New Nikon mirrorless camera ad!
Moderators: rjlittlefield, ChrisR, Chris S., Pau
Steve, Lou,
Agree, that's one feature I wish Nikon had adopted instead of only "In Lens Stabilization" that Nikon and Canon have. Think the In Camera original idea dates back to Minolta, which Sony acquired. Maybe some patent, or licensing issue has prevented Nikon and Canon use
It seems to me that having both In Camera and In Lens Stabilization is the best of both worlds, one could envision that some clever algorithms could utilize both simultaneously for even greater effect.
Soon we shall see
Best,
Agree, that's one feature I wish Nikon had adopted instead of only "In Lens Stabilization" that Nikon and Canon have. Think the In Camera original idea dates back to Minolta, which Sony acquired. Maybe some patent, or licensing issue has prevented Nikon and Canon use
It seems to me that having both In Camera and In Lens Stabilization is the best of both worlds, one could envision that some clever algorithms could utilize both simultaneously for even greater effect.
Soon we shall see
Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike
~Mike
Agree, I cant see any downside to having an on/off feature. Tho i have to also admit that for studio macro/micro i always turn it off.
@Maywatt: sony E-mounts feature both IBIS and OSS which DO act in conjunction. Lens is handling 2 axes, while body handles 3 axes.
Not having G90 macro, i cant speak about it's being usefull in fieldwork, but im quite happy with IBIS alone and Minolta 100 macro.
@Maywatt: sony E-mounts feature both IBIS and OSS which DO act in conjunction. Lens is handling 2 axes, while body handles 3 axes.
Not having G90 macro, i cant speak about it's being usefull in fieldwork, but im quite happy with IBIS alone and Minolta 100 macro.
Last edited by JohnyM on Mon Jul 30, 2018 5:21 am, edited 1 time in total.
Micro four thirds already does combine both methods of stabilization in some bodies, I think. That's originally because Olympus MFT cameras used sensor-shifting stabilization (which also worked on legacy lenses) while Panasonis used in-lens stabilization (which is slightly better for long lenses).
Was the Minolta stabilization for mirrorless cameras? I would imagine that five-axis sensor movements would not play nicely with manual mirror-based focusing, since the image would go out of focus a bit because the mirror does not move. I susoect that is the reason Canon and Nikon did it in their lenses.
Was the Minolta stabilization for mirrorless cameras? I would imagine that five-axis sensor movements would not play nicely with manual mirror-based focusing, since the image would go out of focus a bit because the mirror does not move. I susoect that is the reason Canon and Nikon did it in their lenses.
Minolta cameras featured 2 axis XY stabilisation in DSLR bodies. 3 and 5 axes were introduced during sony government in II generation E-mount bodies, but SLT A99II also features 5 axis IBIS.
Sony IBIS is working with any adapted lens, but is restricted to 3 axes only.
Sony IBIS is working with any adapted lens, but is restricted to 3 axes only.
Last edited by JohnyM on Mon Jul 30, 2018 5:26 am, edited 1 time in total.
-
- Posts: 1511
- Joined: Mon Jan 15, 2018 9:23 pm
- Contact:
IBIS is great, but it comes with lots of issues as well. For my interests, astrolandscapes and macro, it's just counterproductive.Beatsy wrote: Really? I think the IBIS on my Sony cameras (A7rii and A9) is a wonderful thing. You get 4 or 5 stops of stabilisation that even works with adapted, legacy lenses. A little less (maybe 3 stops) at handheld macro scales in my experience, but it at least doubles my keeper rate compared to a fixed sensor. Fewer shots lost to camera shake - or I can lower ISO and shutter speed even further in low light situations and still get the shot. And you can simply switch it off when the camera is tripod or rail mounted.
It genuinely makes a noticeable and valuable difference for me - interested to hear what problem(s) it causes for you.
For ordinary x1-x2 macro in the field, I simply use a small tripod and a DIY stage. I do not handhold. The compensation depends on magnification, if I was to guess, it would be about 2.5 stops at 1:1 macro scale and barely useful at 2:1.
The major issue I have with IBIS is hot pixels, it's a heating element. It will render far more hot pixels compared to an ordinary fixated sensor. This downside is enough for me to not want it at all.
Eh, there is. Turning it off doesn't do anything, it's still there. IBIS bodies still generates far more dark current and yields higher read-out noise compared to ordinary bodies. One contributor to dark currents is interference... how does IBIS function? I'm not talking about "not much of difference", a huge difference already exists at a mere 2-minute-exposure. My stacks are typically exposures of 5-10 seconds.Lou Jost wrote:Steve, I agree completely. There is no downside since you can always turn it off. And that technology of moving the sensor precisely is what makes pixel-shifting possible too.
Also, I don't care about pixel shift. I'd like to see anyone successfully using it for 10x and 20x magnifications, and I'd like to know the technique used.
D810 (above) 707s VS a7m3 (below) 605s
D810 at room temperature 16 degrees, a7m3 at about 24. Not exactly a fair test obviously, I'd be happy for anyone in Australia to post dark frames of 5 and 10 minutes using their a7m3 and a7r3 cameras at the typical winter temperatures.
Assumptions:
Dead pixels not controlled. My D810 is pretty old at this point, the a7m3 is new. Sony also keeps the class A sensors to themselves, so the test is in favour of the a7m3 in this aspect.
These are screenshots at 100% magnification. I used lightroom. The frame size difference is merely due to the screenshot. a7m3 taken from the centre of the frame. For the D810, I couldn't find any hot pixels in the centre, so the screenshot is a mid-right segment.
Post: Highlights+100
The D810 being an entire 100 seconds longer showcased significantly less hot pixel noise.
Same as above but with additional exposure +3
The hot pixels on the a7m3 is so bad, it's visible on thumbnails.
Notice how the D810 is better at +3 EV compared to the a7m3 in the first set with exposure unaltered?
And if anyone is considering the a7m3, I'd give it deeper thoughts. There's a major issue with either sensor luminescence not being suppressed adequately or just shutter curtain lightleaks. This causes one side of the exposure to have such devastating purple pixel noise. According to my friend, he's encountered this in normal shooting, not common, but not rare enough to not be fussed about it. (As a reference, it took about 40k actuations for me to encounter the lightleak issue of the D750 once, and that problem was making multiple headlines).
This is just my preference, I don't want any IBIS flapping behind my sensor, therefore if Nikon does offer such a mechanism in their high MP body, I'm not going to consider it. People who only do typically up to 30 seconds of exposure wouldn't need to worry about anything here. If I was a wedding shooter, I'd get the a7m3 over either a d850 or an a7r3. It's still a really good camera despite all the issues.
Last edited by Macro_Cosmos on Wed Aug 01, 2018 8:19 am, edited 1 time in total.
-
- Posts: 1511
- Joined: Mon Jan 15, 2018 9:23 pm
- Contact:
An additional issue with the D810 here. For the 707s frame, I foolishly forgot to close the viewfinder curtain, which led to this.
Closing the curtain eliminates the leak completely. I wish all cameras have such a mechanism built it. Obviously a piece of gaffer tape does the same thing, but a dial is less of an annoyance.
That's interesting information, I hadn't thought about this aspect before. I now understand your dislike for IBIS, but I don't think it is a big deal for macrophotography, where we usually have plenty of light. In hand-held macro fieldwork under continuous light IBIS really does wonders, far outweighing the drawbacks you mention. Even when using flash, the viewfinder image is more stable and it is easier to get critical focus, though the advantages diminish as magnification increases.
Edit: A mirrorless camera of course will never have the problem you showed in your last post.
Edit: A mirrorless camera of course will never have the problem you showed in your last post.
-
- Posts: 1511
- Joined: Mon Jan 15, 2018 9:23 pm
- Contact:
Yeah I understand, and as I've stated, it's not suitable for me.Lou Jost wrote:That's interesting information, I hadn't thought about this aspect before. I now understand your dislike for IBIS, but I don't think it is a big deal for macrophotography, where we usually have plenty of light. In hand-held macro fieldwork under continuous light IBIS really does wonders, far outweighing the drawbacks you mention. Even when using flash, the viewfinder image is more stable and it is easier to get critical focus, though the advantages diminish as magnification increases.
Edit: A mirrorless camera of course will never have the problem you showed in your last post.
For field work handheld, IBIS is a great tool. Lens stabilisation does virtually nothing at macro distances... well in fact it does the complete opposite. I'm fairly sceptical at a "higher keeper rate" claim. Usually at least for me, a tosser is a complete out of focus shot. The DoF at macro distances is really thin as we all know. I will not go above f8 as diffraction is an annoyance. If a "keeper" includes low noise, adequate exposure, no motion blur due to shakes, then yes, IBIS will bring considerable advantages.
I personally carry 2 tripods, a small "hi-hat" style tripod and a travel tripod, I would go out in the early morning when insects aren't hyperactive. It's usually possible to get about 5s exposures.
I use a halogen fibre optic illuminator for lighting, therefore exposures can get long easily. I'd like to have iso 64 and expose to the right (+0.3 EV usually). One may say I'm way too strict, but that's just me and I have a set of criteria that only applies to my gear.
Yep - I didn't think of the astro long-exposure aspect either. Though even for tracked stuff, I tend to do shorter exposure sub-frames (30-60 secs) at higher ISO and stack them in post, so this isn't an issue I've encountered. Besides, the Sony "star eater" will hide most hot pixels (and a few stars) anyway
It seems the new Nikon mirrorless might have a Z mount 16mm flange to sensor distance, 49mm optical opening and 65mm total Z mount diameter. If so, eventually (when adapters are available) this should be very useful for some of our macro lens assemblies!!
Best,
Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike
~Mike
-
- Posts: 3417
- Joined: Sat Nov 20, 2010 10:40 am
- Location: Santa Clara, CA, USA
- Contact:
Pixel shift would not be very useful at high mags. Getting higher resolution by pixel shifting requires the information presented to the sensor to contain information worth shifting for. A 10x 0.28 Mitty is at f18, so you're not going to be extracting much sub-pixel information from it.Macro_Cosmos wrote:...
Also, I don't care about pixel shift. I'd like to see anyone successfully using it for 10x and 20x magnifications, and I'd like to know the technique used.
- enricosavazzi
- Posts: 1474
- Joined: Sat Nov 21, 2009 2:41 pm
- Location: Västerås, Sweden
- Contact:
I know for sure that the Olympus 300 mm f/4 on E-M1 Mark II body does combine both methods (which adds an extra stop or two of stabilization-equivalence to in-body-only stabilization). I don't remember hearing of any other Olympus lens doing the same, and I don't keep track of what other Olympus bodies do because I use only E-M1 and E-M1 Mark II. Perhaps combining both methods does make a significant difference only for long focal lengths.Lou Jost wrote:Micro four thirds already does combine both methods of stabilization in some bodies, I think. That's originally because Olympus MFT cameras used sensor-shifting stabilization (which also worked on legacy lenses) while Panasonis used in-lens stabilization (which is slightly better for long lenses).
[...]
I remember reading discussions of this combined method stating that the method could not be made any better than it is now, because the rotation of the Earth is detected by the accelerometers and fools the system into trying to compensate for it. This causes the system to take the wrong decision because the subject rotates together with the Earth and the camera, so no compensation is actually required. Canceling out the Earth rotation signal in the calculations is not a trivial problem, since it requires both the direction of shooting with respect to the Earth axis and the latitude to be known.
On the other hand, this is one more proof that the Earth is not flat and immobile at the center of the universe.
--ES
-
- Posts: 1511
- Joined: Mon Jan 15, 2018 9:23 pm
- Contact:
10 60s frames produces similar amounts of hot pixels in totality compared to one 10min frame. This is why people usually take some dark field frames in between to stack those pixels out. The "star eater" is kind of a null issue in my opinion, oversensationalised. I can see why people get fussy about it though.Beatsy wrote:Yep - I didn't think of the astro long-exposure aspect either. Though even for tracked stuff, I tend to do shorter exposure sub-frames (30-60 secs) at higher ISO and stack them in post, so this isn't an issue I've encountered. Besides, the Sony "star eater" will hide most hot pixels (and a few stars) anyway
My typical stacks consist of 200 5s exposures. I'd rather not embed dark fields in between. Occasionally some hot pixels can be found, it has never been an issue. One solution would be speedlights -- not my cup of tea though. My setup is stable enough.
Yeah, that's one point.ray_parkhurst wrote: Pixel shift would not be very useful at high mags. Getting higher resolution by pixel shifting requires the information presented to the sensor to contain information worth shifting for. A 10x 0.28 Mitty is at f18, so you're not going to be extracting much sub-pixel information from it.
Another point would be the precision of pixel shift. Pixel shift as it currently is just cannot handle the precision required for high-mag. work. A friend tried this on his H6D-400C, the results were just abysmal. He didn't even use objectives. The lens was a Hassy 120mm Macro, and that alone presented so many complications leading to a tosser. Basically a nearby construction site also contributed to the blurry exposure... and we're talking about 1:1. 5:1? Yeah... So pixel shift doens't interest me at all. It's a good qwerk, there's however lots of improvements required.