Become a Site Supporter and Never see Ads again!

Author Topic: Tascam DR-70D 4-channel audio recorder (Part 4)  (Read 105444 times)

0 Members and 1 Guest are viewing this topic.

Offline morst

  • I think I found an error on the internet; #UnionStrong
  • Trade Count: (2)
  • Needs to get out more...
  • *****
  • Posts: 5950
Re: Tascam DR-70D 4-channel audio recorder (Part 4)
« Reply #390 on: October 01, 2015, 03:24:16 PM »
A BBB filing?  Seriously?  :facepalm:

Yes.

PS are we trying for page 28?
https://toad.social/@morst spoutible.com/morst post.news/@acffhmorst

Offline voltronic

  • Trade Count: (40)
  • Needs to get out more...
  • *****
  • Posts: 4104
Re: Tascam DR-70D 4-channel audio recorder (Part 4)
« Reply #391 on: October 01, 2015, 06:21:08 PM »
Again, I think Tom's answer on this was satisfactory, in that he stated he doesn't have access to design docs and isn't an engineer.  The statement of "cards slow down over time" absolutely makes sense if you understand how flash memory works and physically degrades over time.  The cache memory on a HDD isn't as direct an analogy as you may think, because a HDD doesn't physically degrade in the same way as flash memory.  A HDD can certainly develop bad sectors, but generally with good disk management the performance of the disk overall isn't going to degrade all that much unless it's an extreme situation.  This is not the case with flash memory - when things start wearing out (and they will), you will start to see performance degradation on a more global scale.  SSD drives employ "wear leveling" to help combat this, but SD cards do not. 

Are you sure about this?  I am virtually certain that modern SD cards, at least those from reputable companies, implement wear leveling.  SanDisk has been doing it for more than decade, I think.  Even if you assume no wear leveling, however, SD card sectors can do several thousand write/erase cycles before failure.  That's equivalent to writing/erasing the card daily for years.  I have no doubt that cards fail, and that they will all fail at some point, but I think the probability that card wear is the issue with so many people/cards experiencing problems is vanishingly small.

Wear leveling is not in the official SD spec as far as I know.  If SanDisk has been using it, it would likely be on their top-line cards. 

I think "several thousand" may be a bit generous - I've seen a variety of estimates as low as 10,000 cycles per sector.  That said, I think card wear is probably not the primary culprit for all of the people having problems, though it may be for some.

If the buffer memory on the 70D is significantly smaller than that of the 680 or other multichannel recorders, then to me it's quite logical that the 70D is thus more "picky" with memory cards because cards because a device with a larger write buffer will be able to better compensate for the reduced physical performance of older and/or new but lower-performing cards that now have fewer good sectors to write to.  That problem is amplified when recording multiple simultaneous data streams as is the case in audio recording.  This is just a theory on my part, but I'd say it's a plausible one.  Remember that a card's rating from the manufacturer is based on burst reads/writes; not continuous streams of data.  Therefore a card may perform like a Class 10 in a DSLR, but absolutely suck wind in an audio recorder because of the difference in how it's being used - even if it's brand new.  That's where media testing becomes important, and where I think Tascam should have had a good list of tested cards ready to go out of the gate upon product release.

I would say that if the buffer is too small to record audio (if that's the issue in the first place), that is a de facto design flaw in an audio recorder.  I wonder if that type of problem would be fixable with firmware?   

With respect to the bolded, these things are usually classed by sustained write speed.  A lot of people use SD cards to record HD video, which is not only continuous, but also considerably more data than even four channel 24/96. 

Interesting to see that a number of class 2 and 4 cards are on the European website.  That would seem to imply that these issues (buffers and write speeds) are not the problem...

Yes, I would agree with you that too small a buffer would be a flaw.  Back in my PC repair / building days I was obsessed with getting the very best CD burners.  One of the things that separated the quality units from the cheap ones was the quality and capacity of their write buffers.

The idea that HD video is more of a data rate than high-res audio is a common misconception.  The individual frames of HD video are pretty small megapixel-wise.  Still high-res photography is probably more taxing, especially if you're shooting in burst mode, and according to Tom, audio recording is more stressful than both.
http://taperssection.com/index.php?topic=173960.msg2160546#msg2160546
http://www.alphr.com/features/380167/does-your-camera-need-a-fast-sd-card

And some good FAQs from the offical SD Card people: https://www.sdcard.org/consumers/speed/max_speed/index.html
I am hitting my head against the walls, but the walls are giving way.
- Gustav Mahler

Acoustic Recording Techniques
Team Classical
Team Line Audio
Team DPA

Offline aaronji

  • Site Supporter
  • Trade Count: (9)
  • Needs to get out more...
  • *
  • Posts: 3861
Re: Tascam DR-70D 4-channel audio recorder (Part 4)
« Reply #392 on: October 01, 2015, 06:41:11 PM »
Wear leveling is not in the official SD spec as far as I know.  If SanDisk has been using it, it would likely be on their top-line cards. 

Yes, their top of the line cards a decade ago.  No, it is not in the official specs for, at least, HD.  I am not sure about SDHC or SDXC.  Not that the spec matters if most companies implement it anyway.

I think "several thousand" may be a bit generous - I've seen a variety of estimates as low as 10,000 cycles per sector.  That said, I think card wear is probably not the primary culprit for all of the people having problems, though it may be for some.

How is several thousand more generous than ten thousand?  At ten thousand cycles, that is every day for more than 27 years...

The idea that HD video is more of a data rate than high-res audio is a common misconception.  The individual frames of HD video are pretty small megapixel-wise.  Still high-res photography is probably more taxing, especially if you're shooting in burst mode, and according to Tom, audio recording is more stressful than both.

Maybe, although I would like to see a reference.  Four channel 24/96 is 1.1 MB/s; I think HD is more.  Of course, if Tom says otherwise...

stevetoney

  • Guest
  • Trade Count: (0)
Re: Tascam DR-70D 4-channel audio recorder (Part 4)
« Reply #393 on: October 01, 2015, 08:00:12 PM »
The idea that HD video is more of a data rate than high-res audio is a common misconception.  The individual frames of HD video are pretty small megapixel-wise.  Still high-res photography is probably more taxing, especially if you're shooting in burst mode, and according to Tom, audio recording is more stressful than both.

Maybe, although I would like to see a reference.  Four channel 24/96 is 1.1 MB/s; I think HD is more.  Of course, if Tom says otherwise...

I chose randomly a hi res video camera and looked in the manual.  Canon Vixia HFS10 has 32gb of built in flash memory
MXP Mode - 2hr 55min
FXP Mode - 4hr 10min
XP+ Mode - 5hr 45min
SP Mode - 9hr 35min
LP Mode - 12hr 15min

So at the lowest resolution, the data consumption rate is something less that 3gb per hour (32gb/12.25hr).
At the highest resolution, the data consumption rate is a little over 10gb per hour (32gb/3hr)

Since audio eats data at around 1hr per gb at 24/48 for 2 channels, at the highest audio rate of 24/96 and 4 channels, the peak data usage rate of the DR70D rate is something close to 4gb/hour.  At 24/48 the rate of course is 2gb/hr.

Conclusion:  There's some overlap, but in general HD video uses more data than HD audio.
« Last Edit: October 01, 2015, 08:02:42 PM by tonedeaf »

Offline voltronic

  • Trade Count: (40)
  • Needs to get out more...
  • *****
  • Posts: 4104
Re: Tascam DR-70D 4-channel audio recorder (Part 4)
« Reply #394 on: October 01, 2015, 08:08:42 PM »
The idea that HD video is more of a data rate than high-res audio is a common misconception.  The individual frames of HD video are pretty small megapixel-wise.  Still high-res photography is probably more taxing, especially if you're shooting in burst mode, and according to Tom, audio recording is more stressful than both.

Maybe, although I would like to see a reference.  Four channel 24/96 is 1.1 MB/s; I think HD is more.  Of course, if Tom says otherwise...

I chose randomly a hi res video camera and looked in the manual.  Canon Vixia HFS10 has 32gb of built in flash memory
MXP Mode - 2hr 55min
FXP Mode - 4hr 10min
XP+ Mode - 5hr 45min
SP Mode - 9hr 35min
LP Mode - 12hr 15min

So at the lowest resolution, the data consumption rate is something less that 3gb per hour (32gb/12.25hr).
At the highest resolution, the data consumption rate is a little over 10gb per hour (32gb/3hr)

Since audio eats data at around 1hr per gb at 24/48 for 2 channels, at the highest audio rate of 24/96 and 4 channels, the peak data usage rate of the DR70D rate is something close to 4gb/hour.  At 24/48 the rate of course is 2gb/hr.

Conclusion:  There's some overlap, but in general HD video uses more data than HD audio.

OK, but is the camera writing that data in chunks or in one continuous stream?  That was the point I was raising - obviously HD video is writing more data over a given period of time on average, but it's not necessarily continuous.  I think that's where Tom was headed with his comment I linked a few posts up.  As I understand it, recording PCM audio is more of the constant-stream variety.
I am hitting my head against the walls, but the walls are giving way.
- Gustav Mahler

Acoustic Recording Techniques
Team Classical
Team Line Audio
Team DPA

Offline voltronic

  • Trade Count: (40)
  • Needs to get out more...
  • *****
  • Posts: 4104
Re: Tascam DR-70D 4-channel audio recorder (Part 4)
« Reply #395 on: October 01, 2015, 08:09:03 PM »
I am hitting my head against the walls, but the walls are giving way.
- Gustav Mahler

Acoustic Recording Techniques
Team Classical
Team Line Audio
Team DPA

 

RSS | Mobile
Page created in 0.069 seconds with 34 queries.
© 2002-2024 Taperssection.com
Powered by SMF