Thank you very much for the report.
I guess the quality of I2S signals of your device is not so bad, none the less,
you have also encountered the issue.
Your finding is very valuable for me.
Thanks a lot again.
Hey RayCTech
Does this mean that you actually successfully managed to play 352.8kHz/24bit files without continuous noise into your ES9018 based DAC using an 80MHz master clock as long as both the transport and the DAC was employing your original JFET regulators?
As you know I have also been participating in the SDTrans 192 project and also actively supported the development of the Fidelix Caprice ES9018 based DAC. As you know I have one SDTrans 192 Rev2.1 modified by yourself with your JFET regulators, but it still is not able to play noiseless with neither the Buffalo nor the Caprice when using an 80MHz master clock.
I apologize to exa065 if this is considered out of topic. I also think that exa065 Post #180 makes clear that the noise problem during playback into a ES9018 based DAC using 80MHz clock remains the same whether the playback transport is an exaU2I or an SDTrans192. (Unfortunately no-one has yet been able to have both devices present at the same time and test under identical conditions. Maybe RayCtech will be the first one? ;-))
I very much support the exaU2I (which I have on order) as well as the SDTrans 192. I think both devices are complementary and for slightly different applications. While the exaU2I is intended for file playback from a computer, the SDTrans 192 allows computer-less, battery-powered file playback. Keep up the good work on both of these great devices
Does this mean that you actually successfully managed to play 352.8kHz/24bit files without continuous noise into your ES9018 based DAC using an 80MHz master clock as long as both the transport and the DAC was employing your original JFET regulators?
As you know I have also been participating in the SDTrans 192 project and also actively supported the development of the Fidelix Caprice ES9018 based DAC. As you know I have one SDTrans 192 Rev2.1 modified by yourself with your JFET regulators, but it still is not able to play noiseless with neither the Buffalo nor the Caprice when using an 80MHz master clock.
I apologize to exa065 if this is considered out of topic. I also think that exa065 Post #180 makes clear that the noise problem during playback into a ES9018 based DAC using 80MHz clock remains the same whether the playback transport is an exaU2I or an SDTrans192. (Unfortunately no-one has yet been able to have both devices present at the same time and test under identical conditions. Maybe RayCtech will be the first one? ;-))
I very much support the exaU2I (which I have on order) as well as the SDTrans 192. I think both devices are complementary and for slightly different applications. While the exaU2I is intended for file playback from a computer, the SDTrans 192 allows computer-less, battery-powered file playback. Keep up the good work on both of these great devices
Last edited:
Yes - I was for a long time not even aware that this was supposed to be a problem, but I was able to recreate circumstances that was able to cause similar problems.
I uses the Sabre32 pinout and not the Sabre8 pinout and functionality of the ES9018 chip and have a completely different programming that also includes support of 352.8k and above even with a 80MHz clock, but I uses a 98.304MHz clock with superior performance compared to a 80MHz clock.
A combination of informations from the datasheet for the ES9018 and then optimal programming and then circuits between the DAC chip and the speakers that have good enough common mode rejection allows use of a 80MHz clock, but when playing at 352.8k and above you should both increase the clock speed and change the programming for better performance.
I found that the combination with both a 98.304MHz clock and the proper programming for 352.8k (352.8k to 1536k) was a real winner.
Yes I believe this (both your and my post) is OT as it does not have anything to do with the exaU2I unit.
I have received my exaU2I unit today, and I completed both my new music server and my new player PC yesterday.
Both are Hackintosh with dualboot (OSX 10.6.6 and Windows 7 in 64bit mode) and OSX defined both systems as Mac Pro 5.1...
The great fun is that the Hackintosh systems now works with both USB 3 and eSATA witch Mac´s does not have..
Hi exa065,
If I use the exau2I in stereo mode with an LFE channel (2.1), is it possible to send all LF content to the LFE channel from software when playing back normal stereo? Reason I ask is because i have a pair of 300B monoblocks driving widerange speakers and a Linn Subwoofer - I'd like to send everything below 60Hz to the sub. I suppose I might need some kind of crossover software, but I'm wondering if there's any driver level configuration for the sub...
-Raja
If I use the exau2I in stereo mode with an LFE channel (2.1), is it possible to send all LF content to the LFE channel from software when playing back normal stereo? Reason I ask is because i have a pair of 300B monoblocks driving widerange speakers and a Linn Subwoofer - I'd like to send everything below 60Hz to the sub. I suppose I might need some kind of crossover software, but I'm wondering if there's any driver level configuration for the sub...
-Raja
ES9018 DPLL Bandwidth Parameter Setting for I2S Input
I'd like to explain one more issue that those who connect the exaU2I devices to ES9018-based DAC modules may experience.
As the most of you may know, ES9018 has a "DPLL Bandwidth Parameter" setting. People agree that the lower parameter value setting will bring the better sound as long as the setting does not cause occasional "unlock" events.
As for S/PDIF input, "the lowest" value can be easily attainable for various transport units. On the other hand, as for I2S input, almost none transport can succeed in using the "the lowest" setting. This situation is well compiled and commented by glt in the following blog pages,
Buffalo II I2S Interface (Again) H i F i D U I N O
More on Sabre32 I2S Locking H i F i D U I N O
One effective way to mitigate the "unlock" events is selecting "the best default" value for the setting as explained in the article above.
Therefore, I'd like to recommend those who connect the exaU2I devices to ES9018-based DAC modules that they make it sure that they set "the best default" for the DPLL Bandwidth parameter.
I do not think this is a kind of defect of ES9018. One very reasonable guess is;
"The parameter value might effect 64 times severely in the case of I2S because DPLL function runs not on Word Clock but on Bit Clock for I2S."
I'd like to explain one more issue that those who connect the exaU2I devices to ES9018-based DAC modules may experience.
As the most of you may know, ES9018 has a "DPLL Bandwidth Parameter" setting. People agree that the lower parameter value setting will bring the better sound as long as the setting does not cause occasional "unlock" events.
As for S/PDIF input, "the lowest" value can be easily attainable for various transport units. On the other hand, as for I2S input, almost none transport can succeed in using the "the lowest" setting. This situation is well compiled and commented by glt in the following blog pages,
Buffalo II I2S Interface (Again) H i F i D U I N O
More on Sabre32 I2S Locking H i F i D U I N O
One effective way to mitigate the "unlock" events is selecting "the best default" value for the setting as explained in the article above.
Therefore, I'd like to recommend those who connect the exaU2I devices to ES9018-based DAC modules that they make it sure that they set "the best default" for the DPLL Bandwidth parameter.
I do not think this is a kind of defect of ES9018. One very reasonable guess is;
"The parameter value might effect 64 times severely in the case of I2S because DPLL function runs not on Word Clock but on Bit Clock for I2S."
There are plug-ins for both foobar2000 and J.River Media Center that will let you do this.
Playing DXD gives a CPU load of totally 4 - 6%.
So far everything have functioned perfect
It's great!
Are you using two exaU2I devices at the same time?
How about 4-channel/one device or 8-channel/two devices DXD surround play?
Raja,
I will create a new area on the exaDevices website where I am going to post sample configurations. I will present options for 2.1 playback for Foobar and J. River.
I am aware of that your advice may be important to Buffalo DAC users but not to ES9018 based DAC´s in general.
The only problem is that for the Buffalo users the solution are a new firmware..
As I am now "toying" with the exaU2I I have discovered that the exaU2I have interesting features that may increase performance on DAC´s that uses the ES9018 in full Sabre32 mode.
Hi Dmitry,
I will open again the discussion about enhancements for version 2 when the waiting list backlog is cleared and all users are taken care of. If you want, you can contact me via PM and explain how you want to see i2c volume control implemented.
With foobar2000, I think you can use a plug-in called the 'Channel Mixer' and with JRMC, you can use a combination of their 'Output Format' and 'Room Correction' tools. However, I have a 5.1 speaker setup and haven't tried it with 2.1. If they don't support a subwoofer unless you have a surround setup, try contacting the developer and explain that you'd like to use it for a 2.1 system and they might be able to make a minor adjustment to their plug-ins.
The problem was reproduced on the following DACs.
A. ESS ES9018 2ch Evaluation Board (40MHz, 100MHz)
B. Buffalo32s
C. Buffalo II DAC (80MHz, 100MHz)
D. Fidelix CAPRICE (96MHz)
You can have one another work-around. Not using "Over Sampling Mode".
If you read the datasheet - ESS recommends 352.8k -> 1536k samplerates to be used with OSF disabled.
If you manages to make a work-around like using a higher clock frequency and thus be able to use the OSF it is up to the designer to solve any issues.
Likewise ESS recommends the "Best Default" setting for the DPLL.
If a product use other DPLL settings it is up to the designer to solve any issues.
So my conclusion are simply:
The products you mentioned are simply not designed for 352.8k -> 1536k use.
Ask the designers of those products to implement support for higher than 192k samplerates...
I'm afraid that your are confusing the 352.8 kHz noise issue with the unlock on the DPLL Bandwidth Parameter of "the lowest" issue.
As for the DPLL Bandwidth Parameter issue, it happens even on 44.1 - 192 kHz sampling rates.
exaU2I basic evaluation
The exaU2I device have been up and running for 24 hours now and I have started the basic evaluations...
The basic evaluation setup are as follows:
Player PC:
Dualboot - 64bit - OSX 10.6.6 and Windows 7 Enterprise.
SSD system disk, 2T local storage, Gigabit ethernet.
With the ASIO driver evaluations I use Foobar2000 and Windows 7.
NAS:
Mac Pro (Hackintosh) running OSX 10.6.6 - 64bit.
SSD system disk, 8T RAID, Gigabit ethernet.
The first evaluated setup are the Japanese Caprice DAC / preamp from FIDELIX Home
Other setup´s and configurations will be evaluated from a technical aspect as I have the time for it....
The exaU2I are connected with a standard USB cable to the PC, and the output side of the GMR isolators and the LVDS transmitters (using HDMI connectors and cable) are powered by a 3.3 volt LIFEPO4 (12Ah) battery.
The HDMI cable connects to the Japanese Caprice DAC / preamp from Fidelix with a integrated HDMI / LVDS receiver with galvanic isolation.
The Caprice DAC uses a ES9018 chip and a 96MHz clock from Fidelix.
This setup might be the second most advanced / sophisticated setup with the exaU2I - disconnecting the USB power and run the exaU2I from two additional LIFEPO4 batteries will possibly be the ultimate solution and will be implemented when I start the listening tests..
The current setup works very well and did so at the very first attempt
44.1k/16bit flac to DXD 352.8k/24bit plays as far as I can evaluate perfectly.
I will test this configuration with 352.8k/32bit and evaluate the various configuration setups of the Caprice DAC before I continues on the next candidates (DAC)..
I will congratulate exa065 with this excellent and very useful product.
EDIT: The used (in picture with the exaU2I) I2S -> LVDS -> HDMI transmitter are also a product from Fidelix.
The exaU2I device have been up and running for 24 hours now and I have started the basic evaluations...
The basic evaluation setup are as follows:
Player PC:
Dualboot - 64bit - OSX 10.6.6 and Windows 7 Enterprise.
SSD system disk, 2T local storage, Gigabit ethernet.
With the ASIO driver evaluations I use Foobar2000 and Windows 7.
NAS:
Mac Pro (Hackintosh) running OSX 10.6.6 - 64bit.
SSD system disk, 8T RAID, Gigabit ethernet.
The first evaluated setup are the Japanese Caprice DAC / preamp from FIDELIX Home
Other setup´s and configurations will be evaluated from a technical aspect as I have the time for it....
The exaU2I are connected with a standard USB cable to the PC, and the output side of the GMR isolators and the LVDS transmitters (using HDMI connectors and cable) are powered by a 3.3 volt LIFEPO4 (12Ah) battery.
The HDMI cable connects to the Japanese Caprice DAC / preamp from Fidelix with a integrated HDMI / LVDS receiver with galvanic isolation.
The Caprice DAC uses a ES9018 chip and a 96MHz clock from Fidelix.
This setup might be the second most advanced / sophisticated setup with the exaU2I - disconnecting the USB power and run the exaU2I from two additional LIFEPO4 batteries will possibly be the ultimate solution and will be implemented when I start the listening tests..
The current setup works very well and did so at the very first attempt
44.1k/16bit flac to DXD 352.8k/24bit plays as far as I can evaluate perfectly.
I will test this configuration with 352.8k/32bit and evaluate the various configuration setups of the Caprice DAC before I continues on the next candidates (DAC)..
I will congratulate exa065 with this excellent and very useful product.
EDIT: The used (in picture with the exaU2I) I2S -> LVDS -> HDMI transmitter are also a product from Fidelix.
Attachments
Last edited:
The first evaluated setup was the Japanese Caprice DAC / preamp and the use of I2S -> LVDS -> HDMI cable -> LVDS -> I2S and battery power of the GMR isolators and LVDS hardware was done to verify that there was no limitations of the possibilities or performance with the exaU2I device.
Buffalo DAC, ACKO DAC, my own DAC and other DAC´s with and without the LVDS / HDMI interface should work just as easy with the exaU2I, and I will post the results as soon as some of the configurations have been tested.
I do not have access to a ACKO DAC so that configuration will not be tested...
Buffalo DAC, ACKO DAC, my own DAC and other DAC´s with and without the LVDS / HDMI interface should work just as easy with the exaU2I, and I will post the results as soon as some of the configurations have been tested.
I do not have access to a ACKO DAC so that configuration will not be tested...
Good news! Several of the users who have received and connected the device have contacted us to say that they are happy with it. I am very pleased to announce that we haven't had any support issues and all devices function without any glitches. We will see more reviews in the near future.
Unfortunately, some users have expressed concerns about sharing their experience on this thread. They say that they were approached, misled and misinformed by third parties that are interested to interfere with the success of exaU2I. In response to their concerns we will create a dedicated forum for supporting the device.
Unfortunately, some users have expressed concerns about sharing their experience on this thread. They say that they were approached, misled and misinformed by third parties that are interested to interfere with the success of exaU2I. In response to their concerns we will create a dedicated forum for supporting the device.