Tír Na HiFi

So after suggesting that, I looked at Volumio's site, and they look like a "real" operation, so may not want to deal with us, unless we provide a larger amount of money. And we would need to somehow get them HW to test with, to make sure it would work.

Here's an option that appears to be in Glasgow, not sure how far away that is from you guys?
http://openhome.org/pages/use/

I haven't tried their SW at all, so I don't know how the UI is, but beggars cant be choosers lol.

Will try to at least download openhome sometime soon to see what it looks like.

Randy

Good to see you are hearing the weakness of computer audio.

Next step will be returning to the phonograph!

One thing I worry about with this card is what i found with the cards I with which I was recording my disks is: they break. The SD card reader part is surprisingly fragile. It will work fine and then it won't work at all. What I see on the yellow card looks exactly the same as what I got with the many cards I bought. One had the spring loaded aspect like the SDTrans (not that it resembled it as far as evident ruggedness) and it broke, too, eventually.

The SDtrans does use a more rugged card reader. The trick would be to fit this reader on to that card.

I guess as long as the mods are easily moved to a new card there is nothing to fear. But it will happen with serious use.

Still think how you put the files on the disk makes a big difference. Most of the ills of USB audio are present if you use a writer that uses USB. Try to find SATA writers. They sound much better. i used the reclocker whose name I have forgot (turned around, of course), and it helped but it wasn't as good as the SATA.

Luckily with LP one has plenty of challenges. But then I am 63 and LP is second nature to me. But I did get tired of the transfer part and with the tiny screen of SDTrans having to rely on memory to know what i was trying to find.

The PCB upon my trick clock was mounted was not easy to keep attached to the SDTrans. Half-moon through holes - it was what the majority of customers wanted even though the sorter thought they were a bad idea. The customer is always right ...

When I put the SDTrans back into service I will use ANDREAS's clock who had a GROUP BUY thread at DIYAudio. Very different approach. I have learned "Jocko" and ANDREAS are friendly. Last I heard not enough demand to buy enough clocks to do any sorting so ANDREAS's clocks could be the best available. Much more complicated. Who knows if it will make a difference but I remember the clock substitution making more than a subtle difference. Maybe not the lowest hanging fruit but should be given a second tier priority.

Computer audio has a long way to go. I still think USB is the biggest part of the problem. Not on any basis of how it works. I have no idea. But I know the SATA connected writers made a significant difference.

I am sorry. I have become a partisan. I wanted digital to be the answer. Not that I ever got rid of my LPs. Whether due to nostalgia and familiarity or because I actually hear the differences I think i do; there is no comparison, yet, between good analogue and good digital. Obviously my experience is tiny. But after playing at it for ten plus years I think I got an idea of what was possible even if I never achieved it but in my most unhumble opinion it was not getting close to good analogue in giving a convincing illusion of music being played. Sure it does a few interesting tricks - it is "quieter" in it own way but analogue noise is mostly right there, it is not hiding and this is, at least for me, easier to mentally workaround. The bass can seem firmer and lower but when I listen to the same music on LP it sounds more like it should. (as I think it should)

Distortion is something we live with everyday. Compromise is the great lesson audio teaches us. We get to pick which distortions we find easiest to ignore.

Right, BBQing yesterday so offline but played the SD player at the BBQ all day on one small LiFePo battery

I think I'll contact Volumio before we kick them to touch - all we want to do is develop a plugin which is optionally included in any person's configuration. Let's see what they say.

I've heard of Openhome before but never looked into them - worth a look.

I'm talking to a couple of different STM programmer guys to feel out what's involved. As I thought, it's not a complex application & there are already library modules or free code covering almost all of the functionality needed - SD card reading; I2S processing; screen interface, etc - what's needed is someone to put the code together into a working whole that works on this hardware (& transferrable to some later, better designed hardware platform). The hard part is convincing these guys that a system & audio synchronous clock is a must have requirement - they can't understand why :)

Rick, are you no longer in love with the SDTrans sound - prefer LP? You're looking into using Andrea_mori clocks - I think some people here bought some early clock modules from him?

I was talking to a Chinese STM programmer this morning & during the discussion he came up with an idea that seems interesting - forgetting about the SD card & using a WiFi enabled ESP as the data interface - audio file transmitted via WiFi to ESP - full audio file transferred into one port of dual port RAM - STM read audio file from other port of RAM - STM can then easily use audio clock as system clock & switch between them

In that configuration, the STM is doing very little other processing while it is handling audio conversion & outputting I2S

The problem might be WiFI data throughput i.e. keeping the dual port RAM half full
Theoretical Actual
802.11g 54 Mbps 20 Mbps
802.11n 600 Mbps 100 Mbps

32bit 384KHz audio throughput is 24.572 Mbps which means 802.11n WiFi would be required

The other possibility would be to transmit the whole audio file into RAM (not dual ported) & then STM processor reads this file out of RAM

jkeny wrote: ↑Tue May 08, 2018 11:46 am I was talking to a Chinese STM programmer this morning & during the discussion he came up with an idea that seems interesting - forgetting about the SD card & using a WiFi enabled ESP as the data interface - audio file transmitted via WiFi to ESP - full audio file transferred into one port of dual port RAM - STM read audio file from other port of RAM - STM can then easily use audio clock as system clock & switch between them

In that configuration, the STM is doing very little other processing while it is handling audio conversion & outputting I2S

think thats a stab into the unknown, sd cards are renowned for low noise, however ram does sound more practical
i suppose it would be good to know how much it would cost to programme and build a prototype
just in case it doesn't work out

nige2000 wrote: ↑Tue May 08, 2018 1:05 pm think thats a stab into the unknown, sd cards are renowned for low noise, however ram does sound more practical

Are they? Sony has introduced a pricey SD card with claimed "Low electrical noise" so I'm presuming the standard cards aren't low noise? But unfortunately, no noise measurements from Sony so it could be all marketing but if SD cards already had a reputation for low noise, I doubt it would make a good marketing plan to target low noise SD card ?
http://www.escapistmagazine.com/forums/ ... ical-Noise

i suppose it would be good to know how much it would cost to programme and build a prototype
just in case it doesn't work out

Sure, that's what I'm working towards finding out - a build price!

Pretty sure that Sony card is not SLC at the price so I would not be inclined to try it out.
The cheapest SLC Micro SDXC card at 8Gb i have been able to get costs €108.
16Gb SLC SD card costs €175, all from Digikey.

These SLC cards do make a significant difference so it is a shame that they still cost so much.
I will bring them along the the next get together so you can hear the difference.

My vote is to stay the current course (at least my course lol), the SD player with a flashair wifi sd card, and some software running on a pc (or mac or pi) that lets me pick a song from music in a NAS, and play the song.

Control of the SD player will be with optocouplers or relays pushing the buttons, through an ESP based device.

The dual port sounds interesting, but that is a major change in direction, and seems like it would require significant effort to make it work.

BTW, going "offshore" for development may be a way to get this done. Either Chinese or Indian. There are pitfalls with going this route, working with someone that does not speak english natively, 1/2 around the world in a different time zone. But lots of companies outsource software to save money.

Randy

randytsuch wrote: ↑Tue May 08, 2018 3:42 pm My vote is to stay the current course (at least my course lol), the SD player with a flashair wifi sd card, and some software running on a pc (or mac or pi) that lets me pick a song from music in a NAS, and play the song.

Yea, I agree that getting the current configuration working satisfactorily is the primary goal but I figure that we can improve on it at some stage & I'm investigating the possibilities

Control of the SD player will be with optocouplers or relays pushing the buttons, through an ESP based device.

The dual port sounds interesting, but that is a major change in direction, and seems like it would require significant effort to make it work.

WIll hopefully get a feel for what effort it requires

BTW, going "offshore" for development may be a way to get this done. Either Chinese or Indian. There are pitfalls with going this route, working with someone that does not speak english natively, 1/2 around the world in a different time zone. But lots of companies outsource software to save money.

Randy

Yea, this guy is about 25, a computer science graduate & speaks good English - he's a contributor on one the the STM32 github projects & has recently joined DIY Audio - he's just getting into the audiophile world so he is interested - talking to him costs nothing & there's no communication barrier - will see where it leads

jkeny wrote: ↑Tue May 08, 2018 3:55 pm Yea, this guy is about 25, a computer science graduate & speaks good English - he's a contributor on one the the STM32 github projects & has recently joined DIY Audio - he's just getting into the audiophile world so he is interested - talking to him costs nothing & there's no communication barrier - will see where it leads

That sounds pretty much like the ideal candidate.

Maybe we can discuss a "trade" with him. We could chip in, and buy a SD Player and flashair sd card for him, mod it and send it to him.
He could use it for the SW development, and then keep it to use. If we included a DAC, its a complete digital playback system, and should sound better than most. Would be a pretty good start into diy audiophile land. I'd wager he'd have to spend a lot more money to get something comparable.

Randy