DxOmark weird results. Pentax K-5 beats the Sony A55 (and the Canon 5DmarkII)
So now I am really to start to be worried about their DxOmark results. They just published the Pentax K-5
I understand that the imaging processor are different but I don’t know if that alone can explain the huge gap!
Via DxOmark
mike
3 years ago |Those scores are stupid and anyone who puts any stock into them is an idiot.
Only way to judge a camera is to use it.
If you haven’t used it you can’t judge it, compare it, bitch about it or praise it.
And I guarantee no one reading this site could ever notice a difference between any of the above listed cameras when it comes to color depth or dynamic range.
Eric
3 years ago |I agree, DXO marks are one of the more worthless things I’ve seen in a while; and this is coming from Pentax fanboy (my first camera in college was a K1000, so I’ve always had a soft spot for Pentax). Cameras have been so good for so long now these lab tests are totally useless to me. There hasn’t been a DSLR/EVIL produced in the last 2 years that can’t produce professional results if you know how to use it. This crap is starting to remind me of golfers buying new clubs instead of going to the driving range. If anyone actually worries about DXO marks when buying a camera I’d be almost willing to bet they’ve never taken a photography course in their life.
Sky_walker
3 years ago |Well, I’m not surprised really.
Do you remember back 2 years ago when Sony and Nikon shared the same sensors and how the noise looked? Nikon was mile ahead.
Now we see similar thing with K5 (and most likely: D7000).
Congrats for Pentax! (Sony – buy this damn company!
)
john
3 years ago |Congrats for Pentax! (Sony – buy this damn company!
That would be WAAAAY cool…slap an A-mount on the K5
Looks like a REAL serious Photographers camera( weather sealed camera and lenses)
Keep all their R&D, Market share etc + Medium format.
Sony could offer Pentax the full frame they’ve been waiting for.
Not sure if Sony could support 3 mounts though.
nuxar
3 years ago |Sony to buy? Crazy idea! If Sony did they would probably made it and SLT, SLU, SLV… anything but not DSLR… better Pentax keep away from Sony… better for us.
john
3 years ago |its not like Pentax can buy Sony..:-)
Simon_P
3 years ago |is that because of the translucent mirror (remember…you can remove it for taking pic)
Simon_P
3 years ago |*cannot
Luis
3 years ago |The difference ISO performance between the A55 and the K-5 is about 1/3 of an f-stop. It is due to the translucent mirror. Expected.
The ISO performance of the K-5 is not as good as the 5DmkII (1162 vs. 1815). Also expected.
The very high dynamic range of the Pentax is a bit surprising, but looking at what the K-x showed with the old sensor, it’s not totally unexpected either.
This new Sony sensor seems to be very good (we’ll see the A580 and D7000 which use the same one and also without the translucent mirror).
GF1
3 years ago |the real story here is that GH2 sensor have lower score than GH1 sensor lol
That’s what you get INTO the MP war panasonic , GH2 sensor use just so many MP in a small area
i thought GH2 suppose to be the best m43 sensor?
Carlos Yamada
3 years ago |You based your opinion with numbers. Search for video files and you will see the GH2 sensors is a big improvement from GH1.
You opinion about GH2 sensor in a small area is ridiculous.
kevin
3 years ago |14 bit vs 12 bit A/D conversion.
qbic
3 years ago |Sony sucks with 12bit….
Jayz
3 years ago |Sony has always used 12 bit A/D conversion, everything from A100 to A900 and in between. I actually don’t get this one; every other manufacturer that uses Sony sensors clock it at 14 bit conversion. So if Sony sensors were designed for 14 bit outputs, why have they never bothered to use it?
Seems dumb to offer a competitive advantage to other companies in this fashion.
john
3 years ago |funny DxOMark scores Pentax 645D ( medium format) at 12.6
Launch date 2010-03-01
john
3 years ago |that’s 12.6 Evs dynamic range
http://www.dxomark.com/index.php/eng/Camera-Sensor/All-tested-sensors/Pentax/645D
compared to 14.1 of the K5
http://www.dxomark.com/index.php/eng/Camera-Sensor/All-tested-sensors/Pentax/K5
mugen
3 years ago |a580 will be close to K-5
Ion
3 years ago |Probably mirror issue looses 1/3 light
Don
3 years ago |The Pentax is the flagship, so it was to be expected. Being someone who actually uses Pentax (in fact, the reason I got into NEX was so I could use it as a bridge between alpha and k-mount glass), I’m not too surprised, but I also agree with the top comment that DxO isn’t that reliable as a tell-all for the camera. For those that do take this seriously: I’m sure once the A77/7xx comes around (Sony’s respective flagship) that’s when all bets are off.
As for the Canon thing, I never was too impressed at the Canon sensor compared to Nikon’s full-frame rendering, but I can’t forget that the 5D is starting to get pretty old now, so it was only natural that something finally outscores it.
Jayz
3 years ago |which explains the rumored 5D mark 3 and updated 24-70 L II
P_Osh
3 years ago |I wonder where they measure resolution. That’s where the Canon 5 Mark II should score.
Dana Kincaid
3 years ago |The Pentax sensor is newer by two years.
Antonio Rojilla
3 years ago |LOL! So DXO was great when they measured the A55 and is faulty now? LOL!
http://www.sonyalpharumors.com/the-sony-a55-has-the-best-aps-c-sensor-says-dxomark/
john
3 years ago |LOL! So DXO was great when they measured the A55 and is faulty now? LOL!
as stated above by several it is understandable it beats the A55 in iso score because of loss of light from translucent mirror.
The part i find weird is that in dynamic range at 14.1, it is not only the best aps-c camera but is higher than any full frame or even MEDIUM FORMAT.
Most dynamic range ever!
Phase One P40 Plus only scores 13.0
iso range 50-3200
smells funny to me
santela
3 years ago |told u those results are bs.
i bet the nikon d7000 is gonna beat the a55 also.
kmfan
3 years ago |Sensor scores are only one part of the whole picture.
An example, the D700 sensor scores are fantastic; BUT,the tested lens scores of the D700 is only at the level of 5D MkI.
It’s great to know that APS-C sensor development are progressing rapidly. Who knows, maybe in two years APS-C sensors will outperform cropped MF digital sensors? The sensor part at least, the lens and resolution is another story.
b shaw
3 years ago |Kudos to Pentax!
I want Pentax to succeed and to offer a great product … because it puts pressure on Sony to improve Sony’s products. (For that reason, I want Nikon and Canon to continue to succeed, too.)
The sensor scores generally make sense when you read the details on how they score them. Non-readers will just be flustered. That being said, there is one part I haven’t been able to understand, the math that transforms the individual scores into the overall score – In the past, it appeared to be slightly inconsistent – but only a matter of a few points, usually less than 5. However, I haven’t looked at those details for a year or so; they may have fixed the issue. Note: They openly admit to skewing the medium format scores. You’ll have to read their site to understand why they do that.
Regarding how the ‘same’ sensor can get a better score in a different camera:
As someone mentioned earlier, Nikon has better scores with Sony sensors than Sony gets with their own sensors. Why? Nikon integrates the sensor differently than Sony, for example they use 14 bit, as someone mentioned. Also, on his website, DK mentioned that Nikon has a slightly different scheme for handling ISO settings and that has helped Nikon tremendously. There could be other differences – for all we know, they may use different material for their Anti-aliasing filter that could contribute to a better performance. Who knows?!
We shouldn’t assume the sensors are completely identical – there is probably some small difference that Nikon and/or Pentax require in their contract with Sony. I doubt if either Nikon or Pentax bought sensors exactly “off the shelf”. They probably have some small changes, but still takes advantage of Sony’s basic sensor architecture.
In this case, it looks as if Pentax has some very smart engineers who put together a great camera. Kudos to Pentax.
Joel
3 years ago |Actually (and this is one of the rare times), I want Sony to be a very popular SLR format. The reason is that there are definite advantages for going with a popular system… renting/borrowing lenses/accessories, better 3rd party support and documentation, and a large used market. Sony is short on all three, and I realized that was something I was going to have to give up going from the D80 to the A55.
But
3 years ago |I see the pixel pitch of K-5 and A55 are different. This means they are in fact not the same sensor as micro metre differences can make things slight changes in the resolution. If these values are not measured by Dxo and are actual values, that means Pentax sensor might have their own specifications although it might have come out of the same Sony factory. Will this be the reason? I do not know. But at least it tells us that they are probably not the same sensor after all.
Toomas Kadarpik
3 years ago |Good results, first APS-C low light iso over iso 1000 and with 16 Mpix, congratulations Pentax and hopefully competition heats up. Did some testing for interior, dynamic range really rocks. K-5 gives iso 6400 images from ordinary interior home lighting.
Same sensor is working just differently, sensor is just matrix of photodiodes and this particular has also A/D on board. Pentax and nikon add their on AA filter, IR filter, Color filters, electronics witch programs and reads A/D, sophisticated algorithms and so on.
Neonsquare
3 years ago |That the pixel pitch is shown as different must not mean, that it really is a different sensor. They may have just calculated it based on sensor size and used pixels there are always unused pixels at the borders of the sensors. The pentax has a small amount of more pixels in width and heigth specs.
daniel
3 years ago |the a55 has a translucent mirror which redirects 33% of the light, ergo, even though the sensor is the same, on the a55 it would need more light to have comparable results.
The a55 also uses different image stabilization, which requires a larger margin of pixels around the usable area than the k5, which means they can use a few more pixels. As far as it beating the 5D, that’s not a surprise, since the sony sensor is a new generation of sensors.
I think if sony got their act together and build a camera with their current technology, sony could have easily made a comparable camera to the k5. instead they are discontinuing the professional and semiprofessional cameras before they have a replacement, while flooding the market with sub 7xx cameras.
Daddy
2 years ago |I’m extremely inspired with your writing skills and also with the structure for your blog. Is this a paid subject matter or did you modify it yourself? Anyway keep up the excellent quality writing, it’s uncommon to peer a great blog like this one nowadays..