fredmiranda.com Login
This topic is in the 2011 archive. You can read it and like its posts here. It is read only.

Archive 2011NEX-7 Review by Steve Huff

  
 
ken.vs.ryu
Offline
Details
p.13 #1NEX-7 Review by Steve Huff
the a77's fw was updated weeks after it's release. same thing can happen to the nex.
Oct 28, 2011 at 10:44 PM
Tariq Gibran
Offline
Details
p.13 #2NEX-7 Review by Steve Huff
uhoh7 wrote:
denoir wroteThe problem is of course that the NEX has no idea which lens you've got mounted. The color cast corrections are entirely lens specific. So I think you can forget about any in-camera software fixes there. And it's actually worse than that - since you don't have any lens metadata, applying the correct cornefix profile may be problematic as you won't know which one you should choose (unless you remember exactly which lens you used for each shot that needs correction).

Leica has lens coding, so the camera knows which lens is mounted and thus knows which profile to apply.
This implies the 5n is dealing with CS in a fundamentally different way than the m9. This is certainly not a forgone conclusion. You don't think it is possible for the camera to detect color shift and respond accordingly?

The m9 needs a profile, but the 5n does not--because of sensor design?


Probably a combination of a more modern micro lens design coupled with the fact that the 5n is a smaller APS sized sensor vs FF M9.


Oct 28, 2011 at 10:48 PM
douglasf13
Offline
Details
p.13 #3NEX-7 Review by Steve Huff
uhoh7 wrote:
denoir wroteThe problem is of course that the NEX has no idea which lens you've got mounted. The color cast corrections are entirely lens specific. So I think you can forget about any in-camera software fixes there. And it's actually worse than that - since you don't have any lens metadata, applying the correct cornefix profile may be problematic as you won't know which one you should choose (unless you remember exactly which lens you used for each shot that needs correction).

Leica has lens coding, so the camera knows which lens is mounted and thus knows which profile to apply.
This implies the 5n is dealing with CS in a fundamentally different way than the m9. This is certainly not a forgone conclusion. You don't think it is possible for the camera to detect color shift and respond accordingly?

The m9 needs a profile, but the 5n does not--because of sensor design?


It's hard to say. From a hardware standpoint, it could be that the NEX-5N has better tuned micro lenses, or, I've read one rather well informed guy's opinion that it isn't better micro lenses. Rather, he states that there was a tiny gap between the IR/AA filter pack and the sensor on the NEX-5, in order to reduce appearance of dust on the sensor, and the gap was shortened on NEX-5N. Who knows?


Oct 28, 2011 at 10:49 PM
uhoh7
Offline
Details
p.13 #4NEX-7 Review by Steve Huff
douglasf13 wrote:
As I just mentioned in another forum, uhoh7, I get where you're coming from, but combine this with the other tests we've seen, and it's hard not to be suspicious of the NEX-7 not performing up to the NEX-5N in terms of wide M lenses.

link to other bad looking shots

I'm hoping that we just have incompetent testers on our hands, but I'm getting nervous.

edit* oh yeah, Steve Huff does apparently have production NEX-7, btw.


The guy in your link is the one who was worried about focus shift in his live view camera and threw up those horrible zm 50 shots.

How do we know Steve has final firmware? No dirsepect to any here, just curious.

The very idea Steve has gone off about CS and is oblivious to the smearing is mind boggling. Sony is sending cameras to the wrong people--well nothing new there i guess.
Oct 28, 2011 at 10:50 PM
Tariq Gibran
Offline
Details
p.13 #5NEX-7 Review by Steve Huff
douglasf13 wrote:
denoir wrote
uhoh7 wrote
sebboh wroteit seems clear that the NEX-7 will be worse than the 5N for color shift and smearing, .
What

Correct me, but the M9 firmware is correcting for colorshift---at least that's what they say in Leica land, while the MLs are dealing with rez.

Does Steve have the final firmware? If not zero conclusions can be made from his tests.
The problem is of course that the NEX has no idea which lens you've got mounted. The color cast corrections are entirely lens specific. So I think you can forget about any in-camera software fixes there. And it's actually worse than that - since you don't have any lens metadata, applying the correct cornefix profile may be problematic as you won't know which one you should choose (unless you remember exactly which lens you used for each shot that needs correction).

Leica has lens coding, so the camera knows which lens is mounted and thus knows which profile to apply.
The GXR apparently allows you to adjust corner color correction in the camera menus. In fact, you can individually adjust each corner:

http://www.grblog.jp/2011/08/09/kyohda/CS02.jpg


That's really cool and would lead me to believe we will be seeing a future, dedicated Ricoh M mount camera given the extent that they have gone to in designing this.
Oct 28, 2011 at 10:51 PM
douglasf13
Offline
Details
p.13 #6NEX-7 Review by Steve Huff
Tariq Gibran wrote:
uhoh7 wrote
denoir wroteThe problem is of course that the NEX has no idea which lens you've got mounted. The color cast corrections are entirely lens specific. So I think you can forget about any in-camera software fixes there. And it's actually worse than that - since you don't have any lens metadata, applying the correct cornefix profile may be problematic as you won't know which one you should choose (unless you remember exactly which lens you used for each shot that needs correction).

Leica has lens coding, so the camera knows which lens is mounted and thus knows which profile to apply.
This implies the 5n is dealing with CS in a fundamentally different way than the m9. This is certainly not a forgone conclusion. You don't think it is possible for the camera to detect color shift and respond accordingly?

The m9 needs a profile, but the 5n does not--because of sensor design?
Probably a combination of a more modern micro lens design coupled with the fact that the 5n is a smaller APS sized sensor vs FF M9.



If Luka doesn't mind, here is an example of his M9 with the ZM 18 and no correction:

http://peltarion.eu/img/comp/nex-m9/A_zm18_M9.jpg
Oct 28, 2011 at 10:51 PM
douglasf13
Offline
Details
p.13 #7NEX-7 Review by Steve Huff
uhoh7 wrote:
douglasf13 wroteAs I just mentioned in another forum, uhoh7, I get where you're coming from, but combine this with the other tests we've seen, and it's hard not to be suspicious of the NEX-7 not performing up to the NEX-5N in terms of wide M lenses.

link to other bad looking shots

I'm hoping that we just have incompetent testers on our hands, but I'm getting nervous.

edit* oh yeah, Steve Huff does apparently have production NEX-7, btw.
The guy in your link is the one who was worried about focus shift in his live view camera and threw up those horrible zm 50 shots.

How do we know Steve has final firmware? No dirsepect to any here, just curious.

The very idea Steve has gone off about CS and is oblivious to the smearing is mind boggling. Sony is sending cameras to the wrong people--well nothing new there i guess.


I hear you. Weird people to give the first cameras to. Steve claims that Sony told him that it was production firmware. Seeing how he's a Leica tester for the most part, I'd have a hard time believing that they'd give him something that performs so badly with wide M lenses. Odd.


Oct 28, 2011 at 10:53 PM
bluetsunami
Offline
Details
p.13 #8NEX-7 Review by Steve Huff
douglasf13 wrote:
denoir wrote
uhoh7 wrote
sebboh wroteit seems clear that the NEX-7 will be worse than the 5N for color shift and smearing, .
What

Correct me, but the M9 firmware is correcting for colorshift---at least that's what they say in Leica land, while the MLs are dealing with rez.

Does Steve have the final firmware? If not zero conclusions can be made from his tests.
The problem is of course that the NEX has no idea which lens you've got mounted. The color cast corrections are entirely lens specific. So I think you can forget about any in-camera software fixes there. And it's actually worse than that - since you don't have any lens metadata, applying the correct cornefix profile may be problematic as you won't know which one you should choose (unless you remember exactly which lens you used for each shot that needs correction).

Leica has lens coding, so the camera knows which lens is mounted and thus knows which profile to apply.
The GXR apparently allows you to adjust corner color correction in the camera menus. In fact, you can individually adjust each corner:

http://www.grblog.jp/2011/08/09/kyohda/CS02.jpg


This is extremely cool. On the software side Ricoh seems like the Anti-Sony. Feature rich, yet easily accessible. I'd love it if you could store each profile in the camera tuned for certain problem lenses and quickly choose the profile in some directory yourself. Seems like a real smart camera in regards to design on many levels.
Oct 28, 2011 at 10:57 PM
uhoh7
Offline
Details
p.13 #9NEX-7 Review by Steve Huff
douglasf13 wrote:
I hear you. Weird people to give the first cameras to. Steve claims that Sony told him that it was production firmware. Seeing how he's a Leica tester for the most part, I'd have a hard time believing that they'd give him something that performs so badly with wide M lenses. Odd.


The picturedesk guy is even worse. Rock Huffwell's mother law I think. Can no one make an equidistant test shot?

But, maybe it's all true. We will know fairly soon.

Hard to imagine. Now I have to buy that 350USD viewfinder and the 5n?

I will say the nex 7 was a bit big for me but then again, this came in the mail today:

http://farm7.static.flickr.com/6099/6289701217_421b0aa4c4_z.jpg



On another note, my new 1937 sonnar 50/1.5 is in the hands of the mad master, Henry Scherer.

"Hello Charlie,

I have an opening while waiting for paint to cure while working on a Contax III and so have moved forward with your lens and it is completely disassembled. It is very dirty but very fine. The lens elements are in perfect condition and so my guess is it's going to be a 10 when its done. It's distinguished by very fine surface oxidation of the front and rear lens elements. This shows it's never been cleaned. Whoever owned it previously cared for it very much. This surface oxidation acts like coating and significantly improves the lens so if I were you I'd invest in a UV filter and would never clean this lens. This surface oxidation is very rare and highly desirable."

I wrote back saying WHAT? (nicely of course, since he has my glass, hehe) I thought oxidation was a BAD thing?

"I can't possibly provide a complete lecture in the history of glass making but here's what you need to know in a nut shell. The history of lens glass is long and begins with plain glass made at very low temperature to make it very clear and very soft so it could be easily ground into lenses by hand through manual labor. A process of continual development and improvement took place in lens glass making it with higher and higher temperatures and it becoming harder and harder as mechanical methods of lens making were developed. Glass corrodes just like everything else. Softer glass corrodes faster and with a coarser grain and harder glass corrodes slower with a finer grain. Finally, Zeiss developed modern optical glass and improved it to the point that it corrodes so slowly and so finely that the corrosion actually improved the performance of the lens because it was thin enough and fine enough to act like a dichroic filter. Zeiss noticed this and began a program of development to duplicate this effect and this led to modern coatings. The corrosion on the surfaces of your lens results in an improvement. The corrosion on the surface of a lens made in 1900 results in the ruin of the lens.

The reason there is virtually no Colonial Era American glass existing today is that it simply corroded into sand because it was made under such low temperatures and was so soft and susceptable to corrosion. Low temperature glass can be made to be very brilliant. Tiffany experimented with it in some of his lamps and these lamps are distinguished by the fact that the glass pieces closest to the light bulb is still present whereas the pieces further away from the bulb have corroded into sand."

So I will just to console myself with these beauties, since my precious Nex-7 is nothing more than a trumped up trollup rudely violated by Rock and his mother in law

PS If anyone wants a 15 heliar, there is one at RFF now and the nice guy offered it to me for 340 shipped. LTM, clean W/O VF.




Oct 28, 2011 at 11:09 PM
kosmoskatten
Offline
Details
p.13 #10NEX-7 Review by Steve Huff
bluetsunami wrote:
This is extremely cool. On the software side Ricoh seems like the Anti-Sony. Feature rich, yet easily accessible. I'd love it if you could store each profile in the camera tuned for certain problem lenses and quickly choose the profile in some directory yourself. Seems like a real smart camera in regards to design on many levels.


You can actually store data for a few lenses, but I've forgotten if you store it on the SD card or in camera.



Oct 29, 2011 at 12:06 AM
michaelwatkins
Offline
Details
p.13 #11NEX-7 Review by Steve Huff
douglasf13 wrote:
Seeing how he's a Leica tester for the most part, I'd have a hard time believing that they'd give him something that performs so badly with wide M lenses. Odd.


Even as NEX-7 prospects cancel their pre-orders, probably, hopefully, somewhere in Sony USA's corporate offices a junior product marketing guy is having his wings clipped even now.

... anyway ...

Has there not yet been done any decent testing against a flat field by anyone? Has smearing on the edges been confirmed or refuted yet? I can live with simple colour shifts if the image quality isn't otherwise being compromised.
Oct 29, 2011 at 12:07 AM
corposant
Offline
Details
p.13 #12NEX-7 Review by Steve Huff
From a (almost) completely detached perspective (I have been playing around with the 5N for about a day now), I have to say the amount of hand-wringing here is really nothing short of incredible.

It's reminiscent of anxious parents/relatives in a hospital waiting for their child to be born. I feel like I have to play the role of the hospital liaison and tell you..."when it finally happens, everything will be alright!"

edit: www.rockhuffman.com is still available.
Oct 29, 2011 at 12:29 AM
michaelwatkins
Offline
Details
p.13 #13NEX-7 Review by Steve Huff
Some prospective parents here would simply like to know that their almost-born will come with all 10 toes in place; we don't care too much if the ones on the outside edge are a little pinker than the ones in the centre.
Oct 29, 2011 at 12:33 AM
sebboh
Offline
Details
p.13 #14NEX-7 Review by Steve Huff
corposant wrote:
edit: www.rockhuffman.com is still available.


no, i don't trust rock huffman. we want rock huffwell.
Oct 29, 2011 at 12:49 AM
ken.vs.ryu
Offline
Details
p.13 #15NEX-7 Review by Steve Huff
michaelwatkins wrote:
Some prospective parents here would simply like to know that their almost-born will come with all 10 toes in place; we don't care too much if the ones on the outside edge are a little pinker than the ones in the centre.


http://en.wikipedia.org/wiki/Throw_out_the_baby_with_the_bath_water
Oct 29, 2011 at 01:10 AM
xbarcelo
Offline
Details
p.13 #16NEX-7 Review by Steve Huff
My guess with the approach Ricoh is taking is that this only apply to jpeg, right? I mean, if we want our raws cleaned up, I suppose that we need to use Cornerfix. Then we're basically in the same position except if we are jpeg shooters. Nothing gained, some (12Mp) lost…
Oct 29, 2011 at 05:35 AM
ken.vs.ryu
Offline
Details
p.13 #17NEX-7 Review by Steve Huff
1080p/24p videos, easy lens adaptability, hires evf, focusing at the working aperture, and 0.02 shutter lag is also lost.
Oct 29, 2011 at 06:38 AM
slungu
Offline
Details
p.13 #18NEX-7 Review by Steve Huff
I don't think everything is lost, only that for the main purpose of such a cam, esp. at that price, would be to be able to drop the SLR with its big and heavy lenses and go with the small and light RF lenses. If this is not the case, you still have a great allrounder that would do great with SLR lenses of any type and would still be small by itself, but not as a kit. If this is ok for the buyer is up for him to decide.

Regards, Stefan
Oct 29, 2011 at 06:46 AM
xbarcelo
Offline
Details
p.13 #19NEX-7 Review by Steve Huff
No, what I meant is that if (and that's a big if) there's no smear in the corners on the NEX7 (also, now I think of it, is there any on the Ricoh? Just wondering), then it really doesn't make any difference whether we use Ricoh or NEX, in terms of wide angle rangefinder lenses, because, if we shoot raw, we'd have to use cornerfix one way or the other.

My assumption here (and it might be wrong) is that if they thought of this solution is because there is a problem with colour shift on the corners. In the NEX 5n there's no problem or less of a problem, and hence you don't need to worry about this. In any case, it'd be nice if there was a solution from ANY brand that was great with rangefinder lenses, good ergonomically and relatively cheap. Then we'd all be happy, wouldn't we?

Cheers,

Xavier

Edited on Oct 29, 2011 at 09:56 AM · View previous versions
Oct 29, 2011 at 09:55 AM
healyzh
Offline
Details
p.13 #20NEX-7 Review by Steve Huff
ken.vs.ryu wrote:
the 15mm is a bad lens

leica tried to correct the red edge with a fw update
Earlier quoteFirmware update 1.162 features a revision of sensor homogeneity for wide-angle lenses. The additional color calibration noticeably reduces chromatic differences between the center and the corners of an image, especially in critical shooting situations. Additionally, the software processes of the Leica M9 have been optimized, significantly reducing the “red edges effect” and improving the menu’s Italian translation.
here's a recent photo with 15mm and the m9 - i'd guess that he has the fw update.
m9 with 15mm


I would have to disagree about the 15mm being a bad lens, I have the older Voigtlander 15mm, and took the "recent photo" you linked to. I can also confirm that you would be wrong in that guess. That was taken with the 1.116 Firmware. I've yet to update my M9.

The 15mm is not a great lens on the M9 (neither is the 28mm Voigtlander f/2 Ultron), but it's not as bad as some people make it out to be. Of course at the same time, it's not a lens that I take seriously, it's for having some fun.

There is some red edging in the shot that I posted, it's light enough I honestly didn't notice it until a friend pointed it out. I want to see what I can do in post processing to correct it. As you might guess that photo already had some pretty serious post-processing.
Oct 29, 2011 at 09:56 AM
FM ForumsLeica & Alternative GearJoin Upload & Sell
  
 

You are not logged inLog inRegister