TheMasterPC, you are welcome!
I would like to clarify two concepts of frequency to avoid confusion as below:
1) The highest sampling frequency at 24Bit/192KHz [Ref#1]
2) The audio frequency of 20Hz to 20KHz for human being [Ref#2]
Take home HiFi system as example, “sampling frequency” refers to the precision of digital signal processing (DSP) reproduced from source (i.e. digital music file) to destination (i.e. DAC chipset) while “audio frequency” refers to the signal frequency that people can hear physically from loudspeaker. (e.g. via digital, analog and electro-mechanical vibration). For this reason, a song with 100Hz audio signal can be DSP sampled by either 24Bit/48KHz or 24Bit/192KHz resolution both under Bit Perfect audio condition. The higher sampling frequency refers to the higher accuracy and precision which is in turn at the lowest jitter. In contrast, a 24Bit/48KHz sampling under “Bit Perfect” audio condition may be better than a 24Bit/192KHz sampling under “non-Bit Perfect” audio condition in term of jitter problem.
(Remark: Although 32Bit DAC chipset such as ESS Sabre ES9018 exists in the market, 32Bit/192KHz digital music file does not exist in the market by download from website. Hence, the highest resolution of source is at 24Bit/192KHz up-to-date unless upgrade in the future)
Sound logic to me,as almost all audio sources are recorded in 16bit/44.1 Khz ,it's not worth the effort to upsample to 24bit/192Khz,anything wich is in plus is rejected at the output.
We still don't have music in recorded at that rate not to mention the DSD,and the high price of the material, imagine if you buy a CD with 10euros or a BLURAY disc with 18 DSD files with 50 euros,the only difference that "you might hear" is the detail of the sound.
Anyway you get the point, 24bit/96Khz is more than enough.
Your point is good to mention, but not recommend to argue because too time consuming if just a paper work discussion. This point refers to “what is the optimum highest sampling rate for DSP in human being?” I have read some renowned audio reviews to mention that 24Bit/96KHz sampling rate is the best for human being and higher sampling rate performs no significant difference to human ear. I believe that it may be based on author’s practical listening testing. Since result may vary to different human ears, it is recognized as a good reference, but not a proved “true” statement. You may do your practical listening testing to verify or based on recognized research paper from authorized organization such as AES in USA. Anyway, either 24Bit/96KHz or 24Bit/192KHz is proved to be optimum highest sampling rate for DSP in human being, and then any further development to higher sampling rate such as 32Bit/352.8KHz becomes meaningless.
As a recording rate, 192Khz has some advantages. But as a delivery bit rate? Not worth it. 96/24 is just fine. Unless you are playing music for bats.
I'd like a device that does 96Khz, but I don't want to pay an extra $200 for it. By the time I have enough material recorded at 96Khz to matter, there will be a $60 player for it.
I'd like a device that does 96Khz, but I don't want to pay an extra $200 for it. By the time I have enough material recorded at 96Khz to matter, there will be a $60 player for it.
I want to add remarks as below:
The statement “the audio frequency for human being is between 20Hz and 20KHz” is based on statistical measurement, take 95% as example to say 95% confidence level that most people fall into this range because it is impossible to measure 100% of population in the world. It still exists 5% of confidence level that some people do not fall within this range. Some loudspeaker designers make his loudspeaker based on 20Hz~20KHz, but some designers make it at 20Hz~30KHz to cover more people at wider range of frequency. It does not matter. Government officer will not enforce it into regulation, but just concerns about the credibility of supplier if the written product specification to be 20Hz~30KHz while the actual product specification to be 20Hz~20KHz.
This principle also applies to the sampling frequency of DSP if an authorized paper is released to support “24Bit/96KHz sampling rate to be the optimum highest sampling rate of DSP for human being”.
1) If we go to 24Bit/96KHz, then it means that 95% of confidence level to cover most people. But, it still exists 5% confidence level that some people can hear the difference for higher sampling rate.
2) If we go to 24Bit/192KHz, then it provide higher range to cover slightly more people under constraint of resource.
Personally, I prefer to go to “24Bit/192KHz” sampling rate due to the following reasons:
1) 24Bit/192KHz digital music file exists in the market and Intel HD codec already supports Bit Perfect audio condition up to stereo 24Bit/192KHz sampling rate.
2) In term of technical capability and cost impact, there is no technical difficulty and without significant cost impact to go to 24Bit/192KHz sampling rate in comparing to 24Bit/96KHz sampling rate.
3) If we buy audio gear to support 24Bit/192KHz, then it covers 24Bit/96KHz file or below. If we buy audio gear to support 24Bit/96KHz, then it doe not cover 24Bit/192KHz file.
4) 24Bit/192KHz sampling rate may be an end point and does not recommend to go to higher sampling rate such as 32Bit/384KHz, etc.. as seems to be meaningless.
5) If you say to me that you recommend at 24Bit/96KHz sampling rate, I personally do not refuse it because it is logical and reasonable with respect to your preference.
The statement “the audio frequency for human being is between 20Hz and 20KHz” is based on statistical measurement, take 95% as example to say 95% confidence level that most people fall into this range because it is impossible to measure 100% of population in the world. It still exists 5% of confidence level that some people do not fall within this range. Some loudspeaker designers make his loudspeaker based on 20Hz~20KHz, but some designers make it at 20Hz~30KHz to cover more people at wider range of frequency. It does not matter. Government officer will not enforce it into regulation, but just concerns about the credibility of supplier if the written product specification to be 20Hz~30KHz while the actual product specification to be 20Hz~20KHz.
This principle also applies to the sampling frequency of DSP if an authorized paper is released to support “24Bit/96KHz sampling rate to be the optimum highest sampling rate of DSP for human being”.
1) If we go to 24Bit/96KHz, then it means that 95% of confidence level to cover most people. But, it still exists 5% confidence level that some people can hear the difference for higher sampling rate.
2) If we go to 24Bit/192KHz, then it provide higher range to cover slightly more people under constraint of resource.
Personally, I prefer to go to “24Bit/192KHz” sampling rate due to the following reasons:
1) 24Bit/192KHz digital music file exists in the market and Intel HD codec already supports Bit Perfect audio condition up to stereo 24Bit/192KHz sampling rate.
2) In term of technical capability and cost impact, there is no technical difficulty and without significant cost impact to go to 24Bit/192KHz sampling rate in comparing to 24Bit/96KHz sampling rate.
3) If we buy audio gear to support 24Bit/192KHz, then it covers 24Bit/96KHz file or below. If we buy audio gear to support 24Bit/96KHz, then it doe not cover 24Bit/192KHz file.
4) 24Bit/192KHz sampling rate may be an end point and does not recommend to go to higher sampling rate such as 32Bit/384KHz, etc.. as seems to be meaningless.
5) If you say to me that you recommend at 24Bit/96KHz sampling rate, I personally do not refuse it because it is logical and reasonable with respect to your preference.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.