CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS

  • December 31, 2015, 11:09:11 AM
  • Welcome, Guest
Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 2 [3] 4

Author Topic: Take a listen #2  (Read 5264 times)

0 Members and 1 Guest are viewing this topic.

gurubhai

  • Ortho Ninja
  • Mate
  • Pirate
  • ****
  • Brownie Points: +104/-0
  • Offline Offline
  • Posts: 317
Re: Take a listen #2
« Reply #20 on: June 27, 2013, 08:18:32 AM »

Preferred the sample A.
It sounds cleaner, more involving and with more accurate timbre esp. during quieter passages & transitions.

Found it difficult to ABX though and had to teach myself the differences before I could ABX with any confidence.
Here are the results :


I found it quite fatiguing. Got the first 3 right, then the wrong one. Concentrated again and got the last two right and was too tired to continue any further.
« Last Edit: June 27, 2013, 08:41:12 AM by gurubhai »
Logged

Marvey

  • The Man For His Time And Place
  • Master
  • Pirate
  • *****
  • Brownie Points: +555/-33
  • Offline Offline
  • Posts: 6698
  • Captain Plankton and MOT: Eddie Current
Re: Take a listen #2
« Reply #21 on: June 27, 2013, 11:57:37 PM »

Using my StratroGrados from the Creative speaker headphone outputs (being fed from my motherboard line out), both samples A and B sound like Mozart's Violin Concerto #4, Allegro / 1st movement - the end of the recap toward the coda where the soloist gets to show off his/her stuff. I'm pretty sure it's #4, if not, it's #3. A lot of Mozart stuff tends to sound very similar to each other. It's been a while.
Logged

donunus

  • Cheapus Sexus
  • Mate
  • Pirate
  • ****
  • Brownie Points: +52/-3
  • Offline Offline
  • Posts: 875
Re: Take a listen #2
« Reply #22 on: June 28, 2013, 02:25:20 AM »

hahahaha
Logged
Team Delicious and Juicy Sound

Sholay

  • Able Bodied Sailor
  • Powder Monkey
  • ***
  • Brownie Points: +9/-1
  • Offline Offline
  • Posts: 43
Re: Take a listen #2
« Reply #23 on: June 29, 2013, 11:35:44 AM »

A = airy, dynamic, some echo                         
B = heavy, no echo at all, not as tall sounding as A, not much air......dense like a jelly
« Last Edit: July 01, 2013, 07:20:55 PM by Sholay »
Logged

AstralStorm

  • Speculation and Speculums
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +250/-164
  • Offline Offline
  • Posts: 559
  • Warning: causes nearby electronics to go haywire
Re: Take a listen #2
« Reply #24 on: July 01, 2013, 09:48:10 AM »

In case you want a good tool, there's one here:
http://www.rarewares.org/others.php

It's actually for ABC/HR, a superset of ABX - it has ABX testing as part of it. Simplifies rating too and provides for easy "warm up".
For correct rating, you need to decide on the total number of trials before starting the test, otherwise the Bayes' rule will underestimate the probability of type I error.
Standard approach would consider a test failed with P>0.05 or 5%.

I should try these samples as well.
« Last Edit: July 01, 2013, 09:54:03 AM by AstralStorm »
Logged
For sale: Hifiman HE-500; Paradox; Brainwavz B2. PM me if you would like to buy them.

ultrabike

  • Burritous Supremus (and Mexican Ewok)
  • Master
  • Pirate
  • *****
  • Brownie Points: +4226/-2
  • Offline Offline
  • Posts: 2384
  • I consider myself "normal"
Re: Take a listen #2
« Reply #25 on: July 01, 2013, 04:59:09 PM »

So... What is/are the difference(s) between the files xnor?  popcorn
Logged

donunus

  • Cheapus Sexus
  • Mate
  • Pirate
  • ****
  • Brownie Points: +52/-3
  • Offline Offline
  • Posts: 875
Re: Take a listen #2
« Reply #26 on: July 05, 2013, 01:08:33 AM »

Both use ASIO yet Jriver is consistently a little bassier and less forward in the upper mids. The difference is there whether I use wasapi, asio, or directsound as if theres a built in non defeatable eq in these players that give them their own flavor. I feel it may be somewhere in another stage of processing and not from the wasapi/asio/directsound stage. I don't know but I feel that these differences between media player's are bigger than the difference in your files.

EDIT: About the media players, It could be placebo due to the different color schemes between them hahaha. Can anyone else compare them to see if I am alone with this one?

I set F2K and JR MC up the same way (WASAPI, 24-bit/96KHz, etc. etc.). I MIGHT have had upsampling turned on in JR MC...or maybe both...or neither. I can't remember. That could definitely throw results off. Or maybe it doesn't make a difference with bit perfect? Either way, I thought I heard a difference similar to what you're describing. For me, given that I am no expert, the difference was small enough that it could have been placebo. I found myself gravitating towards JR MC in the end anyway, mostly due to the UI.

I do notice that the difference between JRMC and foobar become less apparent when using ASIO4All vs when using wasapi. Its as if wasapi still lets more of the players differences through. Foobar is harsher than JRMC with wasapi event style on both players. With ASIO they sound more similar. This is under Windows 7 by the way. The differences that I hear with players are larger to me than any of xnors files from the first and second "take a listen" threads. I also included foobar 0.8.3 in the mix just for the heck of it using its stock waveout output option and it sounded different than the other two players then again. The older foobar sounded more laid back and dynamic with slightly thinner mids than the two newer players both using wasapi but had the smoothness of Jriver.

Anyway back to topic... What was the difference between the two files? was one upsampled and downsampled back to 16/44?
Logged
Team Delicious and Juicy Sound

xnor

  • Pirate
  • **
  • Brownie Points: +39/-50
  • Offline Offline
  • Posts: 154
Re: Take a listen #2
« Reply #27 on: July 20, 2013, 06:56:41 PM »

Sorry for the delay.

*SPOILER*

.
.
.
.
.
.
.

Random dictated that A is the original and B the modified file.

What was done to B?
The bit depth was reduced by about 4.6 bits on average which resulted in about 11.4 bits on average. More bits are used during quiet passages, less during loud ones.

For the cut samples the "real" FLAC bitrates are 722 kbps for A and 356 kbps for B. The entire track had "real" bitrates of 686 and 319 kbps respectively. Yes, the bit depth is so low on average that FLAC can compress the whole track better than MP3 320 CBR.  :)p17

"Real" because I had to add very low level noise to B to avoid FLAC compressing the file better than A, which would've biased comparisons.

In case you don't believe me, download b_real.flac and compare it to B. The only difference is the random noise I had to add at a very low level (about -130 dB in the audition spectrum analyzer with fft size of 65k).

b_real.flac

Also notice the file sizes:
3.02 MB for A and 1.49 MB for the "real" B

Oh, the difference in the highest octave is due to (maybe too) aggressive noise shaping.
Logged
"I'm on a whole new adventure." - "Growing a mustache?"
"No. Bigger than that." - "A beard?!?"

ultrabike

  • Burritous Supremus (and Mexican Ewok)
  • Master
  • Pirate
  • *****
  • Brownie Points: +4226/-2
  • Offline Offline
  • Posts: 2384
  • I consider myself "normal"
Re: Take a listen #2
« Reply #28 on: July 20, 2013, 07:06:19 PM »

Well, it was a close call for me.

I liked B better but it was definitively less dynamic. I was using my KSC75 and my shameless laptop so maybe A demanded better gear to pull through...
Logged

xnor

  • Pirate
  • **
  • Brownie Points: +39/-50
  • Offline Offline
  • Posts: 154
Re: Take a listen #2
« Reply #29 on: July 20, 2013, 07:31:06 PM »

The dynamics you can perceive don't change until you can "hear" low level details being drowned in quantization noise. IIRC nobody complained about the noise floor.

Anyway, my point was to show that 16 bits all the way can be "overkill" even with dynamic classical music. Bit depth can be reduced much further with modern compressed pop or metal to like 8 or 7 bits - still being almost indistinguishable from the full 16 bits.

24 bits ... yep, totally worth it, NOT.
Logged
"I'm on a whole new adventure." - "Growing a mustache?"
"No. Bigger than that." - "A beard?!?"
Pages: 1 2 [3] 4