Do you think Canon will respond to the Nikon D800?

gjl711 wrote in post #16738262
I'm not sure I would classify the 500nm wafer process as slightly out of date. It is quite old being popular about the time the Pentium was first introduced in the mid-1990s (maybe as early as 1993?) In technology terms i would classify it a Jurassic.

Made a correction there for you.
If 150nm* is Jurassic, 500nm* is positively Cambrian...

*nm = my (million years) See what I did there?

I had meant to say 500. I thought it was 150 but had to look it up first, found it was 500 but forgot to change it. in any case, it is very old indeed.

As a semiconductor pro I find it very admirable that Canon uses an archaic process and is still competitive with the best and brightest of today's processes... just imagine AMD still using Athlon 64 5000+ against Intel's 4th gen i7...

This, of course, is not an excuse.

Canon's main advantage at this point is cost control. Their sensors are probably 3-5 times cheaper to make than Sony's, and, keep in mind that Nikon is outsourcing them from Sony which means they are paying even more $$$.

End result in a slumping economy and a shrinking industry is, that Canon is pretty much the only company that makes money.

I'd be a LOT more worried about Nikon's long term viability than Canon's at this point.
Especially after Sony threw the a7(r) wrench into Nikon's plans to erode Canon's market share.
Now there's no need to change systems (CAN -> NIK). Just buy the alpha body and keep using your EF lenses.

If I were to make a 3-5 year prediction, I'd bet that Nikon's position will be a lot worse than it is now.
While they make great cameras using mostly others' sensors, and their lens line is equivalent to Canon's, they are making no inroads towards the INEVITABILITY of still/moving image merge.
If you think that video has no place in dSLRs because you are a purist, well... there are still people using gramophones, Amigas and fax machines somewhere I'm sure...

Technology moves on, keep up or get off.

kin2son wrote in post #16737888
Don't give credit to Nikon where credit isn't due

Nikon can't and doesn't make sensor, they are just re-badging Sony's, they are almost as sad as Pentax.

Nothing wrong with using some other sensor...it has obviously helped Nikon greatly...and Canon making their own seems to have stalled.

Sometimes making all your components in-house is great, many times you just cannot keep up with everything in-house, which is what we are seeing with Canon.

I see nothing sad about this move by Nikon...in fact I see it quite brilliant.

kin2son wrote in post #16737913
Well everyone wants better sensor....but in what way?

D800 has better DR @ base ISO and higher MP, but it has worse ISO performance.

Also because of the high MP, FPS is greatly suffered.

Personally I have minimum to no use of the extra DR at ISO100-400.

Basically one has to choose what's important for their type of photography, weight out all the pros and cons, and pick the best one for his/her job.

Again, you can't have it all, and there's always something better just around the corner

Actually if you dow sample the D800e to the same pixels as the 5d3, they produce extremely close results at high ISO. Much closer than people think.

We would probably not see these issues if Canon did produce the sensors using 150nm (0.15u) geometry. Most probably, their 0.18u fab is good enough, even if it might not have the required capacity.

But one thing to note when discussing how ancient different technologies are - it's basically only high-end processors, RAM and flash that is using really fine geometries. The fabs are so expensive that it almost takes the economy of a smaller country just to finance one state-of-the-art fab.

The end result is that most companies can't afford own fab capacity anymore. They either have to close down, or buy production capacity from the few remaining companies.

A problem here when buying capacity is that sensors mixes analog, digital and optics at the same time. So even if CMOS processes are mature, I don't think the fab owner can't just send a document with the requirements and then wait to receive masks and be able to start producing. Each test production have very long lead times, so it can take a very long time to figure out what tweaks that are needed to create non-standard chips.

Hogloff wrote in post #16738413
Actually if you dow sample the D800e to the same pixels as the 5d3, they produce extremely close results at high ISO. Much closer than people think.

The debate about high ISO performace is mostly centered around people (incorrectly) thinking that Canon is explicitly optimizing for high ISO.

Both Sony and Canon are able to collect a quite decent amount of photons. Sony manages to digitize with more usable bits, so Sony comes out ahead at low ISO. At high ISO, Canon isn't hurt in the same way by the limitations of the read-out electronics, so they end up quite close.

Another thing people don't think about is that we have to look at IQ per mm2 of sensor area, not IQ for individual pixels. A smaller pixel is allowed to have less IQ because it will be projected smaller in the final print. So small but noisy pixels are ok, as long as the noise is random and symmetric around the ideal value, and the noisy pixels aren't clipping without the full dynamic range being possible to capture.

kin2son wrote in post #16738306
That's a terrible setup for airshow.

Why?

Radders wrote in post #16738468
Why?

I wouldn't call it terrible, but there's nothing it can do that the 5D III - 100-400L combo can't (for less money).

kin2son wrote in post #16737888
Don't give credit to Nikon where credit isn't due

Nikon can't and doesn't make sensor, they are just re-badging Sony's, they are almost as sad as Pentax.

Nikon designed the D4 sensor and some other ones I can't remember.

Radders wrote in post #16738468
Why?

Seems like a good plan to me, but kin2son is a seasoned airshow expert, so I guess he'd know.

I just think we put way too much emphasis on technology when we should be focusing on technique, composition, etc...

Today's camera are all very capable. Pick what fits your need. That's. It just a freaking piece of hardware. Its just a tool. It won't take the photo for you. It won't make you a better photograph. I certainly dislike pixel peeper that comes around and talk about details that are insignificant to overall end result.

I think the 5DIII is a good answer to the D800. Somethings the 5DIII does better and somethings the D800 does better. But damn ... the amount of information you are able to pull out from a RAW D800 image PP is incredible ... blows the 5DIII out of the water. I am still not sure if it is simply better DR of the actual Nikon sensor or it if is the way the firmware is setup on the Canon sensor. I still prefer the 5DIII over the D800 if I were to have only 1 camera, but would certainly love an improvement in DR of the Canon sensor!

David Arbogast wrote in post #16738557
Seems like a good plan to me, but kin2son is a seasoned airshow expert, so I guess he'd know.

Even so, I'd still like to know why. To me on paper it sounds like a great set up? Never used either lens or camera so would like to know why it wouldn't be worth going down that route.

Currently use a 60D + 50-500, which to me is brilliant at the moment...

fogboundturtle wrote in post #16738558
I just think we put way too much emphasis on technology when we should be focusing on technique, composition, etc...

Today's camera are all very capable. Pick what fits your need. That's. It just a freaking piece of hardware. Its just a tool. It won't take the photo for you. It won't make you a better photograph. I certainly dislike pixel peeper that comes around and talk about details that are insignificant to overall end result.

Why did you purchase the 5DIII if it would not make you a better photographer?