3.7.3 corruptis 5D images
I remain having problems with 3.7.3 corrputing images. The release build no longer corrupts the original raw files by just touching the image but the thumbnails and the converted images are corrupt and have colored stripes all over the images. DPP does not have this problem with the same images. At least the release 3.7.3 does not destroy the raw original but the problem is not fixed.
0
-
I processed a session last night and the same thing happened to about 20 images. The problem images are each half blacked out with several narrow colored stipes running through them and i've been trying to figure out what i did wrong all day, but they must be corrupt like you say.
I went back into capture one and processed these problem files different ways with different output dimensions; flipping them back to a horizontal state. etc. Nothing has worked yet and i'm running out of time to headache over it much longer.
So what can we do about this? No way to remedy?
And I'm curious why more people aren't having this problem?
patti0 -
I created an incident with support and uploaded samples. It is in engineering now. Hopefully they will figure out this odd one soon. 0 -
I tried to make sense of it. There's really no rhyme to it, you know? Like at first i thought it might be happening just to the images that i rotated. All the images that it did happen to were rotated during editing, but it didn't happen to all of the rotated images. Hmmm,
Anyways, i'll really be anxious to hear what you find out and i'll be following this thread.... thanks again.
(happy new year) 😂0 -
So that is what I'm seeing as well. However it is with a Canon 1DMk2 and RAW, CO Pro 3.7.1 File Version 3.7.1.126
I've had this happen on 5 seperate jobs, and frankly it scares me more than a bit. System info: Dell 9100, XP Home, 2.5gig Ram, using CF reader that is built into their 24\" wide flat panel monitor.
Work flow, download into a folder, make a copy folder as a back up, bring first folder into CO Pro, generate previews. At times 6 to 10 images are corupt bad previews have the color lines, partial images, cross processed looks etc.
At first I thought it was the CF reader, but after deleting what I thought were the bad RAW files from the original folder and then copying the same RAW files from the backup folder into the working folder I'm more often than not getting a good preview so I started to think that CO Pro might be causing the problem.
FWIW, I have not tried Canon's DPP to see if it is working correctly on the same files.
Thankfully CO is not touching the RAW file in this situation TTBOMK. Just about ready to do the upgrade to 3.7.3 and I'm wondering my thoughts are correct, and if so has this issue been fixed in the public release?0 -
i hate to tell you this, but I haven't used anything but 3.7.3 version and I'm having this trouble. (i'm a newby here)
I'm sure the upgrade might solve some other issues, but not this one.0 -
The first post on this problem was on Dec 13, 2005. Today is Jan 09, 2006. As of yet no one from Phase One or Capture One has responded to this thread. 🙄
I'm finding that a bit unexceptable. Anyone else agree?
Phase One HELLO!
The problems with the corrupted images ( previews ) lead to corrupted output and for someone making their living with this program that is not an acceptable part of an efficient work flow.0 -
Hello
Let’s first clear out the terminology.
We are talking about corrupted previews and not corrupted RAW files.
This is quite important so no one feel unease about using Capture One.
Capture One can not corrupt RAW files when it only reads and never writes to RAW files.
Second, Steve Bruns: this is a user to user forum, if you want Phase One to respond to a request then create a support case with Phase One at http://support.phaseone.com just like Mike_Kelly has done.
We do answer the forum quite often but users on our support log system has always a higher priority, which anyone in here is welcome to use!
In this case I would please ask you to do a support case when cases like this is often a longer troubleshoot process involving computer and the software on it which might be interfering with Capture One there can be several reasons to this happening so let's figure out why it is happening on your computer.0 -
Ulf:
<this is a user to user forum,>
I was making the assumption that it was moderated by Phase One and that they would be more involved in it. I did not realize that it was more user based in nature... as to input and problem resolution.
<if you want Phase One to respond to a request then create a support case with Phase One at http://support.phaseone.com just like Mike_Kelly has done.>
I've done that a while ago... Sent out put files and explaination of problems as well system info. Currently I'm waiting on a responce.
FWIW, after thinking about the fact that CO does not do anything to the RAW I was a bit more releived, however this glitch has caused some significant issues with four major jobs and timely delivery of them to clients.0 -
Steve I am sure you will have an answer soon. 😄 0 -
The pre-release 3.7 corrupted my raw files by doing nothing but viewing the images in C1 LE. I don't care if you don't think it is possible. I sent you the corrupted files and told you that there is nothing else in my workflow that could of caused the problem. With the same workflow Canon DPP does not cause the problem. The release version does not do this but the thumbnails and currently viewed image in the workspace are corrupted and can not be processed without the characteristic colored bands being included in the conversion. I have tested this over and over comparing it to Canon DPP. Canon DPP does not have this problem with the same images that C1 LE 3.7 does.
You should check your forums and review the user problems if you care about customer service.0 -
Mike:
What happened? What was done on your end?
On my end I've been going through all sorts of songs and dances in testing this and that, using CO Pro v3.7.3, on one system, CO Pro v3.7.1 and also loaded them to a lap top with CO Pro v3.6, where all of the releases were public releases. Still ran into the problem.
Web based tech to Denmark is not the cat's meow so to speak....but they have been able to help me sort out some of my testing procedures. Also some help from another forum, plus a bunch of testing over a couple of days are narrowing it down to what appears to be corrupt RAW coming off of the CF cards into my systems.
Tomorrow I have a friend bringing over a lap top that has never been near my systems and it has CO Pro v3.7.3 on it. Through a variety of contacts in the industry I've been able to cobble together a group of very knowledgeble folks to help me sort out this problem I'm having.
What I did today was to shoot to two 1gig Ridata's one in each camera. These files were then downloaded to seperate folders. One with the use of the CF reader that is built into my Dell monitor, the other using a Zio CF reader where the files were renumbered to identify them. Either way I ended up with corrrupt previews that were kind of random, but some files were the same. The vast majority of the files processed out not OK.
I also did this on my older backup system, which runs CO Pro v3.7.1, and got the same results.
Then I brought some of the same files into DPP and for the most part DPP indicated corrupt files and processed out bad files, but not always.
I also tried using Raw Shooter Essintial. Some times it worked giving good output from what appeared as corrupted previews, but at other times it did not give good output. BTW RSE is written by the original, or one of the original writers of CO.
This pointed me in the direction of using some brand new Ridata 2 gig cards that have never been used. Ran the same series of test, these all previewed fine. Go figure...
So at the moment I'm left wondering if I have some bad CF cards. Let's see I have about 22 of them to check out.
I also think that the RAW files are comming in with some bad data in my initial tests. Is this due to faulty CF readers, CF readers that run in the monitor? This I do not as of yet have an answer on. When this happens I think at times the RAW processing software is at times able to pull it off, but at other times it fails. This appears to happen with a variety of RAW software, not just CO Pro.
I have calls into Canon Tech to find out if they have had issues with the firmware v1.2.4 for the 1DMk2 and corruption of files. FWIW, if you chimp while the buffer in the camera is still flushing it can cause a RAW corruption as you are trying to read the data as the camera is trying to clear the buffer. It's a lot of data to be process by a very small engine, very fast. However the Canon firmware v1.2.4 is suposed to have fixed that issue. In all of my testing I did not chimp at all...just in case.
If you have access to www.digitalweddingforum.com check out this thread:
http://www.digitalweddingforum.com/ubbt ... e=1#948825
You will find what I've done there. However because it is a private forum you might not be able to access it. I mentioned that thread as I can not, to the best of my knowledge, upload images (screen shots) to this forum, but they are there on that forum.
FWIW, I'd never use a prerelease version of software as it is not ready for prime time playing.. And yes I was involved in some of the beta testing of CO a ways back, but the headaches were not worth it to me. So what I'm saying is that when using a prerelease version of any software your milage may vary greatly.0 -
I truly hate internet \"fly off the handle rants and I just did one in my previous message\". I have gotten attentive support from tier one Phaseone support but the \"it just can be possible\" message from Denmark set me off. I deal with lots of vendors weekly who claim their software or hardware just can't be doing what it is doing.
I fully accept also, as you suggested that beta software is just that and should not be confused as production. I was not ranting about that as much as the claim that it was just impossible for C1 to do that.
I have come home from shoots with the 5D and out of 9, 1gig cards the problems come on a diffrent card on each shoot so I don't think it is the cards. The last time I kept the card loaded and did not delete it after uploading. When I copy straight in via usb card reader and touch it with DPP nothing ever happens to the files. When i copy directly into a folder and look at it with C1Le3.7 release I get the funny thumbnail and current view problems, colored bands, that are converted as seen on the screen to the processed tiff. But the original files are not corrupted. If I then go back to DPP it does not \"see\" a corrupted image and can process the same file correctly.
It seemed proof that C1 was broken. I then proceeded to prove my point by removing my image drive from my workstation and putting it in a firewire external case to change the workflow hardware path. Much to my surprise the symptoms changed. Some images that acted corrupted in C1 before look normal and others look corrupted. It is not random. So it is very hard to blame hardware. With the drive in the fireware case DPP still sees no problem but certain different files are predicatably not displayed properly in C1 and it is not the same files as before but the same files do the same thing repeatably. Very weird. I am now wondering if C1 programmers used some way of accessing the file system, maybe in order to improve performance, that is non-standard and is a little marginal on some hardware? DPP, which is horribly slow, may use standard Microsoft calls to the file system and hence does not experince the problem?
I am tired of this, I have had my 5D since sept. and can't to any serious processing yet.
The only thing worse would be having to use DPP:)0 -
Mike – rant on … as long as it is followed by an Apology and Honesty… I admire and respect this.
We will continue to work on your Support Cases, however since this Forum Thread appears to be getting a lot of attention, I will post here as well.
Actually, the Technical Support is Global and quite a team with the Support in Denmark actually being The Cats Meow, since they have direct access to our brilliant Developers. This is why I was looking forward to working in the Copenhagen office this week and working these Cases with them.
Developers stand firm in that Capture One will not and cannot alter a DSLR RAW file. I’m no Developer so I believe them. I also know that if this were the case, Support would be inundated with many more Users that are having problems such as yours; and Phase One would not be leading the way in RAW development software.
Then there is this Fact: 100% of Support Case accusations that Capture One has corrupted RAW files have been resolved with another application being the culprit.
When this corruption is occurring may be determined by launching the Task Manager and selecting the Processes tab. When transferring or developing images, display this window large enough to show all processes and do a ScreenShot.
The culprit may show itself. Send the ScreenShots in Support Case.
As for DPP having no problem with same RAWs, I honestly don’t know –
• is the software installed on another location ?
• is there another variable here ?
• is there only a conflict between C1 and another software culprit ?
There are so very many variables that can have an ill-affect upon image files throughout a workflow.
Add more devices and drives = Add more variables.
Troubleshooting often involves the elimination of as many of these variables as possible. To absolutely eliminate any in your situations, I suggest the following:Install Capture One at default location:
C:\\Program Files\\Phase One\\Capture One PRO
Set the Preview Cache to default location:
C:\\Program Files\\Phase One\\Capture One PRO\\Previews
Place RAW files only to default location:
C:\\Program Files\\Phase One\\Capture One PRO\\Captures\\ your Session files
Now without using any other software, download image files direct from a CF card to location using a Card Reader (can one in monitors be dependable? ).
Load Session > Browse folders… > clik Insert current folder to Permanent Collection > observe Thumbnails and Previews build > Adjust images > Process image files with Set Process folder… with check at Always process images to associated image folder which will create a subfolder named Develops within your Session folder
Check for any problems whatsoever.
If there are problems, report them in detail in Support Case and use Task Manager to analyze.
Mike - I assumed you were on a Network because the ScreenShot of your file tree shows additional drives of E: G: H: and I: Are these internal hard drives, partitions, other? Depending on how these are configured, the Images are moving to/from various drive locations and this adds variables.
Steve – based on their web site, Ridata appears to be a large global company and seems reliable. But if your cards are suspect, then shoot to another card type or better yet, directly to your hard drive at default location by tethering.
Please understand that we in Support truly want to resolve these issues just as you do. The only way this can happen is with determination, communication, and patience.
So lets put our heads together and see if we All can get a bit Smarter 😉
Very Sincerely,
k c0 -
🤓 ...now I just learned that there is a limit to a post.... Rant Control I suppose...
To finish,
Please understand that we in Support truly want to resolve these issues just as you do.
The only way this can happen is with determination, communication, and patience.
Very Sincerely,
k c0 -
Keith:
<Web based tech to Denmark is not the cat's meow so to speak....>
The only reasons I mentioned this are two:
1. Being the time difference between there and here. Wonder what it would be like if one were 12 hours off of Denmark time? No sleep...;-} Or extended periods of time being required between trouble shooting and solving a problem.
2. The nature of the user interface of the web based support where it is not exactly clear if you are attaching a file OR uploading a file, and the fact that it hangs up with large files. It could be simpler or more clear for the simple minded like me.
Re this forum, it possibly would be helpfull for those that use it to be able to post small images as in screen shots.
'Best0 -
A recent post that went to Phase One Support in responce to what they sent me.
<So they show up corrupt in DPP as well?>
At times yes, at times no, it varies.
<That puts things in perspective.>
Not really, as two of use today worked on it again, comming to the conclusion that it is totally unpredictable.
<CF card reader or (very likley)>
Thought so as well, but eliminated that by using two other entirely different readers today. Problems still happened.
<too some kind of indexing Image handling program on your computer that is causing issues.>
We thought of that one as well, so we turned off XP's making of index thumbs for Windows Explorer in Folder Properties.
<One thing that comes to mind is Microsoft RAW reader program.it does not necessary cause a corruption of the file but it has a tendency to do file locking so other programs have limited access to the file, even reading. Uninstall and reboot will sort it. >
I don't have it on any of my systems.
<It might be CF card as you say>
You are getting close. After working on it all day today, and running other tests, I think that it is bad RAM in the new Dell9100. How I came up with that, I'll elxplain below.
I took the CF's, which had on them the original test shot I shot the other day. Downloaded them to a folder that was on the desktop. This folder was never touched by CO or DPP or any other program.
Burned a DVD with verification turned on at 8x. During the verify process there were several files that failed verification.
Burned a DVD just as above at 4x. During the verification process there also were several files that failed verification.
Burned a DVD of each of the two CF cards directly from the from the CF card reader to the DVD. I did this thinking that the system would not really be touching the RAW files. However when the burn happens it writes to a cache file then to the DVD...so the system touches the files.
The system has two CF readers and two DVD burners on it. Each of these above tests were done trying an either or hardware use. IE this burner, or that CF reader. I still received a failed verification on several files every time. WHAT WAS COMMON is that the files that failed verification never had the same name! Not one file with the same name!
I then took the same CF cards into the older Dell 4550 downloaded just them just as above, and burned DVD's as above. Every time the write DID verify.... Not a single problem. They also opened fine in CO PRO 3.7.1 which is on that machine.
I looked at the properties of each DVD and compared them to the properties of the folders on the two machines. Strangely all the properties matched! Same # of files, same size in gigs, same number of bytes, and same number of bytes on disk.
Well if they did not verify as above, I'd have expected to find different values in the properties, but that did not happen. Maybe the byte order go scrambled or something, but they all were the same # of bytes when compared.
So at this time I'm looking at the Dell 9100 as being the problem. Reason being is that everything that touched that system internally ended up with some sort of a problem, such as files opened in RAW software, burned to DVD, or transferred over a LAN to the backup machines.
I spoke to a friend on one of the private forums whom had mentioned the possibility of a bad mother board bus as an issue, we looked into the Event Viewer in the adminstritave tools of the Dell 9100 looking for refferences to ERROR ID 51 which might point to the machine using a NForce chipset which apparently is a known possible source of corruption. There were a lot of errors in there, but none with ID 51.
We discussed testing the RAM. He told me about \"Memtest86.exe\" mentioned that it is a shame that it can't test CF cards, as I was about to test ALL of my CF's using Card Wiper. The light went on in our heads that said all files that are having problems were in contact with the RAM in the Dell 9100.
The machine came with 512mg RAM in November. I installed an additional 2gig of RAM in early November. The problems started showing up in early November.
So I downloaded \"Memtest86.exe\" and ran it on the system. There were just under 47,000 errors logged by the test!!!! I removed the 2x1gig RAM chips and ran \"Memtest86.exe\" again on the original 512mg of RAM. There were no errors at all after running the test.
Also I loaded the test materials into CO again after the RAM tests and removal of the 2gigs of RAM and things appeared to be fine! SO I'd imagine that the issue is what appears to be bad RAM in the 9100 that is causing the problem.0 -
The machine came with 512mg RAM in November. I installed an additional 2gig of RAM in early November. The problems started showing up in early November.
So I downloaded \"Memtest86.exe\" and ran it on the system. There were just under 47,000 errors logged by the test!!!! I removed the 2x1gig RAM chips and ran \"Memtest86.exe\" again on the original 512mg of RAM. There were no errors at all after running the test.
Also I loaded the test materials into CO again after the RAM tests and removal of the 2gigs of RAM and things appeared to be fine! SO I'd imagine that the issue is what appears to be bad RAM in the 9100 that is causing the problem.
_________________
Steve Burns0 -
After installing a new os on a new hd and then installing COLE3.7.3 everything works with the same files on the same firewire external drive. Something corrupted my system software. Now I get to reinstall 50gb worth of installs from scratch. 0 -
Steve and Mike,
Well, I think we and all the Users that have been following this thread have been enlightened and relieved.
What an outstanding job of your very deep and thorough Troubleshooting methods and sharing the findings with us all.
Thank You, this is very much appreciated.
Re: Support
Please know that between Denmark Time, East Coast US, and Central Time in US (7 hours behind DK ) - the Support System is monitored no less than 18 hours per day and sometimes on Weekends and Holidays.
We continuously 'put our heads together' in trying to resolve Customer problems. When we get over our heads, the Developers and R&D departments become involved until a resolution is found. Our Support is not exactly tiered, but more of a Team Effort.
We constantly work to improve the Support Logging System in all its functions. Recently upgraded was the ability to attach larger RAW files, but we agree there can be improvement in notifying the User the status of an upload. As always, Phase One welcomes and appreciates all Customer input.
Re: Forum
others often do insert images and screenshots, or link to images with Forum Posts. I'll ask the Administrator to post the How To and any Restrictions involved with this.
Thanks again,
k c0
投稿コメントは受け付けていません。
コメント
19件のコメント