[514077]

  Next up... DSD playback

 
1. DSDIFF - "reports" sample rate (to external DAC) based on what you set its one & only parameter: "Max Sample Rate". Oddly, it defaults to 88200 (and max is 192000.)
In the end, I have no idea what Foobar is sending my DAC...

 
2. foo-input-SACD (ver 0.9.11) - obviously designed to do more than just DSD playback but it seems to play well with DSD. Aside from selecting either ASIO or WSAPI (event / push), the only parm (AFAIK) is its "output mode" (PCM or DSD, the latter being the obvious choice). I'm assuming it is a DoP based deal. It correctly "reports" the defined sample rates for both DSD128 & 256 (5.6MHz & 11.2MHz).
Oddly enough, the only issue I've come across is that it does not feed the Visualizations (Spectrum, etc.) when playing DSDs (PCMs are fine)... . Certainly not the worse issue and I prefer it over DSDIFF's poor "reporting"... .
 
I would be curious to know what you think (or know) to be the best DSD solution for Foobar. Important to me is that said solution does not degrade PCM playback in any way as my media in 99.5% PCM. Not that Foobar's Output-Selection feature could be any easier, but I would hope that at the end of this rainbow there is one perfect solution...

I tried version 0.9.7  (latest version) and couldn't get it to work.  At someone's advice, I went to version 0.8.4 and run DSD through DOP.  White light on Hugo and all.
I've been wondering if there's a way to convert DFF (DSDIFF) to DSF, just for tagging purposes.

 

[Traveller]

others will have to help here, I'm a total DSD hater so I never even tried to play any. sorry.

No worries

I suppose there's a thread (or three) with the pros and cons of the two formats...

 

[Soundsgoodtome]

I've asked if Foobar2k's digital volume was 32-bit float. Now I wonder if it is a 64-bit float on the volume? I've been doing some reading and been doing some digital attenuation until a preamp comes in between my dac and active speaker, wondering how much I can drop it before losing data.
 

 

[514077]

 

Originally Posted by castleofargh /img/forum/go_quote.gif
 
asio4all as it's name implies is a tool that pretends to be asio when no proper asio driver are is provided for the DAC. but the actual system used I believe is kernel streaming(I'm not sure of that but I've read it somewhere). so if you already have actual asio drivers for the hugo, it would be meaningless to have asio4all. 

 
That's correct: asio4all makes sense in the best case if your player does not support any bit perfect connections, but otherwise is just an unnecessary complexity and latency to deal with. Chord drivers support native ASIO so there's no need to use asio4all in that case.

Which leads me to a stupid, but I feel necessary question.  What is latency, what does it affect, and what should it be set to?

 

[PleasantSounds]

 
Which leads me to a stupid, but I feel necessary question.  What is latency, what does it affect, and what should it be set to?

 
In generic terms latency is a time lag between cause and effect. In f2k it may surface as a delay when adjusting volume, EQ settings or other DSP functions it performs.
 
You cannot set it to a specific value - it is a side effect of things that have to happen between reading the data from a file and sending it to the DAC. Indirectly you can reduce latency by keeping buffer sizes to a reasonable minimum (without causing pops and clicks) and not going overboard with your DSP chain.
 
While it is quite critical to keep the latency to a minimum in the music production, it is much less important (but still desirable) for casual listening. 

 

[theaudiologist]

does 32-bit(float)/384khz and 24/384 khz make a difference?

 

[PleasantSounds]

  does 32-bit(float)/384khz and 24/384 khz make a difference?

 
Yes it does:
- music in these formats tends to be more expensive and not that easy to find
- it takes 10 x the space of a redbook CD
- it forces your PC to work harder
- pops and clicks are more likely to happen 
That's as much as can be objectively ascertained.
 
The impact on sound quality is debatable at best. Personally I know that with my aging ears on my fairly high end equipment I can't tell them apart from 24/48kHz copies made off the same master. There are those who claim differences from subtle to "night and day", but unless you empirically find out that belong in that camp, I'd recommend sticking to the redbook CD which is capable of delivering stellar quality. Focus on finding the best possible mastered edition instead, as this will make much bigger difference.

 

[514077]

 

 
Which leads me to a stupid, but I feel necessary question.  What is latency, what does it affect, and what should it be set to?

 
In generic terms latency is a time lag between cause and effect. In f2k it may surface as a delay when adjusting volume, EQ settings or other DSP functions it performs.
 
You cannot set it to a specific value - it is a side effect of things that have to happen between reading the data from a file and sending it to the DAC. Indirectly you can reduce latency by keeping buffer sizes to a reasonable minimum (without causing pops and clicks) and not going overboard with your DSP chain.
 
While it is quite critical to keep the latency to a minimum in the music production, it is much less important (but still desirable) for casual listening. 

Thanks.  I think I confused latency with the adjustable buffer size.  In FB2K, it seems to usually come at 1000 (I think) ms.  Can or should it be set lower, and how much is reasonable?

 

[PleasantSounds]

  Thanks.  I think I confused latency with the adjustable buffer size.  In FB2K, it seems to usually come at 1000 (I think) ms.  Can or should it be set lower, and how much is reasonable?

 
This depends on your setup. I have 3 installations and each requires different settings. 1000 ms is a good starting point, but my main PC requires 1200 ms to completely avoid buffer underruns, especially when heavily multitasking.
My tablet is happy with 400 ms, but that's a much newer architecture and I don't use it for things like video rendering while playing music.
 
From my experience the right buffer size has two boundaries: one is when pops and clicks start getting in the audio stream (the lower limit) and the other when the lag between setting controls and hearing the result is more than you can tolerate (happens with too large buffer). The  problem is when the two overlap. Sometimes updating USB drivers helps, sometimes you have to stop some services which get too much in the way (latency monitor is a great tool for that). The last resort is moving your audio setup to a platform that can handle it better.

 

[514077]

 

  Thanks.  I think I confused latency with the adjustable buffer size.  In FB2K, it seems to usually come at 1000 (I think) ms.  Can or should it be set lower, and how much is reasonable?

 
This depends on your setup. I have 3 installations and each requires different settings. 1000 ms is a good starting point, but my main PC requires 1200 ms to completely avoid buffer underruns, especially when heavily multitasking.
My tablet is happy with 400 ms, but that's a much newer architecture and I don't use it for things like video rendering while playing music.
 
From my experience the right buffer size has two boundaries: one is when pops and clicks start getting in the audio stream (the lower limit) and the other when the lag between setting controls and hearing the result is more than you can tolerate (happens with too large buffer). The  problem is when the two overlap. Sometimes updating USB drivers helps, sometimes you have to stop some services which get too much in the way (latency monitor is a great tool for that). The last resort is moving your audio setup to a platform that can handle it better.

Thanks for everything.  I'll try that monitor.

 

[ExtremeGamerBR]

  Next up... DSD playback

 
1. DSDIFF - "reports" sample rate (to external DAC) based on what you set its one & only parameter: "Max Sample Rate". Oddly, it defaults to 88200 (and max is 192000.)
In the end, I have no idea what Foobar is sending my DAC...

 
2. foo-input-SACD (ver 0.9.11) - obviously designed to do more than just DSD playback but it seems to play well with DSD. Aside from selecting either ASIO or WSAPI (event / push), the only parm (AFAIK) is its "output mode" (PCM or DSD, the latter being the obvious choice). I'm assuming it is a DoP based deal. It correctly "reports" the defined sample rates for both DSD128 & 256 (5.6MHz & 11.2MHz).
Oddly enough, the only issue I've come across is that it does not feed the Visualizations (Spectrum, etc.) when playing DSDs (PCMs are fine)... . Certainly not the worse issue and I prefer it over DSDIFF's poor "reporting"... .
 
I would be curious to know what you think (or know) to be the best DSD solution for Foobar. Important to me is that said solution does not degrade PCM playback in any way as my media in 99.5% PCM. Not that Foobar's Output-Selection feature could be any easier, but I would hope that at the end of this rainbow there is one perfect solution...

foo-input-SACD is the best. Just read the readme.txt file and be happy.

 

[ScareDe2]

edited

 

[PinkyPowers]

I'm trying to get Electri-Q to work, using the George Yohng's VST Wrapper.

After hitting Rescan All a few times, it finally finds \dsp_eqfree.dll and CLAIMS to be "Testing" said plugin. But all it's really doing is locking the F up.

Is there by any chance legitimate software that can give foobar a decent EQ?

 

[castleofargh]

is your problem with electriQ while other VSTs work fine? or is it a problem with the wrapper?

 

[PinkyPowers]

Good question. I'm not using any other VSTs, so I don't know.