The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
What is your most reliable laptop config for recording Live classical music DAW Software
Old 19th January 2018
  #91
RPC
Gear Addict
 

Quote:
Originally Posted by studer58 View Post
So, does a Sound Devices or Nagra have more in common with a Zoom H5 (ie a bunch of mic preamps, ADC's and transport, metering etc) or a stripped back computer of any ilk...or are the two species pretty analogous (in a digital kinda way ) ?
I'd say a dedicated recorder (something released in the last few years) is probably more similar to a tablet than a computer, probably an ARM processor or equivalent running a real-time OS (RTOS) and some custom hardware for the audio interfaces. That would allow the vendor to concentrate on their "special sauce."
Old 19th January 2018
  #92
Gear Nut
 

Quote:
Originally Posted by Schoeller View Post

Modern specialized devices like portable recorders are probably typically based on a combination of specialized ICs (Analog Devices, AKM, TI,...) as well as possible some more general purpose processor.
I assume that's true for dedicated recorders. Some small portable digital mixers are running versions of Linux or Unix. The QSC Touchmix 8 and 16 mixers, which are designed for live sound but can also be used for recording (they record 32 bit 44k or 48k wav files directly to an attached SSD), are running a customized operating system that must be based on either Linux or Unix (they won't say), because when you download software updates they are tar.gz files.
Old 19th January 2018
  #93
I'd bet the newer devices are running stripped down and tightly controlled Linux nowadays. My airplanes avionics are running Linux or windows RT.
Old 19th January 2018
  #94
Lives for gear
 

The OS choice as such doesn't matter as long as it's reliable. We're talking about embedded systems which require much simpler OS than a PC or so but OTOH it must mandatorily reliably meet real-time requirements with much lower processor performance. A fast PC may catch up for not so demanding real-time tasks because there's a lot of processing power reserve to compensate lousy designed not real-time-robust software (especially the OS).
A battery powered embedded system has a far smaller processing power overhead which means that the firmware (including OS) must be much more carefully designed. Fortunately it's also much smaller and often also a modular design so each OEM only loads what is required.
Overall it will perform more reliably because the software is designed much more carefully and due to its reasonable size it can be thouroughly debugged, something which is strictly impossible with Tower-of-Babel-like bloated OS like Windows.


The overall design of a portable recorder depends on the product generation.

Maybe we could roughly consider 3 generations of solid-state digital recorders for analog signals:

1) Maybe 25 years ago so.
You designed mostly from the scratch, through-hole technology, mostly basic ICs like high-end instrumentation opamps, very good ADCs, general purpose processors (like Motorola 68k or so) with their associated ICs to handle discrete I/Os, memory, mass storage, LCD text/semi-graphic display, etc., using separate DSP chips, not necessarily audio-specific with mostly in-house developed signal processing application firmware.
Software was either written nearly from the scratch or relying on existing real-time kernels and some additional libraries.
A very small team or even a single person was able to develop a fully functional embedded computer board based on some generic CPU, design was simpler than today because clock frequencies were lower and other reasons but OTOH there were also less capable computer-based design tools and prototyping was more expensive.
Performance was mostly limited by the cost of some strategic components, for example precision low-noise opamps or ADCs. SSD were available but very expensive and not widespread, a couple of specialized but expensive HDDs were also available as of course generic HDDs.

2) Maybe 10 or 15 years ago.
Much more powerful processors, cheaper memory, semiconductors performance has increased while prices decreased.
Possibly still analog frontend based on opamps which are now better and cheaper, the same applies to ADC, digital signal processing etc.
General tasks are handled by some embedded single-board computer (either as COTS board or integrated in the design), software development is easier as better tools and more libraries are available, real-time kernel or real-time OS and mostly applications programmed in something like C/C++.
Specialized audio ICs may be used or not, depends on the required specs as many of those chips are intended for some general purpose volume market with lesser audio quality requirements.


3) Current designs.
More advanced specialized audio ICs are now available with fairtly high performances. A more integrated and more power-optimized design is now easily possible.
Higher-end audio ICs (Analog Devices, AKM, TI, etc.) can now be used because their specs have improved, they allow also much cheaper designs than those based on high-end instrumentation opamps and measurement instrument class DACs.
Processors are much faster but must handle a couple of tricky tasks which didn't exist before, especially USB interfacing and memory card access management Those two are a major challenge for the OEM because of the interface to 3rd party hardware he does not have control over the specs of the externally interfaced hardware. Als there may now be WiFi, Bluetooth, Ethernet and other interconnectivity challenges.

Now I suppose that the ultimately best analog frontend and DAC performance may still be achieved using expensive instrumentation ICs rather than relying on more integrated much cheaper designs based on specialized audio ICs. But that's widely a question of pricing level.

Beyond a DAC, referring to audio performance, it won't matter what's used for digital signal processing as it's only about processing data (of course there can be good and bad digital signal processing). Processing power is cheap so it's possibly even more a question of (battery) power requirements but also here there are now lots of low-power embedded computer solutions available.

Last edited by Schoeller; 19th January 2018 at 08:10 PM.. Reason: Overall edit.
Old 19th January 2018
  #95
Lives for gear
 

Quote:
Originally Posted by TMetzinger View Post
A hardware recorder like a JoeCo or Sound Devices is pretty complex, and a laptop/desktop set up for music is not THAT much more complex.
The complexity is increased with various unrelated companies involved in designing the software and hardware (with fewer limitations of purpose). Operating system + motherboard + audio program = 3 different companies, increased risk.

I've certainly never owned a (Windows based) computer which could run reliably for more than a few months without re-installing the drive image.
Old 19th January 2018
  #96
Gear Nut
 

Quote:
Originally Posted by aracu View Post
I've certainly never owned a (Windows based) computer which could run reliably for more than a few months without re-installing the drive image.
Wow, that seems unusual! I've been using Windows for 17 years now and have never once reinstalled the drive image, although I've certainly had my share of problems with Windows up to version 7. I've found Windows 7 and Windows 10 to run very reliably, at least compared with previous versions of Windows.
Old 19th January 2018
  #97
Quote:
Originally Posted by bradh View Post
Wow, that seems unusual! I've been using Windows for 17 years now and have never once reinstalled the drive image, although I've certainly had my share of problems with Windows up to version 7. I've found Windows 7 and Windows 10 to run very reliably, at least compared with previous versions of Windows.
One of the simulators I use to teach flying is still running XP... no issues. Win 7 and Win 10 are both great.

YES, there are certain jobs where the cost of a failure is VERY high (in dollars or live), and automation of those jobs often get very expensive computers running custom OSes. Audio recording, even broadcast, isn't one of those jobs. We keep going after a glitch (mic fails, wire fails, console channel fails, recorder fails). We put in redundancy where it makes economic sense to do so, but we don't rise to the level of "there can be no failure points in the chain".

I think the advantage of dedicated recorders over multipurpose PCs is that you don't have to know anything about how they work under the skin, and they don't require much in the way of maintenance. On the other hand, for someone who understands computers, it is not that difficult to operate and maintain PCs strictly for audio processing, and you can save a lot of money. Whether your time is worth the money or not is a personal decision.

If my Mom was going to do recording (she understands analog recording very well) - I'd buy her a pair of JoeCos, because she doesn't have the computer skills. For me, a pair of NUCs with only the stuff needed installed has been a great solution.
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump