Why Focus Peaking Would be Useful on a DSLR Camera

[Sator]

It seems that Canon's attitude is that they make autofocus lenses and for that reason they don't need focus peaking on their DSLR bodies. It seems that as far as EF mount lenses go, Canon's attitude is that they make no manual focus lenses.

However, that isn't entirely true. In fact there are several manual focus lenses on offer in their current catalogue, namely their tilt-shift lenses:

TS-E17mm F4L
TS-E24mm F3.5L II
TS-E45mm F2.8
TS-E90mm F2.8

I can imagine Canon's attitude here. You are meant to set them up on a tripod and use live view, zooming in to check your focus as you shoot architecture. The trouble is that this is hardly the only use for a tilt-shift lens. I belong amongst a growing number of people who use tilt-shift lenses for creative portraiture etc.

There are lots of articles on the subject:

https://www.slrlounge.com/creating-artistic-portraits-using-tilt-shift-lens/

https://fstoppers.com/review/my-new-favorite-lens-not-found-very-many-photographer-bags-3186

It is hardly a situation where you would want to be forced to set up on a tripod and use live view to focus. Nor would you want to use autofocus confirm, which just isn't accurate on this type of lens.

The best solution is to get some sort of guidance from focus peaking. While you can install it using Magic Lantern, this may invalidate any warrantee. My solution? Use a Metabones adapter and shoot with Canon tilt-shift lenses mounted on a Sony a7II, and use its native focus peaking ability. But I wish Canon didn't force me to do that. I would prefer to forgo the cumbersome adapter and mount my tilt-shift lenses on my 5DsR.

The bizarre thing is that the EOS M5 has focus peaking even though that mount has not had a single manual focus native lens released for it, whereas the EF mount has several.

That's not all. Canon make macro lenses too don't they? At least I seem to own a couple of them and they still list them in their catalogue. The Canon MP-E65mm macro is a manual focus only model as well bringing the number of manual focus native lenses up to five in their current catalogue. We all know that it's a pretty basic aspect of macro that it is often preferable to override the autofocus and focus manually because the depth of field gets so shallow. Focus peaking isn't perfect either, but it can act as a useful guide.

If Canon are utterly petrified that their users will abandon their native autofocus lenses in droves and replace them with Zeiss manual focus lenses instead, couldn't they at least make it so that the body recognises a native tilt-shift or macro lens and permit focus peaking to be activated only then? Even better if Canon would additionally allow the activation of focus magnification. Unless Canon want to restrict use of their tilt-shift lenses to architecture and landscape only? Don't they want to let people explore creative uses of Canon tilt-shift lenses so that they appeal to a wider audience?

I really wish that Canon wouldn't ignore niche users for trivial reasons and just put focus peaking into all of their bodies. Or maybe Canon want to encourage me to shoot more with my tilt-shift lenses adapted to my Sony? Sometimes, I just wonder...

 

[Busted Knuckles]

Intersecting with your narrative. I use the 90 t/s and an extension tube for angled DoF and focus stacking along with magic lantern focus peaking fairly frequently. It works.

 

[Mt Spokane Photography]

When you say "Canons Attitude", could you provide a link or reference to such a statement?

Too often "Canons Attitude" turns out to be one someone made up, so I'm curious.

Although I do not normally manual focus, since it turns out far worse for me than Autofocus, I'm not against focus peaking.

 

[3kramd5]

My solution? Use a Metabones adapter and shoot with Canon tilt-shift lenses mounted on a Sony a7II, and use its native focus peaking ability.

I have personally found the peeking function on my sony (A7R2, which I suspect has the same implementation as yours) to be fairly useless. I may not have it set ideally (and the manual doesn't lend much insight), but it has a tendency to colorize anything contrasty, even if it's obviously well out of focus. While the obvious stuff doesn't bother me, its worrisome because I find I can't trust the unobvious stuff.

Any tips?

Too often "Canons Attitude" turns out to be one someone made up, so I'm curious.

In fairness, where the OP references Canon's attitude, it's prefaced with "it seems that" and "I can imagine".

 

[AcutancePhotography]

Yeah, I wished my cameras had focus peaking.

Not everyone wants/needs it, but for those who do, it is very handy.

I wonder if there is a significant cost to adding it?

 

[Sator]

It looks like Canon have a patent for PDAF based focus peaking according to the Egami blog:

http://egami.blog.so-net.ne.jp/2016-11-04

BTW for an official Canon USA Youtube video on using t-s lenses see here:

https://www.youtube.com/watch?v=mmPQVvF3EPQ

They mostly show it being used for architectural photography on a tripod. The standard technique is to mount on a tripod, use Live View and zoom in to check focus. But they do briefly touch on creative use where they show us this:



Really??? You are meant to focus like that? Unless this photographer had installed focus peaking via Magic Lantern on that camera, it would be very difficult to get the shot in focus. Very difficult.

Obviously, it's going to be rather hard to put focus peaking into an optical viewfinder. The Sony A mount DSLTs have focus peaking through the viewfinder because they have EVFs. If Canon could deploy their hybrid OVF/EVF patent on a DSLR that would allow viewfinder focus peaking. Even being able to focus peak through the LCD display would be very helpful nonetheless. Architectural photographers too would surely be grateful, especially if it were based on more accurate phase detection autofocus.

 

[rrcphoto]

It seems that Canon's attitude is that they make autofocus lenses and for that reason they don't need focus peaking on their DSLR bodies. It seems that as far as EF mount lenses go, Canon's attitude is that they make no manual focus lenses.

However, that isn't entirely true. In fact there are several manual focus lenses on offer in their current catalogue, namely their tilt-shift lenses:

TS-E17mm F4L
TS-E24mm F3.5L II
TS-E45mm F2.8
TS-E90mm F2.8

weird .. because with a TILT shift lens, you tilt the plane of focus, which makes focus peaking need even more odd.

90% of the use case of these lenses are with tripod mounted cameras.

You have to zoom in anyways for focus peaking to be of good use in critical cases, and if you have to do that, I find that if you can't tell on the LCD it's sharp without focus peaking then focus peaking isn't going to help you much.

focus peaking for fast manual focus primes? certainly. especially with an EVF.

TS-E's ..yeah okay .. whatevs.

especially with the more modern 1.6MP or better LCD's .. and a modern UI / touchscreen that you can quickly scan around the image zoomed in.

I'm also curious how you know canon's "attitude" considering they do have peaking for example on their mirrorless cameras.

for some reason that exists on the powershot firmware base, but hasn't made it over to the DSLR firmware base.

Them, like me.. probably does not see it as an important feature. if you have to use the LCD in either case, you don't need focus peaking.

 

[Sator]

For those who have never used focus peaking on MF lenses, it is not PDAF based so that it will not lose accuracy if the lens plane is thrown off the focal plane of the off-focal plane PDAF sensor. In fact, it remains just as inaccurate with T-S lenses as it is with ultra fast lenses. I have used T-S lenses as well as MF f/0.95 lenses. It is even more important with T-S lenses because in creative use, it often causes the plane which remains in focus to reduce down to a tiny slither. Yes, irrespective of lens type, focus peaking remains only a guide, but one that is extremely useful to have, as anyone who has used a T-S with focus peaking will have found.

This raises the question of how you would implement PDAF based focus peaking on a T-S lens. The focal plane of the lens would fail to match with that of the off-focal plane PDAF sensor. You would probably have to rely on on-focal plane PDAF points after moving your focusing point around, if that were possible, that is.

As for the Canon mirrorless M series, there are no MF lenses for this mount at all. Yet Canon shows gross inconsistency in offering focus peaking because of the bizarre fad for adapting vintage lenses to mirrorless cameras. Nobody says "get a Canon DSLR so you can adapt Nikon lenses to the EF mount" or "get a Leica rangefinder so you adapt lots of non-native lenses to it". The design discrepancy and the particular needs of mirrorless mount designs mean that SLR lenses perform sub-optimally.

 

[3kramd5]

This raises the question of how you would implement PDAF based focus peaking on a T-S lens.

With OSPDAF (such as canons DPAF).

 

[Sporgon]

I have personally found the peeking function on my sony (A7R2, which I suspect has the same implementation as yours) to be fairly useless.

This has been my experience too.

 

[Josh Denver]

Me 3. My A7s has a much worse and much less accurate focus peaking than my 60D with Magic Lantern. Especially that magic stuff called Digic Peaking, lord that makes manual focus A BREATHE

Nonetheless, most of Canon DSLR users are set on AF and I guess that's why Canon has ''forgotten"? peaking feature? But then again, not only as you list Manual focus-only lenses, but actually for video users, all Canon EF Mount glass must be used in manual focus mode so add a 100 to these five if you count video in. Canon engineers, software developers, feedback observers, marketing departments, are less than probable to "forget" the Peaking feature. Especially since they did just add it on the M3 and M5.

It does not cost or takes hardware power at all (based on ML Peaking, that is, which is better than most peaking I've seen, only bettered by f5 and C100II's EVF peaking)

So it's on their mind, doesn't cost them, checks a feature box and a spec line to add, what gives?

Just does NOT make since why at least the latest Canon DSLRs (80D - 5DSR - 5DIV - 1DXII) Don't have a little On/Off focus peaking icon in liveview mode.

Since I've bought my first M42 to EF adapter, I have more manual glass in my posession, and use more manual glass in actual shooting, than Canon EF lenses. These old russian primes from the 50's and 60's that cost nothing are simply GORGROUS. Shame I have no peaking on my Canons which have no ML (My 1100D has ML but my 1200D does not. 1300D, 750D, 760D/T6s, 80D, 7DII, 5Ds, 5DsR, 5DIV 1DXII all have no ML thus no peaking. Virtually all Canon current line up except for the still in-production T3/1100D, 70D, 6D and 5DIII kits still on sale.

 

[j-nord]

Would be great if this was available natively. I just picked up my first manual lens (roki 35 f1.4) and immediately wished my camera had focus peaking.

 

[retroreflection]

For me, using either a dslr or a mirrorless, more than 90% of my shooting is through the eyepiece.
Focus peaking is an option for mf assist, but not my favorite. I think 10X zoom is the best option in an electronic display, but that is just me. An accurate focus screen in an optical viewfinder is also a solution. It might be the best option overall, but it has been decades since I have had the pleasure.
In a real world dslr, for me, focus peaking would be an option that I would probably use once. That is, unless Canon implemented it (in live view) better than I have experienced elsewhere. Even then, I would really want an accurate focus screen for my majority use case (eyepiece) with all those cool mf lenses.
Disclaimer, I use 10X zoom in live view to get accurate focus with my 24 ts-e. Never wished for focus peaking. Have wished for a better focusing screen. Not enough gumption to change it.

So much of that is just me, but the dominance of the ovf in Canon's product line is a significant factor here. No implementation in most use cases diminishes the drive to add the feature.

 

[Mt Spokane Photography]

I've used a focus screen with a split prism on my 1D MK III. Not very accurate on a DSLR though. Its like that focus confirmation light, it tells you that you are approximately focused.

Live view with magnification takes time, but it is accurate, and when tethered using my large monitor with additional magnification, I can focus to an extreme accuracy just where I want it.