Jesus soundcheck... you talk so much bollocks.
How did you cope with the initial learning curve of linux..? Your far from a driver or kernel hacker, but how did you even deal with installing a distribution, when in your head, cheetah blood should be rubbed into the disks to ensure no 'degradation' in 'file quality' occurred...
There are many people around who'd love listen to your findings.
Please don't forget to outline a highly sophisticated scientific
explanation for these. And please attach your measurements.
Then you look for at least 10 people who'd swear by God that you're telling the truth, nothing but the truth. Otherwise nobody will believe you. All of them would have to do a double blind test on at least 10 different top-end systems. It's that easy!
Looking forward to your contributions. (I havn't seen many so far)
Meanwhile, I'll enjoy my ethernet wired, tweaked Touch.
BTW, if anybody is interested to try my tweaks on the Touch -- PM me. I'll sent you a file, which you need to load onto your Touch. It is that simple.
No risk at all.
Cheers
Last edited:
Hi there!
Here we go again! Soundcheck you got my vote! I agree with you. I got the Touch unit some weeks ago...
If you have a high end setup (mine is 50.000USD), there are tweaks that digitally gives a no change (i.e. bit perfect), but that results in clear audible (analog) differences. Call it jitter, call it signal ground pollution, call it what ever you like. The difference is there. If you have a 1.000USD system, you will not hear these differences.
Let's move on with an open mind. That's much more fun!
Cheers! ;-)
Here we go again! Soundcheck you got my vote! I agree with you. I got the Touch unit some weeks ago...
If you have a high end setup (mine is 50.000USD), there are tweaks that digitally gives a no change (i.e. bit perfect), but that results in clear audible (analog) differences. Call it jitter, call it signal ground pollution, call it what ever you like. The difference is there. If you have a 1.000USD system, you will not hear these differences.
Let's move on with an open mind. That's much more fun!
Cheers! ;-)
I'm all for making improvements - that's the fun of being a DIYer.
However, I do think its important that improvements are done based on sound logic and preferably some objective measurements. Else we just get hand waving claims or suggestions that involve time and money but no audible improvement. Or even worse: tweaks that cost money and make the audio sound worse, not better.
So, are there any objective (frequency response, distortion) measurements that support the claim that the tweaks to the Touch result in audible improvements in the output audio?
If the claims are based on listening, were the listening tests done double blind (the gold standard for sorting out what is really happening)? If not, if they were done sighted there are well known placebo effects that mean people think they hear a difference when there is actually no difference.
Also, what is the logic behind the tweaks, the reasoning? If the claims are based on the superiority of the transport protocol (eg Ethernet vs. WiFi) then logic does not support those claims. The same data stream is provided to the DAC by the transport protocol, whatever it is (WiFi, HDD, Ethernet).
If the tweaks are based on the DAC or the subsequent downstream audio circuitry then again either objective measurements or proper listening tests will show if those claims are true.
If the differences are real it shouldn't matter how much a person has spent on their system - the improvements should be audible (with well designed tests). And it is certainly true that the more money you have spent on a system the more you will want to hear differences (even if they are not there).
Indeed.
I doubt that.
As DIYer you should not argue you should try it, especially if it doesn't cost you anything. This is NOT a marketing campaign!!
I am offering a pretty comfortable head start. The whole thing can be accomplished within 5 minutes - if you have a little understanding about computers.
Define "sound logic".
Am I - as a DIY person - supposed to purchase measurement equipment for 20k+$ to proove that a 299$ device sounds better when tweaked? Nonsense. I couldn't care less.
Everybody who is doing DIY for quite some time knows that you need to apply certain modifications to any product you purchase. There is nothing like a perfect commercial product out there. For me that's the same with Linuxed based PCs. The Touch is a Linux based PC. I just had apply all the findings I made over the last 4 years.
I offer my tweaks and you can try it.
The funny thing is that the ones who scream loudest don't even have that device at home.
And I said more than once this is not a hand waving claim. I provided my modification file to several trustworthy open minded persons. They all reported the same result.
TBM doesn't have it yet. Though I know he knows what I am talking about.
And -- it is not about "price" if changes become audible or not.
It is about quality of the system, the engineering behind it and it's about avoiding obvious bottlenecks.
A Sabre 32 DAC at 250$ and a Tripath 2021B amp at 50$ reasonably powered will tell you what's going on. There is no need to spent more on the
electronics side.
Cheers
General comment:
There are IMO 4 main mods to apply to the Touch.
1. Replace the PS
2. Remove the monitor
3. Manage the software
4. Flac decoding to be done on the remote server
Option - This needs a bit more DIY spirit:
(5). For people running analog out - bridging the poor coupling electrolytics
of the DAC output is a must!
What I do under 3.
I mainly shutdown all irrelevant processes - which requires to use the Touch in wired mode and remote server mode. And I do some rt-kernel scheduling adjustments.
For 1. I use a Super Teddy Reg powered by a 12V SLA battery.
For 2. I remove the screen ( Once the Touch is configured you can open the box and remove the front panel - the gain will be as high as the SW mods. Just turning off the screen, with the screen-off screensaver won't do. ) The monitor mod is nothing else then making the Touch a headless device.
You can look up relevant posts in the SB forum. John Svensson ( who's been official beta tester for the Touch) came up with the monitor mod. He also
applied my mods btw. and also confirmed what I am claiming.
BTW: I am preparing right now a Linux related HowTo. This will make life easier for everybody.
Cheers
Last edited:
Hi folks.
I just launched a site where you can download the script and instructions and some other hints.
It is the first drop!
If something is unclear, faulty or missing let me know. I didn't have time to do a lot of testing and proof-reading.
I tried to make it as simple as possible, even for people less experienced
with computers.
I intent to extent that site over time. I do have some more mods in mind.
Enjoy.
I just launched a site where you can download the script and instructions and some other hints.
It is the first drop!
If something is unclear, faulty or missing let me know. I didn't have time to do a lot of testing and proof-reading.
I tried to make it as simple as possible, even for people less experienced
with computers.
I intent to extent that site over time. I do have some more mods in mind.
Enjoy.
It's perfectly reasonable to argue that someone shouldn't try a DIY modification, particularly if:
- its not based on sound engineering principles and might in fact make the sound quality of the device worse, or
- the claimed performance improvements are either not based on objective measurements or well constructed listening tests, or
- the mod might void the warranty or the process of doing the mod might in fact damage the equipment.
By the way, I don't know if your modifications make improvements to the sound of the Touch. I do know that simply claiming that they do without any evidence to back it up is not a sufficient reason to undertake the mod.
How long it takes to make the change is immaterial. Either there is good reasoning behind the change and it is likely to make a difference, or there isn't.
Reasoning that lets us distinguish wishful thinking from claims about the world that may actually be correct.
For instance, I'm sure you would agree that if someone said "Linux runs slower than Windows on the same hardware" you would argue against that based on common knowledge we have of the performance of Linux on PC hardware, and the logic that that is unlikely to be true given what we know about the way Linux is constructed as an operating system.
In the same way we can dismiss claims about improved sound quality that are based on changes to equipment that go against commonly understood engineering knowledge.
Of course you don't have to buy expensive test equipment. But if you're being intellectually honest you could at least do a double-blind test to prove to yourself, as well as others, that the change really is something that's an improvement, as opposed to just wishful thinking.
People don't have to apply mods to equipment that they own. The reason that people apply mods is usually to get an improvement in sound. And at some level the effort required needs to deliver a benefit that exceeds the cost of doing the modification.
See the previous comments on double-blind verses sighted tests. You and others may be very sincere in your perceptions that there is an improvement but you could very well be mistaken about whether an actual improvement occurs.
The other possibility is that the change suggested maybe one where we don't even need to do any tests because of what we know about the engineering processes within the device (eg see the Wifi transport protocol discussion a few posts back).
Agreed.
If those "bottlenecks" are indeed something that has an impact on sound quality - that is still an open question. To date no explanation has been provided about how the load on the receiver or the so-called bottlenecks actually have a demonstrable impact on sound quality.
-----------------
Soundcheck, I'm sorry if my comments come across as overly negative towards you. I have greatly appreciated your advice in the past about Linux systems and it has helped me with my own DIY projects involving Linux and multichannel sound systems.
I do however feel quite strongly (as you can see!) that we need to be careful about the claims we make about doing stuff to well engineered consumer devices.
I couldn't care less about your comments. Not any of them makes me feel that you understood anything what I said or wrote.
People like you are the ones who're undermining the motivation of people who share stuff with the community. You might think about that!
I spent quite some time to prepare stuff for the community.
At no cost and commercial interest.
You bet, I won't spent more time and money to please certain people.
I am one of the few here, who delivers. With a little effort - everybody is able to replicate what I've done. From my perspective this is fair enough.
Instead of writing nonsense posts, buy yourself a Touch and try it and do the measurements. Make as many A/B tests as you like.
Don't forget to hand in your modification proposals soon. We're waiting.
Talking about great engineering: Read the the comments of users - the target group - consumers - at Amazon.de.
Pick the 1* 2* 3* ratings. I'd 100% second those comments over there. If I wouldn't use this device for a "different purpose" I never would have bought it.
The Touch is good as high-end transport when tweaked, stripped down and not used in local server mode. In this config it shouldn't cost more then 100$ though - like the Duet receiver!!
My very conditional recommendation of that box is valid only for that particular use-case!
No further comments on the subject. People know now where to get the stuff.
I am outta here.
Enjoy.
Thank you for graciously sharing your mods, soundchek. The instructions on the site are very clear, and I was able to apply the software mod under 5 mins.
The difference was immediately noticeable, and A/Bing was totally unnecessary. To my ears, the effect of the mod is akin to upgrading a DAC with better power supplies. The sound is more realistic and three-dimensional with no trade-offs. Very, very impressive for a mod that's so quick and easy to apply.
Thank you again soundchek!
For those interested: my super jung-powered SBT feeds SPDIF through a 75Ohm Canare-terminated Apogee coax cable to a Buffalo DAC with Paul Hynes PS mods.
Its perfectly reasonable to question whether a mod is going to make a difference. As far as I know its not a hanging offence (yet
).Are all claims to be automatically believed?
I have three squeezebox receivers at home. If I replaced the power cord on each and then claimed that the sound quality was "easily improved, no need to A/B the difference, anyone can hear it" would you automatically believe me? What if I got 10 other people to say that they too had replaced their power cords and had heard the difference, and had given the mod 5 out of 5 for usefulness. Would you then be convinced?
Its not unreasonable to ask for cogent engineering explanations for why there would be a difference for a mod.
For instance, there seems to me no logic or casual explanation to support changing the power supply of a digital receiver and then expecting it to sound audibly better. But, there may be a good explanation (from an engineering point of view) for the mod - there are always things to learn. I'm curious to know about the reasons behind those sort of claims, but the reasons have not been forthcoming.
Hi guys.
I made an update on my blog. There is a new SW mod.
You turn off the combined digital/analog output. It's gonna be either anlaog or either digital - you choose.
See section 1.3 Alsa modification.
The result is more subtle then the 1.1 mod. You might need to do a bit of A/B testing. Though I bet you'll figure it out.
boconnor will provide the measurements later on I guess.
Enjoy.
Cheers
I made an update on my blog. There is a new SW mod.
You turn off the combined digital/analog output. It's gonna be either anlaog or either digital - you choose.
See section 1.3 Alsa modification.
The result is more subtle then the 1.1 mod. You might need to do a bit of A/B testing. Though I bet you'll figure it out.
boconnor will provide the measurements later on I guess.
Enjoy.
Cheers
A digital stream has two components. The data and the timing.
If, as you say, the data is bit-perfect regardless (and I would expect it to be) then the only factor left is the timing.
So, if as you say, the use of wi-fi, the display and USB-connected hard drive degrade the sound quality of the squeezebox, then they must be introducing large amounts of jitter into the digital out.
Personally, I have trouble believing that.
If, as you say, the data is bit-perfect regardless (and I would expect it to be) then the only factor left is the timing.
So, if as you say, the use of wi-fi, the display and USB-connected hard drive degrade the sound quality of the squeezebox, then they must be introducing large amounts of jitter into the digital out.
Personally, I have trouble believing that.
The person making the claim should be the one to either:
a) provide some objective measurements, or
b) show results from controlled listening tests.
I'm not the one claiming these mods make a difference.
OK, so getting test results is not possible due to a lack of test equipment, fair enough.
How about some explanation about the basis of the mods - how do they work, given what we know about software, transport protocols and DACs. For instance, how specifically does changing the software improve the sound, given that the DAC and the following analogue circuitry is responsible for the sound quality?
Hearing is not believing if the listening is not controlled for bias and wishful thinking.
Hi,
I have been following this thread about the squeezebox touch. I'm thinking of getting one.
@Soundcheck- I read about your proposed mods, if you disable the screen there is no real reason for getting the SB Touch. Is there any other device out there that would do the same job instead of the Touch?
Thanks
I have been following this thread about the squeezebox touch. I'm thinking of getting one.
@Soundcheck- I read about your proposed mods, if you disable the screen there is no real reason for getting the SB Touch. Is there any other device out there that would do the same job instead of the Touch?
Thanks
I'm not the one making the claims about improved sound quality from doing things like disabling a screen.
If there are no test results available then what about an explanation of how, specifically, the mods affect the sound quality? Who knows, it might be relevant for the three squeezebox receivers I have.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.