This was fascinating. I was already confident that Nikon was not exaggerating their claims regarding sensor scan rate, and your experiment clearly confirms it. Others have posted interesting videos regarding other features (e.g., Matt Granger and Ricci Talks) as well as, of course, the beautiful wildlife photos posted by Nikon Canada, Michelle Valberg, and others.Going back to the DP Review chart you showed, I think one of the biggest achievements is in the top line: the highly competitive price. That was a brilliant move on Nikon’s part.

Thank you for one of the most helpful videos on the Z9...

nice video (as a A1 owner) people getting the z9 should not be worried about using electronic shutter. I and many of my A1 friends don't use the mech shutter at all. I do suspect the Z9 to be very slightly faster simply because it has a little less work todo but won't effect the results. I am not sure how you make the call that the cpu vs the cpu's (A1 uses 2) are faster/slower both are new and we are not sure if they are used to their full potential yet (and i say this as a soft/hardware designer) my guess is both can be pushed more over their lifecycle. Lastly why stop at 1/8000 and not go to the max setting of 1/32000 ? I understand you wanted to compare and you clearly proof your point people should not worry but still would be nice to see.

Thanks for taking the time out to explain this.

I tried many times but miserably failed to find a list of "e-shutter rate" or readout speed of different sensors on different cameras... Even on DPReview. Anyone help a nerd get some info?

This is a brilliant video 👍 I liked the idea of using a colour LED fan in this comparison. Thanks very much for making my day.

Fascinating as always and thorough analysis

This is great to see. Be interested in seeing how much the handheld jitter is reduced in video. IBIS has really helped but there are situations where you may need it turned off.

Please do the same experiment with your X1D with Leaf shutter, that should do a real 1/2000 mec shutter, that should produce significant sharper result than any other curtain or e-shutter.

The rolling shutter pic looks quite cool though!

how do you know the processors processing time. Did Nikon and Sony publish the actual CPU spec of Z9 and A1?

@mathphotographer

2 жыл бұрын

No - not that I know. Nikon provided a verbal statement about the sensor read-out time (by the way, good discussions and reviews of the Z9 to be found on dpreview.com) and that includes of course the data processing speed. But camera manufacturers continue to being vague about these datapoints we would love to know.

2016 CMOSIS (Sensor-Manufactur) announced the first BSI CMOS-Sensor with around 48MP with electronic global-shutter (Pseudo-Global-Shutter) . So the disadvantage of the "rolling-shutter effect" was solved, but way too expensive for custom cameras. Panasonic and Sony (2018) did the same btw... The pixel scan-rate of this sensortype is not row by row or progressive. The pixels are scanned all in the same time. So the read-out speed gains a lot. Due to the massive data-traffic, the frames per second count in uncompressed RAW is not so high like competitiors cameras. I ordered two Z9 Bodies for our company, beeing excited to check out, if this is true :-)

An interesting detail - 8000 iso in nikon vs 12800 in sony with the same shutter and aperture. Unless you've changed the light.

@kg22kg

2 жыл бұрын

ISO is not a physical value so it varies between brands.

@schtggl6265

2 жыл бұрын

@@kg22kg Obviously

A readoutspeed of 62ms (Z 7II) compared to only 3ms on the Z9 are a hughe improvement. Looking forward to my new baby early 2022 :D

It's not a real test if you don't have ALL the physical bodies and compare the actual results. Oh and I thought the readout speed means the time to read out alle the single pixels from top to bottom, right? Hmm, the Z9 only has 45 MP, but the A1 has to readout 50MP (much more pixel data). So if I do the math, and downscale the the A1 from 50 to 45MP, it's actually faster than the Z9.

@scb2scb2

2 жыл бұрын

My guess if you balance the extra MP out they will be about the same on the sensor readspeed per pixel. What is unclear is the speed of the cpu's behind them this won't effect this test much but does other things and its a unknown factor at this point.

@godboy159

2 жыл бұрын

You are speaking hypothetically as if the Z9 had the same MP as the A1. Or the A1 had the same MP as the Z9. If this was the case you would be right. But this is reality we are dealing with here and it comes down to how fast the cameras can read the sensor to take an image. Also, the Z9 doesn't read from top to bottom. It's from what I understand reading the entire sensor at the one time. Could be wrong about that though. Do you own an A1?

@wilduntamedphotography1260

2 жыл бұрын

@@godboy159 Own the A1 yes, check my Channel.... Cheers Marco

@godboy159

2 жыл бұрын

@@wilduntamedphotography1260 No thanks.. Just trying to see why you were being nasty. I figured you owned an A1.. Not interested in your content. Take care.

@wilduntamedphotography1260

2 жыл бұрын

@@godboy159 LOL, btw the Z9 does not have a global shutter or readout, what are you dreaming of? The readout works the same in A1, R3 abd Z9... And because the A1 is basically just as quick with more MP tells us a lot 😎 But people like you don't want to hear the truth....

So, where is the Z 9 output?

@SkepticalSteve01

2 жыл бұрын

Roman Barton Give Mathphotographer a chance - his Z9 is undoubtedly somewhere in the supply chain between Japan and Switzerland at the moment. I thought he did very well to devise a demonstration of what can be expected from the Z9’s electronic shutter without actually having one in his studio yet. I expect Nikon will struggle for some while to meet demand for this eagerly anticipated camera.

The Nikon Z9's Low Light Video Performance Assessment & Examples I Would Like to See.

Would be nice if they could come up with a way to control the integration process of the individual photodiodes without having to stop the light exposure and control that on all photodiodes at the same time. But one can only dream right now. But to claim that the camera eliminates the need for a mechanical shutter is still a bit of a stretch especially when it comes to increasing flash sync speeds, if the mechanical shutter can reliably be made faster, which Sony seems to suggest on the A1. However, the question is, have they reduced the life of the mechanical shutter to make it faster relying on the user using it mostly when there is a need to use it with high speed flash photography?

I absolutely HATE the silent electronic shutters; hence why I use the Nikon F2T film SLR which was launched way back in 1971; it’s shutter-curtains and flipping mirror makes loud noises informing me that an exposure has been made; whereas with electronic shutters they are barely audible, and in busy environments you hear nothing. Once I have processed a roll of 35mm film, I scan it using a high resolution scanner, once scanned I can make very good quality 36”x24” prints.

There is something I don't understand. If the Nikon Z9's read out speed is 5 ms how can it achieve 1/32 000 sec exposure time ?

@mikejackson9585

2 жыл бұрын

It's no different than a mechanical shutter which moves at 1/250 can allow for 1/8000. Don't quote me but I believe it is a calculated exposure not an actual.

I like to use M shutter because of the soft bokeh! E shutter makes the bokeh harder.

Nikon does needs to improve their Mirrorless cameras

The read-out time is 3.7 ms.

Shoot ! Just realized you are swiss I knew you had a swiss accent !

All around Nikon Z9 is totally winner. In case if Sony A1 drop, it will damage straight way.

@longrider9551

2 жыл бұрын

and you know this how?

@wilduntamedphotography1260

2 жыл бұрын

BlaBla ....

@lumenspaul

2 жыл бұрын

probably not i expect a lighter body to fair better in a drop test ,although the ibis is locked on z9 so depends although the camera might actually be powered on when you drop it.

Thanks. Great video. But if you want to see the level of rolling shutter on the cameras, you cant use a picture where the fan is level horisontal. This will not give you any rolling shuttes, at the issue with rolling shutter is with movements in a horisontal direction. So the reference image should have been where the blades of the fan was pointing up and down. This would give you a better idea of rolling shutter.

The A1 can go higher that 8000 in electronic, right? how about the z9?

@schtggl6265

2 жыл бұрын

Too

@nyambe

2 жыл бұрын

@@schtggl6265 36000 sorry

So in short image quality sucks, dynamic range on different ISO speeds is a mystery that Nikon chooses not to revile but hype and number of jpg images per second that most of us could care less about is constantly emphasized. Nikon really needs to return to simplicity, quality and value for the money that is the formula for success not the hype.

@godboy159

2 жыл бұрын

How did you get that image quality sucks? And for sports I'm totally interested in the 120fps mode. Everyone uses hype to sell a product. When it gets bad is when something doesn't live up to the hype. Sounds like you will be fine with an older camera. And you definitely seem to not know much about marketing. Nikon needed this camera and it will only get better. Stay salty!

@mikeabc5355

2 жыл бұрын

@@godboy159 It is not sold as sports camera and even for sports it is 120 jpg files not maximum size RAW. It is amazing that the some years back photographers used 3-4 frames per second film cameras and managed to get some outstanding shots. Spray and pray is really not replacement for anticipating moment and clicking the shutter at just right time. Image quality is not what Nikon is pushing you cant even find out what is the dynamic range all that hype is concentrating on is frames per second and supposedly best since sex autofocus.If true grate autofocus would be welcome's but outstanding image quality is even more important and we are not told anything about that.

@markbernhardt6281

2 жыл бұрын

@@godboy159 Don't listen to the illiterate sony fanboy. We know the z9 is 1500 cheaper than the A1 with battery grip. It's value for days.

@w.scholz9705

Жыл бұрын

@@mikeabc5355 " It is amazing that the some years back photographers used 3-4 frames per second film cameras and managed to get some outstanding shots" - You are obviously not a sports photographer ... "Spray and pray is really not replacement for anticipating moment and clicking the shutter at just right time". In sports there is NOT just one "right time" - it is clear to every sports-photographer, that you get far better images with a high frame rate. And ... Compare the "outstanding" images from the days of film with the captures from today and you can easily see a huge difference.

@mikeabc5355

Жыл бұрын

@@w.scholz9705 You are correct I am not sports photographer. Granted very few people need fast frame rate but Nikon ignores dynamic range and photo quality that is important to all photographers and fixated on something that matters to 0.5%.

If Z9 scan speed is faster than A1, why can't it shoot at 30fps compressed raw or high efficiency star? Instead it's capped at 30fps, 45mp jpeg and 20fps, 45mp compressed raw. Interesting quirk if you ask me.

@scb2scb2

2 жыл бұрын

Because you need todo more then just get the image of the sensor (once its in the dram on the sensor) lots of steps before it is stored in your memory card. Somewhere either there is technical or marketing reason for them not todo it.

@mlai2546

2 жыл бұрын

Nikon Z9 processor allocates resources for the EVF dual stream technology and hence is true blackout free without stutter like A1 or R3. Is Sony really 30fps? I thought it depends on lens, auto wb, lossy compressed raw, and several other rules. Also my understanding is that when shooting continuous the evf resolution drops and fps drops to less than 20fps and does not sustain 30fps. It just seems to me Sony is just about specs. Like my Sony Xperia 5 was advertised as the first 4k phone. But guess what, it only displays 4k when viewing pictures. The rest of the time it's 1080p. They don't tell the customers that.

@scb2scb2

2 жыл бұрын

@@mlai2546 How did you decide the A1 is just about specs ? Maybe watch some real reviews like the R3 vs A1 by Jared. The 30fps is real but if a lens can't move fast enough what do you want sony todo all brands will get this issue. The dual stream EVF nikon claims imho is PR bs not sure how sony and canon resolve it but its fine. Yes i own a A1 and i tested getting 30fps is easy if the lens allows to af that fast. Also people (and we will see Z9 owners claim that soon too) should 99% shoot in compressed raw. Uncompressed or even lossless compressed raw is only useful in a very few cases and mostly in single shot mode not bursts.

@mlai2546

2 жыл бұрын

@@scb2scb2 Daniel, go check it for yourself: kzread.info/dash/bejne/Zn-LvLKshMzggtY.html If you don't find this stutter annoying then good for you. I suspect Sony and Canon will eventually adopt the dual stream technology. Nikon's dual stream prevents stutter because it does not need to pause, transfer sensor to CF Express and then resume feeding sensor to EVF. It is working in parallel, both feeding the EVF and CF Express at the same time. Sony's 30fps is compressed lossy. If you want to compare lossy, then you can say Nikon's continuous shooting is 120FPS, that's 4 times more than Sony's. If you are just solely comparing processor power, then Nikon wins in this one as it can record 8k 60fps.

@scb2scb2

2 жыл бұрын

@@mlai2546 I am not sure what is in the video might be because of the hdmi record since that is a extra feed again thats by a 3th chip if you also record the overlay or because they have anti-flicker on (see start of the burst). But i can simply test here myself put the evf in 120hz mode, moved camera about smooth as expected to view, hit burst and just as smooth during the burst. The whole point about lossy confuses me even more since unless i missed something at each speed the sony allows for a better format (30fps and 20fps) then the Z9. The 120fps mode is nice but unless someone convices me its not extracting video frames to get these jpgs. The A1 can do the same you just have todo it at the end of the clip and yes its only 8mp not 11mp because of the 16:9 aspect ratio. Lets see if sony adds the same trick since its nice. The biggest issue with that 120fps mode is there is no clear way to auto-batch or organize them in camera since having say 360 images on a 3second burst doesn't make your life easier. I was hoping they would come up with a smarter way to group and maybe preselect them for transfer. But getting to the dual stream i am just not sure what it effects are. So sensor puts it into the dram on the sensor. Now somehow their is logic in there that takes that now scanned image and feeds both to the expeed 7 (so transfers the same data 2 times to that chip or more likely at different speeds and resolution. The A1 transfers it only 1 time from the dram on the sensor to one of the 2 XR chips. It has to be able to transfer at least some version at 240hz since the EVF supports 240hz mode. We know at least 1 on 8 transfers has to be the full 50mp image since that needs to be processed for the 30fps save. We also know both the A1 and Z9 has to have 120frames inside the Exceed7 or one of the 2 XR chips todo tracking and whitebalance since both do that at 120hz. So in the end i am not saying the Z9 method is not better what i mostly saying we have no idea how the others do it. Clearly nikon felt it was smart to make it press-item that it sends 2 copies from the sensor to the exceed7 while the others seem to only send 1 copy or not we don't know since they have not talked about how they do it and why. As a software and hardware designer by trade i simply can't 100% tell you what makes more sense and why. But i do agree if Z9 method has a huge advantage it will be copied i am just not sure how it compared until i have a Z9 next to my A1 to try it out. Since the Z9 is lower res and lower refresh rate i simply don't know how having the sending 2 copies from the sensor to the cpu helps unless someone really explains what is in these data streams why what happens after they hit the chips. Keep in mind i reacted to this video first as a A1 to simply say to the Z9 owners (that was the goal of the video posted) not to worry about rolling shutter since it won't be a issue as his video clearly shows.

Click Baith...Not a Nikon Z9