-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
E-M1 Mark II Sensor Results
- Thread starter bclaff
- Start date
Martin.au
Forum Pro
LearningForeverIHope
Senior Member
You are both right.
J.
J.
TN Args
Forum Pro
Yes, and those latter people who use DxOMark as a proof of m43 inferiority will have to look at the D500 and the E-M1 Mk2 scores and admit that it indicates:
- overall scores of 84 and 80 are separated by less than 1/3 stop of overall performance (DxOMark say 5 points equates to 1/3 stop)
- the above difference lies entirely in dynamic range ('Landscape') and not in high-ISO ('Sports')
- their sensors are effectively equal for the action photography that so much of their expensive technology is focused on delivering.
... that would ever be true. ; )
To me, the most interesting thing about the MkII's scores are that they appear to show that Olympus is continuing to develop their sensor designs. All the habber-jabber aside, we should all be happy about that.
The problem is that the high-ISO score is very suspect, as it would mean the sensor is unbelievably efficient, more than any other sensor on the market. But Olympus did not announce a uniquely efficient sensor, we know of no technology used to markedly improve efficiency (e.g. BSI) and the direct comparisons around the web do not show any significant improvement compared to previous M43 sensors.
It would be truly fantastic if the new sensor made such an incredible leap, but it seems hard to believe for me.
karlreed
Senior Member
I posted these before, and, the seem to have disappeared. All are 12800iso, the only adjustment is a mild tweak of the exposure in some cases.
D500
D700
D700 cropped
Om-d EM-I mkii
OM-D EM-! cropped
So, while not perfectly scientific, I think they say more than mere argument.
Oh, I don't know the first post was removed..
--
karl reed
D500
D700
D700 cropped
Om-d EM-I mkii
OM-D EM-! cropped
So, while not perfectly scientific, I think they say more than mere argument.
Oh, I don't know the first post was removed..
--
karl reed
Attachments
Last edited:
cameron2
Senior Member
It will never be believed, because common sense tells us that is a marketing lie: Olympus neither designs nor builds sensors. They likely worked with Sony to customize the 20mp sensor used in the PenF, e.g. by adding micro-lenses to support PDAF functionality, but one simply does not "design a totally new sensor".
gary0319
Forum Pro
bclaff
Forum Pro
Just to be clear, on sensor PDAF is a function of the sensor; micro-lenses are one of the "toppings" and may need to be redesigned depending on the PDAF implementation.
(But PDAF is not a function of the micro-lenses.)
Martin.au
Forum Pro
So, where'd it come from.
It has different behaviour than the PEN-F sensor.
Far faster readout rate.
Quite different DXO results.
Cross point PDAF
Assuming the PEN-F behaves like the other Oly's, then the E-M1 II probably also has less long exposure noise.
Last edited:
gary0319
Forum Pro
I believe it has been established that the E-M1 II images on the DPR image comparison are bogus, having been converted with the preliminary version of ACR currently available. I saw the same junky images when I first used ACR with my Pen F as I now get with this version of ACR with my E-M1 II. Right now OV3 does a better conversion, by far, so using the ACR image for comparison is fruitless.
Martin.au
Forum Pro
Agree with that. I got obviously better results than the E-M1 when both were processed with OV3.
Just occurred to me, I might now be able to process them both through DXO.
karlreed
Senior Member
I think you'll find that the actual chip fabrication can be sub-contracted to an appropriate fab plant. Plus, the R&D that is involved exists a a number of levels. conceptual, architecture, semi-conductor physics related to hoto-optics, AF algorithms and technology and so-on.
The designers working in these areas use what I have called "technology-solution-look-ahead". They work out what technology is available, where it is going, and, if they have the clout, even influence its direction. They may even be able to say "I need this please", and the grunts in the form of semiconductor PhD's will trot off and make it happen. Or, tell them it can't be done. (But try telling a marketing person it can't be done!)
Not all of the research requires in-house capability. Putting PDAF sensor on-chip was a conceptual decision made, I think, by Nikon for the V1.
Linking the PDAF and CDAF so that lens calibration can be done the fly, is a conceptual decision. You don't need a lot of money to come up with that, nor to test it.
Happy to expand on this, however, one of the earliest examples was Seymour Cray, who, in the very early 1960's ASSUMED that silicon planar transistor swould become a commercial reality, and, designed the CDC super computers assuming their availability.
So, to suggest that Olympus's scale would mean it cannot do the R&D to develop new senor technology is not really a well-founded suggestion to put it politely.
It's possible that the ONLY thing it might not have is the senor fabrication capability.
As others have said, there are lots of ways to improve sensor IQ as far as the end user is concerned. Perhaps the most fascinating is jpeg rendering, which could see a situation in a few years where NO-ONE will bother much with in camera raw, only very high quality very hi-res jpegs. The OOC hi-iso jpegs from the class leaders are already stunning.
Last edited:
gary0319
Forum Pro
Last edited:
SHood
Veteran Member
Similar threads
Keyboard shortcuts
- f
- Forum