camera) also do a good job of correcting lens weaknesses, at least that is what I find with the Sony lenses that I use. That said, Alamy QC seems OK with a bit of sky noise. However, the images that you uploaded do have a lot more noise than I've seen in the ones from my Sony NEX-6. The somewhat harsh lighting probably doesn't help matters.

 

Here's a download link for the trial version, now called PhotoLab by the looks of it,

Thanks again John, will have a good look at that this week. I did wonder if the lighting may be part of the cause of the luminence noise, I will have to get out a bit earlier in the day. Cheers

 

[george]

33 minutes ago, Betty LaRue said:

I guess I’m blind. I don’t see the noise.  But if you have PS, it is blindingly easy to use the selection tool on the sky and go to filters>noise>reduce noise at whatever % you prefer. I usually do 10.

It can be done with the adjustment brush in LR but I’ve done it so long in PS, that’s the fastest for me.

 

if you don’t mind my saying, your horizons are crooked.  Easily fixed in LR or PS. But you will lose some image. Best to get it right while taking the shot. I naturally tilt the same way you do, so I have to pay close attention or I look like I’m shooting while inebriated! 

Betty

Thanks for taking a look Betty. Its luminence noise as opposed to noise as in from high iso, not sure if you mean that? But i am having to reduce by around 40, I have used the gradient tool on other images before, but just feel it should not be necessary all the time . I think as John has said it may be the harsh light from the time of day possibly? Ha yes, horizons, hehe, fix those when I edit. I did actually wonder if my focus problems as sometimes due to a heavy handbag over that left shoulder (!) but went with a backpack deliberately today to see, but doesn't seem to have made much difference  thanks for looking, appreciate it.

Jenny

[Betty LaRue]

I was looking on my iPad, so I didn’t get a large enough picture to see the luminance noise. Don’t feel bad about the tilting, tilt is my middle name. 

if you have a setting on your camera that gives you a line that turns green when your image is straight, turn it on. I use the heck out of mine.

Betty

[george]

1 minute ago, Betty LaRue said:

I was looking on my iPad, so I didn’t get a large enough picture to see the luminance noise. Don’t feel bad about the tilting, tilt is my middle name. 

if you have a setting on your camera that gives you a line that turns green when your image is straight, turn it on. I use the heck out of mine.

Betty

I will have to check out if its on there thanks. 

[wiskerke]

I don't really see luminance noise either. Some color noise maybe. Also some blockiness in the sky. Could be just jpg-conversion.

Most of the strange colors come from flare. I have the Mk1 and 2, so I cannot comment on the mk3 lens, but mine is not prone to flare. However when it's visible it shows the same shatter colors you have in Peniche 069814.

I have lens shades, but rarely take or mount them nowadays, because of my fear that it will either tear the glued on bayonet off, or it may hurt the focusing mechanism. (When it retracts the lens all of a sudden, while my finger is still in the way.)

Generally I set my color noise slider to 50 and leave the color detail slider at 50 too for the RX100's.

 

I would set up a test to see if the focus is correct. And if it's not motion blur.

These are very light cameras and it's really easy to shift them while pressing the shutter. Your "crooked" horizons point in that direction also.

Firstly I would turn on the Grid Lines in the Menu. I have it set to Rule of Thirds. This way, if you have something lined up when you shoot, but on review it's not lined any more, you'll know for sure that you have moved a lot. I know there's a level as well, but this is about lining things up; not about leveling.

 

Now for the test.

Print 3 Siemens stars on a A4 or Letter size. This one or this one will do for this test.

Either use a stiff paper, or tape them to a piece of cardboard.

Outside.

Place your camera on a tripod or something solid, like a table or windowsill.

Put one target at 5 meter; 1 at 10 and one at 15meter.

Set the camera on one of the Auto settings and ISO 100 and self timer.

Take 4 images 1 at the widest; 1 at 35 eq; 1 at 50 eq and 1 at the longest.

I would pick a place with plenty of detailed structure at all 3 distances plus at infinity. Grass is fine; brick; branches will all work. If the structure fills the frame at all focal lengths even better: that will tell you if the image quality is even. This is something people have complained about re these cameras.

Maybe repeat this series handheld without the self timer.

The stars make them very easy to assess. If you have been able to watch the screen when the camera was focusing, you have seen where the little squares fell. Usually at the closest area with high contrast. In this case the first star. Remember this when you're in the field and there's a subject with strong contrast in the foreground while this is not the main subject of your image. (There are of course multiple settings for AF plus there's always MF. )

 

If you have the camera on tripod, set it to A=Aperture and ISO 100.

Now do the same test for all apertures at all focal lengths.

 

The first test will tell you how sharp the image really is and if there's any difference between the handheld and the supported images. Usually (but not always) it's sufficient to compare the jpg file sizes: highest is sharpest.

 

The second test which cycles through all apertures will tell you for each focal length at which aperture the image gets sharp side to side and at which aperture diffraction sets in.

You could then do a third set with the optimal settings and cycle through a bunch of ISO settings to see which ones are still acceptable for you, with your level of post processing. See if you can beat the in-camera jpgs.

 

I usually have the camera on one of the two all auto settings, but you will have to know when to shift to aperture priority and a fixed ISO.

It took me a while to get satisfying results, especially using the Rx100's along the big Canons. I now press the shutter mostly by squeezing the camera. That has helped a lot. The tiny sensor is quite forgiving obviously compared to full frame, but there's a limit to it.

 

hope this helps,

wim

[george]

19 minutes ago, wiskerke said:

I don't really see luminance noise either. Some color noise maybe. Also some blockiness in the sky. Could be just jpg-conversion.

Most of the strange colors come from flare. I have the Mk1 and 2, so I cannot comment on the mk3 lens, but mine is not prone to flare. However when it's visible it shows the same shatter colors you have in Peniche 069814.

I have lens shades, but rarely take or mount them nowadays, because of my fear that it will either tear the glued on bayonet off, or it may hurt the focusing mechanism. (When it retracts the lens all of a sudden, while my finger is still in the way.)

Generally I set my color noise slider to 50 and leave the color detail slider at 50 too for the RX100's.

 

I would set up a test to see if the focus is correct. And if it's not motion blur.

These are very light cameras and it's really easy to shift them while pressing the shutter. Your "crooked" horizons point in that direction also.

Firstly I would turn on the Grid Lines in the Menu. I have it set to Rule of Thirds. This way, if you have something lined up when you shoot, but on review it's not lined any more, you'll know for sure that you have moved a lot. I know there's a level as well, but this is about lining things up; not about leveling.

 

Now for the test.

Print 3 Siemens stars on a A4 or Letter size. This one or this one will do for this test.

Either use a stiff paper, or tape them to a piece of cardboard.

Outside.

Place your camera on a tripod or something solid, like a table or windowsill.

Put one target at 5 meter; 1 at 10 and one at 15meter.

Set the camera on one of the Auto settings and ISO 100 and self timer.

Take 4 images 1 at the widest; 1 at 35 eq; 1 at 50 eq and 1 at the longest.

I would pick a place with plenty of detailed structure at all 3 distances plus at infinity. Grass is fine; brick; branches will all work. If the structure fills the frame at all focal lengths even better: that will tell you if the image quality is even. This is something people have complained about re these cameras.

Maybe repeat this series handheld without the self timer.

The stars make them very easy to assess. If you have been able to watch the screen when the camera was focusing, you have seen where the little squares fell. Usually at the closest area with high contrast. In this case the first star. Remember this when you're in the field and there's a subject with strong contrast in the foreground while this is not the main subject of your image. (There are of course multiple settings for AF plus there's always MF. )

 

If you have the camera on tripod, set it to A=Aperture and ISO 100.

Now do the same test for all apertures at all focal lengths.

 

The first test will tell you how sharp the image really is and if there's any difference between the handheld and the supported images. Usually (but not always) it's sufficient to compare the jpg file sizes: highest is sharpest.

 

The second test which cycles through all apertures will tell you for each focal length at which aperture the image gets sharp side to side and at which aperture diffraction sets in.

You could then do a third set with the optimal settings and cycle through a bunch of ISO settings to see which ones are still acceptable for you, with your level of post processing. See if you can beat the in-camera jpgs.

 

I usually have the camera on one of the two all auto settings, but you will have to know when to shift to aperture priority and a fixed ISO.

It took me a while to get satisfying results, especially using the Rx100's along the big Canons. I now press the shutter mostly by squeezing the camera. That has helped a lot. The tiny sensor is quite forgiving obviously compared to full frame, but there's a limit to it.

 

hope this helps,

wim

Wow, Wim thank you so much. I will certainly  carry out the test. Very detailed. Thanks. And yes blockiness is exactly how I would describe the sky and had assumed luminence noise, mainly because it or similar appears on the water too. I will get the test done and see the results.

amazing help and thank you sincerely for taking the time.

Jenny

[wiskerke]

Have you considered that the blockiness may stem from your screen having insufficient color depth?

Maybe test this with an image with the Canon and the Sony under the exact same circumstances and then trying to achieve the exact same results (esp saturation) with both files.

 

There are monitor tests as well. Just have none at the ready just now.

 

wim

[Chuck Nacke]

George and all,

 

I know there are many SONY fans, but I have come to hate SONY, even though I worked directly for SONY Records.

I only use NIKON DSLR's and yes I know that SONY unfortunately makes the sensors in the current NIKON's.

I only shoot NEF (RAW) and do my 16bit conversions to TIFF in Lightroom, then to 8bit JPEG's for upload.

The only QC fail I have had since 2003 was one image that I cropped too much and it was under the old

requirement of 48MB.

 

Did I say that I hate SONY......

 

Chuck

[John Mitchell]

5 hours ago, wiskerke said:

I don't really see luminance noise either. Some color noise maybe. Also some blockiness in the sky. Could be just jpg-conversion.

 

 

I don't have a good understanding of noise, but I would have thought that the noise in the sky is chroma noise. Correct?

 

I find that it's often impossible to get rid of it entirely in some images, but QC seems OK with that.

[hdh]

I had a look at your 100% images, 

 

On my 4K monitor, the noise in the sky was hardly visible. 

But it is visible on my 2K monitor.

It may be borderline probably comes from using a small sensor, but I would not mind sending that for QC.  

 

The bigger issue is the lack of a focus point and I agree fully with Wim's very detailed analysis. 

 

1 hour ago, Chuck Nacke said:

<snip>

Did I say that I hate SONY......

</snip>

Also agree with this but not with the Nikon bit

[Alex Ramsay]

I can't see any serious sky noise to speak of, and I've never had one of my RXII shots fail QC on those grounds. Presumably you are exposing correctly in camera rather than altering things in post? And it's not a bad idea with this camera to downsize before submitting to Alamy, too.

 

Alex

[george]

Thanks again to everyone for your comments. The example images I posted were completely unedited. 

One thing that has become obvious is that my moniter needs calibrating again. It is something I have not done for a while and now from comments and seeing my examples on the ipad I realise it needs doing. The noise in the sky has no colour on my monitor but does on my ipad and others see it too. Its good to know also, that most of you don't think the noise as big an issue as I do. 

I will do the lens testing, calibrate the monitor and take it from there. I wonder also if having the onset of catarracts is not helping? But hey, thats another issue all together. 

 

Thank you again everyone who took the time. Wishing you all a good week ahead

Jenny

[John Mitchell]

Good luck. Personally, I'd go back and experiment with retaking these photos at another time of day. I've always found photographing on beaches in harsh light to be challenging, no matter what camera I was using. On my not-so-great monitor, I too see what I always thought was "chroma" or colour noise in the sky in your images. But perhaps I'm all wet on that score. After all, I've never owned a Nikon (or even a Canon for that matter).

[george]

18 hours ago, John Mitchell said:

Good luck. Personally, I'd go back and experiment with retaking these photos at another time of day. I've always found photographing on beaches in harsh light to be challenging, no matter what camera I was using. On my not-so-great monitor, I too see what I always thought was "chroma" or colour noise in the sky in your images. But perhaps I'm all wet on that score. After all, I've never owned a Nikon (or even a Canon for that matter).

yes I think I will do that. There is a change in weather coming so will be ideal (for photographs, not for me, haha)

I have stopped all editing now and waiting for the delivery of a Spyder calibrater, for some strange reason my PC running windows 7 will not allow me to calibrate via Windows calibration on my HP monitor, but Spyder will be better anyway. I am seriously doubting what I am currently seeing now. 

 

[alphaomega]

I have used the RX100M3 extensively since purchasing it a couple of years ago. Never had a problem of the nature listed. Most of my images of architecture, travel or landscape pass through Alamy QC at full size. Sometimes I reduce to 30/35 Mb in overcast weather or the corners are not perfectly sharp. I use Aperty priority around F7.1. This 1" sensor does produces luminance noise in RAW. This is a fact. I simply move the Camera Raw slider until the noise level is gone in the most obvious areas and compensate with the sharpness slider also in Camera Raw. As said, if there is still too much noise in dark/shadowy areas then I simply downsize. Will not go below 24Mb. If I have to go lower I reject the image as being not suitable.

Simply put, if one cannot get this camera to produce acceptable images for Alamy under most conditions, then either read the manual or send the camera back for repair. This is a wonderful tool for getting acceptable Alamy images in a tiny package. Given up on LR as PS Camera Raw will do all I require. I just open in ACR and then go into PS from there as I have my own file system. I am certain that my images would pass in most cases if I used the Jpeg, but I like to do my own adjustments without starting with a Sony Jpeg.

Don't understand how anyone can "hate" Sony. They are now the "mould" breakers. Just look at the reviews of their A9 and A7M3 cameras. Nikon is trailing and only have prof. cameras due to Sony produced sensors. Canon sensors struggle to match Sony sensors for dynamic range. My A6000 with Sony zooms give me DSLR quality at a great weight and size advantage.

 

What is there not to like about Sony?