Excellent news! Well done
Exciting possibilities have opened up
Maybe now we can engage your services
All I did was connect the balanced output from my preamp to the ADC with nothing in between. No buffers, resistors or capacitors. The ref5 does not have any DC on its outputs, so that was not a problem. The preamp output tubes are 6H30's, with a 10uf capacitor to block DC from getting to the outputs.
ADC update #2
Just wanted to report my experience with the ADC using the USBStreamer to go from I2C to USB and then to AIF files. I use Acko's board with his local regulators for the analog stuff. I am using DIYINHK TPS7A4700 boards to provide 5v to the Acko 3.3v analog regulators and to provide 3.3v directly to the digital power for the ADC.
I am able to reliably record audio up to 24bit 192khz, which is the limit for the USBStreamer. I am a bit shocked how good it sounds. I have been making digital files of some of my vinyl that I cannot find on CD.
This has opened my eyes to the world of DAWs and the power that they have. While I have only scratched the surface, I have a lot more respect for recording engineers and what they do.
I tried the follow DAWs for making digital recordings only. I only tried using one of the many plugins once, to open up the sound of a bad vinyl recording.
Goldwave - This was easy to use, but it did not have ASIO support and I found it to sound kind of thin.
Reaper - Free version. This sounded better and was easy to use. But it did not do metadata. It sounded better then Goldwave but not as good as...
Studio One - 30 day full trial, all features work. This is what I used to do all my final recordings. This had the steepest learning curve and was not intuitive. However, once I figured it out it was a pleasure to use and it sounded great. I just made a point of doing all my stuff during the 30 day trial. I would not pay the full price unless I was serious about recording.
Overall the project was a lot of fun and I learned a lot about recording and DAWs. As the saying goes, I know just enough to be dangerous.
Now I have to figure out what to do with the ADC.
John
Just wanted to report my experience with the ADC using the USBStreamer to go from I2C to USB and then to AIF files. I use Acko's board with his local regulators for the analog stuff. I am using DIYINHK TPS7A4700 boards to provide 5v to the Acko 3.3v analog regulators and to provide 3.3v directly to the digital power for the ADC.
I am able to reliably record audio up to 24bit 192khz, which is the limit for the USBStreamer. I am a bit shocked how good it sounds. I have been making digital files of some of my vinyl that I cannot find on CD.
This has opened my eyes to the world of DAWs and the power that they have. While I have only scratched the surface, I have a lot more respect for recording engineers and what they do.
I tried the follow DAWs for making digital recordings only. I only tried using one of the many plugins once, to open up the sound of a bad vinyl recording.
Goldwave - This was easy to use, but it did not have ASIO support and I found it to sound kind of thin.
Reaper - Free version. This sounded better and was easy to use. But it did not do metadata. It sounded better then Goldwave but not as good as...
Studio One - 30 day full trial, all features work. This is what I used to do all my final recordings. This had the steepest learning curve and was not intuitive. However, once I figured it out it was a pleasure to use and it sounded great. I just made a point of doing all my stuff during the 30 day trial. I would not pay the full price unless I was serious about recording.
Overall the project was a lot of fun and I learned a lot about recording and DAWs. As the saying goes, I know just enough to be dangerous.
Now I have to figure out what to do with the ADC.
John
John, many thanks for sharing. Yes, this project has been an eye opener and frankly I did not expect to go this far
Anyway, going further with this ADC the roadmap is clearly indicated in the
manual and elsewhere in this thread. Personally, I would like to see super HiRes DxD recordings 352.8k/24bit but cannot seem to find any capture hardware for it. The idea for a custom designed FPGA one has already been mooted. This will go direct into an SDCard so bypass PC and USB layers. See diagram in the manual. Unfortunately, development costs for this to be professionally done is well over $30k!
Then there is this digital RIAA option. This removes the need for an external pre-amp type analog equalization. The chip default is set for 48k only- may not be very appealing but can be reprogrammed on the fly by an external controller. Of course filter computations and downloading to chip must be done for every possible Fs. Concept for controller is also previewed in the manual accordingly
Again, development cost ~$5-10k is needed.
Not sure where to get all these funding, maybe Kickstarter or by kind donations/contributions
User base is quite small at this stage, so I am afraid I may have to leave it as is for the time being....
Last edited:
Acko,
did you already check this biquad programming functionality and verify with measurements? Build in RIAA at 48kHz works very well but I had problems programming other biquads
kind regards, Daniel
Acko,
did you already check this biquad programming functionality and verify with measurements? Build in RIAA at 48kHz works very well but I had problems programming other biquads
kind regards, Daniel
I haven't tried the others yet, a bit more involved but will dig in and see after the controller design as proposed in the manual is completed.
Yes with the even promised 768kHz
The main issue on this x200 Core will remain:
- no real increase of mips per core
- the power consummation
may sometimes it would be better to build the bit shifts in hard coded HW...
just my 2 cents
hp
Yes with the even promised 768kHz
The main issue on this x200 Core will remain:
- no real increase of mips per core
- the power consummation
may sometimes it would be better to build the bit shifts in hard coded HW...
just my 2 cents
hp
For Vinyl ripping I cannot see why the above mentioned specs. matters?
As for direct recording to disk this has been discussed before:
biquad programming of ES9112 does not work correctly
Dear all,
I've done some more research and found that the biquad programming functionality of the ES9112 does not work correctly. While the Biquad registers can be programmed correctly (verified by reading them back after programming) it seems that they are not correctly latched to the internal biquad chain. The left channel always shows linear, the right channel always standard build-in 48kHz RIAA response
Meanwhile this has also been confirmed by Spiri from AKOUO and Ismosys, the European distributor. My Info is that a successor is to be released around the end of this year adressing this (and other) error.
Now I am wondering whether the top of the line ES9102 has the same issues. Did anyone try to program any custom biquads to the ES9102 and verified the correct functionality? acko, any progress?
kind regards, Daniel
Dear all,
I've done some more research and found that the biquad programming functionality of the ES9112 does not work correctly. While the Biquad registers can be programmed correctly (verified by reading them back after programming) it seems that they are not correctly latched to the internal biquad chain. The left channel always shows linear, the right channel always standard build-in 48kHz RIAA response
Meanwhile this has also been confirmed by Spiri from AKOUO and Ismosys, the European distributor. My Info is that a successor is to be released around the end of this year adressing this (and other) error.
Now I am wondering whether the top of the line ES9102 has the same issues. Did anyone try to program any custom biquads to the ES9102 and verified the correct functionality? acko, any progress?
kind regards, Daniel
Acko,
did you already check this biquad programming functionality and verify with measurements? Build in RIAA at 48kHz works very well but I had problems programming other biquads
kind regards, Daniel
Dear all,
I've done some more research and found that the biquad programming functionality of the ES9112 does not work correctly. While the Biquad registers can be programmed correctly (verified by reading them back after programming) it seems that they are not correctly latched to the internal biquad chain. The left channel always shows linear, the right channel always standard build-in 48kHz RIAA response
Meanwhile this has also been confirmed by Spiri from AKOUO and Ismosys, the European distributor. My Info is that a successor is to be released around the end of this year adressing this (and other) error.
Now I am wondering whether the top of the line ES9102 has the same issues. Did anyone try to program any custom biquads to the ES9102 and verified the correct functionality? acko, any progress?
kind regards, Daniel
Daniel, thanks for highligting this issue. You do have a 9102 module and see if the biquads for digital RIAA works? Programming should be the same as for 9112
I did not get round to check this on the 9102 as the customer who original favored digital RIAA now prefers to do it by the external preamp instead and avoid any chip related issues. There is also the concern that doing filters digitally within the chip will take it away from the important task of doing the AD conversion. But you may have other reason for doing so ...
It is a concern of course if something does not work as specified.
Best I contact ESS directly and explaing this issue. Hopefully they can shed some light and advise us accordingly.
Thats the point, I do not have an ES9102 based module. Hence the question if anyone else tried it
Indeed programming should be exactly the same. I went through both datasheets and all registers are exactly the same for both chips. Seems its only the package thats different. I am using an Arduino for programming and could help out with code snippets.
If the ES9102 would work I could rework my board...
Of course it would be no problem to do the EQ in software afterwards if recording is the goal, however it would be kind of elegant to use the "promoted" feature
Tried to get some feedback from ESS but nothing. Support from Ismosys is very good though but at some point when it gets too far into the technical details they can not help.
kind regards, Daniel
Thats the point, I do not have an ES9102 based module. Hence the question if anyone else tried it
Indeed programming should be exactly the same. I went through both datasheets and all registers are exactly the same for both chips. Seems its only the package thats different. I am using an Arduino for programming and could help out with code snippets.
If the ES9102 would work I could rework my board...
Of course it would be no problem to do the EQ in software afterwards if recording is the goal, however it would be kind of elegant to use the "promoted" feature
Tried to get some feedback from ESS but nothing. Support from Ismosys is very good though but at some point when it gets too far into the technical details they can not help.
kind regards, Daniel
Sorry Daniel, I mistook you for someone else and you are using something you built up on your own with the 9112 chip. I would like to take a look at your code snippet for the biquads. PM me if you do not mind
Interesting.
My Info is that they are working hard on a successor for ES9112 (maybe also 9102) addressing "all the issues"
eek: whatever that means, I experienced quite a few) of the current versions to be released in Q4 (My guess Q1 2016). However Ismosys stated that the current versions will be still available though not recommended for new designs.kind regards, Daniel
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.