canon eos 1ds mk2 verses canon 1d - help

artona

artona

Messages
2,683
Name
stewart
Edit My Images
Yes
Hi

http://i300.photobucket.com/albums/nn10/artonas_photobucket/Jessica-1.jpg

Last weekend I bought an elderly Canon Eos 1d from ebay. Today I took it out to take a few snaps to check it out. Now I normally use a 1ds mk2 and to be honest I did not expect the 1d to be any good but when I opened the above image I was amazed with the sharpness and colour. The shot above was literally a snap, no effort at all and no photoshopping.

So I am now wondering if I have the 1ds mk2 set up right. Any ideas anyone please

stew

Staff Edit : Image(s) removed. (Link(s) left in place). Pictures must not exceed current forum limits as per the rules.
Please feel free to replace this with a fresh/resized image and remove this text
 
nice shot. I am not shocked you find it "sharp out of the box". The 1D was a great camera. The extra pixels of the 1DS M2 would make your shots appear a lot softer when zoomed in.

I would expect your DS to ultimately have better contrast and handle lower light better even with the high pixel count.
 
More pixels = less sharpness.

Not questioning what you are saying but how come? I am the first in line to admit I am a photographic beginner when it comes to equipment

stew
 
1D is 4MP at 1.3x crop
1Ds2 is 16MP 1.0x FF
Even without doing maths it is obvious that 1D has considerably lower pixel density. It tends to mask lens defects and focus error.

1Ds2 has pixel density close to 10D (6mp APS-C). My 40D (10mp APS-C) is extremely sharp at ***pixel level*** with L or macro glass at f/5.6-8. I am really confused why would 1Ds2 would have soft images. So maybe you could:

* tell us what lenses and settings you used
* show some samples and 100% crop from 1Ds?
 
HI

Using 24 - 105L. I will be shooting some shots over the next few days on both cameras with the same lens. Will post asap

stew
 
As I understand it its down to resolving more detail (gradients) which looks less sharp when zoomed right in for a given scene.

Also different anti-alias filters can impact the sharpness of an image.

Look forward to seeing the comparison shots.

I am assuming you are you shooting raw? Jpeg out of the camera could of course have different levels of sharpening and colour settings applied.
 
Up to about a 10in print, 4mp is all you are ever going to see anyway at 250dpi - which is a very good standard. At 72dpi screen res, it's going to be enormous before you max out the sensor, plus you should have good dynamic range from the large pixels, excellent contrast and saturation. Resolution is only part of the story. I'm not in the least surprised :)
 
indolent said:
The extra pixels of the 1DS M2 would make your shots appear a lot softer when zoomed in.
Click to expand...

This is a false theory that is mentioned quite often. Today's high-res cameras like the 5D2 capture so much detail that far more care must be taken when using them to ensure sharpness at 100% magnification. You can't rattle off shots on a 5D2 or D3x like you used to on an old 1D and expect perfect sharpness every time. The detail captured by today's cameras shows up every minor defect when viewed at 100%. In my opinion, cameras like the 5D2 need to be treated with the same time and consideration as medium format cameras. If you know what you're doing, cameras like the 1DsII & 5D2 capture astounding detail.

To illustrate this, here's a shot I took a few months ago. I admit the photo is nothing special but look at the detail in the 100% crop. These images have not been played around with. If you look extremely closely you can just about make out my own reflection in the eye of the bird! Now do you see what I mean?

image.jpg


100% crop from above image:

100.jpg
 
Not yet fully........SO if you took an image from the camera and printed it to what ever size the camera makes - 50"x?...thats when you will get optimum quality. By reducing it to 20x16 the pixels are tripping over themselves.

When you say knowing what you are doing can you expand on that

cheers

stew
 
artona said:
Not yet fully........SO if you took an image from the camera and printed it to what ever size the camera makes - 50"x?...thats when you will get optimum quality. By reducing it to 20x16 the pixels are tripping over themselves.

When you say knowing what you are doing can you expand on that

cheers

stew
Click to expand...


When talking about prints, there is no such thing as "the size the camera makes". That is entirely dependant on what dpi your image is saved at and the size you instruct the printer to output.

The common "ideal" dpi is 300dpi but in reality 250 dpi is more than acceptable. Take the 1D as an example. It's images are 2464 x 1648 pixels. If you determine that 250dpi is good enough for an optimum print you can output a print with a maximum size of 9.86 inches x 6.59 inches. Anything larger than this will break your 250dpi and the image will gradually degrade as the size increases.

You can of course print at any size below 9.86 x 6.59 and the quality will not degrade. Any pixels "tripping over themselves" will not be visible due to the smaller print size.

When I was talking about "knowing what you're doing" I simply meant that users of high res DSLRs should make an effort to apply good photographic technique. Take more time to ensure the focus is perfect (check Live View at 10x if possible), use a tripod more often and try to use the optimum aperture for any given lens when possible. All this extra effort will pay dividends when making large prints or when viewing at 100% magnification.

Naturally, all this advice goes out of the window if you're making prints no larger than A4 from a 5D2 or a D3x as any minor flaws in technique will not be visible.
 
Hi

I see. I shoot mainly portraits and normally produce between 20 - 30 20"x16" canvas prints per week plus about 300 12"x8" dye sub prints.

The way I work a tripod is a no no and most are under studio flash.

stew
 
Edit: Oops, crossed post with Oracle. Looks like we're saying the same thing :)

artona said:
Not yet fully........SO if you took an image from the camera and printed it to what ever size the camera makes - 50"x?...thats when you will get optimum quality. By reducing it to 20x16 the pixels are tripping over themselves.

When you say knowing what you are doing can you expand on that

cheers

stew
Click to expand...

I'll have a go. This can get complicated because there is a lot more to sharpness than resolution (pixels, or dots per inch). Contrast - both colour and tone - is arguably more significant when we look at at image. And with dpi, printers use several different drops of colour to create just one dot.

But at the heart of it, industry standard says that 300dpi is the highest level for critical viewing. If you look at a print under normal conditions, you will be very hard pressed to see detail any smaller than that. In practise, 200dpi looks damn good, but if you go peering at huge prints with a magnifying glass you are changing the rules of the game. That is effectively what pixel peepers are doing and it distorts expectations a great deal.

If you accept that basic assumption of 300dpi, then the maximum number of pixels you can see in a 6x4in print is 2.16m. If you have more pixels than that, you are simply seeing several combined to produce one dot. Increase the print size to 10x7in and you have 6.3m, and anything over that is theoretically wasted; A4 maxes out at 8.7m. It is only when you get really big when you start to see differences, like A3 (420x297mm) which will swallow 17.5m - about where the new Canon 7D sits.

Likewise, cropping ramps up the pixels a lot and with birding for example it is not uncommon to crop out half the image, or even more. That's why birders are so interested in the 7D, because the APS-C sensor gives extra reach, plus the bonus cropping potential. This puts huge strain on the lens however, and an inescapable optical fact is that when resolution goes up, contrast goes down, and there is only so much that can be done in post processing to restore it. In practise, only the best lenses, at optimum aperture, can max out a 7D sensor at any meaningful level of contrast.

In a nutshell, if all you do is view on screen or make smallish prints, then in terms of pixels you frankly don't need many. And unless you print very big, or crop a lot, the same applies. But that is only looking at the pixels side of things - contrast, colour, noise, dynamic range etc, all those things are very important and you get better performance in all those areas with bigger pixels rather than more pixels. For example, if you take a 4mp image from a DSLR and compare it to a 4mp image from a compact (and their pixels are miniscule) the DSLR will blow it clean away.
 
What f-stop, speed and ISO do you normally shoot? 24-105 is sharpest at f/8 as far as I remember... Then focusing is very important. 16 out of focus MP are inferior to 4 well focused MP. 40D is really giving me a headache in many situations at some focal lengths (tracking a dog is a real nightmare for example).
 
artona said:
Hi

I see. I shoot mainly portraits and normally produce between 20 - 30 20"x16" canvas prints per week plus about 300 12"x8" dye sub prints.

The way I work a tripod is a no no and most are under studio flash.

stew
Click to expand...


Don't worry, a tripod isn't necessary in your situation. You will never get camera shake using just flash. I've never really understood people using tripods for taking portraits with flash in a studio unless they need the camera to be in the same position for a long time.

In controlled studio situations it is much easier to get consistently good results. Take your time and get your focus bang on and you'll be fine.
 
daugirdas said:
What f-stop, speed and ISO do you normally shoot? 24-105 is sharpest at f/8 as far as I remember... Then focusing is very important. 16 out of focus MP are inferior to 4 well focused MP. 40D is really giving me a headache in many situations at some focal lengths (tracking a dog is a real nightmare for example).
Click to expand...


Even with today's DSLRs that have great high ISO performance I still try and shoot at the lowest ISO possible in any situation. That guarantees optimum quality for each shot. The 24-105 is actually very good wide open at f4 (at least mine is) but as a general rule, most lenses perform at their best when stopped down a couple of notches from their maximum aperture. e.g. If your lens is f4 shoot at f8, if it's maximum is f2.8 shoot at f5.6. Of course this is a general rule and it all depends on what depth of field you need for the shot at the time.

You're absolutely right. A sharp shot from a digital compact is better than an out of focus shot from a 5D2. HoppyUK also makes a valid point about contrast and colour. Lens sharpness isn't everything. However, any shot showing an obvious focusing error is ruined and may as well be deleted.

When taking shots in quick succession with my 5D2 of moving subjects I find it far better to stop down the aperture and increase the ISO if necessary. The narrower aperture helps keep slight focusing errors in check due to the increased depth of field. That's a good tip to remember.
 
Comparison pics

1dverses1dsmk2.jpg



Taken at F8 200asa on the 24 - 105L lens at 105mm

stew
 
Same photos but I have enlarged the full frame image so the speed sign is a similar size

1dverses1dsmk2_blownup.jpg
 
At these sizes it is hard to tell any difference. The only way is to shoot two identical shots and post 100% crops from each image. If you're shooting RAW then each shot would need to be processed exactly the same.

Also remember that the 1D has the advantage of shooting through the sweet spot of the lens because it doesn't have a full frame sensor. In theory that should mean better quality at the edges of the 1D images. In reality that may not prove to be true.
 
They look very similar, and there are no fine details in the photos. You may wish to try something with fine texture or small details, like brick wall or some city scenes. And definitely shoot RAW.
 
all shots taken on raw and opened the same way, no photoshop
 
artona said:
first pair were full crops
Click to expand...

In that case I'm with daugirdas. It is very hard to see any difference. I'm at work at the moment using a crappy monitor.

Are you absolutely sure they are both crops at 100%? When you say the first pair were full crops do you actually mean full frame?

Look at my pictures of the bird. The first shot is the full frame, the second shot is a full crop (at 100%). We need to see full crops from each camera.
 
what do you mean as full crop? I have taken the whole image and cropped to 20inch at 300dpi so I could see it on screen. I then cropped the whole image to 1024px - the maximum the forum allows.

The biggest difference I see is in th edetail on th ecobweb under the speed sign - the d1 has more detail and the green is greened in the D1


stew

stew
 
artona said:
what do you mean as full crop? I have taken the whole image and cropped to 20inch at 300dpi so I could see it on screen. I then cropped the whole image to 1024px - the maximum the forum allows.

The biggest difference I see is in th edetail on th ecobweb under the speed sign - the d1 has more detail and the green is greened in the D1


stew

stew
Click to expand...

Set image to 100% magnification, and the take a small are of that (like 800x600). Don't resize.

You could also upload your RAW files somewhere and we could have a good look then. rapidshare.com is a good place.

And we really need to see a different image with lots of small things across the frame. Leaves and road signs do not reveal true resolution at all.
 
As daugirdas said. View your image at its maximum resolution on screen and cut out a small section of it. If that section is, say, 500x 500 pixels, cut out an area that is also 500 x 500 pixels from the other image. TRy and take them from the same portion of the frame.
 
You must log in or register to reply here.
Back
Top