5DS scores at DXO **now posted**

[ahsanford]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

Lens Sharpness scores is what I'm most interested in. I like to have the ability to crop or reframe photos. Many time I shoot landscape and crop and frame the shot in portrait to give it a different look.

DXO may be the name for sensor scores, but they are outright batsh-- crazy with lens testing. Their lens findings are inane, contradictory, and eye-rolling. Plastic fantastics outperforming big whites, the same exact lens being stellar on a Nikon rig and disappointing on a Canon one, etc.

Surely, LensRentals, LensTip and Photozone will upgrade their reference rigs to 5DS and report proper resolution testing. I trust those folks far more than DXO.

Here's Roger's basic take on the resolution bumps you might see with a 5DS and 5DSr over a 5D3:
http://www.lensrentals.com/blog/2015/06/canon-5ds-and-5ds-r-initial-resolution-tests

But, if your mind is made up: you may not have your DXO data on lenses tomorrow. As I'm sure you know, DXO tests individual combos of lenses and bodies and may not report that right away. That kind of dole that out over time.

- A

 

[benperrin]

Re: 5DS scores at DXO still not posted

The sensor scores for the EOS 5DS and EOS 5DS R have been completed. They'll be posted tomorrow.

Teaser: They're lower than the D810 and the 5DS ranks slightly better than the 5DS R, mainly because of ISO performance.

It's amazing how bad the dxomark logic is. Most people would take better defined details even if that means the noise is also better defined. What that doesn't take into consideration is that noise reduction software has more room to play with on an image that has more definition. And of course that resolution advantage gets completely lost at 8mp (or whatever they downsample to) which is something you'd probably never do if you were interested in a high mp camera. The logic of trying to make all sensors confine to a magic set of rules is insane at best. We all have different needs for our cameras, some only want to prints 4x5 inch prints, others 3 meter prints, others need to crop heavily and most probably just want to share images on the web. Some only shoot at iso100 others are consistently at iso3200.

Lesson: forget dxomark and find the camera that suits your needs. These magic numbers that dxo pull out of their butts are meaningless in the real world.

 

[PhotoCat]

Re: 5DS scores at DXO still not posted

The sensor scores for the EOS 5DS and EOS 5DS R have been completed. They'll be posted tomorrow.

Teaser: They're lower than the D810

This is to be expected since Canon hasn't made much progress in the dynamic range department for 5DS & 5DS R.

 

[traveller]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

DXO mark's credibility has been damaged even amongst Sony and Nikon users by the way they manipulated their own testing system in order to give their new "ONE" a score of 85.

 

[3kramd5]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

DXO mark's credibility has been damaged even amongst Sony and Nikon users by the way they manipulated their own testing system in order to give their new "ONE" a score of 85.

Did they manipulate their testing system, or did they just decide to score a post-composited image as a "sensor"?

 

[Orangutan]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

...
DXO may be the name for sensor scores, but they are outright batsh-- crazy with lens testing. Their lens findings are inane, contradictory, and eye-rolling. Plastic fantastics outperforming big whites, ...
...

Again, there is no reason why a plastic fantastic cannot outperform a big white.

If I make the perfect 50mm/1.4 lens and put it in a plastic case, why shouldn't it outscore a big white lens that is very good but not perfect?

Aren't the lens elements themselves plastic? Cheap lenses also have cheap coatings. It's not impossible, but if it happened someone would figure out how to use cheap materials to make big expensive telephotos. I.e., it's not physically impossible, but it's nearly impossible economically.

 

[Aglet]

Re: 5DS scores at DXO still not posted

My guess is they'll rate it between 85 and 87. I'll be surprised if it scores higher than 89. Regardless I'll get one later this year, either by mid-September or if that's not possbile then November-December time.

HEHE! Did you have a big bowl of Optimist-Os for breakfast?
That'd be quite a score jump for Canon, I don't think they're quite there yet but, eventually, we'll see.

 

[Aglet]

Re: 5DS scores at DXO still not posted

The sensor scores for the EOS 5DS and EOS 5DS R have been completed. They'll be posted tomorrow.

Teaser: They're lower than the D810 and the 5DS ranks slightly better than the 5DS R, mainly because of ISO performance.

It's amazing how bad the dxomark logic is. Most people would take better defined details even if that means the noise is also better defined. What that doesn't take into consideration is that noise reduction software has more room to play with on an image that has more definition. And of course that resolution advantage gets completely lost at 8mp (or whatever they downsample to) which is something you'd probably never do if you were interested in a high mp camera. The logic of trying to make all sensors confine to a magic set of rules is insane at best. We all have different needs for our cameras, some only want to prints 4x5 inch prints, others 3 meter prints, others need to crop heavily and most probably just want to share images on the web. Some only shoot at iso100 others are consistently at iso3200.

Lesson: forget dxomark and find the camera that suits your needs. These magic numbers that dxo pull out of their butts are meaningless in the real world.

The general advice is good; you're butt-wrong about real-world-meaningless.

 

[benperrin]

Re: 5DS scores at DXO still not posted

The general advice is good; you're butt-wrong about real-world-meaningless.

I'll stand by my statement I think dxomark scores are meaningless. Selection of glass and technique will have more impact on images than picking a camera based on made up scores. Try a camera and if it works for you then great, if not get another. There is no magical number that can tell you what you should buy or use.

 

[Loswr]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

...
DXO may be the name for sensor scores, but they are outright batsh-- crazy with lens testing. Their lens findings are inane, contradictory, and eye-rolling. Plastic fantastics outperforming big whites, ...
...

Again, there is no reason why a plastic fantastic cannot outperform a big white.

If I make the perfect 50mm/1.4 lens and put it in a plastic case, why shouldn't it outscore a big white lens that is very good but not perfect?

Price and size has nothing to do with how good a lens is at drawing an image.

Yes, dilbert...obviously DxO's lens scores make perfect sense to you given what you have in common. DxO gives higher scores to lenses because of more dynamic range...so apparently both you and DxO believe that lenses are cameras.

Does it matter what it is made from?

Shouldn't all that matters when scoring a lens's performance be based on its optical properties?

If a plastic fantastic lens can render an image without CA, no vignetting and be sharp from center to edge, why shouldn't it score really really well?

Oh, so you believe that DxO's Lens Scores are based primarily on the lens' optical performance in metrics like sharpness and CA? Why am I not surprised? : :

 

[bdunbar79]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

The problem dilbert, is that they are NOT following that logic in lens scores. How many times do I need to cite the 500mm lenses comparisons/scores of the Canon vs. Nikon and they specifically said they scored the Nikon lens the same as the Canon, despite it not performing as well as a lens, simply because of the superior DR of the Nikon camera used for the Nikon lens (I think it was the D800 and 5D3).

Obviously that is complete garbage to a scientist and I do contend it has absolutely no real-world usefulness. Or, better, the SCORES don't reflect reality.

 

[3kramd5]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

DXO mark's credibility has been damaged even amongst Sony and Nikon users by the way they manipulated their own testing system in order to give their new "ONE" a score of 85.

Did they manipulate their testing system, or did they just decide to score a post-composited image as a "sensor"?

The latter. The score of "85" for their "ONE" should be excluded from sensor ratings.

What will be interesting to see if if they allow the rumored in-camera HDR from the A7000 to be scored as a "sensor," or whether that's a benefit they'll hold for themselves with some flimsy requirements (like it be stored in a .DNG container).

 

[ahsanford]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

No scores yet. Bummin.

- A

 

[ahsanford]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

DXO mark's credibility has been damaged even amongst Sony and Nikon users by the way they manipulated their own testing system in order to give their new "ONE" a score of 85.

Did they manipulate their testing system, or did they just decide to score a post-composited image as a "sensor"?

The latter. The score of "85" for their "ONE" should be excluded from sensor ratings.

What will be interesting to see if if they allow the rumored in-camera HDR from the A7000 to be scored as a "sensor," or whether that's a benefit they'll hold for themselves with some flimsy requirements (like it be stored in a .DNG container).

Right! Why hasn't Olympus' multi-shot feature been given the same "we'll allow it" consideration? Canon's in-camera HDR? Magic Lantern dual-ISO?

None of it should be reported, including the DXO One's composite score. Tell me how the sensor takes individual shots.

- A

 

[windycitywilly]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

here are the scores i think

 

[ahsanford]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

here are the scores i think

Someone is yanking down the scores as fast as you guys can link them.

These are NOT up on DXO's site right now.

- A

 

[3kramd5]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

DXO mark's credibility has been damaged even amongst Sony and Nikon users by the way they manipulated their own testing system in order to give their new "ONE" a score of 85.

Did they manipulate their testing system, or did they just decide to score a post-composited image as a "sensor"?

The latter. The score of "85" for their "ONE" should be excluded from sensor ratings.

What will be interesting to see if if they allow the rumored in-camera HDR from the A7000 to be scored as a "sensor," or whether that's a benefit they'll hold for themselves with some flimsy requirements (like it be stored in a .DNG container).

Right! Why hasn't Olympus' multi-shot feature been given the same "we'll allow it" consideration? Canon's in-camera HDR? Magic Lantern dual-ISO?

None of it should be reported, including the DXO One's composite score. Tell me how the sensor takes individual shots.

- A

They're hand waving and calling their files "raw" (.dxo super raw plus). Since they clearly are not raw, I presume they carved out a specific file type for which they'll allow composites, a file type only they themselves produce. That being said, even if they start allowing HDR images from other makes, I agree: it's silly if it's not done with a single exposure.

 

[Loswr]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

They're hand waving and calling their files "raw" (.dxo super raw plus). Since they clearly are not raw, I presume they carved out a specific file type for which they'll allow composites, a file type only they themselves produce. That being said, even if they start allowing HDR images from other makes, I agree: it's silly if it's not done with a single exposure.

Well, they are 'RAW' even if only DxO's own software can open them. But the 'regular' RAW files from the DxO One are saved in DNG containers.

 

[ahsanford]

Re: 5DS scores at DXO drop **tomorrow** (i.e. 7/8/15)

They're hand waving and calling their files "raw" (.dxo super raw plus). Since they clearly are not raw, I presume they carved out a specific file type for which they'll allow composites, a file type only they themselves produce. That being said, even if they start allowing HDR images from other makes, I agree: it's silly if it's not done with a single exposure.

Well, they are 'RAW' even if only DxO's own software can open them. But the 'regular' RAW files from the DxO One are saved in DNG containers.

I'd love to see the DXO disclaimers on that score:

1) "If it is windy day, the sensor score is still valid but you will have a blurry shot. That's not our fault."

2) "If your subject is moving -- tell them to stand still. (Our blurry shots will still be gorgeous, though.)"

3) "Using photoshop to composite multiple shots for more dynamic range is cheating. Your camera's sensor score was just docked 5 points for even considering that idea."

4) "I carry an epi-pen because my doctor says I am allergic to impartiality."

- A

 

[ahsanford]

Re: 5DS scores at DXO drop **today** (i.e. 7/8/15)


And it is now official on their site:

http://www.dxomark.com/Cameras/Compare/Side-by-side/Canon-EOS-5DS-versus-Canon-EOS-5DS-R-versus-Canon-EOS-5D-Mark-III___1008_1009_795

- A