Pages: 1 ... 64 65 [66] 67 68 ... 80   Go Down

Author Topic: Nikon’s new mirrorless system, coming in ... late September 2018  (Read 159525 times)

davidgp

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 758
    • davidgp fotografia
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1300 on: October 14, 2018, 06:54:00 am »

I think this shows some banding, look on the bottom and left upper corner of the black door. This is the nef worked in LR, it looks even less obvious on the exported tiff (along with a more natural look). Keep in mind this was shot as a test, has exposure + 1.8, shadows +100, highlights -88 and clarity +21.
I tried to exported at full size with lower quality but the site gives an error.

He updated his tests: https://blog.kasson.com/nikon-z6-7/nikon-z7-banding-revisited/


http://dgpfotografia.com

elliot_n

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1219
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1301 on: October 14, 2018, 08:09:55 am »

Interesting. So you get banding when there is a large brightness range. Which is the exact real-world scenario where you might find yourself pushing the shadows hard (as in Armand's example). Reminds me of the Canon 5D.

--------

Edit to add:

There's a long and ongoing thread discussing Jim Kasson's findings on DPReview:

https://www.dpreview.com/forums/thread/4328936

As of now, the cause of the problem is a mystery. They are currently looking at two images (with a bright light source included in frame), taken one second apart, with identical settings. One displays banding, the other does not.
« Last Edit: October 14, 2018, 11:05:14 am by elliot_n »
Logged

John Camp

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2171
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1302 on: October 14, 2018, 11:52:46 am »

I read through the whole thread on DPR and it's important to note that they're having trouble both forcing the banding and then spotting it. This reminds me a bit of audiophiles wandering around in a room to find the perfect spot to listen to music, when in fact they're not listening to music at all, they're trying to detect technical anomalies in sound waves, and sometimes they agree, and sometimes they don't, on what they find. Whatever the guys on DPR find, it seems as though it won't much affect actual photographs.
Logged

D Fuller

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 608
    • AirStream Pictures
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1303 on: October 14, 2018, 12:25:24 pm »

Interesting. So you get banding when there is a large brightness range. Which is the exact real-world scenario where you might find yourself pushing the shadows hard (as in Armand's example). Reminds me of the Canon 5D.

--------

Edit to add:

There's a long and ongoing thread discussing Jim Kasson's findings on DPReview:

https://www.dpreview.com/forums/thread/4328936

As of now, the cause of the problem is a mystery. They are currently looking at two images (with a bright light source included in frame), taken one second apart, with identical settings. One displays banding, the other does not.

I think I’m beginning to develop a theory for what might be going on with this. It does only seem to occur when there’s a substantial bright area in the frame as well as an underexposed area. A couple of things seem consistent:
  • The dark area has to be so underexposed that the development is really digging it out of the noise floor.
  • The bright area seems to have to either be a substantial part of the image, or it has to be very overexposed.

It seems to me that this effect might be a kind of flare (not sure if that is the right term, but I don’t have another at the moment), where the bright light is bouncing around the inside of the sensor box or reflecting off the cover glass. In that case, what is recorded is a very faint reflected image of the sensor, similar to the familiar red dot problem some cameras exhibit. (The stripes, in that case, might be the reflected pattern of the PDAF sensels.)

If that is the case, the angle of the incoming light would surely be a factor, as might shutter speed and the use of EFCS, as they would change the reflective surfaces inside the camera.

Again, this is just a theory, so take it for what it’s worth.
« Last Edit: October 15, 2018, 10:11:37 pm by D Fuller »
Logged
business website: www.airstream.pictures
blog: thirtynineframes.com/blog

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1304 on: October 14, 2018, 06:17:41 pm »

I read through the whole thread on DPR and it's important to note that they're having trouble both forcing the banding and then spotting it. This reminds me a bit of audiophiles wandering around in a room to find the perfect spot to listen to music, when in fact they're not listening to music at all, they're trying to detect technical anomalies in sound waves, and sometimes they agree, and sometimes they don't, on what they find. Whatever the guys on DPR find, it seems as though it won't much affect actual photographs.

I agree, John. The banding on the Canon 5D, and later on the Nikon D7100 (at least on my model) was something that one didn't have to go searching for, but was something which hit you in the face during normal processing when lifting deep shadows.

The examples I've seen so far, of the occasional, slight banding on the Z7 in certain circumstances which are perhaps not yet fully understood, do not seem to me to be a problem from any practical perspective of taking technically good photos, but are more like a distraction.

Nevertheless, some benefit to basic image quality might eventually arise as a result of the ongoing investigation.
Logged

armand

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5540
    • Photos
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1305 on: October 14, 2018, 06:57:20 pm »

So far I found the banding only when I went looking for it, meaning taking the shot and processing with the purpose of finding banding.

Here is a file where there was some DR challenge but I didn't push the file beyond expectations (no exposure change, - 86 highlights, + 100 shadows, contrast + 26, clarity +10) and I got no banding that I could see.

JohnHeerema

  • Full Member
  • ***
  • Offline Offline
  • Posts: 241
  • Dr. John Heerema
    • http://www.heerema.ca
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1306 on: October 15, 2018, 05:06:07 pm »

Quote
Did anybody tried properly the focus stacking/ shifting? I went through what I did and the results are terrible, maybe a couple of shots in focus and the rest totally out of focus. I'll have to do this on a more organized fashion to see where the problem might be.

This is definitely a bug in the firmware.

I was really keen on focus stacking (focus shift in Nikon-speak) - in fact, it was one of my reasons for getting the Z7. Unfortunately, focus stacking is an ergonomic disaster. It looks like someone at Nikon had read that there was something called focus stacking, and wanted to be able to put it on the feature checklist, but didn't actually know anything about it. Presumably there weren't any smart people available to implement it who knew how focus stacking works.

The most annoying thing about it is that the live view goes away while you're setting it up, and there's no way to actually aim the camera before starting the sequence. This makes handheld focus stacks almost impossible.

This could easily be mitigated in firmware, by assigning the “start” action to a function button.

There’s more. At the default settings, the Z7 will take 100 photos at it’s “middle” focus step. If the 24-70 S lens is focused about 1.5 metres away, the first three shots will be focused on something this side of infinity. The remaining 97 frames will be focused past infinity.

Even on the smallest focus step, only five of those shots will be this side of infinity. Although the documentation claims that the sequence will stop when focus reaches infinity, it does not.

The minimum focus step is much too small.

 I can easily handhold well enough to do half a dozen shots at different focus distances, which is enough depth of field even for the very large canvas prints I make, if combined with stitching. I tried attaching this function to a control ... but I still get taken to the focus stack screen, and so I can't see what I'm trying to focus stack. Too bad - this would be such a simple firmware fix. More sensible control over the focus increments would be really handy too, and would also be easy to calculate in camera (if Nikon folks were interested, I'd be happy to help them out, as I have expertise in the area).
Logged

D Fuller

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 608
    • AirStream Pictures
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1307 on: October 15, 2018, 10:00:52 pm »

Did anybody tried properly the focus stacking/ shifting? I went through what I did and the results are terrible, maybe a couple of shots in focus and the rest totally out of focus. I'll have to do this on a more organized fashion to see where the problem might be.

I did this shot yesterday--Z7, 24-70S. It's a 21-image stack combined in Helicon Focus.

It worked very easily. I set the focus just shy of the nearest point on the orchid, and set the parameters for 25 images, with a focus interval of "1". The camera shifted focus toward infinity as it made the 25 shots. I chose 21 of them and processed in Helicon Focus at its defaults.

I really don't know what I'm doing here, These are the first stacked images I've ever made.

The second one is 100 images, with the same parameters. I don't like it as much as the first, but it certainly worked.



« Last Edit: October 15, 2018, 10:17:26 pm by D Fuller »
Logged
business website: www.airstream.pictures
blog: thirtynineframes.com/blog

D Fuller

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 608
    • AirStream Pictures
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1308 on: October 15, 2018, 10:10:10 pm »

This is definitely a bug in the firmware.

I was really keen on focus stacking (focus shift in Nikon-speak) - in fact, it was one of my reasons for getting the Z7. Unfortunately, focus stacking is an ergonomic disaster. It looks like someone at Nikon had read that there was something called focus stacking, and wanted to be able to put it on the feature checklist, but didn't actually know anything about it. Presumably there weren't any smart people available to implement it who knew how focus stacking works.

The most annoying thing about it is that the live view goes away while you're setting it up, and there's no way to actually aim the camera before starting the sequence. This makes handheld focus stacks almost impossible.

This could easily be mitigated in firmware, by assigning the “start” action to a function button.

There’s more. At the default settings, the Z7 will take 100 photos at it’s “middle” focus step. If the 24-70 S lens is focused about 1.5 metres away, the first three shots will be focused on something this side of infinity. The remaining 97 frames will be focused past infinity.

Even on the smallest focus step, only five of those shots will be this side of infinity. Although the documentation claims that the sequence will stop when focus reaches infinity, it does not.

The minimum focus step is much too small.

 I can easily handhold well enough to do half a dozen shots at different focus distances, which is enough depth of field even for the very large canvas prints I make, if combined with stitching. I tried attaching this function to a control ... but I still get taken to the focus stack screen, and so I can't see what I'm trying to focus stack. Too bad - this would be such a simple firmware fix. More sensible control over the focus increments would be really handy too, and would also be easy to calculate in camera (if Nikon folks were interested, I'd be happy to help them out, as I have expertise in the area).

It certainly would be tricky to use the function hand-held, but on sticks, it's manageable. While you set up the parameters, you can hit the OK button and you'll get Live View back. The camera holds the parameters you've set, so you can get everything set up, check the shot, then start the sequence.

I have no idea what the focus interval scale indicates. But in my attempts, the camera started from where I was focussed and incremented toward infinity, so its actions seemed very logical to me. But again, I have no other experience with focus stacking to compare it to.

It does seem really odd that it requires that you work blind.
« Last Edit: October 15, 2018, 10:15:59 pm by D Fuller »
Logged
business website: www.airstream.pictures
blog: thirtynineframes.com/blog

HSakols

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1239
    • Hugh Sakols Photography
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1309 on: October 16, 2018, 07:57:42 pm »

Thom Hogan said that Nikon does not recommend user sensor cleaning.  I've always used a sensor cleaning brush with my Nikon D800.  What are users supposed to use for the Z7?  There are times as a landscape photographer that I get dust on my sensor even after being very careful with lens switches. 
Logged

armand

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5540
    • Photos
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1310 on: October 16, 2018, 08:31:45 pm »

Thom Hogan said that Nikon does not recommend user sensor cleaning.  I've always used a sensor cleaning brush with my Nikon D800.  What are users supposed to use for the Z7?  There are times as a landscape photographer that I get dust on my sensor even after being very careful with lens switches.

I guess it's similar to the Olympus IBIS. Oly didn't recommend it either. Good that many found magical properties of the dust cleaning function and truth be told I have yet to obviously get my E-M5ii dirty enough to notice it and to require cleaning. One would hope Nikon will be as good.
If not, if the blower fails we face a difficult decision.

armand

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5540
    • Photos
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1311 on: October 16, 2018, 08:40:43 pm »

I got to play a little with the 70-200 F4 via the FTZ adapter. Relatively well balanced, handles well.
For AF-S seems ok although in dimmer light it had issues locking focus, see attached shot where I was focusing on the cloud. For other subjects, see the combine shot, did ok.
For stationary objects with AF-C there was some hunting, will need to experiment more.


Overall the image quality is there. For slow focusing requirements tasks, such as landscape, will do the job. Is it worth it it? The Sony is probably a better value unless you have many Nikon lenses. It feels like it wasn't fully ready for prime time and just rushed it out to stop the leaking to other systems.

Rory

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 528
    • Recent images
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1312 on: October 16, 2018, 10:34:48 pm »

I have no idea what the focus interval scale indicates. But in my attempts, the camera started from where I was focussed and incremented toward infinity, so its actions seemed very logical to me. But again, I have no other experience with focus stacking to compare it to.

It does seem really odd that it requires that you work blind.

The focus interval refers to the amount of DOF overlap between images.  Try 4 - it gives a reasonable overlap.  The distance and aperture will change how many images are required, but the "interval" between images to achieve similar DOF overlap will remain constant.

Compared to the Olympus implementation the Nikon version is pathetic (on the D850).  My EM1.2 shows the progression in the LCD in real time and I can just press the shutter button again when the stacking has progressed through the subject.  However, if your stack has not made it all the way through the subject you can just start another stack and it will continue from where it left off.
Logged
[url=http://www.flickr.com/photos/roryhi

chez

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2501
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1313 on: October 16, 2018, 10:42:07 pm »

I guess it's similar to the Olympus IBIS. Oly didn't recommend it either. Good that many found magical properties of the dust cleaning function and truth be told I have yet to obviously get my E-M5ii dirty enough to notice it and to require cleaning. One would hope Nikon will be as good.
If not, if the blower fails we face a difficult decision.

I've had dust in very humid situations that just stuck to the sensor and even a rocket blower would not budge it. I needed to swab the sensor. I'd really hate to send my camera into Nikon every time I got the sensor dirty...which would be 3 to 5 times a year. I change lenses when needed, even at the ocean shores where there is blowing sand and salt spray. I'd hate to have to skip changing lenses because I might get something on my sensor.
Logged

armand

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5540
    • Photos
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1314 on: October 16, 2018, 11:30:47 pm »

I've had dust in very humid situations that just stuck to the sensor and even a rocket blower would not budge it. I needed to swab the sensor. I'd really hate to send my camera into Nikon every time I got the sensor dirty...which would be 3 to 5 times a year. I change lenses when needed, even at the ocean shores where there is blowing sand and salt spray. I'd hate to have to skip changing lenses because I might get something on my sensor.

Another reason why I like the E-M5ii with the 12-100, I don't really need to change the lens  ;)
Suddenly the lack of IBIS is not such a disadvantage anymore.

For Oly, or all other IBIS cameras, it depends how often it happens. Once a year, fine. Several times, not so much. My guess is that people will bite the bullet and risk voiding the warranty by doing the cleaning themselves. It will be challenging tough, I think for many IBIS cameras the sensor is quite mobile when the camera is turned off. A nice feature would be to "park" it upon shutdown, not only for ease of cleaning but also to avoid damage during transport.

chez

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2501
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1315 on: October 17, 2018, 03:20:09 am »

Another reason why I like the E-M5ii with the 12-100, I don't really need to change the lens  ;)
Suddenly the lack of IBIS is not such a disadvantage anymore.

For Oly, or all other IBIS cameras, it depends how often it happens. Once a year, fine. Several times, not so much. My guess is that people will bite the bullet and risk voiding the warranty by doing the cleaning themselves. It will be challenging tough, I think for many IBIS cameras the sensor is quite mobile when the camera is turned off. A nice feature would be to "park" it upon shutdown, not only for ease of cleaning but also to avoid damage during transport.

I've cleaned my A7R2 sensor that has ibis dozens of times with sensor swabs with no issues. Don't see what the problem is.
Logged

Martin Kristiansen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1527
    • Martin Kristiansen
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1316 on: October 17, 2018, 07:42:07 am »

Also cleaned my IBIS enabled Sony many times. You run a cleaning cycle where the IBIS shakes to clear off dust. Once that finishes the sensor is “parked” and held firmly in place to enable cleaning without it wobbling about. Then you shut down the camera and the sensor is released from its parked mode.

Sure Nikon will have something similar.
Logged
Commercial photography is 10% inspiration and 90% moving furniture around.

armand

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5540
    • Photos
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1317 on: October 17, 2018, 09:04:28 am »

I've cleaned my A7R2 sensor that has ibis dozens of times with sensor swabs with no issues. Don't see what the problem is.

Is it accepted by Sony without risk voiding your warranty?
« Last Edit: October 17, 2018, 10:39:48 am by armand »
Logged

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4388
    • Pieter Kers
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1318 on: October 17, 2018, 10:27:40 am »

The sensor coverglass is expected to be far before the sensor- so less problems with dust .
The coverglass is sensitive and plastic-  vulnerable.

My d850 and j5 did not have to be cleaned after one year of usage other than by the camera or the blower.
and i change lenses very often and use f8-11 a lot.
Logged
Pieter Kers
www.beeld.nu/la

armand

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5540
    • Photos
Re: Nikon’s new mirrorless system, coming in ... late September 2018
« Reply #1319 on: October 17, 2018, 10:45:12 am »

The sensor coverglass is expected to be far before the sensor- so less problems with dust .
The coverglass is sensitive and plastic-  vulnerable.

Are you sure the coverglass is plastic? It doesn't sound right and seems like a bad idea, how is somebody going to clean it without scratching it?
Pages: 1 ... 64 65 [66] 67 68 ... 80   Go Up