NIKON...MIRROR LESS NOW WITH FIRMWARE UPDATE

mhedges

Yep that’s absolutely right. I work in semiconductor manufacturing and its very hard to get major cost savings without die shrink. Not that it never happens but die shrink is always the most significant thing you can do. Since that’s off the table with camera sensors I don’t expect them to become much cheaper over time. And the cost advantage of crop sensor will always be there.

I’m curious myself what kind of defect criteria they impose on sensors. It’s very difficult - near impossible - to make a die the size of a FX sensor without any defects. I assume they do allow a certain number of bad pixels as long they aren’t all next to each other. Is that where the missing pixels between what’s quoted and what you actually get go?

PB_PM

I suspect there is a percentage point that they aim for, and try to avoid large number of bad pixels in an area. The higher the pixel count, the higher the percentage of acceptable dead pixels there will be. That's how manufactures treat LCD monitors typically, so I expect that a similar system is used for sensors. The edges likely get tossed with in camera software, because the edges would be more likely to have issues, particularly if it is from the extreme edge of the wafer.

It's hard to say how bad the sensors really are, because camera makers use algorithms to blend the space dead pixels cover, so that they are never revealed to the end user. The only way to really find out would be to hack the cameras firmware and remove that algorithm.

HankB

I said "Cost of sensors certainly must be declining as a percentage of total cost to manufacture a body, "
The other subassemblies for the bodies and the assembly process itself, as well as distribution costs etc. are certainly going up — probably up faster than the sensor costs.

It would be interesting if we could compare mfg costs of FX vs DX. That would give us the answer as to how much more a body with a 36mm square sensor would be. Though maybe low sales volume would push up the unit cost.

Now here Is an interesting [but misleading] concept: considering how much the potential of all the expensive glass is wasted by the very oblong 36 x 24 format, a square 36 x 36 camera would be cheaper per recorded photon.
[This "statistic" is technically valid, though quite misleading — just like most statistics we are bombarded with each day]

PB_PM

The sensor is easily the most expensive part of any camera body, percentage wise, and that cost is likely on pace with everything else (on a generation to generation cycle), such as materials like plastic or alloys. The process has been highly automated for some time now, so cost savings won't come from that.

The cost of a larger square of silicon would rise the price dramatically, which was the point of my post. Medium format camera are not just more expensive because the brand name, or that the lens mount and assemble on the camera is bigger, it's the silicon inside. Silicon wafers are round, a square die would lead to higher the price per die, since it results in more waste material. If you look, almost every silicon die is rectangular, not square, for that very reason. A die the shape of a modern camera sensor insures more dies per wafer, than a square format would. It's all about efficiency and cost per die.

HankB

Doesn't wafer diameter increase with each newer generation manufacturing facility? Larger wafers would mitigate the effect of die size and shape on waste. Though I have no feel for how frequently new facilities now come on stream in this fairly mature industry, and if we are approaching some practical limit to cost effective ingot and wafer size.

mhedges

It has increased since the early days, but I wouldn't say it increases with each generation. Currently the largest wafer size available is 300mm. There are also quite a few facilities running on 200mm wafers. 300mm equipment is insanely expensive so it's generally only used for the high end, high volume products. There was talk of eventually moving to 450 mm but that appears to have stalled out:

https://www.extremetech.com/computing/242699-450mm-silicon-wafers-arent-happening-time-soon-major-consortium-collapses

WestEndFoto

300mm. ........ now THAT would be large format!

PB_PM

Yeah, and the sensor alone would be $100k.

PB_PM

HankB said:

Doesn't wafer diameter increase with each newer generation manufacturing facility? Larger wafers would mitigate the effect of die size and shape on waste. Though I have no feel for how frequently new facilities now come on stream in this fairly mature industry, and if we are approaching some practical limit to cost effective ingot and wafer size.

Last year Samsung changed the layer process of dies for DRAM (not even enlarging the wafers) at one facility, that alone cost nearly $1 billion, just to give a hint at the type of money going into this system.

rmp

And I have trouble with my budget for cameras.

mhedges

I believe starting price on the newest EUV exposure tools is around $120 million.

Msmoto

I find it fascinating that so many others are really into the mirrorless, yet Nikon is still waiting for who knows what. As I am growing older, and have just about sold my 400/2.8, I want to get into a crop sensor mirrorless by Nikon, but do not have another five years.
Interestingly enough, on my page I see a link to a thread posted several years ago with some images I created in Lightroom, pasting up a Nikon....LOL
https://forum.nikonrumors.com/discussion/3386/nikon-large-sensor-mirrorless-camera-is-a-possibility

OMG, just too much....

Hiker

vtc2002 said:

@hiker I did not perceive your post as being frustrated at all. Compared to some of the comments regarding the potential new lens mount, adapter for f mount lens and lack of direction for the DSLR product line, your comments were pretty tame. I think we all would like to have more information coming from Nikon about the strategy and direction for the mirrorless and DSLR product lines but it is apparent that that's not going to happen. We have some passionate people on the forum and that is a good thing.

Thanks. Much appreciated!! Some will defend Nikon to the death, and that's ok. It's not that I don't enjoy my Nikon, I just don't see them as the future due to their issues. And again, EVERY SINGLE DAY on the Sony A7iii Facebook page, scores of Nikon and Canon people are leaving their beloved Nikon and Canon cameras for Sony, and others to a lesser extent.

BVS

Personally, Sony still seems to have a few too many warts/caveats/footnotes to seriously consider switching yet. However, if Sony eventually gets things like compressed raw working properly and their ergonomic challenges sorted out, and Nikon doesn't come up with a viable alternative, I might have to start questioning my loyalty.

BVS

Hiker said:

Thanks. Much appreciated!! Some will defend Nikon to the death, and that's ok. It's not that I don't enjoy my Nikon, I just don't see them as the future due to their issues. And again, EVERY SINGLE DAY on the Sony A7iii Facebook page, scores of Nikon and Canon people are leaving their beloved Nikon and Canon cameras for Sony, and others to a lesser extent.

You might want to stop reading the Sony A7III Facebook page. Sony has been good at marketing recently and Nikon has not.

HankB

PB_PM said:

Last year Samsung changed the layer process of dies for DRAM (not even enlarging the wafers) at one facility, that alone cost nearly $1 billion, just to give a hint at the type of money going into this system.

Thanks for the insight.

Capt_Spaulding

BVS said:

Hiker said:

Thanks. Much appreciated!! Some will defend Nikon to the death, and that's ok. It's not that I don't enjoy my Nikon, I just don't see them as the future due to their issues. And again, EVERY SINGLE DAY on the Sony A7iii Facebook page, scores of Nikon and Canon people are leaving their beloved Nikon and Canon cameras for Sony, and others to a lesser extent.

You might want to stop reading the Sony A7III Facebook page. Sony has been good at marketing recently and Nikon has not.

Sony produces generally decent products but their forte is marketing them. I've been thinking Nikon should hire Sony to market Nikon's cameras.

mhedges

Sometimes I think a random stranger picked off the street could do a better job with marketing than Nikon does.

sport

mhedges said:

Sometimes I think a random stranger picked off the street could do a better job with marketing than Nikon does.

You summed it up. Although, only releasing one camera a year means that you don't have to do much marketing.

moreorless

mhedges said:

Mmm. The URL says full frame but wouldn't 55-200 imply crop sensor? I don't think I've seen that range on a full frame lens.

I don't think that's impossible, Nikon releasing APSC and FF mirrorless cameras, there seems to be a good market for both. That said it could also be the case that Nikon is playing around with zoom ranges to suit a smaller mirrorless system? a 55-200mm lens is obviously going to be smaller than a 70-300mm one and perhaps could link up with a limited normal zoom? say 24-55mm?

I think its very possible Nikon goes with a two teir focus for mirrorless, a lower cost consumer body with slower/cheaper/small zoom lenses and a higher end body with very fast primes aimed more at the Leica market sold at a premium.

mhedges

Well personally I don't think I would be interested in a 24-55. Just too limited a range. I can't imagine many people would be.

spraynpray

I use a 24-35 and find it a great lens. My most used.

rmp

Over time, I'm finding that the 24-70/2.8 is my most often used lens. I'm also finding I use manual focus and focus-peaking much more often these days. Maybe, I'm just getting older.

sport

I'm with spraynpray, I mostly use the Sigma 24-35 and really don't find it too limiting. 24-55 would really have to offer something big to find a market.

HankB

spraynpray & sport, just curious about 24-35 being your most used lens. Do you heavily crop, and if so, is this with a D850?