HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Feb 06, 2012, 08:35 PM
Registered User
United States, IL, Grafton
Joined Jan 2012
243 Posts
I will probably be a little rusty though We had our own hobby shop when I was growing up, so think it will be fun to get back into it with these cameras!

Quote:
Originally Posted by headlessagain View Post
Just don't forget to video you flight. We all enjoy a good crash video You can have 3 on the plane and one on a hat cam. Should be great footage.
redrider67 is online now Find More Posts by redrider67
Reply With Quote
Sign up now
to remove ads between posts
Old Feb 06, 2012, 08:48 PM
So many planes, So little time
Xflyfun's Avatar
United States, OR, North Plains
Joined Jul 2010
232 Posts
Thanks here's the info.

Quote:
Originally Posted by Isoprop View Post
What is the class rating of your 8GB card? I would like to include this in the list of problematic cards.

Currently there is no solution for your 8GB card. For the moment it's a bit of a gamble on which cards work and which don't. I'm sure this issue will be corrected in a later firmware version, but for the time being your best bet would be to use Sandisk, Transcend or Kingston class 4, but even here there is no guarantee your card will work. However, to date no class 4 cards have been reported as not working and these are also the cards the developer uses for testing.
Sorry I didn't see that you answered my question
I tried a PNY 8GB Micro SD HC class 10
could not do in cam format with it either and only got about 30 seconds of video before shutdown.
Put in a Kingston Micro SD HC 4GB class 4 everything is good, incam format worked and could shoot five minutes plus of video no problems.
Xflyfun is offline Find More Posts by Xflyfun
Reply With Quote
Old Feb 06, 2012, 09:24 PM
Team WarpSquad
Japan, Tokyo
Joined Jun 2011
2,816 Posts
I just posted my microSDHC card performance data to the #1080 thread for those who are interested.

http://www.rcgroups.com/forums/showp...&postcount=367

My findings are similar to Xflyfun above - Class 4 is likely better than Class 10
thwaitm is online now Find More Posts by thwaitm
Reply With Quote
Old Feb 07, 2012, 01:40 AM
Registered User
Joined Nov 2010
2,494 Posts
Quote:
Originally Posted by Xflyfun View Post
Sorry I didn't see that you answered my question
I tried a PNY 8GB Micro SD HC class 10
could not do in cam format with it either and only got about 30 seconds of video before shutdown.
Put in a Kingston Micro SD HC 4GB class 4 everything is good, incam format worked and could shoot five minutes plus of video no problems.
Thanks for this info. I've updated the problematic cards posting.

I think you are the second person to also report you cannot in-camera format the problematic card. I've also added this interesting information.
Isoprop is offline Find More Posts by Isoprop
Reply With Quote
Old Feb 07, 2012, 02:51 AM
Registered User
finster's Avatar
Michigan
Joined Feb 2006
502 Posts
Problematic Cards

All them cards listed are problematic? Hmmm...sounds more like the camera not the cards...I dont remember having problems like that with the #11 cams...
finster is offline Find More Posts by finster
Reply With Quote
Old Feb 07, 2012, 03:02 AM
Registered User
Joined Nov 2010
2,494 Posts
Quote:
Originally Posted by wszim View Post
Only makes sense to write sequential observation. Because this type of record exists in the record stream (and probably for this type refers to the card class). Try to make the test program, Atto Disk Benchmark (freeware) and compare graphs.
I think you are missing something here.

I am no expert in the .mov file format, but common logic tells me that random writes are much more important than sequential writes. The incoming data may be a continuous stream, but a .mov file is certainly not written as such. The data structures are "linked" together, which means previously written data has to be updated. More important, a lot of extra administrative I/O is required to maintain a clean FAT (File Allocation Table). The FAT needs to be continuously read and updated in order to evaluate which sectors are free. As soon as a data block is to be written, the chain of free data sectors has to be updated. For those following the "FULL HD 1080p" thread, this is most probably the reason for the "lost" disk capacity - sectors are marked as being used but they are not indexed.

I would expect even a no-class, first generation card to easily handle the sequential write data speed of at least 10Mb/s but not necessarily the required random write speed. Actually, I would think most cards could not handle the camera's random write speed alone.

CrystalDiskMark is a good utility because it shows sequential and random writes. The camera uses a mixture of both, but since all cards are fast enough for sequential writes I believe the random write speed is more important.

Like I said, I am no expert in the structure of .mov files, but I do know how the FAT is designed to work.
Isoprop is offline Find More Posts by Isoprop
Last edited by Isoprop; Feb 07, 2012 at 04:54 AM.
Reply With Quote
Old Feb 07, 2012, 06:34 AM
Happy when flying!
kjrell's Avatar
France, Aquitaine
Joined Sep 2011
160 Posts
Another video, unfortunately it was very windy

Over the frostbitten lake (FPV 808#16) (7 min 1 sec)
kjrell is offline Find More Posts by kjrell
Reply With Quote
Old Feb 07, 2012, 07:40 AM
Registered User
Polska, małopolskie, Nowy Sącz
Joined Dec 2011
105 Posts
Quote:
Originally Posted by Isoprop View Post
I think you are missing something here.

I am no expert in the .mov file format, but common logic tells me that random writes are much more important than sequential writes. The incoming data may be a continuous stream, but a .mov file is certainly not written as such. The data structures are "linked" together, which means previously written data has to be updated. More important, a lot of extra administrative I/O is required to maintain a clean FAT (File Allocation Table). The FAT needs to be continuously read and updated in order to evaluate which sectors are free. As soon as a data block is to be written, the chain of free data sectors has to be updated. For those following the "FULL HD 1080p" thread, this is most probably the reason for the "lost" disk capacity - sectors are marked as being used but they are not indexed.

I would expect even a no-class, first generation card to easily handle the sequential write data speed of at least 10Mb/s but not necessarily the required random write speed. Actually, I would think most cards could not handle the camera's random write speed alone.

CrystalDiskMark is a good utility because it shows sequential and random writes. The camera uses a mixture of both, but since all cards are fast enough for sequential writes I believe the random write speed is more important.

Like I said, I am no expert in the structure of .mov files, but I do know how the FAT is designed to work.
Isoprop
You have a lot of reasons and a lot of knowledge when it comes to FAT. I am just in this topic I have less knowledge.
Note, however, that in our camera rather not implemented FAT exactly as on the PC, where it is multi-threaded operating system (with multiple access applications to disk). Compatible with FAT, of course, the structure must be maintained because the card must be properly read in an ordinary computer.
However, the implementation of the physical storage of data stream, the solution is probably in a specific way to take into account the scarcity of microcontroller memory, not high core speed etc. How a programmer exactly solved it? - We do not know. I can only guess, because I am in contact with low-level (simple BIOS) programming of microcontrollers that support SD cards. Rather, the system is single-threaded, rather synchronously and asynchronously at most work streams DMA to transfer data (my guesses). In any case, different methods and ways to ensure the smooth entry of data into NAND-flash cards, with little RAM resources, arranging them correctly in the structure of the FAT and others.

In my post I referred to the card class. SD Card Association https://www.sdcard.org/developers/overview/speed_class/
came up with a definition of card speed: "Speed ​​Class and UHS Speed ​​Class symbols indicate the minimum writing performance is the ensure smooth writing with a dry content of streaming video as shooting. There is no definition of it is to take place on the FAT file system or another. There are different ways to improve the transfer of not only the modernization of NAND-FLASH technologies but also by other solutions. Among other things, the use of special processing controllers: CMD25 = Multiple Block Write Operation, ACMD23 = Pre-erase setting prior to a multiple block write operation, the switch to high speed bus to CMD6 mode allowing to work on the bus clocked twice as fast, ie up to 50MHz.
Note on my charts ATTO that using only the large amount of data (ie sequential write) provides a transfer that corresponds to the class. Because only then can use all the clever ways of filling a rather slow NAND-FLASH.

Of course I do not want to outwit. That is not what I mean. I do not have to be right. I care only on the substantive exchange of information.
wszim is offline Find More Posts by wszim
Last edited by wszim; Feb 07, 2012 at 03:16 PM.
Reply With Quote
Old Feb 07, 2012, 09:51 AM
Registered User
Joined Nov 2010
2,494 Posts
Early Cutoff and formatting issues

Tom and I have received an email from the developer concerning the list of problematic cards (certain high-speed cards with which the recording stops after about 50-90 seconds) and has asked if we can provide more information.
This post is mainly for the members linked on the original post, but if others have similar problems with their cards please post. The more information the developer has, the easier it will be to find the problem.
kjrell
jackdaws
downhill
ds64
Xflyfun

The developer is taking the early cutoff issue seriously but needs more information if possible.
1. What kind of computer / OS is used?
2. Were the cards purchased from a reliable source and have they passed the h2testw test?
3. Was the in-camera formatting done correctly?

The developer has also found a small issue with the in-camera formatting which at least one person reported. For some reason, only on some cards, there's no LED indicator while holding the shutter button down during formatting. However, the card is correctly formatted . If you have this problem, here's a temporary workaround:
- Connect the camera as an external drive.
- Copy or create a small file on to the drive - this will force the red LED on.
- While the red LED is on, press and hold the shutter button.
- Keep holding the shutter button and disconnect the camera.
- Still with the shutter button pressed, wait about 4-6 seconds until the red LED turns off.
- Release the shutter button.

So, once again, if anyone else is having issues with early cutoff please post the card manufacturer, the size of the card, the class rating, as much relevant info as possible about the computer and OS, and if the card was purchased from a reliable source (I hope so!).
Isoprop is offline Find More Posts by Isoprop
Last edited by Isoprop; Feb 07, 2012 at 11:05 AM.
Reply With Quote
Old Feb 07, 2012, 11:40 AM
Registered User
Deutschland, BY, Kulmbach
Joined May 2011
37 Posts
I now put a 8GB class6 Transcend card in my #16 an did the in cam formatting. It worked fine So, Im not the problem

traxdata 16GB class6 works in the # 16 if I used the sdformatter, Samsung 16GB class6 does not work with #16 at all

I took a look on the cards after the in cam formatting. The formatting was done. Both 16GB cards are empty.

The developer has also found a small issue with the in-camera formatting which at least one person reported. For some reason, only on some cards, there's no LED indicator while holding the shutter button down during formatting. However, the card is correctly formatted . If you have this problem, here's a temporary workaround:
- Connect the camera as an external drive.
- Copy or create a small file on to the drive - this will force the red LED on.
- While the red LED is on, press and hold the shutter button.
- Keep holding the shutter button and disconnect the camera.
- Still with the shutter button pressed, wait about 4-6 seconds until the red LED turns off.
- Release the shutter button.

this doesnt work with any of my 16GBcards class 6 (Samsung/traxdata)

Ive done the ChrystalDiskMark with every of my three cards after formatting them in the cam although the camera did not indicate the end.The 8GB transcend card did indicate the end of formatting. Maybe some of you can tell me how I have to interpret the numbers

Name: transcend 8GB class6 after in cam from..jpg
Views: 60
Size: 40.5 KB
Description:

Name: Samsung 16GB class6 after in cam form..jpg
Views: 61
Size: 40.1 KB
Description:

Name: traxdata 16GB class6 after in cam form..jpg
Views: 48
Size: 40.6 KB
Description:
downhill is offline Find More Posts by downhill
Reply With Quote
Old Feb 07, 2012, 11:54 AM
Happy when flying!
kjrell's Avatar
France, Aquitaine
Joined Sep 2011
160 Posts
I'm using self built computer with w7 x64
Silicon Power 4GB class 10 has been purchased in a reliable shop
The in-camera formatting was done correctly

This card was previously used in a 808#11 without any problem
Was also working with #16 0.15 firmware, shutting down itself but later like 15+ minutes

Here is Silicon Power 4GB class 10 tests:
kjrell is offline Find More Posts by kjrell
Reply With Quote
Old Feb 07, 2012, 02:14 PM
Registered User
USA, OH, Marysville
Joined May 2010
373 Posts
Has anyone try using a cell phone charger to charge there #16? I have one that is 5v at 850ma will that work or burn up the Lipo battery.
jmdlcar is offline Find More Posts by jmdlcar
RCG Plus Member
Latest blog entry: Binding VLToys V212 Quadcopter
Reply With Quote
Old Feb 07, 2012, 02:30 PM
Dance the skies...
Tom Frank's Avatar
United States, MA, Walpole
Joined Dec 2003
17,471 Posts
Quote:
Originally Posted by downhill View Post
...
Ive done the ChrystalDiskMark with every of my three cards after formatting them in the cam although the camera did not indicate the end.The 8GB transcend card did indicate the end of formatting. Maybe some of you can tell me how I have to interpret the numbers
Attachment 4614869
I also have a Transcend 8GB CL 6 card like yours which functions as it should. I tested mine with both the 32 bit and 64 bit versions of the CDM utility on my W7(64 bit) PC, and posted the results with yours just for a comparison (see attached pic...your test is in the middle). What can be determined by these numbers?... maybe only that seemingly identical cards are not so identical in benchmark test, and probably not when in use either! This seems to coincide with users findings. The test numbers will show differently each time the test is run. I did the in-camera formatting just prior to the each test, so the card was "clean" each time. It's obvious from the sequential read/write tests that our CL6 cards easily surpass CL10 speeds and could be labelled as CL10 if the manufacturer chose to do so.

But we don't know how the camera writes data to the cards. If the consensus thinking is valid, there's a good probability that random writes occur with digital video, especially as the card starts to accumulate data on it. From some routine video I've shot with the camera set at the lower data rate (7Mbps), I've gotten instantaneous brief peak data rates as high as 25Mbps, even though the average for the entire clip was around 8.3 Mbps. The 25 Mbps rate (bits/sec) equates to 3.125MBps (bytes/sec), and even the average data rate equates to just over 1 MBps. So if random writes do occur, it would seem from looking at the 512k random write test numbers that there could be some issues with some writes. That that might show up as a dropped frame, a bit of "garbage" in a frame, or possibly a corrupted file. I have seen a "brief blip of garbage" in a portion of a video frame in rare cases, and even a rare corrupted file (not counting the recent corrupted file problem that was fixed with the v0.18 firmware), but largely have not had any major problems with my cards. But most of my videos do not show average data rates over the normal 7Mbps... it takes bright scenes with lots of detail, lots of contrast, and constant motion to push the data rate up beyond 7Mbps. But for these reasons, I like to see the 512K random write speeds above 1.0. The lower speed class cards typically will noticeably outperform higher speed class cards in the random writes, often by a very wide margin, and this seems to follow the general observation that CL4 cards are not only fast enough, but also have better compatibility than the faster cards. Is this due to much higher to random write speeds?... it's anybody's guess at this point. Since the cards have their own controllers on board that must speak the same language as the camera, my personal view is the compatibility issues are more related to that, and I don't conclude that only the camera is at fault.
Tom Frank is online now Find More Posts by Tom Frank
Last edited by Tom Frank; Feb 07, 2012 at 02:31 PM. Reason: added attachment
Reply With Quote
Old Feb 07, 2012, 02:37 PM
Fidler & twidler
empeabee's Avatar
Cranfield U.K.
Joined Mar 2004
3,986 Posts
Quote:
Originally Posted by jmdlcar View Post
Has anyone try using a cell phone charger to charge there #16? I have one that is 5v at 850ma will that work or burn up the Lipo battery.
Please read Posts #2 to 5 at the start of the thread, they contains answers to most FAQ's
Mike
empeabee is offline Find More Posts by empeabee
Reply With Quote
Old Feb 07, 2012, 02:54 PM
Registered User
Deutschland, BY, Kulmbach
Joined May 2011
37 Posts
@Tom
Many text for me to read. I have to take a dictionary to understand all you wrote. My Englishlessons in school are long before....

I forgot to tell somethig about my OS. I use a Netbook "Medion" E1210, XP.
All cards are purchased from reliable source. And here some more tests:
downhill is offline Find More Posts by downhill
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion The REAL (#11) HD Key Cam Thread (PLEASE! READ POSTS #1-#3 BEFORE POSTING QUESTIONS!) Tom Frank Aerial Photography 9936 Feb 11, 2014 09:56 AM
Discussion Looking to buy #11 HD Key Cam within US Abolfazl Aerial Photography 1 Feb 14, 2012 01:59 AM
Poll Poll For Current and Future #11 HD Key Cam Users Tom Frank Aerial Photography 0 Sep 25, 2011 11:55 PM
Discussion New HD Key Cam CHAMP? 1080x720p 3 HOUR RECORD TIME! Jaybee Aerial Photography 4 Jun 02, 2011 11:39 AM