• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Listen and choose the 8th generation digital copy.

Blumlein 88

Major Contributor
Joined
Feb 23, 2016
Messages
5,131
Likes
3,800
#61
Congrats! :)

Btw, is there any way you can make similar files that differ only in THD but not in frequency response? In ideal case the same recording would have different levels of THD, say 0.05%, 0.1%, 0.5%, 1% etc..
Audacity has a plug in that lets you pick I think 9 harmonics and set the level for each one. It will have that distortion at all levels which isn't really how electronic gear distorts. I've messed with it in the past, but don't have it loaded into Audacity currently. You could do that and see what you think. I think it was part of the LADSPA plug in group though I may have that wrong.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,476
Likes
1,134
Location
Zg, Cro
#62
It would be fair to say that we were actually listening to the ADC in this case as this seems to be the bottleneck in this case and not the DAC.

It makes me wonder how many generations it would take before differences would be audible when say an RME would be used to record the DAC's output ?
IMHO point is that processed file sounded DIFFERENT, not necessarilly WORSE, as it would if only distortion was introduced.

One other thing - I can listen 2 recordings and decide which sounds better to me but it is much harder to test which one is identical to the original.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
2,476
Likes
1,134
Location
Zg, Cro
#63
Audacity has a plug in that lets you pick I think 9 harmonics and set the level for each one. It will have that distortion at all levels which isn't really how electronic gear distorts. I've messed with it in the past, but don't have it loaded into Audacity currently. You could do that and see what you think. I think it was part of the LADSPA plug in group though I may have that wrong.
Excellent idea!

Yes, I could do it myself, but I would rather it is done by more folks so some kind of consensus is reached about distortion threshold that can be heard. :)
 

Blumlein 88

Major Contributor
Joined
Feb 23, 2016
Messages
5,131
Likes
3,800
#64
It would be fair to say that we were actually listening to the ADC in this case as this seems to be the bottleneck in this case and not the DAC.

It makes me wonder how many generations it would take before differences would be audible when say an RME would be used to record the DAC's output ?
In this case yes, its the FR ripple of the ADC. The noise levels are comparable in the two bits of gear with an edge to the DAC. The DAC is lower in 2nd and 3rd harmonic distortion as well. I did this once before with other gear, and thought it was harder to tell. I got lucky in that the DAC had a tiny uptilt the last 5 khz and the ADC had a tiny droop so that even after 8 generations FR was ruler flat. They dovetailed almost perfectly.

I had considered both before and now going for more generations. It appears to me what will become obvious going further is noise floor will intrude first if you have flat FR. I think that would be the case with the RME as well. I think it would be a bother for a non-interesting result. But nothing about it is hard to do if someone has the RME and wants to give it a whirl.
 

gorg

Active Member
Joined
Sep 13, 2018
Messages
121
Likes
55
#66
Bob Marley B
Ry Cooder A
Cowboy Junkies A
Phil Woods A
Jennifer Warnes A

I was rather guessing ;)
 

stunta

Addicted to Fun and Learning
Joined
Jan 1, 2018
Messages
602
Likes
364
Location
Boston, MA
#71
Closest to reference:

Jennifer Warnes: A
Bob Marley: A
Ry Cooder: A
Phil Woods: B
Cowboy Junkies: No diff

PC => Audioengine D3 => NAD HP50

I hope I am wrong on all counts :)
 

stunta

Addicted to Fun and Learning
Joined
Jan 1, 2018
Messages
602
Likes
364
Location
Boston, MA
#72
Thank you @Blumlein 88 for doing this. I got 2/5 correct so I am quite happy that the DA and AD conversions (at least with the gear you used) are a non-issue for me. I wonder if the converters in active DSP speakers (like the JBL 3xx) are of similar quality and if they are, given the low price of these speakers, it would mean AD-DA chips are pretty much a commodity at this point. Very good news for hi-fi IMO.
 

Veri

Major Contributor
Joined
Feb 6, 2018
Messages
1,376
Likes
1,187
#73
I wonder if there's a reason or simply coincidence a lot of people seem to hear a difference in the Bob Marley fragment :)
 

pkane

Senior Member
Joined
Aug 18, 2017
Messages
368
Likes
273
#74
I wonder if there's a reason or simply coincidence a lot of people seem to hear a difference in the Bob Marley fragment :)
I must be the exception... couldn't tell the difference at all and got it wrong by guessing. But got all the others right :)
 

LTig

Member
Joined
Feb 27, 2019
Messages
59
Likes
46
Location
Europe
#75
I'm afraid I need a 16th generation....
Me too. Linux-PC, vlc-player, Edirol UA-25, Sennheiser HD-800, aging ears. OK, I did not invest much time, but it all just sounds very good. This is really fantastic performance, especially when I remember my horrible Philips tape deck (N4420, late 1970s), or the somewhat better Aiwa cassette deck which replaced it (1980s), or the much better Akai GX75mk2 (1990s) which is still in working condition. Event the first generation copy on any of these could easily be indentified.

BTW, the link to Phil Wood is dead.
 

Blumlein 88

Major Contributor
Joined
Feb 23, 2016
Messages
5,131
Likes
3,800
#78
Me too. Linux-PC, vlc-player, Edirol UA-25, Sennheiser HD-800, aging ears. OK, I did not invest much time, but it all just sounds very good. This is really fantastic performance, especially when I remember my horrible Philips tape deck (N4420, late 1970s), or the somewhat better Aiwa cassette deck which replaced it (1980s), or the much better Akai GX75mk2 (1990s) which is still in working condition. Event the first generation copy on any of these could easily be indentified.

BTW, the link to Phil Wood is dead.
Okay try it now.

https://www.dropbox.com/s/b899j2ljmonlxaf/Phil Woods repost.zip?dl=0


Also I put the answer in the first post under a spoiler tag. So once you've listened and made your choices look at the spoiler.

Guys over at CA were mostly scoring 50:50. With all the multiplied distortion, jitter upon jitter, and running through non-descript cables somehow the unaltered bit perfect original didn't stand out all that much to them. Nothing, but SMPS in use too. That alone should have provided the veil dropping night and day difference even before passing thru the $2 USB cables. Heck the DAC was USB powered. All sitting on a table between a pair of laptops emitting noxious horrible RFI/EMI contaminating everything for probably 10's of meters. I was running Windows 10 with no help to lower its perturbations, and then a MacBook sat on the other side.

BTW, having become the owner of a mid-level cassette dubbing deck, I may do the 8th generation cassette tape test. Stay tuned. I expect the smooth musical analog qualities to save the day. :p
 
Last edited:
Joined
Dec 1, 2018
Messages
15
Likes
5
#79
For a test that determines if people are able to discern differences it is perhaps better to give three fragments. Two identical and 1 different. People have to pick the different fragment.
This is a very reliable method to test if you are able to pick up differences, it is used in food evaluation for instance.
However, comparing three audio fragments might be difficult.
Has this type of test setup been used in audio?
 

SIY

Major Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
1,510
Likes
2,174
Location
Phoenix, AZ
#80
Yes, people have used triangle testing. And of course, the objections were made from the usual suspects that this was even more confusing to the listeners than ABX. If one is determined to peek, there is no DBT format that will be satisfying.
 
Top Bottom