[Andrea]

Btw, the AD8058 certainly doesn't have a lot of output current. No more than around 35 mA, you see it from the "output swing vs. load resistance" graph; that's why I inquired about how full it sounded. I think I'll go with the LMH6654, that can output far more current, while maintaining a really low distortion (mostly 2nd order, too). Fortunately the 5 uA input bias current isn't a problem - I'm getting a less than 3 mV DC offset... congrats to Xenos

 

[Andrea]

Don't know what's happened...

The same record by Leonard Cohen, now sounded wonderful through the Xenos and its newest chips. No more slight issues with the tonality - on the contrary, I thought I heard one of the finest combinations of resolution & refinement and naturalness. Leonard's voice sounded so real - very textured & visceral.

The AD8066 may be a keeper. I already know that the LMH6654's are.

 

[Filburt]

How are you calculating it from that chart? Just using ohms' law or something? I'm not sure that will give the correct results. At least, if I were to apply it that way, I seem to get the wrong rating on some other chips that do specify their output current. Maybe I'm calculating it incorrectly.

Even without the more robust PINT power supply and using it in one of my test circuits, I didn't have much trouble with the 8058 amplifying headphones. I found it to have a rather full sound, honestly. I particularly appreciate that it gains some of the definition and punch of the 8397 versus the 6172.

I don't have the LMH6655 so I can't compare it, but I have used the LMH6622 which I didn't consider as good as the 8058.

I think some of the differences here may be due to differences in test methodology. It has become rather clear to me that we don't share that. In some cases, I've found some of the things you've suggested such as perhaps the 8397 being more dynamic in some ways than the 6172. However, some of your comments I've found I cannot replicate. We do use rather different headphones as well, though. I don't particularly like the HD650 so I don't keep a pair around to eliminate that variable, for instance.

As far as the 8065 goes, I found it was one of the lesser op-amps out of the 30 or so single-channel op-amps I tried in terms of resolution and definition, but it was lively enough I suppose. I didn't really appreciate that it seemed to lack a full presentation of the sound in terms of resolving the full audio band in a defined fashion. I tried using it as a direct driver as well as behind another op-amp, and this was a problem in both cases, so I'm tempted to blame the chip. If you like it, though, more power to you I guess.

 

[Andrea]

Yes, ohm's law. I know that applied to other op-amps it doesn't always give the specified data - it often gives more

And if the output current was a strong point of the chip it'd be masochist not to specify it... At least in the normal flow of earthly things

I think that the AD8066 is detailed all round, from low lows to high highs. Perhaps I sense some incoherence in the upper midrange, a little excess of detail / lack of smoothness there. Keep in mind that I've only used the HD485 so far; not the most "monitor" 'phone in the world, but still unexpectedly revealing - particularly tonality wise, one thing that makes Senns so dear to me

It's not really the AD8066's resolution what fails to convince me - it's, just a little, the tonality. Not the most 'varied' in sonic colors that I've heard.


Keep in mind that the Xenos is a particular design. The output driver op-amps are not in the feedback loop of the preamplifier op-amp. The output op-amps are working at a gain of 2, and their bandwidth is limited by a capacitor in parallel with the 1K series feedback resistor; so is the bandwidth of the preamplifier op-amp. I think that this pretty much optimizes the musical performance of all the op-amps used. Or, my ears think so!


I don't know the LMH6622, but the LMH6654 is certainly higher performance, more music-friendly in the tonality, and marginally, higher current, than the LMH6642 (which Xin offers for his latest Supermini, btw). Pity only that it's not rail to rail - oh well, I use it @ 12V.

 

[Andrea]

Just looked at the datasheet of the LMH6622. It would seem at least as good as the LMH6654, save, potentially, the output current. Pity that it's only DUAL

Where exactly do you think the AD8058 improves on the LMH6622? Which would you call "a better buffer" for driving headphones, employed at close to unity gain?


I'm so insistent 'cause I have the AD8057's ready & hot in a drawer

 

[Andrea]

I may run the risk and go with the AD8057's. Mostly for the a little higher output voltage on 150 ohms and more.

 

[Andrea]

I compared the datasheets more accurately, and I believe that the LMH6654 comes nearer to being a "perfect audio buffer" than the LMH6622. And it's single, thus it doesn't suffer from (thermal, also) crosstalk distortion & power imbalances. Oh, you also see that the LMH6654 can output up to around 150 mA to the load. It doesn't have the widest output voltage swing in the chip realm, but you can't have everything, can you?

 

[Filburt]

Well, when I first got both chips, they sort of reminded me of eachother (8058 and 6622). However, I came to find that the 6622 ended up sounding more congested than the 8058 which has more space between instruments and more distinct presentation of each note. The 6622 is supposed to actually be higher performance than the 6654, but maybe National Semi is just blowing smoke; I don't know.

Regarding power, I haven't had much trouble with that thus far. It hasn't clipped for me or gotten a sound to it like it can't drive my headphones. I was concerned about the omission of the output current, too, but I'm finding it sounds as though it is driving my headphones better than the 6172 @ 9V, so I'm thinking it could have 30-35mA if not more. I've seen AD spec as low as 10mA (or maybe it was lower) on an op-amp, though, and not much above that on a few other op-amps, so I'm not sure it being low current would be a reason they'd omit it. Maybe it's just an omission by error on the part of the person making the datasheet *shrug*.

I wonder if perhaps it would be better to replace the AD8066 with the AD8058 than put 8057s in place of the 6654? *shrug*

 

[Andrea]

Thanks... you have convinced me.

If I mod my second 0HA-REP (which hasn't even arrived yet

), it'll almost surely be with the AD8057's. I'm very curious to hear them.


The LMH6654, anyway, has higher slew rate & lower settling time than the LMH6622. It also seems to have a prevalence of 2nd harmonic distortion from 100 KHz downwards, while the 6622 there is more...suspect. From my early listening, it's also evident that it has power and dynamics in abundance.

 

[Filburt]

Quote:

Originally Posted by Andrea Thanks... you have convinced me. If I mod my second 0HA-REP (which hasn't even arrived yet ), it'll almost surely be with the AD8057's. I'm very curious to hear them. The LMH6654, anyway, has higher slew rate & lower settling time than the LMH6622. It also seems to have a prevalence of 2nd harmonic distortion from 100 KHz downwards, while the 6622 there is more...suspect. From my early listening, it's also evident that it has power and dynamics in abundance.

Hmm, sounds like I should try the LMH6654 some time, then! I hope you enjoy the 8057. I'm act least enjoying the 8058 in this amp and enjoyed it in a CMoy-style circuit as well. The more I listen to it, the less I end up wanting to listen to my other PINT, it seems. When connecting each to the headphone out of my MP3 player, I do find the AD8058 PINT seems to be slighty more dynamic and the 6172 PINT seems to actually be less dynamic than the HP out when using my earbuds. Weird. Going to try some other headphones. I just found the MP3 player, so haven't had a chance to test other stuff yet. So this is just a preliminary finding.

 

[Filburt]

Okay, with the Koss A/250, the 6172 PINT gets more intimate presentation and a bit less dynamic than HP out, but maybe slightly extra detail on vocals and violin type sounds. Also kind of hurts my ears a bit. 8058, by contrast, expands the soundstage noticeably and really pulls out a lot of extra detail and nuances in addition to being more dynamic. I find it also passes on more of the HP out's tonality rather than changing it quite a bit like the 6172 PINT. It just opens things up a bunch and brings out detail I'd actually never really noticed on it before. Neat.

Well, it'll be cool if you like it too, but at least I think I've found something that will work for me...for now anyhow

 

[IEATTEFLON]

The issue that I experienced had me perplexed so I did a bit of reading. From the brief observations I did, what I experienced seemed to describe thermal crosstalk. A slight fuzzy sound as if I raised overdrive that attenuated bass and treble definition. I was using a regulated linear 24v 500mah and I recall it being more noticable with the lm6172 than with the ad8620. I sit here wondering if anyone with an lm6172 equipped amp noticed this also.

 

[dhp]

Quote:

Originally Posted by IEATTEFLON The issue that I experienced had me perplexed so I did a bit of reading. From the brief observations I did, what I experienced seemed to describe thermal crosstalk. A slight fuzzy sound as if I raised overdrive that attenuated bass and treble definition. I was using a regulated linear 24v 500mah and I recall it being more noticable with the lm6172 than with the ad8620. I sit here wondering if anyone with an lm6172 equipped amp noticed this also.

you're not listneing to it on battery?

 

[Andrea]

Quote:

Originally Posted by IEATTEFLON The issue that I experienced had me perplexed so I did a bit of reading. From the brief observations I did, what I experienced seemed to describe thermal crosstalk. A slight fuzzy sound as if I raised overdrive that attenuated bass and treble definition. I was using a regulated linear 24v 500mah and I recall it being more noticable with the lm6172 than with the ad8620. I sit here wondering if anyone with an lm6172 equipped amp noticed this also.

I had the same sentiments about the standard Go-Vibe v5 (LM6172). I didn't relate it to thermal crosstalk, but now you've made me think of it, it's a possibility.

 

[Andrea]

Quote:

Originally Posted by Filburt Okay, with the Koss A/250, the 6172 PINT gets more intimate presentation and a bit less dynamic than HP out, but maybe slightly extra detail on vocals and violin type sounds. Also kind of hurts my ears a bit. 8058, by contrast, expands the soundstage noticeably and really pulls out a lot of extra detail and nuances in addition to being more dynamic. I find it also passes on more of the HP out's tonality rather than changing it quite a bit like the 6172 PINT. It just opens things up a bunch and brings out detail I'd actually never really noticed on it before. Neat. Well, it'll be cool if you like it too, but at least I think I've found something that will work for me...for now anyhow

wow, well, seems just like what I'm after, too... we have a common comparison term in the LM6172, which, when push comes to shove, we don't seem to 'hear' that differently! I'll let you know my impressions.