Bill,
Okay, I see the claim about pink noise as an imaging test, which is something I could try.
Regarding LEDR, I am skeptical that that particular illusion would work for me. I have previously tried listening to simulated vertical, overhead, and behind effects played through stereo speakers and through headphones. It never sounded like motion to me, just sounded like varying EQ and an phase stuff, but always coming from the speakers, or stationary inside my head with headphones.
Okay, I see the claim about pink noise as an imaging test, which is something I could try.
Regarding LEDR, I am skeptical that that particular illusion would work for me. I have previously tried listening to simulated vertical, overhead, and behind effects played through stereo speakers and through headphones. It never sounded like motion to me, just sounded like varying EQ and an phase stuff, but always coming from the speakers, or stationary inside my head with headphones.
Is that Dork Audio? I hope you noticed the antenna symbol rather than ground used on the label.
Well spotted! I had to look a second time.
In principle, everything must be measurable. Still need to know what and how.
In this matter, I believe the need is to measure the accuracy of the levels of the first bits. A long work. Do we have to fear any dynamic recovery problems ?
That might be true about the chip but... The dongle, it's the stupidest blitheringly **** slap in the face to the late Steve Jobs as you could possibly make. I can't imagine a more repugnant, anti-apple ethics, thing to do. It's the equivalent of making a Tesla car that has to tow a trailer that's the battery it won't run without.
One of the DACs that loses low-level reverb tails is DAC-3. Measurements performed at Stereophile show ENOB = 21-bits. There are no missing bits as measured however Sterophile does it, and 21-bits is pretty much SOA.
It seems from my experiments with adjustments to dac registers and or AK4137 DSD sample rate, that reverb tail loss is linked to I2S/DSD clock jitter and or dac VCO jitter as function of DPLL bandwidth, at least for S-D dacs of the type most of the tested dacs are. Also, Allo Katana RPi hat DAC reproduces the tails pretty well. Interestingly, the best reverb tail reproduction of dacs included in the listening tests came from a TDA1387T based dac named phiDAC by its designer, our own diyaudio member Abraxalito (Richard). It was clocked by a Chinese CM6631A USB to I2S board. The dac chip itself is obsolete, but Abraxalito has some in stock. He is offering full kits for assembling 10 dacs for $50. So, a nominally $5 dac won that particular competition.
Last edited:
None of those surround soundy things work for me. I mean they sound ... interesting, but fake.
Fake is OK but do they achieve the desired result, the Star Trek franchise makes big use of them. Even on a cheap Kindle in the dark, above, behind, all work.
It's hard to describe - I can sort of hear what it's supposed to do, but the illusion collapses to 2 speakers as soon as I try and listen to it.
I assume it's that something in my brain knows it's a fake, and won't play ball. A bit like, once you have seen past an optical illusion, you can't see the illusion any more.
I assume it's that something in my brain knows it's a fake, and won't play ball. A bit like, once you have seen past an optical illusion, you can't see the illusion any more.
Last edited:
someone here mentioned a DAC that can be had as just pcb all the way up to completed unit without case.
I cant find it ... which was it?
It had some characteristics that should help with low level details like reverb/room sound and decay.
The lack of low level detail of DAC3 might be related to a dynamic range shift upward with the bits to achieve their greater head room at peak levels.
-RNM
I cant find it ... which was it?
It had some characteristics that should help with low level details like reverb/room sound and decay.
The lack of low level detail of DAC3 might be related to a dynamic range shift upward with the bits to achieve their greater head room at peak levels.
-RNM
Last edited:
Is there any reason why this is not level linearity? If you have 9 technically well behaving ones, and the 10th one expose more "reverb tails" I would guess it is because it's compressing (as appose to expanding or linear).
//
Last edited:
Measurements of the same model dac, Benchmark DAC-3, was described on the measurements of page of the Stereophile review as follows:
"Benchmark's DAC3 HGC offers state-of-the-art measured performance. All I can say is "Wow!"—John Atkinson"
The measurements can be found at:
Benchmark DAC3 HGC D/A preamplifier-headphone amplifier Measurements | Stereophile.com
Your understanding is correct, it does not have linearity problems. It also does not reproduce audible low level reverb tails. Clearly the problem is not due to linearity. You appear to me to be thinking like an analog guy, no offense intended, understanding analog is a good thing. However, with S-D dacs it might help to try thinking like a DSP guy.
- Status
- Not open for further replies.