[biz-engineer]

I'm sure all manufacturers do that, but it's not the kind of noise we talk about in terms of software noise reduction.

Exactly. They can add a low pass filter to remove glitches that appear with the higher speed design (which is not present at lower clock speeds), or use an buffer with auto-zero in front of the ADC, that does not remove any noise coming from the pixels such as shot noise or white pixel induced by dark current over long exposure. In fact it's a way to cope with the increase noise due to the requirements of faster clocking for the higher frame rate. Guess why the Sony A9 showed lower DR, without filtering substrate induced glitches, the DR would be worse.

---

Last edited by biz-engineer; 06-18-2017 at 08:17 AM.

[normhead]

This new Sony sensor hasn't as far as I know hasn't been picked up by anyone else. Which begs the question, has it been tested and rejected by others? We have had instances like the delay before the K-3 when the original sensor that was supposed to be supplied by Sony was rejected by Pentax engineers and they had to wait a year for something acceptable. The simple fact is, not every Sony designed sensor has been a winner. So is the lack of this sensor adoption because Sony won't sell it, or because none else wants it? It could go either way.

If nobody else wants it, you sure aren't going to see that in a Sony press release, same as Sony never issued press release saying "The 24 MP Sony sensor supplied to Pentax had so much cross talk at the pixel level, Pentax engineers rejected it." So, ya, it's definte, if there is a problem with this technology that is keeping others from adopting it, Sony isn't going to tell you.

But they might say..."this technology is available only on Sony cameras.". That's the nature of press releases. You make a negative into a selling point.

Marketers are under no obligation to say "exclusive technology" might be technology nobody else wants.

Needless to say, this is all speculation (well except for the dud Sony sensor they tried to sell Pentax for the K-3, that is a matter of public record.) I have no more inside information than the "Sony is the greatest thing since sliced bread" crowd.

[Kunzite]

Sometimes you don't have to be a genius to figure these things out. You just have to listen to the market speak and observe the results.

Or, you have to take astrophotos with the K-5 and see for yourself how the stars are not eaten by some overzealous NR algorithm.
Winder is wrong in assuming that Pentax can't possibly be better than Sony in this regard, to the point of not even (properly) searching information on this matter:
"In August of 2016, Sony made firmware changes to their two flagship a7 series cameras (the a7RII and a7SII). The update (3.30 on the a7RII and 2.10 on the a7SII) was supposed to include improvements for radio controlled lighting and overall camera stability and temperature control. But hidden in these improvements is a change that affects the image quality when shooting long exposures, particularly astrophotos. The problem has been dubbed “Star Eater” by others in the astro community."
The Star Eater Issue: Why I No Longer Recommend Sony Cameras for Astrophotography

So it's something that the camera manufacturer can control. Just like applying lossy compression or no compression to the RAW files.

[normhead]

Or, you have to take astrophotos with the K-5 and see for yourself how the stars are not eaten by some overzealous NR algorithm.
Winder is wrong in assuming that Pentax can't possibly be better than Sony in this regard, to the point of not even (properly) searching information on this matter:
"In August of 2016, Sony made firmware changes to their two flagship a7 series cameras (the a7RII and a7SII). The update (3.30 on the a7RII and 2.10 on the a7SII) was supposed to include improvements for radio controlled lighting and overall camera stability and temperature control. But hidden in these improvements is a change that affects the image quality when shooting long exposures, particularly astrophotos. The problem has been dubbed “Star Eater” by others in the astro community."
The Star Eater Issue: Why I No Longer Recommend Sony Cameras for Astrophotography

So it's something that the camera manufacturer can control. Just like applying lossy compression or no compression to the RAW files.

My wife Tessfully has taken lots of star photos with her K-5. That's what first caught my attention.

[Winder]

OK so there's a quote from your posted PDF. Pretty primitive stuff they are talking about. There is absolutely nothing in here to support the hypothesis that the A9 "star eater"noise reduction, is anything but a processing add on different from on sensor processing used by anyone who uses Sony sensors. The point of the "star eater" discussion is that Sony has added another level of noise reduction beyond that, that can't tell the difference between noise and stars. Now it could be that an add on , or it could be that this new sensor technology they have employed is itself the culprit. We don't know. We just know it sucks.

In either case, it has nothing to do with the K-5 or any other previous camera system and the noise reduction they employ internally, because those systems don't "eat stars". This is some new Sony creation. You're in error in stating it is similar to noise reduction protocols that came before.

Sometimes you don't have to be a genius to figure these things out. You just have to listen to the market speak and observe the results.

It's still unclear to me whether this is the technology that saves Sony, or the technology that kills it. The jury is still out. I'm not sure many of the people willing to pay this much money for a camera will put up with having the decision on how much clarity they will sacrifice to noise reduction taken out of their hands, and I'm definitely opposed to any lossy raw file export. That simply defeats the purpose of raw.

But the camera should appeal to jpeg shooters who will look at lossy raw as superior to what they use now.

Nothing I have talked about has anything to do with the "star eating" problem that the A7 & A9 cameras have. My comment was to "Doctored RAW files". The Pentax K-5 did indeed apply smoothing (NR) to RAW files which is why it did so well at high ISO.

[normhead]

Nothing I have talked about has anything to do with the "star eating" problem that the A7 & A9 cameras have. My comment was to "Doctored RAW files". The Pentax K-5 did indeed apply smoothing (NR) to RAW files which is why it did so well at high ISO.

SO what you are saying is the this is different from what everyone else does? Or what exactly is your point?

The consensus at the time was not that Pentax was doing something no one else did, just that Pentax did it better, leading to higher test scores on DxO and other sites.

The original point, that Sony pushed it's NR so aggressively that it's started to leave important parts of reality was the topic of discussion. At that point noise reduction isn't really noise reduction any more, it's a filter.

[Sliver-Surfer]

Nothing I have talked about has anything to do with the "star eating" problem that the A7 & A9 cameras have. My comment was to "Doctored RAW files". The Pentax K-5 did indeed apply smoothing (NR) to RAW files which is why it did so well at high ISO.

My K5 definitely likes to eat the reflections in compound eyes when I am shooting macro.

[normhead]

My K5 definitely likes to eat the reflections in compound eyes when I am shooting macro.

Well that settles it then, the A9 is a good camera, maybe even better than a K-5.

We'll have to see if in 5 years it's held in as high esteem as K-5s are today.

[Sliver-Surfer]

Sony a9 Sensor One of the Best Ever Tested, DxOMark Finds




---------- Post added 06-19-17 at 05:17 PM ----------

Sony Firmware Updates Solve a9 Heat Warning and ‘Star Eater’ Issue

[normhead]

My comment after looking over the scores was, not even top 10. Funny how some folks see one thing, some see another. I do think it may be the best 24 MP sensor, even if it's not really close to the larger resolution sensors and far behind any medium format camera.

Nope, that honour goes to the 5 year old Nikon D600.

It did somehow tie the Sony Set Alpha 99 at 42 MP. What's with that?

If I hadn't already checked it out to see where it stood you might have been able to deceive me with that post.

It is however the best action camera a quick glance, oops. no, that would be the Nikon D750, but not by much.

[Rondec]

Sony a9 Sensor One of the Best Ever Tested, DxOMark Finds




---------- Post added 06-19-17 at 05:17 PM ----------

Sony Firmware Updates Solve a9 Heat Warning and ‘Star Eater’ Issue

Hmmm. The K-1 which uses a relatively ancient sensor scores a 96. "One of the best ever tested," sounds like some spin to clarify why the A9 is so far behind older (and cheaper) cameras like the D810, K1, A7r II, etc.

[Sliver-Surfer]

DXOMark "The Sony a9’s sensor ranks as one of the best we’ve tested at base ISO, with consistently good scores for dynamic range and color depth, and exceptional results for noise. At those crucial mid to high ISO sensitivities for action photographers, image quality is also excellent and on par with flagship Nikon and Canon sports DSLRs."

[Winder]

You Tube

You Tube

[JohnBee]

Hmmm. The K-1 which uses a relatively ancient sensor scores a 96. "One of the best ever tested," sounds like some spin to clarify why the A9 is so far behind older (and cheaper) cameras like the D810, K1, A7r II, etc.

Isn't the K-1 sensor the same as the one used in D810?

If so, I'd say their has been little progress made since then in terms of IQ characteristics(measurements), in that it would see like the only real changes since then has been in resolution. Whereas the remainder of the attributes seem as though they're being redistributed to favor specific characteristics. ie, trading DR for speed, color depth for DR etc etc.

Whatever the case, the measurements between those sensors(36-42mp) seem so close that I'd venture it would be very hard to distinguish between them in real world applications.

[surfar]

Youtu.be

Youtu.be

---------- Post added 06-20-17 at 09:17 AM ----------

Youtu.be

Youtu.be