Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

13 threads on CORRUPT frames in less than 2 MONTHS

No offense Kim but hire a DIT and stop having problems.

The majority of the issues you are talking about are commercial funded shoots with a budget where the production has cheeped out and payed the price. Would you complain if your shots we're ruined by the fact that you had tried to light with some 500w halogens and be asking RED to state in the manual that your shots may be ruined if the lighting is rubbish? I'm with Roberto on this, what your asking for is Digital Cinematography 101 classes. The first thing you would be told in such a class is that your data is your negative and not to skimp on its management. The same way you wouldn't treat your negative like crap.

If you don't know what causes an issue then you either need to take the time to do the research or hire someone who does know.
 
Problem is that that is not allways an option AND the DIT is the first thing many production companies want to get rid of.

I am NOT saying that is wise, or that I don't agree with the value of a DIT. But you know as well as I that the marketing for RED is "digital awsomness" and not so much "hey, remember to bring an extra man on set, and he better know his shit and be good and in demand - thus have to get paid", sorta.

I don't agree with that to be honest, if your not in a position to hire a DIT for a commercial shoot than your not in a position to offer the service to the client. You wouldn't show up without film to a film shoot, don't show up without a DIT to a digital shoot. And compare the cost of losing the film loader position and losing the film stock costs and the DIT still ends up being the cheaper option.

If the production company drops the DIT then to be honest their just scam artists and unprofessional. And I will stand by the statement that any commercial shoot that does not have a DIT is amateur night, and by DIT that can mean an Owner/DIT, or a dedicated DIT or a Video Assist/DIT/Tech guy, but unless there is someone handling the footage the production is not serious about their job and responsibilities.
 
Jay...
I don't have problems.
Thanks
And as I said, I got your point the first time.
It's not about me it's about everyone out there who comes across this problem and doesn't know the causes, as a lot of threads proof and why I started this thread. But it seems not to be welcome to ask something for someone else.

I am not asking for a 101 on digital Photographie.
I am not treating my Data like crap.
And your halogen bulb comparison really doesn't fit.
And RED does give tips for exposing in the manual.

But my life taught me something else than your "hire someone who knows" philosophie. IMO it's not good to just trust someone blind who says he knows. YOU/I have to know. In my position as Director/Producer I need to know every bit of the craft of filmmaking even the Data wrangling because at the end it's always my fault and my money when someone screws something up. And that's all I'm asking for: Educate and warn the users about this issue.

But well, I'll shut up now, because there seems to be no need for what I'm asking for and a simple sollution seems not to be accurate to save someone (not myself) some serious problems.

Good shooting everyone
 
Kim, you obviously understand the workflow, the only thing we are disagreeing with you on is where the responsibility lies.

If you let a digital-revolution-clueless producer think that a DIT is not a necessity it is your fault. Either don't let them save pennies in less than intelligent ways, or try not to work with people that might give you a bad name from association.

There are a lot of avoidable problems happening with RED, but it is not because of RED, it is because there us a gold rush going on... low budgets trampling over each other in a high IQ frenzy. RED may be creating the gold (evidenced by their 8,000+ cinema cameras sold), but RED doesn't shoot, RED makes tools. It is not RED's fault and if you think it is, then you are confused about this. That is not the flow of energy through the system.

But my life taught me something else than your "hire someone who knows" philosophie. IMO it's not good to just trust someone blind who says he knows. YOU/I have to know. In my position as Director/Producer I need to know every bit of the craft of filmmaking even the Data wrangling because at the end it's always my fault and my money when someone screws something up.

I couldn't agree more with you on this! Look at Soderbergh, he is now going to be a twice-over first shooter of the new generations of RED cameras. And look at his approach with Che where you could say he invented or re-defined the have-enough-media-not-to-offload concept (budget dependent of course). He did that out of necessity, from being the first ever, with the first generation, the first firmware, a complete walk in the dark. What an excellent analogy for where the responsibility lies...

In RED, we don't have some sort of activist organization for the betterment of your co-workers. RED makes cameras.

If you want to be a director or producer who never has problems, then it is up to YOU to educate yourself (like we are doing here). YOU make YOURSELF a better director or producer, and nobody is expected to hold your hand, unless you found a mentor.

If someone can't understand that picking a solid DIT is a must, then it is up to you not to work with someone that will dirty your name by association with costly problems.

If someone can't understand you need to give your DIT enough budget so that he doesn't have to skimp on a ridiculously affordable CF reader --through which all your production effort and $$$$$$$ will flow-- it is still gonna be on you. Run away if you can.
 
always have multiple CF readers on hand... I have 3.. 2 of which went bad and gave me errors in r3d data manager.. I sometimes had to stop the process.. and re-do it until it would work.. I plan on having 4 CF readers at all times.
 
I have had one real data issue in two years of shooting my red. Its was corrupted files shot on a CF card about one year ago. Seems like it was a freak oc.curence. Fortunately its was on a simple pro-bono job. Never happened since.
 
The biggest issue is the card reader. I had 2 go bad on a shoot.
At first I would get one frame out of a shoot with a glitch.
Then a few shoots later it started corrupting the compact flash card, ( had to use Redundead).
I got 2 new readers and everything is fine again.
 
And, there are also 8,000 + RED ONEs out there.

How does that thing go? "For every cockroach you see there are ten"? So 130? That would make it 0.01625% based off 8k.

Or 99.98375% reliability. :)

And I bet you several are pure human error.

And keep in mind that RED is the "future camera" in a way, a camera more widely accepted by people that have a stronger web-presence.

Finally, it would be interesting to find out what the numbers are for hairs stuck in the gate that go un-noticed, shutter malfunctions, or other mechanical deterioration on film cameras with moving parts add up to.

This bears repeating. This is an extraordinary record of reliability. I think Kim Frank's post is illuminating in that there are 13 threads with this issue, but to my knowledge they aren't cross-referenced by geography, serial number, frame rate, weather conditions, power and electrical conditions, Firmware build, Rental facility, Operator and DIT.

I think in order to post such data in the future, Red User must insist that a matrix of information be compiled that would help us further narrow the variables. We could discover that one camera could be to blame for more than one of Kim's threads, or that a specific rental house's batteries are suspect, or the weather was a factor.

Point is: Even with this amazing record of reliability that Roberto calculated, we should be even more thorough when quantifying errors with cameras this complex, because we can't know where the problem really rests until we do.
 
And, there are also 8,000 + RED ONEs out there.

How does that thing go? "For every cockroach you see there are ten"? So 130? That would make it 0.01625% based off 8k.

Or 99.98375% reliability. :)

Bit of an old thread, but I have just found it...

The figures given make 1.625%

You were only out by a factor of 100 :)
 
I had a situation where it corrupted just one last folder of that day. It was weird. Copied everything but the last one. Then, neither Mac nor PC could access it for a while (or even see sometimes). Then I managed to see the folder in Windows and look inside. Couldn't copy any of it though, but did make a screen grab...

RedDriveError.png


It is obvious that it shouldn't look like that! There should've been the one R3D file and the usual MOVs. Fortunately I had only one such error (and only once!) and I am still perplexed as to why it happened.

The next day I had to explain to almost everybody on crew why we had to reshoot it :)

Can't wait to switch to SSD!
 

Attachments

  • RedDriveError.jpg
    RedDriveError.jpg
    19.2 KB · Views: 0
Most of the time it is not up to the CF Cards or drives but to the ofloading Process. In that way you will also with SSD need to visually check your offloaded Footage.
 
I have been reading this thread with interest. I want to add my own point of view here. I recently did my first shoot with a RED MX and had corrupt footage which we are in the process of fixing.
I am also a Software Engineer.

Although I understand the need to use a skilled DIT, it should not be neccessary if the software and hardware were designed correctly. As said up in the thread, a file transfer is a solved problem - I copy 100 gb files across my network when editing and never lose data.

Now, RED is transferring data at extreme rates in a device that is reasonably priced, so I can understand that corners probably needed to be cut in how the data is processed and stored. But that being said, a truly effective solution would result in being able to leave a CF card transferring and never have to worry about corrupt data. This is my expectation of technology.
So I am disappointed that the RED needs this extra attention of a DIT. This need for a DIT isn't in my mind a solution, it's a work-around.

I'm not sure what the true problem really is. I am very surprised to read here that CABLES are a potential problem with reading from the CF cards because I would expect a checksum in the data transfer between the card reader and the computer.
So is the real problem:
- the reliability of flash memory? If so, will frequent formats to identify bad blocks help?
- the card reader - can a card reader misread a flash card that frequently (hence people posting here about using multiple readers)? Are there different qualities of card readers?
- or a RED firmware issue? Is the RED sometimes writing bad data?
 
So is the real problem:
- the reliability of flash memory? If so, will frequent formats to identify bad blocks help?
- the card reader - can a card reader misread a flash card that frequently (hence people posting here about using multiple readers)? Are there different qualities of card readers?
- or a RED firmware issue? Is the RED sometimes writing bad data?

Every single time I have encountered corrupt footage (and sadly I have encountered it a lot) the problem always lies with the transfer to the computer. Whenever the original files have been left on the media, I've been able to get a perfect transfer myself, after other have had problems. So the camera is writing the data just fine, and the flash memory is recording it just fine, but the card reader is misreading frequently.

The idea behind R3D Data Manager is to read every card twice - once to generate checksums, and once to copy the data to the destination. The destination data is then compared to the checksum, and any variation reveals problems in one of the reads from the card, or the write to the destination.

The most common problem I have encountered with 16 GB CF cards is towards the end of a full card, or when a second card is downloaded immediately after the first card, with the same card reader, you start getting a lot of corrupt clips. Picking up that card reader and feeling the heat in it and the card it's been reading will make you think of overheating pretty quickly. Especially if a card read fails miserably with a hot reader, and after letting it cool down for a while you are able to get a perfect read off the same card with the same reader, you'll be convinced.

I've also encountered problems once it's entirely outside of the realm of RED. Let's face it, Finder and Explorer can utterly fail without any warning when copying files. A recent experience of mine was someone transferring CF cards to an internal laptop hard drive, and R3D Data Manager checksum verification passed perfectly. They then dragged n' dropped them to an external eSata hard drive using Finder. When I got the external hard drive, a checksum verification showed 70-80% of the clips were now corrupt. The problem was a MacBook Pro that, due to a design defect, didn't work properly with ExpressCards, and the eSata connection was an ExpressCard. The user had no idea Finder was brutally corrupting his clips, whereas using R3D Data Manager again when making that transfer would have warned him of the problem.

I've become so paranoid about computers now that my Windows PCs are all set up to use Teracopy with checksum verification on every single file copy I do, no matter how mundane. For me, it has completely replaced Explorer for copying and moving files.
 
Oopsie, 98.375% estimated reliability using my cockroach math :)

Not bad at all considering that is assuming only 0.1 of those with Red Ones are saying anything about it here!

Thanks for pointing that out. But hey, at least my brain can survive a nuclear holocaust :)
 
I've become so paranoid about computers now that my Windows PCs are all set up to use Teracopy with checksum verification on every single file copy I do, no matter how mundane. For me, it has completely replaced Explorer for copying and moving files.

And what about a generated checksum by the RED CAM when a new file is stored ? And a software bundled with RED STATIONS that will check the files after each copy (from RED STATION to any other media or HDD) ?
 
And what about a generated checksum by the RED CAM when a new file is stored ? And a software bundled with RED STATIONS that will check the files after each copy (from RED STATION to any other media or HDD) ?

I brought that up in the past (MD5 or SHA1 hash), no response yet but obviously the RED team is slammed just getting everything else done, plus unless you do it smart you may impact write speeds to the media (as you ideally would have to read it back from the media to checksum it and you dont want that to interfere with recording new footage)
 
We see every suggestion. We may not reply - but does not mean it falls on deaf ears. I cannot tell you how many little things we have picked up on REDUSER that we addressed in firmware, software, camera design, etc.

BC
 
...
Now, RED is transferring data at extreme rates in a device that is reasonably priced, so I can understand that corners probably needed to be cut in how the data is processed and stored. But that being said, a truly effective solution would result in being able to leave a CF card transferring and never have to worry about corrupt data. This is my expectation of technology.
So I am disappointed that the RED needs this extra attention of a DIT. This need for a DIT isn't in my mind a solution, it's a work-around.
...

A "DIT" is actually necessary to check what is shot is actually safe, not just data integrity, but image(& audio) integrity, but I get what you mean.
 
This is a professional tool, don't blame the manufacturer because they didn't see it necessary to make a 3rd grade level manual spelling everything out.

While I agree that these things do happen and that's why you always shoot an extra take for safety, RED has always presented the process as being incredibly simple. I remember some of the earliest demonstrations of Ted pulling the RED drive off the camera and saying, "See. It's a FW800 drive, so you can just plug it into the computer and start editing if you want to". Although that is technically accurate, we all know that NO serious production would ever work that way, but I think some people have gotten the impression over the years that it's so easy anyone can do it.

On the other hand, RED has started things like Reducation, demonstrating that you need specialized skills to make the most of the RED experience, but producers tend to not pay attention to those sorts of things and often just look for the cheapest solution.
 
Back
Top