soundcheck's Touch Toolbox 2.0

[magiccarpetride]

Sure. Your point?
Obviously I heard 'something', i.e., the music. But in my case I cannot say with any degree of certainty that I heard a difference in sound before and after soundcheck's mods. I _can_ say with certainty that I did not hear the "staggering improvements in the sound quality" that you and others report hearing (and you hadn't even disabled wifi yet).

OK, let me ask you this: are you, or have you ever been able, to hear ANY differences in the sound quality when comparing two systems?

[Tom186]

I _can_ say with certainty that I did not hear the "staggering improvements in the sound quality" that you and others report hearing (and you hadn't even disabled wifi yet).

Which brand of audiophile cotton swabs is everybody using to clean their ears?

[aubuti]

OK, let me ask you this: are you, or have you ever been able, to hear ANY differences in the sound quality when comparing two systems?

Most definitely. Again: what's your point?

[firedog]

Really? I haven't seen any such definitive measurements, though as a new guy around here I could quite easily have missed something.

I did recently see some info that showed the buffer modification changed the noise induced on the processor power supply (maybe even earlier in this thread), but nothing that showed the change in noise profile made any difference to the analogue output. That said, I'm 100% convinced there are modifications that could improve SQ as yes - the unit will always be a compromise when designed to a budget - but I've yet to see any supporting evidence that these particular sofware modifications make any difference or even credible theories that could explain the claimed improvements, and so ridicule is an entirely natural and, IMO, appropriate response

So in all seriousness, please point me to any such measurements - I'd be very interested to see them and the conclusions drawn.

It seems to me that the measurements Phil and John did (see post around #511 for what John did) show that the mods do have some measurable effect. Certainly it is possible the effects are audible, and we can imagine why these results might result in an audible difference. For me that alone would take the mods out of the realm of deserving ridicule.

I have applied the mods (except for turning off the screen): I'm running wired anyway, and modded my Touch to remove the analogue outs, so the mods don't have any "cost" for me.

I think they do make a small improvement - certainly not the dramatic kind that others have been reporting. But I could be imagining it - I certainly don't have the ability to run an objective test with my setup.

[soundcheck]

@firedog

You run an EA Pacecar (1250$!) according to your signature, right? And your Touch is even slaved to the Pacecar!?!?
The Pacecar is supposed to be doing quite sophisticated reclocking, right?
The only reason for that device to exist is to clean up the mess supplied by the source - any source.

You IMO shouldn't hear any differences with any of my Touch mods assuming that Steve knows what he's doing.

As I always been saying. My mods will have more impact on one system and less on others. Not any system in the world will sound the same.
And arguing about it is waste of time.

[firedog]

@firedog

You run an EA Pacecar (1250$!) according to your signature, right? And your Touch is even slaved to the Pacecar!?!?
The Pacecar is supposed to be doing quite sophisticated reclocking, right?
The only reason for that device to exist is to clean up the mess supplied by the source - any source.

You IMO shouldn't hear any differences with any of my Touch mods assuming that Steve knows what he's doing.

As I always been saying. My mods will have more impact on one system and less on others. Not any system in the world will sound the same.
And arguing about it is waste of time.

First, running an old, second hand PaceCar (not $1250).

Second, I heard the Touch before it was modded and slaved....And I've also had an off the shelf Touch in here since for comparison purposes. Sorry, I guess I should have stated that for ultimate clarity.

Third, you claim that running a Touch over ethernet with wlan circuits turned off makes an audible difference, even on a standard Touch running as a digital transport only. If so, what's your "explanation" for the difference this makes in the resulting SQ? In my understanding, if these mods actually work, they should have some impact in my present setup also. If this is totally incorrect, let me know - I'll remove the mods and only use them when I get a second, off the shelf Touch.

Finally, Klaus, why so defensive? All I said was that I thought your mods make a small positive difference. I can't be sure b/c I don't have the equipment to run a proper blind test. Does it offend you that not everyone describes your mods as making "orders of magnitude" improvements? I've actually defended and recommended your mods on this and other forums, and BTW, I made a small donation at your site in recognition of your efforts.

In my book, once you have good equipment, even small improvements tend to be quite costly. So I think the "small improvements" your mods bring are welcome and significant, especially since they are free. But audiophillia is full of descriptions about upgrades that are wildly significant, when in fact the resulting improvements are actually small or nonexistent. I don't see any reason that I have to be in that camp.

[adamdea]

not trying to be funny, but your stating opinion when you say you did not hear any difference.
Others have tried it and find it worthwhile.
Its ok to disagree

I upsample via SOX in my setup, so all my conversion is done at the server end anyway...

Sorry if this seems pedantic but am not stating an opinion when I say I heard no difference: it is a fact that I heard no difference. I have not actually expressed an opinion on whether there is a difference I know that other people apparently felt a vast difference; if they hadnt I wouldn't have tried it in the first place.

[Tom186]

You aint one of the cynics though Alex!

Update your Squeezebox Server to the latest 7.6 beta...
and update your Touch to the latest 7.6 firmware when prompted.

Do the mods (especially the display off)

Works 100% here(no issues with beta)and...well, I'd be interested on your thoughts on the sound.

Treble is clearer. This is true with mods and without mods, and on or off drugs.

[snottmonster]

Hey - has anyone tried the latest Toolbox recommendation?

"Run your PC or NAS ethernet interfaces at 100MBit/s full duplex."

[soundcheck]

Hi folks.

It seems that somebody is following my DIY-Audio thread.

Just to let you now:

I introduced chapter 1.9 on my blog.

I'm talking about pretty basic (for now) Network Optimizations and different networking implementations on different operating systems. (We touched upon that earlier and the networking HW side I already covered earlier on the blog.) The network handling and setup slightly differs from OS to OS.


Networking/Streaming is a very high prioritized task - for obvious reasons - on the Touch in a media streaming environment.

The ones who managed to understand the logic behind my modifications and optimization proposals, will easily understand that those network originated load conditions might have a direct impact on the device and an indirect impact on the audio stream quality..

What's going on:

1. First the network needs to be managed from the Touch and server side.
There are numerous parameters managing TCP/IP traffic. By default most of those parameters are not properly set. A one size fits all solution.
(You can learn a lot if you follow the online gamer crowd tactics and recommendations btw. Those guys are far ahead on network optimizations)
2. The server (usually equipped with Gbit-Lan) needs to step on it's brakes all the time when feeding the Touch. Otherwise the Touch would be flooded. Or you can turn it around: The Touch is saying "Please stop from flooding me". In TCP/IP terms they are negotiating.
3. And 3rd the streaming buffer needs to be continuously refilled and managed causing a
continuous load on the Touch processor/scheduler.

Note: It seems that certain inefficiencies had been identified by Logitech too. They seem to have optimized traffic handling in 7.6..
There's also a reason, why Logitech prefers to stream compressed flacs by default. An uncompressed PCM stream
at 24/96 is causing a pretty serious load - and resources are more than limited on the Touch...


Yep. There's a lot of ping-pong ongoing in that area. And I can tell you it's not that easy to get it under control properly.

Some of you might be surprised that a default Windows 7 performed best from my and some others perspective. I can tell you I never
expected that myself.

If you find the time -- checkout my Blog. It'll take you no more than 5 minutes. Just a little configuration work to do on the server side.
It should be manageable. I havn't looked into OSX server side optimizations, since I do not own a MAC...




Enjoy.