p.29 #1 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
snapsy wrote:
Because the extra two bits would just oversample noise and thus waste image space, despite how clean the Sony FF sensors are.
If Sony are concerned with wasting space why do uncompressed RAWs use 16 bits per pixel instead of 14?
I think there's some weird technical reason that means they have to run the ADC in 12 bit mode some circumstances (like when the exposure time is not known in advance).
n.b. if you shoot compressed RAW you get at most 11-bits per pixel so the 12-bit drop by itself has no effect.
edit actually you get at most 13-bits per pixel with compressed RAW, so you could detect the 12-bit drop, in theory.
Look at the right side where it talks about 16-bit processing and 14 bit raw files. Asterisks at the bottom.
Why is this a thing? What is the technical rationale as to why data must be thrown away when processing the least time-sensitive (i.e. not 10fps fast) shooting operations. I don't understand why it is even designed this way, at all. If I'm shooting super long exposures, why can I not have the most data that is possible?
Also, I'm assuming that Sony continues to offer true uncompressed raw files that don't even have any lossless compression thus ensuring lots of wasted storage space because they took the customer demand of uncompressed raw to it's most literal sense. ...Show more →
I wonder if this only applies to compressed raw files though, they didn't make that clear, I hope that is the case!
p.29 #4 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
Matt Grum wrote:
If Sony are concerned with wasting space why do uncompressed RAWs use 16 bits per pixel instead of 14?
I think there's some weird technical reason that means they have to run the ADC in 12 bit mode some circumstances (like when the exposure time is not known in advance).
n.b. if you shoot compressed RAW you get at most 11-bits per pixel so the 12-bit drop by itself has no effect.
Because their engineers likely didn't expect the pushback from reviewers and photographers over the IQ compromises of their compression algorithms and thus didn't have any logic in their ASIC to convert the uncompressed files to 14-bit and the general-purpose CPU wouldn't be fast enough to do so either.
That was pretty much to be expected. Some Sony artisans were saying in the launch videos "immediately noticed better DR" but that feel can come from number of things like changed metering, live view histogram calibration or EVF brightness changes that may cause the testers expose RAW files more right.
I'm personally looking forward to test if I can get good ETTR exposures with the new Highlight metering and Exposure Standard Adjustment that can adjust exposure in the 1/6 EV steps for each metering mode.
p.29 #10 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
Matt Grum wrote:
If Sony are concerned with wasting space why do uncompressed RAWs use 16 bits per pixel instead of 14?
I think there's some weird technical reason that means they have to run the ADC in 12 bit mode some circumstances (like when the exposure time is not known in advance).
n.b. if you shoot compressed RAW you get at most 11-bits per pixel so the 12-bit drop by itself has no effect.
This indicate long exposure noise still high with Sony, it make no engineer sense to have 14 bits file with high noise. As for continues shooting, the 12 bit may come from both limited process power and read noise under continue shooting.
---------------------------------------------
To be honest, I would not expect major change in DR department given they keep the 12 bits mode and I don't feel DR is lacking actually from A7RII. It is good enough for me.
Claimed 15 stop DR and better color mostly come from pixel shift I guess, average 4 image to one will have 1 stop DR there, I'd say that is typical market cheating.
It is a solid update i would say but not enough for me to upgrade if I am a A7RII user. I don't need AF and new battery. more responsive and better EVF along with my manual are welcome update. I will just wait for cheap A7RII flood around and pick one again whenever I need it
p.29 #11 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
Good pickup Chris. I wonder how they got the raws? I know this is beta testing, but this is some hard evidence that Sony is lying. There is no DR improvement.
As mostly a landscape photographer, I’m not sure it’s worth the extra money over the A7r2 at the moment.
p.29 #12 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
I've only read through 2 out of the 29 pages of this thread so cannot say for sure. But has anyone mentioned anything about the electronic shutter performance for the r3?
My gut tells me that based on lack of reference to it in the available marketing literature so far, it is not going to be able to do what the A9 does with regard to curing rolling shutter and banding issues.
p.29 #13 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
TakenWild wrote:
Good pickup Chris. I wonder how they got the raws? I know this is beta testing, but this is some hard evidence that Sony is lying. There is no DR improvement.
As mostly a landscape photographer, I’m not sure it’s worth the extra money over the A7r2 at the moment.
It is concerning to see no improvement against the claimed 1 stop but... the same website also measured no change in base ISO DR on the D850 vs the D810. Whilst I have not used the D850 some comparisons in shadow recovery between the two models that I have seen around the web showed a significant advantage for the D850. This is something that Photons to Photos does not seem to capture in its measurements - perhaps the same here?
p.29 #14 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
JohnDizzo15 wrote:
I've only read through 2 out of the 29 pages of this thread so cannot say for sure. But has anyone mentioned anything about the electronic shutter performance for the r3?
My gut tells me that based on lack of reference to it in the available marketing literature so far, it is not going to be able to do what the A9 does with regard to curing rolling shutter and banding issues.
Anyone have any input on this?
My guess we'll have to wait for Jim Kasson to measure it.
p.29 #15 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
Yep, maybe better cleaned shadows, but no extra information.
Aztatlan wrote:
It is concerning to see no improvement against the claimed 1 stop but... the same website also measured no change in base ISO DR on the D850 vs the D810. Whilst I have not used the D850 some comparisons in shadow recovery between the two models that I have seen around the web showed a significant advantage for the D850. This is something that Photons to Photos does not seem to capture in its measurements - perhaps the same here?
p.29 #16 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
TakenWild wrote:
Yep, maybe better cleaned shadows, but no extra information.
Better, cleaned shadows IS more meaningful information, as long as the processing isn't applying putty to the shadows, but is actually lowering the noise floor.
Think if it this way; as you get closer to the bottom (in the shadows), the signal to noise ratio decreases. At some point the actual data in the capture (the signal) is overpowered by the noise in the sensor and signal amps. Somewhere in there, you have to call it quits and concede that you can't get anything more out, even though the sensor is actually recording information.
Take away a portion of the noise, and you will be able to see deeper into the shadows as long as the sensor is making meaningful collection down there. That increases DR and ultimately meaningful information collected.
p.29 #17 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
TakenWild wrote:
Good pickup Chris. I wonder how they got the raws? I know this is beta testing, but this is some hard evidence that Sony is lying. There is no DR improvement.
As mostly a landscape photographer, I’m not sure it’s worth the extra money over the A7r2 at the moment.
It's disappointing to see zero DR advantage from this preliminary report. Sony extrapolated a bit on the marketing hype this time...
p.29 #18 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
JohnDizzo15 wrote:
has anyone mentioned anything about the electronic shutter performance for the r3?
Anyone have any input on this?
Sony have said flash sync in pixel shift life is limited to 1/13s, which suggests that pixel shift uses the fully electronic shutter (sensible) and that the scan time is exactly the same as the A7RII i.e. 1/13s.
p.29 #19 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
Fred Miranda wrote:
It's disappointing to see zero DR advantage from this preliminary report. Sony extrapolated a bit on the marketing hype this time...
Fred,
Sorry if I missed it earlier in the thread, but for landscape shooters like yourself who use mostly MF primes, if this camera isn't offering greater resolution and/or DR, I'm wondering what your decision factors were in deciding to pre-order? Thanks in advance!
p.29 #20 · Pre-orders open! Sony A7R III and FE 24-105mm f/4 OSS lens!
Fred Miranda wrote:
It's disappointing to see zero DR advantage from this preliminary report. Sony extrapolated a bit on the marketing hype this time...
Agreed, but the battery and improved IBIS alone would sell it for me. No more battery grip needed! Also, I have terribly shaky hands, so every bit of stabilization helps.
The improved EVF and screen, UHS-II cards, better buffer, better ergonomics, and new shutter are all gravy.
I'd consider this a mid-cyle refresh and I'm satisfied with what they've offered
