DIYMobileAudio.com Car Stereo Forum banner

Amp S/N ratio- diminishing returns

5.5K views 64 replies 15 participants last post by  Fizzy  
#1 · (Edited)
Brax has the new amp, D class 125 s/n ratio. Their GX line A/B class, 120 s/n ratio. In percentage terms, 4% ‘more’. 4 channels of the D class might end up being say 7k. 4 channels of the GX, 2.5k. Big difference for 4% and some more efficiency when really it’s the subs that are the hungry hippo’s. Those new amps do not seem worth it.
Also, sub amps. If doing a front sub, does the amp matter as in trying to match the highs and mids, or just throw in a decent D class
 
#4 ·
Oy....so first, at what wattage did they measure that SNR at? Often (some say damn near always) manufacturers post SNR at max output. If that's the case, it's a worthless spec. What you want to know is what the SNR is at 1 watt. SNR = hiss at low volume which is where it comes into play. Add in a high efficiency driver (think hlcd's with 110+ db efficiency and this become a huge deal).

As for damping factor: Marketing crap. As soon as you toss actual speaker wire into the equation that is. Countless articles talking about the mythology of damping factor and audibility.
 
#10 ·
"State of the Art" but yet they publish the specs like this:
Signal-to-noise ratio analog input> 125 dB (A-weighted)

Notice the lack of at what power level in that spec.

Hey, I'm just the messenger here. Most of the amp manufacturers inflate their specs with dubious measurement conditions. It's why I spent almost a decade trying to find a low power amp with no audible hiss on my horns. MOST people don't care because they don't use compression drivers (or pro audio drivers such as old JBL midbass) where this becomes a nightmare.

If you want to be snarky, go spend the obscene amount of money I'm sure those amps are being marketed at and HOPE they truly have that high an snr measurement.

Me? I use their DSP Ultra, but that's where the buck stops. Not playing the game of inches with their Brax DSP (don't get me started about audibility differences of dacs in a moving freaking car) nor their overpriced "audiophile" amps.

Want to hear something heartbreaking? The ONLY amp I could find that had a truly inaudible SNR were two modified Soundstream Reference 200 and 300 amps. I've heard rumors of the ultra pricey "true" (if it exists in car audio that is) class A amps that match the SNR of these two amps, but I'm not paying crazy prices for what I have achieved for under $800 for 3 amps.
 
#6 ·
I'm going to go out on a limb here, but I highly doubt these new Brax amps will have any hiss caused by the amp.
 
#8 ·
I'm curious of the price, probably we'll outside my budget for amps.
 
#18 ·
McIntosh would give specifications at a worst-case scenario from .25 watts to rated power and their specification were still better than everyone else's best specs. That's the most useful specification as they weren't playing the best performance game but the worst. That's the way it should be. No one is playing these amps at max output, where many specs are measured, that cares about SQ. Having said that, you bet these manufacturers play the specification game and I assure you, you can't hear the difference between 100 dampening and 10 million, or 0.1 and 0.000001. or 100 S/N and 140. It's just like TV specification marketing and they tout specification above the human eye's perception much as amp manufacturers do for hearing perception. They are just trying to convince you that you need to spend money for the inaudible difference and let placebo differences "psychoacoustics" dictate the differences. It might be a big difference in numbers i.e. 4,000 vs infinite contrast, but you'll never be able to A/B the differences. Hi-Res audio same thing. Also recording studios will master Hi-Res tracks a little difference so you think the difference is related to quality. Buyer beware, it is definitely a marketed solution for a problem that doesn't exist.
 
#19 · (Edited)

POST #9

and...

GORDON TAYLOR's NOTES (The Amp Doctor, U.K.) regarding a Modified SOUNDSTREAM REF 300 Amplifier that ErinH purchased second-hand and used in his old Honda Civic install to power his 5" BMS 5CN162 Concentric midrange/tweeter drivers that had 1.5" HF Compression Drivers with a Sensitivity of 110dB SPL @ 1W/1M, and Erin was having issues with Significant Background Hiss/Noise Floor from the HF compression drivers using his other "typical" car audio amplifiers:


Gordon: "The [SoundStream REF 300] amp is now set up for Horns - I've dropped the power amp gain to improve Signal-to-Noise Ratio which is now nudging 100dBA reference to 1 Watt. As standard, this was around 86dBA, so the 14dB improvement is around a 30x Reduction In Noise.

The preamp gain has also been reduced so you should be able to match levels with your midbass drivers when fed from a similar signal level. It's got one of the new LME chips in the preamp too and bipolar caps.

The Distortion Level at 1 Watt is an infinitesimal 0.0016% as it's still mostly running in Class A at that point. 1 Watt equates to pretty high SPLs with horns (106dB - 110dB SPL @1W/1M), so any distortion won't be coming from THIS Amplifier, and at anything but extreme listening levels you will be enjoying the warmth of Class A through the horns.

I put new connection screws in the 300 as the old ones were rusty and burred. And some pretty blue LEDs."


As is Common Practice, UNLESS they specify otherwise, 99% of manufacturers list their SNR Specs At Full Rated Power.

Given that, IF the BRAX Amplifiers in question have a listed SNR spec of 125/120 respectively, at 1 Watt they will probably be in the region of 14dB-15dB less, which would be 110 SNR & 105 SNR respectively.

THAT is Class-Leading SNR, and assuming proper gain-staging, IF you had ANY Floor Noise in your system, it will not be from these amplifiers.
 
#20 ·
Regarding DAMPING FACTOR, here is my Reference from Benchmark Media Systems' John Siau, that in a joint project with THX, designed and manufactured the AHB2 amplifier, which is one of THE finest audio amplifiers currently available.



Here is an excellent YouTube Livestream interview with John Siau of Benchmark Media Systems discussing the unique design principles of their AHB2 amplifier...

 
#28 ·
Thats a great article that I have read before and DF definitely matters in my opinion. What people seem to forget is yes there may be some marketing involved in this but most of these high end manufacturers are extremely passionate audio people themselves. Now while you have some companies like Critical Mass that are all fluff just trying to get your money you then have companies like Brax, Audiowave, Celestra, Phass, Genesis etc that are literally trying to produce products of the highest order and they don't need or try to achieve crazy high numbers for marketing. They are trying to produce equipment capable of giving the best playback as possible and things like high DF and SNR are a byproduct of this. While it is true that many manufactures are giving SNR A weighted at rated power I don't see a issue as this is the norm in the industry and if I see a amp with a SNR of 120dba I know that I can account for that by taking 15 to 20 db off that number to get a good idea of the power at 1 watt which anything in thar range is phenomenal. It just really comes down to being a informed buyer in this hobby. Bbfoto I just realized I was responding directly to you above everything after agreeing with this article that you posted wasn't directed at you mostly toward gentleman above.
 
  • Like
Reactions: RNBRAD and bbfoto
#21 ·
#23 ·
Yeah, amplifiers using the Purifi Eigentakt Class D modules are about as good as it gets in regards to specs & performance. Multiple big name home audio manufacturers are using them in their flagship products, such as the NAD Masters M33, M23, & M28.

ErinH is using a Pair of the March Audio P501 mono-block amplifiers in his home setup to drive his MoFi SourcePoint 888 tower speakers, as well as being his reference amplifiers for all of the loudspeaker listening tests & measurements for his website and YouTube reviews at Erin's Audio Corner on YouTube .




I'm currently using the Nilai500DIY 250W Stereo Power Amplifier Kit from DIYclassD in my home setup. With the Euro to USD exchange rate at the time it was about $1350 USD. You can easily assemble the amplifier kit in an afternoon, and there is no soldering or special tools required to do so. You can check out the "Assembly/Build" YouTube video for the amp shown below if interested.

Hypex Nilai® Technology at DIYClassD

I began a conversation with Mike at Biketronics regarding the possibility of building a 12vdc PSU for the Nilai500 modules, and he seemed positive and interested in doing so. But life got busy and I never followed-through or followed-up with him about it. I might check with him again in the fall when his business slows down again.

For whatever reason, between the Purifi Eigentakt and the Hypex Nilai500 amplifiers I've heard to date, I've preferred the latter sound-wise...but I admit I haven't done any proper blind A/B/X listening between the two in order to substantiate or qualify my preference.


 
#22 ·
What a post, at the end of the day all you need is a non colored (yea thats a thing too) amplifier from a reputed brand which you can run at 70-80% of the required power and let the tuning do the rest.

Either its old school class a/b or modern class d amplifiers. At a certain value for money the price rise is just not worth the return you are getting. Its just for people with thick wallet to show off their money.
 
#25 ·
Another thing too, a lot of people bridge their amps which changes these specifications. They can certainly drop into that audible range. One example, some of you all buy these class-d mono amps and they may have 50-100 dampening factor at 4ohms, but this rating is cut in half for every half drop in ohms. This is where people can get critically low in DF where it can certainly be audible dropping these amps to 1 or 0.5 ohms. However, the user may not care.
 
owns 2010 Mercedes W212 E550
#26 · (Edited)
Quite a few Class D car audio amplifiers are already configured internally in BTL (bridged) mode and still have fantastic specs and performance. This is especially true for a lot of the Class D amplification used in DSP+Amplifier combination units (some of the A-F HELIX DSP-amplifier models, etc.), and is why they can't be used in a bridged configuration by the end user.

In regards to Distortion, the majority of people who are bridging Class D amps will do so when powering subwoofers or midbass drivers where the distortion spec of the amplifier is much less significant due to our lower threshold for detecting distortion as frequency decreases, i.e. ~10% THD for bass frequencies vs ~3% THD for lower midrange through the upper treble region.

And IMO, at least for a spectrally balanced SQ-centric system, there is absolutely ZERO need to run your amplifiers down to 1 or 0.5 ohm loads. If you want a ground pounder to play at ear-damaging SPL levels, low distortion and "clean/fast/articulate bass" that might be achieved via a high damping factor is a moot point. :p
 
#27 ·
Quite a few Class D car audio amplifiers are already configured internally in BTL (bridged) mode and still have fantastic specs and performance. This is especially true for a lot of the Class D amplification used in DSP+Amplifier combination units (some of the A-F HELIX DSP-amplifier models, etc.), and is why they can't be used in a bridged configuration by the end user.

In regards to Distortion, the majority of people who are bridging Class D amps will do so when powering subwoofers or midbass drivers where the distortion spec of the amplifier is much less significant due to our lower threshold for detecting distortion as frequency decreases, i.e. ~10% THD for bass frequencies vs ~3% THD for lower midrange through the upper treble region.

And IMO, at least for a spectrally balanced SQ-centric system, there is absolutely ZERO need to run your amplifiers down to 1 or 0.5 ohm loads. If you want a ground pounder to play at ear-damaging SPL levels, low distortion and "clean/fast/articulate bass" that might be achieved via a high damping factor is a moot point. :p
Quite a few Class D car audio amplifiers are already configured internally in BTL (bridged) mode and still have fantastic specs and performance. This is especially true for a lot of the Class D amplification used in DSP+Amplifier combination units (some of the A-F HELIX DSP-amplifier models, etc.), and is why they can't be used in a bridged configuration by the end user.

In regards to Distortion, the majority of people who are bridging Class D amps will do so when powering subwoofers or midbass drivers where the distortion spec of the amplifier is much less significant due to our lower threshold for detecting distortion as frequency decreases, i.e. ~10% THD for bass frequencies vs ~3% THD for lower midrange through the upper treble region.

And IMO, at least for a spectrally balanced SQ-centric system, there is absolutely ZERO need to run your amplifiers down to 1 or 0.5 ohm loads. If you want a ground pounder to play at ear-damaging SPL levels, low distortion and "clean/fast/articulate bass" that might be achieved via a high damping factor is a moot point. :p
I would agree for "SQ-centric", but for me, I've never listened to my SQ curves for competition. I always like a little more sub and midbass than the "ideal" curve. I like the tactility of music and it doesn't need to be loud but it certainly takes more power. Power is cheap and today's subs are capable of dipping well into the subsonic frequencies. IMO there's nothing like and there never has been. When I first purchased my SVS SB4000 sub for my home, I knew I wanted that sound and capability in my car.
 
owns 2010 Mercedes W212 E550
  • Like
Reactions: bbfoto
#31 ·
ESS QM chips are only for digital to analog conversion. Analog to digital conversion is a different chipset.
 
owns 2010 Mercedes W212 E550
  • Helpful
Reactions: LowBassSubs
#34 ·
So sounds like the Brax DSP with the updated ESS QM DAC’s, then going analogue into say the GX line amps or the new Pro amps (choice analogue not digital) is optimal. I saw an install online using the Brax gear and the choice was analogue into the Brax DSP to use the DACs from the Astell & Kern SP3000- and it just got me turning over in my head the choices and I knew I had a blind spot, but it’s a blind spot, I couldn’t see the full picture
 
#35 ·
The best scenario is to use a pure direct digital signal into the Brax. If you go analog, that means if you are using a digital source, it has been converted from digital to analog from another piece of equipment. This should be avoided, especially if you have a factory system or an aftermarket CD player and you’re sending speaker level or RCA input in the Brax. Essentially doing this is negating all of the high end components of the Brax and you are relegated to the sound quality of the cheapest DAC in the chain which certainly isn’t the Brax. The DAC conversion you want to happen only 1 time, preferably in the Brax DSP, if not there, 2nd choice would be a Brax amp with high end DAC chips. This process is the most important process for SQ and these details matter. A person starts running multiple equipment with multiple DAC’s, you just add more noise and distortion with every conversion process. Digital input into the Brax eliminates the Analog to Digital conversion process as well.

Some people can’t go digital into the Brax, and if they can’t, I would expect a high end DAC used in the source unit. If there isn’t a high end DAC, I would say it’s a waste of money on the Brax because it’s going to sound like the weakest link.
 
owns 2010 Mercedes W212 E550
  • Like
Reactions: Fizzy
#37 ·
Since no one touched on your math I'll jump in here and post about your original question...

5dB is not 4%

A decibel is a logarithmic unit so each unit is worth much more than the mathematical percentage of the total or absolute value of the total.

Here's a good reference chart and why we strive for every dB of improvement that we can obtain within the realm of cost:

Image


So +5dB is 3.16x times better.

Now, is that beneficial in terms of what you can hear?

Debatable. And this is where the point of diminishing returns comes into question. Additionally, what kind of dynamic range can the rest of your system support and if you are only listening at 95dBZ average then you anything more than 98-100dB SNR is going to get lost in the signal and the background or ambient noise of the listening environment.

Additionally, none of this addresses the issue of what's the possible dynamic range of the digital source cause unless you are using exclusively recorded, mixed and mastered 24 Bit recordings then you are limited to 96dB by the 16 Bit format.
 
#39 ·
By the way, prices have been released for the Brax rx2 pro in Europe

3900 euros for the normal and 4400 from memory for the limited edition founders edition amps, plus 300 or 400 for the analogue or digital input card

The digital having the Ess dac built into it that also does disac2, it works in a slightly different way to the original disac due to the architecture of the Ess dac, technically better than the original disac as found in the mx4 pro 👍🏼

I have 2 rx2 pro coming to me when they are finally built

fun fact… the audio precision apx515 that AF use for every other product isn’t clean enough to measure the Brax DSP with Ess or the rx2 pro, they had to buy several apx555 for that… otherwise they were measuring the output of the audio precision itself effectively
 
#40 · (Edited)
I have all sorts of recordings. 16/44, 24/44, 24/96, 24/192. Some of these are my comfort food music, stuff that never gets old. I listen to the radio also as flicking through I find some country song or chill out station which is outside the body of my recordings. I know there’s Hi-res radio like Mother Earth and many others so could use a small tablet as a Hi-res radio source
 
#42 ·
Just remember that you overall system SNR will be predicated by the weakest link in the chain and that is often the source and quality of the original recording.
 
  • Like
Reactions: Stycker