Based on the charts on the DxOMark site, the E-P3 is no better than the E-5 in terms of noise, dynamic range, etc. Olympus continue to not only disappoint me, but also mislead us with their claims of "a new sensor." If their goal is to anger and alienate their customers, they're doing a bang-up job.
I'm not saying that the E-P3 is bad by any means. I've been impressed with the sample images I've seen. I may still buy one. But Olympus, stop lying to us.
Based on the charts on the DxOMark site, the E-P3 is no better than the E-5 in terms of noise, dynamic range, etc. Olympus continue to not only disappoint me, but also mislead us with their claims of "a new sensor." If their goal is to anger and alienate their customers, they're doing a bang-up job.
I'm not saying that the E-P3 is bad by any means. I've been impressed with the sample images I've seen. I may still buy one. But Olympus, stop lying to us.
- Hal -
I feel disappointed too. I had been told the sensor was improved in terms of RAW-level image quality. I can only think that something was lost in translation. Maybe the case for high expectation didn't help, either.
Like you, I continue to be pleased with the results from the camera, but one has to wonder what it might have been like if the sensor really had been improved at the hardware level.
And let's not forget the considerable step forward that TruePic VI represents, which is great news for those not so interested in RAW processing.
By making the JPEG files effectively superior to raws it seems Olympus are channelling users (of the Pens at least) towards preferring JPEGS. Maybe they feel that with good enough JPEGs raw files will become redundant. It all points towards defining a specific market for MFT.
Perhaps it's because Olympus don't have much experience of designing their own sensors or maybe Panasonic manufacturing is the limiting factor. Either way, could it be time for Olympus to consider Sony as sensor suppliers?
Having said that, it's not a total disappointment, just not quite what we were expecting. The JPEG image quality still speaks for itself.
...why not consider the Foveon sensor?...now that with Olympus optics...
But hey ho, the new PENS are being marketed and if marketing did not raise expectation we'd never buy anything...so dont slag Olympus for being like any other company...and the EP3 is an E5 in a smaller body...maybe that should have been the marketing angle
By making the JPEG files effectively superior to raws it seems Olympus are channelling users (of the Pens at least) towards preferring JPEGS. Maybe they feel that with good enough JPEGs raw files will become redundant. It all points towards defining a specific market for MFT.
Perhaps it's because Olympus don't have much experience of designing their own sensors or maybe Panasonic manufacturing is the limiting factor. Either way, could it be time for Olympus to consider Sony as sensor suppliers?
Having said that, it's not a total disappointment, just not quite what we were expecting. The JPEG image quality still speaks for itself.
As usual, John - your thinking is very interesting.
It has been pointed out to me that the majority of the intended market for Pens will not be interested in messing around with RAW files. And certainly, the improved performance of TruePic VI (considering how good all the previous TruePic engines were for their time) is a remarkable achievement.
We all know that for a given pixel pitch, Sony leads the way in sensor performance at the moment. It would be remarkable if Sony could supply Olympus with a Four Thirds sensor that delivered the same noise and dynamic range performance that Sony's own 16MP cameras currently enjoy. But I think Olympus us tightly wedded to Panasonic sensors for Four Thirds. As I have said before, I don't for a minute believe that Olympus is excluded in any way from using the higher-specification LiveMOS sensors used by Panasonic. Olympus has made a calculated decision in its choice of sensor. It could be simply down to cost or the way the sensors available fit or don't fit into the architecture that Olympus has developed (fitting in to the other circuitry and processors architected by Olympus).
While all this will be added grist to the mill for Olympus critics, it can't take away the fact that the E-P3 is a jolly nice camera and very desirable. Maybe it's a camera that is greater than the sum of its parts!?
One magazine editor did confide in me that he thought the Pens were developing very nicely and that he wasn't concerned about highly technical tests in other magazines that criticised high ISO noise and other extreme image quality measurements.
That said, it would be much better for Olympus if it had been able to use a sensor with improved noise and dr at the RAW level. It would have made it easier to market the camera and add confidence to Olympus' strategy.
I don't buy the Foveon sensor argument. I have nothing against Foveon, but I have yet to see really convincing evidence that Foveon sensors are significantly superior to conventional Bayer Array sensors. I think the latter are developing faster (at least in Sony's hands!) too.
At low ISO, it's does very well for the resolution. At moderate ISO, it breaks up worse than our 12MP sensor. The new foveon apparently is prohibitively expensive - SD1 carries a ridiculous price tag.
The graphs are direct Manufacturer ISO. This may be misleading though. the measured ISO for the Panasonic and even the ep2 was higher than the EP3 values. In fact, in the previous comparison, it was seen that the EP3 shutter speed was slower than the Panasonic or EP2 for the same exposure, at the same ISO.
I've attached the graph of measured ISO, which as you can see is closer.
The new sensor appears to do better at noise - albeit, not by much. The grain does appear easier to work with - but I'll need to play with more RAWs to really commit to that.
At low ISO, it's does very well for the resolution. At moderate ISO, it breaks up worse than our 12MP sensor. The new foveon apparently is prohibitively expensive - SD1 carries a ridiculous price tag.
The graphs are direct Manufacturer ISO. This may be misleading though. the measured ISO for the Panasonic and even the ep2 was higher than the EP3 values. In fact, in the previous comparison, it was seen that the EP3 shutter speed was slower than the Panasonic or EP2 for the same exposure, at the same ISO.
I've attached the graph of measured ISO, which as you can see is closer.
The new sensor appears to do better at noise - albeit, not by much. The grain does appear easier to work with - but I'll need to play with more RAWs to really commit to that.
Statistically, the DxOMark chart you have shown is practically identical (within acceptable margins of error) for all three cameras and this should be no surprise because they all use the same sensor. The DxOMark test does not take in to account factors like anti-aliasing (although this has no relationship to noise anyway. Our DxO Analyzer software does generate data about grain size and intensity, among other things, but at the moment we can only test E-P3 JPEG files.
Statistically, the DxOMark chart you have shown is practically identical (within acceptable margins of error) for all three cameras and this should be no surprise because they all use the same sensor. The DxOMark test does not take in to account factors like anti-aliasing (although this has no relationship to noise anyway. Our DxO Analyzer software does generate data about grain size and intensity, among other things, but at the moment we can only test E-P3 JPEG files.
Ian
I get that - what I wanted to point out was the ISO3200 of one camera doesn't equate to ISO3200 of another.
Based on the DXO data, ISO3200 on the G1, is closest to ISO6400 on the EP3. It's probably not a whole stop different, but there's probably something. If that's the case, then you'll see less of a difference in the graphs from that article.
I get that - what I wanted to point out was the ISO3200 of one camera doesn't equate to ISO3200 of another.
Based on the DXO data, ISO3200 on the G1, is closest to ISO6400 on the EP3. It's probably not a whole stop different, but there's probably something. If that's the case, then you'll see less of a difference in the graphs from that article.
I don't think that's correct. The Pens consistently over-estimate ISO sensitivity and the G2 under-estimates it. If you look at the chart the three plots are almost exactly along the same track. So overall there is little discernible difference.
DxOMark test results for the Olympus Pen E-P3
Linear Mode
