Some explanation from ESS about digital vs analog volume.
And an explanation from John Yang why the L30 II measured worse at low gain than mid gain in Amir measurement.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Topping L30 II Review (Headphone Amp)
- Thread starter amirm
- Start date
And an explanation from John Yang why the L30 II measured worse at low gain than mid gain in Amir measurement.
Bow_Wazoo
Addicted to Fun and Learning
- Joined
- Jan 3, 2021
- Messages
- 784
- Likes
- 684
That's really like going down a rabbit hole.
I found something else from Violectric on the subject.
The experts themselves are probably not in agreement...
"The CD format offers 16 bits which represents a dynamic range of 96 dB, the distortion cannot be smaller than 0.0016 %. A 24 bit signal offers a dynamic range of 144 dB with a theoretical minimum distortion of 0.00001 % - not possible in real life ! For a 32 bit signal the theoretical maximum dynamic range is even calculated to 192 dB.
The best A/D converters available today offer a dynamic range of 120 dB with a minimum THD of -110 dB. The limit of what is physically possible is not far here. There are heaps of losses during recording, mixing, editing ... so that a CD with 16 bit is perfectly sufficient as playback medium.
Much more is not sensibly representable.
The digital volume reduction is achieved by inserting zeros at the top and shifting the rest of the data word downwards. A volume reduction of 6 dB is a complete zero in the top bit.
A CD signal with 16 bits is not only supplemented in Violectric or Lake People converters to 24 bits or 32 bits. This means that 8 bits x 6 dB = 48 dB or 16 x 6 dB = 96 dB volume reduction is available without affecting the original signal at all.
This is a factor of 1:200 (1:65,000) - what do you think you would hear if you reduced the volume in your living room to 5 per thousand?
From the practical limitations of recording described above, we learn that there is a maximum of 20 "meaningful" bits, the rest is literally noise ! Even with such a signal, the volume can still be reduced by 4 bits x 6 dB = 24 dB = factor 1:35 with a 24 bit converter, and by 12 bits x 6 dB = 72 dB = factor 1:4.000 with a 32 bit converter - without touching anything relevant in the actually active bits.
This makes digital volume control the best thing that can happen to a signal - except not being controlled."
I found something else from Violectric on the subject.
The experts themselves are probably not in agreement...
"The CD format offers 16 bits which represents a dynamic range of 96 dB, the distortion cannot be smaller than 0.0016 %. A 24 bit signal offers a dynamic range of 144 dB with a theoretical minimum distortion of 0.00001 % - not possible in real life ! For a 32 bit signal the theoretical maximum dynamic range is even calculated to 192 dB.
The best A/D converters available today offer a dynamic range of 120 dB with a minimum THD of -110 dB. The limit of what is physically possible is not far here. There are heaps of losses during recording, mixing, editing ... so that a CD with 16 bit is perfectly sufficient as playback medium.
Much more is not sensibly representable.
The digital volume reduction is achieved by inserting zeros at the top and shifting the rest of the data word downwards. A volume reduction of 6 dB is a complete zero in the top bit.
A CD signal with 16 bits is not only supplemented in Violectric or Lake People converters to 24 bits or 32 bits. This means that 8 bits x 6 dB = 48 dB or 16 x 6 dB = 96 dB volume reduction is available without affecting the original signal at all.
This is a factor of 1:200 (1:65,000) - what do you think you would hear if you reduced the volume in your living room to 5 per thousand?
From the practical limitations of recording described above, we learn that there is a maximum of 20 "meaningful" bits, the rest is literally noise ! Even with such a signal, the volume can still be reduced by 4 bits x 6 dB = 24 dB = factor 1:35 with a 24 bit converter, and by 12 bits x 6 dB = 72 dB = factor 1:4.000 with a 32 bit converter - without touching anything relevant in the actually active bits.
This makes digital volume control the best thing that can happen to a signal - except not being controlled."
The post in this link compares DR and THD+N for E50 volume and A30Pro volume.
Topping E50 Review (Balanced DAC)
This is a good idea! Can someone share firmware v.1.06, because it's no longer available for download. Attached.
www.audiosciencereview.com
The post in this link compares SINAD for DAC digital volume (without amplifier) and DAC + analog volume.
Cheap but non-degrading solution to add volume control between dac and amp.
I've got a Yamaha AVR as a 2.1 amp, set the AVRs volume level to "0db", use a Topping L30 as volume control, works great.
www.audiosciencereview.com
danadam
Addicted to Fun and Learning
- Joined
- Jan 20, 2017
- Messages
- 999
- Likes
- 1,563
The uploaded file was probably at 0 dBFS and opus compression moved it past that. What did you expect youtube would do in such case?
staticV3
Master Contributor
- Joined
- Aug 29, 2019
- Messages
- 8,030
- Likes
- 12,888
Reduce the signal before compression to compensate, unless opus compression results in unpredictable attenuation or gain.
Last edited:
If you do a direct A/B comparison (perfectly balanced signal vs left or right louder) it's not difficult to hear that difference. I can post some test files if anyone wants to try.
During regular music listening, I don't know if you'll notice something is off. Maybe, maybe not.
Bow_Wazoo
Addicted to Fun and Learning
- Joined
- Jan 3, 2021
- Messages
- 784
- Likes
- 684
Today I played around with the oscilloscope again, and fed the E30II with a 1 kHz tone.
My pot seems to work probably a little over three quarters turned up the cleanest.
Yellow ch.1.
Blue ch. 2.
Max. vol.:
Approx 4 a clock
My pot seems to work probably a little over three quarters turned up the cleanest.
Yellow ch.1.
Blue ch. 2.
Max. vol.:
Approx 4 a clock
Which gain? I just tested my setup with high gain and sweep test was not showing any imbalance issue. L/R was +-0.0. But I can easily fix that because I can bias tubes differently but I don't really want to do that.
staticV3
Master Contributor
- Joined
- Aug 29, 2019
- Messages
- 8,030
- Likes
- 12,888
If your pot is showing 0.0dB L/R delta from 0 to -80dB pot attenuation, then chances are your measurement setup is insufficient.
Pots aren't that good.
Edit: here's one of the best pots that Amir has measured to date:
Roland68
Major Contributor
The Topping L30 II has NO channel imbalance issues. Please consider how quickly such a statement spreads on the Internet and what damage it can mean for the manufacturer.
All regular stereo potentiometers (and quad for balanced) with resistive tracks have variations on the channels all the way through. This is normal and due to production and technical reasons. There used to be corrected and selected pots, but in the price range of a topping L30 II.
That means all devices on the market have this problem if they use a regular pot in the signal path for volume control, whether the device is $50, $500, $5000, or $10,000.
And it always depends on exactly this one measured potentiometer and can be completely different with the next one.
Alternatives are potentiometers with fixed resistors, relay based volume controls, chip based or digital volume controls.
Yup. Not possible to be that good. Not a problem thought on my setup.
i had combo-units (amp\dac) with digital and analog volume controls, i found last one more are convenient to use, never experienced channel imbalance in real life situations since all my headphones wasn't easy to drive, so i've always puched volume knob beyond 12o'clock, L30II is much more friendly amp for IEM's than most of other amp's in this price range (and even higher)
I'm at 9-10 o'clock. I'm happy that it have so much power. With the DX7 pro HP out I cranked potentiometer up to 10-11 o'clock. Sometimes even more because of the EQ compensation.
raif71
Major Contributor
- Joined
- Sep 7, 2019
- Messages
- 2,344
- Likes
- 2,562
9-10 o'clock @high gain? What headphone?
Yes. Nectar HiveX. They require quite much power. Tube amp is approx 56W per channel.
Similar threads
