A77 with and without mirror test

Started Mar 23, 2012 | Discussions
rob asnong Regular Member • Posts: 398
A77 with and without mirror test

Hi all,

I took the time this morning to do the test myself.
and wanted to share the results with you.

My test setup was the following:

1. first i wet cleaned the sensor to be sure that there wasn't any smearing film on it, just to be sure (I clean the sensors every 3 months)
2. shots are taken on a tripod with the 70-400 mm at 300 mm
3. the tree was approx at 100 meter distance

4. i wanted a high contrast scene for eventually see differences in PF (though the G lens is too good for that
5. no PP at all, just using ACDSee 5 Pro for converting the raw files to jpeg's

6. afterwards I noticed that the jpeg's (in standard mode) were shot with sharpness in +1 (I should have avoided that, but it is too late now ;-))

I am looking forward for your conclusions except that the loss in light is noticed in the speed chosen by the cam as it goes from 1/1000 sec to 1/800 sec due to the mirror in place, but this is no news.

By the way, I love this A77 (and I missed him more than any other camera when he was in for a repair to replace the front dialer from which the rubber ring brook and came off).

In the meanwhile I used my A580 (with it's true little more picture quality, just my findings, don't hurt me now ;-)) and my beloved and trusty old horse A700 (with it 130.000 actuations but still going strong) as a back up.

And yes ... I love the EVF more than the OVF, even the FF OVF's. Now that I was 3 weeks without the A77, I missed the EVF so much. Who would have thought that from a 60 years old amateur who lived with OVF's for 40 years!

The results:

This is a 100 % crop RAW without mirror.

The same shot with mirror.

The jpeg without mirror.

The jpeg with mirror.

Don't expect me to react on all your comments, because I have no time for this, but I wanted to share my test with you guys, that's all.

Happy shootings

Rob

 rob asnong's gear list:rob asnong's gear list
Sony a77 II Sony Alpha DSLR-A100 Sony Alpha DSLR-A580 Sony SLT-A77 Sony Alpha a99 +13 more
Sony Alpha DSLR-A700 Sony SLT-A58 Sony SLT-A77
If you believe there are incorrect tags, please send us this post using our feedback form.
mike_2008 Veteran Member • Posts: 5,635
Re: A77 with and without mirror test

Thanks for posting the results from this rather courageous test.

Before commenting on the images, I'm really curious to know if you found any problems resulting from removing and then replacing the mirror, in terms of AF performance or image artifacts. I would have thought that even a small misalignment in the SLT mirror might cause problems.

Your results are interesting and surprising. It almost looks sharper with the SLT mirror. Could small differences in the focal plane explain this?

Going from 1/1000 to 1/800 I guess is pretty much in line with a 30% cut in light, and consequently half a stop longer exposure. So like you say no real surprise there.

-- hide signature --

IQ is not judged exclusively by high iso noise performance

 mike_2008's gear list:mike_2008's gear list
Canon EOS 5D Sony SLT-A55 Sony SLT-A77 Canon EF 50mm F1.8 II Canon EF 17-40mm f/4.0L USM +5 more
tbcass
tbcass Forum Pro • Posts: 46,724
Re: A77 with and without mirror test

I normally don't comment on tests like these because pixel peeping looking for minute flaws isn't my thing but here's what I see.

The RAW with mirror looks better than the RAW without mirror. Did the focus change between shots? The JPG's Look so bad with excessive sharpening and contrast that comparison's are useless.

While this test doesn't prove anything, to my eyes the hit on IQ with the mirror down is negligible to non existent.
--
Tom

Look at the picture, not the pixels

http://www.flickr.com/photos/63683676@N07/
http://www.flickr.com/photos/25301400@N00/

 tbcass's gear list:tbcass's gear list
Sony RX100 Sony RX10 IV Sony a99 II Tamron SP AF 90mm F/2.8 Di Macro Tamron SP 70-300mm F/4-5.6 Di USD +7 more
mike_2008 Veteran Member • Posts: 5,635
Re: A77 with and without mirror test

tbcass wrote:

I normally don't comment on tests like these because pixel peeping looking for minute flaws isn't my thing but here's what I see.

The RAW with mirror looks better than the RAW without mirror. Did the focus change between shots? The JPG's Look so bad with excessive sharpening and contrast that comparison's are useless.

While this test doesn't prove anything, to my eyes the hit on IQ with the mirror down is negligible to non existent.

This is what most serious reviewers and real-life users have been saying about the A77, the only hit to IQ comes from the modest loss of light, there are no artifacts, contrast problems, 'magenta cast' or ghosting in the latest SLT cameras, much to the chagrin of trolls and doom-mongers no doubt.

So I think sony have got some great optical engineers who know how to make very high quality multi-layer optical surfaces. And that's great. What I don't get is why the AF sensor needs 30% light. The new D800 has 1 stop more sensitive AF system than the D700, if sony did this they could reduce the light loss down to 15% and have identical AF performance.

Once we are close to 10% light loss the difference between DSLT and DSLR become effectively invisible in terms of IQ, as this is the kind of variability you get from exposure metering variations (at least).

-- hide signature --

IQ is not judged exclusively by high iso noise performance

 mike_2008's gear list:mike_2008's gear list
Canon EOS 5D Sony SLT-A55 Sony SLT-A77 Canon EF 50mm F1.8 II Canon EF 17-40mm f/4.0L USM +5 more
Dultimate Regular Member • Posts: 115
Re: A77 with and without mirror test

Forgive my ignorance but I keep hearing (reading) about this SLT mirror issue. What exactly is the mirror for if it's functional without it, with a 30% light reduction no less?

 Dultimate's gear list:Dultimate's gear list
Sony a7R III Sigma 135mm F1.8 Art Samyang AF 85mm F1.4 FE Samyang AF 18mm F2.8 FE
Gavin 11 Senior Member • Posts: 2,271
Re: A77 with and without mirror test

They look better with the mirror in!
--
--

"When I warned [the French] that Britain would fight on alone, whatever they did, their Generals told their Prime Minister and his divided cabinet: 'In three weeks, England will have her neck wrung like a chicken.

"Some chicken....Some neck!

 Gavin 11's gear list:Gavin 11's gear list
Sony Alpha DSLR-A700 Sony DT 18-55mm F3.5-5.6 SAM +5 more
FreeRadical009 Regular Member • Posts: 331
Re: A77 with and without mirror test

Dultimate wrote:

Forgive my ignorance but I keep hearing (reading) about this SLT mirror issue. What exactly is the mirror for if it's functional without it, with a 30% light reduction no less?

Since the SLT mirror does not flip up like a regular DSLR mirror (which is actually translucent as well...), it reflects 30% of light upwards into the AF sensor module, so the camera can AF continuously with no problem at the expense of a little light loss. The 30% that's reflected, doesn't reach the sensor, so the exposure requires 0.3 to 0.5 step more to compensate for that. In contrast, a DSLR mirror flips up completely and all the light reaches the sensor, but the AF can't be continuous unless the mirror flips back down, the camera focuses and then swings back up.

Some people can live with it and some just can't and go bonkers about the whole construct.

Back to topic, I don't know about you people, but the shots with the mirror IN look sharper and better all around to me!!

This completely contradicts all the "the mirror was sent here by the devil to ruin our shots and our cameras!!!!!" comments I hear against SLT.

Boy, this is going to tick some people off big time...

reflected_light
reflected_light Regular Member • Posts: 416
Re: A77 with and without mirror test

I'm not seeing any difference in sharpness (but maybe I'm just missing that), but what I do see is that the exposure with the mirror is making the image slightly brighter. The blue of the sky is a little lighter in color, you can see the limbs of the tree just a bit better. So I would say the exposure is more accurate with the mirror in place.

Since the subject is quite a distance from the camera, I would assume the lens was focused at infinity and therefore AF was not used for the shots.
--
r_l

http://www.pbase.com/reflectedlight

Dave Oddie Veteran Member • Posts: 4,153
Re: A77 with and without mirror test

FreeRadical009 wrote:

Some people can live with it and some just can't and go bonkers about the whole construct.

How true!

Back to topic, I don't know about you people, but the shots with the mirror IN look sharper and better all around to me!!

This completely contradicts all the "the mirror was sent here by the devil to ruin our shots and our cameras!!!!!" comments I hear against SLT.

There have certainly been several (to say the least) posts form some people saying they will never buy another Sony because the SLT mirror degrades image quality flying if the face of many examples of excellent quality images posted from all the SLT cameras from the A33 to the A77.

I think its the FUD factor at play.

It's similar to the effect the DPR tests showing the jpegs are not as sharp as raws from the A77. Read that and you would be forgiven for never shooting jpegs again but having shot some myself to use features like multi-frame noise reduction and having left the camera in jepg mode, the ordinary shots look amazing detail-wise anyway.

They are that good I am questioning my need to shoot raw at all except in certain more challenging lighting conditions.

But hey...DPR say the jpeg engine is bad so I must be a complete idiot for believing my own eyes.

It's the same with the effect of the mirror. The FUD factor decrees that for some people the mirror MUST degrade the image to a degree the cameras are fatally flawed.

 Dave Oddie's gear list:Dave Oddie's gear list
Sony SLT-A77 Sony DT 11-18mm F4.5-5.6 Sony DT 16-80mm F3.5-4.5 ZA Carl Zeiss Vario-Sonnar T* Sony 70-300mm F4.5-5.6 G SSM Sony 500mm F8 Reflex +5 more
Ed at Ridersite Forum Pro • Posts: 19,441
Re: A77 with and without mirror test

I agree, the with mirror shot is sharper and jpgs look over sharpened.

Are these from the center? Was there any change in the framing - did you remove camera from the tripod between shots (to remove mirror)? I'm just wondering if the focus point was slightly different? That seems a likely possibility since you would have to at least remove the lens to remove the mirror and even a slight movement in framing would cause focus differences in something like a tree.
--
AEH
http://aehass.zenfolio.com/
Question: What do you do all week?
Answer: Mon to Fri. Nothing, Sat & Sun I rest!

 Ed at Ridersite's gear list:Ed at Ridersite's gear list
Sony RX100 II Sony SLT-A68 Sony DT 16-50mm F2.8 SSM Sony 70-400mm F4-5.6 G SSM II
Vietto New Member • Posts: 4
Re: A77 with and without mirror test

Thaks Rob, form you test seems that photos with mirror are better, this sound strange, the best should be a similar test but in cotrolled light, eg still life, with a subject with many particular.
--
Regards
Marco

dennismullen
dennismullen Veteran Member • Posts: 9,019
Re: A77 with and without mirror test

Even cell phones take great pictures in good light! You lost some shutter speed with the mirror. It's a good thing the tree wasn't moving.

Cheers,
--

"Those who would sacrifice liberty for safety deserve neither liberty nor safety." - Ben Franklin.
You can see larger versions of my pictures at http://www.dennismullen.com .

Karl Scharf
Karl Scharf Veteran Member • Posts: 4,211
Re: A77 with and without mirror test

What surprises me most is the effect that the +1 sharpening has on the JPG image. The image taken with the mirror in place seems quite a bit over-sharpend.
--

Sony SLT-A77 / Rokinon 8mm / Sigma 10-20 f4.0-5.6 / Sigma 18-250 f3.5-6.3 / Sigma 50-500 f4.5-6.3 OS / Minolta 70-210 f4.0 / Minolta 50 f1.7 / Kenko MC4 AF 1.4 / Sony HVL- F56AM flash Karl Scharf

 Karl Scharf's gear list:Karl Scharf's gear list
Sony a7 III Tamron 28-75mm F2.8 III Tamron 17-28mm F2.8 Di III RXD Tamron 70-180mm F2.8 Di III VXD Sigma 100-400mm F5-6.3 DG DN OS +18 more
liquid stereo
liquid stereo Veteran Member • Posts: 7,271
Modest 10%

This "test" suggests 25%. More rigorous tests suggest even more. And you even state that the AF needs 30% of the light. How do you come up with 10% or is that a typo?

It would be interesting if some brave soul would alter their "translucent film" to different degrees to see the degree to which the AF performance decreases.

Cheers!

mike_2008 wrote:

tbcass wrote:

I normally don't comment on tests like these because pixel peeping looking for minute flaws isn't my thing but here's what I see.

The RAW with mirror looks better than the RAW without mirror. Did the focus change between shots? The JPG's Look so bad with excessive sharpening and contrast that comparison's are useless.

While this test doesn't prove anything, to my eyes the hit on IQ with the mirror down is negligible to non existent.

This is what most serious reviewers and real-life users have been saying about the A77, the only hit to IQ comes from the modest loss of light, there are no artifacts, contrast problems, 'magenta cast' or ghosting in the latest SLT cameras, much to the chagrin of trolls and doom-mongers no doubt.

So I think sony have got some great optical engineers who know how to make very high quality multi-layer optical surfaces. And that's great. What I don't get is why the AF sensor needs 30% light. The new D800 has 1 stop more sensitive AF system than the D700, if sony did this they could reduce the light loss down to 15% and have identical AF performance.

Once we are close to 10% light loss the difference between DSLT and DSLR become effectively invisible in terms of IQ, as this is the kind of variability you get from exposure metering variations (at least).

 liquid stereo's gear list:liquid stereo's gear list
Fujifilm FinePix X100 Canon EOS 7D Fujifilm X-Pro1 Fujifilm X-Pro2 Canon EF 50mm F1.4 USM +9 more
FreeRadical009 Regular Member • Posts: 331
Re: A77 with and without mirror test

Dave Oddie wrote:

FreeRadical009 wrote:

Some people can live with it and some just can't and go bonkers about the whole construct.

How true!

Back to topic, I don't know about you people, but the shots with the mirror IN look sharper and better all around to me!!

This completely contradicts all the "the mirror was sent here by the devil to ruin our shots and our cameras!!!!!" comments I hear against SLT.

There have certainly been several (to say the least) posts form some people saying they will never buy another Sony because the SLT mirror degrades image quality flying if the face of many examples of excellent quality images posted from all the SLT cameras from the A33 to the A77.

I think its the FUD factor at play.

Several is understating it mate... The US Library Of Congress would need an entire wing to file the posts all over the DPReview forums about SLT criticism and just plan FUD...

Aside from the purple fringing and ghosting of the A33-A55, I didn't have too many things to criticize the SLT option. Now those two issues have been solved in the A77 and following models, I find less reasons to worry about eventually getting a SLT model.

It's similar to the effect the DPR tests showing the jpegs are not as sharp as raws from the A77. Read that and you would be forgiven for never shooting jpegs again but having shot some myself to use features like multi-frame noise reduction and having left the camera in jepg mode, the ordinary shots look amazing detail-wise anyway.

They are that good I am questioning my need to shoot raw at all except in certain more challenging lighting conditions.

But hey...DPR say the jpeg engine is bad so I must be a complete idiot for believing my own eyes.

To be quite honest, I was able to get by with the JPEG engine of my A700 for a long while and no one ever complained. No one at the exhibits ever said to me: This shot is a straight out of camera JPEG, right? It sucks!!!

Certainly Nikon has a better engine, but if your shot is not strong enough, what's the difference between an excellent and a good JPEG engine? None...

Sony definetely provides better results with RAW. I've seen better results from RAWs across the board of models once you compare them to JPEGs.

I didn't like the idea of messing with RAW, but once I saw all the things I could do and how much it could improve my photography, I went RAW 100%.

That being said, I don't criticize other people for using JPEG only. I used to do the same and I understand why they do it. The fact I shoot RAW does not give me a licence to say other people are idiots for using JPEG. Only pretentious snobs think that way.

Everyone is entitled to use their camera as it works best for them. No one signs a pledge of using RAW always..

Heck, I know people with A200's that use them on the lowest resolution possible, in Standard quality, because that's how the camera best works for them, who am I to call them names for not shooting RAW all the time?

It's the same with the effect of the mirror. The FUD factor decrees that for some people the mirror MUST degrade the image to a degree the cameras are fatally flawed.

I do agree that Sony shouldn't abandon OVF for those that don't like SLT but don't want to lose their money spent on lenses and accesories. That is not a nice move by Sony, but saying that SLT is the doom of photography for Sony users is just as bad, moreover, not true after all.

romicva Contributing Member • Posts: 578
Re: A77 with and without mirror test

+1
same here

mike_2008 Veteran Member • Posts: 5,635
Re: Modest 10%

liquid stereo wrote:

This "test" suggests 25%. More rigorous tests suggest even more. And you even state that the AF needs 30% of the light. How do you come up with 10% or is that a typo?

I was saying that if AF sensors with twice the light sensitivity (like the D800 relative to the D700) were used then identical AF performance could be obtained with half a much light reaching the AF sensor, so the proportion of reflected light goes from 30% to 15%. This could be pushed even further to closer to 10% with a careful balance of the trade-off between low light AF performance and light loss.

Once the light loss is around 10% it becomes irrelevant compared with other variables.

It would be interesting if some brave soul would alter their "translucent film" to different degrees to see the degree to which the AF performance decreases.

Cheers!

mike_2008 wrote:

tbcass wrote:

I normally don't comment on tests like these because pixel peeping looking for minute flaws isn't my thing but here's what I see.

The RAW with mirror looks better than the RAW without mirror. Did the focus change between shots? The JPG's Look so bad with excessive sharpening and contrast that comparison's are useless.

While this test doesn't prove anything, to my eyes the hit on IQ with the mirror down is negligible to non existent.

This is what most serious reviewers and real-life users have been saying about the A77, the only hit to IQ comes from the modest loss of light, there are no artifacts, contrast problems, 'magenta cast' or ghosting in the latest SLT cameras, much to the chagrin of trolls and doom-mongers no doubt.

So I think sony have got some great optical engineers who know how to make very high quality multi-layer optical surfaces. And that's great. What I don't get is why the AF sensor needs 30% light. The new D800 has 1 stop more sensitive AF system than the D700, if sony did this they could reduce the light loss down to 15% and have identical AF performance.

Once we are close to 10% light loss the difference between DSLT and DSLR become effectively invisible in terms of IQ, as this is the kind of variability you get from exposure metering variations (at least).

-- hide signature --

IQ is not judged exclusively by high iso noise performance

-- hide signature --

IQ is not judged exclusively by high iso noise performance

 mike_2008's gear list:mike_2008's gear list
Canon EOS 5D Sony SLT-A55 Sony SLT-A77 Canon EF 50mm F1.8 II Canon EF 17-40mm f/4.0L USM +5 more
linhn Regular Member • Posts: 332
Re: A77 with and without mirror test

I agree. A little surprised but sharpness seems to be better.

Papasha Senior Member • Posts: 2,158
Three things…

1). These pics are too small and you didn’t tell us what part of the image they represent.

2). From the first pair, image with mirror looks better I would guess due to slightly more favorable focusing (from my experience focusing accuracy is a single most quality-effecting factor in such tests)

3). Working with both SLT and SLR models from Sony side-by-side, I stopped thinking about effects of the mirror when I shoot as for me it’s a waste of time, but if you did spend some time and thoughts on putting this experiment together, I would really prefer to read your personal conclusion anywhere along. I could agree or disagree, but at least I would benefit from your time spent and conclusion reached. You worked with full images, you analyzed them, you had to come up with one at least for yourself? Selfish on my part maybe, admitting…

-- hide signature --

Some of my older work: http://www.pbase.com/papasha

 Papasha's gear list:Papasha's gear list
Sony Alpha NEX-7 Sony SLT-A77 Sony Alpha a99 Sony a7R II Sony DT 16-80mm F3.5-4.5 ZA Carl Zeiss Vario-Sonnar T* +10 more
Dave Oddie Veteran Member • Posts: 4,153
Re: A77 with and without mirror test

FreeRadical009 wrote:

Sony definetely provides better results with RAW. I've seen better results from RAWs across the board of models once you compare them to JPEGs.

I didn't like the idea of messing with RAW, but once I saw all the things I could do and how much it could improve my photography, I went RAW 100%.

That being said, I don't criticize other people for using JPEG only. I used to do the same and I understand why they do it. The fact I shoot RAW does not give me a licence to say other people are idiots for using JPEG. Only pretentious snobs think that way.

I am sure you can get more out of the cameras with RAW than JPEG and with Lightroom (which I have) manipulating RAW's is in some ways no harder than JPEG's.

I routinely check all my photos after I upload them and given the size of the 24mp files I am pretty critical of my shots whereas with the much smaller A100 files I tended to discard few. I will then post process the ones I keep such as cropping, correcting exposure, white balance etc so I post process everything anyway so in theory I should be shooting RAW.

There are two reasons why am am less keen to do so than perhaps I should be:

1. I am not confident I have the skills to get the best out of the RAW's so I beat the camera.

2. Storage. I use Zenfolio mainly as an on-line backup for my photos and it does not support RAW files. Now I could simply post process the RAW files and export the resulting JPEG's but I would rather save the original RAW's and simply make sure I backup the LR catalogue with the edits. Otherwise were disaster to strike I don't have the originals but just my post processed JPEG's.

 Dave Oddie's gear list:Dave Oddie's gear list
Sony SLT-A77 Sony DT 11-18mm F4.5-5.6 Sony DT 16-80mm F3.5-4.5 ZA Carl Zeiss Vario-Sonnar T* Sony 70-300mm F4.5-5.6 G SSM Sony 500mm F8 Reflex +5 more
Keyboard shortcuts:
FForum MMy threads