|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
208.46.240.4
I have a friend that built both my linestage and SET amps tell me there is no point to high resolution digital with so many transformers in the signal path. His point being that roll off with these transformers will set in before 20 khz negating any benefit of hi-res files. I don't think he is being inflammatory since I do enjoy 16/44.1 redbook playback a great deal on my system and he uses similar typologies as mine. These aren't budget brand transformers that we've used, in the signal path ranges from Lundahl, Hirata Tango and Tamradio.
I am curious what the Asylum thinks?
Follow Ups:
My suggestion, don't talk design philosophy with your friend, he seems the type to focus on just one issue and make a blanket conclusion based on that one issue, whether it is a major factor or not. You don't change this type of person.
Fly,
My preamp has transformers and goes -1dB at 7Hz and 93KHz. My DAC actually goes about the same but I limit it to 50KHz so the noise doesn't get into the system. My amps are good from -1dB from 11Hz-64KHz (parallel feed). Direct feed go to about 40KHz and below 10Hz -1dB.
My suggestion is either find a better builder or get a better transformer.
Thanks,
Gordon
J. Gordon Rankin
Stereophile measurements seem to disagree
The Stereophile measurements of the Cardinal with the FS-030 output transformer (series feed) show better bandwidth than Gordon claimed, but at lower power levels (as expected).
The Stereophile measurements of the Cosecant DAC are also consistent with Gordon's information regarding bandwidth.
Before trolling a manufacturer, consider clarifying the claims that you're making.
He's not a troll... he's a fly caster.
Signal from DAC goes into LL1676, then to 26 DHT output stage. Transformers block DC offset and filter DAC HF nasties. Never heard better...
I think digital audio of resolution of at least CD quality actually benefits with transformers, because tranformers knock down the RFI transmissions..... In fact, the only time I've come close to enjoying high-resolution digital audio was listening to an SET system with interstage transformers.
Output transformers for my next amp build are nano-crystalline double-c cores. Their specified at -3 db points are 5 Hz and 80 kHz in circuit, but FFT shows better than this, with a nice smooth roll-off in the highs - no "kick" in the curve.
I expect the frequency response effects of even a treated listening room will be more problematic...
Cheers,
91.
"Confusion of goals and perfection of means seems to characterise our age." Albert Einstein
Well the data sheets on my OPTs and interstages all show bandwidth out to near 80 khz, but that is not the measured performance once it is loading a tube.
Agreed. The specs I provided were for loaded. Not sure about the FFT.Cheers,
91.
"Confusion of goals and perfection of means seems to characterise our age." Albert Einstein
Edits: 10/21/15
I hear a significant increase in sound quality when playing hi-res even using contractor grade OPTs.
----------------------------------------------------------------
Big speakers and little amps blew my mind!
I got good result since I replaced my Opt to Finemet Core Transformer.
Frequency Response 25 - 50kHz
A lot of well liked DACs use IV step-up transformers, which do not have a reputation for poor bandwidth. (Likewise for MC step-up transformers)
IMO - output transformers are one of the toughest places to get high bandwidth, but being down less than 1dB at 20kHz isn't challenging, and your speakers will certainly have more roll off than the transformers in question.
As PJ mentioned, fidelity and bandwidth aren't the same thing.
Bifilar wound small signal 1:1 transformers can have bandwidths in excess of 100khz.
And like someone else said on this thread... go check out the "bottleneck" (in terms of freq response) that a speaker will often present.
MSL
Builder of MagneQuest & Peerless transformers since 1989
The point of high resolution is not baseband frequency response.
Then please tell us what the point of it is if not the highest frequency sampled.
That is, what makes it high resolution?
----------------------------------------------------------------
Big speakers and little amps blew my mind!
> > "Then please tell us what the point of it is if not the highest frequency sampled.
That is, what makes it high resolution?" < <
For clarity, sampling frequency is not the highest frequency sampled - there is usually a low-pass filter before the sampling begins. The Nyquist criteria specify a minimum sampling frequency based on assumptions about the signal's bandwidth. Using a higher sampling frequency allows more bandwidth between the highest audio frequency and the Nyquist frequency (half the sample rate). This broader bandwidth in turn allows more gentle anti-aliasing filter slopes. As far as I can tell there is no consensus between the subjective effects and the known science about how valuable that may be - but many including myself have heard improvements with higher sample rates and the associated better filters. I do not pretend to be an expert on digital audio.
The other aspect of the term "resolution" is the number of bits accuracy - typically 16 vs. 24 bits for audio. This is generally regarded as affecting the noise floor rather tan the frequency response. Again, the subjective and objective understandings are not yet in agreement, and again I do not pretend to be an expert.
I practically never go to the Digital forum but I'm sure there are more extensive discussions there, and probably many who do regard themselves as experts ... :^) We at Bottlehead have been working with John Swenson, and I have great respect for his expertise, for whatever that may be worth.
I attended digital audio training at Bang and Olufsen in the mid-'80s. Not that I remember all that much of it, but one thing I know: resolution isn't frequency response. Resolution is the value of the increments available to reproduce the signal, or in other words, bits per volt. Within this definition, analog processing provides near-infinite resolution. Digital audio doesn't even come close. Regrettably, my hearing no longer allows me to detect differences so easily. My daughter, on the other hand, can readily discern the higher fidelity of an LP vs standard CDs, even when reproduced by my lowly tube-type circuitry. :)
--------------------------
Buy Chinese. Bury freedom.
"analog processing provides near-infinite resolution"
Cassette tapes are also analog, no where near the resolution of CD.
.
Digital resolution is strictly limited by the step size, an inverse function of the number of bits. How many bits are there in a 1 volt sine wave from a cassette?EDIT: Just to be clear, there are zero bits creating the 1 volt analog sine wave. The reciprocal of that is infinity. Now, how many bits are there in a CD? Less than infinity? If so, the CD exhibits less resolution than an analog cassette.
--------------------------
Buy Chinese. Bury freedom.
Edits: 10/23/15
This doesn't make sense; a cassette copied 100 (or 1000) times) still has no bits. What will it sound like? Pure crap (my opinion of a "good" factory cassette as well)
Edits: 10/24/15
We weren't talking about sound quality, we were talking about resolution. Resolution is only one aspect of high fidelity reproduction.
--------------------------
Buy Chinese. Bury freedom.
Thanks Paul.
---------------------------------------------------------
Big speakers and little amps blew my mind!
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: