New BSI stills sensor on the way?

RichardC

Mu-43 Hall of Famer
Joined
Mar 25, 2018
Messages
4,546
Location
The Royal Town of Sutton Coldfield, UK.
Real Name
Richard
I do not need it. But I WANT IT!
What is wrong with me? :hmmm:

I confess I've never seen a BSI vs 'normal' comparison and I don't pretend to fully understand even how a sensor works.

Improved low light performance would be nice, but I don't really use my cameras in low light. However, if it's 'faster', that has all sorts of implications for live composite, live ND, HHHR and pro capture.
 

BDR-529

Mu-43 Top Veteran
Joined
Jun 27, 2020
Messages
791
I confess I've never seen a BSI vs 'normal' comparison and I don't pretend to fully understand even how a sensor
IMX472 is stacked BSI-sensor which means that it's technically as good as they come under lowlight conditions.

(I mean mass manufactured camera sensors in reasonable price category)

It's not Dual-ISO or Organic sensor which would be really cutting edge but even now you should be able to crank up the ISO by one or two steps.
 
Last edited:

apete

Mu-43 Regular
Joined
Mar 14, 2018
Messages
101
Location
Poland
Real Name
Piotr
I confess I've never seen a BSI vs 'normal' comparison and I don't pretend to fully understand even how a sensor works.

Improved low light performance would be nice, but I don't really use my cameras in low light. However, if it's 'faster', that has all sorts of implications for live composite, live ND, HHHR and pro capture.
What I hope for is some (probably minor, but still) improvement in noise handling. Minimizing rolling shutter is a good thing as well. I also like the idea of 20/24 mpix, I believe many of us do not need more. I hope Olympus will manage to use the new sensor in upcoming cameras, and will be able to maintain its good color science.
 

Ross the fiddler

Mu-43 Hall of Famer
Joined
May 20, 2012
Messages
4,304
Location
Blue Mountains, NSW, Australia
Real Name
Ross
I confess I've never seen a BSI vs 'normal' comparison and I don't pretend to fully understand even how a sensor works.

Improved low light performance would be nice, but I don't really use my cameras in low light. However, if it's 'faster', that has all sorts of implications for live composite, live ND, HHHR and pro capture.
I dread to think how many more frames I'd have to delete from ProCapture before downloading with that top speed. ;)
 

hoodlum

Mu-43 Top Veteran
Joined
Jul 16, 2012
Messages
566
Location
Toronto Canada
This is not the sensor in the GH6 but still could be used in other bodies. BSI bring very little low light advantage and no DR advantage. Below is a comparison of 20mp 1" sensors when Sony went from non-BSI to BSI sensor. The improvement for the 20mp m43 sensor would be even less as BSI has a larger low advantage with smaller pixels and the 20mp m43 sensor has larger pixels than the 1" sensor. In order to support a stacked sensor you must implement BSI at the same time so they go hand in hand even if BSI provides no benefits.

https://www.dxomark.com/Cameras/Com...ersus-Sony-Cyber-shot-DSC-RX100-IV___957_1033
 

Attachments

  • Capture.JPG
    Capture.JPG
    100.2 KB · Views: 25
Joined
Mar 17, 2021
Messages
145
Location
Solna, Sweden
Real Name
Henrik
This is not the sensor in the GH6 but still could be used in other bodies. BSI bring very little low light advantage and no DR advantage. Below is a comparison of 20mp 1" sensors when Sony went from non-BSI to BSI sensor. The improvement for the 20mp m43 sensor would be even less as BSI has a larger low advantage with smaller pixels and the 20mp m43 sensor has larger pixels than the 1" sensor. In order to support a stacked sensor you must implement BSI at the same time so they go hand in hand even if BSI provides no benefits.

https://www.dxomark.com/Cameras/Com...ersus-Sony-Cyber-shot-DSC-RX100-IV___957_1033
1_jX8qGZC3WOIm1LFvBNL5jA.png
Subscribe to see EXIF info for this image (if available)
 

swifty

Mu-43 Regular
Joined
Jan 30, 2013
Messages
188
Location
Singapore
Real Name
Dave
BSI claws back the photosensitive area you lost to having supporting electronics at the front of the sensor by moving it to the back. If the fill factor was already high on a FSI design, and gapless microlenses already do a great job at helping this, then BSI won’t gain you much at all. The reason there were larger gains observed on very small sensors when they switched from FSI to BSI was because the fill factor were poor on those small sensors so there were more efficiencies to claw back.
Moving the electronics to the back, the type of wiring etc. will improve your read speed. Moving to stacked sensors potentially dramatically further increases your readout speed but generally speaking moving data faster increases your read noise too.
So most of the improvements in sensor tech in recent years have been to increase readout speed whilst maintaining the same IQ levels by keeping electronic noise under control.
We’re not going to see much improvements in IQ, whether high ISO or DR.
What we will see are all the great things they can do with the much greater sensor bandwidth such as AF, subject recognition and tracking, EVF readout (both resolution and refresh rate), all the AI stuff etc. when a current fast sensor is coupled with appropriately powerful processors.
 
Joined
Jan 26, 2021
Messages
122
Location
Northwest Arkansas, USA
Real Name
Mike
  • BSI
  • Stacked
  • Quad Bayer
Didn't see people mentioning quad bayer. What other ILCs even have a quad bayer layout (2x2 grid of same color)?

quad-bayer2.png
Subscribe to see EXIF info for this image (if available)

https://www.gsmarena.com/explaining_huawei_p20_pros_triple_camera-news-30497.php

Has been used in phones for a bit of time, and especially for HDR / computational tricks to leverage the fact that you can sample close-by areas in different ways. But typically at penalty of resolution, so... Maybe somebody has other ideas or info?
 

WT21

Mu-43 Legend
Joined
Feb 19, 2010
Messages
7,280
Location
Boston
Hey - I'm open to ANY attempt at an improved sensor. Faster read out feed could up the Panasonic AF algorithms. Looking forward to this sensor in a G85-type body.
 

agentlossing

Mu-43 Hall of Famer
Joined
Jun 26, 2013
Messages
4,915
Location
Oregon USA
Real Name
Andrew Lossing
BSI claws back the photosensitive area you lost to having supporting electronics at the front of the sensor by moving it to the back. If the fill factor was already high on a FSI design, and gapless microlenses already do a great job at helping this, then BSI won’t gain you much at all. The reason there were larger gains observed on very small sensors when they switched from FSI to BSI was because the fill factor were poor on those small sensors so there were more efficiencies to claw back.
Moving the electronics to the back, the type of wiring etc. will improve your read speed. Moving to stacked sensors potentially dramatically further increases your readout speed but generally speaking moving data faster increases your read noise too.
So most of the improvements in sensor tech in recent years have been to increase readout speed whilst maintaining the same IQ levels by keeping electronic noise under control.
We’re not going to see much improvements in IQ, whether high ISO or DR.
What we will see are all the great things they can do with the much greater sensor bandwidth such as AF, subject recognition and tracking, EVF readout (both resolution and refresh rate), all the AI stuff etc. when a current fast sensor is coupled with appropriately powerful processors.
I think you're right here, it's obvious that the existing 20mp sensors are pretty efficient, since we've seen the M4/3 sensor IQ improve from 12mp to 16mp to 20mp without actual increase in sensor size. I suspect there would be a very slight increase in DR or noise performance, at least stated increase to help justify the extra cost which will doubtless come with this new sensor, since it doesn't have the long-awaited increase in resolution which would be the biggest draw. I could see advertising at least a new top ISO, in addition to speed improvements to allow for better sensor readout, but if it will be enough to make a real-world difference, who knows. I think enhancing the computational capabilities, both of AI focus aids as well as maybe some actual computational enhancements to IQ, will be the main thrust of this sensor. Just as I sit here pontificating...
 

Jon Li

Mu-43 Regular
Joined
Jan 15, 2014
Messages
169
Location
Jakarta - Indonesia
If this stacked BSI sensor will perform anything like that in Sony A9 , I wud be a definite buyer of the camera , 20 mpx is enough but higher is a bonus .
 

saladin

Mu-43 Hall of Famer
Joined
May 29, 2015
Messages
2,263
Location
Melbourne
Real Name
jason
For my use, and more so now that I'm printing a few shots , I'd rather a few more mp than any perceived advantages of bsi. A lot of mft "noise" ceases to exist under an inkjet - or at least doesn't look terrible - and I'm not overly worried about faster readout speeds as I never shoot video and rarely electronic shutter. It might be useful for sports or weddings, I guess.

And I look at comparisons of the Z5 to the Z6 , and in some instances the Z5 looks better (for stills) than the bsi Z6. So I'd want to see actual outputs before rushing out to buy a new bsi equipped model.
 

Ross the fiddler

Mu-43 Hall of Famer
Joined
May 20, 2012
Messages
4,304
Location
Blue Mountains, NSW, Australia
Real Name
Ross
For my use, and more so now that I'm printing a few shots , I'd rather a few more mp than any perceived advantages of bsi. A lot of mft "noise" ceases to exist under an inkjet - or at least doesn't look terrible - and I'm not overly worried about faster readout speeds as I never shoot video and rarely electronic shutter. It might be useful for sports or weddings, I guess.

And I look at comparisons of the Z5 to the Z6 , and in some instances the Z5 looks better (for stills) than the bsi Z6. So I'd want to see actual outputs before rushing out to buy a new bsi equipped model.
How large are you printing them though?
 

John King

Member of SOFA
Joined
Apr 20, 2020
Messages
2,844
Location
Beaumaris, Melbourne, Australia
Real Name
John ...
How large are you printing them though?
Ross, I've printed to A0(+) size using A4 size crops from my E-510 (10 MPx). Still no sign of 'noise' as routinely seen on very low resolution devices (high quality/resolution, 100% aRGB, IPS monitors ... ).

My R3880 has around 1,000 times (approx.) the resolution per square inch of my best monitor!
 
Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Mu-43 is a fan site and not associated with Olympus, Panasonic, or other manufacturers mentioned on this site.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Forum GIFs powered by GIPHY: https://giphy.com/
Copyright © Amin Forums, LLC
Top Bottom