Looks like I may have finally found a USB to SPDIF converter allowing up to 24/96, in the form of the Lilo. I'm not 100% sure it will do 24/96 conversion, or if it is only the DAC that works at 24/96, but I am trying to find out more details.
The only downside is the optical rather than coaxial digital out. Would this be an easy mod to do 9change to coaxial)?
The manufacturer has apparently tried to keep jitter low and it also has inputs for mic and line, as well as a line out and headphone out (all 1/4 inch), so this will come in handy for my laptop music production.
It's cheap too. Here is the link to one retailer. http://vocativeaudio.com/eng_liloiii.html
The only downside is the optical rather than coaxial digital out. Would this be an easy mod to do 9change to coaxial)?
The manufacturer has apparently tried to keep jitter low and it also has inputs for mic and line, as well as a line out and headphone out (all 1/4 inch), so this will come in handy for my laptop music production.
It's cheap too. Here is the link to one retailer. http://vocativeaudio.com/eng_liloiii.html
This is the info I found on Pacific Valve & Electric's site. It mentions jitter reduction a few times. Hope this is of interest?
New from MUSILAND - a PC powered USB
DAC called the LILO. What makes the LILO
so unique is Musiland has completely re-
written the USB bulk protocol which results in
much lower jitter than the common, jitter
prone adaptive protocol. The LILO also offers
DSP capability, advanced conversion
algorithm and accompanying driver that aims
to take full advantage of the newly designed
I/O capabilities of Windows Vista, yet, the
LILO software is flexible enough to operate
efficiently in the XP / 2003 environment.
The LILO uses the advanced German
MICRONAS USB2.0 sensor to drive the USB
programmable controller. Firmware,
specifically built by MUSILAND, fully taps the
potential of the DSP hardware. MUSILAND
pushes the limit with the high performance
24-Bit High independent Burr-Brown codec,
the PCM3010. The MUSILAND LILO also
controls power through an advanced power
purification technology to ensure that as
much as 3 V RMS (8.5V MAX) analog output
level. The LILO has a CIRRUS LOGIC high-
performance CS8420 as an SPDIF digital
transmitter and high-precision optical digital
output.
The LILO software driver is a great and easy
to use software console with up sampling,
volume control, equalizer, bass balance
control and loudness control. The LILO
software and the driver act in tandem to
reduce jitter by redundantly sending packets
and hand shaking on the best ones that it
sees.
New from MUSILAND - a PC powered USB
DAC called the LILO. What makes the LILO
so unique is Musiland has completely re-
written the USB bulk protocol which results in
much lower jitter than the common, jitter
prone adaptive protocol. The LILO also offers
DSP capability, advanced conversion
algorithm and accompanying driver that aims
to take full advantage of the newly designed
I/O capabilities of Windows Vista, yet, the
LILO software is flexible enough to operate
efficiently in the XP / 2003 environment.
The LILO uses the advanced German
MICRONAS USB2.0 sensor to drive the USB
programmable controller. Firmware,
specifically built by MUSILAND, fully taps the
potential of the DSP hardware. MUSILAND
pushes the limit with the high performance
24-Bit High independent Burr-Brown codec,
the PCM3010. The MUSILAND LILO also
controls power through an advanced power
purification technology to ensure that as
much as 3 V RMS (8.5V MAX) analog output
level. The LILO has a CIRRUS LOGIC high-
performance CS8420 as an SPDIF digital
transmitter and high-precision optical digital
output.
The LILO software driver is a great and easy
to use software console with up sampling,
volume control, equalizer, bass balance
control and loudness control. The LILO
software and the driver act in tandem to
reduce jitter by redundantly sending packets
and hand shaking on the best ones that it
sees.
Are you saying the clock in it is not a good one, or are you saying you don't know which clock it has, but it would need to be a good one to make it a worth while purchase over something like the HagUSB?
Is it possible to fit an external PSU to this unit?
Things like the HagUSB run off the USB port and are considered decent devices, so why can't the Lilo?
It would be nice if someone could may be work out a way of using batteries with it.
Is it possible to fit an external PSU to this unit?
Things like the HagUSB run off the USB port and are considered decent devices, so why can't the Lilo?
It would be nice if someone could may be work out a way of using batteries with it.
Kane Williams said:
It may be decent used with the HagDac which has a very efficient clock recovery. On its own it's quite mediocre.
The 12MHz clock shown on the Lilo board does not inspire too much confidence. In fact the only part which seems interesting is the custom USB chip. $60 may be a very good price for that chip alone as everything else will have to be redesigned to meet some basic quality requirements anyway. Or in any case that's what i would do.
Ok, thanks for that info. I'm not actually a DIY person, but I am interested in what you "modders" do.
Looking at the board then, are you sayng you would consider designing a whole new board and just "steal" the USB chip from the Lilo, or do you think replacing other components on their board will result in a decent product?
How much time, effort and money would it result in?
I'm more interested in using the USB to SPDIF aspect of the Lilo above it's DAC capabilities (this is just abonus for use with my laptop). In your opinion, would replacing the clock and adding a good PS be enough to get top quality audio, when used in conjunction with a good ext DAC, such as my Xindak DAC-5?
May I ask a few tech questions?
1. I have been told that jitter can only be measured at the analogue outputs. Is this true?
2. If this is true, does this not mean that in some designs, the DAC may be the main cause of jitter and that before the signal goes through the DAC, it may have been low in jitter? I'm guessing a good clock is still required for low jitter, even in the didital domain?
Looking at the board then, are you sayng you would consider designing a whole new board and just "steal" the USB chip from the Lilo, or do you think replacing other components on their board will result in a decent product?
How much time, effort and money would it result in?
I'm more interested in using the USB to SPDIF aspect of the Lilo above it's DAC capabilities (this is just abonus for use with my laptop). In your opinion, would replacing the clock and adding a good PS be enough to get top quality audio, when used in conjunction with a good ext DAC, such as my Xindak DAC-5?
May I ask a few tech questions?
1. I have been told that jitter can only be measured at the analogue outputs. Is this true?
2. If this is true, does this not mean that in some designs, the DAC may be the main cause of jitter and that before the signal goes through the DAC, it may have been low in jitter? I'm guessing a good clock is still required for low jitter, even in the didital domain?
Kane Williams said:
This is completely irrelevant to me. And i am not a "modder" in the sense that i enjoy building and not modding. Diy very seldom saves money, let alone time and effort. If you want great deals buy used high quality stuff from Audiogon - diy cannot (usually) produce the same value as used, sorry. But it's fun.
As far as jitter goes, very few of us can actually measure it and even fewer can make sense out of the measurements. To me this boils down to following empiric rules which hopefully minimise it. And yes, the dac chip itself would show varying immunity to jitter but it's always a good idea to try and minimise it at every stage.
The clock i was talking of has no effect directly on the dac. It provides 12MHz for the USB decoding and has indirect effect upon the output jitter. The 2707 chips i use for usb are sensitive to the quality of this clock. It seems the Lilo is not using the dreaded isohronous protocol, which us, diyers are commonly forced to use and may in theory be capable of much higher performance. Whether this is true or not i cannot say without actually trying the chip out. Too many possibilities for things to go wrong.
Improving PS is always a good idea. Improving the clock in this case may have no effect at all. Using spdif at any stage is something i religiously try to avoid. Especially the double conversion in this case. Surely there are I2S signals available in the Lilo which you can connect to your dac.
I bet it's fun!
I'd love to modify or make my own gear and I'm sure one day I'll have a go at something like a mic pre amp (probably buy a kit). I have no electronics expertise or anything, so it's a bit daunting.
My DAC only has coaxial or optical SPDIF iputs, so how would I go about trying to get I2S into that? Are people making of modifying DAC's with I2S inputs?
I am tempted to buy the Lilo and compare it to the HagUSB into my Xindak DAC-5. If it is as good with 16/44.1 audio, I would sell my UNUSED HagUSB.
May be down the line, people will come up with some mods for it?
I'd love to modify or make my own gear and I'm sure one day I'll have a go at something like a mic pre amp (probably buy a kit). I have no electronics expertise or anything, so it's a bit daunting.
My DAC only has coaxial or optical SPDIF iputs, so how would I go about trying to get I2S into that? Are people making of modifying DAC's with I2S inputs?
I am tempted to buy the Lilo and compare it to the HagUSB into my Xindak DAC-5. If it is as good with 16/44.1 audio, I would sell my UNUSED HagUSB.
May be down the line, people will come up with some mods for it?
Kane Williams said:
I am tempted to encourage you
Very curious how it compares. And whether hidef does indeed sound better through it.The CS8420 is capable of accepting three different data formats from the USB chip, only one of which is I2S. Other than measuring or just trying this out i cannot offer any way to know for certain it indeed is I2S. Unless, of course it is working in hardware mode 2 and then it's trivial to see the input data format from the board.
Your Xindak has an AD1852 dac, fed either from an ASRC or an input receiver chip. Comparing the pin connections to the datasheet will let you know if it gets I2S data but this is very likely. You then have a choice of either cutting pcb tracks (ugly) or (hopefully) just desolder several series resistors along the I2S lines. The I2S link cannot run long distances; more than 15cm is actually pushing it. It may be possible though to integrate the USB decoder chip inside the DAC case.
I see.
So what you are suggesting is that it may be worth buying a Lilo and use nothing but the USB chip and the software/protocol that Musiland have develloped from it by putting it into an existing DAC box (or build your own) add a USB port and take the I2S signal to the DAC chip in the box?
I wish I was capable of doing this.
This DAC seems to have done something similar http://www.pacificvalve.us/Brigatta.html but it appears they may still convert to SPDIF? May be this is because the PCM1704UK won't accept I2S? I have no idea.
I think I will plump for a Lilo. Only $38 delivered, as long as I don't get stung with VAT/Duty. The only thing that will cheese me off is that I have only just bought the HagUSB, which cost me about £90 and it would be redundant! :-(
"HEY JUST HAD A THOUGHT". What are the chances the software driver that can be downloaded here http://www.pacificvalve.us/LILOSUPT.html will be compatible with any other current USB to SPDIF converterts (HagUSB, Pop-Pulse etc)?
So what you are suggesting is that it may be worth buying a Lilo and use nothing but the USB chip and the software/protocol that Musiland have develloped from it by putting it into an existing DAC box (or build your own) add a USB port and take the I2S signal to the DAC chip in the box?
I wish I was capable of doing this.
This DAC seems to have done something similar http://www.pacificvalve.us/Brigatta.html but it appears they may still convert to SPDIF? May be this is because the PCM1704UK won't accept I2S? I have no idea.
I think I will plump for a Lilo. Only $38 delivered, as long as I don't get stung with VAT/Duty. The only thing that will cheese me off is that I have only just bought the HagUSB, which cost me about £90 and it would be redundant! :-(
"HEY JUST HAD A THOUGHT". What are the chances the software driver that can be downloaded here http://www.pacificvalve.us/LILOSUPT.html will be compatible with any other current USB to SPDIF converterts (HagUSB, Pop-Pulse etc)?
The only think that has stopped me buying a Lilo so far is that it uses an optical connection. Why the heck couldn't they have used a coaxial SPDIF instead?
I know it costs a bit more, but it suggest pro rather than consumer. I mean the Lilo Classic could have had the optical and the Deluxe coaxial. They could have charged an extra $10.
Doesn't an optical connection inherently have higher jitter than coaxial? If so, why would a company go out of its way to try to make a "low jitter" device (as they claim the Lilo to be) and use optical?
I know it costs a bit more, but it suggest pro rather than consumer. I mean the Lilo Classic could have had the optical and the Deluxe coaxial. They could have charged an extra $10.
Doesn't an optical connection inherently have higher jitter than coaxial? If so, why would a company go out of its way to try to make a "low jitter" device (as they claim the Lilo to be) and use optical?
Kane Williams said:
You can try posting on diyhifi.org. Gordon Rankin knows everything there is to be known about USB protocols and supporting chips. He certainly would know about Lilo.
The driver you mention is certainly not usable with any of the isochronous chips. The chip inside the Lilo has been custom programmed to work with that driver.
Thanks for the link. I couldn't find anything about the Lilo or the PCM3010 chip, but I will post something and see who knows what.
I did read this though and think that it is this that the Lilo addresses? So, even using a "standard" clock, may be they can achieve low jitter, as the Lilo is the master and the PC the slave?
I can't remember the name of the person who I'm quoting here, so my appologies.
"The problem with USB is different to that seen with S/PDIF. The issue is that the source clock is in two parts. We get notified of the sample rate, but we then get the samples as lumps every USB send interval. So, in order to keep the buffer happy we need to sync to the send interval. That is 1kHz. Then we can synthesise the sample rate clock. There is no mechanism to send a raw clock with USB other than this. So a second link to slave the DAC would not add anything".
"What is needed - as always - is a way of slaving the sender to the DAC clock. (Which is perhaps what you meant.) To do this does not require another USB cable, merely a protocol to control the USB sender. Sadly there is no such protocol. Not that it would be impossible to create one. An appropriate driver could be written and the sender essentially folded into a PLL controlled by the DAC end. But this would not be plug and play any more. It would solve the problem".
I did read this though and think that it is this that the Lilo addresses? So, even using a "standard" clock, may be they can achieve low jitter, as the Lilo is the master and the PC the slave?
I can't remember the name of the person who I'm quoting here, so my appologies.
"The problem with USB is different to that seen with S/PDIF. The issue is that the source clock is in two parts. We get notified of the sample rate, but we then get the samples as lumps every USB send interval. So, in order to keep the buffer happy we need to sync to the send interval. That is 1kHz. Then we can synthesise the sample rate clock. There is no mechanism to send a raw clock with USB other than this. So a second link to slave the DAC would not add anything".
"What is needed - as always - is a way of slaving the sender to the DAC clock. (Which is perhaps what you meant.) To do this does not require another USB cable, merely a protocol to control the USB sender. Sadly there is no such protocol. Not that it would be impossible to create one. An appropriate driver could be written and the sender essentially folded into a PLL controlled by the DAC end. But this would not be plug and play any more. It would solve the problem".
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.