Timing test results - a poll

[Pdxwayne]

At lower delays, I was going by tonality and ambience created, not distinct clicks.

No, it was more tonality than timing, like Amir said. Almost like the difference between Oh and Ah sounds.

Interesting....I see. A few people mentioned that too. For me, I didn't really sense significant change of tonality with the combos I used.

I wonder if you will hear something different if you use a DAC and amp combo with headphones.

Thanks!

 

[pkane]

Interesting....I see. Some others people mentioned that too. For me, I didn't really sense significant chance of tonality with the combos I used.

I wonder if you will hear something different if you use a DAC and amp combo with headphones.

Thanks!

The three-way test is harder. Here's my attempt at 5ms:

 

[Pdxwayne]

The three-way test is harder. Here's my attempt at 5ms:
View attachment 185112

Still using iMac internal speakers? Thanks!

 

[pkane]

Still using iMac internal speakers? Thanks!

yep

 

[pkane]

Still using iMac internal speakers? Thanks!

Here's 2-way 1ms test with HE6se v2 headphones and ADI-2 Pro @96k:

The tonality/sound pattern wasn't the same as with iMac, but still distinctive enough.

 

[Pdxwayne]

yep

I just tried using my phone's internal speakers just because you tried internal speakers.

: )

Indeed, now I sense tonality difference using my phone's internal speakers.
Interesting.......

 

[Pdxwayne]

Here's 2-way 1ms test with HE6se v2 headphones and ADI-2 Pro @96k:

View attachment 185117
The tonality/sound pattern wasn't the same as with iMac, but still distinctive enough.

Cool. It would seem my Gustard DAC/Amp is the one that is different than the rest. I can't sense any tonality difference for 1ms.

 

[Thomas_A]

I also went for tonality and ambience. The original is darker, dryer and ”echo-free”.

 

[Thomas_A]

Perhaps distortion differences can reveal the signals more in cheaper devices.

 

[Pdxwayne]

Perhaps distortion differences can reveal the signals more in cheaper devices.

The dac and amp I got are supposed to be transparent, which means I should hear no significant differences between them. But not the case when doing the tests, especially for 5ms and 2ms tests.

For 5ms and 2ms tests, the "double click" kind of clues for delay files are obvious for the following:
Windows 10 laptop ->Chrome to the following combos:
KTB->Topping L30
KTB->sabaj a10h
Topping E30->Topping L30
Topping E30->sabaj a10h
Because of the clues, I could easily pass the 5ms and 2ms tests.

As long as I have either Gustard x16 and/or Gustard h16 in the chain, that clues disappeared for 5ms and 2ms. All I could hear are two tones and no more double clicking clues. No tonality change either. So the following are difficult to get good results for 5ms and 2ms:
Windows 10 laptop->Chrome to the following combos:
KTB->Gustard H16
Gustard X16->Gustard H16
Topping E30->Gustard H16
Gustard X16->Topping L30
Gustard X16->sabaj a10h

I wonder if distortions can really explain the differences as all of them are supposed to be transparent and are recommended.....

 

[Jimbob54]

The dac and amp I got are supposed to be transparent, which means I should hear no significant differences between them. But not the case when doing the tests, especially for 5ms and 2ms tests.

For 5ms and 2ms tests, the "double click" kind of clues for delay files are obvious for the following:
Windows 10 laptop ->Chrome to the following combos:
KTB->Topping L30
KTB->sabaj a10h
Topping E30->Topping L30
Topping E30->sabaj a10h
Because of the clues, I could easily pass the 5ms and 2ms tests.

As long as I have either Gustard x16 and/or Gustard h16 in the chain, that clues disappeared for 5ms and 2ms. All I could hear are two tones and no more double clicking clues. No tonality change either. So the following are difficult to get good results for 5ms and 2ms:
Windows 10 laptop->Chrome to the following combos:
KTB->Gustard H16
Gustard X16->Gustard H16
Topping E30->Gustard H16
Gustard X16->Topping L30
Gustard X16->sabaj a10h

I wonder if distortions can really explain the differences as all of them are supposed to be transparent and are recommended.....

check the windows audio settings are the same for all DACs, especially the Gustard.

 

[Pdxwayne]

check the windows audio settings are the same for all DACs, especially the Gustard.

Thanks for the suggestion. I wonder about DAC settings difference too. I will see if Gustard is different.

There is another thing that is strange. As you can see, when using the Gustard amp (h16), I also got similar issue with all 3 DAC. So I wonder what is the difference in the amp as compared to L30 and a10h.....

 

[Pdxwayne]

I am glad I started this thread as I found more interesting differences...

In addition to DAC+amp combos making a difference, I have found out that headphones (in my case K371 with dynamic drivers vs HE400SE with planar-magnetic) also make a difference in term of clues.

I excluded KTB and a10h from my latest round of testing (to save time). For this round of comparisons, I used 2 DAC and 2 amps with two different headphones.

As suggested by @Jimbob54, I started by making sure both DAC are using 16bit 44100 Hz (CD quality) as default output format. I think (don't really remember clearly now as I forgot to write those down) previously my dac were all set to higher bit rate.

My findings:

******K371 headphones***********
*Topping E30 to Topping L30:
5ms - strange clue. Sync file has a strong single click and delay file has light click. Can't sense tonality change. Can't sense double clicks. Can pass using the single strong click clue.
2ms - can't sense clicks clues. Can't sense tonality change. Can't pass.
1ms - Sync file has a strong single click and delay file has light click. Can't sense tonality change. Can't sense double clicks. Can pass using single strong click clue.

*Topping E30 to Gustard h16:
5ms - able to sense double clicks clues and can pass.
2ms - not able to sense click clues. Can't really sense slight tonality change. Can pass about 70%.
1ms - Can't sense clicks clues nor sense tonality change. Can't pass.

*Gustard x16 to Topping L30 -
5ms - strange clue. Sync file has a strong single click and delay file has light click. Can't sense tonality change nor double clicks. Can pass using the single strong click clue.
2ms - slightly stronger single click for sync file. Not as easy. Can do about 70% correct.
1ms - strange clue. Sync file has a strong singleclick and delay file has light click. Can't sense tonality change nor double clicks. Can pass using single strong click clue.

*Gustard x16 to Gustard H16 -
5ms - no obvious clicks clues. A little tonality clue. Delay seems less clean. Can pass.
2ms - no obvious clicks clues. A little tonality clue. Delay seems less clean. Can pass.
1ms - no obvious clues. Get about 60%.


******HE400SE headphones*******
*Topping E30 to Topping L30 - able to sense double clicks clues. Can easily get 10/10 for 5ms, 2ms, and 1ms.

*Topping E30 to Gustard h16:
5ms - Can sense double clicks clues. Can pass.
2ms - not able to sense clicks clues.
Can't really sense tonality change. Can't easily pass.
1ms - strange clue. Sync file has a strong singleclick and delay file has light click. Can't sense tonality change nor double clicks. Can pass using single strong click clue.

Gustard x16 to Topping L30 -
5ms - able to sense double clicks clues. Can easily pass.
2ms - able to sense double clicks clues. Can easily pass.
1ms - able to sense double clicks clues. Not as easy, but can pass.

*Gustard x16 to Gustard H16 -
5ms - no obvious clicks clues. A little tonality clue. Can pass.
2ms - can't sense clicks clues. Sensed a little tonality difference. Can pass less than 70%
1ms - strange clue. Sync file has a strong single click and delay file has light click. Can pass using single strong click clue.


It would seems he400se is the better headphones to get more passing results than k371. HE400se passed 10 of 12 rounds. K371 passed 7 of 12 rounds.

 

[voodooless]

Got down to 5ms with iPhone and Bose QC 35 II over Bluetooth on low volume. Didn’t try lower because the sample buttons are not working properly on the phone browser, so can’t really hear the reference. Probably won’t hear the difference anyway.

 

[Pdxwayne]

Got down to 5ms with iPhone and Bose QC 35 II over Bluetooth on low volume. Didn’t try lower because the sample buttons are not working properly on the phone browser, so can’t really hear the reference. Probably won’t hear the difference anyway.

Thanks for testing. Good to see all sorts of combinations and results.

 

[Pdxwayne]

check the windows audio settings are the same for all DACs, especially the Gustard.

I want to thank you for the suggestion!

Yesterday I was comparing ktb with gustard. Because ktb has no 16bit option in Windows advanced settings, I adjusted Gustard to use 24bit to match KTB setting. I them realized that at 24bit 44100, Gustard behave differently. At 24bit, Gustard hi hat was very soft sounding, but ktb remains strong and sharp.

I did some quick listening tests and indeed at 24bit 44100hz, gustard would not give double click clues when paired with Topping L30, using HE400SE. That matches my observations in post #50.

As you notice in post #53, two days ago, because of your suggestion, I used 16bit 44100 for both Gustard and E30 for more tests. I checked Gustard at 16bit again yesterday and indeed high hat became sharper and stronger. I could again hear double clicks clues when paired with L30. This matches my results for post #53.

So, another variable to consider when doing online blind listening tests. Not only there could be differences between DAC, amp, and headphones, there maybe differences between different Windows audio settings too!

 

[Jimbob54]

I want to thank you for the suggestion!

Yesterday I was comparing ktb with gustard. Because ktb has no 16bit option in Windows advanced settings, I adjusted Gustard to use 24bit to match KTB setting. I them realized that at 24bit 44100, Gustard behave differently. At 24bit, Gustard hi hat was very soft sounding, but ktb remains strong and sharp.

I did some quick listening tests and indeed at 24bit 44100hz, gustard would not give double click clues when paired with Topping L30, using HE400SE. That matches my observations in post #50.

As you notice in post #53, two days ago, because of your suggestion, I used 16bit 44100 for both Gustard and E30 for more tests. I checked Gustard at 16bit again yesterday and indeed high hat became sharper and stronger. I could again hear double clicks clues when paired with L30. This matches my results for post #53.

So, another variable to consider when doing online blind listening tests. Not only there could be differences between DAC, amp, and headphones, there maybe differences between different Windows audio settings too!

No problem. Controlled tests mean controlling for all possible variables. We often focus on level matching and unsighted listening but there are so many more in a digital music chain. One of the reasons they are so hard to do well I am sure.

 

[Cbdb2]

9/10 at 1ms with a surface pc and headphones. Slight change in timbre.

 

[Grooved]

The simple 2 ways timing online blind test is here:

Blind testing a 100 ms Timing Difference (2-way)

Check your listening skills, online.

www.audiocheck.net

I tried a while back using smartphone and could do 10/10 for 10ms. Lower was tougher for me. The site does not show the statistics of all those who tried. So I am curious how ASR members would do.

Just saw your thread now

Very interesting, I will do more tests (and with AKG K371 as you tested this one) as I only had a few minutes available but it seems to match what I found when ABX testing ACC and FLAC : it's easier on speakers for me than on headphones, even on cheap amp/speakers

I started with Tone2 Pro and the Sony MDR-7509 that was the nearest headphones from me : I needed to check sometimes with both samples above, and just missed one at 2ms the only moment with lots of noise coming from the street

Then, just turned on an old Yamaha AVR linked via TOSLINK to the same computer and small Hitachi speakers : I even didn't need to compare with the two samples at the top of the test, just clicking directly between Sync or Delay after the question mark

 

[Pdxwayne]

Just saw your thread now

Very interesting, I will do more tests (and with AKG K371 as you tested this one) as I only had a few minutes available but it seems to match what I found when ABX testing ACC and FLAC : it's easier on speakers for me than on headphones, even on cheap amp/speakers

I started with Tone2 Pro and the Sony MDR-7509 that was the nearest headphones from me : I needed to check sometimes with both samples above, and just missed one at 2ms the only moment with lots of noise coming from the street
View attachment 188545

Then, just turned on an old Yamaha AVR linked via TOSLINK to the same computer and small Hitachi speakers : I even didn't need to compare with the two samples at the top of the test, just clicking directly between Sync or Delay after the question mark

View attachment 188541
View attachment 188542

Thanks for checking!

What were your clues? Extra hit? Tonality change? Something else?