-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Canon G1X gets DxoMarked. Total score of 60.
- Thread starter Jorginho
- Start date
robert1955
Veteran Member
- Messages
- 7,424
- Solutions
- 10
- Reaction score
- 4,964
That seems to be about [camera generated] jpg's only
robert1955
Veteran Member
- Messages
- 7,424
- Solutions
- 10
- Reaction score
- 4,964
Tried with several combinations and:
a) can't get the nikon to score higher than a few of the MF backs
b) can't make it score higher than the 5D2
c) comes about even with the 7D; fits with other non-fanboi observations
a) can't get the nikon to score higher than a few of the MF backs
b) can't make it score higher than the 5D2
c) comes about even with the 7D; fits with other non-fanboi observations
....
Faintandfuzzy
Veteran Member
- Messages
- 3,199
- Reaction score
- 6
It's the Nikon 5100...not 3100. I was typing too fast. Now from the list on the DxO site, you can see the Nikon D5100, Pentax K5 and Nikon D7000 all rank higher than the 5D2 and MFDB.
In fact, the Sony Nex5n is better than the MFDBs. Yup, try that in print and see what the comparison really is!
Faintandfuzzy
Veteran Member
- Messages
- 3,199
- Reaction score
- 6
Figured out that I mistyped Nikon 3100 instead of 5100?
With that in mind, you can now look at the overall ranking list for the following ranking number and camera:
- 7 Pentax K5
- 8 Sony Nex 7
- 12 Nikon D7000
- 13 Sony a580
- 14 Nikon D5100
- 15 Canon 5D2
Now, you need to get back to reading up on how DOF works!
bobn2
Forum Pro
LOL. Think I'll just stick with the physical arguments for the moment. Metaphysics is for forum 1057.
--
Bob
bobn2
Forum Pro
Not a reasonable assumption if he was simply extrapolating from the general quality of argument.
OK, so the task at hand is explaining to you why the D5100 is rated higher by DxO than the
7D: Simple. It's better. It has about the best APS-C sensor there is. The 7D doesn't even have Canon's best APS-C sensor. Still onto the measurements.
7D(68) vs D5100 (80): The D5100 wins on colour depth (+1.5 bits) - we discussed the issue of canon's CFA before, DR (+1.9 stops) - well that Sony sensor has the best DR around and the 7D doesn't, Low light (+329 ISO) - we're seeing the effects of the D5100's slightly larger and more efficient sensor. No error there, the D5100 will give better IQ than the 7D in just about every circumstance, as your test (if you ever did it) would show.
5DII(79) vs D5100 (80): this is in the noise. But, looking at the figures, it's the DR that's doing it, with the D5100 cashing in on its 1.7 stop DR advantage. Now, you might argue that DxO over emphasises the DR, but there is no arguing that the D5100 whacks the 5DII when it comes to DR (in fact even more so than the figure shows, since the D5100 doesn't have the banding issues which limit the 5DII's usable DR)
as for the MF backs, not sure which ones you are referring to, but they'll be losing in the DR and low light department, and again, there is really no arguing that if you are in a low light situation, the D5100 is a better pick than most MF backs.
--
Bob
You mistyped it (several times in several threads) after three people, myself included, told you that the numbers were 79 vs. 67?
BTW, the D5100 does have a better sensor than the 5D2, just not enough of it.
Great Bustard
Forum Pro
- Messages
- 28,369
- Solutions
- 17
- Reaction score
- 34,046
Yep. People still haven't figured out the difference between quality and quantity, and that FF often obtains its superior results from the brute force of more size as opposed to the finesse of superior tech.BTW, the D5100 does have a better sensor than the 5D2, just not enough of it.
They have a big fat disclaimer (at least fat enough for everybody I know that have read their results page to notice) that they estimate what the noise would have been before the noise reduction applied by Nikon.
Should they add something like the license agreements where you have to confirm to have the disclaimer before you can see their estimated results?
I mean how do they know the t-stop of compact camera lenses? They don't, they do not remove the lens from fixed lens compacts, and they thus have to estimate the t-stop. Still makes their results useful, just with a larger error margin.
Faintandfuzzy
Veteran Member
- Messages
- 3,199
- Reaction score
- 6
And some people haven't figured out that the differences of this "brute force" rarely show if ever show in a print. That is why the tests that I posted as well as those of others conformed that even at print sizes of 16x24 and sometimes even 20x30, the differences were invisible. In fact, any differences that do exist are smaller than those of paper differences and post processing differences.
Yep. People still haven't figured out the difference between quality and quantity, and that FF often obtains its superior results from the brute force of more size as opposed to the finesse of superior tech.BTW, the D5100 does have a better sensor than the 5D2, just not enough of it.
That is my point. And that is why when people like Peter make claims of "huge" differences in resolution for landscape prints...in the order of 50% or more....I ask him to post side by side comparison images like I have showing the huge difference. To date....he hasn't.
That is the main reason for his animosity...he resents being proven wrong in a forum. I'm still waitng for these landscape images showing the huge advantage of 50% or more that makes the FF image "pop," etc.
Many have posted samples showing there isn't a visible difference in print. Not one of these FF zealots buzzing around every 7D post have ever provided a sample....just lots of DxO graphs....LOL
You are lying, again. I said "up to 50%" with the upper bound attained with wide open primes. I put the lower bound around 20% - and this is what you get when you shoot landscapes (but still matters).
And I am still waiting for those detailed landscape images from your 7D. Or any image that you took which is not a complete crap.
I have posted quite a few - mine and other people crops. You are lying again.
Does this qualify as a posted example?
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=116&Camera=453&Sample=0&FLI=2&API=0&LensComp=116&CameraComp=474&SampleComp=0&FLIComp=2&APIComp=0
Or maybe this:
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=335&Camera=474&Sample=0&FLI=0&API=0&LensComp=674&CameraComp=0&FLIComp=0&APIComp=0
photo nuts
Senior Member
Unfortunately, it's not as simple and rosy as that. DXOMark reviewers take those assigned scores SERIOUSLY. They even quote the scores when they compare one camera sensor against another, meaning those scores are not used only to appeal to simpletons.
Here is the latest example:
http://www.dxomark.com/index.php/Publications/DxOMark-Reviews/Canon-PowerShot-G1X-Review/Canon-G1X-the-best-compact-cameras
"When it comes to the G1X’s performance, there is no contest here:
A DxOMark score of nearly 20 points more than its closest competitor."
There you go. DXOMark takes those scores seriously even though they are well aware of their pitfalls. Can we therefore blame the 'simpletons' from quoting those numbers blindly when the reviewers themselves do that?
But DxOmark also lay bare all their findings. There is no reason to criticise a newspaper over its editorials, if you don't like them, just stick to the reporting.
photo nuts
Senior Member
Will there be a part 3 to your series on FF vs APS-C?
BTW, to be fair, there's one aspect APS-C cannot compete against FF: DOF control.
BTW, to be fair, there's one aspect APS-C cannot compete against FF: DOF control.
GordonBGood
Veteran Member
The Nikon V1/J1 have a low light score of about ISO 350. That is well below where the compulsory raw Noise Reduction was determined to kick in at ISO 800 and therefore would not affect that score, nor would it affect the Color Depth or Dynamic Range scores which are determined at lowest ISO.
Regards, GordonBGood
Sosua
Senior Member
And realize what they are commenting on.
Honestly, read, listen, learn - all the info is there.
DXO overall score is weighted toward maximum possible image quality (i.e. at low ISO)balanced between color depth, High ISO and dynamic range. This is only useful if you realize how it is calculated.
You are more than welcome to view graphs separately, in print or per pixel - however you choose. Just don't let your own inability to comprehend get in the way.
An all-round performing sensor, such as the D5100 (very good high ISO, very good dynamic range, good color depth) of course will 'beat' a medium format sensor 'overall' which will have very poor high ISO noise in comparison.
Just learn to interpret what you are looking at and glean what you need to.
It isn't that hard, but of course if it tells you your precious camera isn't as good, gnash away. It won't change anything.
--
http://www.samwaldron.co.nz
Honestly, read, listen, learn - all the info is there.
DXO overall score is weighted toward maximum possible image quality (i.e. at low ISO)balanced between color depth, High ISO and dynamic range. This is only useful if you realize how it is calculated.
You are more than welcome to view graphs separately, in print or per pixel - however you choose. Just don't let your own inability to comprehend get in the way.
An all-round performing sensor, such as the D5100 (very good high ISO, very good dynamic range, good color depth) of course will 'beat' a medium format sensor 'overall' which will have very poor high ISO noise in comparison.
Just learn to interpret what you are looking at and glean what you need to.
It isn't that hard, but of course if it tells you your precious camera isn't as good, gnash away. It won't change anything.
--
http://www.samwaldron.co.nz
rhlpetrus
Forum Pro
If they took th etrouble of looking at the various graphs, read the posted reviews and comments at DxO mark, learn what each indicator means, then this debate would be empty. It is obvious that there are some problems, Iliah has called some, others as well, but one can have a pretty good picture from the results there in terms of sensor performance.
Now, they take the final composite score (maybe DxO Mark should leave that out) and use that to disavow all the data collected. The D5100's sensor (Sony's 16MP) is well-known for its excellent base ISO DR and high ISO performance. So, in such terms, it is actually better in sensor tech terms than the Kodak sensors used by many of the MF sensors. What's the problem with that. The MF manufacturers are not competing for high ISO performance like the dslr makers are, so they don't design their sensors, ADC, etc, for such.
Now, since final IQ depends on other issues, and size of sensor plays a big role, a Hassy at base ISO will produce better IQ, if both are printed at the same size, not doubt, and the Hassy's file can still be printed much larger since it has much more pixels. And don't forget lenses, etc. So, final IQ is not related to RAW specs only.
But one could do the following test: crop the Hassy 51MP file to APS-C's size, that is take the 48x36mm sensor file and crop a 24x16mm part of it. That's 22% of the total area, thus would produce a 0.22*51= 11MP file. Print that and the D5110's 16MP file at same final size.
That's how one should compare those two sensors, in terms of technology, removing the size from the equation. My bet is the D5100's will be better, at every ISO. Now, granted, the Hassy 50's sensor is 5-yo tech. The newer model should show better results, in particular for DR.
Those disavowing DxO Mark frequently are likely just unhappy with the results for their brand of preference or are not investing any effort to understand what's at stake. Some that have more serious doubts, like Iliah above, surely are the ones that are able to do their own evaluation. Only problem is that they usually don't publish that for all cameras, systematically, so it remains for us end users, not tech able, to use what is available in the net. DxO Mark has just filled that void. One may choose to ignore them, but most won't, in the end.
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/
Good shooting and good luck
(after Ed Murrow)
Now, they take the final composite score (maybe DxO Mark should leave that out) and use that to disavow all the data collected. The D5100's sensor (Sony's 16MP) is well-known for its excellent base ISO DR and high ISO performance. So, in such terms, it is actually better in sensor tech terms than the Kodak sensors used by many of the MF sensors. What's the problem with that. The MF manufacturers are not competing for high ISO performance like the dslr makers are, so they don't design their sensors, ADC, etc, for such.
Now, since final IQ depends on other issues, and size of sensor plays a big role, a Hassy at base ISO will produce better IQ, if both are printed at the same size, not doubt, and the Hassy's file can still be printed much larger since it has much more pixels. And don't forget lenses, etc. So, final IQ is not related to RAW specs only.
But one could do the following test: crop the Hassy 51MP file to APS-C's size, that is take the 48x36mm sensor file and crop a 24x16mm part of it. That's 22% of the total area, thus would produce a 0.22*51= 11MP file. Print that and the D5110's 16MP file at same final size.
That's how one should compare those two sensors, in terms of technology, removing the size from the equation. My bet is the D5100's will be better, at every ISO. Now, granted, the Hassy 50's sensor is 5-yo tech. The newer model should show better results, in particular for DR.
Those disavowing DxO Mark frequently are likely just unhappy with the results for their brand of preference or are not investing any effort to understand what's at stake. Some that have more serious doubts, like Iliah above, surely are the ones that are able to do their own evaluation. Only problem is that they usually don't publish that for all cameras, systematically, so it remains for us end users, not tech able, to use what is available in the net. DxO Mark has just filled that void. One may choose to ignore them, but most won't, in the end.
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/
Good shooting and good luck
(after Ed Murrow)
rhlpetrus
Forum Pro
Area-wise difference is close to 80%, so there's a loss of performance once you use same print size. For Screen mode, results are almost identical, except for base ISO DR. But that could just be a poorer ADC/electronics grade or QC, not unexpected given it's a lower level in the camera chain.
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/
Good shooting and good luck
(after Ed Murrow)
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/
Good shooting and good luck
(after Ed Murrow)
Faintandfuzzy
Veteran Member
- Messages
- 3,199
- Reaction score
- 6
It is easy. The issue though is that if you're ranking image quality parameters, I'd recommend a new method if you have MFDBs being beaten by consumer grade cameras that actually have poor image quality in comparison.
That is the problem with DxO. They provide a ranking which, let's be honest now, is completely meaningless in the real world.
That is why many, myself included, consider them a joke.
Keyboard shortcuts
- f
- Forum