[adydula]

I asked this question in one of the Schitt threads and was told to come here and play. So here we are.

I have a Schitt Bimby multibit and do not leave it on all the time. I don't really perceive any real world sonic differences when I leave it on for several hours vs 30 minutes.

In some Schitt manuals it states it should and in some others it wont hur t leave the MB dac on all the time.
People have stated its the medical grade dacs that need to stabilize etc...

Well I am trying to understand from an engineering perspective why this might be so?

Is it due to poor circuit design, poor power regulation or unregulated supplies?
What is the actual degradation if not left on all the time? Missing bits, oscillations, noise, hum, distortions??

I have 6 Schitt products and this in no way is stating that these products are not well engineered of built.

Its just I am both an objective person that wants to understand this if at all possible with science and engineering in mind...sans emotion

??
Thanks
Alex

 

[bigshot]

I don't see any reason why it would have to be left on all the time. If it is shifting how it sounds over a half hour of being on, it would have to be pretty poorly designed. Tubes might warm up, but solid state should just work. Sounds to me like some placebo monkey business designed to set up expectation bias.

 

[Strangelove424]

The reasoning for leaving amps or dacs on is to maintain the temperature of electrical components which operate more efficiently at their normal running temps. This was especially true for tube amps, which needed to heat up a filament in order to amplify. SS amps don't rely on heat and gas (or a vacuumed space) to excite electrons, using solid elements with different potentials instead, and are not as sensitive to temp changes. They are also likely to achieve ideal running temperature within 10-15 minutes anyway.

The other side of the argument is that increased temperature causes increased noise in the circuit, and at a certain point (unless equipped with cooling mechanisms) the equipment might exceed its ideal temperature and introduce more noise to the signal. There's also the wear and tear issue of leaving things on 24/7. Like most things in life, I'd approach this with balance.

My rule of thumb is to avoid unnecessary heat cycles (mainly for wear reasons) if I plan on listening to a component again within an hour or two. Otherwise, I turn it off.

 

[astromo]

I say again ...

Thermal equilibrium happens much faster than he thinks. No need to leave it on 24/7. Warm it up. Sure. As a mechanical engineer in the electronics industry who does thermal simulations, I find a lot of this laughable. Save your electric bill. And caps.

Mike has been misunderstood / misquoted. Refer his clarification from an earlier post.

It is not the DSP which needs to thermally stabilize. It is the DAC chip which needs to settle into its INL spec. Regret the slow comment; I have been distracted by the Schiit Show.

 

[amirm]

I have a Schitt Bimby multibit and do not leave it on all the time. I don't really perceive any real world sonic differences when I leave it on for several hours vs 30 minutes.

I was specifically asked about this in my review of the Schiit BiFrost Multibit and I tested it after being on for 5 minutes versus overnight. There was no difference in measured results. The unit temperature doesn't change much anyway so not sure why it would need more time to stabilize.

So I would say your subjective results are correct.

 

[Zapp_Fan]

I found a plausible explanation here: https://www.maximintegrated.com/en/app-notes/index.mvp/id/4672

Although I only looked at it casually, the gist is that a DAC's output can be biased depending on temperature.

"This is of particular importance when a precision DAC is used to set a precise bias value. Any error can be calibrated out at room temperature. However, variations with temperature are far more difficult to compensate for. The errors that drift most with temperature are dominated by offset error and gain error." Looking over the page, this seems pretty credible, basically it's saying that between -40deg C and +104C this particular DAC changes output by about half a milivolt, which actually probably would be audible depending on where it occurs in the signal path. (keep in mind that this is a plausibly audible difference ... if you go from sub-arctic cold to a bit beyond boiling.)

Seems that the DAC can have a specific temperature that it performs theoretically best at, so maybe (speculating) leaving it on can help stabilize at that temperature. This might be considered a design flaw, as this document also notes that it's possible to design the system to compensate for this.

I would say that this is one of those times when something may be better in theory, may or may not apply to the gear in question, and even if it does, wouldn't create an audible difference in all but the most extreme use case.

 

[castleofargh]

as pretty much all DAC chips nowadays use more than one bit(even DSD stuff), I believe it's important to agree that the "multibit DAC" thing is a marketing label.

as for why keep stuff on, well I only do that when I forget to turn something off. ^_^ sure, gears can benefit a little from settling in and having the temperature reach a more stable level. my portable DAC/amp is such a device. but first, in my case the difference isn't much. and second, it's all good after maybe 5mn ON sharing my pocket with my hand. now if money and lifespan of the gear isn't an issue, sure we can leave it ON 24/7 for the convenience of not having to whine for 5 or 10mn, thinking how high end gears should always be at their best. I'm sure it's a rational thought for some.

now if there is another reason, like in the old days when turning stuff on and off could sometimes break them, well the Shiit guys are the ones who would know. but that seems unlikely.

 

[adydula]

Its good to see in this thread I started that there are people who think rationally and reason.....

Amirm, you mention you saw no difference in performance when measured?
Would you share what you measured and what the results were?

I hate to turn my stuff on and off for short listening sessions unless its solid state.....my tube amps , well they don't like this and its part of the tubes and tearing up the surfaces of the internal elements
when cycled on and off etc....just not good practice to me IMO.

But with SS stuff the Schitt and all the other stuff...if its well designed, good power regulation etc after 15 min or so it should be baked enough to play.

Tom at Neurochrome who I respect a lot with his engineering genius with his stuff and especially the HP-1, has stated that you turn it on and it plays well....

My feeling exactly until someone can prove it objectively to me...

Alex

 

[amirm]

Amirm, you mention you saw no difference in performance when measured?
Would you share what you measured and what the results were?

Sure.

This is a test of linearity of the DAC. In other words, whether it outputs the right voltage value relative to the digital PCM value fed to it. Ideal DAC would be a flat red line. Should a ladder DAC improve in performance it would be reflected in this graph:

We see the same failings in both. The DAC loses linearity at around -58 db whether it has been on for a few minutes or hours overnight.

 

[spruce music]

It is super simple. Hot Schiit is smoother than cold Schiit.

What more do you need to know?

 

[adydula]

spruce, repeat after me..."the earth is flat, the earth is flat, the earth is flat"...over and over and over.....lol

amirm, thank you for taking the time to share these plots. Very interesting...I have not heard of the site audio science review...i need to chcek this out....I am so tired of all the warm syrupy adjectives and the countless ooompa luuuumpas that just fall into the trap of it just sounds good if you do this or that without any real cience behind it....

I do like to listen and hear wonderful sounds like most of us, but like to think there are physics and science behond most of what we are experiencing...

thanks much
Alex

 

[SilverEars]

Did this thread get moved here? This question should be out in the open visible to all that are reading about Schiit products.

It's a great question. If engineered by engineers, they should provide reasonable explaination for this. They should explain this and we should question this suggestion of theirs.

 

[adydula]

I was told by a poster in another thread this is science and to come over to this "sandbox"...over in the other thread rather than continue to start a fire fight I left and came here.
I am just interested in the engineering reason of why do MB dacs need to left on all the time for "sonic bliss"? I think this is a fair question and deserves a honest, frank answer.

The stuff on this audio science review site actually shows a few plots of linearity etc.,,and their results makes me concerned and the other things they found also concerns me.

I have no ax to grind here, I have purchased several Schitt devices over the past years, Let the facts stand for themselves and let us decide with our ears, science, wallets etc.

Alex

 

[SilverEars]

Doesn't have to a firefight. They explain the reasons and we further ask questions and determine with all provided information out in the open. After all, we are the ones paying for the products and we need to know if it's just marketing or there is any trace of validity in the statements. This is why it should be visible to all.

We need skeptics and question askers in threads(questionable things asked) otherwise it is not balanced information being put out.

 

[Redcarmoose]

I remember reading all the folks leaving their DACs on when the Schiit DACs first came out a couple of years ago. Expectation Bias is a really strong form of perception changing phenomena. If I told you the jeans were tumbler rolled in imported river rocks for 24 hours and that they used special Arizona river rocks due to the specific results, the cotton would start to feel softer.

Amazingly there was a wine counterfeiting ring selling $60K bottles of wine which were relabeled $20 bottles. The arrest occurred after an owner researched a winery only to find out they were started 5 years later than the year on the counterfeit bottle of wine. The wine company said no that we started in 1935 not 1930 like it says on your bogus bottle. Perception can be swayed very easy, be it jeans, DACs or wine!