5D1 vs 1DsII - image quality

futureal33

futureal33

Messages
2,390
Name
Nick
Edit My Images
Yes
Afternoon all,

I have been doing some testing this afternoon with a few different cameras, to clear a few things up in my head regarding image quality.

I have come to the conclusion that I keep going back to my 5Dmk1 for absolute image quality - it is purely the auto focus which lets it down.

Can anyone who has owned both, comment on the differences between the 1DsII and the 5DI in terms if image quality alone. I know the 5D1 has a weak (or none?) Anti Aliasing filter - which might account for its sharpness, but everything else is also just "right" with it. Colours are rich, vibrant and noise is more like grain - I just love it.

If the 1DsII was simply a 5D1 with decent AF system I would buy one today - can anyone give me any info?

For info, the testing I did today compared it vs the 5Dmk3 and 1Dmk3

Cheers
 
futureal33 said:
Afternoon all,


If the 1DsII was simply a 5D1 with decent AF system I would buy one today - can anyone give me any info?

Cheers
Click to expand...

The answer is "yes", image quality wise both were very capable, and when I had the two to compare with- the 1ds2 actually gave a lot more reliable AF than the 5DMk2. The keeper rate was a lot higher on the 1ds over the 5d when using the outer focus points.
 
For those interested, here are some test results from a productive afternoon (nothing else to do thanks to the weather! haha)

So what we have is a Canon 70-200 f2.8 L IS MkII lens mounted on a tripod.
3 Cameras (5DMk1, 5DMk3 and 1DMk3) each connected in turn to the lens. Settings locked in on manual.

1/30th
f2.8
ISO 6400 (for the 5Dmk1, I used ISO 3200 and then pushed it +1 in Lightroom 4.3)

Focal lengths were adjusted to be the same with the 1.3 crop factor of the 1Dmk3.

10 second timer and mirror lock up. Focus with centre point on all bodies.

Images were all resized to 4368 x 2912px to be a fair test.

Images shot in RAW. No processing (other than the 5Dmk1 being pushed a stop)

100% crops of the focus point (the near eye)

1) 5Dmk1

5D1-crop by futureal33, on Flickr

2) 5Dmk3

5D3-crop by futureal33, on Flickr

3) 1Dmk3

1D3-crop by futureal33, on Flickr
 
To me the 5d3 is sharp at the eyes, where you say the focus point is. The rest of the head is also sharper than the other test pics. Everywhere else looks to be slightly oof/soft, for instance the blue and white scarf is softer than the 5d1.
The 5d1 looks to be focussed on the scarf, the eyes and rest of head are softer than in the 5d3.
The 1d3 looks softer everywhere and doesn't seem to have a clear point of focus anywhere. Or maybe it's just softened by image noise.

The 5d3 pic does it for me out of these three with the 1d3 weakest IMO.
 
stu.artd said:
To me the 5d3 is sharp at the eyes, where you say the focus point is. The rest of the head is also sharper than the other test pics. Everywhere else looks to be slightly oof/soft, for instance the blue and white scarf is softer than the 5d1.
The 5d1 looks to be focussed on the scarf, the eyes and rest of head are softer than in the 5d3.
The 1d3 looks softer everywhere and doesn't seem to have a clear point of focus anywhere. Or maybe it's just softened by image noise.

The 5d3 pic does it for me out of these three with the 1d3 weakest IMO.
Click to expand...

Pretty much what I thought!
The scene is in reality much darker than one is likely to ever want to photograph in normal circumstances, I mean, 1/30th ISO 6400 @ 2.8 is pushing the limits of any focus system - so I am impressed that the cameras even managed to lock focus - although it does seem the 1D3 missed.

Do you know if the 1DsII and 5D1 are comparable in terms of IQ alone?

Thanks :)
 
Sorry, I don't have any experience of the 5d1or 1dsmk2 (used a 300d/30d/7d/5d2 here). The 5d1 gave a good account of itself there though.
 
I preferred the look of the 5dc images over the 1ds2, very very close though. af was better on 1ds2 though. Kept the 5dc in the end as at that time never needed the better af.
 
Last edited:
futureal33 said:
Thanks for replying :)
Could I ask why you preferred the images from the 5Dc? Thanks!
Click to expand...

Seem to remember the 1ds2 being a little noisier and images required a little more work in pp, but there's really not much in it. The 5d screen isn't great but the 1ds2 screen is rubbish IMO, 1ds2 menu system is a little odd but I guess they have a reason for it and some like it.

So I stuck with the 5d for IQ, size, weight, only downer is AF outer points but center point which I mainly used was fine.
 
twist said:
Seem to remember the 1ds2 being a little noisier and images required a little more work in pp, but there's really not much in it. The 5d screen isn't great but the 1ds2 screen is rubbish IMO, 1ds2 menu system is a little odd but I guess they have a reason for it and some like it.

So I stuck with the 5d for IQ, size, weight, only downer is AF outer points but center point which I mainly used was fine.
Click to expand...

very slightly more noise, but it does retain lots more details as well (partly due to 16MP sensor)

It's a noibrainer - 1Ds2 does everything and well, 5D 1 or 2 struggle with AF and build quality
 
daugirdas said:
very slightly more noise, but it does retain lots more details as well (partly due to 16MP sensor)

It's a noibrainer - 1Ds2 does everything and well, 5D 1 or 2 struggle with AF and build quality
Click to expand...

IF you need that tank like build quality, which most dont if you look at the amount of Canon and Nikon D users here VS users of xD and Dxxx.
 
One thing I found this weekend which I didnt like (on the 1Dmk3) is that while the camera has 41 AF points, only 19 are selectable! This seems crazy!

Is the 1DsMkII like that? Or can you select from them all?

And, also, how reliable are the far outer points for locking on in dark conditions? Use would be for weddings, not exactly high speed tracking, but would like to be able to track the bride walking down the aisle in portrait orientation using the top focus point..

Thanks for the help so far :)
Nick
 
http://cpn.canon-europe.com/files/news/pro_lineup/EOS-1Ds-MkII-Whitepaper.pdf

You have 45 and can choose each one. All are very reliable up unlit exposure gets difficult at iso 1600

In practice the difference of af in 1d3 is minimal, it just has slightly less points in the middle but track better at awful light



P.s. I am very surprised to see how little 5d3 evolved over the old tech. I am a bit worried now as I was planning to get one. I can't afford 1dx no chance
 
Last edited:
futureal33 said:
One thing I found this weekend which I didnt like (on the 1Dmk3) is that while the camera has 41 AF points, only 19 are selectable! This seems crazy!


Nick
Click to expand...

Are you talking about how 19 out of the 45 AF points on the 1D3 are cross-type AF points for f2.8 and faster ?

You've probably read it but half way down here might explain better than me typing it....

http://www.imaging-resource.com/PROD...3/E1DMK3A4.HTM

The remaining 26 AF sensors have been demoted to "assist" sensor status. They contribute to AF operation, but can't be independently selected under user control. We don't see this as a limitation of any practical extent, as we've almost never found ourselves caring greatly whether we selected one particular AF point on the EOS-1 cameras or the one next to it. (Actually, we'll welcome only having to scroll through 19 sensors when making our manual AF selections, the selection process would go that much more quickly.)

The legend on the link above makes reference to "f/2.8 sensors" and "f/5.6 sensors," and also notes that the Center AF point is "f/4." Without going into all the details of how phase-detect autofocus systems work, at the heart of it is the need for light rays to reach the AF sensor, passing through different parts of the lens' optics. When the aperture is large, you can separate the light rays more widely, providing a longer "baseline" for the AF system, and in the process improve its accuracy. When the aperture is small, though, the light rays can't reach widely-separated sensor elements. Any AF elements having a wider separation than that allowed by the current aperture setting simply stop working, because they're not seeing any light from the subject at that point.

What this means for the EOS-1D Mark III is that its more closely spaced horizontal-sensitive AF arrays can work at f/5.6, while the more widely-separated (and therefore more accurate) vertical-sensitive arrays can only work at apertures of f/2.8 and larger.
 
daugirdas said:
http://cpn.canon-europe.com/files/news/pro_lineup/EOS-1Ds-MkII-Whitepaper.pdf

You have 45 and can choose each one. All are very reliable up unlit exposure gets difficult at iso 1600

In practice the difference of af in 1d3 is minimal, it just has slightly less points in the middle but track better at awful light



P.s. I am very surprised to see how little 5d3 evolved over the old tech. I am a bit worried now as I was planning to get one. I can't afford 1dx no chance
Click to expand...

I notice on the 1Dmk3 system, they made the outer focus points all X type - whereas on the 1DsII they are normal vertical type sensitive. In reality, in focusing on a persons eye what difference is this likely to make?

In the type of lighting I typically shoot in, the 5Dmk1 outer focus points would simply not lock here, they would hunt back and forth and never achieve lock - do you think the 1DsII would lock?
 
dinners said:
Are you talking about how 19 out of the 45 AF points on the 1D3 are cross-type AF points for f2.8 and faster ?

You've probably read it but half way down here might explain better than me typing it....

http://www.imaging-resource.com/PROD...3/E1DMK3A4.HTM

The remaining 26 AF sensors have been demoted to "assist" sensor status. They contribute to AF operation, but can't be independently selected under user control. We don't see this as a limitation of any practical extent, as we've almost never found ourselves caring greatly whether we selected one particular AF point on the EOS-1 cameras or the one next to it. (Actually, we'll welcome only having to scroll through 19 sensors when making our manual AF selections, the selection process would go that much more quickly.)

The legend on the link above makes reference to "f/2.8 sensors" and "f/5.6 sensors," and also notes that the Center AF point is "f/4." Without going into all the details of how phase-detect autofocus systems work, at the heart of it is the need for light rays to reach the AF sensor, passing through different parts of the lens' optics. When the aperture is large, you can separate the light rays more widely, providing a longer "baseline" for the AF system, and in the process improve its accuracy. When the aperture is small, though, the light rays can't reach widely-separated sensor elements. Any AF elements having a wider separation than that allowed by the current aperture setting simply stop working, because they're not seeing any light from the subject at that point.

What this means for the EOS-1D Mark III is that its more closely spaced horizontal-sensitive AF arrays can work at f/5.6, while the more widely-separated (and therefore more accurate) vertical-sensitive arrays can only work at apertures of f/2.8 and larger.
Click to expand...

Yeah thats right :)
It just suprised me as I didnt realise all 45 wouldn't be selectable - unlike the 5D3 where all 61 are fully selectable - that's what I was expecting to see.

I am glad the 1DsII has all 41 as selectable, even if they are not cross type
 
I use the 5Dmk1 for motorsport, and while it is not the ideal choice, I still manage to use the outer AF points quite regularly with a decent degree of success. Whilst I appreciate that this is not use in the most 'low light' of scenarios, I think that it definitely cannot be ruled that they are inoperable.
Quite simply, it is far from the ideal camera for the job for rallies, but still managed to put in an admirable performance despite its flawed AF system.
Just my 2 cents worth!
Cheers
 
futureal33 said:
I notice on the 1Dmk3 system, they made the outer focus points all X type - whereas on the 1DsII they are normal vertical type sensitive. In reality, in focusing on a persons eye what difference is this likely to make?

In the type of lighting I typically shoot in, the 5Dmk1 outer focus points would simply not lock here, they would hunt back and forth and never achieve lock - do you think the 1DsII would lock?
Click to expand...

In reasonable conditions all points work about as well and are very reliable. When you get a single lightbulb in a room f/2.8 lens scenario your 5DIII will AF a lot faster, but before then there is not a lot difference. Then you need to use One shot AF and that is not great, so no first dance in AI-servo for sure

Probably you need to try if you think you will push it to the borderline of its limits.
 
You must log in or register to reply here.
Back
Top