Jump to content
Sign in to follow this  
Guest guy_mancuso

Results, we want to see results on 1.102

Recommended Posts

Advertisement (gone after registration)

One of my M8s still have that 1 to 2 pixel wide vertical line appearing slightly right of center, mostly at 640 ISO and higher. Barely visible at 320 ISO. No problem at 160 ISO.

Share this post


Link to post
Share on other sites

cherry blossoms, 75mm f1.4 Lux. Colours seem to be better to me. No problems with 1.102.

 

/applications/core/interface/imageproxy/imageproxy.php?img=http://www.pbase.com/jeremystreeter/image/77157945.jpg&key=e18a6ad43db7ae81c97cb93a8c4286bac49ddc236adbfcb5500dbd7f87461f0a">

Share this post


Link to post
Share on other sites

Maybe I am wrong but... I think the noise at ISO 640 has improved a lot...

 

It is your experience too?

Share this post


Link to post
Share on other sites

For results with the B&W 486 filter, see next post.

 

I did a careful white wall study this afternoon. Indirect cloudy daylight, light variation of maybe 1/6 stop across the field of view. Camera was on tripod, about 2 feet from the wall, focused at infinity to blur and to create maximum vignetting. I shot in DNG only, exposure +1 2/3 stops, and the only post processing was to white-balance everything on the center point, and output 900x604 pixel jpegs (in C1) with the default M8 profile.

 

I'll attach the worst case vignetting sample at f/2.8, with lens reco OFF, followed by the cleaned up result, also f/2.8 with lens reco ON (since there was no filter used in this comparison). The first shot shows vignetting overall of about 0.4 (in intensity) averaging over the corners with almost equal effect in the blue and green channels. The red channel is reduced by 0.44 at the corners, an extra 10% due to the angle dependence of the IR filter in the CCD cover glass. The second shot is essentially color neutral and only 0.05 of the overall vignetting remains. (At f/5.6, all vignetting is removed and things remain color neutral across the image.)

 

scott

Share this post


Link to post
Share on other sites

This time I'll compare the vignetting with the non-Leica filter, lens detection off, f/2.8 and then with lens detection on, UV/IR correction added. The correction does not completely remove the vignetting either at f/2.8 or at f/5.6 but the red part of the vignetting is removed. The filter causes an extra decrease in the red channels at the corners. I measured a decease intensity of .38 overall at the corners (in the green and blue) but in the red channel it was .52. After turning on the corrections in 1.102, the overall vignetting only decreases slightly, but the red channel is now slightly increased relative to the green and blue intensities. It looks good to me, but it would be worth trying the Leica filter to see if the differences are worth it. The physical dimensions of the Leica filter (it is supposed to work with the lens hood -- my slim B&W filter prevents the lens hood from going all the way into place) are another reason to try it.

 

scott

Share this post


Link to post
Share on other sites

Advertisement (gone after registration)

. . . The physical dimensions of the Leica filter (it is supposed to work with the lens hood -- my slim B&W filter prevents the lens hood from going all the way into place) are another reason to try it.

 

scott

 

Scott - I've got both the 21 and the 24 so I have two copies of the same lens hood. Last week I attacked one of them with the Dremel tool, grinding off the plastic ring that interferes with the filter. Now it fits fine with the filter on. Of course it no longer keys into the notches on the lenses, but there is enough friction to prevent it from turning easily.

Share this post


Link to post
Share on other sites

I got my ex-demo, silver-chrome, M8 yesterday. I loaded 1.102 last night and went out this morning to try it out around my local railway workshop, which I know to have very mixed, difficult, lighting. Both of the shots below were taken without filters or coding and are amongst the best I've ever achieved with either digital or film under these conditions. I'm impressed...

 

The only downside is that, like others, I find that 640 ISO is as fast as I want to go for normal work. 1250 and 2500 produced a great deal of colour noise.

 

1. Locomotive chassis under mixed fluorescent and mercury vapour lights. CV 28mm f1.9 Ultron at f4 and 640 ISO

 

 

2. Locomotive boiler - mixed daylight and fluorescent/mercury vapour. CV 15mm SW Heliar at f4.5 and 640 ISO

 

Share this post


Link to post
Share on other sites

After a day of snapping with CV 15mm + 1.012 + B+W 386 it seems cyan is now banished.

 

I don't see much dfference between selecting 16mm and 18mm (below, taken with 18mm setting in menu).

Share this post


Link to post
Share on other sites

I am curious about the LCD display:

lshooting RAW/jpg basic, lens detection On+IR, using the 15 coded as WATE and filter (only filtered combo i have), upon first playing or re-playing an image it initially comes up without the cyan correction, then after a second it changes the image to the cyan corrected version.

my question is: where is the lcd getting the un-corrected version from for later re-plays? are both versions stored somwhere? when is the correction actually applied to the image?

Share this post


Link to post
Share on other sites
I am curious about the LCD display:

lshooting RAW/jpg basic, lens detection On+IR, using the 15 coded as WATE and filter (only filtered combo i have), upon first playing or re-playing an image it initially comes up without the cyan correction, then after a second it changes the image to the cyan corrected version.

my question is: where is the lcd getting the un-corrected version from for later re-plays? are both versions stored somwhere? when is the correction actually applied to the image?

I see the same behavior with my coded CV 15 and had the same question for the experts.

Share this post


Link to post
Share on other sites
I am curious about the LCD display:

lshooting RAW/jpg basic, lens detection On+IR, using the 15 coded as WATE and filter (only filtered combo i have), upon first playing or re-playing an image it initially comes up without the cyan correction, then after a second it changes the image to the cyan corrected version.

my question is: where is the lcd getting the un-corrected version from for later re-plays? are both versions stored somwhere? when is the correction actually applied to the image?

Yes, this "corrected" LCD display is a little odd, if not distracting. Would be best to disable this when the system is programmed "Lens Detection = Off" because in this case, the "correction" is just false. Therefore:

 

Lens Detection == LCD corrected display

 

OFF == OFF

ON == ON, no cyan/red correction

ON w/IR cut == ON, with cyan/red correction

 

Or, just let us program what lens we have, and leave the LCD color correction on

 

 

Now, as regards Guy's post:

Since 1.092, and now 1.102 the AWB is much better. I've had no lockups with my "first wave" system; however, when returned from Solms for the h/w fix, I'm hoping not to have a unit with the dodgey transistor[1]. I have never seen the "venetian blind"/"zebra stripes". Do these striped events occur with "Lens Detection" ON, or OFF? I have not used "Lens Detection" yet... too many old Leica lenses, and those not Leica to bother with this "feature" yet.

 

rgds,

Dave

 

[1] Before I sent the system back, a very very small scratch was suffered on the LCD, not there when the camera returned! Also, Mr. Norton's "M8 in pieces" showed that only a few parts of the internals could be cost-effectively replaced... curious.

Share this post


Link to post
Share on other sites
Yes, this "corrected" LCD display is a little odd, if not distracting. Would be best to disable this when the system is programmed "Lens Detection = Off" because in this case, the "correction" is just false. Therefore:

 

Dave

 

From DAY ONE the image displayed on the internal LCD has always had a correction time with it. On every image I shoot and review on the LCD it first comes up and then 2-3-4 seconds later it JUMPS and give a cleaner/clearer version of the image.

I don't normally shoot Jpegs so I'm not sure if it does the same with Jpeg only images.

I would suspect the first view of the image is the RAW data and then after whatever correction the internal engine does it displays that image next.

I get this even when going forward and backwords through all images on the card.

Share this post


Link to post
Share on other sites

"I would suspect the first view of the image is the RAW data and then after whatever correction the internal engine does it displays that image next."

 

doesn't that imply it is storing the un-corrected raw image and only applying the correction upon viewing? seems wacky to me that it can ever go back to the un-corrected version after the inital play

Share this post


Link to post
Share on other sites
Guest guy_mancuso

I think what Leica tried to do or the way they set this up was with instant review so at least you can see the image come up on the screen as fast as possible than it optimizes the raw data with vignetting and cyan corners. They probably did not want it this wayorginally but since the imageif was optimized from the start than it would take a lotlonger to review and than that would have not worked becuase we would think this is too slow to review. So the answer is show it as fast as they can than optimize it and if you went back to shooting it is still doing the optimizing. it does not bother me since i know it has to process a 10mg file with optimizing it and that just takes time . faster cards may help this downthe road also. Also the primary goal is to safely write the data to the card and that should always be first and viewing is secondary

Share this post


Link to post
Share on other sites
After a day of snapping with CV 15mm + 1.012 + B+W 386 it seems cyan is now banished.

 

I don't see much dfference between selecting 16mm and 18mm (below, taken with 18mm setting in menu).

 

I found the 18mm setting better too. With the 16mm setting, the CV 15mm was over corrected and had red corners when using a B+W 486.

Share this post


Link to post
Share on other sites

[1] Before I sent the system back, a very very small scratch was suffered on the LCD, not there when the camera returned! Also, Mr. Norton's "M8 in pieces" showed that only a few parts of the internals could be cost-effectively replaced... curious.

 

Hi Dave

 

Could be that Leica just transferred your top plate to a new body and kept your old internals for further investigation. It wouldn't be the first time they've been known to do this to speed up delivery. IIRC it happened to a certain extent with late model IIIc bodies sent in for upgrading to IIIf in the 1950s.

Share this post


Link to post
Share on other sites
[1] Before I sent the system back, a very very small scratch was suffered on the LCD, not there when the camera returned! Also, Mr. Norton's "M8 in pieces" showed that only a few parts of the internals could be cost-effectively replaced... curious.

Hi Dave

 

Could be that Leica just transferred your top plate to a new body and kept your old internals for further investigation. It wouldn't be the first time they've been known to do this to speed up delivery. IIRC it happened to a certain extent with late model IIIc bodies sent in for upgrading to IIIf in the 1950s.

I don't want to hijack this thread about f/w 1.102, AND I cannot know for certain what actually happened to my system over 7 weeks btw my shipping it out and its return. I have no "banding"/"streaks", and so far f/w 1.102 has not been a step backward from 1.09. Now, if we could get rid of LCD color corrections for "Lens Detect = OFF" that would be swell

 

rgds,

Dave

Share this post


Link to post
Share on other sites

Something else with 1.102--

 

I know Guy said they're still fixing the green streak from bright light, but I'm darned if I can reproduce it now (I could easily before).

 

To me it looks like 1.102 has at least modified the result, if not fixed it. Anyone get it to band with this firmware?

Share this post


Link to post
Share on other sites

Yes, I can get it to band at ISO 2500 but it seems to be harder than before.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue., Read more about our Privacy Policy