Image source: Lesnumeriques
If the NEX-FF will be launched in September one of the big question will be: How good will third party manual lenses work on that camera? So let’s take a look how good (or bad) the current VG900 sensor performs with these lenses. And let’s hope the new NEX-FF will be even more optimized for third party lens solution like the [shoplink 19605 ebay]Leica M Lenses[/shoplink], [shoplink 19923 ebay]Voigtlander M lenses[/shoplink], [shoplink 16763 ebay]Canon FD lenses[/shoplink] and[shoplink 19360 ebay]Zeiss ZM lenses[/shoplink].
I asked all SAR readers to send me their tests to check out the results. This is the first test done by DigitalVersus. It is the first website that compares the new Leica M $8000 camera (here at Amazon) versus the first NEX Full Frame camera (The VG900). They used the same [shoplink 20202 ebay]Leica 50mm Summilux-M lens (here on eBay)[/shoplink] on both cameras. And both cameras have a Full Frame 24 Megapixel CMOS sensor. So what are the results?
Please visit the website to see all image comparisons. Here you can see one of them (Click on it to enlarge):
In short the Leica M is sharper and has richer colors. This may be possible because of the lack of the anti-alias filter. The VG900 is a video camcorder and to have an Anti Aliasing filter makes a lot of sense (otherwise you end up having video clips with lots of moire).
What do you think of the results? Would the [shoplink 20202 ebay]Leica 50mm Summilux-M lens[/shoplink] work fine on the NEX-FF?
60 Responses
Sony still get worse results from their own sensors than Nikon. Making a camera isn’t just building a sensor and programming software features. Sony make some great cameras but it’d be foolish to assume that competing with Leica (Leica!) is simply a matter of compiling a spec list.
Sony fans do themselves a great disservice when they dismiss Lecia as a massive scam. Those lenses are world renowned as the best ever made for a reason. There’s a level of precision and care in assembling a great camera whose results you can see in the files. Think about it. Two identical cameras, one with a mount aligned a fraction of a millimetre off. Which camera’s going to produce sharper files?
I hope Sony give Leica, Nikon and Canon a real run for their money. However there’s nothing inevitable about it. They’ll have to work hard to get there.
We often see these kind of statements but:
– low light performance and DR are within 0.25 EV (without the mirror)@ 14 bit, which is within measurement error
– Sony cameras are often less prone to moire
– all Sony cameras have much better colour response/separation than their Nikon counterparts carrying the same sensor (see DXO measurements confirming this too).
You can’t remove the mirror in an SLT so a ‘without the mirror’ measurement is useless. Less light is less light.
I’m not arguing that Sony are bad. I am stating that they’re less knowledgable than the big two and there’s no shame in that. It’s odd to see Sony owners treating them as equals.
Canon are the #1 camera seller in the world (if you exclude phones) and Nikon are the only biggest cameras-only company. Sony have made huge strides but there’s a way to go.
Also using Nikon as an example. Nikon’s best low-light camera carries their own chip design. It’s foolish to pretend that Nikon are simply relying on Sony sensors to produce excellence. They can clearly do it on their own.
Similarly Leica are much more than sensor-packagers.
The only camera Sony’s produced which could claim to be close to Leica quality or compactness is the RX1. That camera has a large fixed lens slower than Leica’s best and still costs more than all but the two top tiers of DSLR.
We have been told the FF NEX will be more expensive still and that’s accounting for Sony being able to use its own sensors for ‘free’ and not delivering a camera milled from brass with an incredibly complex and expensive manual rangefinder.
I.e. With all the resources at their disposal they’re coming in at about 50% of Leica prices for a camera with a much simpler design (no mechanically linked rangefinder).
If Sony wanted to include a range finder the prices would be even closer.
Sony have a great chance to upset the apple-cart but they’re clearly having to work very hard at it and whatever they release isn’t a Leica substitute, it’s an alternative. Those who love real rangefinders really love them. Fuji’s competing with an electronic rangefinder or sorts, Sony isn’t yet.
Dismissing a companies like Leica, Nikon and Canon just seems foolish and overconfident.
FYI – I own Sony, Leica, Nikon, Canon and Hasselblad cameras. I don’t care about brands but I could happily tell you why I shoot what I do.
When did Nikon became a camera only company ?
Technically you could say ‘optics only’ though Nikon steppers and from a separate company with a shared name. Much as Leica microscopes.
Nikon make no printers, TV’s or other general consumer goods. Their sole focus is lens design and imaging. They make their own glass (actual glass, not lenses).
So much nonsense, it’s hard to figure out where to begin, but here goes.
First, you were talking about sensors and what they are able to squueze out of those based on their knowledge. An argument that isn’t about SLT mirrors.
Second, you can take out the mirror, it takes 10 seconds.
Third, the RX1 and RX1R do not have a mirror and show the bare performance of the sensor in a Sony body is exactly as I described in my previous post.
Fourth, the the A5xx series already showed performance within margin of measurement error compared to Nikon bodies regarding SNR (and better regarding colour separation) and the statement has been true ever since.
Your “less knowledgeable” comment is one of lots of fluff and hot air. The big boys have more knowledge in some areas, less in others.
>>First, you were talking about sensors and what they are able to squueze out of those based on their knowledge. An argument that isn’t about SLT mirrors.
You mentioned mirrors not me.
>>Second, you can take out the mirror, it takes 10 seconds.
Well there goes your credibility. You’re part of the Sony buying camera public who buys a pellicle mirror body and then removes its key feature? Great. A brilliant approach. Do you then put it in an SLR as an additional feature?
>>Third, the RX1 and RX1R do not have a mirror and show the bare performance of the sensor in a Sony body is exactly as I described in my previous post.
Not true. The RX1 is a lens with a camera glued to it. It performs very well but you can’t extrapolate a fixed lens systems performance to a NEX FF. The NEX FF will perform worse given the same size output. Even with an equivalent Zeiss lens (which will be a $2,000 item at retail).
>>Fourth, the the A5xx series already showed performance within margin of measurement error compared to Nikon bodies regarding SNR (and better regarding colour separation) and the statement has been true ever since.
Who cares? No one sane would say that Sony cameras outperform Nikon bodies based on a comparison of prosumer devices. Everyone with eyes and a brain knows the best DSLR on the market has been the D800 since launch and the best low-light performer is the D4 (the 1DX is close). Similarly the D600 outperforms the A99 for IQ – even given that one’s entry level and the other’s flagship.
What next? A comparison of pojnt and shoots?
>>Your “less knowledgeable” comment is one of lots of fluff and hot air. The big boys have more knowledge in some areas, less in others.
Your final mistake. Sony is a ‘Big Boy’. Much, much bigger than Nikon. Massively bigger than Leica. They’re the overdog not some struggling startup.
If you were smarter you’d have read my praise of Sony’s bodies in terms of features and electronics. That’s the easy bit. The harder bit is assembling mechanical devices to consistent tolerances and saying ‘No’ to extraneous features than professionals and serious shooters don’t want.
Limitations are often an asset.
>> You mentioned mirrors not me.
Typical fallacy. You’re including mirrors into the discussion if you’re comparing SLT’s and DSLR’s, when you were initially talking about *SENSOR* performance. I specifically said I left the mirrors out of the discussion, because you made a statement about sensors. Try again.
>> Well there goes your credibility. You’re part of the Sony buying camera public who buys a pellicle mirror body and then removes its key feature? Great. A brilliant approach. Do you then put it in an SLR as an additional feature?
The discussion was about sensor performance, based on their respective knowledge. So why pollute your comparison by including mirrors to begin with? Your grasping at straws left you with zero credibility from the start. Your general statement about Sony *sensor* performance in Sony cameras is suddenly limited to SLT’s vs Nikon DSLR’s. Quelle surpise. Try again.
>> Not true. The RX1 is a lens with a camera glued to it. It performs very well but you can’t extrapolate a fixed lens systems performance to a NEX FF. The NEX FF will perform worse given the same size output. Even with an equivalent Zeiss lens (which will be a $2,000 item at retail).
I replied to your nonsense about sensor perfomance compared to Nikon and again you’re trying to bail out by adding lenses do a discussion about sensors in Sony and Nikon bodies. The RX1 sensor performs within measurement error of the D600 sensor regarding SNR and DR and better in terms of colour separation. Fact. Try again.
>>Who cares? No one sane would say that Sony cameras outperform Nikon bodies based on a comparison of prosumer devices. Everyone with eyes and a brain knows the best DSLR on the market has been the D800 since launch and the best low-light performer is the D4 (the 1DX is close). Similarly the D600 outperforms the A99 for IQ – even given that one’s entry level and the other’s flagship.
Ah, your arguments dried up when you started, so you end up with the expected: who cares?
Apparently you do since you brought up the argument through a false statement to begin with. But let’s point out another fallacy as well, I didn’t make a general statement that Sony makes their sensors perform better than Nikon, as you seem to suggest I did. I said their sensors perform within margin of measurement error and that’s supported by visual evidence in countless controlled comparisons.
Only the SLT’s have a half a stop handicap that has been discussed ad nauseam and that has zero to do with what these companies are capable of with regards to sensor performance.
Try again.
>>Your final mistake. Sony is a ‘Big Boy’. Much, much bigger than Nikon. Massively bigger than Leica. They’re the overdog not some struggling startup.
Ah, one more fallacy. Big boys obviously referred to market share, specifically regarding large sensor cameras which were/are discussed here.
If you were smarter, you wouldn’t make false and hollow statements and then try to defend them with silly fallacies and cute but futile attempts at dodging.
But do try again.
Oh and in case you memory is failing on you, you brought up the “big two” yourself, so my reference was obviously to the two companies you tried to describe. The mistake is as usual, on you.
“With all the resources at their disposal they’re coming in at about 50% of Leica prices for a camera with a much simpler design (no mechanically linked rangefinder).”
So, the mechanically linked rangefinder is 50% of the price? That is what you are trying to say?
And, by “simpler”, you mean adding stabilization and autofocus?
Stabilization for Sony is currently in-lens so adds nothing to the cost of the body. If it does come it’ll be Olympus technology so the cost is implementation – not R&D.
The rangefinder is probably 30% of the price. If you have an idea of how the lens and rangefinder are linked – and how precise that linkage must be while remaining adjustable – you’d know. Leica M’s are also made in Germany and the labour costs are much, much higher.
Autofocus isn’t expensive to implement at all. It’s electronic and every $100 point and shoot includes it at a profit. There is no cheap camera with a mechanical (or electronic) rangefinder.
I think the Sony camera could be fantastic. It’s just really boring to constantly read about the wild pricing and limited features of Leica. The limitations are there on purpose, that’s the point. The price is easy to justify. There’s profit for sure but they’re not selling cameras which cost as much as the A99 and RX1 for massive markups. Leica’s cost more for a host of visible reasons.
@ mrSkelter:
So the Olympus 5 axis IBIS is able to drive a four times larger sensor without any additional R&D? Where did you get that information from?
Sony had to completely redevelop the Minolta APS-C IBIS for the FF A900, according to their own words.
Assumptions, they’re not helping you.
MrSkelter, let me get this straight, the SOny A99 and the RX1 are too expensive, but Leica’s cameras aren’t? That’s new.
“The rangefinder is probably 30% of the price.”
Nice try. No offense, but you should back this number up.
“Stabilization for Sony is currently in-lens so adds nothing to the cost of the body.”
You did mention the A99. Btw, making a FF sensor float would certainly require some R&D. And anyway, the system will cost. Probably around 30% of the price 🙂
“Leica M’s are also made in Germany and the labour costs are much, much higher.”
Too bad for them, I guess.
“Autofocus isn’t expensive to implement at all.”
Sure, developing on-sensor PDAF costs nothing.
“There is no cheap camera with a mechanical (or electronic) rangefinder.”
There isn’t any actually, except Leica. Maybe there’s another reason besides cost?
“It’s just really boring to constantly read about the wild pricing and limited features of Leica. The limitations are there on purpose, that’s the point.”
I have no problem with limitations being there on purpose, I just don’t see one should pay more for that.
“Leica’s cost more for a host of visible reasons.”
No offense, again, but you provided only two half-baked ones – costly rangefinder mechanism (without any idea as to the real cost) and that they are made in Germany.
Nope. Sony uses denser color filters than Nikon and Pentax which pass a little less light, so there’s a slight trade off of noise for better color. It’s been that way since Sony started making DSLRs.
You may have a point but I’ve never heard anyone serious complain that Nikon bodies don’t capture sufficient color information. I’ve heard every reviewer and user complain Sony bodies don’t capture enough light.
It’s a valid approach but I’ll bet Sony change their filters in Nikon and Canon’s direction before they move towards Sony.
All indicative of a company finding its feet.
lots of people discount Leica as scams.
while i agree with you that Leica lenses are good in terms of image quality, it is relatively moot when using a manual focus lenses, as focus is dependent on the user. in terms of autofocus, the contrast detection compensates for the offset in flange focal distances (as long as its small and within limits), this would allow Sony to reduce tolerances and therefore reduce cost (in general, not that im claiming Sony has very slack tolerances)
in terms of competition leica would probably be pretty close (as FF cameras with shortish flange comapred to DSLRS) and Fuji, and both of which are pretty good with customer support, I hope sony would pick up in this area but i don’t think it’d happen overnight.
We know Nikon uses thinner color filter to get better high ISO result, favorable to DxO testing method. Every company has different approach, 1 approach is not necessarily worse, just different characteristic. And this particular FF NEX is known for its soft picture.
I’d say that some of their lenses are worth the praise, but most of their bodies don’t justify their price.
I could get much cheaper R glass (I’d avoid the Jap made ones) to go with the FF NEX instead of new M glass, whose performance is pretty much the same, and lacks options like the 70-180/2.8.
My guess is that the Summilux 50 might work fine on the NEX FF, but, even if it does, it is not enough. Question is, would the 28 Summicron, or, better, the 24 Elmar work fine on the NEX FF?
The NEX FF will cost some 3000$. For that sort of money, if you buy it with the intent of using RF lenses, you have got to be able to go wider than 50mm.
My personal guess: Sony are not interested in making any special provision for it to work well with third-party lenses. They will have the only interchangeable-lens compact FF camera. That is a game-changing feature. When you have that, you can hang on to it your own lenses (Sony, Sony-Zeiss, Zeiss), and not make other people’s (Leica’s) life’s easier by letting them sell lenses on the back of your R&D.
So, if RF wides work well on the NEX FF, it will be because Sony couldn’t avoid it IMHO.
Greetings Philber,
I believe that on APS-C market, Sony wanted to surpass competitors to make their cameras to work with all of possible glasses. The message is clear, if You have already a lenses and You would like to choice our product, You would have it easier (and cheaper) to start 🙂 I believe that Sony don’t want to compete with Leica about most situated customers, not yet. First FF-Nex would be entry level in this class and it would be dedicated for people who may collect money for more expensive body but they wouldn’t be able to afford for 2-3 lenses, each of them has a price at least as a camera body. I hope I am right 🙂
Best regards, WZ.
Improving compatibility with symmetrical type lenses (which is really what’s going on here because most wide rangefinder lenses use this type of design which results in steep light angles for wide lenses) will be a big help to Sony with their own lenses. Remember Zeiss makes/designs most of the optics for Sony’s high end lenses and right now they are rumored to be making ALL the lenses for this new full-frame camera.
Guess what? Zeiss makes legendary symmetrical wide angle lenses. If Sony can figure out this problem they can tap into Zeiss’s renowned biogon designs. I think that’s incentive and it would help with third party lenses like the Elmar and Summicron as well.
With respect, I do not agree. Zeiss Distagons offer better performance than Biogons with short registry systems such as NEX. To wit: the ZM Distagon 18mm f:4.0 functions perfectly on the nEX 5N:5R/6 (no corner smearing, no colour shift), whereas the less-wide ZM25mm Biogon f:2.8.
That’s because the current NEX models aren’t optimized for the use of such lenses. Try these on film rangefinders and you’d see.
Or the Leica M, which I trust still doesn’t support such lens as good as film but is already a big step forward.
The first (rumored) lens will be made by ZEISS, a 3rd party lens maker. So I guess the support for 3rd party is there.
So last year’s Sony is almost as good as this year’s Leica. I wonder how this year’s Sony will be.
i’m not sure how useful a comparison is between a video camera (optimised for video) and a stills camera (optimised for stills).
VG900 is doing better photo than video . 24 millions pixels ? For video ? For what ? Full frame sensor , for video ? For what ? VG900 should’ve never called camcorder !!!
The lack of a anti aliasing filter is a big deal when it comes to sharp image quality. I have found that the 18 mega-pixel M9 ran circles around the 24 mp Nikon when it came to 3 foot + enlargements. I would expect that if the filter is omitted, the Sony will give Leica a run for the money as the R solution and for 35 mm and longer rangefinder lenses.
Concerning the wide end, probably not, as many problems such as image corner darkening etc, are repaired with in camera lens specific software.
If it has a built in viewfinder, maybe not a bad backup, or good long mm lens camera.
Because of the electronic pins and all, it seems like the nex mount is actually smaller than the leica mount in terms of how big of a rear element you can stick inside. Any specs available on this issue?
I haven’t read any rumors at all regarding if this coming Sony FF mirrorless camera is going to use E-mount or a larger mount. The problem with E-mount is that it will make the FF lenses larger because it is a bit tight in this regard.
In case Sony will introduce a new and a bit larger mount for its FF mirrorless camera, the FF lenses can be made smaller and lighter, but some current NEX users would of course scream (not me, though). And it will come with a larger mount, I think we can be pretty sure that they would include the name “NEX” in it. NEX means “New E-mount eXperience”, and it would not be a particularily well choosen designation for a series of cameras not using the E-mount, if that will be the case … 🙂
.
Still no possibility to edit posts!
I wrote:
“And it will come with a larger mount, I think we can be pretty sure that they would include the name “NEX” in it. NEX means “New E-mount eXperience”, …”
But it should of course have been:
And if it will come with a larger mount, I think we can be pretty sure that they would NOT include the name “NEX” in it. NEX means “New E-mount eXperience”, …
.
http://en.wikipedia.org/wiki/Lens_mount
Leica M-mount is 44mm the E-mount is 46.1mm so the E-Mount is larger. Whether the contacts get in the way I don’t know but Leica also uses a sensor above the main sensor so they’ve got a similar issue. I doubt there is a real-world difference in the largest element. In fact the SLR Magic 35mm T/0.95 lens, despite being M-mount, is actually too large to mount in a Leica but will mount on an E-mount camera. It only covers APS-H so it wouldn’t really be useful on the FF NEX but I think it suggests/proves that there’s more leeway in the size of the final element on the E-Mount than the M-Mount.
The M-mount looks bigger because (in this picture) it has a thinner rim around the mount. Optical illusion.
@GunnarK: I think you’re missing the point. A wider mount might help but it isn’t any more of an impediment for the E-mount than the M-mount. Also, while Sony does have a wider mirrorless mount (the FZ-mount look it up) I doubt they will use that for the NEX FF because it is too big and there is only one VERY massive lens designed for it which (to my knowledge) only covers Super 35mm (which is about the size of APS-C).
So despite all the brewhaha there really is *no problem* with the E-mount and full-frame sensors. Sony does not need to introduce another mount. The VG900 proves that. Are there challenges? Sure. Issues with a limited (albeit desirable) subset of lenses (the RF WA)? Yes. Are these insurmountable or caused by the diameter of the mount? NO.
If the mount is wider the lenses will be bigger. If the mount is deeper SOME lenses might be smaller but the final package will be the same size. I’d rather have the flexibility of the shallower mount.
Nicely explained Joel. Maybe you are right that the e-mount has room for a larger rear element than the m-mount, but I had read somewhere else from someone that appeared knowledgable that it was actually less when you took into account the thicker rim and the electronic pins.
I’ve also read whatever there is to be read about the fz-mount which looks interesting. I heard it has an 18mm flange but there is no info I could find about the diameter of the FZ-mount. Did you find that info or have you seen the mount? It doesn’t look too much wider from the pictures but it’s hard to tell. I was wondering if maybe that would be the real FF e-mount they would use since it would seem that it would be easy to adapt it to fit a ff sensor in as well. Perhaps a large mount like that could mean that the lens goes inside the body close to the sensor just as it does in the RX-1.
I do have to wonder about the possibility of Sony introducing a new mount that is shallower than the e-mount, maybe 12mm instead of 18mm, but wider, where they can have one adapter for e-mount lenses and another adapter for a-mount lenses.
It just seems hard to believe they’ll be successful in transforming the current e-mount into a full frame mount that would allow them to build lenses that are small but significantly improved over APS-C lenses – let alone add IBIS which seems bogus considering i’ve read from that there’s no place for a FF sensor to move at all in that mount.
I can’t say for certain but given what SLR Magic has reported with their 35mm T/0.95 lens I am now skeptical about the persistent rumor of the pins and such getting in the way. I am not certain but skeptical. If Sony has this problem, Leica has it worse.
I have only spent a little bit of time with an F5 which uses the FZ mount. Didn’t get to swap lenses around so I can’t say too much there other than the native FZ Zoom is shockingly large!* But then I haven’t seen many other cinema zooms and from what I have seen in cinema primes the size may be very normal. Real cinema primes (no disservice to the Samyang variants but those aren’t really the same thing) are massive.
So while I don’t know the diameter of the FZ mount I am confident it is much larger than the E-mount. Perhaps they could create an adapter like the M42-K that Pentax built when they switched to a bayonet mount. Personally, I think the FZ mount would make sense for the SLR shaped bodies but would be too large for the NEX/rangefinder styled bodies. I explain my reasonings better on my blog: http://camerablog.joelrichards.com/2013/08/ilc-3000-a-sony-mirrorless-full-frame-camera-or-the-super-35mm-answer-to-the-canon-1dc/
I am also skeptical of IBIS on the E-mount; however, from what I have read on shorter focal lengths there is less lateral movement and more rotation which is why the OMD system is so much more effective than past IBIS systems. So IF they can fit a similar system into the NEX body with a full-frame sensor then despite having less lateral range than the Alpha IBIS it might actually be comparable in the real world for focal lengths most likely to be used on a small mirrorless camera (ie less than 100mm). Luckily with an E-mount you can put on stabilized long focal length lenses from other systems if you want to go that route and get better lateral stabilization when it matters more.
*in diameter.
As far as this comparison goes, yes, the Leica is sharper. I believe largely because the VG900 has an aggressive (although ineffective) AA filter. However, the VG900 is still pretty decent especially in the center where it is more a matter of color and exposure differences than absolute sharpness (to my eye).
The two cameras certainly handle exposure and color differently. I wonder if that is partly due to the fact that the raws were processed in Lightroom which is going to have optimized profiles for the Leica but not necessarily for the VG900.
The Leica also did much better in high ISO which is the strongest indicator that something is amiss either with the testing or the quirky VG900 itself. Even IF there are issues with lenses on the full-frame NEX it shouldn’t affect the noise from high-ISO levels on the middle of the sensor! According to DxO Mark the RX1, which shares the same basic sensor but not filter package as the VG900, is slightly better than the M240 at high ISO which is certainly not true here.
I think what this shows is that a full-frame NEX stills camera is possible but it will need to be optimized very differently than the VG900 to really shine. For all their faults I believe Sony can do this much. Also, this comparison really, really makes me hope there is no AA filter.
+1 i would also like to see FF NEX without AA filter. sharpness isn’t everything to what makes a good picture, but it would be nice thing to have 😀
Yup. Yet another advantage to BIS and stacked, is that you can really push the resolution without sacrificing the well fillfactor.
The Sony image looks to be underexposed by about 2/3 of a stop compared to the Leica. I just ran the image through PS and upped the exposure by that amount and the Sony version then looks just as good as the Leica.
Not only is it a little underexposed compared to the Leica, it also has less contrast. Giving it some contrast will help with the sharpness and color. Also Sony, just like Minolta tends to use pretty heavy AA filters on their sensors.
If SONY release the FF-NEX with a thick AA filter that introduces refraction and smearing like in the corners similar to the VG900 with the best wide-angle lenses available they will probably lose half their sales.
Also color cross-talk issues that plague the NEX-7 with even the native 16/2.8 and 10-18 and required a firmware patch to edit RAWs is unacceptable and needs attention at the sensor level (better pixel-well baffles or some combination of backside illumination/organic doping of the photodiodes).
Get this right, and the FF-NEX will immediately win Leica enthusiasts, legacy lens users, landscape shooters, architecture shooters needing a back for T/S and professional Medium-Format shooters requiring a good digital-back for product photog etc, And all the enthusiasts simply wanting a lightweight portable 35mm digital camera.
To elaborate, the reason why backside illumination and organic doping help is because BIL moves the photosensitive area closer to the top of the sensor plane and organics being more sensitive can be reduced in thickness. The consequence is that shallow/high incidence light rays have less chance of penetrating into adjacent well and being registered as a different color component by the electronics/CFA. The same approach also helps with vinegetting.
True(according to my logic, of course). That’s why I’m hoping Sony to come out with 200+mp APSC/FF BSI sensor that produce something like 24mp binned picture.
Comparing two great devices for small differences is so wearisome, prosaic, and insipid.
(However, Howie got it right. Exposure is the main dif.)
In spite of the anticipation the VG900 was correctly recognized to be a disaster for use with fullframe lenses.
Here it is with a specialist landscape wide-angle lens – the CV15 that has edge-to-edge sharpness from wide-open,
http://www.flickr.com/photos/dakw23/8180002529/sizes/o/in/set-72157631993239568/
It’s not even particularly good in the center.
The VG900 works great with all of my full frame PK (Pentax ) mount lenses like my 15mm F2.8 Sigma, 24mm F1.4 Rokinon etc. It’s amazing how the trolls like to cherry pick some WA rangefinder lens that does not work that great in the VG900. This comparison against the $8000 Leica is a joke. Does the Leica have a 1/8000 sec shutter speed in still mode ? Does it shoot 1080p60 ? Does is have clean HDMI output ?
Count yourself fortunate that SLR lenses work well. This is because SLR lenses are designed to accomodate the mirror meaning that light is orthogonal to the capture plane.
However, I disagree that it’s cherry picking to point out that the majority of normal and wide rangefinder lenses experience issues on cameras like the VG900. This is especially so given that rangefinder lenses are smaller designs and will complement the compactness and size of a fullframe NEX much better.
I found the comparison to be interesting, helpful and informative.
Many people are anticipating the FF Nex specifically for using legacy & Leica lenses. By showing samples from the VG-9000 and the actual Leica with the same lens one can see the worst case and best case scenario for how well such lenses may perform.
but there is a demand for this sort of information, just like reasons why tabloids and paparazzi photos are so popular
VG900 was just an attempt to offer something from the Handycam lineup with recycled equipment. Sony does that all the time when they develop a new sensor. Same sensor with small or no tweaks being used across several lineups. I.e., a99, rx1, vg900. The same happened when they launched the NEX-5, a55, vg-10 in 2010 and NEX-7, a77, a65 in 2011.(though there was no handycam with the same 24mp aps-c sensor)
what a meaningless test …
It’s amazing how clueless you people are about the VG900. So what is the problem with Canon FD lenses ?
The Canon 50mm F0.95 is a 50 year old lens.
Sony NEX-VG900 + Canon 50mm f0.95 / first test :
http://www.youtube.com/watch?v=dVBbop5Onws
I was tired to watch this VG900 video with “bokeh” in every scene. Every DSLR with a FF sensor, and with such fast lenses F0.95, has the same bokeh like VG900.
The problem with VG900 is that when used without bokeh … the sharpness across the screen is very soft (like the cheapest camcorders from Sony Handycam line).
But all Sony consumer camcordes and cameras are soft compared with some of the best in sharpness consumer products: Panasonic GH2, HC-X900 (from 2012), HDC-HS900 (from 2011), HDC-HS700 (from 2010) (these results are from 2 web sites with tests about sharpness in camcorders and cameras).
The Lux 50 isn’t the most problematic lens. I’m afraid to see significant smearing with Lux 35 and Super Elmar 18: both these lenses have back element that protrudes quite deeply into a body, see the pic:
https://dl.dropboxusercontent.com/u/12196364/gallery/different/DSC05434.jpg
What does the test tell us? Nothing, sorry for Leike lovers but I can hardly call it a test. Even the test images aren’t the same size. And the VG900 has different coating then an A99, atleast for the dust part that is. Why didn;t they use the RX1R to compare? Thats the potential of the sensor.
This confirms that IQ lies very much, not only in the physical sensor, but in in other electronics and mostly in the code!
Video recording and stills recording is very different and you can never expect optimization to support both 100%. For the VG900 they probably did not try.
This test is not relevant!
Compare a RX1R with a Leica M with 35/2, be it Zeiss or Leica made. This comparison would be a lot more relevant.
“This test is not relevant!” – *facepalm*
– VG900 doesn’t have a dedicated video sensor like FS700.
– “Compare a RX1R with a Leica M with 35/2, be it Zeiss or Leica made. This comparison would be a lot more relevant.” – not at all. You can’t use Zeiss on any E-mount camera, so it’s meaningless when you want to predict NEX-FF behaviour(RX1R got microlenses offset adjusted specifically for this one lens making an enormous difference in performance comparing to any interchangable lens camera). Still though we don’t even know what sensor will NEX-FF have – we’ll see how it goes – but in the end VG900 is as close as it gets to Sony’s FF mirrorless camera and showing all possible issues that might come out with that (eg. rangefinder lenses behaving like garbage)
That seems pretty good. Of course, Leica is sharper, but it’s a photocamera and not a camcorder with a strong AA filter. The main point of this kind of comparison is not sharpness in the center, rightness of exposure or contrast, guys…
The main thing is that the Sony is not bad in the corners. But it was expected with a 50 mm focal length. A comparison with a 35 or 24 mm would be welcome…
Pretty good at f/5.6, but the corner problems are more obvious at wider apertures. It would be nice if they could show us what happens at f/1.4.
You’re right. Wider apertures (and wider lenses, for example 20 to 35) would be far more interesting and more risky for the Sony FF.
I’ve seen some pretty bad results with wide angles mounted on VG900, but I have no clue of how the lens perform on their original mount.
In this linked comparision the Leica 24 smokes the Sony. No comparision, the Sony seems like with a veil, even at high ISO the M24 is better. Uuups…!