Sony.net listed two(!) new Four Thirds sensor that are likely going to be used by Olympus in near future. One is a 16MP Exmor sensor with 23fps. The second sensor is far more interesting and could find his way in a future E-M1II camera: It’s a 20MP Exmor sensor with 27fps and fast sub LVDS 12ch readout. There is yet no way to see if that sensor can also do 4K video…
150 Responses
What does it mean LVDS?
channel readout , i not sure
Low Voltage Differential Signalling – transfer protocol to get data off the sensor block and into the camera processing unit.
Thanks!
Interesting to see how it will affect the IQ (read: noticeable difference or just at pixel peeper level?).
You will have more Resolution & DR with that sensor which Sony is famous with .
ISO could hit 6400 with good IQ then previous sensors .
That is what in my mind .
In my opinion this new sensor will mostly allow em1 II to take high resolution shots handheld.
SONY!!! yes yes yes, can’t hide my excitement yes!
Fanboism doesn’t help.
So people are getting very excited over a sensor that will give a tiny bit of extra resolution and probably at most half a stop extra DR (given that sensor efficiency is already at 50% ).
Still, if it sells a lot more cameras then it will be a good thing but I suspect that most of the people expecting huge improvements are going to be disappointed
How about noise performance at higher ISO’s?
It could be a good thing!
Decreasing pixel size isn’t usually conducive to improving noise.
Hmmmm…..maybe we should not get our hopes up for anything dramatic, then.
Not for current tech, however, if that geomembrane/organic sensor rumour turns out to be true, then you can expect both better dynamic range and lower noise, but that’s because the tech being used will fundamentally change.
DR is related to shadow noise at higher ISOs so there are probably limits to what can be achieved there as well.
I think people have to face facts. We are reaching the point where improvements in sensors will be hitting physical limitations and there isn’t a magic bullet that is going to give us the improvements that we received in the past.
In other words the technology has matured.
Personally I still think it’s pretty amazing what we have already but some people will never be satisfied (Yun, I’m referring to you 🙂 )
I am excited over 27 f/sec. The faster readout gives a better electronic shutter and great possibilities for different high resolution modes.
Yes, that is the most exciting improvement. That is almost twice the speed of the existing sensor
12MP with 60 f/sec please!
I too am excited over 27 f/sec.
Let’s hope it’s BSI too.
Exmor is FSI CMOS sensor Exmor R is BSI CMOS sensor
I think most of us are aware of that but thanks.
With a readout of 27 frames/s at all pixels one could have movie at 24fps without throwing away data. Could give stunning 4k if the processor are up for it (probably not…). But the fast readout certainly gives possibility for a more usable high resolution shot. I would like a 4 shot thing, where the sensor shifts and give me a 20mp foevon-like result. And 8pic shift with 40mp foevon-like result.
The full sensor readout for 4k would also allow the use of the extra pixels for IBIS when doing 4k video
Full sensor readout could bring incredible stabilization!
20 MPixels is not enough anymore. We need a clear differentation as a next step.
– 32 Mpixel for the 600-800 dollar market. GX8/G8.
– 16-20 Mpixel may be reserved for video and action photography. Of course this makes a fast long zoom obligatory.
In this view GX8/G8 would be a jack of all trades master of none in the sub 600-900 €/$ category.
– 32 Mpixels
– 4K but with pixelbining for instance, may be a crop
– 6-8 fps with C-AF, DFD
GH5 would be more geared to video, with 15 fps + DFD. May be 10 bit recording inbody.
Why 32 Mpixels? Go to dpreviews comparison tool, set it tolightbuld mode, chose printsize to equalize all sensors. Now compare GM5, GH4, A6000, RX100III. We see GM5 (also GF7) do very well. More than 1,5 stops better it seems than RX100III. A newer sensor should get us 32 Mpixels and get good ISO 6400 performance.
128MP in FF and good ISO 6400? Unbelievable. Better stay with 16MP an improve in all other IQ deparments. How about a BSI sensor?
I believe crazy amounts of pixels and iso values will be possible in the future. We may even start counting in numbers of 1000 iso (K-ISO?). I don’t know how long we need to wait, but I think I’d like to upgrade by then.
So you think shrinking pixel dimensions not only doesn’t come with any penalties, but that it actually can deliver improvements in dynamic range and such?
More correctly, we don’t really need more than 16-20 MP in m43 so 20 MP vs 16MP doesn’t really matter. 20MP is basically the quality of a good film negative scan. If you want to blow up bigger than 16×24, then pick something else. Shooting in the dark ala the A7S… now that at least opens up shooting options not already present.
I for one would be interested in seeing a sensor that was exactly 8K (about 16.5 MP) that could downsample exactly to 4K/1080p. This would be cropped for 4:3 or similar photography so a little less sampling than what we presently have for photos, but video quality would likely be much more akin to the A7S (still without the insane low light performance) as the sensor wouldn’t have to interpolate for video.
8K FUHD is a resolution of 7680 × 4320 (33.2 megapixels)
Oops, you are correct. My math was off… I was thinking 4K times 2 = 16 MP = 8K. Too bad.
16Mpix allows already to do very fine 4.3×3.5m (meters) size photos, that looks amazing from 1m distance. Seen those, done those, still amazing. Don’t pixel peep large prints too close as you don’t really see the photo, you see pixels. People don’t get it that viewing image files 100% magnification on their computer screen means that their final print image is already around 2-3 meters, and you look such print from 1.5x distance to long edge.
12Mpix allows to make huge, 1.5m long edge prints with quality that you can look from 30-40cm distance and not to see pixels, when you just know from where to order a well done print instead those one hour photo labs.
agree, gearheads and pixel-peepers are the cancer or photography
This is the sensor that should feature in EM5 2 & not EM1 successor .
20 MP Exmor Four Third sensor sounds impressive , pity it just miss the EM5 2 boat .
+1
Shoulda, Woulda, Coulda……..
Finally some people are getting it. This is the same reason why I’ve been saying gearheads shouldn’t pressure Panasonic into releasing a new GX camera right now, at least not until some sort of leap in performance is attainable (organic or not). otherwise we’ll be stuck with a stopgap sensor for at least 1.5 years while Sony and Fuji pummel Panasonic (and Oly with yet another evf-less PEN) to the ground.
Maybe that’s why they called it the e-m5II instead of the e-m6.
20MP is only 11% more horizontal/vertical pixels, its nothing.
100% faster shutterspeed i dim light would be something (ISO6400 = todays ISO3200).
Well, it’s 1.1180 so I’d round it up to 12%… but you’re right, it’s still nothing. If you were happy with a 10-inch-wide print from your 16-megapixel camera, you can now be happy with an 11.2-inch-wide print — big woop. I’d rather have 16mp and a larger pixel size. Unfortunately, a lot of people just buy on numbers…
Id rather have a 12MP with super fast hi-res/elecronic readout. Would be perfect for 4k without line skipping and downsampling.
Maby for the next video optimized E-M5 Mark III
This sensor will only make sense “in a future E-M1II camera”, if it reaches or beats Panasonic’s phase detection AF with DSLR lenses. As of yet, Sony did not show it’s capable of anything like it.
yes please 20 MP, my 40-150 f2.8 outresolves the 16 MPx sensor ! 😀
Likely this is the best 4/3 sensor up to date before the Organic tech hit the market .
Pana should consider this sensor seriously for it’s GX7 successor although my guess this
is very unlikely .
My guess is its ordered for the GX8
Yes .
If Pana is Quality oriented manufacturer then it should use that Exmor sensor
for the goods of the consumers here .
1st: This is Sony tech, Pany will definitively use their own sensor.
2nd: Even if they did use this sensor that would leave the GX line stuck with outdated tech for the remaining product cycle while lesser models get the organic sensor by virtue of being newer
Panasonic develops its own sensors, so it’s very unlikely they’ll use Sony’s. I assume the GX8 won’t have new sensor tech, since it’s only been two years since the last generation. Maybe they’ll push the tech to 20MP, similar to the sensor in the GX1 (16MP but same tech as 12MP). I guess the GH5 will get the next generation sensor, and only the generation after that will possibly maybe have an organic sensor.
Let’s hope your guess is wrong otherwise it’ll be disappointment
in GX7 successor .
Btw , the GX1 sensor is different from previous m43 sensor cause it capable of good IQ in ISO 1600 which previous sensors can’t make it at all .
The GX7 sensor is another breakthrough with good IQ hitting ISO 3200 .
The next target to achieve is clean & clear IQ at ISO 6400 .
My guess , GX7 successor definitely with new sensor with it’s highest grade in m43 CMOS before Pana turn it to Organic tech .
Indeed, the GX7 featured a newly developed sensor array. From ephotozine: ¨The GX7 features a new Panasonic sensor, with improved color
saturation, improved light sensitivity, and improved signal to noise
performance, thanks to improvements made on the sensor, with larger
on-chip lenses, a larger light receiving photo-diode, as well as
improved low noise signal readout circuitry¨
So it’s definitively not a higher Mpx version of the old 12mp sensor, it’s not even one of their 1st generation 16Mp sensors but actually a 2nd generation of the 16Mp sensor Pany was already using by then (GH4 would be 3rd generation). Let’s not forget that the GX line is Pany’s stills flagship model, the GH line sits above it only because its video specs demand more resources, but in terms of segmented hierarchy both sit at the top of their respective niches (i.e. GX7 for stills, GH4 for video); So let’s not assume the GX line isn’t worthy of a new sensor introduction. Also, historically (GX1->GX7) the GX line enjoys a longer product cycle (just like the GH line) so there’s not reason to pressure Pany into releasing a new model yet, specially considering Sony and Fuji are about to renew their competing lines, so it would be unwise for Pany to show their cards beforehand just to be trumped by the competition. Besides, if Pany releases a halfhearted update right now only for new sensor tech to be available in 6 months it will leave the GX line stuck with a useless stopgap of a sensor for at least 2 years, that would mean the death of the GX line (just how Oly is killing the Pen line). If m4/3 turns into a system relegated to big faux-DSLR like bodies then I’m out, there’s no point in having such a compact system if there’s no diversity of possible configurations (small body+big lens/ big body+pancake/ etc) for any possible scenario.
Let’s just be patient and wait for actual new sensors (organic or whatever it takes to see a notorious improvement in DR and high ISO) before pressuring Pany for a meaningless update.
“Olympus may uses that new Sony 20MP Four Thirds sensor!” lol, that headline hurts my brain. Anyways, cool to see new sensor tech but 16 Mp is plenty.
5-axis @ 5160×3870 and with Olympus pushing more into video 😀
In theory this sensor could support some cool tech like:
-4k with no problems
-2560x1080p (21×9) combining 4x pixels and not cropping horizontally
-WHXGA WHXGA (5120×3200) 16:10 ratio
-Ultra Wide TV (5120×2160) 21:9 ratio
-24p @ 4k with anomorphic support of upto 1.75x vertical squeeze 🙂
-12bit RAW sensor readout at 24p
I’d be more interested in greater dynamic range than frame rate. Dropping the pixel size won’t help that. Btw LVDS is the interface not the sensor and it’s pretty old 90’s technology. And 27fps is nowhere near 60fps or even 120fps or more needed for video. Sensor scan read speed is more usually given in Hz than fps.
And you would need a 20Mpx 60fps video for…? 😀 Well obviously it can record more than 27fps with lower resolutions.
How about native ISO 100 (or 50)?
50 iso should really be the maximum minimum iso in mft. 200 is just to high on so many occasions. And nose levels are appreciated.
ISO 50 would be nice…(maybe even 25?)…there should be more attention paid to low ISO – people seem to get focused on high ISO too much (though it has its uses). You can still buy 50 speed film (Ilford), so why not have it available for digital users?
It’s a tradeoff between low and high ISO performance. A sensor with a lower native ISO will require more amplification of the signal to achieve the higher ISO settings, which means more noise in images taken at those settings.
Basically, they have tried to compensate for the smaller sensor size by using a higher native ISO than for larger sensors. And since consumers are so obsessed with low light performance, I don’t think we’ll see a m43 camera with a true ISO 50.
Iso 100 should be possible though.
+1000
+1000 there are many occasions I can’t use max aperture. For example, I recently shot some climbing pictures with E-M5 and was forced to use f3.5 or even f4 because of light instead of f2.8 with the 40-150. I would have prefered f2.8 for thinner DOF. So iso 50 + 1/8000 would be nice. iso 50 is also nice for long exposure…I think m43 really need low iso or very high shutter speed because we really need to use max aperture to achieve thinner DOF and lenses are sharp wide open so low iso makes sense…
Every sensor has it’s optimal ISO rating, which is often NOT the lowest ISO setting a camera can set. So you will get better results by adding an ND filter to your lens than setting your camera to a sub-optimal ISO speed.
My guess is that the E-M5 optimal ISO is probably around 400 ISO, but you can figure this out by shooting your own tests.
If I could only carry one ND filter I would go with a 3 F-stop ND 8 filter.
Why not ND filter? really need to know, thanks
Because it is trouble to add a ND filter most often, that cause sharpness to be lost unless you spend 150-200 euros for good filter, and even then you get easily tint or sharpness lost.
Way to get ISO 25 or ISO 50 would be huge help in many quick situations where you just want to get 1-2 frames that has 1-4 second shutter speed.
And once you have placed ND filter to lens, you are #¤%#¤ if something interesting is going to happen quickly. Even those expensive ND filter plates are too slow to add/remove.
If I’m not mistaken, the IMX159 is the old sensor used in E-M5 and the rest of Olympus current models (except E-M1). So it’s only one new sensor, not two as the admin claims.
No, E-M5 uses IMX109 as chipforce find. So it is a new sensor.
I stand corrected. The IMX159 is a new generation of 16 MP sensor then, hopefully with improved image quality.
“I can pee farther than you” – that’s what increasing MP count actually means. More RAW files with more megabytes requiring more storage space only for the image to be down-sampled and compressed by Flickr 🙂
Anyone have a good example of a next generation sensor performing worse than the previous generation?
Canon 50D vs 40D ?
Fujifilm f30 vs f40 and f50.
20MP? Why? 25% larger files, 25% more power, 25% more heat generation, 25% more noise to suppress, about 0% more detail. Frankly, I’d be happy with 12MP.
sensor doesn’t generate heat as people believe, the CPU and battery (power drawn) are the cause that generate over 98% of the heat.
More pixels doesn’t mean more heat. The noise comes from electrons leakage (jumping).
But you are correct, files are bigger, more processing required and power consumption likely goes up. And we don’t really get ANY benefits from higher megapixel number than marketing propaganda. To really get visible benefit, we would need at least that 36Mpix and that is when cropping heavily or doing very very very large prints (couple meters) that are viewed from under meter distance. To get visible change, we need 64Mpix and then we see difference in larger prints (30″) when viewed from 5″ distance.
I actually think around 20 is the perfect amount of mp. Mp do help in cropping
Roughly calculated, the 16 mp 4/3 sensor already beats the pixel density of the fullframe nikon d810 and canon 3ds, as ist corrsponds to about 64Mpix compared to the fullframe sensor.
Now just imagine the loss of details and picture quality if you reduce the pixel size even more. The 20 mp 4/3 would be like a 80 mp fullframe sensor.
Yep, and I have native m43 lenses and legacy glass that can satisfy a 20MP AA-less m43 sensor, so bring it on. Also, we will likely get better DR when downsampling than with the current 16MP sensor.
In technology that kind of logic is very flawed, you can go and compare cameras that went from 16MP to 24MP and see if that logic applies, it doesn’t, the files, the power consumption, heat, etc. They don’t follow the same proportional increase as the MP count, even more because in that GAP technology advanced.
25% more power and 25% more heat generation?? Yeah… thats not how it works. Heat generation is a function mainly of the image processor, battery and housing of the camera. The actual sensor doesn’t add any significant extra heat… especially not just adding more pixels. That’s like saying a 5″ smartphone with a higher dpi monitor creates more heat than a 5″ smartphone with lower dpi. Also, the power difference, if there is any would be completely negligible. If it was linear like you are assuming, a Sony A7R would get like 1 hour of battery life compared to an m43’s camera.
Well, we won’t see these sensors right away. Chipworks’ images of the Sony IMX109 used in 2012’s Olympus E-M5 seem to indicate that the sensor was first made available in 2010 (see second image here: https://chipworks.secure.force.com/catalog/ProductDetails?sku=SON-IMX109 ). This seems in line with the information I’ve heard before that it takes about two years to bring a camera to market from the R&D stages, and they can’t just plug in a new sensor if it becomes available.
I hope this is not a trade-off of more megapixels for lower quality low light shots (worst ISO performance). Hopefully Sony has made some big strides all around
I hope so too. I even hope they can keep the filesize down no matter how many pixels.
If not then there’s still that new 18mp panasonic sensor they may use.
They have… look at the A7 sensors or the RX1003 sensors compared to the current olympus sensor.
Its 2 stops difference at best, with the A7s pushing closer to 3 when the ISOs get extreme (ie: 25600iso+). This is exactly what is to be expected with a sensor that has 4 times the surface area. There isn’t any special Sony magic going on here, just pure physics.
Huh? The 1″ 20MP RX1003 sensor is competitive with the 4/3 16MP EM-5ii sensor.
That’s “special Sony magic”, AKA Exmor-R.
A pocketable RX100iv with a 4/3 sensor would be pretty interesting if they could keep the lens small enough/collapsible.
No way Sony going to create that camera with 4/3 sensor size except joining m43 .
Would they need to join the consortium if they’re not using the mount? Did they need to join to manufacture the sensor? Sony has been pushing bigger and bigger sensors into smaller and smaller cameras. This seems a natural next step for the RX100 lineup.
I guess because 4/3 sensor size is patented & protected by Pana-Oly .
That’s why we see 1″ sensor size , APSC & FullFrame from various manufacturers but not 4/3 sensor size .
The RX100 can move to APSC but not 4/3 & if Sony did it that will
definitely endanger the NEX line cameras .
The sensor size is NOT protected, it’s the lens mount.
I would not think that an aspect ratio could be patented, nor would a fixed lens compact be seen by Sony as a threat to the NEX lineup. For me, a truly pocketable 4/3 camera would be hard to resist, although I’d much rather have a slightly larger, more ergonomic camera with a next generation sensor! 😉
The sensor size designations (1/2.3″, 1/1.7″, 2/3″, 1″, 4/3″) are based on the diameters of old video camera tubes. Olympus adopted the 4/3″ size for their Four Thirds system, and later for the Micro Four Thirds system, but they didn’t create the format. They only created the lens mount and the electronic communication protocols, and that’s what’s protected.
Yun is confusing the mount standard with the sensor format, two completely different things.
It’s just my guess .
You might right on this one .
No need to join the m43 group, since a compact camera doesn’t need the lens mount specifications.
If it uses a 4/3 sensor which has almost double the area, it won’t remain pocketable, it will have the size of a LX100. The trade off is not worth and I’m pretty sure Sony prefers to keep the size down since that’s the main reason people like the RX100.
Sony could keep the RX100 MKIII in the market and make a LX100-like camera and be aggressive with it, but I don’t think they would go that route.
They always can make the lens darker by a stop. f/1.8 is useless in sunlight without an ND, in-body ND also makes a camera bigger and heavier. A bigger sensor with f/2.5-f/4 instead of f/1.8-f/2.8 would work the same in low light but better in sunlight (potentially higher DR due to higher full well capacity, potentially higher output resolution with the same lens resolution as measured in lines per mm).
Making it darker isn’t really the way when all the cameras in the segment are trying to get brighter.
It is, in combination with bigger sensor, the way to achieve lower size/weight and vignetting while having the same low-light performance and better total DR and sharpness. It is simply a function of current lens and shutter technology – if you need to go below f/2.8 (at least on a zoom), you are better off increasing your sensor size.
I don’t know who made the sensors in the Nikon D7000 or the D7100, but the sensors went from 16 MP to 24 MP and the D7100/D7200 have better ISO performance. So, it can be assumed that 16MP to 20MP wil have a slight iso performance as well.
Sony made those sensors. The D5100, D7000 and D7100 all have amazing high ISO (very clean ISO 6,400) performance. I can’t wait to see what these new sensors do for m4/3. This is wonderful news.
It’s only one new sensor. The 16 MP sensor is already used in several Olympus models.
They are not all from Sony, after the famous 16MP used by so many, I think Nikon only used Sony’s 24MP sensor in the next generation and since then, they are using Toshiba’s 24MP sensor.
Olympus should improve the battery for the E-P5, E-M5, E-M1 and E-M5 II and its successors. A battery with the same interface (shape), but lasting twice as long.
I’m not worried about performance and quality with this new sensor. Moore’s law applies to camera’s as well.
If they had a 100% better battery tech available, they would have used it.
Battery tech is up to the battery manufacturers not Olympus. They can do things to conserve battery but at a certain point it’s just tricks
Yep, battery tech is up to others. More power in a smaller volume is always a challenge. Just look at the history of fire problems with Li-Ion batterie over the years.
No it doesn’t, do you know what Moore’s Law actually means?
A digital camera is a computer with a processor. It converts light to a file on a SD-card. The processor and all other chips are subject to Moore’s law. I assume at least you get newer components in newer camera’s; especially with a 20 MP sensor.
The processor is subject to Moore’s law, but the imaging sensor and battery are not. Also, cameras and electronics other than phones or computers never get the latest and greatest processor from the newest fabrication plants. I think most cameras are using processors from 65nm tech and image sensors are as well, let alone the 14nm tech we have today with Intel’s processors.
Exactly right. Too many people don’t understand what Moore’s Law really means.
Do you think that Sony may use one of these sensors in their rumoured upcoming RX100 mk4? Increasing the size of the sensor much like Panasonic did with the LX100?
A nice thought!
4/3 sensor size is protected & patented by Pana-Oly .
No way Sony going to create camera with that sensor size except joining m43 .
Kodak has a MFT camera. I’m pretty sure it’s an open standard.
And this would not be a M43 camera. It would be a fixed lens compact with a 4/3 sensor.
Did I miss something? “Application: DSLR” is right there in the chart. Fixed lens compact? I hope not.
Yes, it says that, but I doubt that the various categories are specific to particular types. Likely the DSLR category includes all digital cameras with interchangeable lenses and quite possibly also non-interchangeable lenses.
I’m not aware of a current 4/3 “DSLR.” The sensor is most likely for Olympus mirrorless cameras. The conversation threading got chopped up, but we fantasizing about the next generation RX100 getting a bigger sensor but not growing in size. Not likely.
JK Imaging, the licensee of the Kodak brand, is a member of the m4/3 group, so they have access to all the lens mount specifications. They are only shared with members, so it’s not really an open standard to anyone.
It’s irrelevant in this case, though, since we’re talking about a fixed-lens camera without a lens mount.
Kodak (pre-breakup) was one of the original members of the 4/3 consortium and they made the sensor for the E-1 4/3 camera. Who got their membership rights as Kodak gradually broke up I don’t know. Their sensor division was one of Kodak’s early spin offs and that division continued to make 4/3 sensors for industrial purposes.
No, the 4/3 and m4/3 standards are related to the lens mounts and the camera/lens communication protocols. The sensor size is not protected any more than any other sensor format. If Sony can put 1″, APS-C and FF sensors in a camera, they can also use a 4/3″ sensor.
Sony has ALREADY made a m4/3 sensor… the EM-5 mk 1. That was the whole reason it never received focus peaking. Sony apparently required Oly to not add that feature to prevent competition with their own cameras. That aside, I recall sometime ago that Sony agreed to make sensors for Oly and Oly agreed to help make better Sony lenses. Don’t know how it all worked out, but certainly Sony has no problem getting in on the format.
http://digitaljournalofphotography.blogspot.com/2012/07/olympus-om-d-em-5-carries-sony-sensor.html
Tough choice: pocketable RX100m4/3 vs beautiful/ergonomic LX100. I’d love to see it happen
I don’t think that will happen because the majority of people that buys the RX100, they do it because of its size, if the sensor gets bigger, so will the camera, it wouldn’t make sense for Sony to lose that edge and compete with Panasonic in a slightly different segment – they both share part of the same market but there is a different between the two cameras and clashing on that for no reason doesn’t seem likely.
As it has been said: 4/3 sensor in the RX100 line seems unrealistic. This is a lovely pocketable camera and a bestseller. Body size is not an issue, because RX100 is actually larger than Panasonic GM1, but the lens would have to become significantly bigger.
This is, however, a perfect sensor for RX10 successor. This is already a huge camera, so few extra centimeters wouldn’t hurt.
I have to say I love the fact that a company which first made a big fuss about switching from SLR to SLT and then (practically) dumped the latter in favor of mirrorless is putting a “DSLR” in an official table like this one. 😀
Interesting that this comes a few days after Sony sold 50% of its stake in Olympus.
Not really, Sony is Olympus’ pimp. Sony still controls >5% which is a major shareholder and Sony sold the other portion to JPM which makes Olympus a total slave. The original Sony stake was to enter Olympus’ medical device market that they had 70% and also get their technology. Now that they have done this, Sony made tons of cash and no longer needs the shares, still controls Olympus and forces them to buy their sensors AND Sony said they will now use the money they made off of Olympus to further invest to the medical device and consumer camera market to take Olympus’ share. Then…the aftermath is that Olympus must still continue to share its camera technology with Sony.
Hence my comment in another thread and for those that know the phrase.. Sony to Olympus: “All your base are belong to us!!”.
Am I the only one who thinks that cropping deserves some fingers hitting? 🙂
You’re not the only one.
I’m pretty sure the sensor is capable of 4K, it wouldn’t make any sense to Sony, a company pushing for 4K, to not have it in the their new sensor. Will Olympus will use it? That’s another story because it’s not like the current sensor on Olympus’ bodies can’t do 4K to begin with.
I believe that e-m1 & m5II can, but handicapped by FW
I’ve read comments suggesting that they’ve not been able to
solve the heat dissipation issues associated 4k video while continuing to use
IBIS. Don’t know if that’s true or not but I do know heat dissipation is an
issue with 4k video.
The heat problem is not with the sensor but with the processors that process the video…
Sounds right. My Blackmagic Pocket Cinema Cam (BMPCC) is a prodigious heat generator, and needs lots of electricity to generate it.
20mp on a Four Thirds sized sensor?? Ug, NO THANKS.
Nobody here is worried about the Noise levels after iso 800? All those pixels on such a tiny sensor? That spells noise disaster to me.
A newer processing engine would help, but I don’t think it will. Honestly, I want to see a 12mp sensor with no AA filter.
C’mon Panasonic, make it!
Only for you to be whining about not enough megapixels on a tiny sensor? 🙂
Um, how is asking for less Megapixels whining about not enough Megapixels??
No disrespect, But Can You Read?
I can, and I’ve read nearly all of your previous rants. It seems no matter what they do, you find some reason to troll.
Go buy an older m43’s camera with 12MP and see how the ISO performance works out for you compared to today’s 16MP sensors. Oh wait… the older sensors had less pixels AND more noise. There are advantages to more pixels, and sensor engineers are finding ways to increase pixel density AND improve noise over previous sensor designs.
More megapixels is always better.
If it has too much noise for you, you can simply downsample and be done with it. But at daytime with much light more megapixels simply rule…
If the lenses can resolve it ofc, but oly lenses can!
More MP is always better?
Then explain to me why Canikon’s best dslr offerings have lower megapixels than like a Canon 5D MkIII?
AND downsampling? Why would I want to do that?
more MP gives more noise *at pixel level*, but when downsampling you are reducing noise.
so with higher MP you get the choice between more pixels with more noise or equal pixels with comparable noise.
True, although technically it also depends on how much “wasted” (i.e. non-photosensitive area) there is around each photosite (supporting circuitry and such). In ideal case there would be no such areas and your logic would be correct. On the other hand, noise-reduction algorithms trade noise for spatial resolution, so the more resolution they have the more subtle visual artifacts are. But this also assumes that each pixel actually resolves valid piece of information.
I think also Sony mean 4/3″ is a good choice.
Sony still can’t make a EXMORＲ 4/3 or APS sensor. The improvement over the current in DＲwill be minor then,
Unless it has PDAF built-in, it won’t be introduced for the E-M1 MkII, but in two years with the E-M5 MkIII. Don’t forget Olympus said that the E-M5 series will be used to introduce new technologies…
Question: That “fast sub LVDS 12ch readout” what does that do exactly?
Transfers data from the sensor to the processor. “sub” means that it uses even lower voltage than specified by Low Voltage Differential Signaling standard (LVDS), 12ch means it has 12 channels working in parallel (each LVDS channel is serial, meaning 1 bit is transferred at a time).
Sony still doesn’t make a 4/3 or APS-C sensor because the improvement would be minor (compared to the cost increase).
The pixel density isn’t quite high enough yet for BSI to fully make sense.
“There is yet no way to see if that sensor can also do 4K video…”
27fps full-sensor readout of 4:3 sensor means that 19:9 crop of it can be read out at about 35fps or more, more than enough for “4k” 30fps without line skipping. The problem is processor and codec in it, are they able to downscale these megapixels to 4k and encode them?
20 mpix is not exactly competitive in the current market where all competitors are at 24-28mpix.
Exactly…that is falling behind. We need a 32 Mpixel cam for 600-900. Good in everythuing, appealing for anyone who wants detail aswell as laypeople.