Re: mono playback
It could be that the DAC is not being setup correctly via I2C.
What are you using for input?
Is your PIC (IC4) installed correctly?
I would not recommend you use the DAC directly out. It needs a filter.
Cheers!
Russ
back said:
It could be that the DAC is not being setup correctly via I2C.
What are you using for input?
Is your PIC (IC4) installed correctly?
I would not recommend you use the DAC directly out. It needs a filter.
Cheers!
Russ
IVY 2
Hi Guys,
Is there any news about when IVY 2 will be available ?
I was kind of holding out for counterpoint but I guess that could be a long wait.
what do you advise ?
cheers
mike
Hi Guys,
Is there any news about when IVY 2 will be available ?
I was kind of holding out for counterpoint but I guess that could be a long wait.
what do you advise ?
cheers
mike
Re: IVY 2
To quote Brian's post from HeadFi:
They won't be released for another 4-5 weeks though.
I just hope the Twisted Pear gang will at least order enough boards so that I won't have to set my alarm clock at 4 in the morning to grab a piece or two! 😀
mikelm said:
To quote Brian's post from HeadFi:
They won't be released for another 4-5 weeks though.
I just hope the Twisted Pear gang will at least order enough boards so that I won't have to set my alarm clock at 4 in the morning to grab a piece or two! 😀
mikelm said:
Mostly objective. 🙂
Counterpoint does "sound" subjectively excellent to me. But measurement shows that the noise floor is about 8db higher and the THD similarly higher.
Now I actually feel pretty good about this is it is better than I had hoped. 🙂 And if you blindfolded me and asked me which I were listening to, I doubt I could tell you. So at least its not really "worse".
Counterpoint is also just pretty big. 😀 Pretty much the same size as the entire new Buffalo32S PCB. Not that this is a key consideration. Just throwing it out there.
I am not very good at subjective lingo.
Hopefully some of you will try Counterpoint out soon and give me your opinion.
Cheers!
Russ
Output Vrms from Buffalo 32S
I understand that the initial version of Buffalo32S will come as one-board solution including the new IVY II output stage. Will the IVY II stage part of the board also be pre-assembled or come as a kit of components? In case it comes as an entirely pre-assembled unit I would like to know what balanced and single ended Vrms to expect?
Thanks.
I understand that the initial version of Buffalo32S will come as one-board solution including the new IVY II output stage. Will the IVY II stage part of the board also be pre-assembled or come as a kit of components? In case it comes as an entirely pre-assembled unit I would like to know what balanced and single ended Vrms to expect?
Thanks.
Vreg
Could someone sell me Vreg as a kit (pcb+parts)?
My e-mail is antoni.ferragut a.t. gmail.com.
Could someone sell me Vreg as a kit (pcb+parts)?
My e-mail is antoni.ferragut a.t. gmail.com.
Re: sneak peak of BUF32S
It looks very nice and I'm glad to see that there is an option for separate PSUs to each of the analogue channels.
Russ White said:
It looks very nice and I'm glad to see that there is an option for separate PSUs to each of the analogue channels.
Aliasing Problem With Buffalo
While it is great to hear all the good news about the upcoming new products, I still have a problem with the existing Buffalo. Maybe Russ, Brian, Dustin or someone else can come up with a suggestion.
I recently did some measurements on my Buffalo/IVY and discovered significant aliasing in my setup (see attached image). Measurements were done with Visual Analyzer (freeware), a Terratec Soundcard with 96 kHz sampling and a CD on which I burned a number of test tones at different levels. As can be seen, the aliasing component of the 19600 Hz tone is so strong as if there was almost no filtering at all. Note also the relatively strong IM-components at 4900 and 9800 Hz.
Checking the ESS-paper "ESS Technology Sabre 9008 8 Channel DAC Performance Comparison", there is a graph on IM mesurements with 18 and 20 KHz test tones which shows absolutely no aliasing components. ESS was using an 96 kb/s SPDIF input signal for this test; I'm not sure what source they used. My SPDIF source was a Harman Kardon HD 760 CD-Player; but can this account for such a difference?
If I use the CD-Player analog output instead of the IVY/Buffalo output (same test setup), there is absolutely no trace of aliasing! OK, the HD 760 uses the PMD-100 chip, which is supposed to be pretty good, but I can't believe that it is normal for the Buffalo to behave in this way. The difference can also clearly be seen on the scope; it shows some kind of a beating signal with the Buffalo and a clean sine wave with the HD 760. I think I need to solve this in order to utilize the full sonic potential of the Buffalo.
I tried the reset provided on the Buffalo board; no change.
Any ideas?
Kurt
While it is great to hear all the good news about the upcoming new products, I still have a problem with the existing Buffalo. Maybe Russ, Brian, Dustin or someone else can come up with a suggestion.
I recently did some measurements on my Buffalo/IVY and discovered significant aliasing in my setup (see attached image). Measurements were done with Visual Analyzer (freeware), a Terratec Soundcard with 96 kHz sampling and a CD on which I burned a number of test tones at different levels. As can be seen, the aliasing component of the 19600 Hz tone is so strong as if there was almost no filtering at all. Note also the relatively strong IM-components at 4900 and 9800 Hz.
Checking the ESS-paper "ESS Technology Sabre 9008 8 Channel DAC Performance Comparison", there is a graph on IM mesurements with 18 and 20 KHz test tones which shows absolutely no aliasing components. ESS was using an 96 kb/s SPDIF input signal for this test; I'm not sure what source they used. My SPDIF source was a Harman Kardon HD 760 CD-Player; but can this account for such a difference?
If I use the CD-Player analog output instead of the IVY/Buffalo output (same test setup), there is absolutely no trace of aliasing! OK, the HD 760 uses the PMD-100 chip, which is supposed to be pretty good, but I can't believe that it is normal for the Buffalo to behave in this way. The difference can also clearly be seen on the scope; it shows some kind of a beating signal with the Buffalo and a clean sine wave with the HD 760. I think I need to solve this in order to utilize the full sonic potential of the Buffalo.
I tried the reset provided on the Buffalo board; no change.
Any ideas?
Kurt
Attachments
Re: Aliasing Problem With Buffalo
Hi Kurt,
I am not sure what if going on there.
A couple suggestions:
1) Don't reset the DAC without completely cycling the power, as you need to also reset the microcontroller.
2) Try eliminating the IVY from the picture and just measure ES9008 output directly. Try to just FFT the output of the DAC with an RC filter that gives a 30k -3dB point or something.
Cheers!
Russ
Javin5 said:
Hi Kurt,
I am not sure what if going on there.
A couple suggestions:
1) Don't reset the DAC without completely cycling the power, as you need to also reset the microcontroller.
2) Try eliminating the IVY from the picture and just measure ES9008 output directly. Try to just FFT the output of the DAC with an RC filter that gives a 30k -3dB point or something.
Cheers!
Russ
Hi Russ
I did recycle the power. I also checked the output of the Buffalo with the scope: Same thing. No change if I additionally disconnect the IVY. An FFT plot isn't really required, since it shows so clearly on the scope.
The problem must be either on the Buffalo board or have something to do with the SPDIF that is fed into it. Would there be a difference if my CD-player would output SPDIF with 44.1 instead of the 96 that was used for the tests at ESS?
Unfortunately, I don't have any other transports or DACs for comparison. Maybe you could burn a test tone on a cd and check?
I did recycle the power. I also checked the output of the Buffalo with the scope: Same thing. No change if I additionally disconnect the IVY. An FFT plot isn't really required, since it shows so clearly on the scope.
The problem must be either on the Buffalo board or have something to do with the SPDIF that is fed into it. Would there be a difference if my CD-player would output SPDIF with 44.1 instead of the 96 that was used for the tests at ESS?
Unfortunately, I don't have any other transports or DACs for comparison. Maybe you could burn a test tone on a cd and check?
Hi Kurt-
I assume you are feeding your Buffalo SPDIF through he comparator circuit. Looking at the date you ordered your Buffalo, I suspect you have one of the Buffalos with the slower (incorrect) comparator, which could cause anomalies with 96kHz and greater input.
If this is the case, I can replace the part for you for free, or send you a replacement if you can do it yourself.
EDIT: The older comparator will be marked "C20A" and the newer part will be marked "C14A".
I assume you are feeding your Buffalo SPDIF through he comparator circuit. Looking at the date you ordered your Buffalo, I suspect you have one of the Buffalos with the slower (incorrect) comparator, which could cause anomalies with 96kHz and greater input.
If this is the case, I can replace the part for you for free, or send you a replacement if you can do it yourself.
EDIT: The older comparator will be marked "C20A" and the newer part will be marked "C14A".
Javin5 said:
Kurt, one other thing to try is toggling switch #3 and cycling power.
Some of the older firmware has an issue where that switch should be low.
Cheers!
Russ
Hi Guys,
For C13-C16 on the IVY, I have a choice of the following capacitors in my parts bin. Does anybody tried any of these? Please let me know which is a better choice. all are .01mfd.
1. Vishay-Roederstein MKP1837
2. Vishay Roederstein KP 1830
3. WIMA MSK2
If the answer is none of the above, that's fine too. Let me know what I should buy. I bought the kit from somebodyelse, guess these caps were a later introduction and my kit does not have it.
Thanks,
Dinesh
For C13-C16 on the IVY, I have a choice of the following capacitors in my parts bin. Does anybody tried any of these? Please let me know which is a better choice. all are .01mfd.
1. Vishay-Roederstein MKP1837
2. Vishay Roederstein KP 1830
3. WIMA MSK2
If the answer is none of the above, that's fine too. Let me know what I should buy. I bought the kit from somebodyelse, guess these caps were a later introduction and my kit does not have it.
Thanks,
Dinesh
dviswa said:
All of the above are good choices. I have used the WIMA MKS2 so I would probably try that first. 🙂
Cheers!
Russ
Kurt,
I think that the point you are raising (aliasing) is of major concern. However, I am tempted to think (or to hope...whatever) that it is something related to your Buffalo DAC.
Next week, when at home, I'll burn a CD with test tones and undertake a thorough analysis. I encourage Brian, Russ and anyone else to do that test as well... WE need to know...
I think that the point you are raising (aliasing) is of major concern. However, I am tempted to think (or to hope...whatever) that it is something related to your Buffalo DAC.
Next week, when at home, I'll burn a CD with test tones and undertake a thorough analysis. I encourage Brian, Russ and anyone else to do that test as well... WE need to know...
Looking at the ES9008 datasheet, Kurt's response in post #1559 could be correct if the slow roll-off filter was active.
But Russ' response in the previous post does not seem correct either way. If the ES9008 was set up with slow roll-off the 43kHz tone should be down 100dB, and with the fast roll-off filter it should be down 120dB compared to the fundamental.
An explanation from Dustin would be much appreciated.
But Russ' response in the previous post does not seem correct either way. If the ES9008 was set up with slow roll-off the 43kHz tone should be down 100dB, and with the fast roll-off filter it should be down 120dB compared to the fundamental.
An explanation from Dustin would be much appreciated.
Aliasing
Hi Russ: I did try what you suggested with switch 3; nothing changed. In your FFT, the 43 kHz tone would be aliasing if you were using 48 (or maybe 96) instead of the red book 44.1 kHz. To see this effect clearer, you should choose a tone closer to the Nyquist frequency, e.g. about 20 kHz (I used 19.6). But at 43 kHz, this is still quite a significant component. Thanks a lot for forwarding this to Dustin.
Hi Brian: The comparator on my board is indeed still an old C20A. Please send me a new C14A as offered; I will try to exchange it myself. Thanks a lot in advance, I'll mail you my shipping address. Hope this solves the problem; if not, we will at least have eliminated one possible cause.
Hi oj: I also believe that this is something particular to my buffalo board only, but I'm really looking forward to what you'll find out. Please contact me if you have any questions regarding the test setup.
Hi ojg: The fast rolloff was activated, strange, hm ... I will check the solder joints of the switch tomorrow, just to be sure.
In the meantime I had an opportunity to repeat the test with a Sony Blu Ray Player BDP-S350 (plays also CDs). Exactly the same results. This leads me to believe that the problem is caused by something on the board, not by the SPDIF-signal from the transport. And yes, the Sony also reproduces the 19600 Hz test tone absolutely clean and alias-free on its own outputs.
My wav-files are about 21 MB each (2 minutes for each test tone), unfortunately way too large to attach here or to mail (most mail boxes have a 10 MB limit). Any other ideas how to make them available?
Kurt
Hi Russ: I did try what you suggested with switch 3; nothing changed. In your FFT, the 43 kHz tone would be aliasing if you were using 48 (or maybe 96) instead of the red book 44.1 kHz. To see this effect clearer, you should choose a tone closer to the Nyquist frequency, e.g. about 20 kHz (I used 19.6). But at 43 kHz, this is still quite a significant component. Thanks a lot for forwarding this to Dustin.
Hi Brian: The comparator on my board is indeed still an old C20A. Please send me a new C14A as offered; I will try to exchange it myself. Thanks a lot in advance, I'll mail you my shipping address. Hope this solves the problem; if not, we will at least have eliminated one possible cause.
Hi oj: I also believe that this is something particular to my buffalo board only, but I'm really looking forward to what you'll find out. Please contact me if you have any questions regarding the test setup.
Hi ojg: The fast rolloff was activated, strange, hm ... I will check the solder joints of the switch tomorrow, just to be sure.
In the meantime I had an opportunity to repeat the test with a Sony Blu Ray Player BDP-S350 (plays also CDs). Exactly the same results. This leads me to believe that the problem is caused by something on the board, not by the SPDIF-signal from the transport. And yes, the Sony also reproduces the 19600 Hz test tone absolutely clean and alias-free on its own outputs.
My wav-files are about 21 MB each (2 minutes for each test tone), unfortunately way too large to attach here or to mail (most mail boxes have a 10 MB limit). Any other ideas how to make them available?
Kurt
Re: Aliasing
I agree that it is very likely something about the DAC itself.
I have been testing this morning directly out with no output stage. and just a passive 60khz 2 pole filter. It looks the same as with the IVY. so for the rest of my testing I am just leaving the IVY in.
I don't see any significant IMD below the fundamental.
I don't have an older ES9008 eval board or I would test that.
Here is 20khz test tone.
Javin5 said:
I agree that it is very likely something about the DAC itself.
I have been testing this morning directly out with no output stage. and just a passive 60khz 2 pole filter. It looks the same as with the IVY. so for the rest of my testing I am just leaving the IVY in.
I don't see any significant IMD below the fundamental.
I don't have an older ES9008 eval board or I would test that.
Here is 20khz test tone.
Attachments
- Status
- Not open for further replies.