The Camera Store posted a great AF Comparison video between the X-T1, E-M1, A6000, GH4, and D4S

[napilopez]

The results may surprise you =]

https://www.youtube.com/watch?v=up8K_xd_iwU

Spoilers:

To supplement the video with some of my own thoughts, and as I mention in my review thread, the GH4 supports live view during burst shooting if you set it to medium speed (which will still deliver the max 7ish fps for AFC), something it inexplicably doesn't at the high speed setting. That alone ultimately makes the shooting experience that much more DSLR like than the with X-T1 and A6000 (and I think the E-M1 as well, though I don't have one of those on me). That plus the added AF zone options of the GH4 make it by far the most versatile of the tested cameras.

I found the results of the E-M1 vs A6000 for S-AF a little surprising, as I generally find the A6000 to be a teensy bit slower than my even older E-M5 when I test it--might simply be a matter of lens choices and environments. Either way, they're both super fast and it's probably a negligible comparison in the real world.

The X-T1 is a bit of a finicky action camera. It performs excellently in certain scenarios, and poorly in others. On one hand the predictive AF is probably the best of any of these cameras when an object is running straight at you. On the other hand that only works for approximately the central 1/9th of the image frame. Also, as mentioned, the slow/uncertain initial focus often makes changing from subject to subject a bit of a pain, and the evf freezes momentarily when acquiring focus. But most of the time, it's very fast.

And as I keep saying over and over again, the GH4 deserves to be regarded as just as exciting for stills camera as it is for video. But the most important part of the video is Chris' takeaway: all these cameras now match or better DSLRs in their price brackets for action shooting (arguable, but they're competitive at the very least). All mirrorless users win.

 

[OzRay]

I'll take your word for it, since I can't see a link or anything.

 

[napilopez]

Try refreshing? It wasn't working originally. Just incase, here's the URL

"https://www.youtube.com/watch?v=up8K_xd_iwU"

 

[wildwildwes]

I found the results of the E-M1 vs A6000 for S-AF a little surprising, as I generally find the A6000 to be a teensy bit slower than my even older E-M5 when I test it--might simply be a matter of lens choices and environments.

Yup. Agreed. I'm wondering how the E-M1 would've faired if the reviewer's were using the Oly 12-40 on BOTH the GH4 & E-M1... (as opposed to the Pany 12-35 they used in the review tests). Right?!?

 

[OzRay]

Providing an actual link is always helpful, as I don't have embedded video enabled.

 

[OzRay]

First off, I was really impressed that the reviewer actually called the Nikon a 'traditional SLR' not 'traditional Dee-SLR' as every other reviewer does. I don't know why, but this constant reference to 'D' as in digital annoys me as film, for the average pundit, is long dead. I wouldn't even use DSLR anymore, but in photo forums and the like, readers would still think of a film camera if I mentioned SLR.

OK, minor diversion over, what amazed me was that the E-M1 C-AF came out the best, I would never that expected that in a million years. Then what surprised me was that the S-AF of the E-M1 was so comparatively slow in this test. This is completely opposite to anything that I would have expected, and floors me as to how this came about. With my 4/3 lenses, I find the E-M1 incredibly fast and was always of the understanding that with m4/3 lenses, it would be much faster. Very strange indeed.

 

[napilopez]

Yup. Agreed. I'm wondering how the E-M1 would've faired if the reviewer's were using the Oly 12-40 on BOTH the GH4 & E-M1... (as opposed to the Pany 12-35 they used in the review tests). Right?!?

Yeah, I mean it's always hard to test with these things--you can't possibly check for every scenario. Another thing to note is that they chose to match aperture values rather than depth of field ones, although in my testing I've tried to match DoF and results are essentially the same.

OK, minor diversion over, what amazed me was that the E-M1 C-AF came out the best, I would never that expected that in a million years. Then what surprised me was that the S-AF of the E-M1 was so comparatively slow in this test. This is completely opposite to anything that I would have expected, and floors me as to how this came about. With my 4/3 lenses, I find the E-M1 incredibly fast and was always of the understanding that with m4/3 lenses, it would be much faster. Very strange indeed.

Yeah, well let's be sure to keep things in perspective, if the E-M1 is slower than the A6000 but faster than the X-T1, than that still makes it a VERY fast camera. But Sony did do an amazing job with the A6000's focusing system. This is, at the end of the day, just one particular scenario.

 

[OzRay]

I wonder if the latest firmware update would change some of the outcomes for the E-M1?

 

[Reflector]

I'm curious if the high EVF refresh rate mode was enabled on the E-M1. The E-M5 gains snappier S-AF from my experience when the high refresh rate option is enabled, but it seems to have issues with AF reliability, whereas leaving the EVF on the normal refresh rate gives me AF (significantly) slower than the GX1's S-AF but the hit rate is practically very high on everything short of extremely dim indoor lighting. Switching to the high refresh rate means that certain low illumination things are a no go.

 

[darrellc]

I have the GH4 and the EM-1, and also find that the S-AF on the GH4 is significantly better (faster and more accurate) than the EM-1. I was surprised at the difference - I didn't expect that. The GH4 AF is quite amazing, and I really like the focus peaking (a joke on the EM-1) and other manual focusing aid implementation. Just wish the GH4 had IBIS! I haven't really used the C-AF on either - I should, but I've been conditioned by previous mirrorless cameras I've owned (except for the Nikon V1) not to expect much, so I haven't really gotten around to trying it.

 

[val]

well Panasonic and Sony should run an advertising campaign where they show their cameras in an actual sporting scenario

 

[pdk42]

But the most important part of the video is Chris' takeaway: all these cameras now match or better DSLRs in their price brackets for action shooting (arguable, but they're competitive at the very least). All mirrorless users win.

Having used a number of Canon DSLRs over the years, I think this is a good point. Sure, a 1D mk IV will be much better at action AF performance than CAF on u43, but the same will not be true of a Rebel/xxxD camera at all. Even the xxD range have limitations. I'm not saying that u43 cameras don't have some way to go before they can be considered good sport//BIF cameras, just that DSLRs are only significantly better once you're very high in the model range.

 

[napilopez]

Having used a number of Canon DSLRs over the years, I think this is a good point. Sure, a 1D mk IV will be much better at action AF performance than CAF on u43, but the same will not be true of a Rebel/xxxD camera at all. Even the xxD range have limitations. I'm not saying that u43 cameras don't have some way to go before they can be considered good sport//BIF cameras, just that DSLRs are only significantly better once you're very high in the model range.

Yep, as I've also said before, I feel people overestimate the performance of most DSLRs for this. There isnt a mythical PDAF advantage by default. They make mistakes too =P

 

[John M Flores]

Yup. Agreed. I'm wondering how the E-M1 would've faired if the reviewer's were using the Oly 12-40 on BOTH the GH4 & E-M1... (as opposed to the Pany 12-35 they used in the review tests). Right?!?

I think they are using the Panny 35-100 in the tests as that situation (and many sports situations) require longer glass. It was good that they discussed differences in usability, as minor annoyances can mean the difference between getting the shot and not. I recently had my Panasonic GH2 at a youth baseball practice and it was comically inadequate. Amazing how much progress has been made in such a short time.

I do wish they had included a Nikon 1 camera in the mix, as it's my mirrorless benchmark.

 

[jurgen]

Interesting test, and good video!

As a current GH4 owner (and former E-M1 owner), my experience more or less (if you'll excuse the turn of phrase) "mirrors" that of the TCSTV guys. The GH4 is ever-so-slightly quicker, and ever-so-slightly better in some situations at locking continuous focus. It's close enough, though, that if I were only a stills shooter, I would be inclined to save the ~$400 and buy the E-M1. I think Olympus has the two best bang-for-buck cameras on the market in the E-M1 and the E-M10, and I don't really think the marginal improvement in S-AF speed of the GH4 over those two models warrants the difference in price.

If you shoot serious video, I think the choice is clear, as you're getting the best of both worlds. If you really only care about still though ... the E-M10/1/A6000 warrants a serious look.

 

[Kilauea]

Yep, as I've also said before, I feel people overestimate the performance of most DSLRs for this. There isnt a mythical PDAF advantage by default. They make mistakes too =P

Its very interesting. I thought mirrorless were further away from SLRs than that. I just thought that while the AF for moving targets was decent, even a recent budget SLR would do considerably better than my E-M5.

 

[gcogger]

Interesting test, and good video!

As a current GH4 owner (and former E-M1 owner), my experience more or less (if you'll excuse the turn of phrase) "mirrors" that of the TCSTV guys. The GH4 is ever-so-slightly quicker, and ever-so-slightly better in some situations at locking continuous focus. It's close enough, though, that if I were only a stills shooter, I would be inclined to save the ~$400 and buy the E-M1. I think Olympus has the two best bang-for-buck cameras on the market in the E-M1 and the E-M10, and I don't really think the marginal improvement in S-AF speed of the GH4 over those two models warrants the difference in price.

If you shoot serious video, I think the choice is clear, as you're getting the best of both worlds. If you really only care about still though ... the E-M10/1/A6000 warrants a serious look.

What's the difference in price between the E-M1 and the GH4? In the UK they're about the same price.

 

[jurgen]

Where I am (Austin, TX, United States), the GH4 runs ~$1800 after tax. The E-M1 can be had new for ~$1350 after tax (GH4 retail $1700 and E-M1 $1250).

 

[napilopez]

Interesting test, and good video!

As a current GH4 owner (and former E-M1 owner), my experience more or less (if you'll excuse the turn of phrase) "mirrors" that of the TCSTV guys. The GH4 is ever-so-slightly quicker, and ever-so-slightly better in some situations at locking continuous focus. It's close enough, though, that if I were only a stills shooter, I would be inclined to save the ~$400 and buy the E-M1. I think Olympus has the two best bang-for-buck cameras on the market in the E-M1 and the E-M10, and I don't really think the marginal improvement in S-AF speed of the GH4 over those two models warrants the difference in price.

If you shoot serious video, I think the choice is clear, as you're getting the best of both worlds. If you really only care about still though ... the E-M10/1/A6000 warrants a serious look.

While I'm inclined to agree with you and my next camera purchase will probably be an E-M1, from what I think I know, the GH4 has some advantages in the flexibility of its AF system that I can see being very useful for an action shooter, and I cant help but always be impressed at how panasonic managed to improve performance so drastically, to arguably the best among mirrorless cameras, while only using CDAF. I shot a wedding today, side by side with an X-T1 and my E-M5 (not really action, but there was a lot of dancing =P), and being able to use the touch screen to set my AF point while looking through the EVF and set custom AF points were both incredibly useful. The former was kind of like manual focus tracking. That said, Olympus' face detection is still the very best out there. Panny's is still good, though.

The X-T1 plus 56mm F1.2 was a bit of a frustrating experience with false positives and slow focus. I ended up reverting to manual focus or just using the 45mm f1.8 on the GH4 or E-M5 most of the time. Shame, because the 56mm is a beautiful lens.

Also, I really had to give kudos to Panasonic's work with battery life. The X-T's battery fizzed out at 601 frames (it was on high performance mode for about 50% of the time), the E-M5's died at 651 frames. Meanwhile, the GH4 showed 2/3rds of the battery left after 1299 frames. It only dropped down to that level at the very end, AND it spent most of its time in C-AF+Tracking or plain C-AF (the others were mainly in S-AF). I'm sure there's more to it, like most of the GH4s shots were shot sequentially rather than intermittently, but there's no question I spent substantially longer shooting the GH4 in a theoretically more power consuming mode(plus some WiFi too), and it still outlasted two other cameras combined. It's not a DSLR, but it sure is closer.

Its very interesting. I thought mirrorless were further away from SLRs than that. I just thought that while the AF for moving targets was decent, even a recent budget SLR would do considerably better than my E-M5.

Well even a budget DSLR is better than the E-M5 for C-AF, almost certainly, but then the E-M5 does way better than a budget DSLR at S-AF, particularly of you care about accuracy.

 

[jurgen]

The X-T1 plus 56mm F1.2 was a bit of a frustrating experience with false positives and slow focus. I ended up reverting to manual focus or just using the 45mm f1.8 on the GH4 or E-M5 most of the time. Shame, because the 56mm is a beautiful lens.

I keep hearing this, and it's too bad, because it seems like a lovely lens otherwise. The non-optical components of Fuji lens performance are what holds me back from buying into the system I think (that, and some reports of poor quality control).

And yes, some of your points are things I failed to consider. The battery life advantage of the GH4 really can't be overlooked - it's nice to know you've got enough juice to cover extra-long walks, or nights out when you forget a spare. Carrying a spare is a trivial thing (especially if you're carrying a bag with extra lenses/accessories), but some days I just like to take off with my GH4 and a prime and not worry about the gear. It's nice to know you've got some cushion with your battery.

I think my overall view on it is that while there's nothing the E-M1 really does better than the GH4 (save IBIS, which I don't find practically all that useful anyway), the advantages of the GH4 aren't worth an extra $400-$500 for a stills-only shooter. If you've got the money, buy the GH4, but if you've got budget constraints - say, if you've only got $2000 to spend - I'd prefer the E-M1 and a couple Oly primes over the GH4 and one of the (still very nice) $200 Sigma A lenses.