[ElJamoquio]

lol have you seen the sensor performance?

'lol' ?

Yes, the SNR is the same. ... so, are you saying I'm right?

And if there is no difference in noise between high and low MP, why is the Nikon D4 so low? Meh

. The D4 has the same performance as the D600 and the D800. I think that's pretty remarkable because I believe the D4's chip was a Nikon chip, right?

In any case, the D4/D600/D800 is just another in a long list of examples... there's no improvement to lower resolution other than frame rate and storage size.

---------- Post added 04-13-14 at 01:32 PM ----------

IF the sensors have the same DR and S/N ratio. More mega-pickles is better if all else is equal.

That's true. It's possible the 12MP sensor is a new generation sensor. Of course, there hasn't been an improvement in SNR since the K-5 so I'm not holding my breath.

If the 'new + improved' sensor part is true, then the reason it would have better performance would be that it's a newer design, not that it has lower megapixels. If that's true then Pentax should spend the $ to be the first to introduce the 40MP version of the chip.

---

Last edited by ElJamoquio; 04-13-2014 at 04:03 PM.

[Winder]

If the 'new + improved' sensor part is true, then the reason it would have better performance would be that it's a newer design, not that it has lower megapixels.

Did anyone say it would be better simply because it has fewer MPs?

[ElJamoquio]

Did anyone say it would be better simply because it has fewer MPs?

Several people said they'd like less MP and better SNR. That's demonstrably wrong.

No one has seen that the 12MP sensor has any better than the 36MP, FWIW. We've seen some pics flying around but nothing head-to-head at equal (measured) ISO. To my knowledge there's nothing from which we can conclude the 12MP has better performance than the 36MP.

[DSims]

Several people said they'd like less MP and better SNR. That's demonstrably wrong.

No one has seen that the 12MP sensor has any better than the 36MP, FWIW. We've seen some pics flying around but nothing head-to-head at equal (measured) ISO. To my knowledge there's nothing from which we can conclude the 12MP has better performance than the 36MP.

Now you're venturing into nonsense territory - saying there's no conclusive evidence.


A sharp 12MP photo is better than a blurry 36MP one - it's that simple. Plus what twitch said. And what others said. Plus a few things we didn't even think of.

[Na Horuk]

No one has seen that the 12MP sensor has any better than the 36MP

Right. Except the videos and reports of its performance. And why would Sony go out of their make a 12MP sensor if the 36MP one is just so much better? They already have two A7 models, why would they make a third with 12MP if it is so much inferior? Clearly its not the same wafer

It's for high FPS, not for noise.

That is part of it, but so many people talk about the D4's noise performance.

I think the problem in this thread a) is that some people are talking about SNR per pixel and others are talking about absolute detail/noise in a given photo.
and b) some people think the new 12MP Sony sensor is the exact same as the other sensors we have seen in the past, just lower MP

But I do agree with the niche niche thing. This is why I said that a 12MP Pentax camera shouldn't be the "flagship", but it would be a great addition and something I would be interested in.

[ElJamoquio]

Right. Except the videos and reports of its performance.

You could say the same about the Fuji X100, and it had cruddy performance when it was actually analyzed.

And why would Sony go out of their make a 12MP sensor if the 36MP one is just so much better?

1) Video. It could have a framerate 3x as fast.

2) People who are misinformed about SNR across different resolutions.

They already have two A7 models, why would they make a third with 12MP if it is so much inferior? Clearly its not the same wafer

The 24 and the 36 have the same performance. The 12 might have better. Might. But that's speculation.

That is part of it, but so many people talk about the D4's noise performance.

'People Talk' about a lot of things. The D4's performance is equal to the 36 at equal magnification.

I think the problem in this thread a) is that some people are talking about SNR per pixel and others are talking about absolute detail/noise in a given photo.
and b) some people think the new 12MP Sony sensor is the exact same as the other sensors we have seen in the past, just lower MP

Indeed. I've asked a couple of times for workflows in which SNR per pixel is relevant, and I'm still waiting for an answer.

---------- Post added 04-13-14 at 04:24 PM ----------

Now you're venturing into nonsense territory - saying there's no conclusive evidence.

I'm annoying many people by telling them things they don't want to hear.

There's no reliable evidence on a sensor that doesn't exist in the marketplace yet, and hasn't been tested by, e.g., DxO. Sony could be pulling a Fuji for all we know.

A sharp 12MP photo is better than a blurry 36MP one

All else being equal, a picture made with 36MP will be better than the picture made with 12 MP. Every time. People believe otherwise because they magnify the 36MP picture 3x as much as the 12MP one.

[DSims]

All else being equal, a picture made with 36MP will be better than the picture made with 12 MP. Every time. People believe otherwise because they magnify the 36MP picture 3x as much as the 12MP one.

Ha Ha. I'm not sure if this is getting funny or annoying. You're acting ignorant.


The scenarios are so common you already know them. It's any time you need a sharp, moderate (or lower) resolution photo, in all kinds of conditions. Especially if you need to publish it quickly.



Often you're avoiding motion blur - either from subject movement, or because of low light, or both. SR is of no help in many of these scenarios.

Journalism. Sports. Events and more.



But in each case, it allows you to get the shot when you otherwise wouldn't. So studio comparisons are irrelevant.

[normhead]

It's for high FPS, not for noise.

What does Nikon say?

Nikon flagship ISO performance
Noise-free shooting from ISO 100 to 25,600

Photographers go where their assignments take them, and this often means working in the high-contrast glare of the midday sun, the low-light at dawn or the poorly illuminated interiors of a sporting venue. Nikon flagship D-SLRs have long been recognized by professionals for their clean performance at high ISO settings, and D4S takes that performance to the next level. With a specially designed 16.2MP FX-format CMOS image sensor engineered to collect more light and render higher image quality plus the power of EXPEED 4, you can shoot sharp photos from ISO 100 all the way up to 25,600. The ISO range can be extended to a mind-boggling ISO 409,600 equivalent at Hi 4 for overcoming the most challenging lighting conditions. Enjoy great flexibility in your approach to fulfilling challenging assignments.

D4S Nikon Digital Camera | Flagship HDSLR Camera from Nikon

So unless you're saying that a D800 has noise free shooting up to 25,600... I'm not really sure what you're saying. It's odd having to quote the Nikon website, a place anyone can go to check their facts.

From sonyalpharumors.com

A7s characteristics
ISO 50-409600
extreme high dynamic range and color fidelity (So Sony says without to share numbers yet)
4K recording with no pixel binning
4K recording on external records only (internal recording only Full HD)
4:2:2 color sampling with external recorder
The “S” stays for “sensivity
Price will be announced later and release in “Summer”. Some other sites said price should be $1,699. But take this with a huge grain of salt!

http://www.sonyalpharumors.com/sony-a7s-announced/

I'm not sure what the issue is with larger sensors that collect more photons having higher sensitivity than what can be attained with smaller sensors would be? I think the problem some people may be having is they equate absolute resolution with highest IQ, as if the two are somehow synonymous.

[DSims]

It's odd having to quote the Nikon website, a place anyone can go to check their facts.

+1

[oxidized]

The noise performance is no better than 36MP or 24MP if you post-process.

I have no idea how in the world you can say this when you have never seen raw files coming out of this camera before. You are also ignoring the dynamic range advantages of such sensors. (And no I dont think that 12MP will be enough for my kind of photography so I am not one the proponents "for" this sensor)

[johnmflores]

Niche*Niche = 2nd order niche, which is a bad idea, which is what I was referring to. Niche is OK. Double niche is not.

A niche of a niche has worked quite well for GoPro, Red, BlackMagic, etc...

[normhead]

Yeah I'm saying the D800 is noise free at 25,600

Whatever the marketing says, the D4S and D800E are near as identical in SNR with the D800E with an edge at low ISo and the D4S < 1/10 stop edge at high ISO. However the D4S shoots at 11 FPS and the D800 at 4 FPS.

And you're going to tell us how you know this?

[normhead]

DXO has both sensors

So does Imaging Resource...



The D4s image is sharper and has more Dynamic Range... at 25,600 ISO you can believe the images, or believe the frauds.

[DSims]

Whatever the marketing says, the D4S and D800E are near as identical in SNR with the D800E with an edge at low ISo and the D4S < 1/10 stop edge at high ISO. However the D4S shoots at 11 FPS and the D800 at 4 FPS.

Not only would the 11 FPS help, but you might miss a few shots with the D800E because it can't go as high in ISO.


But I just looked at DPR's preview of the D4S with the low light test scene. The D4S appears to be about a stop better than the D800E would be at the same resolution, and about a 1 1/2 stops better on a per-pixel basis. This is based on looking at the RAW results, but the JPG results also back this up (assuming Nikon optimized JPG reasonably well on both cameras - and why wouldn't they?). The fact that the image gets blurred significantly more indicates that there indeed must be more noise to cover up on the D800E's ISO 25,600 image. And indeed, the JPGs are exactly what matters in many of these shooting scenarios.


So what are you basing this statement about SNR on? And what is its relevance? Do you think it's more important than observable differences? And if so, could you explain why? I'd like to hear your thoughts.

[Rondec]

So does Imaging Resource...



The D4s image is sharper and has more Dynamic Range... at 25,600 ISO you can believe the images, or believe the frauds.

SNR for the two cameras is equivalent. The D800 is a lot better at low iso, however the D4s does hold on to dynamic range better at high iso, such that it is about one stop better than the D800 over iso 12,800. Of course, these are Nikon isos we are talking about, so over stated by about a half stop, for what that's worth.

I still would prefer the D800 for what I shoot and believe that if you shoot RAW and process, you wouldn't see huge differences between these two sensors.

And yes, I do believe DXO Mark. Their measurements seem to jive with what I see in real life shooting.