Squeezebox Touch -- Modifications

ttscreen mods

Hi Soundcheck,

Thanks for all the contribution you have done so far to the audiophile Touch community! I am looking forward to test TT3.0!

I'd like to ask your feedback about the impact of software screensaver disabling in conjunction with ttscreen: are the SQ improvement of the mods cumulative or does ttscreen de facto also disable the screensaver modes as well as the screen HW? In the second case, this would make screen SW mods unnecessary if screen HW is disabled, am I right?

The reason of my question is that I would like to leave the screensaver on for userfriendlyness while the screen HW is on and only use ttscreen for use with remote on Android phone when SQ matters.

Best regards
 
The screensaver is a function which runs as a timed task.
"Screensaver off" is that what would be the best option. That's what you'd do on an audiophile PC based setup too.
I'm not sure if that option exists though. I don't think so. I'm using "screen off" if I recall it correctly -- The Logitech way. ;) That function won't stop neither the screensaver task nor does it turn the screen of.

Perhaps you try if there is a difference between different screen saver options..


Turning the screen off with ttscreen is a complete different story.
You'll see that at the power consumption of the box. It'll drop instantly.

Cheers
 
Thanks for your quick answer.

If I understand you well, the ttscreen does not impact the underlying processes running to manage the state of the screen and would be still running even if screen HW is turned off. So, both SW and HW screen mods have to be applied together for better SQ. Correct?
 
Other thoughs on another topic.

Did you ever look at what cost effective solution could be used as low cost, low power embedded Windows 7 hardware dedicated to running Squeezebox Server, since it seems te be a better solution for SQ compared to Linux?

I am currently using Synology NAS 211j but on top of being unable to tune it for SQ, I am also annoyed by performance bottlenecks causing sound interruptions when other tasks are perfomed in parallel such as file downloads or jpg archiving.

Therefore, I would consider building or buying my own fanless cost effective low consumption barebone around dual core Intel Atom but this open many questions as which WIN7 version to use, type of storage (HDD, CF or SDD), power supply, network adapter,... any advice from the community is welcome.
 
Actually, I see two distinct issues:
1) there is a bottleneck that prevents running other CPU consuming tasks on NAS together with listening
2) the NAS is running Linux that is not the optimum for SQ

I am wondering if there is a third issue I ignore: does the CPU power itself have an impact on sound quality? I would guess not, since I didn't experience CPU bottleneck to handle audio streaming properly when there are no other processes running on the NAS, but I did not compare with more powerful servers.
 
Hi guys.

Somehow my (and some others) W7 platform seems to handle the ethernet traffic best. My Ubuntu Server install required serious tweaking to get close.

Dont ask me what makes the difference. And why it makes a difference. I never expected the server to make any difference at all. The problem: It does
make a difference.

Just give it a try.


In general I do not recommend to use NASes, because these are ususally outdated as soon as you got them hooked up.
Updates drop in for a while with a pretty high delay.
If you're lucky your device will be supported for the next 2 years. ;)
NASes are inflexible black boxes. They are usually pretty slim dimensioned.

My recommendation: Stay away from them!

What I do:

Right now I run a Windows 7 platfrom acting as HTPC/DLNA-Server/Squeezebox Server/iTunes base on an Asus ION platform.
I can watch Blu-Rays, stream video and audio material, can run my Squeezebox stuff and get my Apple devices managed - all on one box.
Not to forget -- it also acts as a normal PC.

What I can't do is high quality Video manipulation/filtering ( e.g. JRMC - Red October HQ mode won't work). That would require a bit more juice.
Beside that I got "my choice" of two speedy "eSata" WD 2TB Caviar Black hooked up to make sure my data are save and managed the way I'd like to do it.

A NAS for sure will limit you on all this.

BTW: Linux also failed on providing above solution. That's why I'm back to Windows.


I'm pretty much enjoying latest JRMC. Since recent times it plays even HD video pretty well. Even on my ION platform. The only thing missing is a nice iPad remote app for JRiver.

To be honest. The only reason why I'm still running a Squeezebox as audio renderer is "iPeng" and its slightly less costly achievable SQ.

I do not intend to seperate video and audio any longer. I'll be talking Multimedia.

Off-Topic:

Just to let you know: The other day we compared the well regarded 580$ Audiophillio II USB-SPDIF interface (+CPLAY/CMP2 setup on an XP PC) against my modded SB Touch (TT3.0 and some HW&PS tweaks). Both were feeding a highly modded TP Buffalo 2.

Result: There is absolutely no reason to spent 570$ for that device if you can live with a a little squeezebox.
I'm not saying that the Audiophillio II is bad. I'm just saying that my solution ( a tweaked reatime Linux + slightly tweaked SB Hardware) IMO performed clearly better. Unfortunately we didn't have time to hook up the Audiophillio to the SB Touch. I'm sure that could have been another intersting test scenerio.
Not to forget to mention that the test also showed again how poor the TP Buffalo isolates from incoming stream related distortions. A properly done DAC
shouldn't show any differences, no matter what's feeding it, as long as the data are the same.


Cheers
 
Last edited:
Thanks for your quick answer.

If I understand you well, the ttscreen does not impact the underlying processes running to manage the state of the screen and would be still running even if screen HW is turned off. So, both SW and HW screen mods have to be applied together for better SQ. Correct?

Yep. ttscreen acts towards the kernel.

Luckily the Squeezebox application doesn't seems to recognize a problem.
Otherwise I guess you'd end up with a restart or simliar.

I'm not sure if the screen related SW part do plays a big role after ttscreen is enabled. Just try it. And let us know. ;)

I just turn everything off what can be turned off. If it doesn't contributes anything -- who cares. I just can't afford to test every little parameter change anymore. Meanwhile there are just too many parameters which I'm addressing.

Cheers
 
Thanks for your quick answer.

If I understand you well, the ttscreen does not impact the underlying processes running to manage the state of the screen and would be still running even if screen HW is turned off. So, both SW and HW screen mods have to be applied together for better SQ. Correct?

Yep. ttscreen acts towards the kernel.

Luckily the Squeezebox application doesn't seems to recognize a serious problem.
Otherwise I guess we'd end up with a (cyclic) restart or simliar.

I'm not sure if the screen related SW part do plays a big role after ttscreen is enabled. Just try it. And let us know. ;)

I just turn everything off what can be turned off. If it doesn't contributes anything -- who cares. I just can't afford to test every little parameter change anymore. Meanwhile there are just too many parameters which I'm addressing.

Cheers
 
Ho Soundcheck,

very interesting to read - i´m really looking forward to your TT3.0.

What makes me think is that you are running your Win based PC as
SBT-Server as well as iTunes-Server. What kind of compression do
you use? iTunes can´t play flac. Jow did you import your iTunes DataBase
into your SB-Server?
I´m asking because my SB still isn´t playing my music-files i bought via
iTunes.

Greetings ulf
 
Ho Soundcheck,

very interesting to read - i´m really looking forward to your TT3.0.

What makes me think is that you are running your Win based PC as
SBT-Server as well as iTunes-Server. What kind of compression do
you use? iTunes can´t play flac. Jow did you import your iTunes DataBase
into your SB-Server?
I´m asking because my SB still isn´t playing my music-files i bought via
iTunes.

Greetings ulf

I use flac and apple lossless. I keep the iTunes/Apple related data 100% seperated from the rest.
I actually do rip CD's again just to put them on my Apple devices.


It shouldn't be a problem to get your iTunes db connected to SB db. I never tried it. Here is a link
 
I found an interesting HW to build up a low cost media server: the fanless shuttle XS35V2 with dual core 1,8 GHz Intel Atom 525. I am thinking to build it up with 500 GB HDD, 2 GB RAM and Win 7 64 bits for about 400€.

Given Soundcheck feedback, I would however also consider the GS version for HQ HDMI rendering to my TV, since I bought recently an Xtreamer for that but it is also quite limited in usage and support and I could merge both devices in that way and even hide it behind the TV screen with VESA support.

Anyone on the forum with experience with shuttle as media server?

There remains an open question for me: is it worth keeping the NAS for RAID1 network media storage and not much more?
 
I was experimenting with different network setups.

My SBT has all Soundcheck (thanks !!! ) tweaks, PS with Super Teddy Reg, (phantastic !!!) Buffalo 2 DAC (the best DAC ???) and IMHO good system with very good tube pre (with NOS Telefunken tubes) , diy poweramp based on Hackernap and Spendor speakers.

I tested three setups:

1. SBS Notebook with W7 (OS on SSD and second internal 1TB HD with flacs) and Touch connected directly (SBT cable is ~5m long) with good cables to the 100MBits router ports of D-Link 524UP.

2. SBS Notebbok connected to the router and close to SBT there is one small switch refreshing the signal and a short unscreened cable between small switch and SBT

3. SBS and SBT are connected to a small standalone switch and this switch is connected to the router. Between switch and SBT there is 5m Cable and the screen is "off" just before SBT.

So IMHO absolutely best quality has number 3, than 1 and than 2, and there is a huge difference between 3 and 1 and a ver small between 1 and 2.

I have two different switches - nice few years old Netgear FS-105 100MBits with small linear powers supply (Netgear was selling this switch with linear PS) and one new Cisco SD2005 with switched ps.

With Cisco there is more space, deeper more precise bass and the music flow is better, but the sound is less pleasant and little agressive (some distortions appears ?) than with Netgear. Maybe the reason is the switched ps of Cisco, i want to try Cisco with a linear ps, when i will find or built one ...

So my conclusion is - take a small nice switch, connect it to your router and than connect your SBS and SBT directly to this switch. Don't buy the cheapest one.

Make your own tests, maybe it also depends on your environment, router type, etc, but using a dedicated switch for data flow between SBS and SBT seems to be the best solution.

Next i want to try Cisco SD200-8 which i can borrow from my work.

There is one thing which is a completely mystery for me - SBT is buffering a LOT of data.

When you will disconnect the switch betwen SBS and SBT playing 24/96 files the music will play for ~10 seconds and when playing 16/44 the music will play for ~30 seconds.

If so much data is buffered, why different network configuration has such an impact on sound quality ?

:) :) :)
 
My buffer is already set to 3600. It looks, that they are two buffers in SBT, or ?

when i type ttsat in my SBT, than i am getting this:

# ttstat
________________________________________________________________

soundcheck's SBT Toolbox 2.00: Modification Status
________________________________________________________________

Modification WLAN: enabled
Modification Watchdog: enabled
Modification Jive2: enabled
Modification Buffer: enabled
Unknown HZ value! (92) Assume 100.
Buffersize: 3600us
Modification Screen: enabled
Audio outputs: Analog: off Digital: on USB: off
Modification Kernel: enabled
________________________________________________________________

???