Neuros Technology Forums

I just looked at the blurb for the OSD and it seems like a cool gizmo, but it's disappointing that it doesn't have firewire so the only way to convert mini-DV camcorder video is through analog conversion, losing quality.  It would be great if future versions had a firewire port that could plug directly into a miniDV camera and convert the DV25 video to the other formats.

I'm resisting the X5L because of its battery.  The only hard disk audio player I'd buy right now is the long-discontinued Archos Jukebox Recorder 20, available on ebay used for $100 or so with 20gb disk.  It can run open source software (Rockbox).  It uses standard batteries (4aa NiMH with internal charger).  It uses a standard 2.5" laptop hard disk, i.e. it's expandable to 160gb (maybe 200gb) today, probably 300gb by end-2007.  Given that 16gb flash cards are available today, there's no point messing with the hassle and power consumption of a hard disk if it's just a wimpy 30gb or 60gb.  I feel likely to pull the trigger for a 16gb Nexblack in the next few weeks though, depending on whether a certain hoped-for improvement to my finances comes through. 

It sounds like nothing's been happening with the N3, which is just as well.  I can't understand Dragonwisard's point that

I completely agree that the N3 should focus on Audio first and foremost. However I think the platform is capable of much more and we should take advantage of that so long as it doesn't degrade the quality of the primary function (which is Audio)

What else is there for the N3 to do besides audio?  The only thing I can think of is video, which requires a power-hungry screen, which in turn mean special batteries, which definitely degrades the device as an audio player.  I -refuse- to buy any audio player that needs proprietary batteries.  I  -only- want to buy a player that uses AA's or AAA's.  If a mid-2007 N3 has a HD and uses some reasonably standard, user replaceable lithium battery then I suppose I'll consider it out of solidarity but it will be a big minus.  "Reasonably standard" means it's used by some popular model of cellular phone or digital camera, so replacement batteries are available from lots of different suppliers and will stay available for many years.  I'd still much rather have flash and AA's though.  For now, I don't feel I need more than 16gb capacity in an audio player, and would far prefer flash to a HD at that level despite the cost premium.  Keep in mind that while flash costs more than a HD, it makes the rest of the unit cheaper by getting rid of the lithium battery and charger.

I just don't see what's so hard about this.  Really, it's almost an industry conspiracy that nobody is making a sensible DAP right now.  That means G3 sized, powered by 1AA or 1AAA (makes player thinner), with a monochrome display, low power consumption, and an SDHC slot (8GB SDHC-2 cards are $129.95 at Newegg as of today).  Better yet, two SD slots, giving 16gb capacity today (with bigger cards coming) and allowing copying from one slot to the other (product differentiation, heh heh).  Google "minty mp3" to see how simple a basic DAP can be.  I'd make one of those myself except it has no display.  The Minty player already has a CF slot--just adding a display and replacing the CF with two SD's (or heck, leaving it CF) doesn't sound like a multi-year engineering project.  The firmware (Rockbox) is already done, you just need to use a Rockbox-capable cpu and mp3 chip and connect up some buttons and stuff.

Another player that I'm amazed nobody makes is a DVD-MP3 audio player, basically the Giga MP3 feature found in some car stereos.  This is the basic, ancient, butt-standard MP3 CD player as made by Panasonic and every other company, powered by AA cells, but using a DVD drive instead of a CD drive, so you can have 8.5 GB of music on one disc (dual layer DVD+R).  Yeah there are portable DVD players but they all have video playback which means color screens and, you guessed it, proprietary lithium batteries.  I'm talking about an audio-only player that uses AA's.  I guess it's not Neuros's approach, but I wish somebody would make this.

To Dragonwisard - the N3 issue isn't lateness, it's that technology changes over time.  The original 5GB Archos Jukebox was a breakthrough product and a huge success when it came out.  Those days are gone now.  Imagine the yawns if Neuros were to launch a 1 pound, 5GB hard drive player today.  Well, larger hard drive players (e.g. 30GB) are still interesting today, but my claim is that by a year from now, flash will be cheap enough that every HD audio player will be silly.  So, continuing development on such a player that won't be ready for a year is out of step with where technology and the market is going.

Also, re it being an "open source product", the issue isn't the software, it's the hardware.  Yes I'd love to have an open source audio player that I could put my own codecs (ogg, flac, etc.) on.  Remember though that I'm viewing the N3 as an audio player, something to listen to while coding or driving, NOT a video player, I don't care at all about video; video capability imposes hardware requirements in terms of cpu speed, power-hungry backlit color displays, and the battery systems needed to run all that.  Maybe it's just me, but watching a movie on a 2 inch screen is IMO totally unattractive and if I ever buy a video system, it's going to use a DLP projector so the picture fills a whole wall.  (Or maybe it could use those LCD eyeglass thingies).  Anyway that gets way off-topic regarding the N3.

For the N3/NX to make sense as a pure-audio product, it should not sacrifice any battery power or hardware size to implement video features that an audio-only player no use for.  The Cowon G3, already considered an obsolete player, is years ahead of the hypothetical N3 in terms of power consumption--it runs 40 hours on one AA cell (and yes, the G3 plays Vorbis, though not FLAC).    So I'm suggesting to Joe that any forthcoming Neuros audio player should be AA-powered and use flash instead of a HD.  That is a matter of hardware, not programming.  I realize Joe's plate is full and audio devices are backburnered at Neuros and that's ok.  But no contribution that I could make to OSD software development will get the audio hardware out the door any sooner.  All I can do is make suggestions about what the audio hw should be like when Neuros eventually has resources available to build it.

I'm the one who lobbied for including SPDIF-in on the N3 and more recently posted a rant on the N3 wiki discussion page.  I hope you guys don't get upset but I now think the N3 project should be abandoned.  With the new projected date, the N3 will be obsolete before it is introduced.  HD audio players in general are already in their last throes.  Flash memory is in the $10/GB range, 8GB SD cards and 16GB CF cards are starting to show up on ebay (still $$$$), and there's soon going to be no market left for a non-video HD player.  And video means big, power-hungry screens and batteries.  The N3 successor (call it NX) should be a flash player.

That the iPod is in its 5th generation doesn't mean it's all refined or evolved, it just means Apple is pathetic taking this long to get it right.  Sony got the cassette Walkman essentially right on the first try in the 1970's, digital watches had some technological evolution since LED's but they matured quickly when LCD's came out; cassette and CD portables haven't changed much in the past 20 years and neither have watches.  The portable digital audio player (DAP) is not that complex a product and it should be mature too.  That means Neuros can and should be making DAP's designed to still be in use 20 years from today.  Well, ok, make it 10 years since tech development has sped up so much, but this is still doable.  1980's vintage cassette walkmen are still viable.  The characteristics of a mature DAP are (to me) exceedingly obvious, but EVERY manufacturer is choosing to screw up one thing or another in the name of product differentiation.  Stop with that, just do a straightforward implementation of the obvious stuff and gain differentiation by making a high quality product.  Inspirations in the digital world should include the HP-12C calculator (1990's), the Canon S100 digital camera (circa 2000), and the basic Casio digital watch, all of which are still quite usable (as in, there's not that much reason to upgrade one you already own, even if current technology is better).  In the audio world, look to the Sony TC-D5M portable cassette recorder plus the obvious Walkmen and Ipods.  Look also at the Archos Jukebox series with Rockbox firmware.  Let us figure out the characteristics of the NX:

1) power: this is a no brainer, it MUST run on AA cells, or AAA's if size is critical.  No lithium ion.  There are many 1990's DAT recorders and laptop computers collecting dust because the proprietary nicad packs are now unavailable or not worth the cost.  The NX should preferably run on one AA, like the Cowan/iAudio G3 or the iRiver T10.  It should be designed to run satisfactorily on every AA from NiMH to crap-quality zinc-carbon AA's (the most common battery in the world). The player's form factor should be like the Cowon G3 mentioned above, which is business card sized x 2/3" thick or so.  The iRiver T10 is also interesting and different.  If you absolutely need higher internal voltage and can't use a dc-dc converter (analog noise or whatever) then use two AAA's.  If you want, include an internal NiMH charger powered by the USB port.  Finally, look at (but don't imitate) the Sandisk Sansa M260, a 4GB player built like a cheap digital watch.  Again, with the exception of some firmware issues, the M260 is an example of maturity: we could easily imagine the same player being sold for $14.95 in a few years once its internal flash memory (it has 4GB) is essentially free.  The NX should have similar principles but much higher build quality.

2) Basic features: see the Frontier Lab Nexblack which is finally out.  Frontier has screwed up execution (thing is a year late) but basically have the right idea: high quality analog audio section, relatively powerful headphone amp, FM tuner, and a CF slot.  These days I'd say SD is acceptable (CF seems on its way out; SD is more rugged and keeps size down).  But expandability is a must.  I'd suggest two SD slots, giving 16GB total capacity with today's 8GB cards.  Maybe one slot could be internal to save space.  Skip UI gimmicks, just use Rockbox, with the most frequently used controls on good physical hardware switches, not GUI widgets or dinky membrand buttons.  The Nexblack uses two AA's and is fairly power hungry, but I think that's because it uses old electronics.  The G3 and the T10 get ~40 hours on one AA and should be examined as an example of how to do this stuff.

3) Special feature: one thing every DAP should have is active noise cancellation, which is getting more popular in specialized headphones.  But it should be built into the player unit instead, so you can use your favorite headphones or earbuds.  I posted to the wiki about how to do this.  This is an essential feature for listening on public transport or airplanes, without having to crank the volume to ear damaging levels.  It should practically be legislated into players.

I'll add some more to this later, I have to go do some stuff.