[Bhargu]

Do I feel any part of the response sounds augmented? No.

By neutral, I just mean no emphasis in any regions of the FR graph. So, basically what you said here. I can expect the typical details of a Sabre, without the glare/harshness, right? At least that's what I have been hearing about the 9038 based DACs till now.

Yes. This has been covered in the 'DAC for Peanuts' thread. I use iSilencer, iDefender, LT3045 power supply. All ESS dacs are sensitive to power supply from my experience.

Sorry, I didn't see that. What do you think of the iPurifier series of PS from iFi? I have been thinking of getting iSilencer or something similar. I had tested before for ground-loop issue, but haven't noticed it till now. Will I still need something like iDefender, in case of something which is not really audible, but still there? I believe iDefender also lets you switch out the USB power channel with separate cleaner power supply, right?

Which regulator. There are a bunch.

I was talking about the Tesla regulator (big black one) that I thought you had removed in the image that you attached.

If you like Burr Brown get the iFi.

The problem is, I have not listened to Burr-Brown DACs. I have used ES and AK based ones, but never the BB ones.


Thanks for your help with this. I was thinking of getting an iFi Nano BL (mostly because of the portability, being a single USB, battery operated with built-in iPurifier) when I came across this. I had a Hifime 9018 sabre DAC before (used it for my mobile) and this seemed interesting, the cheapest 9038Pro DAC. I have been interested in 9038 as it apparently reduces the typical brightness of Sabre chips.

[Bhargu]

Regardless of USB only or external, there is a pretty noisy regulator that converts the power to +/- voltage for the opamps. There's no way to avoid it.

Do you remember which regulators were used in the unit, before you changed them?

[wushuliu]

Do you remember which regulators were used in the unit, before you changed them?

The opamps are powered by an LT1054. The rest are generic 3.3v and 1.2v LDOs. I did not change them. I avoided them completely by either removing them or by removing the components after them so I could access the pads for my external power supply. You mentioned the Tesla in the Hifime promotional photo, but that was not used in the final design. My dac looks nothing like that internal photo. Hifime made a lot of changes, even after the first batch went out in February.

I have modded the hell out of it and am all done now.

[Bhargu]

The opamps are powered by an LT1054. The rest are generic 3.3v and 1.2v LDOs. I did not change them. I avoided them completely by either removing them or by removing the components after them so I could access the pads for my external power supply. You mentioned the Tesla in the Hifime promotional photo, but that was not used in the final design. My dac looks nothing like that internal photo. Hifime made a lot of changes, even after the first batch went out in February.

I have modded the hell out of it and am all done now.

Are the modded results significantly better than the stock experience? Also, do you use this as one of your daily drivers? I am considering this vs Topping D50, and am finding it hard to decide.

[wushuliu]

Are the modded results significantly better than the stock experience? Also, do you use this as one of your daily drivers? I am considering this vs Topping D50, and am finding it hard to decide.

Yes, significantly better. But I like modding and learning along the way. If that's not for you then get the D50. I know UDA38Pro feedback is hard to find. I am the only person talking about it in length. Not sure why. Hifime doesn't seem interested in promoting their product or sharing more information. I have a feeling the dac does not 'measure' well even though it sounds excellent in stock form.

Tough decision, I know. Whichever choice will be impressive.

[envydd]

In real life how much of a difference does the 9038 pro make? I am on 9018 on my NuPrime gear and have a SABABJ DA3 for my hifiman .... I dont perceive much of a difference - or maybe its my ears.

[envydd]

I would like to know if the volume control has become better. For example how would it compare with a good DSP-ed volume control. Are the new ESS DACs' volume control as good as the software ones like Roon https://community.roonlabs.com/t/roons-64-bit-floating-point-volume-leveling/25547/2

[wushuliu]

In real life how much of a difference does the 9038 pro make? I am on 9018 on my NuPrime gear and have a SABABJ DA3 for my hifiman .... I dont perceive much of a difference - or maybe its my ears.

That's not a question I can answer for you. You will just have to listen to one. Engineering-wise the 9038 is a *significant* step up. These new chips are complicated with many many options for implementation. How good a 9028/38 based dac sounds will depend on what options were selected and how everything is designed around it. The more one researches these chips, the more surprises you find. Like an easter egg hunt. Which is why it's taking some of these companies a longer amount of time to design new products around them (that optimize everything). And since there is a NDA for these chips, there's only a certain amount of info that can be shared. Here is a good primer on the new chips courtesy of Benchmark:

https://benchmarkmedia.com/blogs/application_notes/dac3-introducing-the-new-es9028pro-converter

The performance and complexity of the support circuitry outside of the ES9028PRO can be configured to match a manufacturer's budget and space constraints. When Benchmark designed the DAC2 we started by using all of the ESS-recommended performance options and then we added our own set of enhancements. This same strategy is carried forward to the DAC3. Our performance-driven strategy greatly increases the number of support components, but these components add important performance improvements. For this reason, it is important to understand that there can be a wide variation in the performance of an ES9028PRO converter solution. The overall performance will be determined by the manufacturer's budget, goals, and skills.

As for the volume question, you've asked this before and I'm not sure what you're looking for beyond what was already said. The ESS volume control is integrated *into* the chip. You can't beat the volume processing of the chip itself in terms of signal/bit integrity.  Plus for the 9028 and 9038 there are additional factors that play into the quality of the analog volume output. There are probably additional improvements but you'd need to sign an NDA and get the datasheet. If you really want nitty gritty answers I'd suggest asking the guys at Twisted Pear. They are the only diy folks working with these chips and talking about them.

Now whether or not you will hear a difference, only you can say.

[Bhargu]

Yes, significantly better. But I like modding and learning along the way. If that's not for you then get the D50. I know UDA38Pro feedback is hard to find. I am the only person talking about it in length. Not sure why. Hifime doesn't seem interested in promoting their product or sharing more information. I have a feeling the dac does not 'measure' well even though it sounds excellent in stock form.

Tough decision, I know. Whichever choice will be impressive.

I am not sure about the quality of LT1054 (noise should be higher than a typical LDO), but OPA1622 has a nominal PSRR of about -140dB. Any half-way decent regulator should have been good enough, right? But from what I heard, ESS has been recommending ES9311Q for the left and right channel voltage rails for 9038Pro. It is supposed to have a noise characteristic of 1μVrms. I don't know if HiFime has used a regulator of similar performance, but I don't think they would at this price range.

The only measurements that I have seen of a HiFime product was of their 9018 DAC, which was not flattering. But that could have easily been due to USB noise, which can be rectified at user-end. Though, exotic sounding DACs generally don't seem to measure well.

I haven't done any DIY projects in more than a decade, but if I get this, I was planning to move this device to a new housing so that I can include the regulators and USB isolator inside the housing itself (maybe also a few batteries with charging and protection circuits). I was thinking of a housing form factor similar to iFi micro series.

[fredgarvin]

This is another DAC that I'm interested in but it's missing some feature or other. In this case it's in not having coaxial input.

[Roen]

Yes, significantly better. But I like modding and learning along the way. If that's not for you then get the D50. I know UDA38Pro feedback is hard to find. I am the only person talking about it in length. Not sure why. Hifime doesn't seem interested in promoting their product or sharing more information. I have a feeling the dac does not 'measure' well even though it sounds excellent in stock form.

Tough decision, I know. Whichever choice will be impressive.

Unfortunately, unless Amir over at ASR has a bad unit, it definitely doesn't measure well for being neutral.

But hey, people might not want a neutral, reference DAC to begin with.

[wushuliu]

Unfortunately, unless Amir over at ASR has a bad unit, it definitely doesn't measure well for being neutral.

But hey, people might not want a neutral, reference DAC to begin with.

That's a false premise. Since Amir himself can't subjectively differentiate most of the DACs he measures, then there is no basis to say whether this or anything he measures sounds 'neutral'. You just have a cult of measurement junkies ranking components as being awesome or not based on one person's assessment - which, hey, whatever floats your boat; but if he himself can't actually hear a difference then what's the point? In fact my takeaway from his work is that distortion isn't a big deal, since he doesn't seem to notice it even when it's pretty bad. Same with the other metrics. IOW, anyone who isn't invested in the tribal mindset over there would have to come to the conclusion that an $80 DAC will do as well as a $1K DAC unless defective or has features you want. Not that I believe that, but that's the logical outcome of his approach. I'll stop there and won't say any more on the subject.

[cujobob]

Unfortunately, unless Amir over at ASR has a bad unit, it definitely doesn't measure well for being neutral.

But hey, people might not want a neutral, reference DAC to begin with.

I’m beginning to think Amir works for Topping. The company has QC issues galore and has swapped out parts with those of unknown quality. He also rants constantly about his measurements that haven’t found to correlate with sound quality make one device better than another which makes no logical sense. He also has something against Schiit, in particular.

[fredgarvin]

I’m beginning to think Amir works for Topping. The company has QC issues galore and has swapped out parts with those of unknown quality. He also rants constantly about his measurements that haven’t found to correlate with sound quality make one device better than another which makes no logical sense. He also has something against Schiit, in particular.

Agreed, and he has been banned from other forums for his outright trolling.

[pinkfloyd4ever]

I'm glad I'm not the only one who finds most everything at audiosciencereview absurd, maybe even bordering on cult-like.

Seriously, who listens to music with an oscilloscope? Your ears are not a distortion analyzer, and music is not a perfect sine wave from a function generator. Who gives a rat's ass about the difference between a noise floor at -110dB vs -130dB?

None of the criteria he uses to rank DACs really matter IMO. I never see any discussion on ASR of how DACs actually sound when listening to music.

[Letitroll98]

I think it's a good source of measurements if that's all you take away from it.  I don't read it as a final authority, but I do like his comparisons.  And the replies on the blog occasionally fill in some of the subjective areas.  I guess there are cult followers there, I just don't notice them.