Jump to content
IGNORED

Compulab Airtop-G - new interesting, fast industrial server


Recommended Posts

Hi Guys - I just received a PM from Compulab asking if I want to review the Airtop-G.

 

It looks like a really good HQPlayer machine and should have CUDA offloading.

 

Anyone see a reason why I should or shouldn't get one in for review?

 

Go for it. I'd be very interested to know how it does with HQP upsamplampling. It seems there are a large number of HQP users - and others - here who'd be very interested in just such a small quiet, yet powerful server .

 

It would also be interesting to know how it sounds in general as a music server platform, and not just with HQP.

 

I actually have a Compulab i7 fanless device as my desktop work computer for several years. Very happy with it.

 

The only issue I have with it is that Compulab sometimes does proprietary things with drivers, because they may have unique HW solutions in their PCs. For instance, I tried upgrading my unit to Windows 10 from Windows 7, and everything worked except the audio out. Compulab don't have a solution, and I couldn't find an audio driver that worked with Windows 10 on that particular model.

Main listening (small home office):

Main setup: Surge protector +>Isol-8 Mini sub Axis Power Strip/Isolation>QuietPC Low Noise Server>Roon (Audiolense DRC)>Stack Audio Link II>Kii Control>Kii Three (on their own electric circuit) >GIK Room Treatments.

Secondary Path: Server with Audiolense RC>RPi4 or analog>Cayin iDAC6 MKII (tube mode) (XLR)>Kii Three .

Bedroom: SBTouch to Cambridge Soundworks Desktop Setup.
Living Room/Kitchen: Ropieee (RPi3b+ with touchscreen) + Schiit Modi3E to a pair of Morel Hogtalare. 

All absolute statements about audio are false :)

Link to comment
I am interested to know which off the shelf linear power supply are you using for your DSD512 up-sampling PC ?

 

Where did I say I'm personally upsampling to DSD512?

 

In any event, the SGM 2015 would do the job, and so too would my tricked-out PC with its Terradak linear ATX.

 

Moreover, when I used a hdplex linear power supply (in a dual Jplay setup), I found it restricted dynamics

 

Not sure about your case, but the effects of electrical noise, from a SMPS say, can often be confused with a 'more dynamic' presentation. Again, IME...

 

Mani.

Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs

Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers

Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro

Link to comment
Where did I say I'm personally upsampling to DSD512?

 

 

You didn't. That was more of a question than a statement :-) I am looking at this PC in context of just an HQPlayer up-sampling engine capable of doing DSD512 with the toughest filters/modulators to work in conjunction with a HQPlayer NAA device, like a microRendu.

 

In any event, the SGM 2015 would do the job, and so too would my tricked-out PC with its Terradak linear ATX.

 

Yes, I know that SGM is the only server that can do DSD512 with passive cooling and linear power supply but folks who are looking at Airtop PC wouldn't possibly be interested in getting the SGM.

 

Not sure about your case, but the effects of electrical noise, from a SMPS say, can often be confused with a 'more dynamic' presentation. Again, IME...

 

Maybe, maybe not. Jussi wouldn't recommend it if the presentation is hyped/artificial due to smps noise. Anyway, I am behind a conditioner and my experiments showed that it didn't matter if I had an smps supply in the chain. Alex from Uptone also uses a balanced power to isolate smps noise from the main components.

Link to comment
I am looking at this PC in context of just an HQPlayer up-sampling engine capable of doing DSD512 with the toughest filters/modulators to work in conjunction with a HQPlayer NAA device, like a microRendu.

 

In that case, having a silent, small footprint PC might not be so important, as you could put the PC wherever you want (e.g. in the basement, if you have one). That's the beauty of the whole NAA concept. (Incidentally, I set my first NAA up back in 2012.)

 

Anyway, I am behind a conditioner and my experiments showed that it didn't matter if I had an smps supply in the chain. Alex from Uptone also uses a balanced power to isolate smps noise from the main components.

 

Yes, but no conditioner or transformer is going to isolate the PE from the SMPS. In my experience, any SMPS in the chain destroys the sound. Let's just leave it at that...

 

Mani.

Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs

Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers

Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro

Link to comment
In that case, having a silent, small footprint PC might not be so important, as you could put the PC wherever you want (e.g. in the basement, if you have one). That's the beauty of the whole NAA concept. (Incidentally, I set my first NAA up back in 2012.)

 

I am very much aware of the NAA concepts (and has been using it for a while now) but the trouble is I have to keep the PC in the audio room.

 

Yes, but no conditioner or transformer is going to isolate the PE from the SMPS. In my experience, any SMPS in the chain destroys the sound. Let's just leave it at that...

 

Mani.

 

Yes, we all have different ears and brains to judge things. I agree - lets just leave it at that :-)

Link to comment

Going back to CPU speed, I asked Jussi for a possible benchmark. He said that a good candidate would be POV-Ray benchmark which he used it himself for benchmarking.

Other good candidates may be c-ray, NAMD Molecular Dynamics and NPB Fluid Dynamics from the Linux-Bench suite.

 

According to POV-Ray benchmark,

i7-5775C = 1541

i7-6700 = about 1720

i7-6700K = 1974

 

In view of these values it has to be seen if an Airtop-G is capable of upsampling to DSD512. The GeForce GTX 950 would speed things up anyhow.

 

A few other POV-Ray benchmark values from Legitreviews.com:

 

povray-695-x-645x740.jpg

 

PS: I deduced i7-6700 value from Anandtech.com

Link to comment
Dev,

with a single review you have no comparison to a computer with a 6700K Skylake. Maybe another computer/chip offers better SQ in the same price range.

 

Matt

 

Slapping hardware together is something anyone can do for a PC... at least the CAP server projects were geared towards educating DIY members or off the shelf purchasers on some principles of better audio PC design. I could care less about another "me too" manufacturer unless there is something novel about the motherboard that a review could prove/disprove as add on to the body of knowledge for a better audio PC design.

Regards,

Dave

 

Audio system

Link to comment

From what I've read, the hardware in the Compulab is oriented more for video and 3d performance. The 5775 is apparently better than the 6770k in those areas. Doesn't mean the PC isn't good for audio, just means it wasn't especially designed for audio.

Main listening (small home office):

Main setup: Surge protector +>Isol-8 Mini sub Axis Power Strip/Isolation>QuietPC Low Noise Server>Roon (Audiolense DRC)>Stack Audio Link II>Kii Control>Kii Three (on their own electric circuit) >GIK Room Treatments.

Secondary Path: Server with Audiolense RC>RPi4 or analog>Cayin iDAC6 MKII (tube mode) (XLR)>Kii Three .

Bedroom: SBTouch to Cambridge Soundworks Desktop Setup.
Living Room/Kitchen: Ropieee (RPi3b+ with touchscreen) + Schiit Modi3E to a pair of Morel Hogtalare. 

All absolute statements about audio are false :)

Link to comment
From what I've read, the hardware in the Compulab is oriented more for video and 3d performance. The 5775 is apparently better than the 6770k in those areas. Doesn't mean the PC isn't good for audio, just means it wasn't especially designed for audio.

 

Yes, the Compulab and the best Intel NUC are gaming computer but that does not matter. Even if there may be not much difference in performance there is a price difference of 1k, enough for me to chose the cheaper one and buy an external high quality LPS for it.

 

Matt

"I want to know why the musicians are on stage, not where". (John Farlowe)

 

Link to comment
That's not the only way of looking at it, if CUDA offload to a Graphics card enables it to do DSD512 upsampling, then why not prefer passive cooling?

 

Maybe, what I dislike is that it uses a 5th generation processor and you can change pretty nothing.

 

Matt

"I want to know why the musicians are on stage, not where". (John Farlowe)

 

Link to comment
  • 3 weeks later...
We'll see. I just got one to try out. DSD512 upsampling ain't of great interest to me but I'll give it a shot.

 

Thanks Kal. Appreciate your willingness to devote time to the community. I'll be interested in hearing your impressions and results.

Main listening (small home office):

Main setup: Surge protector +>Isol-8 Mini sub Axis Power Strip/Isolation>QuietPC Low Noise Server>Roon (Audiolense DRC)>Stack Audio Link II>Kii Control>Kii Three (on their own electric circuit) >GIK Room Treatments.

Secondary Path: Server with Audiolense RC>RPi4 or analog>Cayin iDAC6 MKII (tube mode) (XLR)>Kii Three .

Bedroom: SBTouch to Cambridge Soundworks Desktop Setup.
Living Room/Kitchen: Ropieee (RPi3b+ with touchscreen) + Schiit Modi3E to a pair of Morel Hogtalare. 

All absolute statements about audio are false :)

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...