While sending the unit to Amir , I have done one more experiment .
We stand now at -115 THD+N on RCA output and THD was slightly improved (still around -125)
More important I was able to reduce the H5 (one of our main issue and delays) to -132db (and I am talking about bad channel , good one is at -138)
All H higher than 10 were reduced by 1 db at least . There is no harmonic over -130 (any channel)
All units will incorporate this change .
We stand now at -115 THD+N on RCA output and THD was slightly improved (still around -125)
More important I was able to reduce the H5 (one of our main issue and delays) to -132db (and I am talking about bad channel , good one is at -138)
All H higher than 10 were reduced by 1 db at least . There is no harmonic over -130 (any channel)
All units will incorporate this change .
Last edited:
Its an impressive number. Hopefully it sounds as impressively as it measures
I spend a long time on sound quality of this unit . Many PCB revisions.
My model is to test for everything and any deviation (noise) as long as I can measure its fair game (Bruno P. quote) while I will not comment if you can hear it.
Now sound quality has an subjective side. I did my best to incorporate audiophile "myths " into the unit and the result was (for most of the times) a clear reduction in H. Some changes (like RFI removal) were very slightly visible in the AP but clearly very audible.
I also incorporated work from Rasmussen / Abraxalito (IV pre filter) but I used my own schematics and tests
At last ,,, I want to clearly say that low impedance power supply is one of the important aspects of Rev . Now only we used lots of lytics on the PCB , but also we used a lot of supercaps in AVcc (Rasmussen/ Ians pioneering work) and many subjective impression point to the massive importance of having enough electrons on board (especially for Avcc and opamps)
Revolution Dac is made for audiophiles , but not against THD. Who said that great THD means low SQ ?
Maybe a Straw Man said it?
The real question, I think, is how good can extraordinary design effort make a Sabre dac sound?
Underneath it all, there is still a Sabre dac and what's going on inside the dac cannot be changed by the board designer. One can argue that any errors must be below human limits of audibility, but down the road when a better chip eventually becomes available we always find out that wasn't true.
Still, a great dac for a great price available today is something to rightfully proud of. I would like to hear Rev, but not looking for a new dac at the moment. Still hoping you sell a lot of them.
Last edited:
I found that whats called a Sabre Dac sound is only the textbook schematic implementation . The main issue is datasheet schematic and everyone that made a copy
We used a different filter scheme with cutoff frequency that was optimized. There is no glare.
Now I am not saying there are no better DAC ic. Still what I have found is that is far more important implementation than actual IC... unless you are looking for great numbers with minimum design research .
Next challenge. One of the 3 best AMPs in the world for a price thats under...hehe
I had a few ideas and next PCB I will be testing. If that works (and it should) I will post more details (but I am gunning for -106 to -110 THD from 0-20Khz) / 5-25W with a maximum power at 105W
We used a different filter scheme with cutoff frequency that was optimized. There is no glare.
Now I am not saying there are no better DAC ic. Still what I have found is that is far more important implementation than actual IC... unless you are looking for great numbers with minimum design research .
Next challenge. One of the 3 best AMPs in the world for a price thats under...hehe
I had a few ideas and next PCB I will be testing. If that works (and it should) I will post more details (but I am gunning for -106 to -110 THD from 0-20Khz) / 5-25W with a maximum power at 105W
Generally Allo hardware tests well with Amir's instrumentation, but he is definitely of the opinion that a good DAC implementation is agnostic to the power supply feeding it, and also apparently that all you need to know about the DAC's quality is in his metrics. So, in reviewing, he may not use the optimal PSU, or acknowledge that it has an effect.
On the other hand, since he started speaker testing, he has had to admit that some gear may test well, but sound mediocre (and vice versa).
Interesting. Last time I was over there (not very often), I noticed a few people had the temerity to say how something sounded and came away unscathed, without threats of banning, no ridicule, etc.
Still, I would say Amir is only just starting to learn about listening.
That's the aim though the Allo amp is class D whereas Neurochrome are class A/B.
Should be an interesting comparison.
So you think you can keep it that low in the higher frequencies as well at 25W? You would be doing better than xrk971, drmordor and the TI reference implementation have been able to do!
That would be an impressive achievement and I sincerely wish you well with it. I’m all for technological innovation and progress.
I somehow hoped that the measurements wouldn't be THIS good and the amp would be just a level above Allo Volt+ D. It doesn't look that I will be able to afford it. A budget option in between would be fantastic.