Schneider 60XL on 60mpx chip (Credo60/IQ160 etc) samples

[torger]

Well, I received my 70 HR W and can say that it eats my 55mm APO-SD alive!

Interestingly, I can shift the back horizontally by 17mm AND drop the back by at least 12mm and still not see the hard vignette. This is when focusing at about 20m distance and at f11. Brilliant!

The 70HR has 100mm image circle, that is 10mm more than the shorter focal lengths in the series so it does provide a quite large range of movements even for 645 fullframe.

(I've talked with Paula and my SK60XL lens is going back to Schneider, I shall do some additional tests first though.)

 

[tjv]

An update:

I've been told by Paula that she hopes her stocks of the 60XL will hold for another 10 months, current demand continuing.

Either way, I can't afford to buy the 60XL right now, especially considering the courier has lost my 90 HR-W that was being sent for service :-(

In the meantime, I have ordered the centre filter for my 55mm APO-SD to see how that improves things. I notice it's 2.5 stops! Indicates, as the LCCs also suggests along with final inspection of images, that noise on my pano stitches is really killing detail. I'll update this when I get a chance to test...

 

[torger]

Gaahh! Thanks for reminding me, I haven't sent my decentered SK60XL back for fixing yet. I was just about to make the final tests they required, and then other stuff came inbetween. Maybe this week....

 

[tjv]

An update.
I've had the CF for my 55mm APO-Sironar Digital for a while now, but haven't had time to really put it to the test except for a quick demo in the back yard. Linked here are some LCC samples (before LCC corrections applied) and corresponding histograms, plus a couple of reference shots to see the noise difference on the far left of a frame shifted 17mm, with 5mm rise to the back. Both crops are 200% and around centre of the edge of frame (if that makes sense?) Anyway, the noise difference is quite noticible. Not a magic bullet, especially conidering the 2.5 stop loss of light, but should make for far better results in some situations.

For reference, these images are shot on my Credo 60 with my Linhof Techno, there was +50 dialed into the shadow slider, +15 Highlights, +10 contrast, +25 brightness (although +20 for the non-CF shot as the light changed). Colour was quite different between the two and the non-CF shot was warmer, so I cooled it down with -10 on the blue channel in levels. Sharpending and NR were as follows:

Amount: 125
Radius: 0.6
Threshold: 1.0

Lumanance: 15
Details: 35
Colour: 35
Single Pixel: 1

Focus was about 4 meters from camera (which helps vignetting somewhat, so probably not the best test setting). I did a small amount of extra sharpening in LR after C1 conversion (I'm not great with C1.) The forum downsamples for viewing, so control click on them and open in new tab so you can view at 100%. Again, the screen shots are viewed at 200% to magnify differences.

Full image, pano stitch. I should mention too that I shot a contrasty scene on purpose, as I'm often battling to protect highlights and lift shadows, which is not really the domain of best use for a CCD sensor:

 

[tjv]

I've just been found out that for some reason the forum software doesn't treat PNG files very well, so that explains a little why the crops above aren't showing up as detailed as I see them on screen. If anyone would like me to send them the files directly, just let me know. Either way, the CF makes a huge difference all be it at the cost of 2.5 stops of light.

 

[Pelorus]

That last shot in Post 64 must be an NZ backyard surely tjv?

 

[tjv]

That last shot in Post 64 must be an NZ backyard surely tjv?

Sure is! Last days of Summer here in Christchurch...

 

[Pelorus]

Sure is! Last days of Summer here in Christchurch...

It just seemed so NZ that I thought it must have been. Christchurch and not a sign of liquefaction in the back yard

 

[tjv]

It just seemed so NZ that I thought it must have been. Christchurch and not a sign of liquefaction in the back yard

It only took about three weeks to get rid of it... Not helped by the car loads of young men driving past, hanging out of windows and photographing us as we shoveled it all out onto the road for councel pickup. There were many parasites in those months after the earthquakes.

 

[torger]

I finally got around sending back my SK60XL lens to Schneider-Kreuznach to have a look at it.

After analyzing it they claim that it performs as it should, but found that the lens plate is askew. However only 8um left-to-right and I doubt that would cause the somewhat weak performance I get.

The good news is that the check was only 90 euros, which I find perfectly reasonable.

I don't really know how to proceed now, but I'll ask Paula for advice.

 

[MGrayson]

My 35XL developed a soft right side. Off it went to Cambo. They sent it back with a new mounting plate ... and it performed worse than before. Back again, this time to SK, they replaced the shutter. Lens was ok not perfect afterwards. I was strongly not impressed.

But the 60XL should be razor sharp.

Well, I'll have to make do with Leica S glass - until a focus motor fails and Leica takes 6 months to replace it

--Matt

 

[tjv]

Seems odd, especially after looking at the sample RAW files you sent me. They weren't very good and in fact I'd say my 55mm Rodenstock performs better. Actually, now that I have the centre filter for it, I have almost zero interest in upgrading to the SK60XL. I'm sure a good sample would be better than my RS55mm, but as others have indicated above, just how much better in real world is hard to tell. The only thing that annoys me is losing 2.5 stops of light to compensate for light falloff.

Could your problem be due to parallism issues with your Techno, exacerbated with large shifts? Have you seen issues with other lenses? (I'm guessing not.) I thought I had some problems with my 90mm HR-W a while back (before it was stolen by some low life courier driver on the way back to Paula for a trade,) but it turned out I'd mounted the back on the sliding back slightly off kilter. That was a really sharp lens, all the way out to the edge of the image circle. I wish I still had it, although hardly ever need that FL.

I finally got around sending back my SK60XL lens to Schneider-Kreuznach to have a look at it.

After analyzing it they claim that it performs as it should, but found that the lens plate is askew. However only 8um left-to-right and I doubt that would cause the somewhat weak performance I get.

The good news is that the check was only 90 euros, which I find perfectly reasonable.

I don't really know how to proceed now, but I'll ask Paula for advice.

 

[torger]

Seems odd, especially after looking at the sample RAW files you sent me. They weren't very good and in fact I'd say my 55mm Rodenstock performs better. Actually, now that I have the centre filter for it, I have almost zero interest in upgrading to the SK60XL. I'm sure a good sample would be better than my RS55mm, but as others have indicated above, just how much better in real world is hard to tell. The only thing that annoys me is losing 2.5 stops of light to compensate for light falloff.

Could your problem be due to parallism issues with your Techno, exacerbated with large shifts? Have you seen issues with other lenses? (I'm guessing not.) I thought I had some problems with my 90mm HR-W a while back (before it was stolen by some low life courier driver on the way back to Paula for a trade,) but it turned out I'd mounted the back on the sliding back slightly off kilter. That was a really sharp lens, all the way out to the edge of the image circle. I wish I still had it, although hardly ever need that FL.

It's ongoing, we'll see how it ends up. I've given Schneider-Kreuznach a bit more material. Linhof Studio is very supportive as always and has offered to lend me another 60XL to compare, but I'm in no particular hurry and we'll see what Schenider responds to the test images I've now sent them.

I don't have problems with my other lenses, so I'm quite sure it something with this one. Schneider did discover some skewing in the lens plate, but only 8 micrometer left-to-right and that would not cause the type of performance issue I see.

 

[torger]

It's ongoing, we'll see how it ends up. I've given Schneider-Kreuznach a bit more material. Linhof Studio is very supportive as always and has offered to lend me another 60XL to compare, but I'm in no particular hurry and we'll see what Schenider responds to the test images I've now sent them.

I don't have problems with my other lenses, so I'm quite sure it something with this one. Schneider did discover some skewing in the lens plate, but only 8 micrometer left-to-right and that would not cause the type of performance issue I see.

Despite the added screenshots Schneider-Kreuznach says the same thing, nothing wrong with the lens. Despite that I insisted on getting a comment on my screenshots if it represents acceptable image quality from that lens or not they did not answer to that. I've insisted once more and see if I can get a comment from them. I simply want to know if they believe there is an error in my camera system, or if they think that my lens is performing "within spec" despite this result, and they haven't provided me with that information.

I also asked them if they had a paid service to simply take the lens apart and remount it again with highest possible precision and said I was prepared to pay for that even if they didn't consider the lens to be in error. They did not offer such a service, or well avoided answering that question too.

In any case I'm getting the lens back now without any action and I will then redo my tests and try a new lens plate, mount it upside down etc, tests that I haven't done prior to sending it in. It may still be some issues with my camera, although I tried to rule it out the first time around by testing my other wide angles and not having issues with them.

This is my first experience with Schnieder-Kreuznach repair services and so far I'm not impressed, even if the error is on my side they haven't been helpful at all in providing information I would need to solve the problem.

Fortunately I bought this lens new at Linhof Studio and they're very supportive so I'm confident that it will be resolved in one way or another in the end. I'm getting a new lens plate and a proper lens wrench to make some proper testing to rule out any lens plate/camera issues. If I'm lucky it really is the lens plate, and currently that seems like the most likely reason for the problem as it would explain both that the optics is without error and that I have no issues with my other lenses. However the error doesn't look as a lens plate skew to my amateur eyes, and I was hoping Schneider-Kreuznach could have commented on that, but unfortunately not :-/

 

[tjv]

Hi Anders,
This isn't good, I'm sorry to hear it. I had a similar experience with Rodenstock regarding my 90mm that was exhibiting an enormous amount of flare on overcast days. They said nothing was wrong with it and it turns out they were probably right as the 90mm is notorious for crazy flare (actually, all the retrofocus RS lenses are too,) and more diligent use of a hood fixed the problem in most cases. Paula at Linhof Studio was excellent in supporting me with it, even shipping me a Lee universal hood free of charge to do some of my own testing. Great service.
Let us know how your test get on...
T

 

[torger]

I got a reply from Schneider regarding looking at the images, and they simply said that they never look at images, and stated again that the lens performs as it should, or actually "even better". I do hope they're right and something is wrong with the lens plate or similar.

I sort of can understand that they don't look at images provided by users as it's so many parameters in it they can't control, but for me as a user it's frustrating of course. I guess they're afraid that users will turn any comments/guesses against them, in all they feel very defensive in their communication. Oh well, less-than-excellent service from manufacturers when it comes to repairs is more of a rule than an exception, and then it's nice to have a dealer like Linhof Studio in between. In this case I'm actually getting a loaner lens from them, and I even didn't suggest it.

With a lens wrench, new lens plate, and a loaner SK60XL I will have all possibilities to really figure out where the problem sits. I'll do my very best. It will take some time until the stuff gets to me here, but hopefully I can start testing early next week.

 

[dchew]

I had a similar experience with my 100hr. I swear it was decentered. Good at f/11 but not good at f/4. You could focus so one side was good but the other wasn't, then focus the other side. But never both sides.

Rodenstock looked at it and said it was fine. Ugh.

Dave

 

[torger]

I haven't got back my lens yet, but I got a loaner SK60XL today which I've tested shooting +/-25mm sideways shift at f/11. I get better results than I got with my lens, but still the right side was blurrier (or rather it focuses closer), so I suspected something skewed in the camera. I then remounted the lens upside down on the lens board (in place without refocusing) and then the left side got blurrier and the right became good.

This makes the lens the prime suspect as if the camera would be skewed the error would be the same with the lens mounted upside down.

I've attached a 100% crop (normal USM sharpening) showing the difference between the lens mounted upright or upside down at the 100mm image circle. The sharper result I'd say is really good for f/11 at ~100mm, while the other is not great but I expect it to clean up to okay when shooting at f/16 as I would do with this extreme amount of shift. The lens seems to be much better than mine, but I shall make a side-by-side test when I get it back.

I'll give the final verdict after f/16 shot, but I expect that I will consider this lens to be fine. I'm no extreme perfectionist and have come to realize that even at this level there is some sample-to-sample variation and I'd go crazy if I would only accept perfect no difference between left-and-right at extreme shift because then I don't think any lens would pass.

(I rarely go beyond 15mm shift in my shooting style)

 

[torger]

Indeed at f/16 the bad and good side are almost identical, so this loaner lens seems to be okay.

I'm not sure what expectations one should have though. I've got the feeling that "sharp" lenses are always supersharp in center (which this is), but performance at extreme shifts can vary to a quite large extent. In 135 format sample variations are quite well documented (lens rental stats), it would be great to have the same stats for tech lenses. I suspect that precision of these lenses are not as good as their reputation. What do you think?

 

[torger]

Attached a shot showing the loaner lens performance summarized. As you can see it's very sharp in center, and at 100mm image circle it's sharper on one side than the other, clearly seen at f/11. However at f/16 the difference is not too large. The loaner lens is performing better than my copy.

I also dug out a SK60XL test with the IQ180 probably made by a dealer (Doug perhaps), and it shows the same, that is at f/11 one side is considerably sharper than the other at 100mm image circle. So all three SK60XL lenses I've pixel-peeped all show a clear left-to-right difference at f/11, which leaves me with the conclusion that it should be expected with this lens, but that at f/16 the good copies should not differ much.

TJV's Rodenstock 55 shots earlier in this thread they show crops at about 80-85mm image circle at f/16. I don't show that in my attached screenshot so you have to trust me on this: the SK60XL on its good side makes a much sharper at f/11, but on the bad side it's about the same as the Rodenstock at f/16. Stopping down the SK60XL to f/16 the differences evens out almost completely, but the result is only marginally better than the Rodenstock 55. Possibly if we go out to 100mm the SK60XL would win.

If it's possible to get an SK60XL copy that really is as sharp on both sides as this one is on its good, then you could shoot your panoramas at f/11 and clearly outperform the Rodenstock 55. However seeing three SK60XL copies and all three having pretty strong left-right differences at f/11 it seems unlikely to get a copy that performs at that level.

I suspect that my lens that is about to come back from Schneider-Kreuznach without action will show that it performs as bad as it did when I sent it in. They did not find any error and even said that it performed better than their spec. If this is true it means that if you have a bad copy it won't be possible to get it aligned either as the repair shop don't aspire to that level of quality. So far that's speculation though, but it doesn't look too good.

ALPA is said to have custom calibration of their lenses. Anyone who knows who does that, and if it has any effect, or if it's just market speak? If it's Schneider-Kreuznach repair shop that does it, it doesn't look promising... I would certainly be prepared to pay for a remounting service if you could get the lens to perform at the same high level both on the left and right side. One of original reasons for using medium format is to get better precision in lenses, but the observations I have on this lens so far makes me doubt that there really is any truth in that.

I'm not sure the image is displayed correctly here in this forum, so here's a temporary link to the image http://torger.dyndns.org/sk60xl-loaner-analysis.jpg