O2 AMP + ODAC

headwhacker said:

 

silverears said:

  Guyz, I've tried feeding other DACs to O2, and it's not the same as ODAC feeding it.  I just tried feeding it DX90 and X5 line-outs. And, also C5D in the past. It doesn't sound as detailed and clear as when ODAC is feeding it.  Do you think there was something on the ODAC that matches well with O2?  U know who probably knows something we don't about the match up.  I need to read up on it, maybe he talks about it.

Click to expand...

 


I'm curious to hear the ODAC as well with O2. DX90 already sounds detailed and clear with O2.

Click to expand...

If you can get a hold of an ODAC, you should try it.  You may hear what I'm talking about.
 
 

I haven't tried the DX90, but the X5's DAC sounded coloured to me. Likewise for the C5D, but I vastly prefer the C5D's DAC to that of the X5. I'll be receiving the Geek Out soon, so it'll be interesting to compare that to the ODAC.
http://geek.lhlabs.com/force/geekout/898-geek-out-vs-the-others-3rd-installment

Click to expand...

I don't find the X5 colored.  I was thinking the opposite, quite neutral.  reason why I think it's neutral is it now forward in any way.  The highs are not forward, there isn't bass quantity like the DX90, and not punchy sounding either.  I had the C5D and it sounded unnatural, the bass was too tight and smooth.  The sound was too smooth to my ears, and lacked detail.  I think it was on the colder side.
 
It's just that either DX90 and X5 doesn't have a clean LO or O2.  heaphone out of X5 sounds much clearer than feeding X5 LO to O2, it sounds not as clear.
 
Why do you think X5 is colored?

miceblue said:

Sure it can. That's what Fluke is for.

 

Click to expand...

I didn't think that Fluke worked with iTunes > 10.7, it used to and I'd done it... 
 

 

1. Does BitPerfect support high-res FLAC playback?

Well, yes and no.  But mostly no.  The problem is importing FLAC files into iTunes in the first place.  Fluke used to be an option to import high-res FLAC files into iTunes.  However, Fluke does not seem to work any more under OS X 10.7 (Lion) and above - and Fluke's developer is no longer active supporting the product.  At this time there is no alternative solution.

Click to expand...

http://bitperfectsound.blogspot.com/p/faq.html

Click to expand...

Click to expand...

 
If you still can, that's awesome... and I could play them with BP then.

I've been building my digital music collection for 14 years. When I started out flac was the way to go; lossless & relatively compact. Like most I started out on Windows but when Windows 8 was leaked I jumped ship to Apple. It made sense at the time as we already had an iPad & iPhone in the house. I still have the iPad but the iPhone is long gone (replaced with a Note 3).

Irrespective of what proportion of users choose flac I can't see any commercial sense in Apple not supporting us.

I also much prefer navigating my music via folder structure. I have never in my life made a Playlist by genre. I know what I want to listen to and I know where to find it. Perhaps this is why I keep Cog installed.

Anyway, my collection is now sitting at something like 100,000 tracks and the majority are flac. There's no way I'm going to start converting them to alac just so I can use the overblown iTunes interface. In fact, the only reason I bought JRiver is because of the remote apps that are made for it.

silverears said:

I don't find the X5 colored.  I was thinking the opposite, quite neutral.  reason why I think it's neutral is it now forward in any way.  The highs are not forward, there isn't bass quantity like the DX90, and not punchy sounding either.  I had the C5D and it sounded unnatural, the bass was too tight and smooth.  The sound was too smooth to my ears, and lacked detail.  I think it was on the colder side.

It's just that either DX90 and X5 doesn't have a clean LO or O2.  heaphone out of X5 sounds much clearer than feeding X5 LO to O2, it sounds not as clear.

Why do you think X5 is colored?

Click to expand...

"Neutral" is so subjective in audio it's not even funny. People say the HD800 is neutral, others say the LCD-2 is neutral; they sound completely different from one another. XD

Yes, the highs are not forward, but they're a bit soft on the details to me and like FiiO's typical house sound, the sense of soundstage is narrow. The bass is not forward either, but it seems a bit bloated and thick (not punchy), which isn't unusual for a FiiO product from my experience. From my experience with FiiO's products, they tend to have a shouty upper-midrange that bothers me, and that was definitely evident to me with the X5 and is one of the reasons why I don't like it.

Objectively, FiiO's own published measurements of the X5 are nearly perfectly identical between the headphone out and the line out.
http://fiio.com.cn/products/index.aspx?ID=100000055517771&MenuID=105026016

greenkiwi said:

miceblue said:

Sure it can. That's what Fluke is for.

 

Click to expand...

I didn't think that Fluke worked with iTunes > 10.7, it used to and I'd done it... 

Click to expand...

I'm using the latest version of iTunes with OS X 10.8.5. You have to set iTunes to run in 32-bit mode. Granted, Fluke + iTunes is pretty much completely useless since you can only add one music file at a time, but the ability to play FLAC in iTunes is still possible nonetheless.

true, neutral is very subjective without a reference. DX90 might sound like having a boosted bass but it's possible the X5 rolls off the bass a bit. ( I have not heard the X5 yet).

As for me Rockbox is a good know flat/neutral reference. When I compare DX90 with my rockboxed DX50 I hear just about the same bass quantity except DX90 has slightly more bass extention (a bit more rumble). Especially with the slow roll off filter and driven by O2 using my T1 or DT250.

Wow....
 
I don't use Apple at all.
 
I use only FLAC, lossless and bitperfect playback.
 
Simple and it works..
 
Apples world is all about making money. Which is fine. So is Windows but to me its a much more relatively open OS.
You have a lot less issues with new stuff etc....and being the majority of the world base out there STILL, usually programs etc will work first time..or better the first time etc.
 
I just don't like or need the Apple closed world.
 
JRiver is a great simple to use player and works very well in the Windows world.
 
With all the talk about getting the most out of your stuff, and worrying about how to set up the correct Bitdepth and sample rate....why would you even consider anything but a lossless file format???  Even 320 bits...OMG.
 
DBPoweramp works very well with ripping to Flac, Foobar is a good alternative, but not as user friendly as a windows gui program like JRiver.
 
Each to his own....
 
Alex

I converted to the dark side a couple of years back from Microsoft to apple and I haven't looked back. Call them what you like, I admit they have me , I'm in their pocket but my 'digital ' life has never been more carefree and easy . I have a mac / iPad/ I phone / Apple TV and Everything works with continuity and without hassle . I want computers and devices to make my life easier and to just enjoy using them. I never enjoyed using windows it was so high maintenance .

greenkiwi said:

BitPerfect also does this really well, and costs $10.

http://bitperfectsound.blogspot.com/

Click to expand...

Thanks for posting this. I downloaded it (no brainer for the price) and it's fantastic. Gets out the way and just does what it needs to. Can't believe I hadn't discovered this before. Pure music has more features but I don't care for them and it's more intrusive on the I tunes ui.

tdockweiler said:

BTW I really love the O2 with the LM4652. I'm using that instead of the OPA2227 and stock op-amps. For JUST the Q701, the OPA2227 would be better, but it's not a perfect match for the HD-650. LM4652 sounds more "neutral" (yes, I know).

Click to expand...

 
I was never a fan of those JRCs either, i highly recommend you look into the Booster Board with OPA827's...

hybridcore said:

  Would be awesome to see JDSLabs include booster boards on their O2s.

Click to expand...

 
agdr kindly provided us with a boost board in exchange for dScope analysis. I'm impressed with his progress, unfortunately, this modification must remain a DIY project in its current state. This is not agdr's fault, but rather a result of agdr choosing to adhere to the CC BY-ND 3.0 Attrib license on O2.
 
agdr has posted initial findings to his main thread. We still have more questions to answer. Key results so far:
 

1. Max output power at 150-600 ohms is identical to a stock O2. The booster can accommodate increased supply rails from +/-12V to +/-15V, but thermal dissipation in O2 is insufficient for smaller loads, so this mod is not production acceptable.
2. Max output power at 32 ohms is ~35% stronger than a stock O2 in bursts, however, the booster's LME49600's overheat within 60-90 seconds of continuous maximum load. The thermally overloaded chips quickly exceed 60C, causing THD+N to rise dramatically before thermal shutdown. This is obviously not production acceptable.
3. THD+N is limited by the NJM2068 with booster installed. A stock O2 already performs well here.
4. Turn on/off relay works perfectly. Nice addition!

 
In short, the base O2 circuit needs to be abandoned to thermally satisfy LME49600s.

Thanks for sharing your finding. O2 has already proven that it's as good as it gets from the design target.

The O2's good but it has its limitations, as was recognised at the outset by the designer.  I salute agdr for the work he did but repeat what I said a couple of 100 posts back in the thread:  with his talent I'd opt for revolution and a new design on my own terms rather than incrementalism and respecting the constraints of the existing product.  My hat off to him, I hope he carries this forward.

Damn!!! This combo is rediculously transparent!!!  I just got a TOTL CIEM called Hidition NT-6 .  My god, never heard details like this, and the objective combo provides.  I hear everything on Kind of Blue.  Everything!!!

jseaber said:

• but thermal dissipation in O2 is insufficient for smaller loads, so this mod is not production acceptable.
• Max output power at 32 ohms is ~35% stronger than a stock O2 in bursts, however, the booster's LME49600's overheat within 60-90 seconds of continuous maximum load. The thermally overloaded chips quickly exceed 60C, causing THD+N to rise dramatically before thermal shutdown.

Click to expand...

So under normal operating conditions, people with the booster board have been listening to their music with a bunch of added THD+N because of the thermal instability?

miceblue said:

So under normal operating conditions, people with the booster board have been listening to their music with a bunch of added THD+N because of the thermal instability?

Click to expand...

No, you'd only encounter those conditions using low-impedance headphones at very high (unlistenable) sustained volumes.  At loud volume absorbing a peak without extra distortion but while delivering extra power would be possible.  Wether wise for your ears is a different matter.