Nikon Rumors: Z8 and Z9 Firmware updates in December?

[Cameron T]

To sum it, I am 100% correct. Pre capture does eat in into the buffer.
One might say that I’m still getting the 4 seconds, because pre capture adds the 1 seconds, followed by the next 3 on the shutter.

And I was referring to pre capture RAW with 30FPS RAW, it would be almost impossible to utilize it before the buffer is backed up.

Regarding the question as to why I would keep the shutter pressed for 4 seconds? That’s a different topic. But in short: Conowingo.
The critical fish run takes between 4-6 seconds.

Your original post was still incorrect.

You're more correct here, but but the camera starts dumping as fast as possible to your card once you press the shutter. You could easily do 30 in he/he* for a limited amount of time, but it's always about tradeoffs. You could probably do 18 forever with a fast card.

 

[JoelKlein]

Your original post was still incorrect.

You're more correct here, but but the camera starts dumping as fast as possible to your card once you press the shutter. You could easily do 30 in he/he* for a limited amount of time, but it's always about tradeoffs. You could probably do 18 forever with a fast card.

Ideas on how to maximize the duration, is also another discussion. Best cards, HE*, lower fps. I get it.
I was mainly referring to the hard core physics. “IF” 30fps becomes a reality, and “IF” pre capture gets RAW, then the shooting duration would be almost none.

 

[Cameron T]

Then you'd still have the option of jpg, or shooting at 20fps the whole time manually.

People have gotten these shots for years without all of the fancy tech. It's still doable

 

[Joel N]

The 5 min limit is likely a “just in case” soft cap to prevent possible issue from memory leaks if there is or more importantly overheating with accidental prolong half press etc in camera bag like. Nikon is being careful with fundamentals which I feel is good.

 

[Joel N]

That is possible. Nikon could have more chip real estate inside to unlock as they see fit… maybe they have the global shutter capability too. Who knows?!

Well global shutter is impossible on the z9 or z8 unless the sensor (hw) and the architectures design (hw) is set for that. Even if it is, Nikon won’t be that nice to give it to an existing body. Also sensor is from Sony, what is the chance? It is nil with stamp.

 

[John Navitsky]

The 5 min limit is likely a “just in case” soft cap to prevent possible issue from memory leaks if there is or more importantly overheating with accidental prolong half press etc in camera bag like. Nikon is being careful with fundamentals which I feel is good.

yes, and you want such a protection because the memory likely can wear out. typically this lifespan is much higher than you'd ever counter in real life, even with high usage, but accidentally letting it just hammer the memory for hours straight (until the battery wears out) might actually make a dent in it.

 

[Wes Peterson]

To sum it, I am 100% correct. Pre capture does eat in into the buffer.
One might say that I’m still getting the 4 seconds, because pre capture adds the 1 seconds, followed by the next 3 on the shutter.

And I was referring to pre capture RAW with 30FPS RAW, it would be almost impossible to utilize it before the buffer is backed up.

Regarding the question as to why I would keep the shutter pressed for 4 seconds? That’s a different topic. But in short: Conowingo.
The critical fish run takes between 4-6 seconds.

Conowingo is the last place you would need to use precapture. What do you want to do? shoot the second before the bird even starts its dive? In all honesty if youre shooting even 1 second before the pickup youre wasting your time and buffer. This entire argument is moot because of HE* anyway.

 

[Geoff_G]

Well global shutter is impossible on the z9 or z8 unless the sensor (hw) and the architectures design (hw) is set for that. Even if it is, Nikon won’t be that nice to give it to an existing body. Also sensor is from Sony, what is the chance? It is nil with stamp.

Agreed - there is no chance for a global shutter on the Z8/9. First off the sensor is not designed for that. Secondly, at 45 Mp the amount of data coming off the sensor would require a huge bandwidth increase.

 

[John Navitsky]

Agreed - there is no chance for a global shutter on the Z8/9. First off the sensor is not designed for that. Secondly, at 45 Mp the amount of data coming off the sensor would require a huge bandwidth increase.

to clarify. it’s a large step up from what sony has currently accomplished with the a9iii which does mean it will probably be a while before we see a 45mp global shutter. the bandwidth is the same. 45mp is 45mp. a gs 45mp sensor doesn’t need any more bandwidth to transfer an image from the sensor to the camera vs a 45mp stacked sensor. it may of course support faster reads which in turn may enable more fps which would increase bandwidth but it doesn’t mean that is the case and there is nothing reported with the a9iii that suggests it reads any faster than the z9. iirc, they talk about 120fps af calcs, which is the same as the z9 and that’s a good indication that we can only get the frame off the sensor about 120 times a second i think*.

* even that is fuzzy because it’s likely the af sensors can be accessed asynchronously from the rest of the sensor so you could potentially do af calcs without reading the sensor. however we know the z9 (and a9iii i think) can update the evf at 120fps which further reinforces our idea it can read the sensor that fast but even that isn’t known for sure because it could be doing line skipping or something.

 

[Geoff_G]

to clarify. it’s a large step up from what sony has currently accomplished with the a9iii which does mean it will probably be a while before we see a 45mp global shutter. the bandwidth is the same. 45mp is 45mp. a gs 45mp sensor doesn’t need any more bandwidth to transfer an image from the sensor to the camera vs a 45mp stacked sensor. it may of course support faster reads which in turn may enable more fps which would increase bandwidth but it doesn’t mean that is the case and there is nothing reported with the a9iii that suggests it reads any faster than the z9. iirc, they talk about 120fps af calcs, which is the same as the z9 and that’s a good indication that we can only get the frame off the sensor about 120 times a second i think.

A global shutter reads all of the data at the same time and has to process all of that before the next shot. The Z9 sends data off the sensor sequentially and not all at once. There is a big difference in necessary bandwidth to process it. I do agree this is transparent to the user but I seriously doubt a 45MP sensor with Global shutter is in the near future.

 

[John Navitsky]

A global shutter reads all of the data at the same time and has to process all of that before the next shot. The Z9 sends data off the sensor sequentially and not all at once. There is a big difference in necessary bandwidth to process it. I do agree this is transparent to the user but I seriously doubt a 45MP sensor with Global shutter is in the near future.

i think it might be better to say a gs CAPTURES all the data at the same time. i think they use some sort of cache memory built into the sensor. i don’t think it inherently says anything about how the data gets from the sensor to the rest of the system

but we agree, i doubt we’ll see a 45mp gs soon

 

[Geoff_G]

i think it might be better to say a gs CAPTURES all the data at the same time. i think they use some sort of cache memory built into the sensor. i don’t think it inherently says anything about how the data gets from the sensor to the rest of the system

but we agree, i doubt we’ll see a 45mp gs soon

Yes, it captures all the data at the same time but then it has to process all that data. Compared to reading part of the data vs time.....that's a big difference in the bandwidth (notably for the sensor).

 

[John Navitsky]

Yes, it captures all the data at the same time but then it has to process all that data. Compared to reading part of the data vs time.....that's a big difference in the bandwidth (notably for the sensor).

you need to _read_ the data, but you need to read the data with the stacked sensor as well.

i'm not sure the rest of the system really sees a difference if you're reading directly from the sensor or from the sensor cache memory. and i also am not sure there is a difference from the sensor's perspective either.

 

[Nimi]

Well global shutter is impossible on the z9 or z8 unless the sensor (hw) and the architectures design (hw) is set for that. Even if it is, Nikon won’t be that nice to give it to an existing body. Also sensor is from Sony, what is the chance? It is nil with stamp.

In astronomy world there is at least one sensor that can do both.

The issue is noise and the resulting lower dynamic range. That's why the native ISO on the A9iii is 250 and that's why no YouTuber nor the company released dynamic range data. It won't be great (maybe around 10 stops?). The noise is the result of pixel well depth which is shallow in high MP sensors. That's why most cinema cameras abandoned global shutters in favor of fast-readout sensors and rolling shutters and kept the MP low even as they adopted full frame sensors.

 

[Wes Peterson]

In astronomy world there is at least one sensor that can do both.

The issue is noise and the resulting lower dynamic range. That's why the native ISO on the A9iii is 250 and that's why no YouTuber nor the company released dynamic range data. It won't be great (maybe around 10 stops?). The noise is the result of pixel well depth which is shallow in high MP sensors. That's why most cinema cameras abandoned global shutters in favor of fast-readout sensors and rolling shutters and kept the MP low even as they adopted full frame sensors.

That is rather interesting. ALL of the youtubers who went to the event for the release got to keep the camera they tested. Fro, even took his to africa. No word from any of them about image quality....

 

[John Navitsky]

That is rather interesting. ALL of the youtubers who went to the event for the release got to keep the camera they tested. Fro, even took his to africa. No word from any of them about image quality....

image quality will be fine.... when comparing apples to apples and when you're inside of the native iso range.

ie, 250 on the a9iii vs 250 on the a1 or z9 (more or less).

but you loose out on both ends of the iso range. the a9iii isn't going to compare well at iso 250 to the z9 at iso 64. and likewise the z9 has more iso range on the top end.

so basically, if you have enough light, it's going to look fine.

as far as i can tell, none of the youtubers are really exercising the actual global shutter feature of the a9iii. yah, sure, you can take it on safari and it'll take pictures of lions at iso 1600 just fine.

it'll take great pictures of fast moving subjects just fine.

i think most of these folks are confusing the next gen af performance for the global shutter features. because the a9iii isn't just a gs camera, it's also the first cut of their next gen processor/sw. which is a feature of the a9iii of course, but it's not enabled by the global shutter particularly.

to show off the a9iii we need to see some interesting strobe work and some video stuff.

and then we need to see stuff like pushing the high iso capability to see how much we loose as a trade-off.

 

[Cameron T]

That is rather interesting. ALL of the youtubers who went to the event for the release got to keep the camera they tested. Fro, even took his to africa. No word from any of them about image quality....

If it had great iso, Sony would be crowing about it. They aren't.

It's not that it's a bad camera by any means, but it'll be noisier and less general purpose than even a stacked sensor camera (z9/8, which had slightly less Dr than a non stacked sensor eg the z7)

 

[Geoff_G]

you need to _read_ the data, but you need to read the data with the stacked sensor as well.

i'm not sure the rest of the system really sees a difference if you're reading directly from the sensor or from the sensor cache memory. and i also am not sure there is a difference from the sensor's perspective either.

I'm still covinced the sensor has to have more bandwidth with a Global Shutter than a non global shutter. It has to handle all of data at once vs sequentially reading data for a non global shutter.

But let's drop this....we both agree a global shutter is not going to happen with a firmware update (which is what this thread is about.) And we both agree there isn't going to be a 45mp global shutter cam anytime soon...

 

[John Navitsky]

we both agree a global shutter is not going to happen with a firmware update (which is what this thread is about.) And we both agree there isn't going to be a 45mp global shutter cam anytime soon...

agreed!

 

[Nimi]

image quality will be fine.... when comparing apples to apples and when you're inside of the native iso range.

ie, 250 on the a9iii vs 250 on the a1 or z9 (more or less).

but you loose out on both ends of the iso range. the a9iii isn't going to compare well at iso 250 to the z9 at iso 64. and likewise the z9 has more iso range on the top end.

so basically, if you have enough light, it's going to look fine.

as far as i can tell, none of the youtubers are really exercising the actual global shutter feature of the a9iii. yah, sure, you can take it on safari and it'll take pictures of lions at iso 1600 just fine.

it'll take great pictures of fast moving subjects just fine.

i think most of these folks are confusing the next gen af performance for the global shutter features. because the a9iii isn't just a gs camera, it's also the first cut of their next gen processor/sw. which is a feature of the a9iii of course, but it's not enabled by the global shutter particularly.

to show off the a9iii we need to see some interesting strobe work and some video stuff.

and then we need to see stuff like pushing the high iso capability to see how much we loose as a trade-off.

Reading into the press release of having "no worse" IQ than the 9ii, and taking it a step further and adding "at same ISO," that's 9.85. That's a full stop lower than the 9ii at its native ISO of 100 and about two below A1/Z9 et al. About the same as an MFT sensor although the much bigger pixels may eliminate the noise issue inherent to small sensors.

As we've known, it's an incredible specialist camera for sports photographers. It won't be a good video camera, presumably.

I'm keeping my name on the wait-list for now.

 

[John Navitsky]

As we've known, it's an incredible specialist camera for sports photographers. It won't be a good video camera, presumably.

other than the lack of rolling shutter

 

[Nimi]

other than the lack of rolling shutter

Really not an issue at 24-30fps and 1/50th of a second. Motion blur for the most part hides the rolling shutter effect. For specialized applications, yes. That's why the most expensive cinema cameras continue to use rolling shutters, including the newest Sony. Arri has one with a full-sensor capture, so global in that respect, but it uses a traditional mirror shutter.

 

[Michael Hall]

I just hope Pre capture raw at 1/15 is possible so handheld lightning shot are possible again, 1/60 is just too quick.

 

[Zof9]

I reckon tomorrow is the last day they can announce new firmware before Chrimbo???
Staff holidays over the festive period and all that...

 

[Wes Peterson]

I reckon tomorrow is the last day they can announce new firmware before Chrimbo???
Staff holidays over the festive period and all that...

You forget they were shipping Z9s for people to have on dec 23rd