Starke Sound AD4.320 Review (Multichannel Amplifier)

[greenpsycho]

This sort of reminds me of schiit's first response when amir's testing initially came out. They just denied everything off the bat and totally discounted all the measurements. Then they actually bought an audio precision machine, started testing, and their products improved exponentially.

If I made something that was advertised to weigh a hundred pounds, and someone else weighed and it weighed 50 lb, I would be curious about why that was and reach out, I wouldn't call his measurements crazy or totally discount them

 

[JRaab]

This sort of reminds me of schiit's first response when amir's testing initially came out. They just denied everything off the bat and totally discounted all the measurements. Then they actually bought an audio precision machine, started testing, and their products improved exponentially.

If I made something that was advertised to weigh a hundred pounds, and someone else weighed and it weighed 50 lb, I would be curious about why that was and reach out, I wouldn't call his measurements crazy or totally discount them

Yeah, my amps are 1 year and 7 months old respectively. So I'm not thrilled with their response. Depending on their official response and findings I may ask for a full refund just out of principle. Don't make claims you can't backup immediately and with un-biased proof. We'll see though, if they do find an issue and offer free repairs then I might give them a second chance. Screwing up is bad, but admitting to it and making it right goes a very long way to showing pride in your work. And that I can respect.

 

[greenpsycho]

Mine is coming up on a year old - I wonder if I'll have any recourse...

 

[RichB]

@amirm,

What is the bandwidth for the power measurements?

Could it be that the manufacturer is limiting the bandwidth at a lower frequency than ASR?

- Rich

 

[amirm]

@amirm,

What is the bandwidth for the power measurements?

It is only 22.4 kHz. An AES-17 pre-filter is used to cut out ultrasonics above 200 kHz. Then the input gets filter again with the ADC in the Audio Precision (48 kHz sampling, 22.4 kHz bandwidth).

Different bandwidth would change the measured distortion but would not cause the severe clipping that indicates power limit has reached.

 

[amirm]

Hi All,

Amirm I love the forums and have made purchase decisions in the past based off your measurements so I really appreciate all your work.

I created an account specifically for this thread cause I have 2 Starke AD4.320 bi-amped to a pair of Totem Hawks and a Totem Rainmaker center (not using bridge mode, just have #4 connection open). I reached out to Starke back on the 28th, when I saw the review, and this was their response.

"Hi Jeff -
Thanks for reaching out. Yeah, we saw the review and are quite disappointed with ASR. Their measurements are misleading and are not the case. All of our published spec's are true and real. Armin’s using some bizarre means of determining peak power output that is constantly running at 1/3 or ¼ of actual power – and it makes no sense to us at all. We're going to publicly release our measurements and the test scripts we used to obtain them so anyone can verify them. We'll have that available in a few weeks time. We'll also have an independent third party perform the same test."

Like John I haven't received an update yet but since it's been 2+ weeks since Amirm's findings and the products been out a year+, they should have had something put together already. The casual tone of the email wasn't great, but least that means a real person wrote it and not some auto-generates PR release.

Thanks for checking. The tested unit was shipped to them a while ago. By now they should have received and tested it. So very surprised we still have no information from them.

Forgetting the retesting, where are the R&D or production test measurements that would ostensibly exist prior to my review? They post one graph they removed so clearly some measurements were made and would have been quick to post.

 

[JRaab]

Thanks for checking. The tested unit was shipped to them a while ago. By now they should have received and tested it. So very surprised we still have no information from them.

Forgetting the retesting, where are the R&D or production test measurements that would ostensibly exist prior to my review? They post one graph they removed so clearly some measurements were made and would have been quick to post.

Hey Amirm,

My thoughts exactly. If you had done testing throughout then you'd have the results readily available. Especially since the NS600 modules are supposedly custom made in house. Wouldn't you have been testing them the entire time you were designing them? The way I look at it, the longer they take the less likely they are to have been honest and are more likely talking with lawyers regarding responses that open them up to the least level of liability for selling a product not up to their claims.

 

[amirm]

They are also proving why it is futile in most cases to "check with manufacturer" before publishing reviews. Would they have made me wait nearly 3 weeks with no response?

 

[JRaab]

They are also proving why it is futile in most cases to "check with manufacturer" before publishing reviews. Would they have made me wait nearly 3 weeks with no response?

What? You mean you don't pull an Adam Schefter and let the subject proof your article before posting? What a novel idea. I'm willing to give them the benefit of the doubt for the moment, the broken binding might indicate a busted unit that slipped through QC, but the longer they take to respond, the more skeptical I get. I'm a Safety and Risk director, so admittedly I've used those tactics myself

 

[sam_adams]

I had pretty much written this amp off as another device that was more marketing than engineering given its poor performance in @amirm's test. Man, was I wrong about this one. This has become quite the controversy both here and in another @amirm-bashing thread over on AVS.

I'm going to have to stay tuned on both of these to see how this plays out.

 

[Shanman]

Well, I can only imagine it takes some thick skin and big cojones to tell it like it is. I appreciate that.

 

[MajorJuggler]

Hi All,

Amirm I love the forums and have made purchase decisions in the past based off your measurements so I really appreciate all your work.

I created an account specifically for this thread cause I have 2 Starke AD4.320 bi-amped to a pair of Totem Hawks and a Totem Rainmaker center (not using bridge mode, just have #4 connection open). I reached out to Starke back on the 28th, when I saw the review, and this was their response.

"Hi Jeff -
Thanks for reaching out. Yeah, we saw the review and are quite disappointed with ASR. Their measurements are misleading and are not the case. All of our published spec's are true and real. Armin’s using some bizarre means of determining peak power output that is constantly running at 1/3 or ¼ of actual power – and it makes no sense to us at all. We're going to publicly release our measurements and the test scripts we used to obtain them so anyone can verify them. We'll have that available in a few weeks time. We'll also have an independent third party perform the same test."

Like John I haven't received an update yet but since it's been 2+ weeks since Amirm's findings and the products been out a year+, they should have had something put together already. The casual tone of the email wasn't great, but least that means a real person wrote it and not some auto-generates PR release.

This implies that Starke's rated power is short-term non continuous power. It's not necessarily a peak power rating (peak during an individual cycle), but rather peak RMS power that the ClassD amp can put out for a short period of time before the power supply can't mantain the supply voltage.

An appropriate follow on question is how much continuous RMS power their toroidal power supply can output. It's likely right around the output numbers measured here divided by the Class D efficiency.

If true, then I imagine that Starke's argument is that most program material never stays close to 0dBFS for any meaningful period of time, so it has negligible impact in real world usage.

 

[audio_lover]

so it has negligible impact in real world usage.

This 100%, I do have the amp and it just sounds wonderful and rivals my Nord NC500's.
One thing that Amir is correct about is the low gain, have to crank up the level in my pre-amp quite a bit with the Starke compared to the Hypex

 

[amirm]

If true, then I imagine that Starke's argument is that most program material never stays close to 0dBFS for any meaningful period of time, so it has negligible impact in real world usage.

I test for burst power. It is the graph on the right:

That is still way too little power.

 

[voodooless]

This implies that Starke's rated power is short-term non continuous power. It's not necessarily a peak power rating (peak during an individual cycle), but rather peak RMS power that the ClassD amp can put out for a short period of time before the power supply can't mantain the supply voltage.

But Amir already measures burst power:

The peak power is already a burst measurement. And it's also not bizarre as claimed, it's according to a standard: CEA-2006/490A. Which is something like 20ms@1Khz (correct me if I'm wrong). At 50 Hz, 20ms is just one period. So one can assume that in bass frequencies, these measurements will actually be worse. And yes, if you actually hit these power levels, you will notice in practice. That can also be seen here:

In the Bass, distortion rises much earlier with a lot of power, and available clean power is at 20 Hz is only half. So power saturation at low frequencies is severely compromised.

So why then are 9 out of 10 power measurements done in reviews according to the specs of the manufacturer? If somebody uses bizarre means, it's them, not @amirm.

Edit: Amir beat me to it

 

[MajorJuggler]

Ah-ha, well there goes that theory. 20ms is very short, so that doesn't explain the discrepancy.

I didn't catch the spec CEA-2006/490A. Back when I was doing amplifier development we would always test max continuous power, but it was for military applications, so testing was done vs program requirements, not consumer specs.

In any event, I'm curious what their test methodology is that could yield such a large difference.

 

[ColonelJoy]

Measurements Smeasurements, I was skeptical when I ordered this amp to audition in replacement of an 11K Hegel H590 I regretted selling as soon as it was out the door... I was disappointed with combos of a modified Lumin T2 driving amps like the Benchmark ABH2, and Rogue ST 100 powering my Vandersteen 2ce. I found I could equal the Lumin performance with a $399 iFi streamer feeding a $499 Pro-Ject S2...win for the budgetphile! But I couldn't find an amp that move my bi-wired Vandersteen 2Ce sig IIIs to the heights I wanted.... I've only had the Starke a day....Holly shit! It's a unicorn amp! It sounds so good and so way above its price point, it's really hard to describe. I would say that Jay's review, if anything, doesn't do it justice. Considering how thoroughly the amp moves the vandersteens, I would guess that Amir certainly got a faulty unit. I mean currently my amp/DAC/preamp combo.. whole front end is $2,300...and it sounds better than any system I've had the pleasure of listening to in either a home or an audio shop. The amp does seem to have a hybrid solid state/tube voicing... Heard those claims a lot over the years of course... But this is legit. This is the only amp to consider if you're not spending under 10 grand, as far as I can tell.

This is a unicorn amp!

Matt

 

[RichB]

Measurements Smeasurements, I was skeptical when I ordered this amp to audition in replacement of an 11K Hegel H590 I regretted selling as soon as it was out the door... I was disappointed with combos of a modified Lumin T2 driving amps like the Benchmark ABH2, and Rogue ST 100 powering my Vandersteen 2ce. I found I could equal the Lumin performance with a $399 iFi streamer feeding a $499 Pro-Ject S2...win for the budgetphile! But I couldn't find an amp that move my bi-wired Vandersteen 2Ce sig IIIs to the heights I wanted.... I've only had the Starke a day....Holly shit! It's a unicorn amp! It sounds so good and so way above its price point, it's really hard to describe. I would say that Jay's review, if anything, doesn't do it justice. Considering how thoroughly the amp moves the vandersteens, I would guess that Amir certainly got a faulty unit. I mean currently my amp/DAC/preamp combo.. whole front end is $2,300...and it sounds better than any system I've had the pleasure of listening to in either a home or an audio shop. The amp does seem to have a hybrid solid state/tube voicing... Heard those claims a lot over the years of course... But this is legit. This is the only amp to consider if you're not spending under 10 grand, as far as I can tell.

This is a unicorn amp!

Matt

I am sorry, unicorns are not real but measurements are.
You are happy with your amp, so rock-on.

- Rich

 

[sarumbear]

Measurements Smeasurements, I was skeptical when I ordered this amp to audition in replacement of an 11K Hegel H590 I regretted selling as soon as it was out the door... I was disappointed with combos of a modified Lumin T2 driving amps like the Benchmark ABH2, and Rogue ST 100 powering my Vandersteen 2ce. I found I could equal the Lumin performance with a $399 iFi streamer feeding a $499 Pro-Ject S2...win for the budgetphile! But I couldn't find an amp that move my bi-wired Vandersteen 2Ce sig IIIs to the heights I wanted.... I've only had the Starke a day....Holly shit! It's a unicorn amp! It sounds so good and so way above its price point, it's really hard to describe. I would say that Jay's review, if anything, doesn't do it justice. Considering how thoroughly the amp moves the vandersteens, I would guess that Amir certainly got a faulty unit. I mean currently my amp/DAC/preamp combo.. whole front end is $2,300...and it sounds better than any system I've had the pleasure of listening to in either a home or an audio shop. The amp does seem to have a hybrid solid state/tube voicing... Heard those claims a lot over the years of course... But this is legit. This is the only amp to consider if you're not spending under 10 grand, as far as I can tell.

This is a unicorn amp!

Matt

Said the member on his second post since joining seven months ago.

The excuse is the ever so predictable: @amirm got a faulty unit.

 

[RichB]

Said the member on his second post since joining seven months ago.

The excuse is the ever so predictable: @amirm got a faulty unit.

Well, that is better than the AVS thread where some get personal.
How dare you gore my ox!

- Rich