FLAC vs. 320 Mp3 - Page 7  

Quote:

Originally Posted by BlindInOneEar 

No disrespect, but the fact that a forum poster convinced YOU that he could tell the difference between high bit rate lossy and lossless hardly constitutes "scientific fact" that high bit rate lossy is audibly distinguishable from lossless, or that there is a select group of individuals who can make the distinction.  The incident may be interesting.  It may even be worth investigating further.  What it's not, though, is proof.

Again - no disrespect - but it goes both ways.  With bigshot and Lan647 saying conclusively that it is indistinguishable - but someone like DeadlyLover showing Foobar logs on volume matched samples that say they can distinguish - then I guess we are left with the premise that most people seem unable to show proof that they can differentiate (I certainly can't), however it seems possible that a small percentage of individuals may be able to differentiate.

Either way - I agree - the current data seems inconclusive :)

Quote:

Originally Posted by Brooko 

 

Again - no disrespect - but it goes both ways.  With bigshot and Lan647 saying conclusively that it is indistinguishable - but someone like DeadlyLover showing Foobar logs on volume matched samples that say they can distinguish - then I guess we are left with the premise that most people seem unable to show proof that they can differentiate (I certainly can't), however it seems possible that a small percentage of individuals may be able to differentiate.

 

Either way - I agree - the current data seems inconclusive :)

I agree.  The "data," if we want to give that level of dignity to claims made on internet forums, is inconclusive on both sides of the argument.

Quote:

Originally Posted by Clarkmc2 


You would think that the distortion analyzer had never existed. I use my ears, I have been known to enjoy distorted reproduction, but I trust in o scope traces for facts.

I'm not sure what your point is.  I thought the argument was whether differences could be heard between high bit rate lossy files and lossless files, not whether or not differences existed.  I think we'd all agree that there are differences between the two formats.  The dispute is whether or not they are audible. 

Quote:

Originally Posted by bigshot 


There is one fact that is conclusive... If in fact a difference does exist, it's small enough to be considered insignificant. For all intents and purposes, the sound quality of high bitrate AAC and lossless is equivalent.

You've the misfortune of having staked out the unprovable side of the argument.  However large our data set, however many people we test that can't hear a difference, that does not mean there isn't a paragon somewhere out there who actually can hear a difference.  As you say, we can be very confident that the two formats are equivalent, but we can't be certain.  The other side of the issue has a much easier case.  All they have to do is come up with one person, just one, who can demonstrably hear the difference and their proposition is proved.  That's why I pointed out to nanaholic that though one internet poster may have convinced nanaholic the difference is audible, that's not actually proof that the difference is audible. 

I just use flac because I have enough hard drive space to hold as much music as I could ever want and I also wanna give my gear the best possible chance of sounding great

Quote:

Originally Posted by Clarkmc2 


Sorry to have been unclear. I was not speaking in the particular, but in general. That said, my tie in to this issue is that a distortion analyzer - at this point most known as an old HP piece of gear by some - shows how far down, in dB if so set up, the distortion is. If someone says they hear it and it is 50 dB down, I would be doubtful, to put it kindly.
The distortion analyzer was used for amps and such, but I have no doubt similar osilloscope based equipment could be set up to measure most things we debate here. Need I mention that a scope is the easiest way to do a null test as well? That is the long version of what I meant. If I'm flat wrong about that, I'm sure someone will let me know.
Listening tests are like arms control to me. Trust but verify.

Thanks for the explanation!  That makes things clearer for me. 

I'll take your word that an oscilloscope is the easiest way to do a null test, assuming you have one lying around!  Not being so blessed I've had to limit myself to fiddling with the Audio Diffmaker.  Even there I'm only using the provided dyf files as I'd have no idea how to record my own.  Still, just that little introduction has been a revelation for me.  Soft sounds are hard to hear when loud music is playing!  People often quote the decibel scale as if the ear can easily hear the full range of it at once, from 0 db to 130 db.  I wonder if there is actually a much narrow range of loudness that the ear can hear at the same time.  Sure, you can hear leaves rustling on a quiet day, but I'll bet you can't while you're at a rock concert.  I'll also bet that the ability to hear noise or distortion drops off pretty quickly below -60 db or so.   

Quote:

Originally Posted by BlindInOneEar 

Thanks for the explanation!  That makes things clearer for me. 

 

I'll take your word that an oscilloscope is the easiest way to do a null test, assuming you have one lying around!  Not being so blessed I've had to limit myself to fiddling with the Audio Diffmaker.  Even there I'm only using the provided dyf files as I'd have no idea how to record my own.  Still, just that little introduction has been a revelation for me.  Soft sounds are hard to hear when loud music is playing!  People often quote the decibel scale as if the ear can easily hear the full range of it at once, from 0 db to 130 db.  I wonder if there is actually a much narrow range of loudness that the ear can hear at the same time.  Sure, you can hear leaves rustling on a quiet day, but I'll bet you can't while you're at a rock concert.  I'll also bet that the ability to hear noise or distortion drops off pretty quickly below -60 db or so.   

-60dB is WAY down! Not to take it entirely off topic, but oscilloscopes for audio are cheap used. The required bandwidth is very low and digital display is unnecessary. A scope off eBay is frequent at $100 shipped and a couple of probes sourced from the bay are maybe $40. A B&K 2120 is fully featured for this work. How to use it? Well, that's what the internet is for. Here are a couple available now, as current examples.

http://www.ebay.com/itm/B-K-BK-Precision-2120-20MHz-Oscilloscope-tested-working-/200786205635?pt=BI_Oscilloscopes&hash=item2ebfca5bc3

http://www.ebay.com/itm/B-K-Precision-2120-20MHz-Dual-Trace-Oscilloscope-Scope-/110913712540?pt=BI_Oscilloscopes&hash=item19d2f8f19c

This level of rig will handle most electrical signals. Getting into displaying live acoustic sound is pricier. A calibrated microphone and a power source for it are necessary - you are after all trying here to do better than PC based tools. I imagine a signal generator from PC software would be fine. Getting this far into it and a bit of practice will end a lot of arguments that tend to add post count to audio forums. Admit it, it would be great to not have to trust others who may be talking off the top of their head or recycling third party information which is not about your specific gear anyway. (The one thing that is very difficult to do cheaply is headphone measurement. I trust Tyll at Innerfidelity to do that for us.) IMO the scope and probes will get you there for seeing what it going on with electronics.

I would probably take more heat for recommending Ethan Winer's book to give this all context in the real world of audio science and practice. The Audio Expert is about $60 shipped from his website with a lot of nice extras included. The scope of the book is unique and I highly recommend it. It is very easy to read and understand.

Quote:

Originally Posted by Clarkmc2 

-60dB is WAY down! Not to take it entirely off topic, but oscilloscopes for audio are cheap used. The required bandwidth is very low and digital display is unnecessary. A scope off eBay is frequent at $100 shipped and a couple of probes sourced from the bay are maybe $40. A B&K 2120 is fully featured for this work. How to use it? Well, that's what the internet is for. Here are a couple available now, as current examples.

 

http://www.ebay.com/itm/B-K-BK-Precision-2120-20MHz-Oscilloscope-tested-working-/200786205635?pt=BI_Oscilloscopes&hash=item2ebfca5bc3

http://www.ebay.com/itm/B-K-Precision-2120-20MHz-Dual-Trace-Oscilloscope-Scope-/110913712540?pt=BI_Oscilloscopes&hash=item19d2f8f19c

 

This level of rig will handle most electrical signals. Getting into displaying live acoustic sound is pricier. A calibrated microphone and a power source for it are necessary - you are after all trying here to do better than PC based tools. I imagine a signal generator from PC software would be fine. Getting this far into it and a bit of practice will end a lot of arguments that tend to add post count to audio forums. Admit it, it would be great to not have to trust others who may be talking off the top of their head or recycling third party information which is not about your specific gear anyway. (The one thing that is very difficult to do cheaply is headphone measurement. I trust Tyll at Innerfidelity to do that for us.) IMO the scope and probes will get you there for seeing what it going on with electronics.

 

 

I would probably take more heat for recommending Ethan Winer's book to give this all context in the real world of audio science and practice. The Audio Expert is about $60 shipped from his website with a lot of nice extras included. The scope of the book is unique and I highly recommend it. It is very easy to read and understand.

Wow, I thought oscilloscopes were a lot more expensive than that. What won't a 20MHz scope be able to handle? Obviously that's more than enough for audio applications, but what couldn't I use it to measure?

Quote:

Originally Posted by OJNeg 

 

Wow, I thought oscilloscopes were a lot more expensive than that. What won't a 20MHz scope be able to handle? Obviously that's more than enough for audio applications, but what couldn't I use it to measure?

I'm not sure because I only use it for audio. Current scopes tend to be in the gigahertz range. I suppose very fast ICs might have performance parameters in that range. Radio, obviously. In audio high frequencies are encountered in unwanted oscillation of circuits, but I imagine a 20mHz instrument will find it. I have read that it is usually still in the kilohertz range.

Interestingly, most tubes we use for audio are good to up in the megahertz range but are described as low frequency amplifier tubes. That is because they can't handle the gigahertz radio applications. So 20 megahertz is considered low frequency.

Today's op amps can be pretty fast, but the manufactures' data sheets have the specs presented clearly and thoroughly.

Quote:

Originally Posted by Clarkmc2 

 

I'm not sure because I only use it for audio. Current scopes tend to be in the gigahertz range. I suppose very fast ICs might have performance parameters in that range. Radio, obviously. In audio high frequencies are encountered in unwanted oscillation of circuits, but I imagine a 20mHz instrument will find it. I have read that it is usually still in the kilohertz range.

Here someone found the AD797 used instead of a "slower" op amp oscillating at 43 MHz. It is safest to use parts that have no gain above the bandwidth of the oscilloscope.