• 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!

Sabaj D4 Review (DAC & Amp)

Rate this DAC & HP Amp

  • 1. Poor (headless panther)

    Votes: 10 6.3%
  • 2. Not terrible (postman panther)

    Votes: 103 64.8%
  • 3. Fine (happy panther)

    Votes: 45 28.3%
  • 4. Great (golfing panther)

    Votes: 1 0.6%

  • Total voters
    159
Good question... But, engineering in itself doesn't sell stuff to people, while marketing does.
We are buying things we don't really need all the time, cars, houses, phones - such "overkill" is written in our genes probably.
But we also like to limit our spending, so "most bang for the buck" is probably the optimum.
Sure, I myself use a E30 DAC to be honest... But I certainly didn't buy it because I feel it shows Engineering prowess. I bought it because it's cheap and it does the job well but there's not much engineering involved in a product like that. It's the Angle here that I disagree with, as if the Sinad chart would be a measure of the engineering hability. The guys that put together the first CD players and smartphones, the guys that have patents on some technologies, the guys that releases products with useful features like DSP, network streaming and such, that to me show great engineering, releasing an other 110 dB SINAD Dac, not really.
 
Sure, I myself use a E30 DAC to be honest... But I certainly didn't buy it because I feel it shows Engineering prowess. I bought it because it's cheap and it does the job well but there's not much engineering involved in a product like that. It's the Angle here that I disagree with, as if the Sinad chart would be a measure of the engineering hability. The guys that put together the first CD players and smartphones, the guys that have patents on some technologies, the guys that releases products with useful features like DSP, network streaming and such, that to me show great engineering, releasing an other 110 dB SINAD Dac, not really.
For the average user, probably all above 100 dB SINAD will be at least good enough.
On the other hand, which other single differentiator would be better than SINAD?
Nobody has the resouces to test "a hundred points", probably.
And even if he did, somebody would still say, "but you didn't test my 101st".

Ceterum censeo: Mission impossible.
 
For the average user, probably all above 100 dB SINAD will be at least good enough.
On the other hand, which other single differentiator would be better than SINAD?
Nobody has the resouces to test "a hundred points", probably.
And even if he did, somebody would still say, "but you didn't test my 101st".

Ceterum censeo: Mission impossible.
My point is that it's a problem trying to judge a product by a single differentiator, not that any other would be better. The average user should look at their needs, not at who wins the Sinad competition.
 
My point is that it's a problem trying to judge a product by a single differentiator, not that any other would be better. The average user should look at their needs, not at who wins the Sinad competition.
OK, a) you have to judge by something, and resources are limited ("one man show"), b) even the way it is, the tests (and the discussion after) are really helpful to the readers. There are many more passive readers than registered members I think.
 
OK, a) you have to judge by something, and resources are limited ("one man show"), b) even the way it is, the tests (and the discussion after) are really helpful to the readers. There are many more passive readers than registered members I think.
Oh I was not saying that the scope of what's measured here was not appropriate. Are you saying that Amir (That's what you mean by one man show?) judge by only one metric and that his Sinad chart could be relabeled engineering capability chart and that it means the same thing because that's how DAC are judged here? I was not suggesting that.. I was debating the point that SINAD at 1k shows great engineering.
 
What I'm saying is, Amir couldn't satisfy everybody by working alone.
What would be the other points to be considered, across the range of products?
Does SINAD show great engineering - I can't say, I'm not an engineer.
But is it possible to prove the opposite: do any badly engineered products with "great" SINAD exist?
 
What I'm saying is, Amir couldn't satisfy everybody by working alone.
What would be the other points to be considered, across the range of products?
Does SINAD show great engineering - I can't say, I'm not an engineer.
But is it possible to prove the opposite: do any badly engineered products with "great" SINAD exist?
Ok, my critic was at the statement, not at Amir's process. Yes there are plenty. Products that breaks, products that over heat, products that fails, products that you have to hard reset every other days, products that glitches, click, pop, products that are not plug and play. Product that don't follow security standards or that don't support popular formats. All can have great SINAD and are poorly "engineered" But then again, the real engineering here is at AKM and ESS, I just don't make this relation, there is not that much engineering in products like that. generally, and by that I don't want to minimise the value of competent design and yes I am not saying they are not engineers. It's just that "great engineering" I see it much more personally in other aspect of the trade than putting a reference design out that perform well on the AP
 
@amirm
Any chance you could measure new version of good old FiiO E10K? They made E10K-TC based on XMOS XUF208 + PCM5102
That sounds cool. Sadly FiiO doesn't provide the THD specs like they have with other devices.

This is under $110, can add discount code $5: mandycode
Overall not too bad considering.
Sure, I myself use a E30 DAC to be honest... But I certainly didn't buy it because I feel it shows Engineering prowess. I bought it because it's cheap and it does the job well but there's not much engineering involved in a product like that. It's the Angle here that I disagree with, as if the Sinad chart would be a measure of the engineering hability. The guys that put together the first CD players and smartphones, the guys that have patents on some technologies, the guys that releases products with useful features like DSP, network streaming and such, that to me show great engineering, releasing an other 110 dB SINAD Dac, not really.
It is a different type of engineering. Just because they didn't make the DAC chips themselves doesn't mean they aren't engineers. I mean the reality is that many other companies haven't done this and in the case of Topping; no company unseats them in terms of sonic performance and that has been proven so far.... so that doesn't mean that it "isn't hard" or every company would offer devices like this and the entire hifi industry basically wouldn't exist.
I actually test probably as many products as Amir does these days and have considered investing in some type of analyzer or ADC setup where I can measure things too and maybe even post those results here. My results are all "subjective" but I have developed quite standardized and objective testing requirements when it comes to listening tests.
I can tell you that the number of normal audio products with very easily identifiable audible distortion is shockingly high. Some of these products even use audiophile buzzwords or advertise high end DAC's like Sabre 9028, AKM449X, etc
Yet the performance is TERRIBLE, and the device sounds like you are playing through a wet paper towel. Sometimes it is fixed with an external DAC; other times the performance just sucks no matter what it is. Then a product that costs 1/5th will come along and destroy the other product... because it is indeed about engineering. Parts sums don't make good products the same way that ingredients don't make good food. Balance and understanding do.
 
That sounds cool. Sadly FiiO doesn't provide the THD specs like they have with other devices.

Overall not too bad considering.

It is a different type of engineering. Just because they didn't make the DAC chips themselves doesn't mean they aren't engineers. I mean the reality is that many other companies haven't done this and in the case of Topping; no company unseats them in terms of sonic performance and that has been proven so far.... so that doesn't mean that it "isn't hard" or every company would offer devices like this and the entire hifi industry basically wouldn't exist.
I actually test probably as many products as Amir does these days and have considered investing in some type of analyzer or ADC setup where I can measure things too and maybe even post those results here. My results are all "subjective" but I have developed quite standardized and objective testing requirements when it comes to listening tests.
I can tell you that the number of normal audio products with very easily identifiable audible distortion is shockingly high. Some of these products even use audiophile buzzwords or advertise high end DAC's like Sabre 9028, AKM449X, etc
Yet the performance is TERRIBLE, and the device sounds like you are playing through a wet paper towel. Sometimes it is fixed with an external DAC; other times the performance just sucks no matter what it is. Then a product that costs 1/5th will come along and destroy the other product... because it is indeed about engineering. Parts sums don't make good products the same way that ingredients don't make good food. Balance and understanding do.
I agree with all that. I didn't say they weren't engineers. But can you give me a few AK449X our ES90X8 stand alone DACs designed in the last 2 years that sound horrible? I am mainly talking about the basic combination of a XMOS Receiver and control with off the shelves hi performance DAC ICs.
 
Last edited:
I agree with all that. I didn't say they weren't engineers. But can you give me a few AK449X our ES90X8 stand alone DACs designed in the last 2 years that sound horrible? I am mainly talking about the basic combination of a XMOS Receiver and control with off the shelves hi performance DAC ICs.
I haven't listened to many DACs from the past 2 years.
However if you look at the performance of some of the devices on the market; the mfg specs are terrible...
ifi gold bar was just released and according to their own specs it would only get around 90db SINAD best case.
And FiiO has a Q3 even with MQA;
This device published specs from FiiO are the same thing... 80's SINAD just by their THD numbers...

I'm sure there are many products which don't specifically sound good, but most people don't know about them. I tested a sound bar with an AK4452 (I think it was this one) and they obviously did it wrong because it sounded super flat.
I'm sure if you dig through amazon & ebay; you will find random companies with these DAC chips inside of their products which may perform like garbage.... they just haven't been measured here.
 
I haven't listened to many DACs from the past 2 years.
However if you look at the performance of some of the devices on the market; the mfg specs are terrible...
ifi gold bar was just released and according to their own specs it would only get around 90db SINAD best case.
And FiiO has a Q3 even with MQA;
This device published specs from FiiO are the same thing... 80's SINAD just by their THD numbers...

I'm sure there are many products which don't specifically sound good, but most people don't know about them. I tested a sound bar with an AK4452 (I think it was this one) and they obviously did it wrong because it sounded super flat.
I'm sure if you dig through amazon & ebay; you will find random companies with these DAC chips inside of their products which may perform like garbage.... they just haven't been measured here.
FIIo specify THD+N of 0.0012%.that's 98.5 dB SINAD. I don't know where you took your 80's SINAD, there are even measurments on their site, look at 1K:

1659954098386.png
The AK4462 has a specified THD+N of -107 dB, not a reference class chip to start with. So all trough the Q3 there is a penalty of 8.5 dB in noise or distortion, but You have a non Bypassable Volume pot, there is Bass boost option which will always need headroom in the chain, , and it's a portable product. I don't see a problem here at all. But, I am not talking about Sound bar obviously. I was not talking about small dongles neither, there is obviously compromises to put 300 mW in a Dongle like the gold bar. I think you are quick to judge it's poorly engineered, first I am confident they may sound excellent, and second it just show my point that the absolute SINAD don't always have to be the absolute design goal. Is this documented that at 0.002% THD+N the degradation is audible? I was talking about DACS using high performance reference designs from XMOS and AKM or ESS. It's pretty hard to mess up with good construction practices, unless you are trying to do something special with them. You'll get 2-3 dB SINAD differences between Manufacturers, all below audibility. To be honest, even this Sabaj D4 here,2018 design, low cost chip, below SOTA numbers, I am not even so sure it sounds that bad.

"Parts sums don't make good products the same way that ingredients don't make good food. Balance and understanding do". That's the thing tough, your food analogy is an interesting one, because those chips manufacturers, they don't only sell you the chip. They literally give you the recipe with it that you just have to follow.
 
Last edited:
If you're an impecunious student and you just found yourself a deal on some HD 650s used and need a really cheap, not terrible DAC/Amp to pair with them, this guy would seem like a good choice. As long as the amp can drive the phones, the Sabaj unit should have acceptable if not cutting edge performance.

Kinda ugly, though.
 
First of all, it should be obvious that my comment was tongue-in-cheek. Second, it is not true, after the Dave there were measured the Topping E30 II and the SMSL D-6, which both scored a higher SINAD. Before the new and cheap Sabaj ones, all better. So? It was only the also not very cheap Naim Atom that performed worse.
Shanling UA2 was the next. Obviously any good new DAC like those from Topping or SMSL will have measurements that smash anything as old as DAVE. The tongue in cheek occurred to me but I figured it must have been based on some mistake either way.
 
Shanling UA2 was the next. Obviously any good new DAC like those from Topping or SMSL will have measurements that smash anything as old as DAVE. The tongue in cheek occurred to me but I figured it must have been based on some mistake either way.
Well, the Shanling was first of all a remeasurement and furthermore is a dongle and not a desktop DAC, a different category. The age does not count if the item is still for sale as the actual flagship, which is the case for the Chord DAVE. If you are saying it is obvious that any cheap chinese DAC beats the Chord flagship, what does this say about Chord?
 
Well, the Shanling was first of all a remeasurement and furthermore is a dongle and not a desktop DAC, a different category. The age does not count if the item is still for sale as the actual flagship, which is the case for the Chord DAVE. If you are saying it is obvious that any cheap chinese DAC beats the Chord flagship, what does this say about Chord?
Maybe it say about Chord that they feel these few extra dBs of SINAD don't make their product any better?
 
Well they are still charging top dollar for something that is worse than a dongle lol
as long as they can get it. what is wrong with that. As far as i know they are a "for profit" company.
Kudo's to them for pulling the wool over peoples eyes. Buyers do not have a gun stuck to their heads. It is by choice (at least in the US)
 
Back
Top Bottom