Lens for 5Ds?

[Squidy]

Hi all,

I've recently upgraded to a Canon 5Ds and loving the resolution, however I've noticed photos with my 70-200 2.8L non IS aren't as sharp as they used to look. I'm guessing it's the higher resolution in the sensor mixed with slight camera shake.

I've got the Tamron 24-70 2.8 VC and it seems ok so far, do you think if I upgraded to the IS version of the 70-200 I should notice an improvement or should I perhaps look at another lens?

On screen the pics look crisp, but I had some blown up of my grandfather in an ANZAC day march and his face was blurry at larger sizes whereas my landscapes on the 24-70 at 10x18 inches look crisp (enough) all the way through.

[peety3]

http://www.the-digital-picture.com/News/News-Post.aspx?News=15356

[Squidy]

Yeah I read through that earlier, but I was just wondering if the lens upgrade would be beneficial or if I was potentially doing something else wrong

[DavidEccleston]

I'm looking at the Image Quality Tool pages. First, choose your lens, and the 5Ds. If you're seeing something softer than what's there, then there's something else wrong. If you're seeing a similar level of softness, and it's unacceptable, then keep reading.

The 70-200 f/2.8 L IS USM you're considering was only ever tested on a 1Ds2 and 1Ds3. So, choosing 1Ds3 for both lenses, I see that the IS version is quite a bit softer at 70mm (and that softness will be magnified on the 5Ds or any crop body). At 200mm, they look about the same, but the IS has less vignetting.

Right, so the IS version I isn't going to help... at the short end it will make things worse, so that leaves "other lenses". Without much guidance it's hard to suggest other lenses. I'm assuming you're not considering the version II lens based on price, so that also rules out things like the 100-400mm version II as well.

Getting sharp, cheap, means compromising elsewhere:

Aperture
Flexibility (using primes)
Going non-Canon, and potentially having weaker AF
Used, Old lens designs... I picked up a 70-210 f/4 push-pull zoom for cheap for Bella. Seems sharp to me, but focus is noisy and slow.


You can consider the 70-200mm f/4L IS USM lens. It's sharper than the f/2.8 versions, and you'll maintain your fast focus.
The 200mm f/2.8L is supposed to be fantastic (except when compared to the f/2L!)
I thought you could go 70-300mm f/4-5.6L IS USM, but it looks like, when stopped down to match apertures, your lens is already sharper (but it has more Chromatic Aberration).

[Squidy]

Thanks for the info

I don't mind springing for the 70-200 IS mk II, but at $2800 I just want to make sure it's right :)

So you'd recommend the F4 IS over the 2.8 IS? I'm planning on being a second photographer at weddings and low light is a must, but if it's sharper and not much difference in low light I'll definitely look into it

[clemmb]

Have you made AFMA?

[Kayaker72]

First, I would check and re-check your AFMA for each lens on the new body. It would seem odd to me that a lens would perform worse on the 5Ds. Rather, I would expect you to maybe not achieve everything the 5Ds resolution is capable of, but not worse than before given viewing at the same size (like a 10x18 print). At the pixel level, sure. Put another way, I would still expect the lens to resolve a 10 x 18 print. If it did it before, it should still be able to do it.

Second, especially considering the price, you might consider renting a lens like the 70-200 II first. Or maybe buying one used which you can then sell at, hopefully, even money.

[HDNitehawk]

With the 5Ds you should be seeing improved resolution over other bodies. With certain lenses you will see less improvement than others.
For instance I see less improvement from my 500mm than I do from my 24-70 II. I would have thought it would be the other way around since the 500mm is arguably Canon's sharpest lens.

In your case your base line for improvement should be from a 6D or 5D III, not one lens to the next. You should see some, but definitely not less.

The camera shake / sensor resolution theory that keeps getting thrown around with the 5Ds is a crock in my opinion. It just hasn't panned out in my experience. It gets thrown out quit a bit by individuals looking for a reason to not buy the 5Ds, far less by those that own one. If you are getting camera shake with the 5Ds you were getting it with your other FF and it is poor technique not resolution that causes camera shake.

[DavidEccleston]

Agreed that the lens should be performing equal or better to previous bodies for a print of the same size. But, when Squidy mentions zooming in and not seeing clear details, then that's him looking for details in the extra resolution the 5Ds offers, and wondering how to get those details. That's the question he's asking. Would the new lens improve sharpness/detail at the pixel level... and that's all my above answers were about.

If you aren't concerned about the extra detail at the pixel level that the 5Ds offers, then you don't need a 5Ds. The whole point of the 5Ds is to get those extra details. If you're fine with the pixel level details being soft because it will print the same as your 5D3, then stick with the 5D3.

Which bring the argument from 'same photo, printed same size, looks the same' to 'I want more detail when pixel peeping'.

I also agree that camera shake isn't a good reason to not get a 5Ds. It will affect the sharpness at the 100%. Faster shutter speeds or better technique or a tripod are more important to the 5Ds than a 5D3. Then again, your old technique may have been good enough. It's only a concern when your technique/shutter speed was only marginally okay on the lower density sensors. But, if you really want those details, you'll bump the shutter speed, or invest in a good tripod, or both.

AFMA is a good point too. I adjusted a few lenses on my 7D2, and it made quite a difference, even if the change was only +1 or +2.

[Squidy]

Hi everyone,

Thanks for the feedback. I also think the camera shake thing is a crock as it doesn't make sense to me. I've never had an issue with camera shake on any of my bodies previously so I don't know why I'd get it now.

I had a closer look at the image, and why the focus point was exactly over my grandfathers face it was zoomed to 200mm and 1/200s so I'm just going to assume that I should have used a faster shutter speed (as this has no IS)

As for AFMA I haven't made any, I don't have the gear required to do so.

Also, the only reason I got the 5Ds was I recently bought a 5D MK III and had to return 2 of them as the mirror kept getting little "pings" in it from smacking something inside the camera, and the mirror slap sound also sometimes sounded like a "ting" so I knew it wasn't right. The store sent them back to Canon but couldn't get any more in for a few weeks so they made me an offer on paying a little more to upgrade to a 5Ds so I figured it's a newer camera with higher res so it'd be good for blowing up landscapes etc so I decided to upgrade. I'm not a pixel peeper as such but I must admit I do sometimes like zooming in just to make sure it's all sharp, even just to train myself on what I could potentially be doing wrong if not.

In regards to clear details, it's more that I was focusing on my grandfather's face and it turned out blurry... mind you the rest of the shot is pin sharp... It's only in the one shot so far so I might put it dowen to something I did wrong

[DavidEccleston]

At 1/200s it could be your movement, his movement, or both.

I did AFMA the low-tech way. In good light (ie: a bright day), shoot a flat object next to a slanted one at typical a distance. A hardcover book and a meter stick at a 45 degree angle with a specific mark (ie: 50cm) aligned with the front of the book. Review on camera at 100% zoom. Judging from the slanted part, did it focus on, in front, or behind the flat object, where I focused. If it's off, adjust AFMA and and repeat.

Sure, it's not automated, but it didn't take long, and gave good results... I haven't tried to AFMA any zoom lenses though. Newer cameras allow for different AFMA settings for wide and tele ends of the lens. It *should* be as easy as doing the steps above for each end of the zoom.

[Busted Knuckles]

I rented the "R" version and found that my technique had to be spot on to get the resolution the sensor offered. I did several multi shot panos and wouldn't lock down for each frame and relied on the tripod and shutter speed - 1/200 and several of my shots upon close inspection had camera motion blur from me mashing the shutter in a hurry to take the 200 frames.

The lesson - technique, technique technique - a gentle shutter, good support and faster than you think shutter speed.

Mike

[Squidy]

Well, I went out tonight and shot a band with the Tamron 24-70 and the focus was spot on for the most part so I'm pretty happy... The noise on the 5Ds though at anything over ISO 1600 is absolutely insane. I mean it's grainy as... Well, an incredibly grainy thing. This definitely won't be my go to camera for low light.

[DavidEccleston]

When scaled to the same size as the 5D3, the image *should* look similar, noise wise. But yeah, at a pixel level, it will quite a bit more noisy than you're used to. As noisy as a yucky crop body. ;)

I love my noisy 7D2... most of the time.

[Squidy]

Ahhhh ok, I was just viewing it in Lightroom, haven't done any scaling yet

[jrw]

Suspect that AFMA and perhaps using a tripod will help. One easy way to field test is to use live view to focus on a target. Turn off live view and focus on subject. If the distance scale shifts then you have a problem and know which direction to adjust.
You 'blew the image up'. If this involved resizing the image what settings were used in what software? This could seriously affect things.
Were the images printed? Were they sharpened sufficiently for the media in use? What was the PPI count? Different textures and finishes do want different treatments.
Just a few other items to consider in the workflow where softness can creep in.

[neuroanatomist]

My recommended 'DIY' AFMA setup needs a printed focus target (http://www.bealecorner.org/red/test-patterns/star-chart-bars-full-600dpi.png), a measuring tape, box and a chopstick or pencil.

http://media.the-digital-picture.com/Images/Tips/AF-Microadjustment/AFMA_Image4.jpg

[John Coster]

Good morning! (Well it is here in the UK!!)

I also have a 5DS-R and an EF70-200 f2.8L IS II, and here's what I have learnt. The 5DS's high resolution put high demands on lenses and user and it shows up any weakness. I also learnt that pixel peeping can be depressing verses reviewing a printed image for normal review. But overall, I'm staggered at the image detail this camera and lens combination can achieve.
Practical things I've learned with the 5DS-R:-
I use the silent shutter mode to help reduce shutter vibration;
Use the back button for focusing, personal thing, but for me seems to aid a smoother shutter press;
I have also invested a good tripod, something else this combination, due to weight/size and resolution, demands.

Another element in this thread, is the live music, I also enjoy taking pictures at local music events, normally unsigned cover bands in local pubs where the light can be terrible. For this I enjoy the 16-35 f/4 for the wide angle if not the f/4 limit and the 24-70 f/2.8 helps with focal range and low light, but my favourite lens is the Sigma 50 f/1.4, which works well in low light and provide a great DoF to pick out an individual, although focusing in low light with the Sigma is often a slower affair.
I've also got over worrying about high ISO and Noise with the 5DS-R, especially with live music. Yeah sure, if you Pixel peep - loads of noise! But I'm sharing images with unsigned bands, who are really pleased. Not once have I ever had "that's noisy" - I have occasionally had "I like the grain!!!"

No real advice in my post, other than try to develop a rock steady technique and be less sensitive about high ISO Noise!

I'm still learning, and this forum and resources have been a great help over the years.

PS - Happy New Year!