8 ms·

Its DX0 score is 98

MBCook 9y ago
Which is meaningless. They have declared themselves the standard but that number doesn’t actually measure anything. I mean the fact they sell cameras alone means they’re not unbiased. https://daringfireball.net/linked/2017/09/22/dxo-ratings-are-horseshit https://daringfireball.net/linked/2017/09/22/dxo-ratings-are... Wait for real camera reviews.
- endorphone 9y ago
  John Gruber's stance on DXO is utterly meaningless, and it's farcical that you seriously cite it. Measuring the technical capabilities of an imaging device is patently obvious and justifiable. From dynamic range to optical and electronic resolution, noise at different light levels, etc. You can question their methodology, but saying that it "doesn't measure anything" is nonsense, and citing a guy who laughably claims that you can't measure the "art" is uproarious. Gruber has less than zero legitimacy in this discussion, and his sole motivating intention is defend Apple in all facets.
  - oflannabhra 9y ago
    Gruber's conclusion is correct, even though his reasoning is not. DxO scores are crap, because they do not adequately control scenes in their comparisons. Most DSLR reviews have a consistent scene that pushes the envelope of what a sensor can capture, and all cameras are benchmarked against that scene. Here's an example: https://www.dpreview.com/reviews/nikon-d7500-review-speed-and-capability/8 https://www.dpreview.com/reviews/nikon-d7500-review-speed-an...
    - endorphone 9y ago
      I agree entirely -- their methodology is sloppy and what they do have is poorly documented. I certainly don't hold their scores supreme or canonical. Complaining that it can be done with more rigor and consistency, and with more objective openness, is valid. Gruber's hand-wavy art complaint is not.
  - PascLeRasc 9y ago
    Did you read that blog post? He was criticizing DXO for giving the iPhone 8 the highest score ever, not defending Apple. How do you objectively measure bokeh effects? And if you're so defensive of DXO, can you explain why their photo subscores range from 51-89 but that adds up to a 96?
    - jimmies 9y ago
      Should we run it through a machine learning algorithm to see how the hell any of the "component" score adds up to a 96 overall score? And why only those components, but not others? I want my dog nose filter score in too. Snake oil is what that number is.
    - endorphone 9y ago
      Gruber had previously shat on DXO when the iPhone was bested by the Pixel. He took this opportunity to say "Oh btw the iPhone is now best...but it's still horseshit look at how unbiased I am". And I'm not "defensive" of DXO scores any more than I'm defensive of dpreview. DXO analyzes a variety of real world technical capabilities of imaging systems -- most certainly in an imperfect way and needing more rigor -- but to call it "horseshit", or to ridiculously claim that it measures nothing, is specious.
  - MBCook 9y ago
    > Measuring the technical capabilities of an imaging device is patently obvious and justifiable. From dynamic range to optical and electronic resolution, noise at different light levels, etc. I agree completely. DPreview and other sites do that. > but saying that it "doesn't measure anything" is nonsense Then please tell me what that one final number measures. How do they get to it? It doesn’t seem objective at all. Other camera review sites give a bunch of objective measurements then a subjective opinion about how the camera related to others. DXO wants you to think their arbitrary number is scientifically rigorous. > and citing a guy who laughably claims that you can't measure the "art" is uproarious. Are you claiming you can measure art? That’s what DXO seems to do. Having a single number final score means that if camera A had a higher score than B then it must take objectively better pictures. Or we have to admit their score is subjective. > Gruber has less than zero legitimacy in this discussion Why? > and his sole motivating intention is defend Apple in all facets. He’s complaining that the iPhone had the highest score ever. If you were right he would be writing about how DXO said it was the best camera ever, not that DXO is a sham. If you can’t look past Gruber’s byline I imagine there are other articles online about DXO and their questionable practices. I’ve seen others before, this was the first Google turned up.
    - endorphone 9y ago
      "Why?" Because Gruber is a biased individual whose professional existence is to pander to a narrow crowd. He has nothing broadly interesting to say about cross-cutting concerns. "He’s complaining that the iPhone had the highest score ever." He previously dismissed DXO when the iPhone was beat. Now he wants to simultaneously crow about the iPhone taking the victory while claiming that he totally doesn't care about it anyways. People buy this teenager level nonsense? "Are you claiming you can measure art?" This is absurd. DXO is a broad, generalist, imperfect measure of imaging devices, and paradoxically there is broad agreement that the devices that do really well generally are capable of the best photos. But you can take a great photo with a pinhole camera, from an artistic perspective. Does this make a pinhole camera the best? That is nonsensical. Don't believe DXO. Don't believe Consumer Reports. Don't believe metacritic or star reviews or RT or whatever. But to claim it measures nothing, or to cite some biased player, is not credible.
- Grazester 9y ago
  It might be meaningless but given the Pixel/Google's recent push with image based ai I would not be dismissive of the phone's camera prowess
  - MBCook 9y ago
    I don’t doubt the camera is good. The dual pixel thing sounds a little iffy to me in that I wonder if it can really duplicate the portrait mode of the iPhone. I’ll be very interested to see REAL reviews. DXO’s numbers mean as much as a five star system. They’re not quantitative but want to be treated like they are.
jimmies 9y ago
I own a Pixel but this DxO stuff is really dodgy. In the portrait pic of the girl, the difference between the Pixel 2 [1] and the iPhone 8 [2] is night and day, with the Pixel 2 looking like shit. She so pale, and the fake bokeh looks like a 5 year old smearing all over, it's simply not pleasant to look at. And the whole push on adding the fake bokeh """scores""" to justify changing the scores so every fucking newly released device gets the highest score is stupid. Thanks for scoring the bokeh, if I had really wanted bokeh, I would have pulled out my bulky mirrorless camera. I bet when a new phone comes out and they pay the DxO guy enough, they will add a dog nose filter score to make the new camera score the highest. 1: A https://cdn.dxomark.com/wp-content/uploads/2017/10/ref1_Bokeh-Indoor_GooglePixel2-e1506967768712.jpg https://cdn.dxomark.com/wp-content/uploads/2017/10/ref1_Boke... B https://cdn.dxomark.com/wp-content/uploads/2017/10/ref2_Lowlight_GooglePixel2.jpg https://cdn.dxomark.com/wp-content/uploads/2017/10/ref2_Lowl... 2: A https://cdn.dxomark.com/wp-content/uploads/2017/10/ref1_Bokeh-Indoor_ip8Plus.jpg https://cdn.dxomark.com/wp-content/uploads/2017/10/ref1_Boke... B https://cdn.dxomark.com/wp-content/uploads/2017/10/ref2_Lowlight_ip8Plus-e1506980596843.jpg https://cdn.dxomark.com/wp-content/uploads/2017/10/ref2_Lowl...
- deanCommie 9y ago
  Uhhhhhhhhhhh are you serious? The iPhone 8 results look SO much better...
  - jimmies 9y ago
    Exactly, the iPhone 8 pictures look so much better. You don't have to quantify shit to realize this.
- monort 9y ago
  It's post-processing, not the quality of the sensor, photo from iPhone have warmer color balance. If you compare the right eyebrow, you will see that [1] have more details, than [2]
  - jimmies 9y ago
    That might be true, but I'd say if they say their expertise is to have artificial intelligence, then by that I'd like to have the picture that is the most pleasant to look at. By that benchmark, and I can confidently say that 1 is definitely not more pleasant to look at than 2. Remember, the picture is always a depiction of what our eyes see, and that our eyes see is only a part of the reality. My eyes don't see bokehs. My eyes don't focus the same way the camera does. My eyes don't see in 3500K or 6500K or whatever the hell that is. My eyes don't care how many strands of hair there are in her eyebrows and how detailed they are. My eyes don't see in black and white, either. Yet those are what you see in pictures. Focusing on the little detail "accuracy" and forget what it really matters at the end is dangerous.

Its DX0 score is 98

Which is meaningless. They have declared themselves the standard but that number doesn’t actually measure anything. I mean the fact they sell cameras alone means they’re not unbiased. https://daringfireball.net/linked/2017/09/22/dxo-ratings-are-horseshit https://daringfireball.net/linked/2017/09/22/dxo-ratings-are... Wait for real camera reviews.

endorphone 9y ago
John Gruber's stance on DXO is utterly meaningless, and it's farcical that you seriously cite it. Measuring the technical capabilities of an imaging device is patently obvious and justifiable. From dynamic range to optical and electronic resolution, noise at different light levels, etc. You can question their methodology, but saying that it "doesn't measure anything" is nonsense, and citing a guy who laughably claims that you can't measure the "art" is uproarious. Gruber has less than zero legitimacy in this discussion, and his sole motivating intention is defend Apple in all facets.
- oflannabhra 9y ago
  Gruber's conclusion is correct, even though his reasoning is not. DxO scores are crap, because they do not adequately control scenes in their comparisons. Most DSLR reviews have a consistent scene that pushes the envelope of what a sensor can capture, and all cameras are benchmarked against that scene. Here's an example: https://www.dpreview.com/reviews/nikon-d7500-review-speed-and-capability/8 https://www.dpreview.com/reviews/nikon-d7500-review-speed-an...
  - endorphone 9y ago
    I agree entirely -- their methodology is sloppy and what they do have is poorly documented. I certainly don't hold their scores supreme or canonical. Complaining that it can be done with more rigor and consistency, and with more objective openness, is valid. Gruber's hand-wavy art complaint is not.
- PascLeRasc 9y ago
  Did you read that blog post? He was criticizing DXO for giving the iPhone 8 the highest score ever, not defending Apple. How do you objectively measure bokeh effects? And if you're so defensive of DXO, can you explain why their photo subscores range from 51-89 but that adds up to a 96?
  - jimmies 9y ago
    Should we run it through a machine learning algorithm to see how the hell any of the "component" score adds up to a 96 overall score? And why only those components, but not others? I want my dog nose filter score in too. Snake oil is what that number is.
  - endorphone 9y ago
    Gruber had previously shat on DXO when the iPhone was bested by the Pixel. He took this opportunity to say "Oh btw the iPhone is now best...but it's still horseshit look at how unbiased I am". And I'm not "defensive" of DXO scores any more than I'm defensive of dpreview. DXO analyzes a variety of real world technical capabilities of imaging systems -- most certainly in an imperfect way and needing more rigor -- but to call it "horseshit", or to ridiculously claim that it measures nothing, is specious.
- MBCook 9y ago
  > Measuring the technical capabilities of an imaging device is patently obvious and justifiable. From dynamic range to optical and electronic resolution, noise at different light levels, etc. I agree completely. DPreview and other sites do that. > but saying that it "doesn't measure anything" is nonsense Then please tell me what that one final number measures. How do they get to it? It doesn’t seem objective at all. Other camera review sites give a bunch of objective measurements then a subjective opinion about how the camera related to others. DXO wants you to think their arbitrary number is scientifically rigorous. > and citing a guy who laughably claims that you can't measure the "art" is uproarious. Are you claiming you can measure art? That’s what DXO seems to do. Having a single number final score means that if camera A had a higher score than B then it must take objectively better pictures. Or we have to admit their score is subjective. > Gruber has less than zero legitimacy in this discussion Why? > and his sole motivating intention is defend Apple in all facets. He’s complaining that the iPhone had the highest score ever. If you were right he would be writing about how DXO said it was the best camera ever, not that DXO is a sham. If you can’t look past Gruber’s byline I imagine there are other articles online about DXO and their questionable practices. I’ve seen others before, this was the first Google turned up.
  - endorphone 9y ago
    "Why?" Because Gruber is a biased individual whose professional existence is to pander to a narrow crowd. He has nothing broadly interesting to say about cross-cutting concerns. "He’s complaining that the iPhone had the highest score ever." He previously dismissed DXO when the iPhone was beat. Now he wants to simultaneously crow about the iPhone taking the victory while claiming that he totally doesn't care about it anyways. People buy this teenager level nonsense? "Are you claiming you can measure art?" This is absurd. DXO is a broad, generalist, imperfect measure of imaging devices, and paradoxically there is broad agreement that the devices that do really well generally are capable of the best photos. But you can take a great photo with a pinhole camera, from an artistic perspective. Does this make a pinhole camera the best? That is nonsensical. Don't believe DXO. Don't believe Consumer Reports. Don't believe metacritic or star reviews or RT or whatever. But to claim it measures nothing, or to cite some biased player, is not credible.
Grazester 9y ago
It might be meaningless but given the Pixel/Google's recent push with image based ai I would not be dismissive of the phone's camera prowess
- MBCook 9y ago
  I don’t doubt the camera is good. The dual pixel thing sounds a little iffy to me in that I wonder if it can really duplicate the portrait mode of the iPhone. I’ll be very interested to see REAL reviews. DXO’s numbers mean as much as a five star system. They’re not quantitative but want to be treated like they are.

jimmies 9y ago

I own a Pixel but this DxO stuff is really dodgy. In the portrait pic of the girl, the difference between the Pixel 2 [1] and the iPhone 8 [2] is night and day, with the Pixel 2 looking like shit. She so pale, and the fake bokeh looks like a 5 year old smearing all over, it's simply not pleasant to look at. And the whole push on adding the fake bokeh """scores""" to justify changing the scores so every fucking newly released device gets the highest score is stupid. Thanks for scoring the bokeh, if I had really wanted bokeh, I would have pulled out my bulky mirrorless camera. I bet when a new phone comes out and they pay the DxO guy enough, they will add a dog nose filter score to make the new camera score the highest. 1: A https://cdn.dxomark.com/wp-content/uploads/2017/10/ref1_Bokeh-Indoor_GooglePixel2-e1506967768712.jpg https://cdn.dxomark.com/wp-content/uploads/2017/10/ref1_Boke... B https://cdn.dxomark.com/wp-content/uploads/2017/10/ref2_Lowlight_GooglePixel2.jpg https://cdn.dxomark.com/wp-content/uploads/2017/10/ref2_Lowl... 2: A https://cdn.dxomark.com/wp-content/uploads/2017/10/ref1_Bokeh-Indoor_ip8Plus.jpg https://cdn.dxomark.com/wp-content/uploads/2017/10/ref1_Boke... B https://cdn.dxomark.com/wp-content/uploads/2017/10/ref2_Lowlight_ip8Plus-e1506980596843.jpg https://cdn.dxomark.com/wp-content/uploads/2017/10/ref2_Lowl...

deanCommie 9y ago
Uhhhhhhhhhhh are you serious? The iPhone 8 results look SO much better...
- jimmies 9y ago
  Exactly, the iPhone 8 pictures look so much better. You don't have to quantify shit to realize this.
monort 9y ago
It's post-processing, not the quality of the sensor, photo from iPhone have warmer color balance. If you compare the right eyebrow, you will see that [1] have more details, than [2]
- jimmies 9y ago
  That might be true, but I'd say if they say their expertise is to have artificial intelligence, then by that I'd like to have the picture that is the most pleasant to look at. By that benchmark, and I can confidently say that 1 is definitely not more pleasant to look at than 2. Remember, the picture is always a depiction of what our eyes see, and that our eyes see is only a part of the reality. My eyes don't see bokehs. My eyes don't focus the same way the camera does. My eyes don't see in 3500K or 6500K or whatever the hell that is. My eyes don't care how many strands of hair there are in her eyebrows and how detailed they are. My eyes don't see in black and white, either. Yet those are what you see in pictures. Focusing on the little detail "accuracy" and forget what it really matters at the end is dangerous.