Sony a7s - highest Professionally usable ISO?

[Telecaster]

There's another 55mm lens, the Otus, that has very little character, too. I think that's a feature, not a bug.

I like having both options, character & lack thereof, available. That's one of the advantages of the A7 series…lots of (sometimes accidental) optical gems from the past at your disposal.

-Dave-

[Jim Kasson]

1) Isn't it pretty common to see image degradation in M-lenses in the A7r, thought perhaps to be due to the cover glass?

That is certainly correct, but the comparison I linked to was the 50 'lux on the M240 vs the Zony 55 on the a7. Both 24MP, both FF. The Leica body has no AA filters, tho'. The 50 was designed for no cover glass at all, not the thin M240 glass, and the Zony for exactly the cover glass it has in any of the a7 cameras.

2) I truly have no idea where you focused the -lux.

On the grape vines. The M240 LV is not as accurate as the a7 LV, but I could focus wide open at f/1.4 with the Leica lens. Because it's a RF lens, I don't expect much focus shift upon stopping down, and I haven't seen much. The Zony was focused at f/1.8. Don't forget you're looking at a 2x image blowup in both cases.

Was thinking of renting an OTUS 55 and comparing it to my 50-cron on the A7s

Here's a comparison of the 55 Otus and the 50 'lux on the a7S:

http://blog.kasson.com/?p=6357
http://blog.kasson.com/?p=6382

One thing for sure, there are very significant differences between the 3 items in the "A7" designation

Indeed.

Jim

[allegretto]

That is certainly correct, but the comparison I linked to was the 50 'lux on the M240 vs the Zony 55 on the a7. Both 24MP, both FF.
Jim

Now I'm truly concerned. Have shot many an image with an M240 with both my 50's, a -cron and a Noctulux and would be very disappointed seeing that image in my monitor.

Again, no insult intended but there that's a bit amazing. I must say that the LV of the M240 is no place for critical focus. The outboard EVF is is far better, especially with mag. But for a 50 I would always use the RF if possible. Nothing in my experience is as accurate and if you're shooting wide open, especially a Noctulux the light is quite low so sometimes I can't use the RF and default to the EVF. Never ever LV, even with mag though... they end up looking like that far too often

OK, checked out the OTUS vs. -lux 50.

- I have never shot with an OTUS so cannot comment completely on this particular test from that standpoint.

- the 1.4 -lux is not the sharpest of the 50mm m offerings. Better the -cron. I won't hold out for the new APO -cron since it's very rare and even if it was in stock I'd have serious considerations about plunking down N of $8K+ for that package

- Finally, it occurs that the -lux images are once again OOF. Sorry, but I expect much more from the M-glass than your test demonstrates.

Turns out I've done some head to head with the A7s and the M240 with a couple of M-lenses. Will look around and try to figure how to post to this site successfully.

No question of your integrity and knowledge, but that's just OOF until proven otherwise to me...

[Jim Kasson]

I must say that the LV of the M240 is no place for critical focus. The outboard EVF is is far better, especially with mag.

By LV, I meant with the semi-brain-dead rebranded Olympus EVF.

Jim

[allegretto]

By LV, I meant with the semi-brain-dead rebranded Olympus EVF.

Jim

Sorry, posted over you. Yeah the Oly EVF is no Sony item... but RF is RF for a reason. When I shoot a 50 or the 90 -crons I'm amazed by how thin the true DoF is and how much difference a few mm can make

[allegretto]

use of the problems wit the A7s and M-adaptation is that you get NO LENS information with the Voiglander or Meatbones adapters

I have some photos, but cannot commit to them since I could be in error. I think not, but it's better for me to just shoot a new batch tomorrow in the light. I have a favorite target that seems to get to the bottom of focus, color cast and even some much loved DR

For the record, I will shoot M-lenses on both the M240 and the A7s (returned the loaner today and my B&H care package arrive almost simultaneously) So I have a brand new A7s.

I also have some great Zeiss EF- lenses that really rival anything even though they are not OTUS (21mm 2.8/ 100 mm f2 Makro-Planar) but the Metabones Mk IV for EF- to E-mount was out of stock but i may gt mine this week later

But I will be testing M-lensee on both cameras at 200 ISO wide open. Will also go to 3200 or so just to look at what happens. If you have any other requests for a side-by-side please let me know. I love this kind of stuff. No spec sheets or baloney... just look at the images and make the call

[Jim Kasson]

Finally, it occurs that the -lux images are once again OOF. Sorry, but I expect much more from the M-glass than your test demonstrates.

There are a lot of variables here: demosaicing, sharpening, JPEG compressiong, etc. If the M-glass images were OOF, wouldn't there be a dramatic improvement as you stop down?

One thing I did notice about the 'lux 50 and LV focusing: it's really sensitive. Just removing your hand from the focusing ring the wrong way can cause the image to go OOF. I blame the short throw of the Leica lenses. By comparison, it's a joy to focus the Otus, with its long throw.

But I will be testing M-lensee on both cameras at 200 ISO wide open. Will also go to 3200 or so just to look at what happens. If you have any other requests for a side-by-side please let me know. I love this kind of stuff. No spec sheets or baloney... just look at the images and make the call

You can't use too stiff a tripod. Unless you always shoot wide open, I think a full aperture series would be a good idea. Remember to turn off LV before tripping the M240 shutter release (I use the self-timer) to avoid shutter shock. Use EFCS on the a7S (and the a7, if you;re going to test it). If you use an a7R, avoid shutter speeds between 1/250 and 1/15. If I'm interested in field flatness, I just focus in the middle and then look in the middle and the corners. Otherwise, I focus in the corners, too, although you can't do that with the M240. Make sure your adapter is flange-parallel. If you have tripod heads that support this move, a quick way to test is to make two exposures of a planar subject and invert the camera in between. An ISO 12233 chart makes a good target for this. You can use a mirror to make sure your camera is square to the target. That's not a bad idea in general, but it doesn't work well with natural, 3D, targets.

I apologize if you know all that already.

I'm looking forward to seeing your results.

Jim

[Manoli]

If I'm interested in field flatness, I just focus in the middle and then look in the middle and the corners. Otherwise, I focus in the corners, too, although you can't do that with the M240.

Another practical and simple way to test for field flatness and corner resolution is to focus on a horizon line and then tilt the camera sideways so the horizon extends from lower left-hand corner to upper right-hand corner (or vice versa). First saw this in an earlier test series by another LuLa contributor - jerome_m, (if I remember correctly) so credit to him for the idea.

[jjj]

Another practical and simple way to test for field flatness and corner resolution is to focus on a horizon line and then tilt the camera sideways so the horizon extends from lower left-hand corner to upper right-hand corner (or vice versa). First saw this in an earlier test series by another LuLa contributor - jerome_m, (if I remember correctly) so credit to him for the idea.

Neat and simple.   

[Jim Kasson]

- Finally, it occurs that the -lux images are once again OOF. Sorry, but I expect much more from the M-glass than your test demonstrates.

I went back to the M240//lux Ps stack and zoomed in on the center of the f/5.6 image. I see pixel-level details, which wouldn't happen if the image were OOF.



Jim

[Telecaster]

Yeah the Oly EVF is no Sony item... but RF is RF for a reason. When I shoot a 50 or the 90 -crons I'm amazed by how thin the true DoF is and how much difference a few mm can make

Just to note: the EVF in Oly's E-M1 (made by Epson?) is in a different league to the M240's "Visoflex" finder. IMO it's at least as good as the Sony's, better contrast-wise in bright light. Leica gets major demerits from me on both the EVF and the 240's inability to move the magnified point-of-focus around in the finder. Agreed on RF focusing, at least for lenses 50mm & under.

-Dave-

[Jim Kasson]

Just to note: the EVF in Oly's E-M1 (made by Epson?) is in a different league to the M240's "Visoflex" finder. IMO it's at least as good as the Sony's, better contrast-wise in bright light. Leica gets major demerits from me on both the EVF and the 240's inability to move the magnified point-of-focus around in the finder.

Right you are. And the EVF picked by Leica wasn't even the best Oly shoe-mount EVF available at the time.

Agreed on RF focusing, at least for lenses 50mm & under.

I hear you , but my eyes are too old to love the RF experience. I do find the M240 RF an improvement over the M8 and M9 ones, though.

Jim

[JV]

Leica gets major demerits from me on both the EVF and the 240's inability to move the magnified point-of-focus around in the finder. Agreed on RF focusing, at least for lenses 50mm & under.

Add the EVF of the Fuji X-T1 also to the list of much better EVF...

With the next M Leica really needs to step up and improve big time on the EVF and on the focusing aids other than RF.

You already see a pattern right now where people complement their M8 or M9 with a Sony.  They don't contemplate the CMOS M.  That trend is only going to get stronger IMO.

You see the same with other models like the X Typ 113.  Sexy camera but really, no built-in viewfinder?  The Fuji X100S has two... and costs half of the price...

IMO it is really simple.  If Sony delivers on the lenses and improves on the bodies with the next generation they will take most of the mirrorless market and make serious inroads into the DSLR market.  

[allegretto]

Ack, who's saying that!? The 35 & 55mm FE lenses are both very good indeed, with the 55 well into excellent territory. The 24–70mm zoom is just average, though. This is on an A7r. On the s I can't imagine them being worse.

-Dave-

Hey Dave. Had time to form my own opinion on this with the Sony 55mm 1.8/24-70 4.0/70-200 4.0

They are fine for all but the most demanding circumstances

My 6d with a 50mm 1.4/24-105 f4/70-200 4.0 is that the Canon is significantly but perhaps not so importantly... faster in AF acquisition. BY a noticeable amount. Again though, not many situations will use that difference.

Amazing what the 6D does with so many "inferior" parts and systems. Really is. That's another very underrated camera. Some feel the 5D3 somewhat superior. Construction... yes. Output... no.

[Jim Kasson]

Hey Dave. Had time to form my own opinion on this with the Sony 55mm 1.8/24-70 4.0/70-200 4.0

They are fine for all but the most demanding circumstances.

I haven't tested the 24-70, but in my experience the 55 and the 70-200 are in different area codes. The 70-200 is about the same, and maybe a bit better than my current-rev Nikon 70-200 f/2.8. My Zony 55 is amazing. I've only tested the one sample, though.

By the way, on the subject of multiple sample testing, I wish I could do it, but it's just not practical for me. I do have two copies of the Sigma 50mm ART lens, and they turn in materially identical optical performance.

Jim

PS 1000!

[Telecaster]

My 6d with a 50mm 1.4/24-105 f4/70-200 4.0 is that the Canon is significantly but perhaps not so importantly... faster in AF acquisition. BY a noticeable amount. Again though, not many situations will use that difference.

Yes, the Sonys aren't the speediest cameras AF-wise. And PD-AF systems do prioritize speed over accuracy. (D-SLR dig required to meet monthly quota.    ) I'll be much happier with the FE lineup when I have 28 & 85mm lenses to complement the 55 & 35mms. Until then it's mostly manual focus for me.

-Dave-

[andyptak]

A lot of interesting information here, but to come back to my initial question - ya'll remember that don't you?

After a bit of testing and some guesswork, I think I'd have to push the ISO to at least 3200. Can the 7s cut it and produce decent work at that range? Obviously it would have to be cleaned up in post, but is that even possible with this camera? Thanks.

[Jim Kasson]

A lot of interesting information here, but to come back to my initial question - ya'll remember that don't you?

After a bit of testing and some guesswork, I think I'd have to push the ISO to at least 3200. Can the 7s cut it and produce decent work at that range? Obviously it would have to be cleaned up in post, but is that even possible with this camera? Thanks.

IMHO: Yes! Better than the D3s at that ISO, if you've used that camera.

Jim

[allegretto]

A lot of interesting information here, but to come back to my initial question - ya'll remember that don't you?

After a bit of testing and some guesswork, I think I'd have to push the ISO to at least 3200. Can the 7s cut it and produce decent work at that range? Obviously it would have to be cleaned up in post, but is that even possible with this camera? Thanks.

 Oh yes, I recall

at 3200 you need do nothing. 6400 too . Not till 12800 do you have to start lifting in PP. A good 4 -stops on your MF

Further, the files are just... pretty. Big pixels do that. My D4 made files that killed the 7100 in spite of the latter's denser pack. it wasn't noise either. It killed at 1600 too. Very sweet. Even ate my Fuji X-Pro which I liked the look of.

What I like about Sensorgen.info is that they give you very complete ISO/DR files as some but they also give you e- for noise and sat capacity. Note the A7s just blitzes the field and I think that may contribute to what I "see'