"Audiophile Optimizer"....fish oil...?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Well-meaning automagic system "optimization" tools of any kind tend to create more problems than they solve. Besides, if you've got playback that already is free from stuttering and the like, there is no way it could get any better. Audio per se is in no way particularly taxing (heck, a 386 would output 16/44 audio via ISA DMA at rather negligible system load back in the day), it merely is a bit time-critical.

If you do have problems, there are multiple things you can investigate.
Download and run DPC Latency Checker. You may find the occasional high spikes. The most frequent culprits are network (particularly WLAN) drivers, so try updating these. Some systems also have issues with USB (and thus USB sound devices) when changing processor power states; one would hope that this is a thing of the past now.
You can also try tweaking the network throttling option.
Finally, you can investigate whether your playback application supports invoking the MMCSS service (Foobar2000: Preferences --> Advanced --> Thread Priority).

In the olden days it would also not have been unusual to shuffle around expansion cards to resolve cases of unlucky interrupt sharing (for example, an ES1370 based soundcard didn't get along with an LSI U2W SCSI host adapter, while it didn't mind a Matrox Millennium II). In my old 440BX machine, I additionally had to set the PCI latency timer for the USB controller to 0 in order to get rid of crackling in sensitive cards.
 
Last edited:
Services, drivers, processes....anything that takes cpu cycles other than audio playback.

If you've got anything better than a Pentium Dual Core you shouldn't worry about freeing up cpu cycles for music.

Unless the system is so loaded that the song is really skipping, getting 0% usage on a quad core is not going to do anything.

This is about as silly as audiofools saying cables are directional......

Well-meaning automagic system "optimization" tools of any kind tend to create more problems than they solve. Besides, if you've got playback that already is free from stuttering and the like, there is no way it could get any better. Audio per se is in no way particularly taxing (heck, a 386 would output 16/44 audio via ISA DMA at rather negligible system load back in the day), it merely is a bit time-critical.

If you do have problems, there are multiple things you can investigate.
Download and run DPC Latency Checker. You may find the occasional high spikes. The most frequent culprits are network (particularly WLAN) drivers, so try updating these. Some systems also have issues with USB (and thus USB sound devices) when changing processor power states; one would hope that this is a thing of the past now.
You can also try tweaking the network throttling option.
Finally, you can investigate whether your playback application supports invoking the MMCSS service (Foobar2000: Preferences --> Advanced --> Thread Priority).

In the olden days it would also not have been unusual to shuffle around expansion cards to resolve cases of unlucky interrupt sharing (for example, an ES1370 based soundcard didn't get along with an LSI U2W SCSI host adapter, while it didn't mind a Matrox Millennium II). In my old 440BX machine, I additionally had to set the PCI latency timer for the USB controller to 0 in order to get rid of crackling in sensitive cards.

Thanks for backing me up. I completely agree, back then you did have many little problems that were hardware specific usually, but as you said, " if you've got playback that already is free from stuttering and the like, there is no way it could get any better."
 
I'm very skeptical of the whole thing, naturally...even more so that it costs 100 EU and you can't trial it!! A lot of people who use it, swear by it...but this is very subjective.
I understand what you're saying about CPU load and the like...but I've seen powerful PC's cough and hick up and stutter even with CPU load very low and latency at a reasonably, if not ideal, level.
I wonder if many setups maybe "borderline" in this respect but not consciously noticeable...? Until the system is optimized a little more....?
 
There is a free script that you can download that will stop most unneeded services, disable the page file, change the priorities of important interrupts (timer, soundcard, usb controller, etc) and a few other things on a Windows 8 machine, with the purpose of optimizing the machine for music playback. No need to pay over $100 for that.
 
This Audio Optimizer script is design to work best with a 2 PC setup and use JPlay as the network transport. When I mention DNLA instead of JPlay (because JPlay is also another fish oil potion!), I got very negative remarks about using Dnla. I wouldn't be surprised that there isn't some financial benefit/ agreement between this Swiss magic script and JPlay.
When I talked about how this script must be altering the music output (as there as several options to choose from for dialling in the sound), I was told that the data is NOT altered in any way...the difference in sound from the different options is from the different optimizations settings within the OS...
No other elaboration was given, only to say that a magician doesn't tell how their tricks are done!
A paid, 14 day trial is available now however...
 
Hi

There is no DSP or alteration of the source signal. As you know my software makes permanent changes to the OS, but there is no EXE or something else running during music playback. So there just can't be any DSP of any form.

What i do is change settings of the operating system which will directly influence sound reproduction, but there is still not a single bit changed in the source file or output.

...fishoil? :) definitely not -> http://supratekaudio.blogspot.com.au/2014/03/finally-beautiful-digital-sound.html


Best,
Phil
 
with 100 euro I will buy a BBB and donate the rest to the Open Source community. at least it will be some good spend money


I agree.

Sure using a server OS is going to use less resources but Highendaudiopc please please tell me how using less ram and cpu cycles is going to give me a bigger soundstage, well defined bass, etc.

And if that is the key, explain why you must use expensive closed source windows server, and not a simple command line install of Debian, or your favorite distribution of free open source GNU/LINUX?

It is simply placebo, if anything.
 
What is really interesting is that every and each reviewer of your product forgets to add the final price (software and hardware) included for the whole system.

By the way did you made a (I was going to say blind but...) test between a arm hw platform (let's say i-cubox with mpd and alsa) and your "product" (installed on 2 computers with crippled operating systems etc etc )? Please do and come with a nice review, or pay someone to do it, I presume giving a free "license" of your product will do it also.

You can try to spam in this thread also http://www.diyaudio.com/forums/lounge/234829-funniest-snake-oil-theories.html.

Sure you will get a a lot of feedback.
 
i won't tell you no nothing about how it works, so easy it is.

no, there is no placebo. but you guys know it better anyway.

NO MATTER if you have tried it or not. Sorry, but this is a
waste of my precious time my friends...

learn to read -> Highend-AudioPC | Reviews


Sorry but I really do feel that you are just someone else trying to pull one over on those less knowledgeable with computers. I'm sorry to say but I'm not one of them.

I've been repairing, and assembling computers and networks for about 6 years including quite a few high end gaming pc's, I've run servers for both games and web (with php, mysql etc). I've used windows server, various linux distros, I can feel my way around HTML, PHP, C++, Assembly, and more.

I'm also sorry to say it but surely you must know how easy it is to fake a few dozen reviews. I'll even give you the benefit of the doubt and say that perhaps you did get all of those reviews and only doctored them a bit.

You can not tell me that using a different OS, stripped down, on any powerful modern computer will improve the sound in any way.

I mean what argument could you be trying to make, perhaps with less load on the PSU there is more wattage available to the sound card? Gimme a break


And as cmiu007 pointed out, tell me again how much legitimate Windows Server 2012 licenses are. Even for the "Essential" edition it is around $400. Then hardware, which I'm sure you recommend nothing less than a huge X79 rig with a 4930k but maybe I'm wrong.

Then if you still want to even keep going you have to pay for your program too! HAH
 
Last edited:
Foobar + WASAPI or ASIO ... that is all you need, any moderm computer can run with very low levels of latency.
There is a few tweaks u' can try to lower the latency:
>disabling services, drivers you don't use and energy saving
>enabling HPET timer in BIOS and windows, disabling CPU energy saving options (EIST, C1, C3 ect)
>If you have W8 this command sometimes works bcdedit /set disabledynamictick yes

I'm pretty sure this "Snake-oil optimizer" do a few of this ....
But there are more things at hardware level that affect sound in a bigger grade. The noise generated by the disks in the 12v line, some video cards or motherboard generate audible interferences, PSU ripple ...

BTW: IN W8 DSP Latency Checker reports
imprecise values of latency use LatencyMon
 
Phil's optimiser package - in the context of JPlay and two PCs running Server 2012 - does offer sonic benefits, despite what any rational engineering approach might have us expect. A more relaxed, organic and detailed sound can be achieved.

I tested it during the development phase as I was intrigued by early reports. It is worth noting that the developer is himself a software engineer. I walked away from it when the price of the Optimiser was set simply because it was pricey, and required 2 MS Server licenses on top of the hardware and running costs.

Swapping from such an elaborate (and quite expensive) arrangement to a RPi running MPD over a direct I2S into the same DAC produced a subjectively more appealing sound.
 
despite what any rational engineering approach might have us expect.

It is not despite rational engineering approach, if you take off the HDD from the same computer and run a voyage mpd from a USB stick and you will have a lower electrical noise due to several factors.

Take for example any linux distribution run X and some programs in the same time with mpd. Do a "cat /proc/interrupts" stop everything apart from mpd and do the same and compare the results, you can take this example a little bit forward and record the values and draw an nice comparative graph using time as axis. Unfortunately I don't know any program to record the no of interrupts in windows in order to conclude the exercise.

The only new thing (in windows ecosystem) the this product is introducing is the headless server with a separate client, but this is old news if you compare it with mpd.

You can "improve" the sound using mpd + brutefir with a carefully tweaked impulse response filter and not paying for jplay+jriver+windows+ and the magic script that is not doing anything else apart for stoping some services and adjust the length of the USB buffer. I don't deny that it takes a lot of trial and error to get this blindly using windows but really this is not coding is scripting.

After Jriver took a stand against jplay, the guys from jplay took any oportunity to keep the myth going and even they hosted the development thread of a.o. on their forum.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.