iFi micro iDSD

[jhwalker]

That's too bad. I have had one for over a month and I have none of these issues. Once I switched to native DSD playback vs DoP all of the unexplained noises went away. Mine works pretty much flawlessly. Hopefully they get these ironed out. iFi is pretty top notch with customer support so I would encourage you to give them another chance. If it's a defective new unit I would think there would be no problems with an exchange.

 

Same here: very soft "pop" (like a needle drop) when I turn it on. No hiss, no other issues. No issues with charging, playback whatever - just wonderful sound

 

Sorry to hear others have issues.

[wisnon]

The one that assured me, there is nothing wrong with it, was only you. Up until now on forums I've encountered two other people that experienced "the extra hiss" they are probably overblowing and nitpick too.

I do agree that it does sound incredible when there is no hiss. I wish I could compare it to another unit, but there is no stores in NYC where I can go and do it.

 

BTW, I did not mean to stir trouble, I just wanted to convince myself to buy it again, since I did like the sound and the versatility of it.

 

No problem with my unit either.

[bmoura]

No problem with my unit either.

 

I haven't experienced any problems with my iDSD Micro.

 

But I can say that the first iDSD Nano I received didn't work correctly and I had to replace it. The second iDSD Nano I received has been fine.

 

This may be a case of a bad batch. We'll see as the comments come in.

[Ned Kelly]

No problems with my Nano - never heard the Micro.

[sebassie]

The only problem I had was a volume knob with an asymmetrical insert. The knob touched the rim when turning. iFi replaced the knob within one week!

[Ned Kelly]

Hmm - re-reading some of the hiss comments dredges up a problem I've had with the HRT microStreamer (same Burr-Brown DAC chip fwiw) from day one, but I suspect mine may be a faulty unit - if it's not, I would have read more reports of the problems I've had with the line-out. Basically that tiny metal casing seems to leave it vulnerable to RMI/EMF noise leaving me with issues around hiss and serious SQ degradation. Using the iUSBPower with iPurifier into the microStreamer this morning I was stunned by just how good this little gizmo is used purely as a DAC - crystal clear and showing the kind of low-end kick that I've come to expect from the iDSD Nano. Still not sure that the line-out on mine is 100% - moving the mini cable connecting it to my amp messes with the sound in ways that my other portables with a line-out do not. I just used the headphone-out with my IEMs and forgot about the line-out prior to buying the iFi stack - plugging the DAC straight back into the USB port on my ASUS laptop the line-out was completely unlistenable and that's not the case from my Mac Mini.

[Miska]

When you understand BitPerfect is a marketing term for Non-Oversampling, it is easier to understand why there may be slightly more noise. That is the nature of NOS filtering.

 

Yes, when "BitPerfect" is selected with PCM input, the DAC operates in NOS mode and thus outputs huge amount of ultrasonic distortion. It will also move the delta-sigma modulator's noise shaping noise to much lower frequencies than it normally would be (peaking at 30 kHz for 44.1 kHz input rate).

 

I wouldn't recommend using BitPerfect setting for PCM inputs below 352.8 kHz sampling rate.

[g3act]

Unfortunately, after over a week of trying to live with it, I'm returning my as well.

I like the sound, it is great, but there are lot's of issues. I'm not sure it is my luck or others just don't pay enough attentions. There are clicks, pops during the on/off, slight channel imbalance at low but still listenable levels, but the biggest one for me and I just want to ask everybody here with one to check -

 

Bit-perfect Filter with 44.1kHz-88kHz lossless files. Do it in A+ or iTunes with BitPerfect, start playing, pause it, and rotate the pot to the max. What my expectation is/was - science, but what I hear is hiss, white noise, randomized noise level, whatever you want to call it. You do not need to max the volume to hear it, but it is easier to prove it. It is not there in other filter setting, but only in recommended by iFi preferred Bit-perfect.

This noise makes it, to say the least, less then 'bit-perfect'. Lot's of people say, on head-fi forum, they prefer Bit-perfect setting as it is more analog sounding and I wander if they mistake this white noise overplayed with music as analog sound. I compared it with dacs that cost less then half the price of Micro and there is no hiss, Dragonfly is silent, so is ODAC.

Even if not to take the price into consideration, I decided, that I do not like my gear to hiss. Now I just wander if it was a faulty unit and I should get another one or look elsewhere.

 

I also experience hiss with 44.1 and 48khz music when using the Bitperfect filter. Disappears with higher sample rates.

 

I use the minimum phase filter with the red book stuff, bit perfect with HiRes. I listen to classical mainly, so it is very noticeable during quiet passages.

 

Hope iFi can work out a fix, but think it is just a symptom of a NOS approach. Wasn't that the reason over sampling was introduced back in the day? To combat the issue of noise?

[G0bble]

I use a NOS dac and never had a problem with hiss... Strange to say that NOS results in hiss.

[mraulino]

Listen for yourself and make up your own mind. The most reliable method. I own a Chord Qute EX and a Lampi. Lampi is one of the best PCM around and is king of DSD. I actually like Chord design, so maybe I am in the minority. FPGA for the moment seems better with PCM. Perhaps they should all do like Lampi and iFi and keep the PCM and DSD signal paths separate (2 Dacs in a box).

 

 

Hi Wisnom,

 

I'm curious because you mentioned the QuteEX, and I was just wondering if I should get my QuteHD upgraded to EX, or spend the money on an iDSD micro instead (it's almost the same $$$ here in China).

 

Understood from your post above that the iDSD handles DSD much better than the QuteEX, but in your opinion how these two compare in terms of PCM?

 

And, in case your QuteEX is an upgraded HD, have you noticed any real difference in SQ between the HD and the EX in regards to both PCM and DSD (besides of course the fact that the EX can play higher-def files)? (I know this may seem out of topic, but the context is that, if PCM is the same on QD and EX, but the iDSD outperforms both in DSD, then I think investing in an iDSD would be the most logical decision for me...)

 

Thanks,

Mario

[wisnon]

Hi Wisnom,

 

I'm curious because you mentioned the QuteEX, and I was just wondering if I should get my QuteHD upgraded to EX, or spend the money on an iDSD micro instead (it's almost the same $$$ here in China).

 

Understood from your post above that the iDSD handles DSD much better than the QuteEX, but in your opinion how these two compare in terms of PCM?

 

And, in case your QuteEX is an upgraded HD, have you noticed any real difference in SQ between the HD and the EX in regards to both PCM and DSD (besides of course the fact that the EX can play higher-def files)? (I know this may seem out of topic, but the context is that, if PCM is the same on QD and EX, but the iDSD outperforms both in DSD, then I think investing in an iDSD would be the most logical decision for me...)

 

Thanks,

Mario

 

If you are looking for DSD performance, then go with the Micro. Cant yet say for sure if the Micro matches the Chord for redbook. i would need to study that more later on.

[LewinskiH01]

I'm a bit confused. Several people here are commenting on what seems like quality control issues on iFi products. But are these widespread among all products, or poping up on Nanos only, iDSD only? I'm seriously considering an iDSD, but can't really tolerate bad quality control in this day and age.

[labjr]

DSD 256 & DSD512 seems to be a big selling feature for the Micro iDSD. I wonder if it really makes any difference. From what I've read DSD512 seems like it's beyond the point of diminishing returns. Quantization noise is completely separated from the audio band at DSDS256. And doesn't clocking have to be twice as good, each time the sample rate doubles?

 

I think Andreas Koch has said there are other problems beyond DSD128 with clocking, propagation delay, ground bounce etc. with currently available chips? And higher sample rates can actually sound worse than DSD128.

[One and a half]

I think Andreas Koch has said there are other problems beyond DSD128 with clocking, propagation delay, ground bounce etc. with currently available chips? And higher sample rates can actually sound worse than DSD128.

 

Yup, beyond (real) DSD128 was tried a few years ago in the lab with little success. I am really questioning what methods are used to produce DSD256 let alone DSD512.

[bmoura]

Yup, beyond (real) DSD128 was tried a few years ago in the lab with little success. I am really questioning what methods are used to produce DSD256 let alone DSD512.

 

DSD256 was an undocumented feature on the ESS chip used by exaSound and Pyramix for DSD256 playback. But it is available and Pyramix and exaSound led the way with this feature. They've been followed by others like the iFi folks that also offer DSD256 products today.

 

The Pyramix work in this area is particularly important since their DSD256 pro products allow for new DSD256 recordings in Stereo and Multichannel DSD like the one recently released by Eudora Records.

 

And Pyramix's recent release of the HAPI product allows both AD and DA conversions in 8 channels up to DSD256. It should prove quite popular with record companies looking to record new albums in DSD (Stereo & Multichannel). Not to mention collectors and audiophiles who may want to archive their valued Open Reel Tape and Vinyl LP collections at DSD128 or DSD256.

 

Interesting products indeed!

[Miska]

DSD256 was an undocumented feature on the ESS chip used by exaSound and Pyramix for DSD256 playback. But it is available and Pyramix and exaSound led the way with this feature. They've been followed by others like the iFi folks that also offer DSD256 products today.

 

I think ESS supports DSD at least to DSD512 and PCM up to 1.5 MHz. TI chips used for example by iFi have always officially supported DSD256 and PCM at 768k, so there it is not so much big deal. I think the biggest missing item was USB implementation for it. It is more challenging to implement at DAC side with DoP than with ASIO-style driver because of 33-50% DoP overhead amount of extra data that would flow over I2S when DoP is decoded only after the interface in FPGA or similar.

[wisnon]

DSD 256 & DSD512 seems to be a big selling feature for the Micro iDSD. I wonder if it really makes any difference. From what I've read DSD512 seems like it's beyond the point of diminishing returns. Quantization noise is completely separated from the audio band at DSDS256. And doesn't clocking have to be twice as good, each time the sample rate doubles?

 

I think Andreas Koch has said there are other problems beyond DSD128 with clocking, propagation delay, ground bounce etc. with currently available chips? And higher sample rates can actually sound worse than DSD128.

 

The big selling points are the TONS of features it offers. It was a crowd design unit. The DSD high rates are just future proofing and in any case, DSD bypasses the "chip" decoder and is low pass filtered with SS integrated components.

[G0bble]

The big selling points are the TONS of features it offers. It was a crowd design unit. The DSD high rates are just future proofing and in any case, DSD bypasses the "chip" decoder and is low pass filtered with SS integrated components.

I didn't get it - Can you elaborate? Are you saying the decoding function is not chip based but part of another circuit in the dac? If yes does that solve a problem with chip based decoding?

 

Thanks

 

Sent from MiPad

[wisnon]

I didn't get it - Can you elaborate? Are you saying the decoding function is not chip based but part of another circuit in the dac? If yes does that solve a problem with chip based decoding?

 

Thanks

 

Sent from MiPad

 

The way iFI implements the hybrid BB1793 allows them to bypass the dac chip decode for DSD, ie a separate pathway for DSD which is simply low passed filtered (still inside the chip).

 

Thus, you have native DSD treatment with minimal processing.

ML Grado explains it better than I can over at Head-fi.org forum.

[labjr]

The big selling points are the TONS of features it offers. It was a crowd design unit. The DSD high rates are just future proofing and in any case, DSD bypasses the "chip" decoder and is low pass filtered with SS integrated components.

 

Having more features isn't really future-proofing if it's not well implemented. I can't imagine a $500 MSRP product has properly implemented DSD512 when it's doubtful that the technology exists at any price.

[wisnon]

Having more features isn't really future-proofing if it's not well implemented. I can't imagine a $500 MSRP product has properly implemented DSD512 when it's doubtful that the technology exists at any price.

???

 

You are mixing up what I said. The features are the other stuff, like powerful amp, myriad of settings. ability to charge peripherals, multiple filters, EQ features lie X-Bass and X-feed and of course the easy connectivity to iDevices.

 

The future proofing is with thehigh rate DSD. People have been upsampling to 256 for sure and I think Miska has stated that he has now enabled 512 playback in HQPlayer. he also confirmed that the chips can handle the rates and it was the USB tech that was lagging. He says Sabre can do up to DSD1024.

[labjr]

???

 

The future proofing is with thehigh rate DSD. People have been upsampling to 256 for sure and I think Miska has stated that he has now enabled 512 playback in HQPlayer. he also confirmed that the chips can handle the rates and it was the USB tech that was lagging. He says Sabre can do up to DSD1024.

 

Doesn't mean it sounds better than DSD128 and may actually sound worse. The implementation and support circuitry would need to have proportionately lower jitter to be better. Probably doesn't yet exist or companies like Playback Designs would be doing it. Personally, I see no need to go beyond DSD256 since DSD256 already has a ridiculousy high sample rate as far as audio goes, and moves the noise away from the audio band. File size would be ridiculous too.

 

The only future-proofing is bragging rights which seems to be the selling point since everyone touts the feature when they write about it.

[bmoura]

Doesn't mean it sounds better than DSD128 and may actually sound worse. The implementation and support circuitry would need to have proportionately lower jitter to be better. Probably doesn't yet exist or companies like Playback Designs would be doing it.

 

Remember that the Playback Design units have been on the market for awhile now. Undoubtedly Andreas is working on a next generation series of DACs which have, as yet, an unannounced feature set. Ditto for other companies that have come out with DACs before the iFi DSD Nano and DSD Micro.

 

We'll have to wait and see if these DACs include newer features including ASIO drivers (vs. DoP) and resolutions above DSD64 and DSD128.

 

Based on recent announcements from other companies that are supporting DSD256 and even DSD512, I think iFi and exaSound will have more company/competition at DSD256 and above soon!

[labjr]

Just last year Andreas told me that they could easily convert their DACs to play DSD256 but there was no gain, and actually had other signal artifacts which made it worse. I have great respect for Andreas and find him to be a no nonsense type of guy.

[Kelly]

Having more features isn't really future-proofing if it's not well implemented. I can't imagine a $500 MSRP product has properly implemented DSD512 when it's doubtful that the technology exists at any price.

 

Don't discount the micro iDSD (or nano iDSD) just because of the price especially when you haven't heard it. The same basic implementation will be used in the upcoming $1500 mini iDSD and then later in AMRs DP-777. They aren't investing all of this time and effort in the micro iDSD as a standalone product. They are going a little backwards in introducing their newest tech in the lower line products first, but then again perhaps that is a way to get all of the wrinkles ironed out before the more esoteric stuff get it.