JBL SDR-35 AVR Review

[Dzhaughn]

Just Because Lackluster
Sound Doesn't Reduce
Audiophile Valuations, Really.

 

[G|force]

I voted poor with an eye on not terrible.
For the size and functionality of the thing I suppose it would be attractive to some, but the value to price seems to be inflated 3 or 4 thousand dollars. I'm just a basic 19th century stereo guy, so this is not kit I would buy.

Is there one processor and a stack of cables and 8 stereo amplifiers that can meet or exceed the performance and price without giving up anything? I honestly don't know.

 

[Spocko]

Looks like Anthem AVRs are the way to go since they have 2x independent bass management using ARC which is about as good as Dirac (Onkyo, Pioneer, Arcam and NAD AVRs).

 

[KMO]

It seems like there is a pattern of avr manufacturers specing high gain amp sections and low voltage output DAC sections. These don't seem well matched with external components when trying to use preouts or subwoofer outputs. Any idea why manufacturers are choosing this type of low gain DAC/high gain amp?

AVRs are switching devices. They accept both analogue and digital sources, and the analogue sources can either be digitised or passed straight through. To keep all that manageable, the DACs have basically the same level as the analogue inputs (ie topping out in the 1Vrms-2Vrms range - normal consumer line level). And then presumably for cost reasons they do their full 29dB of gain in the power amp step, so the "pre-outs" are roughly line level - there is no separate "pre-amplification" step to give higher-level pre-outs.

Block diagram and gain structure from an old AVR here - new ones haven't changed much:



But at this point, the main answer has to be because it's what they've always done and what their competitors are all doing. Standards are standards, and standardising on a gain structure seems reasonable, allowing straightforward matching with external amps. Someone upthread suggested THX specified 29dB as external amp gain, but I think they were probably just restating what most AVRs were already doing.

Put the other way, why are people making external components that don't match well with AVRs?

And are they, really? Is this a new problem? There seem to be plenty of amps out there with sensitivity at or close to 29dB that pair fine with AVRs.

Maybe my view is skewed my mainly looking at multichannel amps, who will naturally be trying to be AVR-compatible.

I think on this site you see quite a lot of amps with only 26dB, based on the Hypex/Purifi reference designs, but I don't think that's representative of the general market. Eg the Purifi 1ET400A-based NAD M23 has selectable 19/24/29dB gain.

I think mainstream manufacturers know AVRs need high gain.

 

[Vacceo]

Looks like Anthem AVRs are the way to go since they have 2x independent bass management using ARC which is about as good as Dirac (Onkyo, Pioneer, Arcam and NAD AVRs).

I hope Amir can measure the AVM90 to see if there is an improvement pattern in Anthem. That is not only interesting for consumers, but as a statement to the whole industry showing that technical excellence matters.

 

[Spocko]

Is there one processor and a stack of cables and 8 stereo amplifiers that can meet or exceed the performance and price without giving up anything? I honestly don't know.

I don't think so because it's all about priorities - all products, from sports cars to washing machines, have to decide which features are more important than others as sometimes they are mutually exclusive because elevating the performance of one aspect necessarily attenuates something else. If your focus and emphasis is the all might SINAD measurement, then too much DSP necessarily reduces some of that signal and yet room correction is essential when you have a multi-channel 9.2.4 movie setup! Additionally, dedicated home cinema setups are very different than multi-channel music because the audio source for movies is less about "audiophile" lossless zero distortion music playback and all about creating the immersion from subwoofer power to ambient cues to channel separation to height effects. The crazy level of acceptable distortion during explosions and action sequences where the SPL is over 82db (with peaks over 95db) means the AVR/PrePro really doesn't have to spend much money on best in class SINAD because what's the point? The louder you play, the less audible the distortion!

It's like complaining that one's pimped out Toyota Landcruiser can't run as fast as an el cheapo 20 year old Miata through tight hairpin turns. Not to say it isn't possible to create a dynamic suspension system to do this but that's why the Lambo Urus is so expensive but it still has limitations despite the premium paid, I mean, what about fuel efficiency - I want at least 35MPG from my Urus if I'm paying $200K right?!

 

[amirm]

AVRs are switching devices. They accept both analogue and digital sources, and the analogue sources can either be digitised or passed straight through. To keep all that manageable, the DACs have basically the same level as the analogue inputs (ie topping out in the 1Vrms-2Vrms range - normal consumer line level). And then presumably for cost reasons they do their full 29dB of gain in the power amp step, so the "pre-outs" are roughly line level - there is no separate "pre-amplification" step to give higher-level pre-outs.

Block diagram and gain structure from an old AVR here - new ones haven't changed much:

View attachment 226101

But at this point, the main answer has to be because it's what they've always done and what their competitors are all doing. Standards are standards, and standardising on a gain structure seems reasonable, allowing straightforward matching with external amps. Someone upthread suggested THX specified 29dB as external amp gain, but I think they were probably just restating what most AVRs were already doing.

Put the other way, why are people making external components that don't match well with AVRs?

And are they, really? Is this a new problem? There seem to be plenty of amps out there with sensitivity at or close to 29dB that pair fine with AVRs.

Maybe my view is skewed my mainly looking at multichannel amps, who will naturally be trying to be AVR-compatible.

I think on this site you see quite a lot of amps with only 26dB, based on the Hypex/Purifi reference designs, but I don't think that's representative of the general market. Eg the Purifi 1ET400A-based NAD M23 has selectable 19/24/29dB gain.

I think mainstream manufacturers know AVRs need high gain.

This doesn't answer why internal amps are allowed to run hard even when you have an external amplifier.

 

[amirm]

And why they allow higher voltages but at severely distorted conditions.

 

[Jbrunwa]

AVRs are switching devices. They accept both analogue and digital sources, and the analogue sources can either be digitised or passed straight through. To keep all that manageable, the DACs have basically the same level as the analogue inputs (ie topping out in the 1Vrms-2Vrms range - normal consumer line level). And then presumably for cost reasons they do their full 29dB of gain in the power amp step, so the "pre-outs" are roughly line level - there is no separate "pre-amplification" step to give higher-level pre-outs.

Block diagram and gain structure from an old AVR here - new ones haven't changed much:

View attachment 226101

But at this point, the main answer has to be because it's what they've always done and what their competitors are all doing. Standards are standards, and standardising on a gain structure seems reasonable, allowing straightforward matching with external amps. Someone upthread suggested THX specified 29dB as external amp gain, but I think they were probably just restating what most AVRs were already doing.

Put the other way, why are people making external components that don't match well with AVRs?

And are they, really? Is this a new problem? There seem to be plenty of amps out there with sensitivity at or close to 29dB that pair fine with AVRs.

Maybe my view is skewed my mainly looking at multichannel amps, who will naturally be trying to be AVR-compatible.

I think on this site you see quite a lot of amps with only 26dB, based on the Hypex/Purifi reference designs, but I don't think that's representative of the general market. Eg the Purifi 1ET400A-based NAD M23 has selectable 19/24/29dB gain.

I think mainstream manufacturers know AVRs need high gain.

The AV40 has the following input config option (per input), assuming the JBL models also:

"Input Trim – Sets the maximum analogue input signal level (sensitivity) on this input before the ADC (Analogue- to-Digital converter) signal path clips. Options are 1, 2 and 4 volts RMS maximum input. The default is 2Vrms maximum.
For example, analogue sources with low output levels may benefit by choosing the 1V maximum setting. This helps maximise signal-to-noise performance of the Receiver and also helps keep the various analogue sources sounding about the same level for any given Receiver volume control setting."

 

[Vacceo]

I don't think so because it's all about priorities - all products, from sports cars to washing machines, have to decide which features are more important than others as sometimes they are mutually exclusive because elevating the performance of one aspect necessarily attenuates something else. If your focus and emphasis is the all might SINAD measurement, then too much DSP necessarily reduces some of that signal and yet room correction is essential when you have a multi-channel 9.2.4 movie setup! Additionally, dedicated home cinema setups are very different than multi-channel music because the audio source for movies is less about "audiophile" lossless zero distortion music playback and all about creating the immersion from subwoofer power to ambient cues to channel separation to height effects. The crazy level of acceptable distortion during explosions and action sequences where the SPL is over 82db (with peaks over 95db) means the AVR/PrePro really doesn't have to spend much money on best in class SINAD because what's the point? The louder you play, the less audible the distortion!

It's like complaining that one's pimped out Toyota Landcruiser can't run as fast as an el cheapo 20 year old Miata through tight hairpin turns. Not to say it isn't possible to create a dynamic suspension system to do this but that's why the Lambo Urus is so expensive but it still has limitations despite the premium paid, I mean, what about fuel efficiency - I want at least 35MPG from my Urus if I'm paying $200K right?!

AVR's are inhently limited by design due to dsp's, sure. That limitation is a necessary evil because 7, 9, 11 and more channels do indeed need adjustment.

However, the basic rule of hi fi stays no matter the content: play the source as close to what is there. True, when the content goes all explosions, distortion is not as terrible as playing music, but at this time and age, we should expect a minimum of clean reproduction. Keep also in mind that these systems are often used with a multitude of content, so certified clean stereo or just TV news (or radio/podcast) is not being picky.

While expecting 115 db sinad (I'm sure even Benchmark would have issues reaching that in an AVP/AVR), room correction, a multitude of connections and all the extras you can add at a low price is unfeasible; I think we can agree that a certifiable clean sound with a sinad around CD quality (96 db, but to give a nice, clean figure, let's push it to 100 db) can be expected and looked for. It has been archived, so not a hard ideal to stick to.

Considering what you do, that assurance comes with an extra: you can judge how good or bad the AV content has been mastered. Discounting the gear as a source of wacky sound (or image) is pure peace of mind.

 

[GXAlan]

I voted poor with an eye on not terrible.
For the size and functionality of the thing I suppose it would be attractive to some, but the value to price seems to be inflated 3 or 4 thousand dollars. I'm just a basic 19th century stereo guy, so this is not kit I would buy.

Is there one processor and a stack of cables and amplifiers that can meet or exceed the performance and price without giving up anything? I honestly don't know.

Right now the X8500H is pretty darn good with the A110 if you can find it, a nice upgrade.

I think everyone is waiting for 48GBps HDMI chips before taking the next step.

This doesn't answer why internal amps are allowed to run hard even when you have an external amplifier.

In theory, it's as simple as a relay to provide power/no-power to the amplifier section, right?

 

[Jbrunwa]

I think everyone is waiting for 48GBps HDMI chips before taking the next step.

According to Harman support, their HDMI 2.1 board supports all HDMI 2.1 specs including speeds to 48 gpbs, although I have no way to independently verify.

 

[KMO]

This doesn't answer why internal amps are allowed to run hard even when you have an external amplifier.

That's just saving being cheap and not having extra mute/cut circuits, I guess?

Presumably, just muting the input to the on-board amp would suffice. The current top-end Denons show it can be done.

And why they allow higher voltages but at severely distorted conditions.

What do you mean by "allow"? What's the alternative? Hard clip rather than distort?

Or do you mean they shouldn't allow the gain to be turned up that high - greater than unity?

I'd say it's better to have the option of high gain for low-level signals than to limit gain to make distortion on a max-level signal impossible.

If you don't want that, then I think basically all AVRs will let you set a maximum master volume, so you could block approaching or exceeding 0dB.

 

[Spocko]

...
Considering what you do, that assurance comes with an extra: you can judge how good or bad the AV content has been mastered. Discounting the gear as a source of wacky sound (or image) is pure peace of mind.

You're correct, sometimes ignorance is bliss as you can only tell if a movie track is "wacky" if you have a reference to compare it to! I'm absolutley a big fan of well engineered gear and you bring up a great point about discounting the "source of wacky sound". I'm currently reviewing some $2K flagship soundbars with wireless surround satellites and a sub (Samsung Q990B, Sony A7000, etc.) and they're shockingly great values when I listen to them without any context. However, when compared to my reference Trinnov/Genelec 9.2.4 system - their biggest shortcoming is the clarity/smearing/muddiness of certain high bass and midrange frequencies when listening to all three, 5-second snips back to back to back (yes, my review will include these clips for viewers to assess); this issue is likely due to the lack of room/speaker correction. But then, without a reference system, I wouldn't have noticed these shortcomings as they are quite impressive for such affordable packages.

 

[Jbrunwa]

I wouldn't buy an integrated processor/amp at this price point from any brand, so this model is not designed for me. But could two of the noted deficiencies be improved by configuration changes?

1. High analog input noise when fed a low voltage input signal. It would be interesting to see if changing the input config input trim setting from the default 2V to 1V would improve the measurement.
2. Changing protection mode from default high to medium. It would be interesting to see if this would inhibit shutdown long enough to complete measurements.

 

[Vacceo]

You're correct, sometimes ignorance is bliss as you can only tell if a movie track is "wacky" if you have a reference to compare it to! I'm absolutley a big fan of well engineered gear and you bring up a great point about discounting the "source of wacky sound". I'm currently reviewing some $2K flagship soundbars with wireless surround satellites and a sub (Samsung Q990B, Sony A7000, etc.) and they're shockingly great values when I listen to them without any context. However, when compared to my reference Trinnov/Genelec 9.2.4 system - their biggest shortcoming is the clarity/smearing/muddiness of certain high bass and midrange frequencies when listening to all three, 5-second snips back to back to back (yes, my review will include these clips for viewers to assess); this issue is likely due to the lack of room/speaker correction. But then, without a reference system, I wouldn't have noticed these shortcomings as they are quite impressive for such affordable packages.

You made me remember something interesting about references. It was not till I listened to some of my fave records on a studio's Genelecs that I truly realized that some of those records could have been better done.

Of course, those "bugs" have not changed the enjoyment of those records. I'm sure that you will like Event Horizon on an amazing OLED with the Trinnov and the Genelecs, but you can critically asses all the visual and sound mistakes.

I think that is what many mistake: the source can be enjoyable and artistic even if the mastering could have been better executed. That is precisely what good engineering gives you, the capacity to asses.

 

[Spocko]

You made me remember something interesting about references. It was not till I listened to some of my fave records on a studio's Genelecs that I truly realized that some of those records could have been better done.

Of course, those "bugs" have not changed the enjoyment of those records. I'm sure that you will like Event Horizon on an amazing OLED with the Trinnov and the Genelecs, but you can critically asses all the visual and sound mistakes.

I think that is what many mistake: the source can be enjoyable and artistic even if the mastering could have been better executed. That is precisely what good engineering gives you, the capacity to asses.

heh, this is why I love FM car radios so much - you forget about audio gear "performance" and only focus on the music. I was soooo happy my rental moving truck had a built-in radio for me and my son to enjoy tunes while we drove for hours. I just prayed for music with no static!

 

[beagleman]

That's just saving being cheap and not having extra mute/cut circuits, I guess?

Presumably, just muting the input to the on-board amp would suffice. The current top-end Denons show it can be done.



What do you mean by "allow"? What's the alternative? Hard clip rather than distort?

Or do you mean they shouldn't allow the gain to be turned up that high - greater than unity?

I'd say it's better to have the option of high gain for low-level signals than to limit gain to make distortion on a max-level signal impossible.

If you don't want that, then I think basically all AVRs will let you set a maximum master volume, so you could block approaching or exceeding 0dB.

My one AVR goes up to 0db, but allows it to also be set to +15db, which is INCREDIBLY high and loud......

It can also limit the volume to lower levels below zero.

 

[dlaloum]

AVR's are inhently limited by design due to dsp's, sure. That limitation is a necessary evil because 7, 9, 11 and more channels do indeed need adjustment.

However, the basic rule of hi fi stays no matter the content: play the source as close to what is there. True, when the content goes all explosions, distortion is not as terrible as playing music, but at this time and age, we should expect a minimum of clean reproduction. Keep also in mind that these systems are often used with a multitude of content, so certified clean stereo or just TV news (or radio/podcast) is not being picky.

While expecting 115 db sinad (I'm sure even Benchmark would have issues reaching that in an AVP/AVR), room correction, a multitude of connections and all the extras you can add at a low price is unfeasible; I think we can agree that a certifiable clean sound with a sinad around CD quality (96 db, but to give a nice, clean figure, let's push it to 100 db) can be expected and looked for. It has been archived, so not a hard ideal to stick to.

Considering what you do, that assurance comes with an extra: you can judge how good or bad the AV content has been mastered. Discounting the gear as a source of wacky sound (or image) is pure peace of mind.

All the main AVR's seem to achieve CD grade Signal/Noise, and a bit more.... (so circa 100db+) - the issues seems to be Jitter and Harmonics - Harmonics should not be a problem, even in an AVR.... and Jitter is purely in the digital domain, and has long been demonstrably resolved...

Yes 100db SINAD should be far more common in AV PrePro's and the pre stages of AVR's

 

[jam]

Thanks for the two days of hard work you put in to review this AVR Amir, it's very much appreciated. Alas, such a disappointment from JBL/Harman on most fronts.

Considering that the main use of an AVR is home theater and that folks that are looking for the royal immersive treatment have to resort to the pre-outs, where they're getting only 84dB from the HDMI inputs at max volume (2V) or 0.7V to 0.9V for a somewhat decent 100dB, which is insufficient to drive many external amps to adequate levels, while Blu-ray HD audio codecs are mastered at 24 bits of dynamic range and this unit only delivers from 14 to 16 bits of resolution, you end up also loosing too much definition.

Taking into account the overall poor measurements, the technical issues or the implementation bugs and the high sticker price, this AVR doesn't make the cut in my book.