Become a Site Supporter and Never see Ads again!

Author Topic: Portable Larvy AD10?  (Read 23988 times)

0 Members and 1 Guest are viewing this topic.

Offline Javier Cinakowski

  • !! Downhill From Here !!
  • Trade Count: (11)
  • Needs to get out more...
  • *****
  • Posts: 4325
  • Gender: Male
Re: Portable Larvy AD10?
« Reply #60 on: April 09, 2008, 09:58:54 PM »
Quote
Of course, the user would have to know the bit depth and sample rate. Where is that information stored? Is it in the WAV file itself? I would hope such is the case.

yep, the wav file has a header that indicates bit depth and sample rate....  Whatever you record on your device will transfer to windows without problems
Neumann KM185mp OR DPA ST2015-> Grace Design Lunatec V2-> Tascam DR-100mkIII

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (15)
  • Needs to get out more...
  • *****
  • Posts: 15700
  • Gender: Male
  • "Better to love music than respect it" ~Stravinsky
Re: Portable Larvy AD10?
« Reply #61 on: April 10, 2008, 12:10:51 AM »
Quote
Of course, the user would have to know the bit depth and sample rate. Where is that information stored? Is it in the WAV file itself? I would hope such is the case.

yep, the wav file has a header that indicates bit depth and sample rate....  Whatever you record on your device will transfer to windows without problems

and that file is transfered like any other data at a rate determined by the memory read & bus speed.  But only after it has been recorded in the WAV format with proper header info. Not in real time.
musical volition > vibrations > voltages > numeric values > voltages > vibrations> virtual teleportation time-machine experience
Better recording made easy - >>Improved PAS table<< | Made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<< (note: This is a 1st draft, now several years old and in need of revision!  Stay tuned)

Offline it-goes-to-eleven

  • Trade Count: (58)
  • Needs to get out more...
  • *****
  • Posts: 6696
Re: Portable Larvy AD10?
« Reply #62 on: April 10, 2008, 07:33:47 AM »
Slap on the screen, expansion board for storage and some buttons.  It's making the leap from the output of the internal AD to something like the gumstix is where I get lost.

Yeah, so does M-audio.

Offline mblindsey

  • Trade Count: (23)
  • Taperssection All-Star
  • ****
  • Posts: 1245
Re: Portable Larvy AD10?
« Reply #63 on: April 10, 2008, 08:12:12 AM »
Slap on the screen, expansion board for storage and some buttons.  It's making the leap from the output of the internal AD to something like the gumstix is where I get lost.

Yeah, so does M-audio.


+T

I forgot to mention that QNX is probably suitable for building the type of device I was mentioning:

http://www.qnx.com/products/middleware/multimedia.html

--Michael
« Last Edit: April 10, 2008, 09:35:22 AM by mblindsey »
Mics:  Microtech Gefell m200/M20/nBob Actives>PFA, CA-11
Pre's: USB Pre2, 1x V3, 2x V3 w/optimod, MP2, Church Ugly
Decks: SD MixPre 6 II, R44 Oade Concert Mod, M10
Playback: Grace m9xx->Sen HD 650, Fostex TH-X00, HIFIMAN HEXX
Mixing: RME Fireface UFX->Reaper/Izotope->Yamaha HS8

Offline Dan Lavry

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 19
Re: Portable Larvy AD10?
« Reply #64 on: April 10, 2008, 01:34:13 PM »

I've often wondered how difficult it would be for a manufacturer of a pre/ad to add the "storage" component to an existing device.  I mean, there are already companies building very small embedded Linux devices.  Here is one:

http://gumstix.com/platforms.html

Slap on the screen, expansion board for storage and some buttons.  It's making the leap from the output of the internal AD to something like the gumstix is where I get lost. 

The Rockbox folks have probably already written the code to write all the important file types, as well as, handle most all the other functions we care about.  I haven't gone through it, but it's here: http://build.rockbox.org/

inside one small case => {nice pre's > Larvy AD > gumstix running modified rockbox > flash}

Does this sound far fetched? 

--Michael

Thanks for the post. Here are my initial thoughts:

The modules you point out seem to be aimed at very high speeds, at lower capacity. For audio storage, the speed for say 96KH stereo at 24 bits (which is really 6 bytes in "computer language") is near 600KB/sec. Given that one wants some margin, I would want at least 1MB/sec write speed. There is no need for 400MHz clock speed...
Also, there are all sorts of memory types. Most of the modules oriented for computer applications are volatile RAM. Volatile means the data is lost when you turn the power off, which is not good for a recorder. RAM means "random access memory", which is not needed for audio, because we "park the data" sequentially ("first come first served").
The type of memory needed is non volatile and sequential. The memory size of a standard CD (16 bits stereo at 44.1KHz) is around 650MB. So for the same CD length of time, at 24 bits you need 1/3 more, and for 96KHz 24bits you need around 2.5 times more memory... you are quickly approaching 2GB capacity...

Hard drives have huge capacity, but they tend to have limited life, and are a bit on the fragile side, though things have been improving a lot due to advancements in laptops. I like memory sticks and other flash memory technology, though they have their issues as well. With most memories, one can simply re write new data over the old data. But with flash, one needs to first erase the intended area (sort of like formatting the old floppy).

There are many other issues under consideration. It is not all that simple to design a digital device inside analog circuits.

It is always a challenge to put digital circuits near sensitive analog circuits, and with a mic pre, the challenge is the greatest. The electrical "environment" has to be extremely free of electrical interference.   
As a rule, digital circuits are "noise machines", which is not an issue for all digital gear, such as computers.
But it is an issue for converters and even more so for micpre's. When your micpre is set to say 60dB gain, an input signal of 1mV (from the mic) will be amplified by 1000  (you get 1V). Clearly a 1mV digital noise getting in to the input will yield a 1V of unwanted noise... It is always a challenge to put digital circuits near sensitive analog circuits, and with a mic pre, the challenge is the greatest.

There are many more considerations and issue, but the post is getting pretty long. As you probably expected, some of the issues have known solutions, others require innovation and/or a lot of care.

Regards
Dan Lavry     
www.lavryengineering.com

Offline Dan Lavry

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 19
Re: Portable Larvy AD10?
« Reply #65 on: April 10, 2008, 01:38:49 PM »
Quote
Of course, the user would have to know the bit depth and sample rate. Where is that information stored? Is it in the WAV file itself? I would hope such is the case.

yep, the wav file has a header that indicates bit depth and sample rate....  Whatever you record on your device will transfer to windows without problems

Thank you.

That is good news. I will look into the details of the format.

Does anyone use the Mac format (AIFF)? Does the Mac accept WAV files?

Dan Lavry
www.lavryengineering.com

Offline bgalizio

  • Trade Count: (3)
  • Needs to get out more...
  • *****
  • Posts: 3555
  • Gender: Male
    • http://www.archive.org/bookmarks/spyboychoir
Re: Portable Larvy AD10?
« Reply #66 on: April 10, 2008, 01:40:38 PM »
Quote
Of course, the user would have to know the bit depth and sample rate. Where is that information stored? Is it in the WAV file itself? I would hope such is the case.

yep, the wav file has a header that indicates bit depth and sample rate....  Whatever you record on your device will transfer to windows without problems

Thank you.

That is good news. I will look into the details of the format.

Does anyone use the Mac format (AIFF)? Does the Mac accept WAV files?

Dan Lavry
www.lavryengineering.com

Mac audio editing software will accept WAV files.

Offline mblindsey

  • Trade Count: (23)
  • Taperssection All-Star
  • ****
  • Posts: 1245
Re: Portable Larvy AD10?
« Reply #67 on: April 10, 2008, 02:17:35 PM »

Also, there are all sorts of memory types. Most of the modules oriented for computer applications are volatile RAM. Volatile means the data is lost when you turn the power off, which is not good for a recorder. RAM means "random access memory", which is not needed for audio, because we "park the data" sequentially ("first come first served").
The type of memory needed is non volatile and sequential. The memory size of a standard CD (16 bits stereo at 44.1KHz) is around 650MB. So for the same CD length of time, at 24 bits you need 1/3 more, and for 96KHz 24bits you need around 2.5 times more memory... you are quickly approaching 2GB capacity...

I was thinking that you would need both RAM and flash.  A small Operating System of some sort on EEPROM/flash (like on most portable multimedia devices...iPod, M-Audio Mircrotrack, Edirol R-09) that runs in RAM, and removable Flash to write the *.wav file to.  You'll need some form of an OS to display of recording time to a screen, start/stop writing of files, let user delete files, etc.

With most memories, one can simply re write new data over the old data. But with flash, one needs to first erase the intended area (sort of like formatting the old floppy).

Are you referring to the wear-leveling algorithms built into flash memory?  Most flash memory sectors can be written to a finite amount of times.  This causes problems for things like file allocation tables that get written to frequently.  Most of these devices act like they are normal IDE/ATA devices, and they abstract the fact that they are using a wear-leveling algorithm to write data evenly across the whole device.  I don't think you have to erase before you write. 

Side Note:  This has big implications for forensics on these types of devices as you may think you are overwriting a file when in reality the wear-leveling algorithm is writing your bits somewhere else on the drive...thus, leaving data you thought was overwritten still around for investigators to find.  :o

--Michael

Mics:  Microtech Gefell m200/M20/nBob Actives>PFA, CA-11
Pre's: USB Pre2, 1x V3, 2x V3 w/optimod, MP2, Church Ugly
Decks: SD MixPre 6 II, R44 Oade Concert Mod, M10
Playback: Grace m9xx->Sen HD 650, Fostex TH-X00, HIFIMAN HEXX
Mixing: RME Fireface UFX->Reaper/Izotope->Yamaha HS8

Offline Dan Lavry

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 19
Re: Portable Larvy AD10?
« Reply #68 on: April 10, 2008, 04:20:01 PM »

Also, there are all sorts of memory types. Most of the modules oriented for computer applications are volatile RAM. Volatile means the data is lost when you turn the power off, which is not good for a recorder. RAM means "random access memory", which is not needed for audio, because we "park the data" sequentially ("first come first served").
The type of memory needed is non volatile and sequential. The memory size of a standard CD (16 bits stereo at 44.1KHz) is around 650MB. So for the same CD length of time, at 24 bits you need 1/3 more, and for 96KHz 24bits you need around 2.5 times more memory... you are quickly approaching 2GB capacity...

I was thinking that you would need both RAM and flash.  A small Operating System of some sort on EEPROM/flash (like on most portable multimedia devices...iPod, M-Audio Mircrotrack, Edirol R-09) that runs in RAM, and removable Flash to write the *.wav file to.  You'll need some form of an OS to display of recording time to a screen, start/stop writing of files, let user delete files, etc.

With most memories, one can simply re write new data over the old data. But with flash, one needs to first erase the intended area (sort of like formatting the old floppy).

Are you referring to the wear-leveling algorithms built into flash memory?  Most flash memory sectors can be written to a finite amount of times.  This causes problems for things like file allocation tables that get written to frequently.  Most of these devices act like they are normal IDE/ATA devices, and they abstract the fact that they are using a wear-leveling algorithm to write data evenly across the whole device.  I don't think you have to erase before you write. 

Side Note:  This has big implications for forensics on these types of devices as you may think you are overwriting a file when in reality the wear-leveling algorithm is writing your bits somewhere else on the drive...thus, leaving data you thought was overwritten still around for investigators to find.  :o

--Michael




I have designed with all sorts of RAM and various types of ROM in the last 30 years, but my experience with flash is limited. I do use flash memory inside my Altera CPLD to store operating parameters, which are user settable, and one does need to erase a whole sector before writing anything (even a single bit). You can alter a single 0 into 1, but can not alter a single 1 into a 0 (I may have it upside down but that does not matter). You must "reset" a whole sector to all 1's (or maybe it is all 0's) before you write into it.

Yes, there are limitations regarding the number of times you can re write. I believe the specific hardware I use allows over 10000 writes, and that is plenty good for parameter settings such as how many bits, what sample rate you use... You can change it 3 times a day for 100 years.... But such devices with so many cycles offer very tiny amount of memory, and are also very slow. Many of the larger capacity devices have far fewer cycles, and recording music may be done much more often then altering a converter setting such as - to turn on or off a "peak hold" function, an internal/external clock mode....

That "wear leveling" may be good when you have a lot of memory, and use only some of it at a time. But say you have 2GB and you use 24bit/96KHz stereo for an hour. You used half the space, and the next time you get to use the other half... The third time you are already rewriting...

So why not go for say 20GB with "wear leveling"? It may be great but there are tradeoffs - First, it would increase cost a lot, and second, it may increase power consumption, which is not ideal for battery operated gear. I am not making final decisions here, because I still have a lot to learn here, and the flash technology is evolving very fast.

I really like the little portable memory sticks. I can backup whole projects, stick the device in a usb port, and there it is, ready to work with… The cost of a 2-4GB device has gone way down in the last couple of years. Too bad that they are too slow for uncompressed music.

I also often work with other type novram (non voltile ram) that does not require pre erasing and offers almost endless cycles, at tiny power and good enough speed. There is tons of such stuff on the market, But the memory capacity is way too tiny.   

Regards
Dan Lavry   

« Last Edit: April 10, 2008, 04:23:18 PM by Dan Lavry »

Offline mblindsey

  • Trade Count: (23)
  • Taperssection All-Star
  • ****
  • Posts: 1245
Re: Portable Larvy AD10?
« Reply #69 on: April 10, 2008, 04:35:02 PM »

What's your opinion of building a device that uses the QNX real time OS?

http://www.qnx.com/products/middleware/multimedia.html


--Michael
Mics:  Microtech Gefell m200/M20/nBob Actives>PFA, CA-11
Pre's: USB Pre2, 1x V3, 2x V3 w/optimod, MP2, Church Ugly
Decks: SD MixPre 6 II, R44 Oade Concert Mod, M10
Playback: Grace m9xx->Sen HD 650, Fostex TH-X00, HIFIMAN HEXX
Mixing: RME Fireface UFX->Reaper/Izotope->Yamaha HS8

Offline scb

  • Eli Manning should die of gonorrhea and rot in hell. Would you like a cookie, son?
  • Trade Count: (11)
  • Needs to get out more...
  • *****
  • Posts: 8676
  • Gender: Male
Re: Portable Larvy AD10?
« Reply #70 on: April 10, 2008, 04:58:08 PM »
But here we are just copying files so I would want to provide up to 24 bits, at sample rates from 44.1 to 96KHz. Of course, the user would have to know the bit depth and sample rate. Where is that information stored? Is it in the WAV file itself? I would hope such is the case.

the wav format stores bit depth and sample rate in the header info

check this site for info on the header format:

http://ccrma.stanford.edu/CCRMA/Courses/422/projects/WaveFormat/

Offline Dan Lavry

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 19
Re: Portable Larvy AD10?
« Reply #71 on: April 10, 2008, 09:27:39 PM »

What's your opinion of building a device that uses the QNX real time OS?

http://www.qnx.com/products/middleware/multimedia.html


--Michael

I do not know much about QNX.

I rather use some "plug in" portable small device, and whatever it may be, it should take the data in at "real time". I would like WAV (and possibly AIFF for the Mac), since it does not require special drivers to be installed.  When transferring the data from memory to computer, real time is not needed. In fact it would be best to go as fast as possible, as long as the transfer is reliable (that is a whole other issue).

Do you think the QNX offers something that I missed?

Regards
Dan Lavry

Offline mblindsey

  • Trade Count: (23)
  • Taperssection All-Star
  • ****
  • Posts: 1245
Re: Portable Larvy AD10?
« Reply #72 on: April 10, 2008, 11:05:24 PM »

What's your opinion of building a device that uses the QNX real time OS?

http://www.qnx.com/products/middleware/multimedia.html


--Michael

I do not know much about QNX.

I rather use some "plug in" portable small device, and whatever it may be, it should take the data in at "real time". I would like WAV (and possibly AIFF for the Mac), since it does not require special drivers to be installed.  When transferring the data from memory to computer, real time is not needed. In fact it would be best to go as fast as possible, as long as the transfer is reliable (that is a whole other issue).

Do you think the QNX offers something that I missed?

Regards
Dan Lavry


I guess what I see are two possible options:

1.  Build a device that outputs both analog and digital out in the ways people have expressed they would like to get their output in this thread.  Your device would pretty much be in direct competition with the Grace Designs Lunatec V3 and Apogee's MiniMe (some others).  In this scenario, you wouldn't have to worry about providing storage, and I assume, would be quicker to market.  Buyer's would purchase a recording device that has either S/PDIF in or Line In that would write the file to removable media.  Most here have that already.

2.  Try to build a device that incorporates storage in addition to the pre's and AD.  I think this is more difficult.  I'm not an expert in this area, but I think you would have to incorporate some type of OS, be it embedded Linux, something custom, or something like QNX.  The embedded OS would be responsible for displaying things to a screen, play/stop/record/pause/ff/rw, writing files to a removable media, allowing a user to delete files, keeping track of/displaying time, possibly displaying levels, etc.  A device like this would possibly compete with the Edirol R-4/R-44/R-09, M-Audio Microtrack, Sound Devices 7xx, Korg MR-1, Marantz PMD 660, Fostex FR2-le, Tascam HD-P2/DR 1, and a host of others.  If you build the device in option number one, it would go in front of a device like those mentioned in the previous sentence to provide boutique pre's and/or AD.

I mentioned QNX because I know it to be a small, fast, extremely reliable OS that is used in critical commercial applications, and it looks like they are now branching into audio.  If I were building a recording device, I would look to something like QNX to keep development time to a minimum.

I don't want to discourage building an "all in one" solution, but if you want to use some "plug in" portable small device, and whatever it may be, it should take the data in at "real time"...you can do that by building a quality device that provides both analog and digital outs.  There are devices in the market that will take the feed and write the file in multiple formats to removable media in real time (see those in option #2).  I'm guessing that both option #1 and option #2 will probably sell well here.  I think people would be more excited about a quality "all in one" solution, though. <- warning!, opinion.

I'm not the only customer here, though...so, team..chime in and correct me where I've misspoken.

--Michael

p.s.  I'm no E.E/C.E, so I may be missing an obvious option that doesn't involve an OS, per se....well, this is where I get that free Internet education  :)  
Mics:  Microtech Gefell m200/M20/nBob Actives>PFA, CA-11
Pre's: USB Pre2, 1x V3, 2x V3 w/optimod, MP2, Church Ugly
Decks: SD MixPre 6 II, R44 Oade Concert Mod, M10
Playback: Grace m9xx->Sen HD 650, Fostex TH-X00, HIFIMAN HEXX
Mixing: RME Fireface UFX->Reaper/Izotope->Yamaha HS8

Offline taylordb

  • Trade Count: (0)
  • Taperssection Member
  • ***
  • Posts: 256
Re: Portable Larvy AD10?
« Reply #73 on: May 06, 2008, 09:09:56 PM »
Topping this just to keep my hopes up  ;D

Offline eric.B

  • to the side qualified
  • Trade Count: (0)
  • Needs to get out more...
  • *****
  • Posts: 2796
Re: Portable Larvy AD10?
« Reply #74 on: May 06, 2008, 09:30:41 PM »
Quote
code to write all the important file types, as well as, handle most all the other functions we care about.  I haven't gone through it, but it's here: http://build.rockbox.org/

inside one small case => {nice pre's > Larvy AD > gumstix running modified rockbox > flash}

Does this sound far fetched? 

--Michael

Thanks for the post. Here are my initial thoughts:

The modules you point out seem to be aimed at very high speeds, at lower capacity. For audio storage, the speed for say 96KH stereo at 24 bits (which is really 6 bytes in "computer language") is near 600KB/sec. Given that one wants some margin, I would want at least 1MB/sec write speed. There is no need for 400MHz clock speed...
Also, there are all sorts of memory types. Most of the modules oriented for computer applications are volatile RAM. Volatile means the data is lost when you turn the power off, which is not good for a recorder. RAM means "random access memory", which is not needed for audio, because we "park the data" sequentially ("first come first served").
The type of memory needed is non volatile and sequential. The memory size of a standard CD (16 bits stereo at 44.1KHz) is around 650MB. So for the same CD length of time, at 24 bits you need 1/3 more, and for 96KHz 24bits you need around 2.5 times more memory... you are quickly approaching 2GB capacity...

Hard drives have huge capacity, but they tend to have limited life, and are a bit on the fragile side, though things have been improving a lot due to advancements in laptops. I like memory sticks and other flash memory technology, though they have their issues as well. With most memories, one can simply re write new data over the old data. But with flash, one needs to first erase the intended area (sort of like formatting the old floppy).

There are many other issues under consideration. It is not all that simple to design a digital device inside analog circuits.

It is always a challenge to put digital circuits near sensitive analog circuits, and with a mic pre, the challenge is the greatest. The electrical "environment" has to be extremely free of electrical interference.   
As a rule, digital circuits are "noise machines", which is not an issue for all digital gear, such as computers.
But it is an issue for converters and even more so for micpre's. When your micpre is set to say 60dB gain, an input signal of 1mV (from the mic) will be amplified by 1000  (you get 1V). Clearly a 1mV digital noise getting in to the input will yield a 1V of unwanted noise... It is always a challenge to put digital circuits near sensitive analog circuits, and with a mic pre, the challenge is the greatest.

There are many more considerations and issue, but the post is getting pretty long. As you probably expected, some of the issues have known solutions, others require innovation and/or a lot of care.

Regards
Dan Lavry     
www.lavryengineering.com
[/quote]

great post.. thanks! 
We have a system that increasingly taxes work and subsidizes nonwork.  ~Milton Friedman

 

RSS | Mobile
Page created in 0.109 seconds with 43 queries.
© 2002-2024 Taperssection.com
Powered by SMF