23 replies to this topic

[mateus]

So, i have a Nikon D750 and i've noticed some really strange things about it.

 

There's some serious swatches and weird collor patterns (specially on the red channel) and they vary in different ISO settings.

Also a white circle/haze around all my images, even with dew heater on and covering the viewfinder to protect the sensor from reflections

 

Following image was taken with a red cat 51 at ISO 400. 55 minutes of integration time. Histogram view in Siril. Color callibration + DBE
Here's the stacked file - https://drive.google...iew?usp=sharing

Attached Thumbnails

• 

[Dynan]

Just dust on the sensor. Clean it carefully. There are many ways to properly do it described here on CN.

 

I believe you'd need a reducer/flattener to remove the severe vignetting you're seeing.

[mateus]

Just dust on the sensor. Clean it carefully. There are many ways to properly do it described here on CN.

 

I believe you'd need a reducer/flattener to remove the severe vignetting you're seeing.

The pattern is different when using different ISO settings...

Reducer for the RedCat? 

[TOMDEY]

Follow Dynan's advice and see what you get. The sensor itself is probably fine.    Tom

[sharkmelley]

The dust shadows can be fixed by using flats.  The saturated white circle is a processing problem (a more careful DBE is required) and flats would help enormously.

 

Unfortunately you are also seeing typical problems of Nikon cameras.  The vertical green stripe in your image is caused by the Sony split sensor (the full-frame sensor operates in 2 halves).  The coloured rings are almost certainly caused by Nikon's infamous hardcoded image correction which applies crude corrections to the raw data.  For the same length of exposure the spacing of the coloured rings will be dependent on ISO.

 

Up until now, I have not had any firm data on the Nikon D750 but yours would appear to be behaving like the others.

 

It would be very useful if you could make available one or two raw NEF files, especially if they are at different ISOs.  It would give me the evidence needed to update the Nikon D750 information on my website.

 

Mark

Edited by sharkmelley, 03 September 2023 - 01:17 PM.

[mateus]

The dust shadows can be fixed by using flats.  The saturated white circle is a processing problem (a more careful DBE is required) and flats would help enormously.

 

Unfortunately you are also seeing typical problems of Nikon cameras.  The vertical green stripe in your image is caused by the Sony split sensor (the full-frame sensor operates in 2 halves).  The coloured rings are almost certainly caused by Nikon's infamous hardcoded image correction which applies crude corrections to the raw data.  For the same length of exposure the spacing of the coloured rings will be dependent on ISO.

 

Up until now, I have not had any firm data on the Nikon D750 but yours would appear to be behaving like the others.

 

It would be very useful if you could make available one or two raw NEF files, especially if they are at different ISOs.  It would give me the evidence needed to update the Nikon D750 information on my website.

 

Mark

Would you be able to walk me through how to do a better DBE? Also, i'd like to know if there's anything i can do to prevent the Sony split sensor thing

 

After using your Nikon Ring Removal program, the white circle looks exactly the same as i had with my Rokinon (i posted about it some time ago), so i think it's a camera problem

 

45 second sub at ISO 800 - https://drive.google...iew?usp=sharing
60 second sub at ISO 400 - https://drive.google...iew?usp=sharing

Unfortunately, i don't have subs with the same exposure time at different ISO settings

[sharkmelley]

The Nikon Ring Removal program works only for the rings caused by Nikon's faulty lossy compression algorithm.  However, the raw exposures you uploaded use lossless compression (which is the right thing to do), so they are not affected by this problem.

 

The rings you are seeing are caused by the Nikon hardcoded correction (probably a correction for sensor colour shading).  In the case of the D750 this will be applied by Nikon's firmware whether a lens or a telescope is being used.  There is no program that can remove the rings caused by the hardcoded sensor correction.

 

For your stacked image here is the result of dividing the red channel by the green channel and stretching:

 

 

The rings and the vertical band can be clearly seen.  They were also visible in a single raw exposure.  Unfortunately, the rings are obvious even though your back-of-camera histogram is nearly halfway across.  It's possible there's a exposure sweet spot where the rings in the red channel will disappear or almost disappear.  This would be at low ISO with a much shorter exposure e.g. 20seconds at ISO 200 or ISO 100.  But this might trigger rings in the blue channel.  The only way to know is to try it.

 

As for the vertical band, this is caused by the split sensor - it's one of the worst split sensor artefacts I've seen and I've no idea what effect short exposures at low ISO will have.  In general, split sensor artefacts vary greatly from copy to copy of the same camera.  Another D750 might have no split sensor artefact at all.

 

I'm sorry to say that ultimately this will always be a frustrating camera to use for deep-sky astrophotography.

 

Mark

Edited by sharkmelley, 03 September 2023 - 04:11 PM.

[mateus]

The Nikon Ring Removal program works only for the rings caused by Nikon's faulty lossy compression algorithm.  However, the raw exposures you uploaded use lossless compression (which is the right thing to do), so they are not affected by this problem.

 

The rings you are seeing are caused by the Nikon hardcoded correction (probably a correction for sensor colour shading).  In the case of the D750 this will be applied by Nikon's firmware whether a lens or a telescope is being used.  There is no program that can remove the rings caused by the hardcoded sensor correction.

 

For your stacked image here is the result of dividing the red channel by the green channel and stretching:

 

NikonD750_HardcodedCorrection.JPG

 

The rings and the vertical band can be clearly seen.  They were also visible in a single raw exposure.  Unfortunately, the rings are obvious even though your back-of-camera histogram is nearly halfway across.  It's possible there's a exposure sweet spot where the rings in the red channel will disappear or almost disappear.  This would be at low ISO with a much shorter exposure e.g. 20seconds at ISO 200 or ISO 100.  But this might trigger rings in the blue channel.  The only way to know is to try it.

 

As for the vertical band, this is caused by the split sensor - it's one of the worst split sensor artefacts I've seen and I've no idea what effect short exposures at low ISO will have.  In general, split sensor artefacts vary greatly from copy to copy of the same camera.  Another D750 might have no split sensor artefact at all.

 

I'm sorry to say that ultimately this will always be a frustrating camera to use for deep-sky astrophotography.

 

Mark

Ah alright, thank you so much for the help.
I was having the same problem in the past but a simple camera firmware update fixed it... how come it be so bad now?

Would it be worth it to sell the camera and buy a astro-dedicated ZWO? I'm so annoyed by the fact that a $1600 is able to produce artifacts worse than my $100 13 year old DSLR... Never did i expect to have such results with it. People i know who have it say they had no problems at all - from split sensor to weird rings. Nothing

Edited by mateus, 03 September 2023 - 04:40 PM.

[sharkmelley]

Ah alright, thank you so much for the help.
I was having the same problem in the past but a simple camera firmware update fixed it... how come it be so bad now?

Would it be worth it to sell the camera and buy a astro-dedicated ZWO? I'm so annoyed by the fact that a $1600 is able to produce artifacts worse than my $100 13 year old DSLR... Never did i expect to have such results with it. People i know who have it say they had no problems at all - from split sensor to weird rings. Nothing

There's a lot we don't know about the hardcoded sensor correction.  For instance, it's entirely possible that an earlier firmware version does not have the same issues.  

 

A dedicated ZWO will not have the same problems but full-frame dedicated astro-cameras are much more expensive than full-frame consumer cameras.

 

Mark

[mateus]

There's a lot we don't know about the hardcoded sensor correction.  For instance, it's entirely possible that an earlier firmware version does not have the same issues.  

 

A dedicated ZWO will not have the same problems but full-frame dedicated astro-cameras are much more expensive than full-frame consumer cameras.

 

Mark

Some weird pattern were also visible before i updated my firmware (C 1.12)

About dedicated astro-cams, i was taking a look at the ZWO 533mc Pro and the Uranus-C Pro. The sensor size really fits most targets i'd shoot very well in the frame

Edited by mateus, 03 September 2023 - 10:18 PM.

[sharkmelley]

About dedicated astro-cams, i was taking a look at the ZWO 533mc Pro and the Uranus-C Pro. The sensor size really fits most targets i'd shoot very well in the frame

The ZWO 533MC Pro is a good solution, if its small sensor is large enough for you.  But I have no experience of dedicated astro-cams because I switched to Canon.  Unfortunately Sony, Nikon and FujiFilm insist on performing undocumented processing on the their raw data which makes many of their cameras unsuitable for deep-sky astrophotography. 

[mateus]

The ZWO 533MC Pro is a good solution, if its small sensor is large enough for you.  But I have no experience of dedicated astro-cams because I switched to Canon.  Unfortunately Sony, Nikon and FujiFilm insist on performing undocumented processing on the their raw data which makes many of their cameras unsuitable for deep-sky astrophotography. 

Ight, thank you so much for your help. 

[mateus]

The ZWO 533MC Pro is a good solution, if its small sensor is large enough for you.  But I have no experience of dedicated astro-cams because I switched to Canon.  Unfortunately Sony, Nikon and FujiFilm insist on performing undocumented processing on the their raw data which makes many of their cameras unsuitable for deep-sky astrophotography. 

Hi Mark. I'm still trying to figure out what to do. I have a lot of Nikon equipment, so i've been taking a look at some of their newer cameras. The z5 seems very tempting. Do you know if it also has those problems we discussed on this topic? May be a really stupid question, but I can't really find anything about it in the internet

[vidrazor]

Hi Mark. I'm still trying to figure out what to do. I have a lot of Nikon equipment, so i've been taking a look at some of their newer cameras. The z5 seems very tempting. Do you know if it also has those problems we discussed on this topic? May be a really stupid question, but I can't really find anything about it in the internet

If your Nikon camera works OK terrestrially, just use it that way and get the Uranus-C. The Uranus-C is a great inexpensive all-around camera you can use for DSO, planetary, and lunar imaging. So far I've only used it for planetary, but if weather permits next week I'm going to finally get a chance to shoot DSO. If you're curious what it can do, just go check out the images over at AstoBin. It has an active cooling option, but honestly you don't need it. It's a Starvis 2 gen sensor and it's really efficient. No need for darks either.

[mateus]

If your Nikon camera works OK terrestrially, just use it that way and get the Uranus-C. The Uranus-C is a great inexpensive all-around camera you can use for DSO, planetary, and lunar imaging. So far I've only used it for planetary, but if weather permits next week I'm going to finally get a chance to shoot DSO. If you're curious what it can do, just go check out the images over at AstoBin. It has an active cooling option, but honestly you don't need it. It's a Starvis 2 gen sensor and it's really efficient. No need for darks either.

My D750 works perfectly for photography, but i really don't use it that often. I do want a camera for photography, but at that price, i want one that i can use for astro too.

 

About the Uranus-C, i was taking a look at the Pro version. It is very interesting, and a great option as i do planetary too. 

 

I think it really comes down to use now, and i'll maybe go with the Uranus-C.

 

Thanks

[vidrazor]

About the Uranus-C, i was taking a look at the Pro version. It is very interesting, and a great option as i do planetary too.

Honestly I don't see a need for the Pro, unless you live in hell.   The fellow in that video shot that DSO with an un-cooled Uranus-C for 6 hours in a heat wave where his observatory was at 104° F(!), and got the footage you see in that video.

[unimatrix0]

Has anyone ever informed Nikon about their firmware coding is having issues? There has to be a reason why they are doing it (and Canon does not, as far as I know) and I'm pretty sure the reason has something to do with daytime photography and results from it.  But as it seems, this issue has been around from older cameras and all the way to recent ones, so it's like they just keep plugging in the same firmware and pre-processing technique they use for all their cameras. 

 I wonder if daytime photographers noticed anything strange stuff or did we get these adjustments, because they complained about something, so Nikon hardcoded something that may fixed that issue, but ruined another (causing these circles)? 

[mateus]

Has anyone ever informed Nikon about their firmware coding is having issues? There has to be a reason why they are doing it (and Canon does not, as far as I know) and I'm pretty sure the reason has something to do with daytime photography and results from it.  But as it seems, this issue has been around from older cameras and all the way to recent ones, so it's like they just keep plugging in the same firmware and pre-processing technique they use for all their cameras. 

 I wonder if daytime photographers noticed anything strange stuff or did we get these adjustments, because they complained about something, so Nikon hardcoded something that may fixed that issue, but ruined another (causing these circles)? 

My D750 works perfectly fine for day-time photography. Those issues are only visible when i stretch my images considerably.

I don't think it has anything to do with solving a problem for day-time. Just my opinion about it.

 

I emailed Nikon about it a week ago. No response yet... If i do get one, i'll update you guys

[mateus]

Nikon was generous enough to not actually answer me, but ask me to rate their customer support 😂

[unimatrix0]

Nikon was generous enough to not actually answer me, but ask me to rate their customer support

I think both Nikon and Canon are a bit over themselves, even though the camera market has been shrinking significantly every since the cell phones. The only growth outside of cellphone imaging is astrophotography and security camera tech. They should be reading the signs on the wall, but it seems like they aren't paying attention. 

Both of these companies (not mentioning Olympus and Pentax for now) should get on the train before it leaves the station.  Their loss is others' gain (ZWO) , but any major camera brands like Canon or Nikon on board could significantly reduce the price and level the field. 

It would be interesting to see a Canon branded cooled-astro camera, but I have a feeling that may never happen. 

Edited by unimatrix0, 28 September 2023 - 12:59 PM.

[primeshooter]

Has anyone ever informed Nikon about their firmware coding is having issues? There has to be a reason why they are doing it (and Canon does not, as far as I know) and I'm pretty sure the reason has something to do with daytime photography and results from it. But as it seems, this issue has been around from older cameras and all the way to recent ones, so it's like they just keep plugging in the same firmware and pre-processing technique they use for all their cameras.
I wonder if daytime photographers noticed anything strange stuff or did we get these adjustments, because they complained about something, so Nikon hardcoded something that may fixed that issue, but ruined another (causing these circles)?

I contacted them; as I believe several have here (see the nikon ring thread). They replied to me saying they would submit it for consideration. I got a buddy to submit a ticket too; but he got the idiot response of "there is no issue, the rings are just something wrong with the lens vignetting or something like that."

[mateus]

Update: I went to a bortle 4 and took some widefield images with the camera. No issues at all

Maybe it's caused by the abundance of light in my bortle 7 home? Maybe its due to the exposure and ISO settings i used?

 

*The images i took were with a 14-24mm lens and a nifty fifty, not with my Roki or RedCat.*

 

Just note that the images were taken with 'unusual' settings. I did some 2 minutes exposures at ISO 1600, some at 800 and 30 seconds exposures at ISO 3200...

Even though different settings may fix the issue, i can't use them at home. The sky is simply too bright. I'd end up overexposing.

 

If needed, i can send the stacks i got so y'all can take a look at them.

[sharkmelley]

If needed, i can send the stacks i got so y'all can take a look at them.

I would certainly be interested in seeing a handful of raw files.

[mateus]

I would certainly be interested in seeing a handful of raw files.

 

Single 120 sec sub at ISO 1600 - https://drive.google...iew?usp=sharing

Stack - https://drive.google...iew?usp=sharing

 

Single 120 sec sub at ISO 800 - https://drive.google...iew?usp=sharing

Stack - https://drive.google...iew?usp=sharing