Sangean HDT-1 Review

[Tom Wells]

Let's repeat that test, but a little differently. Say you can REALLY hear aac+ processing on a 96kbps HD broadcast? PROVE IT! Before are four files. All are uncompressed .wav. Two of them are straight cd rips from Working in a Coal Mine, cut 1 on Harry Connick Jr's new album "Oh My NOLA". One of them was encoded in AAC+ at 96kbps, then decoded back to uncompressed .wav. The other was encoded in AAC+ at 48kbps, then decoded back to uncompressed .wav (simulating an HD2 stream). Encoding was done with MediaCoder (I don't have aac+ on my workstation program...Adobe Audition).

If you really can tell such a huge degradation from aac+ coding on HD radio, PROVE IT! Tell me which two of these files are uncompressed cd rips, which one was compressed to aac+ at 96kbps, and which was compressed at 48kbps. If you don't get all four then you've proven one thing...you're only guessing! Science...what a concept!

I'm at the airport now, and can't wait to get home and try with headphones.

But do us a favor, and find some music where you also have the original on viynl, so we can start with a source which originally was not digital to begin with. Even a pure, single note at 22,050 hz, the theorectical upper limit of the CD medium, will not be reproduced properly on a CD, since phase-timing may mean that we can measure the waveform only twice in one cycle, and if this falls on zero crossing, there is no reproduction of the note. If the phase timing falls on sine peaks, the two points measured then reproduce as a sawtooth wave, or the reconstruction of the wave must assume sine waves. If the original source as non-sinusoidal, as most speech and music is, we get phase distortion.

I suspect the above recording will sound like the previous WDAV clip, good, but with super-fine-tooth comb effects strangely accentuating
material near the upper-end resolution of the sampling.


If not Harry C, perhaps something

 

[hipporadio]

I just found on Napster (the legal one!) four different stereo versions of "I saw her again"... All imaged boy singers left, girl singers right. Interestingly enough, that's reversed from WNMB. Obviously WNMB's channels got "flipped" somewhere.

Mike... I’m happy you’re helping to keep Napster legally in business! THANK YOU for helping me discover that one of the Denon CD players in my studio was channel-reversed for years. I had posted the opposite observation... You are correct... In “I Saw Here Again”, it’s “boys—left & Mama—right”.

With all due respect, Mike—the issue here is not the quality of good ‘ole-fashioned AM-stereo as presented on a receiver designed specifically for that function. Those of us who haven’t taken the HD Accurian plunge are discussing the merits of the claim that it and the new Sangean HDT-1 do double-duty as a credible C-QUAM AM-stereo receiver. Neither am I debating FM IBOC bit-rates or am interested in the 96k vs. 48k vs. .wav file shell game you’re inviting us to play with audio files on your website (we’ve already been down that road). Many here (including me) are repeatedly on record surmising that FM HD audio quality is not the issue. Refer to my own comment on this tuner’s review:

So here’s the Sangean HDT-1 scorecard as I decode this review: Very good digital FM

...And IBOC FM HD audio quality in general from my post on the FMExtra topic:

I am prepared to concede: The current [96k] bandwidth... [HD FM] CAN provide a fulfilling experience for most in the marketplace.

The initial question remains, and was posed by SUPERCASTER—and answered by audiophile and R.F. Burns (catchy “handle” BTW!)

So Sangean stopped providing (analog?) AM [C-QUAM] stereo reception on the HDT-1 tuner?

For the record, I know for fact that the two of us that bought the Accurian on black Friday, neither one receives C-QUAM AM stereo under real world conditions with plenty of signal.

I can say that the radio does do C-QUAM, but the signal needed to activate the C-QUAM detector must be much higher than what it took to say make my Sony AM stereo radios to open up.

Many weeks ago, speculation began flying here that the RS HD Accurian was even more of a “miracle machine” because of its hidden ability to resurrect and reproduce C-QUAM AM-stereo. A brief aircheck of C-QUAM station WNMB was compressed to an mp3 and made accessible from a link posted here. Frankly, it was very unimpressive...

The only music included was the Mamas & Papas’ “I Saw Her Again”... The demodulated audio was NOT compelling... Nearly NO content over about 2.5k [YUCK]... And I was not convinced that it’s even “real” (hard left/right) stereo. I’m wondering if it’s merely an interesting “image” created by a confused DSP. It sounds like a low-bandwidth version of synthesized stereo from the infamous Orban box of the 1980s used by early-adopters of the stereo television technology. So... Does HD radio REALLY offer genuine Motorola C-QUAM capability?

rbrucecarter5 inquired about a comparison of the streamed file to a genuine stereo copy of that recording:

If you can post that recording of "I Saw Her Again" - I can compare it with the original stereo version on vinyl to see if the separation is really there.

I responded after comparing the streamed aircheck with a CD STEREO copy of the song (off the “Millennium Collection”)...

The WNMB aircheck presents a surround-effect, but NOT “hard” left/right separation. My CD copy of this song features Mama Cass Elliot’s vocal “hard left”... [note correction] —NOT SO on the WNMB sample. Understand that we have no darn idea what WNMB’s source is for the song—it may be “synthesized stereo”... IF their copy IS the post-1999 remaster—the HD Accurian IS NOT properly decoding C-QUAM AM stereo.

Replay A/V was used to capture the WNMB stream from the provided link. I instructed it to “record” to a stereo 44.1khz 16-bit PCM .wav file for maximum integrity. Although plenty of “spatial effect” was noticeable—the playback was NOT in DISCRETE STEREO. Does your file differ from the one posted earlier... Was the stream possibly in mp3Pro format which is inactive in my current version of WinAmp? Could you repost the source of this stream, or better yet—arrange with your contact there to provide a new and more convincing example. If HD AM radios (even select models) receive and faithfully reproduce GENUINE C-QUAM AM-stereo—I’m sure many would like to acknowledge and give these radios their due!

 

[Mike Walker]

Tom, with all due respect, unless you're 17 years old, or a girl, you can't hear 20khz anyhow. Get over it.

As for starting with something on vinyl...hey that's a great idea. Nothing gives scientific creedance to a test like compounding the noise and distortion of a vinyl recording, with the artifacts of digital. Good call! Then we get the best of both worlds! I mean the worst.

The idea that vinyl is some kind of purity test really pisses me off. Talk about delusional! The linear velocity of vinyl is very high on the first song on a side. It makes for teriffic quality. Too bad linear velocity steadily falls as the stylus tracks toward the center, to the point where distortion is orders of magnitude higher on the last cut than the first. Just where music traditionally reaches it's loudest climax, with the most heavily modulated grooves, the medium is least capable of handling such densely packed information. Something's gotta' give, and it does. Mastering engineers apply compression, limiting, and equalization specifically to make up for the limitations of vinyl record tracking. This is why people hear "more air and ambience...more of the trailing edges fading toward silence" on analog recordings. DUH! The dynamic range has been reduced from 96db to less than 60. You hear low level detail more clearly, because it's been compressed....IT'S FREAKING LOUDER! Look at a VU meter on the same piece of acoustic music on vinyl and cd. Adjust it so that the loudest passages are comparable, and watch as the music fades away in quiet passages. SEE HOW MUCH QUIETER THE CD GETS? Low level detail on digital recordings is actually (gasp) LOW IN LEVEL!

I'm tired of the "analog is better than digital" crap. You flat-earthers lost. Decades ago. Time to move on!

(And NO I don't hate analog. I have a turntable, and still listen to vinyl. It sounds quite nice. But I never fool myself into thinking it's more accurate. That's both objectively and subjectively so easy to prove wrong.

 

[Tom Wells]

Mike, I have spent 15 minutes carefully comparing these clips, I can honestly say the difference is infinitesimal.
You have done a much better job of encoding than any of the results I've heard from HD FM in my limited experience.
I would hope HD FM can sound as good as what you've offered, but none I've heard had the "air" that these samples have, nor
are they as natural in the sibilants.

The sampling rate sampler reads to my ears as:

No. 1 uncompressed CD rip.

No. 2 48K compression

No. 3 96K compression

No. 4 uncompressed CD rip.

I submit that if you offer a similar selection of 4 with the same encoding chioces, the result would be more readily apparent
if you began with a high quality viynl source, where the upper frequency resolution does not suffer phase distortion.

If the source material had no info above 10khz, the various schemes would be absolutely indistinguishable.

update.. just read the last post and at 45 still do hear the FM 19 kc, occaisional old-fashioned burglar alarms at who knows what, and the TV
horizontal oscillator still give me headaches to the point that I built our TV into a sound-proof wall enclosure.

And when I listen to viynl, it is first deeply cleaned, as it's usually dirty, and I have click removal, and variable cut hiss removal (DNR), and
use moderate expansion. I understand your points about compression, groove velocity and equalization, and still prefer the higher resolution
of continuous measurement (analog).

I admit most people will not be able to get the same results without similar measures.
I also admit I have no way at present to post a link or file to offer a sample of how good viynl can be.

I am sorry that any aspect of this pisses you off, and hope that this issue only stems from the fact that most people can't hear the difference.

I am truly not a flat earth person, but I do have a good deal more respect for the laws of physics than any computer, processing scheme, or codec.

And I find radio to be an art of physics, as is music.

Computers, while wonderful, are still not the medium best suited to store what starts out as analog, and will be heard by eardrums in analog.

It is merely an extra step of complication rife with opportunities for new types of distortion.

Please explain how you can properly define the nature of a wave form when it is only sampled twice per cycle.
In a CD, this is 22.05Khz. Would we assume that wave is always a pure sine wave? What if it's not?
This is why all CDs sound as though they were recorded through the same microphone, but a very, very good one.

 

[audiophile.]

Lets start off by saying the the four-clip audio test by Mike Walker IS A TRICK!

THERE ARE NO DIFFERENCES IN THE FOUR FILES. THEY'RE EXACTLY THE SAME! :

First I listened to all four with speakers, then headphones, then I put them in Cool Edit studied them a bit, and then ran a custom program that did a binary comparison. Result: Exactly the same! (It's no wonder Tom Wells said they're some of best AAC he has ever heard!)

Now that said, I will give Mike Walker the benefit of the doubt that this is an honest mistake. For instance unless you a compressed close a file in Cool Edit / Audition it continues using the cached uncompressed copy.

Either way you can't fool me!

 

[Mike Walker]

I don't mean to be too fussy Tom, but please explain how you can properly sample a waveform with a moving rock that will ALWAYS overshoot (continue moving once the signal is removed...because all things in motion want to continue moving), is tracked on a pivoted arm despite the fact that the cutting stylus was in a tangential arm, and is in a groove of constantly falling linear velocity? There's far more random phase information generated by the moving system and resonances in a phono cartridge than a digital system (and of course linear PCM has it's limits). Say only the left groove wall is modulated. The stylus in theory will move only the magnet (or coil) on the left side. In reality the magnet will vibrate, but so will the stylus' suspension, the structure it's mounted to, in turn the entire cartridge body and tonearm will vibrate, transferring these now delayed (hence phase-shifted) vibrations back into both the magnetic assembly on both channels (resulting in crosstalk AND phase distortion) as well as the stylus, where they make the rounds again (if bad enough resulting in feedback).

The moving system in a phono cartridge, even a very good one, is no marvel of phase integrity. Having said all that, what's miraculous is that it sounds as good as it does. Often VERY good. Extremely fun to listen to, but it shouldn't be anyone's reference. It's too easy to point to dozens of measurable and audible problems with the system.

 

[Mike Walker]

You're absolutely right audiophile. It was a trick. That others DID hear "differences" proves the point I've been trying to make repeatedly. The only meaningful tests are double-blind tests. Otherwise people claim to hear differences where none exist.

The files were IDENTICAL. I lied about the process involved in creating them, but hopefully you can see why I did it.

Good job Audiophile!

 

[hipporadio]

Mike, I have spent 15 minutes carefully comparing these clips, I can honestly say the difference is infinitesimal.

Tom... Obviously, you are a very polite and accommodating man and you have “taken the bait”. Over the years, I have learned that good engineers have a high threshold for “proof”—many times it is absolute. No person (including Mike) should be assumed “dubious”, BUT we have NO DIRECT knowledge of the circumstances under which he procures and produces these audio files he has come accustomed to post. This is the third time he has either promoted or posted files to advance his position here.

Let me contrast my own situation. I have a very nice home studio with no shortage of professional “toys” and a mere 10-minute drive to unfettered access on a T-1 circuit, yet I’ve never posted a “here’s-proof file” with the expectation that I should invite zero scrutiny.

Mike began this a month back with a demo of the quality of several neighboring HD stations he could receive on his new HD radio with only an indoor antenna at his home in Wilkes County, NC. Several of those files were compelling, and I had no reason to question their authenticity. I acknowledged his effort and gave HD radio its due, but one instance aroused reasonable suspicion—an off-air example of the fragile HD-2 service from WLYT whose TX is southwest of Lake Norman near Charlotte. 102.9’s HD-2 IBOC carrier needs to travel over 50-miles of very difficult terrain to reach his radio. TWO engineers with HD experience I know gave me a slanted grin over this assertion. I spent six months in Charlotte in 1999, and three I know in radio there report unreliable reception at their nearby homes in Dillworth, Matthews, and Tega Cay. A couple weeks later, Mike advanced the rumor that his beloved RS Accurian could competently reproduce C-QUAM AM-stereo and promoted a link to an online audio file. In this very thread, you can see this fable unfolding.[I read audiophile's post at this point] Now the results of his latest adventure in online file-sharing...

Tell me which two of these files are uncompressed cd rips, which one was compressed to aac+ at 96kbps, and which was compressed at 48kbps.

Neither am I debating FM IBOC bit-rates or am interested in the 96k vs. 48k vs. .wav file shell game you’re inviting us to play with audio files on your website (we’ve already been down that road).

Lets start off by saying the the four-clip audio test by Mike Walker IS A TRICK! THERE ARE NO DIFFERENCES IN THE FOUR FILES. THEY'RE EXACTLY THE SAME!

The files were IDENTICAL. I lied about the process involved in creating them, but hopefully you can see why I did it.

I’m sorry Mike, but I CAN’T understand your use of intentional trickery in a forum generally composed of adult professionals with industry experience, history, and well-grounded opinions (whether you choose to accept them or not). This ISN’T a room dominated by jealous and embittered DJs—I would hope to find a standard here that surpasses that so often demonstrated elsewhere on this site. You have done your IBOC brethren no service.

* I began composing this post earlier this morning (before audiophile's "discovery")—often taking a break for other chores and events... I decided to post it anyway.

 

[Chuck]

Mike,

I think you have done yourself and the Pro-IBOC folks a great disservice. I never took your latest bait. My life is just too busy for arguments about how many angels can dance on the head of a pin. On the other hand, I did listen to your clips of alleged HD radio reception at your home. I was impressed.

I hope they were the real thing, but now I seriously doubt your findings, or at least I question the methods used to acquire them. Other than taking your word for it, I have no way of verifying that the recordings were made in the deep fringe areas of the stations who's signals you posted. For all I know, the recordings were made next to the transmitter site, or perhaps, directly off the Board at the stations in question. They may have even originated in your home studio. As I take it, radio production is what you do for a living, and doing some dummy samples should not be very hard for you to do.

In any case, you've blown your creditability with me. It will take quite a bit for you to earn it back. I think that is too bad.

 

[Mike Walker]

When I perpetrated a hoax, ONCE to prove a point, I told you that I had, and why. It would have been so much easier to let others continue to extoll the "differences" they heard. THAT WAS MY POINT. Which was proven. Subjective listening tests under uncontrolled conditions are BULL.

Hell, I could take my video camera and show you the face of my Accurian receiving a station 80 miles away in HD, then with it running walk outside and show that I'm in the middle of friggin' nowhere (the cows in the pasture next door might be a giveaway). But I have the flu, and it's 25 degrees outside. So that won't happen today.

Know how you can trust me? The only time I ever did anything that wasn't above board, I told you so. I have a conscience. Otherwise, what was my incentive to tell you I had posted fake files? That I was anxious to make myself look bad? Get a clue! It was a demonstration. Some, particularly those who "heard differences" are pissed. GOOD! That was the point.

 

[audiophile.]

Only one person took the bait, but even that was tempered with several comments that essentially that it didn't add up and there was no difference:

Tom Wrote: Mike, I have spent 15 minutes carefully comparing these clips, I can honestly say the difference is infinitesimal.
You have done a much better job of encoding than any of the results I've heard from HD FM in my limited experience.
I would hope HD FM can sound as good as what you've offered, but none I've heard had the "air" that these samples have, nor are they as natural in the sibilants.

Bottom line, the shell game is over! Sorry to say this Mike, but you lose...

PS infinitesimal is: an infinitely small number.

 

[Tom Wells]

I will ascribe any difference I strained to discern in the highest frequencies to the following effect.
In the same way the wagon wheel in the movie sometimes spins forward and smetimes backward, the phase of the sample to the event
makes a difference at some particular speed ( or frequency ).

I the resolution is low enough, we can't tell which way the wheel is turning.
4 movies taken at the same moment, unless synchonised, show "wagon wheels" with varying spin "distortion".

This is then, still a battling codec phase issue which accounts for any audible difference.
I would noy expect a wave form analyzer to find the difference as it likely samples in similar resolution.

Please, find one of your best audiophile half-speed mastered 33s with low productions, something we all have heard with headphones,
and make a true sampler with 5 or 6 different schemes including one "transcription" quality large file.

I must ask, Mike, if in fact you checked your own owrk and satisfied youself that the files were indeed different enough to be proven
as coming different event dubs if we were to check for different checksums.

Or if you just submitted the same file 4 times, I am dissapointed in your debating tatics.

A open ended survey regarding the 4 would still have me reporting no significant difference.

Exactly as there should be, it seems.
Didn't we get into this mess by asking faulty questions and providing a slanted answer?

 

[Mike Walker]

Tom...I also posted the files on another board (not from this site). It was there that folks went off on "this one sounds better", etc. I finally felt guilty, and called it off...which is why I called it off on Radio-Info as well.

I was trying to prove a point, but in retrospect could have maybe found a better way to do it.

 

[700WLW]

Tom...I also posted the files on another board (not from this site). It was there that folks went off on "this one sounds better", etc. I finally felt guilty, and called it off...which is why I called it off on Radio-Info as well.

I was trying to prove a point, but in retrospect could have maybe found a better way to do it.

Just, as with HD Radio - too-little-too-late.

 

[Chuck]

When I perpetrated a hoax, ONCE to prove a point, I told you that I had, and why. It would have been so much easier to let others continue to extoll the "differences" they heard. THAT WAS MY POINT. Which was proven. Subjective listening tests under uncontrolled conditions are BULL.

I'll be the first to acknowledge that a lot of audiophiles go way over the top with their claims of superior audio performance. Not every “golden ear” really is. There is quite an industry selling folks "high definition cables" and the like. In the 35+ years I spent in the audio biz, I can tell you that people hear what they want to hear. If you spend $300 on a speaker cable, you will probably be convinced that it sounds better than 14 gauge lamp cord, whether it does or not. Your “joke” reinforces that, but it did it quite an expense. A fake is a fake, and when you do that, it makes everyone wonder whether or not your other claims are equally bogus.

 

[Mike Walker]

I didn't have time to explain this in my own words. Fortunately I was but a google search away from an explanation that is better than mine would have been anyway.

Here's why it's just a myth that 44.1khz sampled digital audio can't properly reproduce a 20khz sine wave. Anyone with Adobe Audition, an oscilloscope, cd burner, and cd player can prove this as well. Here's how it works, proof that it does, and information about how anyone with the above tools can prove it for themselves.

Science, what a concept!

http://www.audioholics.com/techtips/specsformats/DigitalAudio.php

 

[Mike Walker]

Chuck, with all due respect, it was an experiment, not a "fake". The files WERE fake, but the demonstration itself would have been a "fake" only if I'd left the impression that some of the files had been compressed. I didn't.

It WAS sneaky, and I can see how you would be pissed. But before becoming too critical, remember who it is here who actually goes to the trouble of posting recordings, files, and documentation demonstrating the validity of their opinions, not just links to other people's "me too" posts. I have literally spent hours preparing material to this board (and others I participate in), the goal always being to ask for, and present EVIDENCE to support my conclusions and point of view. Have those who criticize done as much?

Like it or not, asking people to point out difference where there were none, and then watching them point out those differences is a GREAT way to point out that the emperor (those who claim to hear differences where none exist) is quite naked, and shivering from the cold! (see this link) http://www.avsforum.com/avs-vb/showthread.php?t=799176

I re-assert...people can't make scientifically valid listening comparisons when they know what they're comparing! This is why most subjective tests are pure b.s.! If you know you're comparing, say, an NAD receiver with a separate vacuum tube preamp/amp/tuner from Cary Audio and Magnum Dynalab, and you stongly believe that tubes sound better, and that more expensive components sound better, then it's highly likely that you will choose the Cary/MD rig as "best" even if the output of the two systems is electrically identical. The proper way to conduct such a comparison would be with a device called an a/b/x comparitor, a switchbox with three switches. Switch a would be one system, switch b the other, and switch x would be the same as EITHER a or b. Those involved in the listening test would NEVER be asked "which sounds better?" Instead they'd be asked whether switch "x" was the same as "a" or "b"...because if they couldn't reliably tell (beyond what statistically would be mere guesswork), then it doesn't matter which they perceive to be "better"...they've proven they actually heard no difference.

If you were me, and believed that people frequently hear "differences" where none exist, can you think of a better way to demonstrate this on the internet than the one I chose?

 

[700WLW]

Chuck, with all due respect, it was an experiment, not a "fake". The files WERE fake, but the demonstration itself would have been a "fake" only if I'd left the impression that some of the files had been compressed. I didn't.

It WAS sneaky, and I can see how you would be pissed. But before becoming too critical, remember who it is here who actually goes to the trouble of posting recordings, files, and documentation demonstrating the validity of their opinions, not just links to other people's "me too" posts. I have literally spent hours preparing material to this board (and others I participate in), the goal always being to ask for, and present EVIDENCE to support my conclusions and point of view. Have those who criticize done as much?

Like it or not, asking people to point out difference where there were none, and then watching them point out those differences is a GREAT way to point out that the emperor (those who claim to hear differences where none exist) is quite naked, and shivering from the cold! (see this link) http://www.avsforum.com/avs-vb/showthread.php?t=799176

I re-assert...people can't make scientifically valid listening comparisons when they know what they're comparing! This is why most subjective tests are pure b.s.! If you know you're comparing, say, an NAD receiver with a separate vacuum tube preamp/amp/tuner from Cary Audio and Magnum Dynalab, and you stongly believe that tubes sound better, and that more expensive components sound better, then it's highly likely that you will choose the Cary/MD rig as "best" even if the output of the two systems is electrically identical. The proper way to conduct such a comparison would be with a device called an a/b/x comparitor, a switchbox with three switches. Switch a would be one system, switch b the other, and switch x would be the same as EITHER a or b. Those involved in the listening test would NEVER be asked "which sounds better?" Instead they'd be asked whether switch "x" was the same as "a" or "b"...because if they couldn't reliably tell (beyond what statistically would be mere guesswork), then it doesn't matter which they perceive to be "better"...they've proven they actually heard no difference.

If you were me, and believed that people frequently hear "differences" where none exist, can you think of a better way to demonstrate this on the internet than the one I chose?

You have lost all credibility in this forum - as with HD Radio, too-little-too-late.

 

[hipporadio]

Know how you can trust me? ...I told you so. I have a conscience. Otherwise, what was my incentive to tell you I had posted fake files? That I was anxious to make myself look bad? Get a clue!

Clue well taken, Mike... I suspect that you just got caught (by audiophile), and your mission was instantly “undressed” and foiled.

Had I chosen to participate (without a bit or checksum analyzer) I would have first simply compared THE SIZE of the allegedly-differing files—but I’m sure you took care of that one during your pursuit. Did you re-convert the smaller aacPlus files back to .wav to camouflage your activities? If so—the “science” of your “innocent test” instantly became doubly-defective... Oh, I forgot—you believe that once in the digital realm—no material can ever “go wrong” or become colorized... There’s no truth like digital truth—right Mike?

NO... In reality it was none other than Tom Wells who without the assistance of Cool Edit or a “bit scrubber” passed your test with flying colors...

I have spent 15 minutes carefully comparing these clips, I can honestly say the difference is infinitesimal.

God gave us analog ears—but he also gave (some) a very intuitive mind!

I suggest that you leave these “double-blind tests” to the involved professionals such as those at Coding Technology who have a concern far more altruistic than yours.

 

[Chuck]

Chuck, with all due respect, it was an experiment, not a "fake". The files WERE fake, but the demonstration itself would have been a "fake" only if I'd left the impression that some of the files had been compressed. I didn't.

<snip>

It WAS sneaky, and I can see how you would be pissed. If you were me, and believed that people frequently hear "differences" where none exist, can you think of a better way to demonstrate this on the internet than the one I chose?

I wouldn't do it at other people's expense. "Fool me once, shame on you. Fool me twice, shame on me."