Audio by Van Alstine ABX Comparator Review, Part 3: New Twists & Conclusion

Is something broken here?

As I discussed these things with my audiophile buddy who did the listening on the two systems for the Burn In Audiophile Law article, he insisted the ABX Comparator was defective. He would hear nothing of the sort that the ABX Comparator might be doing its job perfectly, and that as a class of components amps might be different than all other components and speakers. He dismissed the idea summarily, declaring that if the human ear can hear differences between amps in systems, then if those distinctions disappear as heard through the Comparator , it is junk.

But, from my perspective, it’s not so simple. The ABX Comparator is in essence a series of smart switches, and if it doesn’t work, nothing will be heard. The ABX was given rigorous listening testing by Van Alstine to ensure that it would leave a vanishingly thin footprint on anything associated with it, and I found it to not hinder effectual retrieval of an audio component’s native sound. All other components and speakers sent through the Comparator were assessed, and I was able to pass ABX trials, so the argument that the ABX Comparator doesn’t work is weak. Furthermore, the sonic signature of components and speakers when the Comparator was removed did not change appreciably; the ABX Comparator doesn’t wreck their sound.

My contention is that amps as a class of component are tremendously beholden to level matching, and that it is all but determinative as to their sound. The reason we hear such differences as I discussed previously, which is an “every day” occurrence for the audiophile, is because we are hearing them in a state being level unmatched. I know that if I were to remove the ABX Comparator and run the rig with the Innamorata, and then swap it out for the Phi 200, I would hear distinct differences. I have for years, as I have been using both ongoing for system building and reviewing.

But that begs the question; to what would those differences be due? That is what has thrown my world into turmoil, because the ABX experience strongly suggests those differences are due to the playback not being level-matched. So, is it that simple? Can you really obtain identical performance from any amp? Despite the results, I am not ready to conclude that, and the predominant reason is that we don’t typically have an ABX Comparator stuffed into our systems! It may be so that when there’s an ABX box in the rig one can make two amps sound identical, but if there’s no ABX box, i.e. in the normal system, those amps sound different! I have not been “hearing things” for 25 plus years, and tens of thousands of audiophiles are not all wrong. Amps have characteristic sounds, and unless one is doing level matching the outcome is a moot point. When the average guy puts in a Class D amp and pulls it out to hear a Class A amp, he will hear a distinction. Is it due merely to not being level matched? I cannot argue the outcome of the testing, but aside from such testing it’s a moot point.

Further, if not level matched, I have strong preferences as to the class and brand of amp I use. I desperately wished I had a Class D amp on hand for further ABX comparison, because I have tried several Class D types and found them not as enjoyable as Class A or A/B. What would I say if I had compared via ABX testing and found them to sound identical when level matched? Suffice to say it’s a conundrum to me. I don’t wish to run a giant switch affecting every component in every system I build. So, for now, and perhaps forever, I will go on comparing amps by ear and suiting my tastes as to their qualities. If that is denial, so be it, as I have had many a glorious change to a system from amp substitutions, several times with both the Innamorata and the Phi 200.

The weirdest thing

Why should I stop with such a bizarre revelations? Why not push it all the way into the audio Twilight Zone? Ok, I will! On one hand I was unnerved by the experience of amps, even ones of different topologies, sounding identical when level matched. I also was energized, for this was a new experience that was stretching me and forcing me to work through the problem.

After I concluded the amp comparison testing, I moved on to comparison of cables. The ABX Comparator allows for easy sighted or blind comparison of power cords, interconnects and source cables. What better way, I thought, to see what’s going on here than to switch interconnects on the amps. Remember that I had been using the same Clarity Cable interconnects on both amps previously, but now I made one simple change; I used a Morrow Audio set of interconnects on the Innamorata and a pair of Silnote interconnects on the Phi 200. In the past, in non-level matched listening were I to swap interconnects on an amp I would hear distinct differences between such cables, and even between power cords. Would those differences now evaporate as heard through the ABX Comparator, even as the distinctions between amps had?

I presume this was the logic of Frank Van Alstine in regards to power cords? It would seem to hold that if an amp could not sound significantly different, how could a cable? But, if it had not been tested, how would one know? I had also learned from Mary at AVA that not much had been done in comparison of cables; I think my theory was correct about the assumptions made about cables at Audio by Van Alstine. Thus, I had to press on to determine if a simple set of interconnects might effect a detectable difference. What would be the conclusion if level matched amps didn’t sound different when using the same interconnects, but did with different interconnects?

Care to guess the outcome? No need to guess; I did the blind testing, and here are the results. In the first trial I aced all 8 tests with a score of 100%. Then, I switched the interconnects on the amps and repeated the testing. Result; 6 out of 8 for a score of 75%. Finally, I put in the most recently arrived interconnect, the TEO Audio Liquid Silver Splash, in place of the Morrow Audio. Now I would be comparing two extremely fine cables which theoretically, based on listening to them individually in non-level matched systems, would perform more closely to each other than different. Once again, the result of the Trial was a perfect 8 out of 8, 100%. Remember, this was in the blind testing mode! In some respects these were the easiest of all trials, as I found the sonic characteristics readily discernible! Sorry, cable naysayers.

What is one to conclude from such results – that interconnects are more important than amps? I’m not ready to make such a declaration, especially in light of the fact that normal listening is not done level-matched. However, I pose this question: If amps failed to distinguish themselves when level matched, what significance might be credited to cables when they easily distinguished themselves in level-matched ABX testing? The answer to me is obvious, and one I have declared for many years – cables matter tremendously!

Conclusions

As I look back on the weeks of listening and many trials, some thoughts percolate to the surface, among them what terrific fun it is to conduct blind testing in one’s room, unhindered by distractions and expectations of others. This is a carefree way (Aside from system setup!) to explore ABX testing.

It also is a great way to build confidence in one’s ability to hear and discern differences in gear; that is, if you can pass the blind tests. I assume a goodly portion of the community might struggle with ABX and would not score well. If so, then use that as a gauge for how much you should be investing in the equipment. Due to age and environmentally related hearing loss, poor room conditions, etc. poor ABX scoring might validate one’s choice to limit the investment into the system.

For me, validation of my hearing acuity was important. I thrive on my desire to hear different systems, and I believe I can tune rigs quickly, pushing them toward the sound I desire. I trust my ears, which is something every audiophile concerned about performance wishes to do. I learned that I should trust my ears most of the time. It became obvious through this experience that as fallible humans everyone will have moments of misjudgment, and no one’s opinion is to be held as absolute. It reinforced the decision I made decades ago, that I would not be anyone’s acolyte in audio, but would chart my own course according to what my ears heard. I suggest you also do not glom onto any reviewer’s opinion, yes, even mine, absolutely, but use them as guides and expect occasionally to hear things differently.

Systems are as unique as people, but as you can identify people, so also you can identify level matched systems and components. However, it’s a lot harder, perhaps more along the lines of trying to discern between groups of peoples’ voices on the phone. The differences between systems seemed to shrink significantly when hearing them level-matched. The supposed transformational, epical, monumental, and jaw-dropping changes were rendered impotent to impress, but rather I had to work at distinguishing them. Level matching seems to level the playing field so much that comparing systems and components that way is more like a tug of war between two heavyweights than a boxing match between Evander Holyfield and me.

ABX testing, I assume, forces the most work by the listener to discern differences when the components are similar. Would my scoring have been better if I compared one of the larger floor standing speakers with much better LF extension and a more sizable soundstage to one of the smallish floor standers I used? I would think so! But, if in very close systems I could score better than average across a majority of trials, then ABX testing can be passed, and people can discern between level matched components and systems.

Except for amps, and I’m still trying to gain a definitive perspective on that.

Focusing while listening, or critical listening, in assessment of gear matters a lot! I was miffed when I scored low due to overconfidence. But typically after taking a break and returning refreshed and motivated, I scored well. It seems that besides the Orientation Error, tiredness, overconfidence, and distractions all play determinant roles in poor ABX results.

The music used in assessment may play a larger role than assumed. I love guitar, as I took lessons as a child, and it strikes my ear as an instrument having a phenomenal range of expression. However, solo guitar and solo voice were tough to assess in ABX. More complex music seemed to work better for trials. This may inform a choice of music to take to a dealer or show when considering purchase of components.

If I weren’t a reviewer…

If I weren’t a reviewer I likely would buy the Audio by Van Alstine ABX Comparator. That may be the strangest endorsement of a product yet. Here is my rationale. I have to compare systems continuously without extraneous devices in them, and I’m almost always conducting reviews. Perhaps 10% of the time I get to build a rig for casual listening without reviewed components. As such, the Comparator is not ideal for me. However, as a system builder, if I were seeking components for verified improvement of my rig I would find the ABX Comparator a compelling purchase. It would help me quickly assess the proffered improvements of components and cables. I found that I was able to isolate pleasurable effects of electronics and cables quickly and consistently. If I had a device like that years ago I would have had a means to definitively select gear for my rigs.

The components I preferred with the ABX Comparator in the system I also preferred without it. So, for an audiophile who wished to demo in home a $20K preamp versus a $10K preamp, or even a $3K preamp, imagine the potential cost savings the ABX Comparator could yield. It wouldn’t be a guess, but a comparison that informed the selection. Of course, it would hurt the budget if the $20K preamp was revealed as your favorite, but then you would have opportunity to decide if the difference was enough to motivate you to purchase it. I found myself asking such questions of review pieces, and in some instances I heartily endorsed the extra dollars for the sonic benefit.

The rub is that ABX testing is as much a curiosity as an aid to system building. Once the unit is removed, the seeming huge differences return and it does seem primarily due to non-level matching. Yet, it was helpful to have chosen my favorite components in real time fashion, knowing that I was selecting a preferred performance. Winding down the last session with the Comparator, counting the 24 pages of notes accumulated over the weeks, I sat listening to Alison Krauss sing “I Don’t Know Why” and thought that there is so much about our hearing and discernment of sound in this hobby that we do not know! I don’t know why ABX testing is so doggone tricky; my theory is that our ears are exquisitely sensitive to sound, but our acoustic memory is a mess, even in the short term. I have thought this for a long time, and my experience with the ABX Comparator seems to confirm it.

Like Alison, I have a lot of questions. I can’t say exactly why I love audio so much more than video. I don’t know why I am driven to hear different sounds. I don’t know why I love the gear as much as the music. I don’t know why ABX generates such visceral reactions in audiophiles. I can relate to her sentiment, “If there were no music, I would not get through…” I love this hobby and I love the fact that artists like Alison Krauss make enjoyable music, people in the supply chain like bring it to market, industry members produce and distribute the products which bring it into our homes, and people like Richard Van Alstine make devices to explore the peripheral, or to some, elemental, mysteries of audiophilia. Kudos to Richard for the ABX Comparator; long may it serve the curious and committed among us to better their systems and their zeal for listening!

Addendum, April 20, 2015: In a conversation with Frank just before this article was sent off, I discussed with him my findings regarding the ABX Comparator and amps. As I had reached my conclusions in the article and he had not seen the finished article, I have not gone back to edit the article, but instead add the discussion here.

Frank gave insight about the uniformity of amplifier sound, noting that in his shop he uses two very different speakers, through which differences in amps can be heard. However, he noted that the differences are very subtle, and the speakers he uses are quite different; Philharmonic Audio’s Philharmonic 3 and Janzen zA2.1 electrostatic speakers. Recall that I commented that it would be expected for different technology and/or sized speakers to sound obviously different. This would shed light on my findings, that with nearly identical speakers amps could not be distinguished, but with Frank’s obviously different speakers they could.

I find this additional information to dovetail with my article above. Indeed, my revised conclusion is that it takes an exceptional difference in speakers to hear the subtleties of different amps. Perhaps I should have set up the Kingsound King III Electrostatic speakers in comparison to the Vapor Audio Nimbus White? Would I have heard the difference between the level matched amps? Perhaps so! But, then I think the distinctions between all other ABX testing would have become more like child’s play, hardly effort at all. Perhaps in all other testing I would have scored so well that the results would be suspected as fabricated. We will never know, but I’m content the way the tests were conducted.

So, perhaps I should have softened my declaration that all amps through the ABX Comparator are identical sounding. They were the same sounding with similar speakers, and perhaps it’s good that the review went that course, for it might inform audiophiles who wish to transform their listening experience to select a significantly different speaker for maximum change to their system, rather than merely another similar speaker. It might also imply that it would be more difficult to transform a system through change of an amp than a speaker.