Debunking myths: The translucent mirror reduces (slightly) the image detail!
A big question we all have in these days is: Will the NEX-7 deliver better image results than the A77 because it has no translucent mirror? And it seems like the answer could be…YES!
Our friend Ray from Thesybersite.com (Click here). He used the Sony A55 to take a couple of images with the translucent mirror and an image without the mirror (yes he removed the mirror!). In short:
1. The mirror reduces the amount of light reaching the sensor by about 1/2 stop.
2. The mirror reduces image detail (Around 5 % loss of detail).
3. Purple fringing is not influenced in any way by the SLT mirror.
Please take a look at his test at Thesybersite.com (Click here) to see the images and make an opinion for yourself!
ALL Preorder links:
Sony A77 with 16-50mm lens at Amazon, Adorama, Jessops UK and Sonystore.
Sony A77 body only at Amazon, Adorama, Jessops UK and Sonystore.
Sony A65 with 18-55mm lens at Amazon, Adorama, Bhphoto (notification only), Jessops UK and Sonystore.
Sony A65 body only at Amazon, Adorama, Bhphoto (notification only), Jessops UK and Sonystore.
Sony NEX-7 body only at Amazon, Adorama, Jessops UK and Sonystore.
Sony NEX-7 with 18-55mm lens at Amazon, BHphoto (notification only) and Sonystore.
Sony NEX-5N with 18-55mm lens at Amazon, Adorama, Bhphoto, Jessops UK and Sonystore.
Sony NEX-5N at Amazon, Adorama, Bhphoto, and Sonystore.
Zeiss 24mm f/1.8 E-mount at Amazon, Adorama, Bhphoto (notification only) and Sonystore.
Sony 50mm f/1.8 at BHphoto (notification only), Adorama and Sonystore.
Sony 55-210mm E-mount at Amazon, Adorama, Bhphoto (notification only) and Sonystore.
Sky_walker
2 years ago |Nah, that’s old stuff, and… well.. different tests show different results. This isn’t anything professionally made so… I wouldn’t put much weight on it.
pancanikonpus
2 years ago |I think is 5% loss quality detail is not much as what we see in a77 0.xFW ISO to 5N. the comparison high ISO is loss > 5%. Is nex7 experience buffer issue like a77?
alphafan2011
2 years ago |no, whith A77 and NEX 5N its not the main difference the mirror – its the different resolutions. the only realistic comparison would be between NEX 7 and A77
Steve Jones
2 years ago |This has been around for a while. It would be nice to see DXOmark perform a test with and without the SLT mirror in place. As it is, the DXOmark tests show the A580 about half a stop better than the A55 on DR and noise. That’s entirely consistent. That will reduce detail at high ISO, however under half-decent lighting the SLT is going to make a vanishingly small difference to image quality. Unfortunately I do take a lot of high-ISO images at gigs, so it is of some concern, but it still looks way better than my A700.
I’ve no idea where the idea that the SLT caused “purple fringing” came about. That particular phenomenon has been around since the dawn of DSLRs. Here’s a test for that on the A100 (I only chose that model as I can’t be bothered to find anything older – anybody can do a search).
http://www.cameralabs.com/reviews/SonyA100/page4e.shtml
Sky_walker
2 years ago |Th idea came from here: http://kurtmunger.com/sony_translucent_mirrorid181.html There was a different test methodology though and very different conditions. But AFAIK it’s like with ghosting – a phenomena that sometimes occurs but mostly do not. So IMO: not much to bother.
Never the less Kurt Munger seem to take more care in test and actually do something above “taking off mirror and shooting 2-3 samples”.
Steve Jones
2 years ago |I’m afraid that series of photographs fails the first hurdle of performing controlled tests. You simply can’t use cameras with completely different sensors and blame the differences of purple colour fringing on the presence of the SLT mirror. The great advantage of testing the A77 with and without the SLT mirror is it removes that variable.
I’m a bit more convinced about the ghosting demonstrated, but even then tests with and without an SLT mirror in place on the same camera would be better.
Sky_walker
2 years ago |1) Guy at Thesybersite wasn’t testing it with A77.
2) This would be a valid concern if not the fact that only cameras having some reported issues with purple fringing are SLT – and that’s from all using this very sensor, including both classical DSLRs and mirrorless cameras from various brands.
Though I agree that the test methodology of mr. Munger isn’t perfect – it’s far closer to being a valid test than what author of Thesybersite did. Photos made in same place under same conditions don’t prove anything to a phenomena that’s suspected to occur only in specific conditions.
So I’m more convinced to these results than the test made to show no purple fringing at all, yet for certain some complex test would be good to have in order to put out final verdict including possible description of conditions where camera would pronounce the effect and when it would not.
Steve Jones
2 years ago |I know he didn’t test an A77, but the the removing the SLT mirror test can be done on any SLT camera (if you are brave, and it’s easily removable).
Purple fringing is a classic problem caused by many things. Here’s one cause at the microlens level.
http://www.dpreview.com/learn/?/Glossary/Optical/chromatic_aberration_01.htm
pancanikonpus
2 years ago |5n dxo hit 95 points
Steve Jones
2 years ago |I can’t find any NEX-5n results on the DXOmark web site. It says no measurements available. The Nikon D3x rates at 88, so 95 would be simply amazing for an APS-C sensor.
at
2 years ago |The result is expected, but I would think NEX users will use E-mount lenses while SLT users will use A-mount lenses in most cases, and current E-mount lenses are not of very high quality, so the difference in image quality between NEX and SLT can be ignored when making the decision of choice, anyway.
Steve Jones
2 years ago |I think the really serious NEX shooters after the best quality are using top-quality manual lenses with those for the Leica M mount being preferred. It may be the new Zeiss lens is the first really top-notch native A-Mount lens to rival them.
Sky_walker
2 years ago |If the difference in image quality falls below 1/2 EV than it doesn’t matter really and all the other camera features come in play.
But well… there always will be some people who gonna whine about pixel-peeping and some other who shall scream that Nikon is the best because it’s Nikon. I wonder if Nikon would come out with SLT system what would happen? Would would be uber-happy and see all the advantages of it. Just like it was with tilt-able LCDs and live view which were “tools for school kids mothers” till both Canon and Nikon released the cameras, and now suddenly no good DSLR can come without them. LOL. Hypocrisy of Internet crowd is beyond my understanding.
SRL
2 years ago |Yes there are so many variables in camera/sensor testing.
Optics can cause a huge difference; that’s what a camera body is a platform to hold various optical devices, that’s why whenever possible, I put slightly more credence in tests that use a third-party lens like a Sigma so camera/sensor caparisons are slightly less tarnished by Sony vs. Nikon or Canon lenses into the mix.
Mike
2 years ago |I think the loss of light is negligible. What worries me far more is the loss of resolution. Why spend $$$ on an expensive lens that gives you these extra 5% of resolution compared to a much cheaper lens when the SLT mirror then eliminates the gain in resolution?
Sky_walker
2 years ago |The really big question is if this effect will limit the resolution of A77 to the point where there will be a notable gap between it’s image quality and the image quality of NEX7.
Steve Jones
2 years ago |You do realise that loss of resolution is going to be completely invisible unless you print to a huge size – maybe 40 x 60cm and even then you will have to look closely. I’d argue if you really need output that size, then you are heading into at least full-frame, and very probably MF territory. It would be interesting to know just how many times photographers will work with prints of that size at exhibition quality, although I suppose some crop down to those levels.
The loss of light issue will probably affect more photographers as, speaking from experience, indoor available-light photography of action sequences is highly demanding and it’s the loss of dynamic range, rather than detail which can be critical.
Mike
2 years ago |The loss of light will affect you only in one situation: if the 1/2 stop you just lost makes the difference between a usable shot and an unusable one. I argue that in most situations, the amount of light will either be so small that 1/2 stop more or less makes no difference and the shot can’t be made anyway. Or that there’s enough light anyway, even if it’s reduced by half a stop. So unless you are in exactly the situation where the half stop makes the difference, you won’t notice.
Compare that to the reduced resolution: this will affect every single picture you take. Even if you don’t print it now, you might want to print it later or take a crop. Why would I buy a 24MP camera if I didn’t intend to use this resolution?
Sky_walker
2 years ago |“You do realise that loss of resolution is going to be completely invisible unless you print to a huge size” – I don’t care, recently I don’t print images at all.
But I watch them on a PC where I actually sometimes view the image 1:1.
“I’d argue if you really need output that size” – sure I do. Cropping. That sole reason make nonsense-megapixel-cameras actually have some point.
“it’s the loss of dynamic range, rather than detail which can be critical.” – That’s true, very very true. But if someone buys 24 MPx camera it would be good if it wouldn’t be crippled to 16 or 18 MPx thanks to SLT mirror, don’t you think?
Zonkie
2 years ago |The test was on a 16MP sensor. And to my eyes, the loss of detail already looks quite visible (less so in the shot at f/11, but that’s because at f/11 the resolution is already limited by diffraction).
On a 24MP sensor, as long as you use high quality glass and shoot at an aperture bigger than f/7, the loss of detail will be quite higher than 5%. If it’s going to matter in real life or not is debatable, but then so is the difference between a 24MP camera and a 16MP camera, between a $1500 lens and a $400 one, etc…
Pascal645
2 years ago |So good to hear about the NEX-7. I thought it had all just been a rumor
Lot’s of noise until Aug 24th, then … NOTHING. Since then, the NEX-7 has all but vanished. Boooo
Alpha Mark
2 years ago |For a 24mp sensor, lost little detail is still more detail than my A700 12mp. The maximum size i upload on web is only 1200 by 1200 at 72dpi. So its not going to be a big issue.
Bugger
2 years ago |Not surprising because usually it’s these 3 things:
1. quality of sensor
2. optimisation of sensor/processor
3. quality of lens and/or filter
Now you have another object in the path of capturing the light.
Maybe they should make a zeiss mirror. ehehe.
Mist
2 years ago |I am thinking about a crazy idea, can sony remove the mirror, let all of the light enter the main sensor, then using an EVF to project the image from the main sensor to the phase detection auto-focus sensor to make phase detection possible.
Is it possible?
Steve Jones
2 years ago |That is a crazy idea – it simply won’t work, as a PAF sensor requires two perspectives on the same target, much like stereoscopic vision. You simply can’t do that from the main sensor in the way you outline. If you could, then you could use the image direct from the sensor. Note there are attempts to put PAF on the main sensor, but not very successfully.
Here’s a useful explanation of how it works
http://graphics.stanford.edu/courses/cs178/applets/autofocusPD.html
However, if 3D stereoscopic photography ever become successful, then that will solve the problem as the two sensors can provide PAF.
Mist
2 years ago |i see, lol
David
2 years ago |Well, 5% loss of detail would never be noticeable in the real world.
I’d say that I have more missed, poor-quality shots from things like motion blur, camera shake, missed-focus. They are much bigger “problems” than an absolute lack of detail.
I think people would be surprised how many of their images will be *slightly* out of focus. If you stop, live-view zoom and manually focus everything, things are sharper. It’s just more time-consuming.
What I’m saying is that there are a TON of factors important in a good quality picture. And a 5% reduction in detail will never be noticeable in real-world conditions.
Steve Jones
2 years ago |Indeed, I see that from my own photos which are often of performers. However, for those focussing still scenes who do want the utmost accuracy, the availability of focus peaking on the A77 could be a big winner. For ultimate accuracy, nothing beats contrast-based focussing. Manual focussing using focus peaking is relatively fast and much more accurate than doing it unaided unless you have exceptional eyesight.
However, much more important, is that if you do need that 5% detail, then a tripod or other such support is essential. Quite simply, in the vast majority of situations people can’t hand-hold cameras to the required exacting standards even if using image stabilisation and using conservative shutter speeds. Anybody taking hand-held action shots and simultaneously complaining about the theoretical loss of 5% of detail is talking total nonsense.
David
2 years ago |Absolutely.
I despise tripods, and I’ll happily accept a bit of IQ loss in exchange for not carrying one.
It should be the same for SLT. There are a *TON* of benefits. And if that “costs” you 5% IQ, it is still a very good deal.
For moving subjects, the time between focus lock and shutter opening can still mean subjects which are slightly out of focus. Also, for people using focus-recompose, it will introduce minor focus errors too.
People should work on their own photographic technique first, before complaining bitterly and emotionally about 5% hypothetical loss of image quality on a stationary test target.
Nico Foto
2 years ago |+100, very well said. By what i read over the internet, it looks like everyone is a super pro photographer shooting the second arrival of Jesus, or the first alien contact, or (insert your one time only pic that will change the course of the world as we know it).
I would like to understand how a 5% IQ loss affects people’s work, and how they would measure or notice it…Get out and shoot!!!
David
2 years ago |lol, and funnily enough, most of the most famous pictures are terrible in terms of compostion, lighting and image quality! That’s not what matters, and I doubt the a77 image quality would be a factor which holds ANY photographer back creatively.
But all those other features DO help you creatively.
Zonkie
2 years ago |Yes, but then why are people spending so much money in gear that will give them that slightly better image quality?
At normal web sizes and print sizes (what most people use), you can’t tell the difference between a CZ prime and a Tamron zoom. Hell, even if you need a large depth of field and shooting in daylight (landscape photography), at small sizes you won’t be able to tell the difference between an image taken with a Leica S2 (medium format) and a Canon G12 (compact camera).
So you are right in a way, but then people like to get the most out of their gear, even if they don’t really need it (not talking for myself, I just have a Panasonic LX3 and I’m happy with it).
David
2 years ago |For me it’s more about the useability of the gear
I pay for for fast/reliable focusing. Good build quality. Weather resistance. Buttons in positions that I like. And also the fun and features of shooting.
Even entry-level dSLRs have incredible amounts of detail now, and with post-processing the limitations shrink even more!
David Roberts
2 years ago |I really hope someone comes up with a proper A77 review soon. All this speculation/navel gazing is getting quite old.
Sky_walker
2 years ago |+1
Jonas
2 years ago |because the design has taken into account, it should be a mirror,, you have to change more than just remove the mirror to avoid the 5% loss!
tipper
2 years ago |When the A55 loose 5% the A77 will lose 10% when the translucent mirror is the same because of the higher resolution sensor.
The main question for me is what about the microlense-grid before the sensor in the NEX-7? How will this influence the resolution of the different lenses?
I’m waiting with visible impatience for the first NEX-7 tests…
the666bbq
2 years ago |how many of us still carry UV or other cheap “protective” filters in front of all our lenses ?? how much image quality would we gain by removing those first (permanently) ?
Wheelus
2 years ago |I was thinking the same thing. How many people use neutral density filters and polarizing filters, etc. on their cameras. What affect does that have?
David
2 years ago |At least ND and polarisers *DO* something
UV filters do nothing at all.
Greg
2 years ago |Is the A77 translucent mirror the same? Where has it been said, that it is the same mirror?
tipper
2 years ago |A filter in front of the lens has much less optical effects then a filter behind the lense!
Rog
2 years ago |As I am in the market for a SLT I note with concern the 5% loss in detail but I also note the author had mentioned at the end that he had removed, flexed, fingerprinted and cleaned the SLT mirror with cotton swabs.
Oswald
2 years ago |Per Sonys website the a77 mirror is 2nd generation. So yes it should be different than the a55s.
Buel
2 years ago |No surprise. I’m a confessing pixel peeper. Good is not good enough when better is possible. Sony should use a translucent but removable mirror in its high end cameras. For video and when there is enough light or the speed is required it is perfectly fine to keep the mirror in place. But when the best iq is required the mirror has to move away.
David
2 years ago |Then why are you even looking at APS-C cameras?
The whole purpose of APS-C was compromise. Giving an acceptable level of good image quality in exchange for a smaller, lighter, cheaper camera.
If you’re a pixel-peeper, go at least medium format if you demand the absolute quality
b shaw
2 years ago |+1
Buel
2 years ago |Of course you always have to deal with compromises. MF is simply to large and heavy. And high iso performance is not that good, btw.
It wouldn’t be that complex to use a removable translucent mirror to get (even) better iq. so the portable A77 kit could have been slightly better in iq than it is now.
But actually I’m not looking for APS-C, but for the coming full framers. I’m using a Canon 5DII, but have still some nice Minolta lenses I would like to use again. So I’m hoping that the full frames will have a removable mirror …
Rob
2 years ago |So if I stop down from f/3.5 to f/4.0 will I lose 5% detail?
And . . . will I *really* be losing 5%, or will it be more like 4.74% or maybe 5.82%, or some other number plucked from someone’s imagination?
And . . . how much detail do I get added to a hand-held shot at a slow shutter speed by the absence of mirror slap?
I’d rather spend my time maximizing what my equipment *can* do instead of speculating about what *might* be its shortcomings.
Yeah, it’s raining here . . .
Mike
2 years ago |No because f/3.5 is not half a stop faster than f/4, even if the difference is 0.5. First, get your maths right, then you can bother if it’s 4.74% or 5%.
The question people are trying to answer here is not what they can do with their existing equipment but if they should invest in new equipment.
Linh
2 years ago |Ditto! And the most recent debate is whether or not to buy NEX-5N or NEX-7.
David
2 years ago |The nex7 and 5n are very different. NEX7 looks nicer to use, better to handle etc. That should count more than some tiny difference in image quality measured on a test chart.
Those real world a77 pics from the concert look pretty damn good. And it focused corrected, dealt with the funny white balance and pulled out plenty of details in a real world environment under difficult lighting conditions. That counts for more than grid-lines, test charts and other measurebation nonsense.
Rob
2 years ago |It’s a rhetorical question, corollary to an arbitrarily-arrived-at quantity. Get an understanding of empirical data vs. contrived speculation and then you can “bother” with commenting on what I wrote.
Maybe I should have added that I just got an a35 in part *because* of its apparently controversial innovations, and that those specific features in fact augment my photography experience.
It’s sunny today…I’ll be out shooting. Hereafter, too. Better to leave discussion to the measurebators.
Wheelus
2 years ago |Yes. I’m amazed that in the days of 35mm film and ASA 400 film (look up ASA if you are to young to know what it is) that any photographer ever shot a decent image. No image stabilization, no 1/8000th of a second shutter speed, no lenses with 35 elements in 400 groups (that’s a joke). Rob, you are correct, work with what you have. Each camera will have its own faults. Work with them and you will understand what photography is all about. An how did those photographers with 4X5 view cameras ever function shooting news and then having to process their film.
And today we complain that that we can’t shoot at an ISO of 125,000, have zero noise, and make a 30X40 FOOT print.
Please learn your craft and quit depending on technology to do everything for you.
Andrew
2 years ago |Well put!
Cliff
2 years ago |I think a good head to head test between the A77 and NEX7 with the same lens would end much of this debate. (Then take the winner and test it against the NEX 5n)
Cliff
acolyte
2 years ago |with the SLT adapter or normal adapter? :p
Cliff
2 years ago |Lmao…both, I guess!
LuckyNumberSlevin
2 years ago |If high quality uv filters leave their mark, you can bet your behind that a translucent mirror will.
Wayne
2 years ago |Why is the loss of some light on the sensor, with its impact on detail and noise so hard to accept? Yes, the Sony marketers have used all kinds opf hyperbole to describe the A77 IQ. That’s what they are paid for and no self respecting business in it for the money would do otherwise. But if you got one of their engineers at lunchtime and asked them about real world IQ impacts due to the SLT design I’d bet they would completely agree that there are some compromises that can’t be overcome (yet).
With every new, emergent technology there will always be something given up, especially with early iterations. Two things: 1) Anyone expecting the “production firmware” images from the A77 to match or exceed the high ISO quality of cameras like the K-5 or D7000, are likely going to be disappointed 2) Those who are already writing the A77 off as a severe disappointment, as if the very fact that a small compromise in IQ is a personal affront, are missing the point entirely. What’s the point then? The A77 is a game changer. Whether we like it or not, the market is evolving. Market demands call for better video implementation (fast/reliable AF during video – what a novel concept!!!), improved continuous AF for the masses, good live view experience,etc etc. Sony has done their homework here, they know what the market wants, and they have a forward looking vision (Not that Canon and Nikon don’t, but they certainly seem to be more willing to apply traditional assumptions of film camera engineering to DSLR construction). Unless these other players respond to this emerging technology with something similar, or an alternative that will answer some of those same market demands, Sony WILL gain a sizable foothold in this space and they will take market share from Canon and Nikon. Large chunks of it.
For those of us ONLY looking for maximum IQ in an APSC sized camera, don’t care for next gen video capability, and the thought of an EVF that gives us a live pre-view of the final image output scares us to death (and we aren’t willing to accept some inherent EVF weaknesses), I’d suggest either the A580, D7000, or K-5 as an upgrade. The A77 is not not for you, it represents something new that brings a wealth of game changing features to the table. That either resonates with you, or it doesn’t.
Cliff
2 years ago |The bottom line issue in all this is;
Just how important is super fast auto focus to you?
This whole worldwide debate was caused by a bunch of Sony developers that think that fast auto focus is REALLY important to us. That’s it!
Cliff
Andrew
2 years ago |Speaking for myself, AF speed is of minor importance. I could care less about what the SLT design does to advance AF. The real benefits of the SLT, as I see it, are the absence of a moving mirror, and believe it or not, the evf. Don’t get me wrong, nothing will replace the joy of using a good optical viewfinder…and on that note, I will continue to enjoy my a850 for years to come. However, if change is eminent, to me, mirrorless cameras are the future. The NEX-7 shows more future promise in my eyes. The SLT design is simply a stepping stone for those who are reluctant to abandon a traditional DSLR design completely.
Edward
2 years ago |Agreed. CDAF is more than fast enough for 99% of what I shoot. That is why I wish Sony could provide us with 2 models, one SLT and one mirrorless with on sensor CDAF, or make the mirror user removable.
Wayne
2 years ago |As much as many of us would like, We’ll never see a manufacturer sell two cameras that fill roughly the same price point-functionality-user niche, other than a key functional difference such as SLT or lack thereof. The two cameras would canibalize each other’s sales. There are always differing functional needs when you look at the more granular user segments. I love the idea of having PDAF for improved contiuous AF for action shots and reliable and accurate AF for video (in which CDAF simply doesn’t come close), in a pro-sumer level body. But many others here don’t need that. On the other hand, I have a lesser need for critical CDAF for macro work/still shot situations. A common need however is that we all want the best IQ possible, so we have to decide what we are willing to give up IQ wise for the other features we want/need.
I do agree with the gist of your post though – ideally we should have an option in the same body. To take your removable mirror idea further (I’m sure others have suggested this) why not make the SLT mirror rotate so it can lock up for those times you don’t need/want PDAF and you want the best IQ possible.
One other note: recent comparisons of the Nex7 and the Nex5n suggest (subject to pre-production firmware disclaimer) that the IQ degradation in the A77 samples is not entirely due to the SLT, but also due to the new 24 MP sensor. **I’ll withhold my opinions on going that high with an APSC sensor, when the 16 MP version seems to perform wonderfully.
Wheelus
2 years ago |I have seen tons of discussion and reviews of the A77/65 about ISO 3200 and above. Has anyone looked at ISO images of 50?
Back in the “good old days” if you were a professional photographer, say for National Geographic, you shot as much as possible with Kodachrome (ISOs of 32 and 64) and Pan-X and Plus-X for black and white. It was rare to go above an ISO of 400, and then it was “push-processed” but only for unusual situations.
How does the new mirror and noise look in the ISO 50 mode?
If there is a loss of 1/3 to 1/2 stop with the A77/65 while shooting someone in a black suit in a coal mine at midnight, might there be some advantages with shooting scenic photographs at noon in Arizona in the middle of summer and less of a need to shoot with neutral density filters?
Andrew
2 years ago |I second that! Give the camera a chance to play to its strengths. I’m very keen on knowing, at it’s base ISO, what this new sensor is capable of. I plan on purchasing the NEX-7 regardless, as it looks like it will be a highly versatile and usable camera. However, if this new sensor can deliver similar results as the a850/900′s at base ISO, the NEX-7 would make a fantastic camera for tripod work.
rttew
2 years ago |not a math wiz but let’s say the 5% loss in resolution is in fact true. if you take a nearly 25 megapixel image (A77, el at) and reduce it by 5%, that gives you 23.085 instead of the 24.3 it’s advirtised to be. talk about pixel peeping…geez, when is there ever enough resolution for some people???
Edward
2 years ago |The correct answer is 21.66 mp. 5% drop in resolution would make the sensor 3800×5700 pixels. That’s a huge difference IMHO. Now the 5% number is not based on any measurements so it’s just an unreliable estimation. However from the samples shown, I do find a clear image degradation, and I find that unacceptable.
David
2 years ago |Then again, why are you looking at enthusiast level aps-c cameras?
edward
2 years ago |Because the future FF from Sony will most probably feature an SLT design. I want to know what to expect from this technology in terms of image quality.
Mike
2 years ago |A 5% drop in resolution is NOT a 5% drop in pixel count.
The picture will have exactly the same size and amount of pixels. But you’ll simply see less details.
John
2 years ago |Could we not test this using the same lens and camera, just different lens adapters. LA-EA1 and LA-EA2.
EA2 has the mirror in it.
Werner Ruotsalainen
2 years ago |Folks,
I’ve posted a comparison between the Canon 60D and the A77 to http://forums.dpreview.com/forums/readflat.asp?forum=1037&message=39230692 . The full message:
The 60D resolves detail much better even with its lower-resolution sensor. A PNG-based crop of the resolution chart in the top right corner, without any processing, is as follows:
http://dl.dropbox.com/u/13100693/html/a77res.png
http://dl.dropbox.com/u/13100693/html/60dres.png
BTW, speaking of this chart, resolution degradation is certainly visible even at ISO800 compared to ISO50/100.
As you can see, the 60D is able to resolve the (semi-)parallel lines to about 8 (or even more), while no parallel lines can be seen on the A77 over about 6.5 … 7. That is, however much the A77 has a higher-resolution sensor, it still delivers worse actual resolution because of the very strong NR.
Wheelus
2 years ago |You find clear image degradation that is unacceptable when compared to what? I have no problem whether it is 21, 22, 23, or 24 when compared with 16 or 18. You may find it unacceptable but I’ll take it…
the666bbq
2 years ago |funny how emotional some discussions get quickly, A77 surely does something to people
frankly I don’t see the fuss about fast or even continuous autofocus. When the autofocus is accurate and flexible, why wouldn’t the processor handle ‘sharp’ better than me in most situations (without the oldschool two-half-bubble I can be a lousy manual focusser).
I’m curious about the EVF (only have tried A55 yet). I have to do my stuff now through the pinhole viewfinder (already magnified) of the A300, so a bigger than average viewfinder with 100% accurate coverage sounds cool. For macro, being able to enlarge my focal point (for manual focus control) appears a treat.
Still not sure how the DOF preview is handled/offered on A77, but WYSIWYG could be nice (A300 without DOF button is a trial and error nightmare sometimes if you dial away from the green button).
There shouldn’t be too much lag for my likings though, I have other and more pleasant ways to get drunk than to look through a viewfinder.
The EVF for reviewing photos ? no Sony, give me tethering instead please, I know you can do that…
Noise .. yes, had too much of that on some occasions (like that day when I forgot to zero out the ISO 3200 from the night before) but than again, I see fellow photographers in the field, including myself, who can easily improve a few stops with technique (holding camera, positioning, breathing, chosing the ‘silent’ moment, … you know those things that make your pictures better than those of others – your partner, the excellent camera you selected, will thank you for letting him shine
Will I become a movie-man, probably not, will I use the excellent video features (look at swiss guy) : of course, our Panasonic handycam is not that handy (says the dust on its case). It is clear that this new technology does a good job at that (hope I don’t get the videocam-that-can-do-stills feeling when I give the A77 a fair try-out)
Wheelus
2 years ago |Sony advertises on the A65/77 Multi-frame Noise Reduction: “Experience incredible low-light shooting without a flash. The camera captures six images in a fraction of a second. Combining the data from all six, it creates a single image with a reduction in noise equivalent to two additional steps of ISO sensitivity. Sensitivity selectable up to ISO 25600. (Recommended for still subjects.)”
Has anyone seen any photographs taken in this mode?
Barbarous Waytel
2 years ago |The “test” doesn’t consider that that 5% blur is used as part of AA filter compensation. So you can put a weaker AA filter.
Compare A55 resolution (at base ISO) with other cameras with the same sensor.
A55 resolution score is better than D7000
It’s better than A580
And far better than K-5
You can find these numbers in IR review for these cameras.
JustAThought
2 years ago |I’ve seen a few results from the A77 now and, although it is not yet the production model, they seem uninspiring.
Is this a sensor or a software matter? If it is the first than Canon may end up as the big winner this autumn since Nikon is highly likely to use the new Sony sensor or Nikon has to stick with its current senors (which I believe are from Sony).
Renaud
2 years ago |I had 2 Minolta X700. I own several Sony : A100, A700, A55.
Quite disappointed by the first reactions and tests of the so long awaited A77. Where is the beef ? Why is the AF area so tight ?
I like my A55, but I’m disappointed by the slight delay I have with my A55 : what I get was not what I saw when I pressed the shutter button, and heard the shutter.
And by the comparison before-without mirror. See the photos on thesybersite.com : we sure lose a lot of details and contrast.
I begin to wonder if the all “translucent + EVF” strategic choice is not a big mistake, at least for the serious/enthousiast user. Not speaking about the lack of any FF foresight.
And I begin to wonder if it’s not the time to jump out of the wagon, drop the consumers electronics company that Sony has always been, and switch for a serious cameras and lenses maker, I mean Nikon or Canon, maybe Pentax. Not speaking about Oly and Panny.
Renaud
2 years ago |I meant D700.. I also own a GH1.. which is -despite the not-so-fast 14-140- better when making clips : fast and silent autofocus, far, far better face detection.
A55 has a poor face detection, especially for close-up portraits..