Topping L30 Headphone Amplifier Review

[Mr:River]

Well that's good to hear about the Hifiman's, I really want to try magnetic planars.

On Amazon, the HE-400i is 169.00 and the Deva's are 219.00. Can you tell me where you found the Deva's for 169? From what I can tell, they are the same headphone technically, just design feature differences.

Sorry sorry
I changed the values.

 

[Mr:River]

@MechEngVic

Hifiman (Open box)

https://store.hifiman.com/index.php/open-box.html

 

[Deleted member 20267]

Do you have such an impedance graph with 1400Ohm peaks? I didn't see such one. I have seen this one from https://www.0db.co.kr/index.php?mid=REVIEW_0DB&category=182&document_srl=575546

Better to compute requirements from an average impedance than from peaks.

It was in a head-fi post, and they just mentioned 1400ohm peaks, nothing else. So you are probably right in that it was 1st gen.
And I disagree with the avg impedance thing. Of course it'll mostly run 600ohm, but if I have something that can't handle the voltage the low end impedance hump requires then I'd imagine my bass would end up under driven right? It's just to cover all bases pretty much, that I'm doing it

 

[Deleted member 20267]

An amplifier is a voltage source. The higher the impedance the less current is drawn.
This combo can reach 6V out which can bring it to 119dB peak levels.
These are impressively loud levels distortion free.
As said impedance doesn't matter at all.

Well, until your amp caps out voltage wise haha.
Figured it'd work since it gets well past 5v in real world testing and the t1v2 even at 1400ohm just needs 3.74v, just wanted to make sure

 

[solderdude]

As said, the impedance is completely irrelevant. Only the voltage is.
Regardless if it is 120, 300, 600 Ohm, 1400 Ohm or even 2kOhm (old version of HD414)

at 3.7V it will reach 115dB peaks (not average level)
The higher the impedance the higher the dB/mW efficiency of the headphone and the easier the load (less current at the same voltage)
Nothing is under driven.

 

[Deleted member 20267]

I see, thank you.

 

[odyo]

Sorry you missed the point. Either you will feed the DAC with the source material sample rate (usually 44.1k) and DAC will oversample up to MHz level of sample rates, or you will (at least partially) upsample in PC and leave less on the DAC to continue. In delta sigma DAC there is no way with PCM input to avoid increasing sample rate into MHz region because such sample rates are required for delta sigma modulator to reach the desired precision. That's the principle how delta sigma DAC is working. Either you do a part of that job in PC, where you can do that at better quality level, or you leave all on DAC. The process performed in DAC or alternatively in PC is in principle the same, the only difference can be different quality. The quality of the first up- or oversampling step (from the lowest sample rate) has the biggest impact on the result.
With DSD input and direct DSD mode it is different, because you can perform complete DSP in PC and DAC will not need to increase the sample rate anymore.

Ifi's burr brown Zen dac is interesting though. It's hybrid multibit and delta sigma.

 

[solderdude]

fi's burr brown Zen dac is interesting though. It's hybrid multibit and delta sigma.

All Delta Sigma DACs can be considered hybrids as they use mutibit and DS combined.

I assume you mean this DAC ?

 

[Veri]

It's hybrid multibit and delta sigma.

Pure marketing is what it is.

 

[bogi]

Ifi's burr brown Zen dac is interesting though. It's hybrid multibit and delta sigma.

That's only iFi marketing language how they describe the Burr-Brown DSD1973 TI chip functionality with PCM input. With 24bit PCM input, top 6 bits are processed as in multibit PCM DAC and all the remaining lower bits are processed as usual by oversampling and delta sigma modulation. The analog section then joins outputs of both parts.

That 'multibit' terminology has nothing to do with the direct DSD path which DSD1973 TI provides similarly like also CS chips and AKM chips. With direct DSD path no multibit PCM processing, no PCM oversampling and no delta sigma modulation happens, since DSD is already delta sigma modulated.

 

[JohnYang1997]

That's only iFi marketing language how they describe the Burr-Brown DSD1973 TI chip functionality with PCM input. With 24bit PCM input, top 6 bits are processed as in multibit PCM DAC and all the remaining lower bits are processed as usual by oversampling and delta sigma modulation. The analog section then joins outputs of both parts.

That 'multibit' terminology has nothing to do with the direct DSD path which DSD1973 TI provides similarly like also CS chips and AKM chips. With direct DSD path no multibit PCM processing, no PCM oversampling and no delta sigma modulation happens, since DSD is already delta sigma modulated.

Have a read of this.
https://www.audiosciencereview.com/...qui-dac-and-streamer-review.10770/post-300866

By converting to DSD and using DSD direct mode you are essentially forcing the DAC to work under a non-ideal mode. People like it because they think it's "pure" to not have any processing. It's just not the case. What really matter is to actually design the DAC to match the chip specification (usually better in PCM mode and often in 44.1khz/48khz) and try to improve upon the spec. It's not rocket science nor it's about what you believe. There IS the reality. We can MEASURE the output of the DAC.
Are there exceptions? Of course yes. Some DAC chips don't have good enough DSD processor. But the idea is the same. What you measure is what you get not what you think it is.

 

[bogi]

Have a read of this.
https://www.audiosciencereview.com/...qui-dac-and-streamer-review.10770/post-300866

OK, you are pointing to adaption to the internal data format DAC is using. Some DACs are really 1bit sigma delta in their nature, but mostly is the internal format multibit. So yes, my above description contains simplification to be enough easy and to point to a more basic principle.

Every DSP results to some level of distortion. In real world no filter has infinite time to process a sample. So what is to compare are benefits or drawbacks of 2 paths: Conventional oversampling + modulation in DAC, or upsampling + modulation in software + adaption of 1bit DSD signal to internal format of DAC.

People like it because they think it's "pure" to not have any processing.

I humbly disagree. It is the difference in sound quality what forces audiophiles to use HQPlayer. All other things are significantly less important. If there would be a cheap way to get the same quality easier and without software, they would do it. It is not primary intention of people to complicate things. And it is really not about a "sensational but mistaken information" what forces people to do high quality upsampling. Please don't underestimate technical background and listening experience of people. This is the place where you re simplifying things.

We can MEASURE the output of the DAC.

Yes. Take high quality test signal, setup HQPlayer for the highest PCM to DSD conversion quality, play the test signal in HQPlayer converting it to high rate DSD on the fly and do a wide band measurement like author of HQPlayer is used to do.

Like this measurement of RME ADI-2, which uses the same AKM4493 DAC chip as Topping E30 (and thus the same internal data format):
https://audiophilestyle.com/forums/...-rme-adi-2-dac/?do=findComment&comment=926151

 

[JohnYang1997]

OK, you are pointing to adaption to the internal data format DAC is using. Some DACs are really 1bit sigma delta in their nature, but mostly is the internal format multibit. So yes, my above description contains simplification to be enough easy and to point to a more basic principle.

Every DSP results to some level of distortion. In real world no filter has infinite time to process a sample. So what is to compare are benefits or drawbacks of 2 paths: Conventional oversampling + modulation in DAC, or upsampling + modulation in software + adaption of 1bit DSD signal to internal format of DAC.



I humbly disagree. It is the difference in sound quality what forces audiophiles to use HQPlayer. All other things are significantly less important. If there would be a cheap way to get the same quality easier and without software, they would do it. It is not primary intention of people to complicate things. And it is really not about a "sensational but mistaken information" what forces people to do high quality upsampling. Please don't underestimate technical background and listening experience of people. This is the place where you re simplifying things.



Yes. Take high quality test signal, setup HQPlayer for the highest PCM to DSD conversion quality, play the test signal in HQPlayer converting it to high rate DSD on the fly and do a wide band measurement like author of HQPlayer is used to do.

Like this measurement of RME ADI-2, which uses the same AKM4493 DAC chip as Topping E30 (and thus the same internal data format):
https://audiophilestyle.com/forums/...-rme-adi-2-dac/?do=findComment&comment=926151

RME adi2 dac/pro has an ASRC chip in the front and a DSP chip after that. This is very not representative for majority of DACs. Also are you talking about the out of band noise? That's how measurements mislead people....You should be take the in band measurements. THD IMD Multitone noise in 20KHzBW.

 

[JohnYang1997]

Very small amount of people can differentiate HIRES and CD quality in blind listening tests. But this is actual HIRES. If you upsample, the best you can get is from the source. If you use CD then if a DAC has higher performance than CD quality then the source is the best that you can get.
There's distinct difference between actual DSD source and upsampled DSD source. DSD just won't clear the 24bit PCM performance.

 

[odyo]

All Delta Sigma DACs can be considered hybrids as they use mutibit and DS combined.

I assume you mean this DAC ?

Yes. Is akm and sabre doing same thing ?

 

[ElNino]

RME adi2 dac/pro has an ASRC chip in the front and a DSP chip after that. This is very not representative for majority of DACs. Also are you talking about the out of band noise? That's how measurements mislead people....You should be take the in band measurements. THD IMD Multitone noise in 20KHzBW.

There is no ASRC in the RME ADI-2 pro or non. They use a steerable clock generator (DDS) instead.

You must be thinking of another DAC.

 

[bogi]

Very small amount of people can differentiate HIRES and CD quality in blind listening tests. But this is actual HIRES. If you upsample, the best you can get is from the source. If you use CD then if a DAC has higher performance than CD quality then the source is the best that you can get.
There's distinct difference between actual DSD source and upsampled DSD source. DSD just won't clear the 24bit PCM performance.

That's your opinion, based on quality of upsampling you were facing with in the past. Most probably you don't have any experience with quality of HQPlayer filters. Then we are not talking about the same thing. Most probably you are underestimating the result of many years long development, when the developer is long time specialist on low frequency DSP (not only in audio area) and, regarding HQPlayer, concentrated himself mostly on the digital filter development and less on things like GUI and user convenience.

Yes, it is very hard to differentiate PCM HiRes vs CD quality if 44.1k/16bit recording was created using high quality downsampling from HiRes recording.
But when doing it the other way - from CD quality recording and when we include also delta sigma modulation in the path, then the two paths (take into account also the difference how DAC processes them) are much easier distinguished by HQPlayer users. When skipping both complete DAC oversampling and delta sigma modulation, it is possible to find differences between upsampling filters and delta sigma modulators. This is the most usual case how HQPlayer is used. You may not trust me but if you didn't try it (it requires some time to setup things and then some quiet hours for comparisons), then it is hard to find understanding, you will repeat your position and I mine.

 

[bogi]

Yes. Is akm and sabre doing same thing ?

No. When processing PCM input, they are pure delta sigma without that additional multibit PCM part of Burr-Brown (TI) chips.
But don't overestimate that multibit part. The top 6 bits represent only very small subset of all values you can get from 24bit data. All the fine resolution is done by delta sigma part. Sabre and AKM have newer generation of higher performing chips. If I well remember the one newer from TI is PCM1795 with some improvements against the older ones.

 

[Veri]

No. When processing PCM input, they are pure delta sigma without that additional multibit PCM part of Burr-Brown (TI) chips.
But don't overestimate that multibit part. The top 6 bits represent only very small subset of all values you can get from 24bit data. All the fine resolution is done by delta sigma part. Sabre and AKM have newer generation of higher performing chips. If I well remember the one newer from TI is PCM1795 with some improvements against the older ones.

You are entirely misguided on these things I'm afraid. What is "pure" delta sigma, anyway.

 

[frogmeat69]

You are entirely misguided on these things I'm afraid. What is "pure" delta sigma, anyway.