*** 5Dc vs 60D resolution test ***

Started Mar 17, 2013 | Discussions
Great Bustard
Forum ProPosts: 23,258
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to schmegg, Mar 18, 2013

schmegg wrote:

MAC wrote:

Photozone correctly says -- 11 times they say it on one page -- you can't compare their numbers across different formats

Correct - partially because the detail present at the sensor plane is different due to the tests begin conducted at different distances. The very thing the macro test eliminates!

When I compare with those sites -- I stay in the same format

Me too. Anyone who doesn't is making a mistake - though telling them this often proves incredibly difficult! Hehe!

PZ makes that claim, but, in my opinion, it's a CYA move. Basically, Klaus is saying that different RAW conversions were used, so cross-system comparisons are not valid. However, in my opinion, it's pretty easy to normalize for this by sharpening the output consistently across systems, as opposed to using default settings or "sharpening off" (when we all know that "sharpening off" doesn't always mean "no sharpening").

In other words, while the numbers may not be exactly comparable, they will, in my opinion, likely be quite close (say within 10%, and I would be suprised if someone could see the difference, for example, between a photo that resolved 2700 lw/ph and a photo that resolved 3000 lw/ph, even at the pixel level, much less care about the differences even if they could).

Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,810Gear list
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to schmegg, Mar 19, 2013

schmegg wrote:

In the previous test, the OP very wisely provided the raw files with full EXIF attached. This makes his results verifiable by others.

So, where are the raw files qianp2k?

I said I don't have a site that can hold RAW files.  But willing to upload my files to a site anyone can provide.

Without them I afraid that your copious outspoken posts leaning in only one direction on this topic pretty much negates anything the images may or may not show.

You can do similar tests provided you don't have agenda.

For a start - those images are too small. I do believe it was you that kept calling for 5000 pixel wide images for a decent comparison.

They are 100% cropped and I even upsampling 5D files to 60D size.

We need to see the raw files for ourselves (with full EXIF) if you are going to convince anyone of anything here. Otherwise, I'm afraid that it's just more of the same from you - despite what these images seem to imply (which is totally opposite to what you have been claiming all this time).

So give a link.  I just don't have.

BTW - where is the macro test? I still await a test from you that delivers exactly the same image detail to each sensor - the only sure way to tell which sensor can resolve more detail.

Might do later.  however since my Sigma 150/2.8 OS is a prime so I have to adjust distance between 5D3/5D to 60D to frame into the same AOV as in real world I'd do on a flower for example.

Old Canon 100/2.8 macro (just sold more than I bought, same for 50/1.4 and 17-40L as all those lenses I bought in Dec 2007 - 1st half year of 2008) test shots on 10mp 1D3.  I remember you used to try to compare from your old 50D.  My ones are obviously sharper   I believe you cannot separate sharpness from resolution as you cannot separate details from noise.  They are related.



-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,810Gear list
Like?
Re: Well done
In reply to MAC, Mar 19, 2013

MAC wrote:

Sovern wrote:

Dave Luttmann wrote:

I see the 5D has very slightly better corners, and the 60D is slightly better in the center and middle areas. This definitely shows that you never saw an advantage in resolution or sharpness after the files were reduced to 2000 pixels wide.

That said, the test seems well executed...at least at my cursory glance. Not sure why you chose DPP as it is probably the worst raw converter for detail and would harm the results from both the 5D and 60D.

In all fairness, I think you did a good, and fair job. That said, it shows the 5D resolves less....so please, no more posts about the rez.

Well done.

I disagree. The 5Dc very obviously resolves more detail than the 60D.

Remember, I use a crop factor camera...I am not biased.

The differences are quite large on my large monitor.

no, just like the prior thread in the lens forum

The crop sensor resolves more in the center

and the 5dc at the edges

T4i is amazing camera btw -- Bryan at The Digital Picture Site talks about how Canon tuned the Raw files at 1dx sharpness levels.

These are processed -- zero on the LR sharpness slider

After I returned from Cancun trip, and temperature warming up. I will do again in closer distance in the school park by moving just cross the school building. It will be a perfect test site as I will face entire walls of bricks. Will test 24-70L II from about only 30 meters away. Will bring 70-200L II to test.

My dollar bill was taken in macro mode just outside the min focus distance. What you see is a tiny center area that only frames the bill (I cropped out 2/3 outside frame to save space). I will test again by shooting from a bit far distance, and also with 70-200L II.

All my tests in OP on tripod. 60D photos were taken by 10x LV precise MF focus, and I also tried AF, and fired 3 shots on each lens-camera combo. So I had 6 60D with 70-200L/4.0 IS for example, and they are all look identical. Absolutely no miss-focus issue and actually 5D has disadvantage because it doesn't have LV.

In your test, 5D center is about the same as T4i at this size. I don't see much difference.

5D side is noticeable sharper and at least the same detailed. I can see purple fringe (CA) in T4i easily.

-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
MAC
MAC
Forum ProPosts: 12,268Gear list
Like?
Re: Well done
In reply to qianp2k, Mar 19, 2013

qianp2k wrote:

MAC wrote:

Sovern wrote:

Dave Luttmann wrote:

I see the 5D has very slightly better corners, and the 60D is slightly better in the center and middle areas. This definitely shows that you never saw an advantage in resolution or sharpness after the files were reduced to 2000 pixels wide.

That said, the test seems well executed...at least at my cursory glance. Not sure why you chose DPP as it is probably the worst raw converter for detail and would harm the results from both the 5D and 60D.

In all fairness, I think you did a good, and fair job. That said, it shows the 5D resolves less....so please, no more posts about the rez.

Well done.

I disagree. The 5Dc very obviously resolves more detail than the 60D.

Remember, I use a crop factor camera...I am not biased.

The differences are quite large on my large monitor.

no, just like the prior thread in the lens forum

The crop sensor resolves more in the center

and the 5dc at the edges

T4i is amazing camera btw -- Bryan at The Digital Picture Site talks about how Canon tuned the Raw files at 1dx sharpness levels.

These are processed -- zero on the LR sharpness slider

After I returned from Cancun trip, and temperature warming up. I will do again in closer distance in the school park by moving just cross the school building. It will be a perfect test site as I will face entire walls of bricks. Will test 24-70L II from about only 30 meters away. Will bring 70-200L II to test.

My dollar bill was taken in macro mode just outside the min focus distance. What you see is a tiny center area that only frames the bill (I cropped out 2/3 outside frame to save space). I will test again by shooting from a bit far distance, and also with 70-200L II.

All my tests in OP on tripod. 60D photos were taken by 10x LV precise MF focus, and I also tried AF, and fired 3 shots on each lens-camera combo. So I had 6 60D with 70-200L/4.0 IS for example, and they are all look identical. Absolutely no miss-focus issue and actually 5D has disadvantage because it doesn't have LV.

In your test, 5D center is about the same as T4i at this size. I don't see much difference.

the whole card is in the center of the frame.  you see what you see because you are looking for sharpness versu resolution, which are two different parameters.  Look at the fine print which is clearly more easily resolved in the center of the overall shot on the T4i.

So much for dxo!

dxo rates 7d + 70-200 F2.8 L = MPix 8

dxo rates 5d + 70-200 F2.8 L = MPix 11

How can this be?  The T4i like your 60d out resolves the 5dc in the center!

3 MPix difference?

dxo is a joke if you try to line up values across platforms

5D side is noticeable sharper and at least the same detailed. I can see purple fringe (CA) in T4i easily.

I have no problem removing

look at the fine print on the 5dc in te center of the fram -- it turned more colors than the fine print for the T4i.

For a sensor 2.56 smaller, the T4i does very well against the 5dc.

You were attributing a crop penalty when one doesn't exist -- because you were relying on numbers from sites that are nonsense when you use those sites to compare resolution across different formats.  Even TDP site's visuals across formats should not be used -- they were taken at different distances across formats.

your process of looking at these sites have confused you.

Now you and I and the fella from Bulgaria all get the same results -- and this trumps MPix - which should not be used across formats

-- hide signature --
 MAC's gear list:MAC's gear list
Canon EOS 5D Canon EOS 650D Canon EOS 100D Canon EF 70-200mm f/2.8L USM Canon EF 100mm f/2.8L Macro IS USM +5 more
Reply   Reply with quote   Complain
Keith Z Leonard
Senior MemberPosts: 4,151Gear list
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to qianp2k, Mar 19, 2013

Google drive or sky drive are free & could hold your raw files.

 Keith Z Leonard's gear list:Keith Z Leonard's gear list
Canon EOS 5D Mark III Canon EF 70-200mm f/4.0L USM Canon EF 100-400mm f/4.5-5.6L IS USM Sigma 50mm F1.4 EX DG HSM Canon EF 24-105mm f/4L IS USM +13 more
Reply   Reply with quote   Complain
schmegg
schmegg MOD
Senior MemberPosts: 5,002
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to qianp2k, Mar 19, 2013

qianp2k wrote:

Might do later. however since my Sigma 150/2.8 OS is a prime so I have to adjust distance between 5D3/5D to 60D to frame into the same AOV as in real world I'd do on a flower for example.

The test involves rendering a very finely detailed subject at 1:1 magnification - that is what macro is.

To do this, you will need to set the lens at the minimum focus distance and then photograph the subject with both cameras to record it at 1:1 magnification.

This way the detail you present for each sensor to record will be identical - and from this it should be easy to see which sensor is able to resolve more of the same available detail, and hence, which camera resolves more detail.

Old Canon 100/2.8 macro (just sold more than I bought, same for 50/1.4 and 17-40L as all those lenses I bought in Dec 2007 - 1st half year of 2008) test shots on 10mp 1D3. I remember you used to try to compare from your old 50D.

Someone else? I've never owned a 50D.

My ones are obviously sharper I believe you cannot separate sharpness from resolution as you cannot separate details from noise. They are related.

I've tried to explain this to you many times now, but, as I've failed on every occasion, I'll have another go ...

Yes - they are related. Sharpness depends on resolution and acutance.

It's possible to have low resolution and high acutance and end up with an image that is perceived to be sharp even though it actually resolves less detail. That's what you have with a 5D compared to a 60D, where the resolution is higher but the acutance is lower.

The big difference between resolution and acutance is that resolution is set at capture time and can only ever be decreased. Increasing resolution after capture involves "inventing" detail ... that is, having a guess at what might have been there.

Acutance, on the other hand, can be altered after capture time by re-sampling to a lower resolution (for instance, 18MP to 12MP) and by post processing sharpening algorithms.

Many feel that it's better to capture with more resolution and lower acutance as the lower acutance can be recovered effectively to a great degree during post processing.

Of course, even better is to capture with high resolution and high acutance. And that's exactly what Canon's 20-odd MP full frame sensors are very good at.

I hope that's the last time I have to type that for you!



Nice, but they aren't really macro photos, just close-ups.

Here's another not-macro, this one from a crop ...

Reply   Reply with quote   Complain
Steen Bay
Veteran MemberPosts: 6,188
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to schmegg, Mar 19, 2013

schmegg wrote:

qianp2k wrote:

Might do later. however since my Sigma 150/2.8 OS is a prime so I have to adjust distance between 5D3/5D to 60D to frame into the same AOV as in real world I'd do on a flower for example.

The test involves rendering a very finely detailed subject at 1:1 magnification - that is what macro is.

To do this, you will need to set the lens at the minimum focus distance and then photograph the subject with both cameras to record it at 1:1 magnification.

This way the detail you present for each sensor to record will be identical - and from this it should be easy to see which sensor is able to resolve more of the same available detail, and hence, which camera resolves more detail.

The framing of the two images will be different, so the detail presented for each sensor won't be identical if using the same lens (but it would be, pretty much, if using the 60/2.8 on 60D and the 100/2.8L on FF). Not really needed to make such a comparison. Nobody disputes that a crop camera has more 'reach' (or higher 'magnification') if shooting with the same lens from the same distance (if the crop camera has a higher pixel density).

Old Canon 100/2.8 macro (just sold more than I bought, same for 50/1.4 and 17-40L as all those lenses I bought in Dec 2007 - 1st half year of 2008) test shots on 10mp 1D3. I remember you used to try to compare from your old 50D.

Someone else? I've never owned a 50D.

My ones are obviously sharper I believe you cannot separate sharpness from resolution as you cannot separate details from noise. They are related.

I've tried to explain this to you many times now, but, as I've failed on every occasion, I'll have another go ...

Yes - they are related. Sharpness depends on resolution and acutance.

It's possible to have low resolution and high acutance and end up with an image that is perceived to be sharp even though it actually resolves less detail. That's what you have with a 5D compared to a 60D, where the resolution is higher but the acutance is lower.

The big difference between resolution and acutance is that resolution is set at capture time and can only ever be decreased. Increasing resolution after capture involves "inventing" detail ... that is, having a guess at what might have been there.

Acutance, on the other hand, can be altered after capture time by re-sampling to a lower resolution (for instance, 18MP to 12MP) and by post processing sharpening algorithms.

Many feel that it's better to capture with more resolution and lower acutance as the lower acutance can be recovered effectively to a great degree during post processing.

Of course, even better is to capture with high resolution and high acutance. And that's exactly what Canon's 20-odd MP full frame sensors are very good at.

That makes good sense. Worth to remember though that sharpening also increases the noise.

Reply   Reply with quote   Complain
schmegg
schmegg MOD
Senior MemberPosts: 5,002
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to Steen Bay, Mar 19, 2013

Steen Bay wrote:

schmegg wrote:

qianp2k wrote:

Might do later. however since my Sigma 150/2.8 OS is a prime so I have to adjust distance between 5D3/5D to 60D to frame into the same AOV as in real world I'd do on a flower for example.

The test involves rendering a very finely detailed subject at 1:1 magnification - that is what macro is.

To do this, you will need to set the lens at the minimum focus distance and then photograph the subject with both cameras to record it at 1:1 magnification.

This way the detail you present for each sensor to record will be identical - and from this it should be easy to see which sensor is able to resolve more of the same available detail, and hence, which camera resolves more detail.

The framing of the two images will be different, so the detail presented for each sensor won't be identical if using the same lens (but it would be, pretty much, if using the 60/2.8 on 60D and the 100/2.8L on FF).

Yeah - perhaps it's my way of trying to describe things that is confusing things here.

Imagine you have a very small creature (not a flower or something large) and you wish to get the most detailed image you can of it.

The way to do this is to select your macro lens (lets assume you only have one - I think that's probably realistic enough for many).

You mount the lens, set it to it's minimum focus distance and then move to the subject until it is in focus. Then take the shot.

Now, if you do this with, say a 5D, and then again with a 60D, you will effectively be recording exactly the same amount of subject detail with each camera (providing you got the subject in the centre and at MFD).

In this case, each camera has recorded the exactly same projected image from the same lens and at the same exposure (if you are doing this as a test). And it should be relative straight forward to compare the images and see which camera sensor resolved more of the detail projected onto it.

Not really needed to make such a comparison. Nobody disputes that a crop camera has more 'reach' (or higher 'magnification') if shooting with the same lens from the same distance (if the crop camera has a higher pixel density).

Peter does - he continues to claim that a 5D out-resolves an 18MP crop.

When he is confronted with the test above - which actually properly compares the resolving capabilities of the two, he avoids it.

He has now done other, and as you hint, possibly more practical (for certain shooting situations - though not all) tests. These are closer but still don't show any advantage in resolving capabilities for the 5D - but the amount of detail the cameras have been provided to record is different, so care should be taken in drawing conclusions regarding his 'out-resolving' claim.

In any case - it would seem to me that the above macro test (or, in fact, any fixed lens/exposure/subject distance test) is the best to settle the truth of the claim that a 5D out-resolves an 18MP crop.

Old Canon 100/2.8 macro (just sold more than I bought, same for 50/1.4 and 17-40L as all those lenses I bought in Dec 2007 - 1st half year of 2008) test shots on 10mp 1D3. I remember you used to try to compare from your old 50D.

Someone else? I've never owned a 50D.

My ones are obviously sharper I believe you cannot separate sharpness from resolution as you cannot separate details from noise. They are related.

I've tried to explain this to you many times now, but, as I've failed on every occasion, I'll have another go ...

Yes - they are related. Sharpness depends on resolution and acutance.

It's possible to have low resolution and high acutance and end up with an image that is perceived to be sharp even though it actually resolves less detail. That's what you have with a 5D compared to a 60D, where the resolution is higher but the acutance is lower.

The big difference between resolution and acutance is that resolution is set at capture time and can only ever be decreased. Increasing resolution after capture involves "inventing" detail ... that is, having a guess at what might have been there.

Acutance, on the other hand, can be altered after capture time by re-sampling to a lower resolution (for instance, 18MP to 12MP) and by post processing sharpening algorithms.

Many feel that it's better to capture with more resolution and lower acutance as the lower acutance can be recovered effectively to a great degree during post processing.

Of course, even better is to capture with high resolution and high acutance. And that's exactly what Canon's 20-odd MP full frame sensors are very good at.

That makes good sense. Worth to remember though that sharpening also increases the noise.

And also worth remembering that noise often increases perceived sharpness.

Reply   Reply with quote   Complain
MAC
MAC
Forum ProPosts: 12,268Gear list
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to schmegg, Mar 19, 2013

schmegg wrote:

Here's another not-macro, this one from a crop ...

you out -did gianp2k's closeups imo - and you did it with a crop!  Nice! Folks should spend more time making these type of images instead of investing $15K in heavy paper weights that will eventually be stolen while they are on vacation or eventually make their wife fed up as she carries the heavy gear!   Sorry, couldn't resist! 

btw - It is my opinion that Canon gave T4i owners the best RAW output accutance of all their 18 mpxl cameras -- as Bryan describes in The Digital Picture review of the T4i.

 MAC's gear list:MAC's gear list
Canon EOS 5D Canon EOS 650D Canon EOS 100D Canon EF 70-200mm f/2.8L USM Canon EF 100mm f/2.8L Macro IS USM +5 more
Reply   Reply with quote   Complain
MAC
MAC
Forum ProPosts: 12,268Gear list
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to schmegg, Mar 19, 2013

schmegg wrote:

And also worth remembering that noise often increases perceived sharpness.

and noise redux often makes things mush -- not usable, no matter what the resolution advantage.   note, I never went above iso 1600 on my 40d -- see below.  Whereas my T4i and 60d seem to handle iso3200 well and even iso 6400 when I have the light or am zoomed out

Bob adkins describes below for iso3200:  "With some noise reduction applied, the 50D can show less noise then the 40D, but still maintain an edge in resolution as shown below:"

yeah right -- look at the mush

 MAC's gear list:MAC's gear list
Canon EOS 5D Canon EOS 650D Canon EOS 100D Canon EF 70-200mm f/2.8L USM Canon EF 100mm f/2.8L Macro IS USM +5 more
Reply   Reply with quote   Complain
schmegg
schmegg MOD
Senior MemberPosts: 5,002
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to MAC, Mar 19, 2013

MAC wrote:

schmegg wrote:


And also worth remembering that noise often increases perceived sharpness.

and noise redux often makes things mush -- not usable, no matter what the resolution advantage. note, I never went above iso 1600 on my 40d -- see below. Whereas my T4i and 60d seem to handle iso3200 well and even iso 6400 when I have the light or am zoomed out

Bob adkins describes below for iso3200: "With some noise reduction applied, the 50D can show less noise then the 40D, but still maintain an edge in resolution as shown below:"

yeah right -- look at the mush

Just a couple of observations.

1. Try resizing the non-NR'ed 50D to the same size as the 40D - then see how it compares.

2. Please note I said "perceived sharpness", not "resolution". They are very different things!

Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,810Gear list
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to schmegg, Mar 19, 2013

schmegg wrote:

Peter does - he continues to claim that a 5D out-resolves an 18MP crop.

I'm busy at the moment so unable to reply in details. Nevertheless please don't twist what I said. I said clearly throughout posts in many threads that depend on lenses. With all lenses I have before such as 24-105L and 70-200L/4.0 IS...yes 5Dc does outresolve 60D. I have some new lenses now such as 24-70L II and 70-200L II and I will do more tests later. With 24-70L II we can see they are very close and I also want to see with 70-200L II. I believe it's more meaningful to test resolution from a distance at least 50X focus length (outdoor tests are more meaningful in real world photos) not in very close distance (24-70L II is not a macro lens anyway for example). As I said DXOMark might use AVERAGE MTF resolution to calculate its P-MPix number. From so far what I have seen, I don't have indications that DXOMark tests are wrong and actually I think pretty reflect what I have found.

-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,810Gear list
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to Keith Z Leonard, Mar 19, 2013

Keith Z Leonard wrote:

Google drive or sky drive are free & could hold your raw files.

Thx info.  I will create an acct and post RAW there.

-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
schmegg
schmegg MOD
Senior MemberPosts: 5,002
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to qianp2k, Mar 19, 2013

qianp2k wrote:

schmegg wrote:

Peter does - he continues to claim that a 5D out-resolves an 18MP crop.

I'm busy at the moment so unable to reply in details. Nevertheless please don't twist what I said.

Would you agree that you said that in almost all cases the 5D out-resolves an 18MP crop. I can only recall you mentioning the 300/2.8 IS L II as an exception.

I said clearly throughout posts in many threads that depend on lenses. With all lenses I have before such as 24-105L and 70-200L/4.0 IS...yes 5Dc does outresolve 60D.

No - it doesn't. And so far, from what we can tell from the comparison you posted, your own testing confirms this.

I guess you are still in that river in Egypt then.

I have some new lenses now such as 24-70L II and 70-200L II and I will do more tests later. With 24-70L II we can see they are very close and I also want to see with 70-200L II. I believe it's more meaningful to test resolution from a distance at least 50X focus length (outdoor tests are more meaningful in real world photos) not in very close distance (24-70L II is not a macro lens anyway for example). As I said DXOMark might use AVERAGE MTF resolution to calculate its P-MPix number. From so far what I have seen, I don't have indications that DXOMark tests are wrong and actually I think pretty reflect what I have found.

So you still refuse to provide the same detail to each sensor to see which one resolves more?

As the old proverb says .. "The truth is like a door, and you yourself must choose to enter"

Once again - DxO are conducting lens sharpness tests, not comparing the resolving abilities of different sized sensors.

Here  is their testing protocol ... please take note of the target and then explain how this enables them to determine the resolving abilities of different sized sensors.

I await your enlightenment!

Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,810Gear list
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to schmegg, Mar 19, 2013

schmegg wrote:

qianp2k wrote:

schmegg wrote:

Peter does - he continues to claim that a 5D out-resolves an 18MP crop.

I'm busy at the moment so unable to reply in details. Nevertheless please don't twist what I said.

Would you agree that you said that in almost all cases the 5D out-resolves an 18MP crop. I can only recall you mentioning the 300/2.8 IS L II as an exception.

Yes but based on what lenses I used to have and according to DXOMark.

I said clearly throughout posts in many threads that depend on lenses. With all lenses I have before such as 24-105L and 70-200L/4.0 IS...yes 5Dc does outresolve 60D.

No - it doesn't. And so far, from what we can tell from the comparison you posted, your own testing confirms this.

Yes from both DXOMark and from my own tests. My own tests actually confirm what DXOMark tests.

I guess you are still in that river in Egypt then.

I have some new lenses now such as 24-70L II and 70-200L II and I will do more tests later. With 24-70L II we can see they are very close and I also want to see with 70-200L II. I believe it's more meaningful to test resolution from a distance at least 50X focus length (outdoor tests are more meaningful in real world photos) not in very close distance (24-70L II is not a macro lens anyway for example). As I said DXOMark might use AVERAGE MTF resolution to calculate its P-MPix number. From so far what I have seen, I don't have indications that DXOMark tests are wrong and actually I think pretty reflect what I have found.

So you still refuse to provide the same detail to each sensor to see which one resolves more?

As the old proverb says .. "The truth is like a door, and you yourself must choose to enter"

Once again - DxO are conducting lens sharpness tests, not comparing the resolving abilities of different sized sensors.

It's not. DXOMark P-MPix is based on MTF data. I happen save some old DXOMark tests snapshots in MFT that shows 5Dc with 24-105L outresolves with 7D.

The door bill was taken in very close distance (macro mode) just outside the min focus distance, and still is only a very small center area (while I cropped out all 2/3 outer areas). Outside tests in the school park for example is more meaningful to represent real world photos. It's just too far away to see the details clearly at 38mm FF focus length on the schoold building that is 150-200 meters away.  Next time as I said I will test just cross the building from 30 meters away to see clearly. Even I upsampling 5D photo to 60D's size with 24-70L II, I can see they are pretty the same in center but 5D is sharper and at least has the same details at sides, and 5D photo is cleaner and much less CA. DXOMark's only give 1 P-MPix difference with 24-70L II between two cameras and they might use average MTF resolution cross entire frame (center, mid-range, edges and corners) to calculate P-MPix, and I believe their results are accurate. You see with 70-200L/4.0 IS that narrow down the building much "closer" we can see 5D is obviously better (as I said there is absolutely no mis-focus issues from 60D that only enjoys advantage thru precise 10x LV MF and I also tried AF).

Here is their testing protocol ... please take note of the target and then explain how this enables them to determine the resolving abilities of different sized sensors.

I await your enlightenment!

-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
schmegg
schmegg MOD
Senior MemberPosts: 5,002
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to qianp2k, Mar 19, 2013

qianp2k wrote:

schmegg wrote:

qianp2k wrote:

schmegg wrote:

Peter does - he continues to claim that a 5D out-resolves an 18MP crop.

I'm busy at the moment so unable to reply in details. Nevertheless please don't twist what I said.

Would you agree that you said that in almost all cases the 5D out-resolves an 18MP crop. I can only recall you mentioning the 300/2.8 IS L II as an exception.

Yes but based on what lenses I used to have and according to DXOMark.

I said clearly throughout posts in many threads that depend on lenses. With all lenses I have before such as 24-105L and 70-200L/4.0 IS...yes 5Dc does outresolve 60D.

No - it doesn't. And so far, from what we can tell from the comparison you posted, your own testing confirms this.

Yes from both DXOMark and from my own tests. My own tests actually confirm what DXOMark tests.

Not from what I've seen - though it's hard to tell the actual difference at the size you've posted them.

I guess you are still in that river in Egypt then.

I have some new lenses now such as 24-70L II and 70-200L II and I will do more tests later. With 24-70L II we can see they are very close and I also want to see with 70-200L II. I believe it's more meaningful to test resolution from a distance at least 50X focus length (outdoor tests are more meaningful in real world photos) not in very close distance (24-70L II is not a macro lens anyway for example). As I said DXOMark might use AVERAGE MTF resolution to calculate its P-MPix number. From so far what I have seen, I don't have indications that DXOMark tests are wrong and actually I think pretty reflect what I have found.

So you still refuse to provide the same detail to each sensor to see which one resolves more?

As the old proverb says .. "The truth is like a door, and you yourself must choose to enter"

Once again - DxO are conducting lens sharpness tests, not comparing the resolving abilities of different sized sensors.

It's not. DXOMark P-MPix is based on MTF data.

Mathematically derived from acutance tests. Do you even know what this means with reference to how much a sensor can resolve?

I happen save some old DXOMark tests snapshots in MFT that shows 5Dc with 24-105L outresolves with 7D.

No - they show that a 5D with a 24-105 shooting a target from a different distance has higher acutance. They do not, however, make any definitive statement about which sensor can resolve more.

And who mentioned anything about the 7D? It's irrelevant which 18MP crop is used. Could it be your irrational dislike of the 7D creeping in here? Hmm.

The door bill was taken in very close distance (macro mode) just outside the min focus distance, and still is only a very small center area. Outside tests in the schoo park for example is more meaningful.

Is it? Why? What if you wish to shoot a dollar bill near minimum focus distance and resolve the most detail? Which camera would you use?

It's just too far away. Next time as I said I will test just cross the building from 30 meters away to see clearly. Even I enlarged 5D to 60D's size with 24-70L II, I can see they are pretty the same in center but 5D is sharper and at least has the same details at sides.

Sharpness and resolution are not the same thing! But you use them interchangeably!

First you talk about the 5D out-resolving an 18MP crop, then you say it's sharper! They are different things! As I said before (did you even read it?), it's quite possible for an image to be sharper and have less resolution. (and round and round we go!)

It would be fair to say from you images that the difference is not significant either way.

DXOMark's only give 1 P-MPix difference with 24-70L II and they might use average MTF resolution cross entire frame (center, mid-range, edges and corners) to calculate P-MPix, and I believe their results are accurate.

Probably. They are probably accurate mathematical derivations of acutance tests - yes. They do not, however, tell you which sensor is capable of resolving more detail, nor is that their purpose.

You see wiht 70-200L/4.0 IS that narrow down the building much "closer" we can see 5D is obviously better (as I said there is absolutely no mis-focus issues from 60D that only enjoys advantage thru precise 10x LV MF and I also tried AF).

Lets see the raw files shall we.

Here is their testing protocol ... please take note of the target and then explain how this enables them to determine the resolving abilities of different sized sensors.

I await your enlightenment!

Nothing on this then?

Perhaps you need more time - I'm probably rushing you a bit - my apologies.

Reply   Reply with quote   Complain
Steen Bay
Veteran MemberPosts: 6,188
Like?
Re: *** 5Dc vs 60D resolution test ***
In reply to schmegg, Mar 19, 2013

schmegg wrote:

Sharpness and resolution are not the same thing! But you use them interchangeably!

Well, don't almost everybody do that? Photozone call it "MTF (resolution)". Isn't it OK to call it "the MTF-50 resolution", if that's what it is? Would it be better to measure e.g. the MTF-10 resolution? On DxO we could see the full MTF graphs before, but unfortunately not any more.

Reply   Reply with quote   Complain
Mako2011
Forum ProPosts: 15,486
Like?
relate
In reply to Steen Bay, Mar 19, 2013

Steen Bay wrote:

schmegg wrote:

Sharpness and resolution are not the same thing! But you use them interchangeably!

Well, don't almost everybody do that?

No. They define the two very differently and explain how the relate to each other.

Photozone call it "MTF (resolution)". Isn't it OK to call it "the MTF-50 resolution", if that's what it is? Would it be better to measure e.g. the MTF-10 resolution? On DxO we could see the full MTF graphs before, but unfortunately not any more.

Because they score a not well understood hybrid subjective measurement now.  Good for comparing like formats or different lens on same body. Less useful as an absolute but kind of an "in the ballpark" result.

-- hide signature --

My opinions are my own and not those of DPR or its administration. They carry no 'special' value (except to me and Lacie of course)

Reply   Reply with quote   Complain
qianp2k
Senior MemberPosts: 9,810Gear list
Like?
Re: relate
In reply to Mako2011, Mar 19, 2013

Mako2011 wrote:

Steen Bay wrote:

schmegg wrote:

Sharpness and resolution are not the same thing! But you use them interchangeably!

Well, don't almost everybody do that?

No. They define the two very differently and explain how the relate to each other.

Actually DXOMark uses the two words interchangeably but they don't mean acutance. DXOMark's P-MPix is based on MTF data.

Photozone call it "MTF (resolution)". Isn't it OK to call it "the MTF-50 resolution", if that's what it is? Would it be better to measure e.g. the MTF-10 resolution? On DxO we could see the full MTF graphs before, but unfortunately not any more.

Because they score a not well understood hybrid subjective measurement now. Good for comparing like formats or different lens on same body. Less useful as an absolute but kind of an "in the ballpark" result.

I agree DXOMark should publish the formula for how they calculate P-MPix, maybe based on average MTF cross entire frame (center, mid-range, edges, corners) mixed with some other tested data in some ohter areas?

For example DXOMark just tested Sigma 35/1.4 on D800 with 23 P-MPix vs 17 P-MPix on 5D2. I think it's about right.

http://www.dxomark.com/index.php/Lenses/Compare-Camera-Lenses/Compare-lenses/(lens1)/1056/(brand)/Sigma/(camera1)/0/(lens2)/1057/(brand2)/Sigma/(camera2)/0

-- hide signature --

My opinions are my own and not those of DPR or its administration. They carry no 'special' value (except to me and Lacie of course)

-- hide signature --
 qianp2k's gear list:qianp2k's gear list
Sony RX100 Canon EOS-1D Mark III Canon EOS 5D Mark II Canon EOS 5D Mark III Sony Alpha 7R +20 more
Reply   Reply with quote   Complain
Mako2011
Forum ProPosts: 15,486
Like?
Based on
In reply to qianp2k, Mar 19, 2013

qianp2k wrote:

Mako2011 wrote:

Steen Bay wrote:

schmegg wrote:

Sharpness and resolution are not the same thing! But you use them interchangeably!

Well, don't almost everybody do that?

No. They define the two very differently and explain how the relate to each other.

Actually DXOMark uses the two words interchangeably but they don't mean acutance. DXOMark's P-MPix is based on MTF data.

Based on but we have know idea how it is weighted. Like a the Move Lincoln...based on but not all that accurate in the scheme of things.

Photozone call it "MTF (resolution)". Isn't it OK to call it "the MTF-50 resolution", if that's what it is? Would it be better to measure e.g. the MTF-10 resolution? On DxO we could see the full MTF graphs before, but unfortunately not any more.

Because they score a not well understood hybrid subjective measurement now. Good for comparing like formats or different lens on same body. Less useful as an absolute but kind of an "in the ballpark" result.

I agree DXOMark should publish the formula for how they calculate P-MPix, maybe based on average MTF cross entire frame (center, mid-range, edges, corners) mixed with some other tested data in some ohter areas?

Agreed... at lest go more this direction

Sharpness

For example DXOMark just tested Sigma 35/1.4 on D800 with 23 P-MPix vs 17 P-MPix on 5D2. I think it's about right.

About right...in the ballpark...could be useful...might be OK. Who knows?

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads