A Pro's opinion of EVF vs OVF

Started May 30, 2013 | Discussions thread
chlamchowder Senior Member • Posts: 2,083
Re: Uh, no. Not at all.

My oh my, how convenient. I can still see (today) subjects in some of my camera EVFs that can't be seen in my DSLR's OVFs because it's too dark. Today.

There is a certain range if lighting conditions where live view/EVF can deliver a brighter image, but OVFs are definitely still usable. In extremely dark situations, I can still use the OVF (just let my eyes adapt), while live view/EVFs show nothing but noise.

And again, if you want what you get with an EVF with any modern DSLR today, just hit the live view button. Traditional DSLRs deliver everything an EVF camera can, and more. This is the key point - live view means you lose nothing with traditional DSLRs, and that traditional DSLRs will always offer more flexibility than EVF cameras.

Bad assumption. I missed nothing. A large DoF will indeed mask focusing errors if you don't look too closely, but they're still very visible when you do.

No, you missed the point entirely. Stopping down (especially when stopping down from f/2.8 to f/22) can make small focusing errors invisible, even when pixel peeping at 16 MP or 24 MP images.

In the real world, no one uses the "is it perfectly focused" standard because nothing's ever perfect. The "is it tack sharp" standard is what's interesting. And stopping down increases the number of tack sharp pictures because it masks AF inaccuracies.

The point that you missed entirely (if you even bothered to consider it) is that the AF accuracy has nothing to do with the aperture used or with the DoF. All cameras stop trying to refine the focusing accuracy beyond a certain point. With the PD AF sensors we're talking about it's almost always done using the f/5.6 part of the lens, although with some of Nikon's latest models in some cases it uses the f/8 part instead.

Yes, which is why fast lenses are more challenging for the AF system. We don't know how the V1 will handle fast long lenses, right?

Changing the aperture of your DSLR to f/11 doesn't help at all.

Yes, it does help, because it'll mask inaccuracies.

The AF accuracy is the same as it would be at any aperture from (depending on the lens) from f/2 to f/22 because the AF sensor uses the same f/5.6 part of the lens (while the mirror is down, between shots) to do its AF measurements and then adjust the lens's focus. And it doesn't care at all what the DoF might be.

But the key point that you're missing is that DoF is not the same, and therefore subject sharpness is not the same. What I'm saying is that we can't determine from the reviews that the V1 tracks as well as a traditional DSLR, because of how much depth of field it has. Can I say that compact cameras track well too, because everything's in focus anyways? Like it or not, DoF matters.

Also, I'll bring up another point: right now, on-sensor PDAF is completely unusable in low light. From the DPR review, " In our testing, at the long end of the 10-30mm kit zoom (which has a maximum aperture of 5.6 at 30mm) the switch happens at brightness levels between 9-10EV, which is roughly equivalent to a dull overcast day".

...so you can't even track on a dull overcast day because PDAF fails and it switches to CDAF. Indoors, or at nighttime sports games, on-sensor PDAF would be completely out of the question.

Should I also point out that there's no example of on-sensor PDAF with large sensor cameras (APS-C or full frame) that really works well? Like the Canon 650/700D (poor performance) or a99 (can't even select on-sensor points, and on-sensor points don't even work with most lenses)?

So now what you're really saying is that your A580's AF accuracy while tracking wasn't stellar, but at smaller apertures the greater DoF masked its inaccuracies. How does that in any way tell you that the V1's accuracy is any worse or any better?

I was using my experience with the a580 as an example of how more DOF masks focusing errors. I think that's exactly what's happening with reviewers saying that the V1 tracks well. It might be tracking well, but we don't know from those tests because of how much DOF it has. I can stop a 300/2.8 lens down to f/22, and probably get a very high hit rate in AF-S (not AF-C) mode because focusing errors will be masked.

How hard is it to understand that a very small focus miss (say, by 1 cm or so) is visible at f/4, but not at f/8?

Since Nikon has stated that the V1's AF tracking algorithms were based on the D3's algorithms, it's safe to conclude that they are very, very good. It's probably safe to assume that the A580's AF tracking algorithms aren't quite as robust. If they were, you might have had a greater hit rate even when shooting at f/4.

Right, but is the V1 using the same autofocus sensor? No matter how many lines of code you write, garbage in = garbage out. The V1's PDAF doesn't even work at low light levels (stuff that 30 year old SLRs could handle), for example.

But if you want to continue arguing about things that we can't know, enjoy.

So basically, we don't know that the V1 actually tracks as well as a DSLR with dedicated PDAF sensors, because no one has compared hit rate with comparable depth of field. Tighter DOF sets a far higher standard for focusing accuracy. You therefore can't make the claim that on-sensor PDAF delivers comparable performance to PDAF with a dedicated sensor.

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow