TheCameraStoreTV team tested the latest generation mirrorless camera autofocus performance and compared it to the best AF cameras of the world, the Nikon D4s (a $6,500 camera!). According to Chris the Sony A6000 is definitely the best bang for the buck. And along the Panasonic GH4 the performance is quite close to that of the exactly 10 times more expensive D4s!
Cannot wait to see the newer generation FF E-mount cameras having the same (and maybe even improved) af performance!
Sony A6000 store link list:
Sony A6000 at Amazon, Adorama, BHphoto, SonyStore. In Europe at Sony DE, UK, FR, IT, ES, AT, NL, BE, CH, FI, SE, PT, WexUK.
Silver A6000 at Amazon US, BHphoto, WexUK.
Sony A6000 with lens at Amazon, Adorama, BHphoto, SonyStore, Sony Canada, WexUK.
Silver A6000 with lens at Amazon US, BHphoto, WexUK.
Sony 70-200mm FE at Amazon, BHphoto, SonyStore. In Europe at Sony DE, UK, FR, IT, ES, AT, NL, BE, CH, FI, SE, PT
0 Responses
1st blood! ^^
If there is one new ILC FF with A6000 AF system and new video format, expect it will sell like hot cake since it will close the gap for the ‘quite bad’ low light AF test by A6000. cant wait for it!
if only there was A6000 AF on the A7s 😀
Slower sensor readout of larger FF sensors ensure that smaller sensors will always have an advantage assuming process and lens motors can keep up.
Not only that, the smaller sensor DOF is pretty wide, making proper focus way easier
DOF is not the only thing impacting AF speed. A slower lens on a larger sensor while increasing DOF also passes less light to the on-sensor PDAF pixels of mirrorless sensor. This would offset the improvement from DOF by causing more missed focus in lower light vs a faster lens
And the Sony lens in this test has an f4 max aperture too..
The Nikon D4 was still faster than any of the 4/3 sensor cameras and accomplishing the task with a much more critical (smaller) depth of focus. Most of the advantage with the tiny sensors is because of the depth of focus difference, comparing all three size sensors at the same F Stop is not a valid test.
No matter how fast the micro 4/3 cameras focus, you still end up with a inferior quality image compared to APS-C or Full Frame.
More size envy from the micro crowd.
I don’t think so. There is a speed from which going faster would mean nothing.
@phenoy You can simply stop down on a bigger sensor and get the wider DOF. There is no advantage to the smaller sensor here.
A DSLR/T will always acquire focus at the lens’ largest aperture.
Readout speed is primarily an issue of the number of pixels that have to be read, fed through the ADC and processed than the physical size of the sensor. The D4S manages the highest frame rates despite being FF but the fact it has “only” 16MP is surely material.
However, there is one curious issue in that we might expect to the forthcoming A7S to have a high frame rate in view of it’s low resolution, yet it’s limited to 5FPS. I wonder if that’s simply because of a physical shutter limitation, which it probably shares with the A7 and A7S. There’s probably not room in the very compact body for the more robust shutter which would be required to operate at higher frame rates (and here smaller sensors do have a considerable advantage).
Readout speed is dependent on the size of the sensor. That is why the Nikon 1 sensor can readout at 80fps while the APS-C can only do 15 fps and m43 around 25 fps. Processor speed impacts how fast the # of MP get processed.
Yes, and the EPIC Red Dragon can do 85fps with a much large sensor. That’s what happens when sensors are optimised for video. It’s also important to check what is actually implemented rather than just the sensor specs. As it happens, whilst the Nikon 1’s Aptina sensor is rated at up to 80fps, as implemented in the Nikon 1 V3, it will have a maximum full resolution frame rate of 20fps using an electronic shutter. For HD video it offers the same maximum frame rate (120fps) at the same resolution (720p) as does the Sony A7S
We know that the A7S will offer only 5fps (with a mechanical shutter), but there are rumours that it might also support electronic shutter modes in the manner of the Nikon 1. We also don’t know what the full technical capability of the Sony’s sensor is, we only know what the camera will offer. In the real world, things are limited by issues such as heat (cooling is a bigger issue on large sensors), power consumption and data processing rates.
There is one sense in that FF sensors might require more read time in that, if the increased DR is to be taken advantage of, it demands more precision in the ADC phases along with higher data rates due to the increased bit rate.
Yes, the dragon does take it to a new level but it will likely be a while before the tech from this $30k video camera can be feasibly done at a low cost. I suspect much of it is hardware related so that may take while. And the same technology could be applied to all sensor sizes so this doesn’t change the inherent sensor readout speed advantage of smaller sensors unless it only gets implemented for FF.
The limitation of 20fps for the Nikon 1 electronic shutter is likely limited by processor as it cannot process the AF information at faster frame rates. With just AF for the first frame the Nikon 1 can support 60fps at full resolution. Here are the specs for the sensor that show the readout speed.
http://www.aptina.com/products/image_sensors/ar1411hs/
We will know more about the A7S in time but I have not seen any new technology mentioned that would suggest a big leap in AF over the A7.
I’m having a problem viewing my A7 image thumbnails on windows 8.1. is there any solution to this? It’s perfectly showing up in lightroom when im importing but it can’t be viewed on windows photo viewer.
Can’t give a solution without having all info needed but..
You can right click on a thumbnail then select “open with” then choose another program to open the images other than windows photo viewer.
I didn’t ask if the images are in RAW format or JPG !
If you are not an expert in computer stuff then just download Picasa and set it to be the default photo viewer
hi and thank you for the reply. the A99’s .ARW and my other sony camera using the same .ARW files is showing up as thumbnail on windows photo viewer except A7’s .ARW
I’m not having any problem with my other cameras so I’m thinking my A7 has a problem? I tried changing cards and formatted them and tested them and im still getting the same result. no thumbnail showing up on photo viewer when its from A7.
That means the photo viewer you are using does not support RAW format to show thumbnail, try another software.
I’m using Picasa and it is working fine.
Try searching google for Picasa
picasa only reads out the preview jpeg thats included within the raw file 🙁
otherwise it would be nice and fast solution.
if someone knows a workaround to work with raw-files directly in picasa iw ould be very thankful.
greetings michael
I found something that might help you, before you download Picasa.
Try this codec from Microsoft it should work.
http://www.microsoft.com/en-us/download/details.aspx?id=26829
Thank you so much for the help bro! I just downloaded the raw driver directly form sony esupport for windows 8.1. I blame Microsoft for not updating their raw codec. it has been 6 months and sony a7 is still not getting any support from Microsoft. grr!
man, it must be that xbox vs ps4 thing and I hope us sony shooters will not be part of the casualties of this on goin war lol
Try downloading and installing the latest Microsoft Camera Codec Pack at this link: http://www.microsoft.com/en-us/download/details.aspx?id=26829
Even though it seems to be the same kind of RAW format, the camera model identifier is different, and it’s enough to break all existing RAW readers. It’s a shame that we have to go through this with every single new camera release. There are better viewers out there than Microsoft’s.
If Lightroom can see them. It is not a camera issue. I don’t use Microsoft, but sounds like Windows 8.1 doesn’t have a plugin for it yet. That would be what I would check
What Chris is saying panasonic GH4 is very close to Nikon’s flag ship. Not a6000. Gh4 is the best for the buc.
A6000 was below oly e-m1 for the single shooting focus but then above with the contionious focus test.
I as a oly e-m1 owner, acknowledge a6000’s advantages. And expected Fuji doing better than all of us, but GH4 is clearly a whole different animal.
Did you even watch it? He says they are all very close to each other and they aren’t far behind the Nikon. He clearly states why he is naming the GH4 the winner above the other 3, and why he’s calling the A6000 ‘the best bang for the buck’.
I will give you a hint: it has something to do with the price.
I will look forward to seeing how the A77II AF goes head to head with Canon and Nikon
I wish Sony could scale down the a7S sensor to APS-C and produce an a7000s, keep it 12MP for 16:9 4K full pixel readout without cropping! Deploy the a6000 AF system – larger pixel size better AF sensitivity and cleaner pictures! Make a vertical grip(with mSATA SSD slot) for 4K recording in 10bit 422 and SLOG2, it will kill GH-4 in no time!
m43 will always have an inherent AF speed advantage over APS-C due to the faster readout of a smaller sensor. This assumes the processor and lens motor are fast enough to process the information.
Low light AF is also impacted by how the information is processed. Panasonic still has a large lead in this area, even over Olympus with the same sensor. Here is a low light comparison between the A6000, GH4 and X-T1.
http://www.youtube.com/watch?v=TB1sGrrWZEI
Thx for the info. The faster readout is about the amount of data or actual real estate? I believe it’s the former, so readout of 12MP data from an APS-C sensor should be faster than 16 MP from m43? But the heavier lens element of the APS-C lens may offset the advantage…
Low light AF would be impacted by the algorithm and more importantly the number of photons reaching the sensor for the algorithm to process, with much larger size than A6000, I trust that AF response would be much improved – maybe not up to the level of a dedicated AF module like D4s for A-B focus, but it could beat DSLR AF in terms of object recognition & tracking!
Real Estate is the main impact on sensor readout. You also need a faster processor if readout is faster. You won’t need as fast a processor if you have fewer MP but sensor readout doesn’t change.
I expect the A6000 to still have significantly faster AF than the A7S. The main advantage of going with 12mp with the A7S is to reduce the amount of scaling required for video which has higher frame rates. Less scaling also reduces the amount of moire and Anti-aliasing.
You mean GX7, not GH4
Yes, although they are basically the same sensor with the GH4 having more advanced processor for video and DFD.
You are basically asking Sony to skill their first FF E-mount cameras, they are not going to do it. I wished there was an A7000 as well, it would be a great camera, but I’m pretty sure that the omission of it is due to the FF counterpart, therefore, it won’t come so soon.
But you are expecting too much of it anyway, it wouldn’t have all those things, it would not only hurt the FF but also their video cameras as well. That’s not how Sony usually does it.
Yes, I know I’m asking too much!
But remember Sony lost the battle of Liveview/DSLR video on a700/a900 almost led to the extinction of the A-mount!
Technically there’s nothing new, Sony owns all these technologies.
If Sony don’t do it, I’m sure Canon/Panasonic/Samsung can do it simply becoz they do not have a ‘F5’ grade Cine Camera.
Canon? That’s nonsense. Canon hasn’t been doing anything new in the DSLR front in quite a while and they do have their EOS-C C100, C300, C500, etc. which guarantees ZERO innovation in video for DSLR. Samsung is far from doing anything at all, all I’ve heard from Smasung is that their goal is to be #1 but they haven’t been able to execute that so far. Panasonic also have some more PRO video cameras but anyway, I wasn’t talking about F5 or F55, but more like FS camera line up.
Yup, Canon is all about marketing not about innovation! But, Magic Lantern RAW do have an impact on DSLR video.
Samsung got a whole series of 4K TV, making 4K video cam can cross-sell their products, but it will take a few years to catch up.
Panasonic can develop a video grip w/SSD for GH-4, without going thru the external HDMI port the bitrate is almost unlimited!
It’s just like VW putting a 400HP engine into the new Golf R.
Golf R is still a Golf, It won’t hurt the sales of Lambo or R8s!
Anyone have any experience with similar continuous and single AF tests in low light? Any opinions on this?
Look on Youtube, there are some tests.
Yeah, I’m not interested in someone taking pics of a random wall, and then a tree, and then a random wall. This test at least tested a moving object moving towards the camera at a good speed. There doesn’t seem to be an AF test for that in low light.
Shooting at f4 on all cameras gives advantage to smaller sensors on Panasonic and Olympus. Effectively this is f8 DOF witch makes it easier for AF, and if there is small error is not going to show up us much as on APS-C size sensor
But you can’t take away that as an advantage, it’s part of the trade offs of having a smaller sensor. You will never see someone saying that you have to shoot f/8 against Micro4/3 f/4 because FF has better light sensitivity and that this would be more fair, it’s part of FF’s advantages to have a better clean high ISO. In a shootout type of test, you won’t change the parameters just so you can match the differences, it would be a complete contradiction.
FF only has that advantage if you don’t bother to (or, because there might be none available for a certain FF lens, are not able to) select an accordingly faster lens for the FT sensor. And then there’s DoF – if I have to carefully choose the aperture, it’s most probably because I need a specific depth of field, and then it’s f/4 for FT and f/8 for FF again, too.
Yeah but this wasn’t a DOF shootout, just AF.
If you are talking about an APS-C size sensor, then surely it isn’t f8.
Simply stop down your FF lens and you will make it easier for the AF. What you call an advantage for MFT is simply lack of an advantage for FF.
Even an APS-C size sensor gets effected by this so that will not have the DOF of a f4 lens either. It will be f5.6. So the difference is only one stop compared to the a6000
Why exactly is the FF E-mount AF so slow? It shouldn’t be.
If you configure the A7 correctly, it isn’t slow at all. On the same level as a D600 or 6D. A pity that in most configurations it is mixing PDAF with CDAF which makes it slow.
For the slow CDAF on full frame:
– full frame sensors take longer to read out
– thinner DOF requires more accurate focusing
– more glass to move
What is the ‘correct configuration’?
You need to make sure only PDAF points are used and not a combination of PDAF and CDAF points, and of course that PDAF is enabled. On top of that you have to keep the aperture below f/8.
You can turn on a menu option to mark the area with PDAF points. It’s roughly the APS-C portion of the frame. Set to a single point inside that area, keep aperture below f/8, and you’re golden.
I found the AF to be quite fast then. I can only directly compare to my D7000 and 6D, and it seems to be about the same.
AF speed in mirrorless cameras is more dependent on sensor. Larger sensors will always have slower sensor readout which impact AF speed. DSLRs don’t have this issue as the PDAF module is separate from the image sensor so they are not reliant on sensor readout speed for AF.
So Nikon should always have faster faster AF than m43 and m43 faster than APS-C, etc. There will also be some variations dependent on sensor technology, processor speed/algorithms and lens motor. That is why within each sensor size we see variations in AF speed between manufacturers. But a Sony FF mirrorless will always have slower AF than a Sony APSC mirrorless unless Sony reserves some significant new tech for just FF that can overcome the sensor size disadvantage.
By*, not bt.
Any camera with a $1500 lens attached to it better be good.
Would like to see it with the cheaper 55-200.
Panasonic GX7, X-T1 and Sony A6000 in low light conditions.
http://www.youtube.com/watch?v=TB1sGrrWZEI