CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS
Lobby => Amp and DAC Measurements => Topic started by: ultrabike on July 15, 2014, 06:37:36 AM
-
Have had these for quite a while now. Thanks Raif! :)p5
Keep in mind that when describing how an Amp/DAC sounds, to me it is not always at the same level as with headphones. In most cases differences are more subtle... sometimes. With that in mind, these are IMO warm yet very detailed. I think it complements well headphones such as the HD600. It's a pretty powerful little gadget, and it gets a bit warm...
The Geek (1000 mW) is also a pretty practical device and it indeed handles all the way to 192 kpbs through it's ASIO driver. I feel it can render dynamics pretty well. I found this little amp quite engaging indeed.
What I did not like AT ALL is the volume control. HORRIBLE. The volume setting needs to be saved somewhere inside the device. ALWAYS make sure the volume is turn down before playing anything through. Default is full powa.
As far as measurements is concerned, take them with a grain of salt. I'm not using an Audio Precision or a Prism Sound or a whatevers... Just a set of laptops and good 'ol REW. Results I get seem a little worse than what the LightHarmonic guys get, but not bad at all IMO.
Test were done with a 33 ohm load. The Geek is able to deliver 4 Vrms into 33 ohms, which translate into about ~500 mW. I'm confident that with a 16 ohm load (if current is not limited by the USB source) the Geek will deliver 1W... This is about 250 mA, and I'm definitively not confident that the Geek will deliver these 1Ws in "Class A" mode.
Here are the 1 kHz THD plots under different output power considerations and using 192 kHz option in the Geek (Focusrite 2i2 capture is set to 96 kHz which is the highest it can do):
NOTE: Please ignore the y-scale, as the dBFS are off by about 5 dB. However, the distortion % are IMO accurate. Full scale to me is 4 Vrms (relative to the Geek), but the 2i2 inputs are not calibrated to that. The title of the plots specify what I think is the actual dBFS.
1 kHz THD (-20 dBFS)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6973;image)
1 kHz THD (-10 dBFS)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6975;image)
1 kHz THD (1 Vrms)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6977;image)
Same situation, but Geek is set to 44.1 kHz
1 kHz THD (1 Vrms)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6979;image)
The following are the 500 Hz THD numbers for 44.1 kHz and 192 kHz Geek settings @ 1 Vrms
500 Hz THD (192 kHz)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6981;image)
500 Hz THD (44.1kHz)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6983;image)
-
Moving on to the IMD plots @ 192 kHz and 1 Vrms
SMPTE (60 Hz + 7 kHz 4:1)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6985;image)
DIN (250 Hz + 8 kHz 4:1)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6987;image)
CCIF (19 kHz + 20 kHz 1:1)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6989;image)
IMD plots @ 44.1 kHz and 1 Vrms
SMPTE (60 Hz + 7 kHz 4:1)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6991;image)
DIN (250 Hz + 8 kHz 4:1)
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6993;image)
CCIF (19 kHz + 20 kHz 1:1) - Note the un-attenuated image not present in the 192 kHz case
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6995;image)
-
One of the things that seem to stand out about the Geek, besides the formats and the power it can provide for its size, is that it seems to come across a bit warm. Therefore a set of custom bass target IMD plots: 30 Hz + 100 Hz (4:1) & 20 Hz + 70 Hz (4:1) where generated and compared to the Focusrite 2i2 at 0.4 Vrms... (Kinda nice output level for the Focusrite since it cannot give out as much juice as the Geek 1W)
IMD 30 Hz + 100 Hz (4:1) Comparo - 2i2=green, g1k=yellow
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6997;image)
Geek is about 0.0804% while 2i2 is about 0.0256%
IMD 20 Hz + 70 Hz (4:1) Comparo - 2i2=red, g1k=red
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=6999;image)
Geek is about 0.0787% IMD while 2i2 is 0.0298%
The a bit higher IMD of the Geek may (or may not) account for the perceived extra warmth or bass slam or [insert your favorite term] when using the Geek 1W.
-
Summary: Geek Out 1K SQ wise kicks ass and is the schiit yo... Interface volume control is poo poo poo poo poo poo
... also the FR is uber flat into 33 ohms from the 0.47 ohm connector... Ignore the dB scale as it's a bit arbitrary:
44.1 kHz
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=7001;image)
192 kHz
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=7003;image)
-
Summary: Geek Out 1K SQ wise kicks ass and is the schiit yo... Interface volume control is poo poo poo poo poo poo
Agreed that it sounds great and is a great product, and also that the volume control is wonky. The good news is they are posting a firmware update in the near future to correct that. I have just used my system volume controls (on a Mac, I go to the Audio MIDI panel and drag it down quite a bit). The SQ can't be changed much after it ships but the volume control can! So hang in there.
-
Do you already have a 450 coming? Both CT and I find it slightly less warm than the 1000. I'm actually using it as my work DAC with the Wyrd.
-
Do you already have a 450 coming? Both CT and I find it slightly less warm than the 1000. I'm actually using it as my work DAC with the Wyrd.
Yup, gonna pick one up this week and hand it off at some point.
-
I can do the tests on the 450 as well, and provide them both in a few days to Mike so he can make some comparos...
I feel as if this one was kind of like the anti-Vali in terms of sound signature, but in a good way. This one is obviously more compact, but it's solid state. Both are smallish for what they are. Also this is more than an Amp, and the DAC section is pretty versatile. I would say this is ideal for portable and practical applications... if one overlooks the power consumption. The 450 might be more practical in that sense... Different operating points might also change the distortion scene, but dunno.
A good driver (SW) should turn this into an even shinier object.
-
Wut bad about vol control? Plz explain
-
Welp... Happily plug in the device, then the cans, fire up the music... and...
(http://static.music.cbc.ca/v2/modularpages/0/31/images/bttf1_1114101853426.gif)
QUICKLY PRESS STOP! Lower the volume on the device or ASIO... Back to normal.
Day passes, plug in the device again, then the cans, fire up the music... and... fvck! Didn't I turn down the volume before! Apparently not!
Annoyingly realize the thing doesn't save volume level...
-
The manual says to make sure you do NOT have your earphones in when you press play.
-
It was a "perk." :)p3
-
Didn't have the manual :(. On the bright side I didn't use earphones, just regular cans...
Still, IMO documenting this "feature" in the manual is probably not sufficient. Less than 16 ohm cutetips and cans will likely experience the full power of the dark side.
... God forbid one has them in cuz one forgot :-0
-
That image of Marty getting his ass blown across the room shares my sentiments when I had the GO 450. It's the only USB DAC that I know of that behaves like that. After owning it the second time around, thinking "I'll be sure to not have my iem/ciem plugged in, and/or manually lower the volume (WIN machine) prior to plug in" it was the only time my boys heard me yell, "F#$!", and my ears were ringing on/off for hours after those incidents.
I hope that once LH gets off their "pie in the sky" aggressive approach to their crowd funding projects, which IMO has been half-baked with no end in site. Then maybe I'll pick up another GO once they fix that infamous lethal and deafening volume feature.
-
http://geek.lhlabs.com/force/geekout/1382-beta-test-for-new-firmware-v1-1.html
I hope it's out soon but there is Larry posting about the beta.
-
Hopefully Mike can check the new F/W and confirm this "feature" has been "improved".
-
Hopefully Mike can check the new F/W and confirm this "feature" has been "improved".
Cool! I can't wait to brick Raif's GO!!
-
Cool! I can't wait to brick Raif's GO!!
Ha! :)p1
By the way G, awesome in-depth analysis of the GO1000.
-
Thanks mate! ;D
-
Interested by your thoughts on the GO popcorn.
Just came across their introduction post for the Geek Source on CA and my first thought was: who the hell to these guys think they are? Making such high claim as the BEST SOURCE and a more than suspicious product comparison (with the yet-to-be-release Aries and a stock Mac Mini using itunes...).
I'm open to hear all comments!
-
Interested by your thoughts on the GO popcorn.
Just came across their introduction post for the Geek Source on CA and my first thought was: who the hell to these guys think they are? Making such high claim as the BEST SOURCE and a more than suspicious product comparison (with the yet-to-be-release Aries and a stock Mac Mini using itunes...).
I'm open to hear all comments!
Actually I have notes that have been sitting around for more than a month comparing the GO1000 to GO1000+ UHA6Smk2, ODAC and Chord Hugo. Been waiting for the 450 to sort out if there are any discrepancies between the 1000 and 450 which should happen in the next week or so.
Till then the impressions have been favorable.
-
Actually I have notes that have been sitting around for more than a month comparing the GO1000 to GO1000+ UHA6Smk2, ODAC and Chord Hugo. Been waiting for the 450 to sort out if there are any discrepancies between the 1000 and 450 which should happen in the next week or so.
Till then the impressions have been favorable.
Favorable to the Hugo objectively (technically) or subjectively?
I own the Hugo currently. It is a very fine DAC technically (great details, smooth highs, good bass pitch/definition), but I find myself preferring the warmer / more fun sound of the quteHD (which I re-acquired recently).
The Hugo is slightly too thin and treble-oriented for my ears and my setup (Hifiman headphones are already that... On paper, Audeze headphones are a better pairing). That's my subjective preference at play. Technically, it is better than the quteHD, no question (except maybe bass, where I actually find the qute to reach lower and with better details. Could be because of the higher level in that region as opposed to the Hugo focusing more on the other end of the spectrum.)
-
I and a couple of other people have heard a significant amount of hiss from the GO 450 using IEMS, but the LH people are insisting that it is dead quiet on their systems.
Not sure if it is my source (have the same issue using a laptop or the computer and regardless of whether the power is plugged into the laptop or running from battery), but have put myself on the list to swap the GO 450 for a GO 100 (hopefully this will eliminate the hiss). Unfortunately I don't have an LPS or other power conditioning system at hand, but given this is meant for portable use, it's not really practical carrying a LPS, the computer and the GO 450 all at once!
Also have you tried the 3D awesomifier effect, because I and other people on the forum have found it unusable because it creates a recurrent buzzing sound that is about twice as loud as the background hiss with the 3D awesomifier effect turned off.
I am really really hoping the GO 100 will fix these issues. I only want to use it for IEMs (I have other amps for headphones and don't intend to use big cans on the move ... )
-
Didn't have this problem with the GO 1k. Maybe a product defect? I used mostly cans instead of cutetips though... Which IEMs did you try?
-
GO prototype 100 I had for a bit was quieter than my 450 but still the <slightest> bit of hiss with UERM. Good enough that I wanted to buy the Proto, though. My UERM have higher impedance than many IEMs. But talking to Gavin last Saturday, the production 100s will be quieter a than the prototype I heard and may even fix that 3D buzz. (I, too, find the 3D buzz with IEMs makes that unusable on 1000/450/proto 100.)
-
I just tested my GO 450 on a Lenovo Yoga 13 laptop and a Macbook Pro using my Noble K10's which I think are 30 ohms. I get no noise at all. It's dead quiet.
When I first plug in the GO450, I do hear noise, but I think that's from the GO powering up and initializing. After that it's quiet.
Also turning on the 3D awesomefier does not cause any noise or distortion.
Silly question, but you are using the 0.47 ohm output, not the 47 ohm output, right?
-
I have tried it with the Shure E530, Sony XBA-H3 and the Zero Audio Carbo Tenore (the only IEMs I have handy).
All of them have audible hiss, with hiss being most obvious with the Shure, then the Sony and finally the Tenore. With the tenore it's nearly inaudible. This is perhaps consistent with Ultrabike's measurements of the Tenore having a bit of a treble roll-off?
On all of them the 3D buzzing is obvious and makes 3D mode unusable. I am REALLY hoping the GO 100 fixes these issues. I have put myself down for one.
Interestingly I had the GO 720 which I swapped down to the 450, and overall I think I prefer the sound of the 720 even though it was noisier :( But this may be placebo though.
-
I've a Geek Out 450 on Loan since today...
- One the worst noise floor I've ever heard (Shure SE535)
- it runs very hot .
- This little thing go always to 100% volume when I reboot the PC or when I plug off and in the GO. high risk of Ears damage ... p:8
I'me using Ubuntu. I thought this dac/amp should work plug & play. was I wrong or is it the normal behavior of this thing? walk the plank
-
That is currently "normal" behavior.
-
Sadly it is a known issue.
The 450 would come with a product sheet warning you about the volume and not to have your have your IEMs in your ear or player at max when you start the music.
I think there is a FW update for volume and they are working on other fixes as well.
As for noise, I didn't notice anything with the UERMs on the 450 but they aren't 535s. I can check later with my ES5s.
-
So, curious, I have 2 450s and one has the noise with TF10s while the other doesn't. There's something strange here.
-
So, curious, I have 2 450s and one has the noise with TF10s while the other doesn't. There's something strange here.
Model numbers?
I did notice a sonic difference between two 450s at the last meet but found out it was temperature related once the more strident one got warmed up.
-
They're at home and I'm in an airplane with the 720 and my HD800s in my bag.
So, LH got the first few batches of the ESS mobile chip line and ESS is notorious for being opaque with the product. There might be some weird production variant.
-
I've had two 450 units and they were not the same with regards to noise. One was silent with UERM while the other I can hear the the noise.
-
My early unit was silent and the recent unit is the noisy one.
-
Same. Early unit silent. New, retail package unit sounds like aliens are beaming encrypted messages into my brain.
-
As for noise, I didn't notice anything with the UERMs on the 450 but they aren't 535s. I can check later with my ES5s.
Thks I haven't any Windows on the hand to test it against my Ubuntu. Volume management seems to be awful: Software volume settings is not kept by ALSA or PULSE AUDIO , Hardware settings are always back to full power , sometimes the sound disappears (but not the Noise) ... A very dangerous little piece of crap .
Sadly when I tried ( very carefully with my hand on my IEM in order to pull out them out of my ears quickly) it seems to sound good.
Jeez... how this thing can be legally sold ? it's a noisy lethal weapon .
-
I think my eardrums were damaged at least three times with my GO yesterday. For apparently no reason, the indicator lights went berserk; it showed the music as DSD 3x (when I was playing 16/44.1) and then the volume went way down and then screamed all the way up. Man, that is indeed a lethal weapon. Each time, I was playing it through my MBA with Amarra v3.0.2. Now I'm beginning to doubt whether the Amarra may be buggy to cause this.
-
Despite this happening a few times with my GO 450, which pissed me off to hell and back enough to sell them off, and once with my current 720, I have my fingers crossed that a new firmware that is planned to repurpose the two physical volume buttons as 1)On/Off 3D enhancer; 2) On/Off digital filter respectively. That's all fine and dandy if and when that happens.
Still puzzled with the two outputs and wish that the lowest output was hiss free, at least with my UERM. I guess that's where the GO 100 comes in, no? Anyone have in en route or plan to get one?
-
An update, I have the GO100, no noise. Thing goes berserk when I plug out the line out to plug in my IEM but not volume/DC bursting. Sometimes the device becomes unrecognized when you plug it in and out. Keep the drivers on hand for that to remove and reinstall if that happens.
Sound wise this is pretty amazing for the price I got it for($119). Minumum phase is pretty full bodied. Sounds harsh on warmup as the clocks are not fully set in. FRM filter clips badly out of the box(distortion reaches -40 or -30dbFS) but settles in though still looks bad compared to the TRM filter, looks like a firmware fixable issue.
Out of the box jitter looks bad, there is no difference between 16bit and 24bit data as its all swarmed with noise, the clocks or something needs warmup. Sounds harsh too before full warmup.
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=8119;image)
After it warms up the harshness disappears together with the jitter.
(http://www.changstar.com/index.php?action=dlattach;topic=1678.0;attach=8121;image)
-
That seems to jibe with subjective impressions myself and others had about different sounding units when cold. Nice data.
-
Keep the drivers on hand for that to remove and reinstall if that happens.
Pain in the neck. But more important, is it not class-compliant, at least for lower sample rates?
If not, then that is big-black-marker through the name for us Linux users --- whether it sounds good or not.
-
Pain in the neck. But more important, is it not class-compliant, at least for lower sample rates?
If not, then that is big-black-marker through the name for us Linux users --- whether it sounds good or not.
Its not at all, requires drivers to be recognized.
-
Its not at all, requires drivers to be recognized.
Pretty sure it is, worked perfectly fine on my MacBook Pro right off with no drivers.
Edit: huh, just read what you wrote in your GO100 thread, the 1000 I tried in Tokyo back in April worked fine with no drivers on my mac.
-
Hmm I thought it did, better correct that.
-
I wonder if Sony MDR MA900 will blow up when connected to geekout 1... this headphone hjas input impedance of 12 ohm.
-
Just want to add that the slightly excessive warmth of the geekout 1000 disappeared after a few weeks of use.