E-M5 video quality
My friend Andrew from EosHD (Click here) agrees that the E-M5 is an absolute great camera. But he focused on the video quality of the camera and writes: “The Olympus approach to video on the other hand is the opposite of masterstroke. If this camera was hacked to support high bitrate 24p it would be a GH2 beater. Given the shear responsiveness of the camera, the speed of the AF, the fast scan of the sensor for minimal rolling shutter and the high resolving power in video mode – it clearly packs some powerful hardware under the hood and a very capable sensor. It is just a shame we yet again don’t see the best of it in video mode because of decisions made in firmware.“. Agree, Olympus could have done better! They could make it good by releasing a firmware update which increase the video quality. Are you listening Oly? 🙂
E-M5 sensor with moirè?
The french magazine Chassimages claims to have found the answer about the E-M5 sensor. Pierre Lepidi (Click here) posted the text from the magazine (you can’t find the Chassimages text online): “its the same GX1 sensor but what changes is that the E-m5 sensor has a lighter Anti Aliasing filter. Images are sharper although you have a higher risk to find moiré“. This seems to be confirmed by the comparison posted by DSRLcheck (Click here). There is visible more reoslution and moirè when comparing the E-M5 with the GX1.
E-M5 news roundup:
TVsmith (Click here) has been asked at the race track if the OM-D can do action… The pictures may show you the answer 😉 OM-D in-body stabilisation versus Pana OIS @ 300mm at Snappydiablog (Click here). OM-5 initial reaction at Footlessfish (Click here). The latest Apple RAW update brings support for the E-M5 and GF5!
E-M5 third party battery:
A review of the third party E-M5 battery at Clubsnap (Click here): “These batteries will only hold 84% of the charge of a real Olympus 1220mAh battery. (Assuming you get 250-shots out of an original battery, this will give you 209-shots before it gives up on you.) It is also only 70% of the claimed 1450mAh.“. You can buy those batteries on eBay:
[shoplink 30960 ebay]One Olympus batteries and NO(!) charger (Click here to check the price on eBay). You need to have a compatible charger![/shoplink]
[shoplink 30954 ebay]One Olympus battery + 1 chargers (Click here to check the price on eBay)[/shoplink]
[shoplink 30955 ebay]Two Olympus batteries + 1 chargers (Click here to check the price on eBay)[/shoplink]
[shoplink 30961 ebay]Three Olympus batteries + 1 chargers (Click here to check the price on eBay)[/shoplink]
James: “Hi, I see you have a lot of love for the Olympus Om-D, so do I. If your readers like they can subscribe to my facebook as I’m travelling around Europe for the rest of the year with my OM-D, 7-14mm, 12mm, 25mm & 45mm. I’m trying to post photos for my friends and family as much as possible. Just thought it might be something your readers could be interested in. the url is www.facebook.com/coxy84”
E-M5 orders (Click on the names of the stores to see the product page):
Black E-M5 body at Amazon, Adorama, BHphoto, J&R, Warehouse UK, Digitalrev and Amazon Japan.
Silver E-M5 body at Amazon, Adorama, BHphoto, J&R, Warehouse UK, and Amazon Japan.
Black E-M5 body with 14-42mm lens at Amazon, Adorama, BHphoto , Digitalrev and J&R.
Black E-M5 body with 12-50mm lens at Amazon, Adorama, BHphoto, J&R, Jessops, Amazon UK, Amazon Deutschland , Digitalrev and Amazon Japan.
Silver E-M5 body with 12-50mm lens at Amazon, Adorama, BHphoto, J&R, Jessops, Amazon UK, Amazon Deutschland and Amazon Japan.
39 Responses
OK, very handy test by our friend Voiceoverman……I have been thinking of getting a longer lens…but now I might just get the E-M5 earlier than planned. I have a 200/F3 that would be fun at 1/80th! Very rarely is it sharp, handheld on the E-PL1.
How did Voiceoverman manage to take these shots in the future (23rd November 2012)? I presume his E-M5 also has a future Firmware update! 🙂
thats the time travel option buried deeply in the menu.
@ GreyOwl
“Use the Force, Luke.” (the mft Force)
I could tell you how it’s done but then I’d have to kill you.
If EM-5 indeed use G3 and GX-1 sensor then that’s explain why EM-5 doesn’t include the 24P as an option in movie mode. Maybe Panasonic forbid it for avoiding EM-5 in head to head with GH-2 or G3. Anyway an article in UK fourthird user pretty sure that the sensor is different since Panasonic sensor division is different from camera division and they never mind if Olympus use the 16 MP sensor since PEN third generation and the insider from Panasonic also confirm that EM-5 using different sensor.
I’ve used the E-M5 quite a bit for action already with the Panny 100-300mm! I wish that lens was as fast as the Zuiko 50-200, or that the Zuiko 50-200 worked worth a damn on the m4/3 bodies…
I can confirm the moire.
When I do macro I often have trouble with moire
on dragonfly compound-eyes.
Besides that issue, the E-M5 is great.
i think moire can be reduced during shooting if you try to rotate a bit along the z-axis, so shoot more of the same but at at slightly different angle, but this is just theory havent tried it my self, but its the technique i used during scanning magazine photos.
You could claim a new species? You could name it olymoir.
That Angle taken for the E-M5 just look like a Machine Gun XD
moon shoot 75-300mm M43 lens… with Em5 body
http://www.facebook.com/photo.php?fbid=3670832082389&set=p.3670832082389&type=1&theater
Slightly OT?
I think the E-M5 dynamic range measurements done so far tends to show that it is not the G3/GX1 sensor. Dxo will give a final answer by showing signature curves…
E-M5 sensor = GX1 sensor? But what about the very significant difference in Dynamic Range? E-M5 has significantly more range in the highlights (2/3 EV) and in the shadows (1/3 EV). It has also been shown that lifting the shadows in the E-M5 at base ISO can be pulled off much better than in the GX1 and GH2.
Not that it matters much who makes the sensor, or how they do it: the E-M5 easily gives the best m4/3 IQ so far: ‘fingers in ze nose’. That’s what counts.
DR comparison of Nikon D3X v.s. Sony A900:
http://goo.gl/BT9sQ
Sensors in D3X and A900 are well known of the same. But from DXO’s measurement, there is about 1EV DR difference.
Good example, and there are several examples of cameras using the same or very closely related sensors that still measures differently. Usually not by as much as 1EV, but alot of RAW-files have been processed to some degree which can explain differences. Sony got at lot of negative press for the A700 which had some NR applied to the RAW-files and I think they came out with a firmware update that made it optional (could be wrong, it’s several years ago).
Anyway, I wouldn’t rule out that it’s a G3/GX1 sensor in E-M5 and it will be interesting to see what DxO find.
If one outputs 12bit RAW and other 14bit RAW sensors are hardly same.
Also live view of Nikon (lacked by A900) gives its own requirements for sensor.
So while light gathering photosite itself might be same otherwise Nikon has different sensor with obviously improved read out circuitry and ADC.
Also Nikon’s lower base ISO helps to give higher DR.
D300 could do both, do you think there are two sensors in there?
Olympus do something different from Panasonic even if it’s the same sensor. I can’t say if it’s a different sensor or not, but I certainly can’t rule out that it’s basically a GX1-sensor based on what I’ve seen of reviews so far.
Nikon D3X (or A900) is analog output sensor.You can use 12-bit, even higher than 14-bit on-board ADC depends on sensor output signal quality and cost.If sensor output noise is larger than specified value, ADC higher than 14-bit is meanless.
And ISO sensitivity measurement of sensor depends on the exposure to reach sensor saturation. Some manufacturer may do design choice to avoid saturation in higher exposures, but main structure of sensor still the same.
D3X & A900 are analogue output sensor, which bits output of RAW depends on on-board ADC, not sensor.
Nikon using 14-bit ADC means that the sensor output quality is better than 12-bit ADC requirement. But for cost or other concerns, Sony only uses 12-bit ADC.
And sensor ISO sensitivity measurement depends on exposure to reach sensor saturation. Some manufacturers make design choice to keep some margins to avoid saturation in high exposure, and cause lower base ISO. For product management, same structure but different design choice usually have same product code, and called “same product” but provide to different customers in bussiness.
The basic sensor may be the same but the electronics that control sensor output may be different. They could be different under a turnkey supply contract that specifies a change from the original manufactures specification and it’s just bolted in place. They could be different because the sensor was built upon by the purchaser using a different methodology. They will also have different firmware driving them.
Car engine makers might make an engine that is used by a family car. Sometimes that engine is used by another company and has some specifications changed that change its characteristics and it is put into a sports car. Basically the same but different.
Same can be applied to sensors, memory and much much more.
Por fin se han dado cuenta los que están probando la Olympus E-M5 que es una maravillosa cámara de fotos pero de video le falta mucho para compararse con la GH2 de Panasonic. Le falta mejorar el códec de video, un mayor britate 35 Mgps sería el ideal, e implementar en la cámara los 24p y 25p que la harían realmente profesional para los cineastas. Tiene mayor nitidez que la Panasonic GH2 porque el filtro que evita el moiré es más suave.
Por favor Vitaily hackea esta cámara que los de Olympus no hacen nada por mejorarla.
http://www.eoshd.com/content/8149/olympus-om-d-e-m5-first-impressions
por favor hablar en ingles porque la mayoría de personas aqui son internacionales
Coño
OM-D in-body stabilisation versus Pana OIS @ 300mm at Snappydiablog
Wait a moment.
Wasn’t the whole premise of the in-lens IS that it can be finely tuned to the given focal length(s)?
Yet general purpose IBIS wins over the in-lens IS??
> There is visible more reoslution and moirè when comparing the E-M5 with the GX1.
I call it BS. I can’t believe that simple AA-filter tweaks (or even manufacturing tweaks) can be responsible for greater push-ability of the RAWs. Even if this is a G3/GX1 sensor, it is definitely manufactured using a better facility than that of Panasonic. But G5 and GH3 releases would tell us more. Anyway I’m happy that Oly finally managed to catch up in the sensor area. Basically, from now on, one can disregard the sensor differences when comparing m43 cameras.
OM-D in-body stabilisation versus Pana OIS @ 300mm at Snappydiablog
Wait a moment.
Wasn’t the whole premise of the in-lens IS that it can be finely tuned to the given focal length(s)?
Yet general purpose IBIS wins over the in-lens IS?
There is visible more reoslution and moirè when comparing the E-M5 with the GX1.
I call it BS. I can’t believe that simple AA-filter tweaks (or even manufacturing tweaks) can be responsible for greater push-ability of the RAWs. Even if this is a G3/GX1 sensor, it is definitely manufactured using a better facility than that of Panasonic. But G5 and GH3 releases would tell us more. Anyway I’m happy that Oly finally managed to catch up in the sensor area. Basically, from now on, one can disregard the sensor differences when comparing m43 cameras.
Is this from your twin or is the preseeding one from your twin?
What sensor? The Oly may have the same physical sensor, and there are tweaks in firmware and image processing (+AA-filter) that enhances IQ. – Normal that the latest version is better.
Regarding DR the above is applicable + reminding you Pekka Potka mentioning the jpg-engine also contributes to the improved DR.
It is also obvoius that the RAW files aren’t as raw as we believe; some “magic” is applied here also.
The sensor + processing trend is, and always was, more and more pixels + better pixel quality. – And it will stay that way. (Not applicable to compacts though)
> The sensor + processing trend is, and always was, more and more pixels + better pixel quality. – And it will stay that way. (Not applicable to compacts though)
Except that sensor tech would have to develop noticeably just to keep pixel quality same when cramming more Marketing Pixels into same area.
Laws of physics don’t stop working when class of camera is changed from compact to system camera and effect of decreased pixel signal quality shows most at where manufacturers want to give impression newer and higher MP count sensors are better: At higher ISOs.
If you look DxOMark’s actual measurement results instead of dumbed down scores you’ll notice that.
For example 16MP Sony NEX-5N has higher SNR than 24MP NEX-7 at same sensor sensitivity. (so does even old tech 14MP NEX-5)
And if we disregard DxOMark as synthetic testing same shows also in DPreview’s studio scene shot comparison widget.
At 6400 NEX-7’s heavier NR processing makes image in overall no more detailed than NEX-5N’s (despite of 50% more pixels) and at 12800 NEX-5N has more lower contrast details. With non-NR’ed RAW image quality difference is even more noticeable with noise literally flowering lot more in higher MP count sensor image.
There simply isn’t any substitute for pixels having adequate quality to start with: You can’t make party cake out of shit no matter how much of it you have.
If sensor fundamental technology stays the same then what you say is mostly correct and they will creep along getting better in a slower and slower manner. However the ways of putting it together may change and the quality take a leap forward. Likewise a new method of capturing an image electronically may suddenly make an entry. (maybe shit on a cardboard base, funny, but stranger things have happened).
When valves were used people were throwing $m into improving them then along came the transistor out of left field. Millions were poured into improving aero piston engine fuels (lead). Then along comes the jet running on kero. The CRT was used for many years as the visual output of TV and computers, now gone for general use, replaced by a later technology.
Who knows what tomorrow may bring until it gets here.
What if the Oly sensor is the first of something entirely new, it has been released now because it has a slight edge, yet has more room for improvement than the current crop of offerings. Highly unlikely but…..
DxO will have to change their whole methodology or will they?
Panasonic [cameras arms] did already vaguely admit it must have been a Panasonic [semiconductor arm] sensor.
Although those huge reported improvements in DR could suggest otherwise. Will wait for dxo for further tests but the final word will come from Chipworks or whoever else is the first to tear the thing apart and look at the chip 🙂
i would love to see a 24p hack, but those guys can only hack it to increase bitrate. remember all those guys who wanted 25p or 30p for the gh2? they didn’t get it until panasonic released an official firmware update supporting those modes.
so until olympus gets off their ass and give us official support, i doubt you’ll see 24p/25p or anything else.
I really wish Olympus had included 24/25p and variable bit-rates, in the OMD EM-5.
Is it really possible to fix this in firmware? Can a hack really provide this???
It seems like the only aspect of this camera which is a bit of a let down. Everything else about it is completely awesome.
Please, Olympus, fix this!!!
Those French fries aren’t proving anything else than lack of knowledge.
Anti-aliasing filter is entirely different part from sensor semiconductor itself and can be even left out completely like Nikon did in D800E which is otherwise identical to D800 except for lack of AA-filter in front of sensor.
And if you’re interested what lack of AA-filter causes second picture in here is extreme example about colour artefacts caused by moiré:
http://sulanto.blogspot.com/2012/04/nikon-d800e-onko-se-jarkeva-hankinta.html
There may also be other factors. I get moire when viewing some shots at a certain size, change the size and they disappear, go back to the original size and they are back. Symptoms seem to be the same as camera induced moire, maybe just a frequency v software v codec thing more than we realise.
I was thinking about this some further and it could be the GX1 sensor – if and only if, the GX1 sensor has DR like the GH2. Would like to see how the EM5 really scores on DXo. Would also love someone to send me a side by side shot with the GX1 and EM5.
The only thing keeping me from the Oly OMD owner’s club is the poor movie mode. I’m hoping they come around and offer more frame rate options and increased bit rate encoding, but it seems pretty unlikely.
http://www.youtube.com/watch?v=qN-oIcSTfHA
BAD VIDEO COMPRESSION
OLYMPUS TO INCREASE TO 35 MBPS BITRATE LEAST. AND PUT THE 24P and 25P URGENTLY WITH A NEW FIRMWARE.