Very serius bug in CO 4.0 final relase
Im discovered crital bug in CO 4.0 final.
1. On the preview window zoom over 50% lost any detali in highlight
2. All output files jpg and tiff 8 or 16 bit color mode don't have any detali in highlight
3. Only DNG output have full information about processed file.
Only in this files are detali in highlight.
This is serius bug.Please fast relase correct update...
Preview window full detalis in highlight :
http://www.vpx.pl/up/20071231/ml514750f54abbcebb0de7cd55ce249d8f8fd257943.jpg
Lost detali output :
http://www.vpx.pl/up/20071231/ml514797b49655747396ebe9689ce931d04f84c82475.jpg
Sample image by Jacek Góźdź
http://linuxphoto.org
More detail on highlight recorvery in varius programs :
http://linuxphoto.org/test_highlight.html
Visible detail zoom 50 % :
http://www.vpx.pl/up/20071231/ml515198576dd65f2747bb24439aa3a051a732c192491.jpg
Lost detail zoom 67 % :
http://www.vpx.pl/up/20071231/ml51522ce28acbae87d2283141dc3e6f69bb643147062.jpg
1. On the preview window zoom over 50% lost any detali in highlight
2. All output files jpg and tiff 8 or 16 bit color mode don't have any detali in highlight
3. Only DNG output have full information about processed file.
Only in this files are detali in highlight.
This is serius bug.Please fast relase correct update...
Preview window full detalis in highlight :
http://www.vpx.pl/up/20071231/ml514750f54abbcebb0de7cd55ce249d8f8fd257943.jpg
Lost detali output :
http://www.vpx.pl/up/20071231/ml514797b49655747396ebe9689ce931d04f84c82475.jpg
Sample image by Jacek Góźdź
http://linuxphoto.org
More detail on highlight recorvery in varius programs :
http://linuxphoto.org/test_highlight.html
Visible detail zoom 50 % :
http://www.vpx.pl/up/20071231/ml515198576dd65f2747bb24439aa3a051a732c192491.jpg
Lost detail zoom 67 % :
http://www.vpx.pl/up/20071231/ml51522ce28acbae87d2283141dc3e6f69bb643147062.jpg
0
-
Markus,
I went through the links in your posting. Read it all. With all respect I think I don't I get any idea of what you intend to say. Please do not be offended, but for reporting a 'critical bug' there is a serious lack of detail in your report.
Since this is a user-to-user forum, I advice you to report bugs to Phase One in a support case as well.0 -
If You don't trust me.I can doogin it self.
Please download http://linuxphoto.org/other/exposure16.mrw
After open this press F7 (auto setting) and see....
Also You can see post on this forum...
http://manufactura.home.pl/p-pl/viewtop ... f77#164234
First post and pictures...0 -
Markus
I'm with Paul on this. I don't see what the problem is.
I loaded your RAW into C1, reduced the exposure by half a stop and applied 30 Highlight recovery. No blown highlights at all.
However, what I did notice was a total lack of the 'jaggies' that I see at 50% on my Canon RAW files.
Can you post a link to one of your RAW files that has more lines in it, preferably diagonals?
Thanks,
Iain0 -
Im test overexposed raw my friend.
Link is below.
Other people have the same problems...
See You link of Pentax forum...
Link is below..0 -
Markus
Any chance of a link to another RAW file from you that has some lines in it?
Thanks.0 -
Sorry my pef (RAW) are not supported.
My camera is Samsung GX-1S....0 -
Hi Markus,
I downloaded your exposure16.mrw file (and exposure0.mrw), which is a 1.6 EV overexposed file as is stated. I opened it in CO4 (both Win and Mac) and ACR 4.3.1 and noticed the blown highlights. Next I converted it to DNG, as you stated, and opened it in ACR4.3, as I did in with the original mrw file and with the same result.
For me it is clear that the highlights have fully clipped in the sensor of the camera already due to the overexposure. That does not come as an surprise. You also presented a nominal exposed file (exposure0.mrw). The blown out highlights are very well recovered by CO4 (and ACR as well) with the Auto setting (Ctrl-F on Win). Easy job.
No bug whatsoever I can put my finger on, to the contrary! CO4 does a splendid job where with CO3.7.x I would have to take more steps. Thanks for sharing though. And a very happy new year to you.0 -
No bug whatsoever I can put my finger on, to the contrary! CO4 does a splendid job where with CO3.7.x I would have to take more steps. Thanks for sharing though. And a very happy new year to you.
Did the same comparison, here, and had no problems. Excellent recovery
and very usable results.
Rob0 -
How to reproduce this bug:
1) open RAW file from http://www.linuxphoto.org/other/exposure16.mrw
2) recover highlights in "fit" zoom mode (see the recovered details?)
3) change the zoom into 100% - the recovered details are gone
4) pan the image - there are details for a short while and then nothing
In my opinion picture what you view in preview window CO 4.0 without any zoom is not 100 % identical to compare jpg output...
Please compare view in aplication window and output jpg or tiff....0 -
[quote="Markus_pl" wrote:
How to reproduce this bug:
1) open RAW file from http://www.linuxphoto.org/other/exposure16.mrw
2) recover highlights in "fit" zoom mode (see the recovered details?)
3) change the zoom into 100% - the recovered details are gone
4) pan the image - there are details for a short while and then nothing
In my opinion picture what you view in preview window CO 4.0 without any zoom is not 100 % identical to compare jpg output...
Please compare view in aplication window and output jpg or tiff....
I just did another test with a combination of exposure reduction and
highlight recovery, which produced an image that looked pretty decent
in the "fit" preview. I converted the file as an 8-bit tiff and viewed it in PS7. It
didn't look anything like the corrected preview image. All the areas that were
adjusted turned a light/medium gray.
So, at least from the conversion I just viewed, it looks like there is definitely
a problem.
Rob0 -
Hello all,
Just to give an update, There is a support case for this that has been forwarded along to our official test department so that an error can be identified, clarified, and fixed.
Based on my tests the problem seems to occur when images reach the limits of data in overexposure, however I believe it is clear in this situation that some information is getting clipped in the end result.0 -
Markus / Robert,
With Markus' latest explanation I understand the issue better. I acknowledge that in the converted file output detail is missing that seems to exist in the preview (as long the preview is not zoomed in at 50% or more). Also when zooming in to 50% and beyond, detail gets lost which show up initially.
In addition I can confirm that this behavior also occurs in the CO4 / Mac version likewise.0 -
Thanks all for attention 😊
Happy New Year 2008
Special tnx to Jacek Góźdź.
He supply test image and help me explain bug for all.
Im have one's wish :please fast addition my Samsung GX-1S support 😊
In my free time I will be beta tester 😊
Best regards Marek0 -
Happy new year to you too Markus! 0 -
Where is new version CO 4.0 with correct this problem?
Im download today from eu server new CO 4.0 and this problem is not solved...0 -
i have exactly the same problem;
yet converting the picture either in C1 3.6, capture NX or ACR works fine;
will there be a fix? thanks0 -
Im wait for fix this problem 1 month.
PhaseOne is very slow.0 -
Perhaps getting 3.7.8 out for the "pros" has taken priority over 4.0 for a while. 😊 0 -
[quote="Markus_pl" wrote:
Im wait for fix this problem 1 month.
PhaseOne is very slow.
Well Markus, 4.0.1 is out and I hope it was worth the wait.0 -
TNX for info 😊
Wait a moment Im test it ..0 -
Im find next error 😊
Output files is ok it contains data in highlights.
Im can't process single file when Im click on process button, Im not process now image.
Image have auto add to batch quene.
Only in batch mode Im can process image and write output files...
My camera Samsung GX-1S is still not supported ☹️(
What you are doing? Sony A700 is supported 1 month afer premiere and my
Samsung is not supported after 2 years after premiere ????!!!!!
Im worse photographer because im use Samsung camera???0 -
[quote="Markus_pl" wrote:
Im worse photographer because im use Samsung camera???
Yes. (just kidding!!!! 😊 )
Support for cameras is determined based on demand. Now with the 4.0 platform we can support new cameras quicker, but priority will will always be based on popularity.
you can make requests by creating a support case.0 -
Im thing more people on the world have Samsung DSLR than Sony A700 😊 and more profesional photographers have Nikon or Canon DSLR than Samsung DSLR 😊 not but that users of Samsung should be ignored..
Im request for support 2 times and many times of this forum...You need pef for correct support ? write to me and Im send to you this via e-mail or ftp..
Return theme, after restart computer after instalation Im can process single file 😊
In first run via instaler check "run program" Im can't process it...
Sorry, Im be nervus 😊 program work now correct and Im don't work with my pefs because it not supported ☹️( Samsung pef is very, very silimar to pentax pef.
Best regards Marek0 -
I have just downloaded v4.01, and have exactly the same problem. As soon as I zoom an image to above 50%, the image looks way over-exposed. I have logged the issue.
I use a Canon 40D0 -
The Phase One team have solved my problem. I have been using highlight tone priority, which version 4.0 did not process correctly. v4.01 has corrected this, but I had some old cache files in my photo folders that were causing the problem. Deleted them, and problem solved.
Thanks to the support team.0 -
Create folder cache in all viewed folders by CO is stupid...
On disk after it is many unnecessary folders....
My propose is simple: aplication make one!!! cache folder on disk location definied by user..
This folder must be purge via CO menu :automatic one at week or definied period or manually by user..
It is simple and perfect....0 -
[quote="Markus_pl" wrote:
Create folder cache in all viewed folders by CO is stupid...
On disk after it is many unnecessary folders....
My propose is simple: aplication make one!!! cache folder on disk location definied by user..
This folder must be purge via CO menu :automatic one at week or definied period or manually by user..
It is simple and perfect....
Hmmm, wasn't that the way it worked in v3? Now there's a thought! No-one's mentioned that before!?!0 -
works fine for me (NEF - D70) after the update; thanx 0
Post is closed for comments.
Comments
28 comments