Go Back   Team-BHP > Under the Hood > In-Car Entertainment


Reply
 
Thread Tools Search this Thread
Old 9th September 2008, 14:04   #1
Senior - BHPian
 
kpzen's Avatar
 
Join Date: Oct 2004
Location: Faridabad
Posts: 5,545
Thanked: 1,598 Times
Default 1-bit vs 24-bit DAC in HU's???

I'm in search of HU for my Getz ICE. While doing some research work i noticed that few HU's have 1-bit DAC while some newer HUs have 24-bit DACs.
Will this make that much of a difference in sound quality/power?
Basically, will I regret purchasing a 1 bit DAC system instead of a 24-bit DAC system?
kpzen is offline   Reply With Quote
Old 9th September 2008, 14:09   #2
Senior - BHPian
 
Join Date: Mar 2006
Location: Faridabad
Posts: 6,814
Thanked: 294 Times
Default

In simple language NO.

I will not make much of a difference.

Choose that HU which you like in features, sounding, looks. Other than this is all marketing crap.
low_bass_makker is offline   Reply With Quote
Old 9th September 2008, 14:11   #3
Senior - BHPian
 
greenhorn's Avatar
 
Join Date: Apr 2007
Location: TN-14
Posts: 6,838
Thanked: 1,318 Times
Default

no I dont think you'd regret it . there are good examples of both cases. as always , let your ears be the judge
greenhorn is offline   Reply With Quote
Old 9th September 2008, 14:17   #4
Senior - BHPian
 
kpzen's Avatar
 
Join Date: Oct 2004
Location: Faridabad
Posts: 5,545
Thanked: 1,598 Times
Default

Ok! Just for my knowledge.
What exactly is this 1 bit and 24bit DAC??
Is it related to sound output/quality or resolution?? To my knowledge it is the resolution though i might be wrong.
kpzen is offline   Reply With Quote
Old 9th September 2008, 14:28   #5
Senior - BHPian
 
greenhorn's Avatar
 
Join Date: Apr 2007
Location: TN-14
Posts: 6,838
Thanked: 1,318 Times
Default

they relate to how the analog signal is reconstructed from the digital info. in 24 bit DAC's , each individual sample is created independently, while in a 1 bit DAC, the signal is recontructed by increasing and decreasing the output as required. It decreases the number of steps the DAC can take , but compensates by increasing the number of steps it takes. Thus is solves the problem of non linearity with high resolution DAC's For example , if you had a digital signal like this

Time. 0 5 10 15
Value 0 1 6 10

the output of a 24 bit dac would look exactly as the above values.

a 1 bit dac working at 5 times the speed would look like
time. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Value0 0 0 0 0 1 2 3 4 5 06 06 07 08 09 10

so effectively it takes steps of 1, but faster, instead of the full 24 at one go

Last edited by greenhorn : 9th September 2008 at 14:30.
greenhorn is offline   Reply With Quote
Old 9th September 2008, 14:35   #6
Senior - BHPian
 
Join Date: Mar 2006
Location: Faridabad
Posts: 6,814
Thanked: 294 Times
Default

It means one and the same thing. KP dont go into such technology stuff rather look for features like Dolphins.. they make you relax along with music
low_bass_makker is offline   Reply With Quote
Old 9th September 2008, 14:51   #7
Senior - BHPian
 
greenhorn's Avatar
 
Join Date: Apr 2007
Location: TN-14
Posts: 6,838
Thanked: 1,318 Times
Default

to make an analogy , You have to create a mark the height of 10 bricks. You can use 10 different bricks , or 1 brick 10 times. The former can cause small differences because of the slight variations in size between the bricks which is avoided by using the same brick 10 times. The same principle is used. but practically you'll find that it introduces its own set of problems
greenhorn is offline   Reply With Quote
Old 9th September 2008, 22:13   #8
BHPian
 
Join Date: Jun 2007
Location: Pune
Posts: 268
Thanked: 10 Times
Default

e.g.

In one second, If you have to make 10 samples of audio (frequency) and 0 to 1 volt is range then

1. for 1 bit DAC, at each sample interval it could change by 1/10th V or no change. Because it will denote logic 1 to up the level by 1/10 V and 0 to lower the level by same amount. Error remains there as actual audio variation could not be multiple of 0.1V. So maximum error of 0.1V in case of each samples.

2. For 24 bits that sample value at each interval could be 2^24 times acturate means it could represent 1/(2^24) V change at each level. This means at each level change could be multiple of 1/16777216 V. So error minimizes to 1/16777216.

Still you have lost details between two audio samples and just considering that there is steady slope/change in current value of sample from previous sample.
So it depends more on sampling rate. more samples = nearly equal presentation.

But for sample frequency of 100KHz and 20KHz audio, variation in audio frequency is not more. Means 1 bit will have 1/100000 the change represented and 24 bit will have 1/16777216. So gap is narrowed between two methods.

So there could be minimal effect to human ears for 1-bit v/s 24-bit for high sampling frequency. Practically we listen upto 16KHz and sample rate is 192KHz.
omishra is offline   Reply With Quote
Old 9th September 2008, 22:25   #9
BHPian
 
valhallen.282's Avatar
 
Join Date: Aug 2008
Location: Bangalore
Posts: 315
Thanked: 129 Times
Default

ok in simple words the 24 bit dac will have better detailing than a 1 bit dac. essentially it will give you better headroom and clarity but in very miniscule differences and only with very high res audio. yes you do get better quality all the way upto 320kbits at 48khz or even 6 channel audio. the difference is more noticeable in home theatre amps where you are paying absolute rapt attention to the audio. some of the better new ones come with upto 4 dacs. but if you are listening to a lot of mp3 dont really bother. if you can get all the features and the price and the 24bit dac all in one it'll be a jackpot.
valhallen.282 is offline   Reply With Quote
Old 9th September 2008, 22:59   #10
Senior - BHPian
 
greenhorn's Avatar
 
Join Date: Apr 2007
Location: TN-14
Posts: 6,838
Thanked: 1,318 Times
Default

Quote:
Originally Posted by omishra View Post
1. for 1 bit DAC, at each sample interval it could change by 1/10th V or no change. Because it will denote logic 1 to up the level by 1/10 V and 0 to lower the level by same amount. Error remains there as actual audio variation could not be multiple of 0.1V. So maximum error of 0.1V in case of each samples.

2. For 24 bits that sample value at each interval could be 2^24 times acturate means it could represent 1/(2^24) V change at each level. This means at each level change could be multiple of 1/16777216 V. So error minimizes to 1/16777216.
.
for any given input sample rate, the sampling rate of the 1 bit DAC will be much higher.. so by varying the width of the pulse, you can achieve comparable accuracy, its pretty much the same principle as PWM
greenhorn is offline   Reply With Quote
Old 10th September 2008, 09:30   #11
Senior - BHPian
 
kpzen's Avatar
 
Join Date: Oct 2004
Location: Faridabad
Posts: 5,545
Thanked: 1,598 Times
Default

Thanks Greenhorn.
I got a 1bit DAC HU
kpzen is offline   Reply With Quote
Old 10th September 2008, 09:38   #12
Senior - BHPian
 
greenhorn's Avatar
 
Join Date: Apr 2007
Location: TN-14
Posts: 6,838
Thanked: 1,318 Times
Default

Quote:
Originally Posted by kpzen View Post
Thanks Greenhorn.
I got a 1bit DAC HU
Which one ?
PS: So do i
Quote:
Originally Posted by kpzen View Post
Basically, will I regret purchasing a 1 bit DAC system instead of a 24-bit DAC system?
Further PS: Huge sense of Deja VU : someone in texas had exactly the same doubt 2 years ago

Last edited by greenhorn : 10th September 2008 at 09:43.
greenhorn is offline   Reply With Quote
Old 10th September 2008, 10:05   #13
Team-BHP Support
 
navin's Avatar
 
Join Date: Mar 2004
Location: mumbai
Posts: 22,509
Thanked: 3,885 Times
Default

Quote:
Originally Posted by kpzen View Post
What exactly is this 1 bit and 24bit DAC??
Quote:
Originally Posted by low_bass_makker View Post
It means one and the same thing. KP dont go into such technology stuff rather look for features like Dolphins..
no it does not and beware of dolphins keep you eyes on the road.

Quote:
Originally Posted by valhallen.282 View Post
ok in simple words the 24 bit dac will have better detailing than a 1 bit dac.
not really.

A Fundamental Introduction to the Compact Disc Player

this harks back to the days when CD players started out with sampling rates of 44.1 kHz (often called Nyquist rate - google this and there are many explantions of it).

When the music signal is sampled at 44.1kHz on needs a filter (a steep filter) that can remove the artifacts of sampling. THis filter introduced its own set of problems. So someone upped the samplng rate (called over-sampling) to 88.2 and then 176.4 and even 352.8kHz and all of a sudden everyone who prescribed to the Julian Hirsch school of Audio (measurements is king) were screaming sampling rates and hence lower slopes for the low pass filter (at 20kHz). This is akin to the Megapixel race in digital cameras today.

See some genius in the marketing department (I believe the person had worked for voice telephony before - read as AT&T) figured that if sampling rates can be used to sell a CD player why not have a sampling rates of 1Mhz (after all 1Mhz clocks are cheap since PC-CPUs were now faster than 10Mhz). They even coined a word for it "Bit Stream".

Since we have a sampling rate of 1Mhz why have 8-12-16-24 bit of information we could use Delta Modulation and only use a 1 bit sample (this later came to known as Sigma Delta Modulation.

SDM has been around since the days of PCM - SDM lost out to PCM (Pulse Code Modulation) because the CPU clocks were not fast enough at the time of PCM and the fact that building a practical homogenous quantiser to "quantise the deltas" was expensive if not impossible). PCM really took root after Mitsubishi (X800) and Sony (3324) digital tape recording equipment became the vogue replacing everything from Akai and Technics to Revox and Tascam.

PWM is really a special case of PDM. I believe that PDM is probably the most efficient method to sample a signal (kinda like todays VBR vs CBR for MP3s). The algorithim is mathematically simple and that means fewer errors and fewer components required (hence less distortion) to implement the circuit.

oh my I'm rambling again. Sorry.
navin is offline   Reply With Quote
Old 10th September 2008, 10:10   #14
Team-BHP Support
 
navin's Avatar
 
Join Date: Mar 2004
Location: mumbai
Posts: 22,509
Thanked: 3,885 Times
Default

Quote:
Originally Posted by greenhorn View Post
Further PS: Huge sense of Deja VU : someone in texas had exactly the same doubt 2 years ago
Greenhorn,
I'd like to know what the person who could hear differences between the 20bit and 24 bit processors was smoking. Most likely it was differnces in the DAC that he heard.

If you want to know more about this stuff Moulton Labs had a nice set of explanations. I cant google like LBM can but here is their home page Moulton Laboratories

Last edited by navin : 10th September 2008 at 10:11.
navin is offline   Reply With Quote
Old 10th September 2008, 10:15   #15
Senior - BHPian
 
esteem_lover's Avatar
 
Join Date: Jul 2006
Location: Madras/Py
Posts: 7,554
Thanked: 445 Times
Default

Quote:
Originally Posted by low_bass_makker View Post
KP dont go into such technology stuff rather look for features like Dolphins.. they make you relax along with music
Lol !

Quote:
Originally Posted by kpzen View Post
Thanks Greenhorn.
I got a 1bit DAC HU
With the dolphin ? Which set is that ?
esteem_lover is offline   Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Amplifier DAC & DSP suggestion rahuls21297 In-Car Entertainment 15 1st June 2009 13:54
Hello All, Just a bit of a Ramble and a Little Intro. Big Kahuna Introduce yourself 13 16th December 2005 20:05
New games, are they being a bit too arcadish??? Gordon Shifting gears 15 23rd November 2005 16:32
can 32-bit ecu's be fitted in e-2 cars speedsatya Technical Stuff 4 12th March 2005 10:55


All times are GMT +5.5. The time now is 04:36.

Copyright 2000 - 2017, Team-BHP.com
Proudly powered by E2E Networks