• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Upgrade Your Speakers

Status
Not open for further replies.
Joined
Jan 28, 2021
Messages
845 (0.71/day)
I've been out of audio for awhile but have been getting back into it and though YouTube found GR Research is selling DIY upgrade kits for many commercially available speakers. Klipsch, B&W, Polk, ect.

A bit of an introduction to the concept...

Speaker upgrades are not a new concept for anyone into the DIY aspect of audio. All speakers are built to a price point so unless you are spending 5 figures there is always going to be room for improvement with better components, cabinet modifications and crossover network changes, the opportunity has always been there. The issue is most people with the means to analyze a speakers shortcomings and address them would just be better off building something better from scratch, which is what most of what the DIY market is. Publishing these videos and making the kits available opens up this concept to tons of people that want to upgrade what they have, get into a new level of sound quality that would normally otherwise be financially out of reach, and learn something in the process.

About GR Research...

GR Research is one of the companies out there thats been selling kits, raw drivers for quite while, and (in the past at least) did design work for larger speaker manufactures. I'm not sure who builds the drivers but I believe they are their own design. They also sold their speaker designs in partnership with AV123 which got good reviews and praise for their value but many may be more familiar with due to their dramatic impulsion. Their kits and designs have always been well regarded from what I remember (the AV122 thing happened right when I was getting into audio) and it looks like they are still selling some of those same designs from then as kits.

It looks like there is a pretty huge library of upgrade videos on YouTube for various speaker upgrades from budget to high-end, and some vintage stuff as well. They sell a lot of the more popular redesigns (popular speakers that make good candidates) as kits but you can follow his work from the videos or it looks like if you reach out he's willing to work on something if you want it upgraded. Seems like a pretty good way for someone make a pretty significant upgrade to their sound and learn something in the process. All you need is a soldering iron, the ability to read and electrical diagram, and the time.

Fixing a popular commercial design...

This is a long form video of the entire process of upgrading the Klipsch RP-600M (the others are more concise, and look at just the redesign) which is a solidly reviewed mid-range speaker. Its a decent speaker but its pretty clear looking at the crossover components and the measured response that concessions were made to hit that price point and corners were cut during the cabinet construction.

Before and after measurement results... Stock frequency responses are on the left. notice the bump at around 800Hz and the huge dip after 1Khz. This is not a intentional design decision but cost concession for mass produced product and/or done for product segmentation to protect the high-end line. Horizontal and vertical greatly extends the dip in response at the crossover point as the phase issues are exaggerated.

Measurements with new the crossover design on the right, all the response issues resolved, off axis response greatly improved. Impedance is also smoothed as well as improvements to the spectral decay of the speaker.


600M 1.jpg

600M 2.jpg



About $300 for a new redesigned crossover topology with vastly improved components, cabinet dampening, cable and connectors. It will sound similar in sense that its using the same drivers but correcting the crossover flaws and upgrading components and you will end up with essentially a new speaker that would be several price tiers up.

***EDIT TO ADD BEFORE AND AFTER RESPONSE PICS BACK IN***
 
Last edited:
Joined
Jul 15, 2022
Messages
696 (1.05/day)
Ultimately, the most important of all is software. Both the audio stack of your operating system has a strong impact, as well as its settings, as well as the exact audio app and the settings of this app.

While most Klipsch speakers are better than the F&D F550X from a purely technical point of view, my F&D F550X sounds much better in reality because I use FreeBSD + bit-perfect mode + vchans disabled + Audacious + 24-bit tracks.
This sounds of a completely different level than what can come out of windows wasapi.

Here is my sound demo of these speakers: https://drive.google.com/file/d/1jRYGgpwM24xgAZFJhyaqw_NgRFBYMMo5/view

The brands popular with my acquaintances are mainly B&W and Infinity, but none of their speakers sound more clinical than my cheap F&D. I'm sure it has to do with the quality of the source signal going to the speakers.
 
Joined
Jun 2, 2014
Messages
416 (0.11/day)
Location
Midwest
System Name Core
Processor Intel 12700k @ 5.0/4.0
Motherboard ASRock z690 Steel Legend
Cooling Artic Cooling Freezer 420 AiO
Memory GSkill 64GB 3200 cas 14 b die
Video Card(s) Asus Nvidia RTX 4070 Super OC
Storage Optane 900p x2, SK Hynix p41 Pro
Display(s) ACER 250hz 1080p 25" IPS display x2
Case Phanteks p500a with all Noctua Chromax/Arctic fans
Audio Device(s) Focusrite interface, Presonus Studio Monitors and Subwoofer
Power Supply Seasonic 850w plat with cable mod cables
Mouse Glorious Model O
Keyboard Corsair mech k65
Software Win 11 Pro
Benchmark Scores 3dmark TimeSpy 20240-rtx 3090FE/12700k/Optane 3dmark TimeSpy 21862-rtx 4070Super/12700k/Optane
I considered GR mods, but then I go against what Klipsch intended in their product and well, I can just EQ the dip out, which I have done and I still prefer it off, with the dip. I up/side graded from Kef Q 150s because I wanted more dynamic sound and Klipsch delivers. That said, I don't deny upgrades can help, but you want some balance between what the equipment says and your brain/ears are listening to.

As for Windows tweaks, wow, you have quite a few. I do think some of those will surely help the quality. I use Foobar>fibre optic to Schiit Dac>unbalanced cables>Yamaha 2 Channel Amp/Plate sub amp>custom 10ga speaker cables to Klipsch RP500Ms and a Kef subwoofer. Id like to do some more tweaking...

If Im going to serious listen to music, I go into my home theater and listen to Mission Argonauts and 4 twelve in subwoofers, lol. Thats my reference.
 
Joined
Dec 8, 2020
Messages
2,777 (2.23/day)
I have researched a design that eliminates as much analogue as possible, in short, working with and pushing digital all the way up to the speaker driver, and then a PowerDAC.
It's a little bit complicated, but if you would like to know more about the idea and some of the specifications, I could post more about it if you like.

A short version, the idea is called 'Smart Speakers', which uses optical from the main digital amp to each speaker, plus power out (separate).
The main unit also accepts optical in at the current 15 channel standard, any amplification will be done in PCM form.

It's possible for OEM's to build a 15 channel main unit, and the consumer purchases the number of speakers required, adding more later is also an option.
 

Attachments

  • Smart Speakers.png
    Smart Speakers.png
    16.8 KB · Views: 65
Joined
Jan 28, 2021
Messages
845 (0.71/day)
I considered GR mods, but then I go against what Klipsch intended in their product and well, I can just EQ the dip out, which I have done and I still prefer it off, with the dip.
The stock crossover has flaws in it, nobody intentionally puts design flaws in a product. Also you can't EQ most of this because the problems lie in the crossover region between the two drivers, not even room correcting DSP can do anything about this. Only breakup modes of the drivers and to a certain extent cabinet resonances can be compensated for with EQ or room correction processing.

I'm not sure what happened to the pics but I'll see if I can get them back in.
As for Windows tweaks, wow, you have quite a few. I do think some of those will surely help the quality. I use Foobar>fibre optic to Schiit Dac>unbalanced cables>Yamaha 2 Channel Amp/Plate sub amp>custom 10ga speaker cables to Klipsch RP500Ms and a Kef subwoofer. Id like to do some more tweaking...
I have no idea what the above post is about but assuming you are starting with good source material (a high bit rate or lossless file) sound reproduction is an electromechanical thing. That means your speakers / headphones, amplifier, and source (DAC, turntable) are responsible for what you are hearing, not software tweaks and they certainly are not overcoming physical and electronic limitations of cheap multimedia speakers.

I have researched a design that eliminates as much analogue as possible, in short, working with and pushing digital all the way up to the speaker driver, and then a PowerDAC.
It's a little bit complicated, but if you would like to know more about the idea and some of the specifications, I could post more about it if you like.
Analog is not the limiting factor so being all digital is not the answer to any of these problems aside from the fact that flaws like whats present in the Klipsch could be corrected via software.
 
Joined
Dec 8, 2020
Messages
2,777 (2.23/day)
I pretty much thought the thread was more genetic than a specific model, I missed that, I apologize.
 
Joined
Jan 28, 2021
Messages
845 (0.71/day)
I pretty much thought the thread was more genetic than a specific model, I missed that, I apologize.
Oh, it is meant to be a more of a generic topic, that Klipsch is just particularly bad offender.

To the general topic at hand going 100% digital is not the answer because whatever you do in the digital domain is binary approximation of something that is inherently analog. Its (digital) is perfect for storing and transmitting data (digital music files, and streams) but as soon as you start transforming it backing into something analog (DAC) everything you do is best effort compromise to get as close as you can to the OG analog signal, and despite what some people think class D does not stand for digital (there is no such thing as digital amp). DSP based crossover networks can be very good and it allows you to iterate different designs as often (and at no cost) as you want but passive (capacitor, inductor, and resistor) crossover networks are not going to be made obsolete anytime soon.
 
Joined
May 13, 2022
Messages
12 (0.02/day)
Processor AMD R7 5700X
Motherboard MSI B550 Bazooka
Cooling Noctua NH-U9S
Memory Patriot Viper 16g @ 3200mhz
Video Card(s) Sapphire Pulse RX 7900XT
Storage Intel 512gb M.2 + Samsung 860 EVO 2TB + Crucial P2 2TB M.2 + Seagate BarraCuda 4TB HDD
Display(s) Sony XBR-55X930D + ViewSonic OMNI 32" 1440p@144hz
Case NZXT
Audio Device(s) Yamaha A780, SVS Prime BS + Prime Center + Prime Sats. Dayton Atmos. SVS PB-1000 + SVS PC-12NSD
Power Supply Gamemax 800 watt
Mouse Steele Series 3
Keyboard Steele Series 3
Software W10
The last thing someone interested in good audio would want is tons of post processing.



A good goal is to get your audio as flat and colorless as possible. Then the recording can do its thing as intended.
 
Joined
Aug 22, 2019
Messages
174 (0.10/day)
System Name Donnager
Processor 13900KS, lapped and contact frame
Motherboard Asus Z790 Hero
Cooling Heatkiller IV CPU block, Heatkiller V GPU block, GTX 480mm radiator, D5 pump
Memory 32GB Kingston Fury 7200C38
Video Card(s) eVGA RTX 3080Ti FTW3
Storage Optane 380GB M.2 OS drive, M.2 2TB game drive
Display(s) Alienware 34" Ultrawide 120Hz 3440x1440
Case Fractal Meshify 2 XL
Audio Device(s) Outlaw RR2150 stereo receiver driving DIY kits, Schiit Asgard for Sennheiser HD6XX headphones
Power Supply Seasonic Prime 1000W
This is a pretty cool area for GR to expand into. I built a pair of their AV1's years ago, and still enjoy them.

Prebuilt speaker manufacturers using value crossover components is no surprise and nothing new. Upgrading select drivers isn't a bad thing either. I do wonder at the value aspect, though. Why not just buy better speakers to start with, or build your own kits? With prebuilt or flat pack boxes, building speakers is easier than ever.

My father is currently upgrading his 1980s era Klipschorns. After several iterations, it has all new crossovers, new midrange and tweeter drivers and horns. The original spec components were very modest by today's standards. Horns and speakers have come a long way since the 1940s. And yes, they do sound better. More efficient, flatter response and less distortion.
 
Joined
Jan 28, 2021
Messages
845 (0.71/day)
The last thing someone interested in good audio would want is tons of post processing.



A good goal is to get your audio as flat and colorless as possible. Then the recording can do its thing as intended.
I know less about that but I think something like Dirac Live via Roon would be fine, its just fancy room correction based EQ, its not doing any processing to the content of whats being played back. Either way its something I'd like to mess with down the road when all the easy hardware upgrades are done.
This is a pretty cool area for GR to expand into. I built a pair of their AV1's years ago, and still enjoy them.

Prebuilt speaker manufacturers using value crossover components is no surprise and nothing new. Upgrading select drivers isn't a bad thing either. I do wonder at the value aspect, though. Why not just buy better speakers to start with, or build your own kits? With prebuilt or flat pack boxes, building speakers is easier than ever.

My father is currently upgrading his 1980s era Klipschorns. After several iterations, it has all new crossovers, new midrange and tweeter drivers and horns. The original spec components were very modest by today's standards. Horns and speakers have come a long way since the 1940s. And yes, they do sound better. More efficient, flatter response and less distortion.
It seems like Danny likes the challenge of fixing some of the commercial designs short comings whether that be cost concessions or design flaws.

All the speakers I own aside from the active monitors on my desk (and I plan on getting rid of those too) are speakers I've built from kits or published designs. I don't think I'd be as interested in audio if it was just stuff was buying off the shelf. Any speaker really is subject to cost compromise but when it comes to mid-level stuff from a huge company like Klipsch in the RP-600 they are going to cut cost wherever they can while still keeping the quality reasonable, and that means the cheapest crossover components you can get and cabinets that require as little effort to manufacture as possible. You just have to hope the underlying design is good (and in this case it isn't). But even if the crossover didn't have flaws compare the quality of crossover components with even an entry level kit like the C-Note I'm building which uses all poly caps, air core inductors, and large sand cast resistors to the Klipsch's iron core inductors, and electrolytic caps, and tiny resistors. The C-Note is a $150 per pair for reference... DIY from the start is clearly the way to go but if you already have something like these this is still probably well worth the effort.

The Klipschorns sound like a fun project. Yeah driver technology has moved on a ton since then, even with old school paper cone drivers like that, and I'm sure any electrolytic caps that were in there have drifted way off their original values. Is he taking measurements before and after?
 
Last edited:
Joined
Aug 22, 2019
Messages
174 (0.10/day)
System Name Donnager
Processor 13900KS, lapped and contact frame
Motherboard Asus Z790 Hero
Cooling Heatkiller IV CPU block, Heatkiller V GPU block, GTX 480mm radiator, D5 pump
Memory 32GB Kingston Fury 7200C38
Video Card(s) eVGA RTX 3080Ti FTW3
Storage Optane 380GB M.2 OS drive, M.2 2TB game drive
Display(s) Alienware 34" Ultrawide 120Hz 3440x1440
Case Fractal Meshify 2 XL
Audio Device(s) Outlaw RR2150 stereo receiver driving DIY kits, Schiit Asgard for Sennheiser HD6XX headphones
Power Supply Seasonic Prime 1000W
We have a calibrated mic setup to get the crossovers right. He's doing the crossover design himself, and has the modeling software to get the filters correct.

Here's where he's getting the guts from. Super nice stuff.
 
Joined
Jan 28, 2021
Messages
845 (0.71/day)
We have a calibrated mic setup to get the crossovers right. He's doing the crossover design himself, and has the modeling software to get the filters correct.

Here's where he's getting the guts from. Super nice stuff.
Man, those are some high roller components, thats gona be sick. Pretty much a completely new speaker just in the spirit of the original.

At some point I'd like to get a calibrated mic and get more involved but for now I'm sticking to kits and proven designs. The highest-end design I've built is the Singularity, which is full range 8" (modern) driver in 90L transmission line cabinet, high-end parts but relatively cheap because there are so few of them. Pretty sure my limiting factor is my electronics (Pioneer A9 Elite integrated) so I'm working on that while making plans for new builds, currently have parts for 6.5" Peerless two way with a NE180W and XT25.
 
Joined
Aug 22, 2019
Messages
174 (0.10/day)
System Name Donnager
Processor 13900KS, lapped and contact frame
Motherboard Asus Z790 Hero
Cooling Heatkiller IV CPU block, Heatkiller V GPU block, GTX 480mm radiator, D5 pump
Memory 32GB Kingston Fury 7200C38
Video Card(s) eVGA RTX 3080Ti FTW3
Storage Optane 380GB M.2 OS drive, M.2 2TB game drive
Display(s) Alienware 34" Ultrawide 120Hz 3440x1440
Case Fractal Meshify 2 XL
Audio Device(s) Outlaw RR2150 stereo receiver driving DIY kits, Schiit Asgard for Sennheiser HD6XX headphones
Power Supply Seasonic Prime 1000W
Yeah....my dad is having fun, lol. Calibrated mics are stupid cheap now. I was surprised. Such a useful tool for dialing in speakers and figuring out the room response.


Nice designs on Curt's page there. I'm a fan of the MTMWW ones.
 
Joined
Dec 8, 2020
Messages
2,777 (2.23/day)
I guess I have a different opinion, since digital form is lossless, id prefer to keep it as lossless as possible, that means the only analogue is right in front of the driver.
A digital amp (using lossless) will always be better than an analogue one (not handling lossless, added THD THD-N to amplify).

1. PC > DAC > Analogue out > Conductive cable (EMI other) > Receiver > Amp (also amps any THD THD-N) > Conductive cable (EMI other) > Speakers.
2. PC > Digital out > Optical cable (no EMI) > Receiver > Amp (digital) > Optical (no EMI) > [PowerDAC > Speaker].


Intel.png

2. will produce better audio, and it should also be cheaper than 1, less parts (grounding other).

A £300 sound card will invest all of that into analogue parts, for better audio........

----

1663933887274.png
 
Last edited:
Joined
Jan 28, 2021
Messages
845 (0.71/day)
I guess I have a different opinion, since digital form is lossless, id prefer to keep it as lossless as possible, that means the only analogue is right in front of the driver.
A digital amp (using lossless) will always be better than an analogue one (not handling lossless, added THD THD-N to amplify).

1. PC > DAC > Analogue out > Conductive cable (EMI other) > Receiver > Amp (also amps any THD THD-N) > Conductive cable (EMI other) > Speakers.
2. PC > Digital out > Optical cable (no EMI) > Receiver > Amp (digital) > Optical (no EMI) > [PowerDAC > Speaker].


View attachment 262733

2. will produce better audio, and it should also be cheaper than 1, less parts (grounding other).

A £300 sound card will invest all of that into analogue parts, for better audio........

----

View attachment 262748
You are operating under misconceptions. How can converting to analog degrade the quality when ultimately its all analog anyway? I mean from from Intel's perspective who isn't in the business of making anything audio related, sure if you run analog out of your $150 motherboard to your high-end amp / speaker setup thats going to degrade quality vs. even the relatively loosely SPDIF output of dekstop boards and using an AVR with a built in DAC.

You can process the signal and simulate everything a passive crossover does actively with a DSP but how cleanly it can do that is still a limiting factor and ultimately not as good as passive corssovers. You can look at the reviews of the KEF LS50 and Wireless version (which I believe is using active crossovers) has a built in DAC, class D on the woofer and A/B on the tweeter, most people seem to prefer the regular LS50.

I'm not sure what the image / article is talking about but it sounds like some kind of proprietary design that closely couples the DAC and amplifier which sounds like its class D = "Power DAC"? Any kind of amplification is analog this is just coupling the amplifier that much closer to the DAC and eliminating analog stages with proprietary digital ones There may be some merit to it but digital still has limitations in how accurately it can simulate an analog signal which is essentially what this is doing. You also still have the distortion of the amplification stage; that dosen't change and is ultimately only as good as the amplifier. Making it all digital up to that final stage is a cool achievement I guess but how much does that really benefit things vs really well designed conventional DACs and amplifiers?
 
Joined
Dec 8, 2020
Messages
2,777 (2.23/day)
From reading it produces crystal clear sound, and won awards, but the original company that designed a full PowerDAC system went bust (interestingly), however the design is slightly different from the idea I have.
Personally, I would prefer a digital amp that works directly with PCM, If it makes audio better in general, even by a small 2%, but costs 1/2 the price, we would be better off.

There are a few PowerDAC's going around that drive speakers directly, using SPDIF-TOSLink, 2.12 Ohm, with minimal distortion.
Speakers with high SPL (>= 90dB) are recommended with the PowerDAC-S, they sell 102 dB SPL, 2 Ohm speakers for it.

I think PowerDAC's are new enough to not know about them, and also not many exist (at this time).

====

A side note about technology, sometimes it comes too early (Game Gear - colour, TV adapter, console adapter), or they go bust but it becomes mainstream later (Ageia) < I owned the original one xD.
 
Last edited:
Joined
Jun 21, 2021
Messages
2,790 (2.66/day)
System Name daily driver Mac mini M2 Pro
Processor Apple Silicon M2 Pro (6 p-cores, 4 e-cores)
Motherboard Apple proprietary
Cooling Apple proprietary
Memory Apple proprietary 16GB LPDDR5 unified memory
Video Card(s) Apple Silicon M2 Pro (16-core GPU)
Storage Apple proprietary 512GB SSD + various external HDDs
Display(s) LG 27UL850W (4K@60Hz IPS)
Case Apple proprietary
Audio Device(s) Apple proprietary
Power Supply Apple proprietary
Mouse Apple Magic Trackpad 2
Keyboard Keychron K1 tenkeyless (Gateron Reds)
Software macOS Ventura 13.6 (including latest patches)
Benchmark Scores (My Windows daily driver is a Beelink Mini S12. I'm not interested in benchmarking.)
I'm not sure what the image / article is talking about but it sounds like some kind of proprietary design that closely couples the DAC and amplifier which sounds like its class D = "Power DAC"? Any kind of amplification is analog this is just coupling the amplifier that much closer to the DAC and eliminating analog stages with proprietary digital ones There may be some merit to it but digital still has limitations in how accurately it can simulate an analog signal which is essentially what this is doing. You also still have the distortion of the amplification stage; that dosen't change and is ultimately only as good as the amplifier. Making it all digital up to that final stage is a cool achievement I guess but how much does that really benefit things vs really well designed conventional DACs and amplifiers?

While this might be interesting in terms of proof of concept, I don't see the appeal to Joe Consumer. Whether one can do something and whether it is commercially viable are two separate things.

Let's face it: today's contemporary popular music is written, performed, recorded, mastered, distributed, and converted back into analog sound waves with the expectation that Joe Consumer will likely be using a cheap pair of wireless earbuds via Bluetooth Audio from their smartphone or their car's entertainment system while they are doing something else.

TPU's Audio, Video & Home Theater subforum is littered with various discussions about ways to send multi-channel high-resolution audio signals down whatever physical interface but Joe Consumer doesn't really care. 256K AACs from the iTunes Store (as it was formerly known) was good enough and today whatever standard the streaming services use seems to keep Joe Consumer happy.

Even if you can hear the difference between uncompressed and 320K audio, 99.99% of the music listened to today won't be in conditions that really allow the listener to discern the nuances.
 
Joined
Jan 28, 2021
Messages
845 (0.71/day)
From reading it produces crystal clear sound, and won awards, but the original company that designed a full PowerDAC system went bust (interestingly), however the design is slightly different from the idea I have.
Personally, I would prefer a digital amp that works directly with PCM, If it makes audio better in general, even by a small 2%, but costs 1/2 the price, we would be better off.

There are a few PowerDAC's going around that drive speakers directly, using SPDIF-TOSLink, 2.12 Ohm, with minimal distortion.
Speakers with high SPL (>= 90dB) are recommended with the PowerDAC-S, they sell 102 dB SPL, 2 Ohm speakers for it.

I think PowerDAC's are new enough to not know about them, and also not many exist (at this time).

====

A side note about technology, sometimes it comes too early (Game Gear - colour, TV adapter, console adapter), or they go bust but it becomes mainstream later (Ageia) < I owned the original one xD.
What is a "PowerDAC" though? It sounds like just a closely coupled DAC to some sort of amplifier topology, almost certainly class D. At some point it has to be make the transition to analog and I get them calling it a "Power DAC" but its still an analog amplifier at its output stage no matter what. Its a interesting concept but beyond that if it was the way forward others would have adopted it. The audiophile market has no limits when comes to spending money on that last .1% of perceived quality.

Ultimately its kinda pointless to try to keep everything digital down the path as long as possible when there are gains to be made literally everywhere else. Here are two of the latest Class D amp designs so there is plenty of progress to made before we try to tie everything into a single unified topology.
TPU's Audio, Video & Home Theater subforum is littered with various discussions about ways to send multi-channel high-resolution audio signals down whatever physical interface but Joe Consumer doesn't really care. 256K AACs from the iTunes Store (as it was formerly known) was good enough and today whatever standard the streaming services use seems to keep Joe Consumer happy.
I think thats why I post here and not on audio forums or Reddit. Everyone there is already in their associated camps of "high-end audiophile gatekeeping", or the "only thing that matters are measurements of the AP 5000". For someone that wants to get into something higher-end or back into (I grew up with a pretty mid/high-end system in the 90s) making sense of the cluster fuck of a landscape of subjective opinion and technical measurements is impossible if you try to get started from either one of those entrenched perspectives.
 
Last edited:
Joined
Jun 21, 2021
Messages
2,790 (2.66/day)
System Name daily driver Mac mini M2 Pro
Processor Apple Silicon M2 Pro (6 p-cores, 4 e-cores)
Motherboard Apple proprietary
Cooling Apple proprietary
Memory Apple proprietary 16GB LPDDR5 unified memory
Video Card(s) Apple Silicon M2 Pro (16-core GPU)
Storage Apple proprietary 512GB SSD + various external HDDs
Display(s) LG 27UL850W (4K@60Hz IPS)
Case Apple proprietary
Audio Device(s) Apple proprietary
Power Supply Apple proprietary
Mouse Apple Magic Trackpad 2
Keyboard Keychron K1 tenkeyless (Gateron Reds)
Software macOS Ventura 13.6 (including latest patches)
Benchmark Scores (My Windows daily driver is a Beelink Mini S12. I'm not interested in benchmarking.)
Ultimately its kinda pointless to try to keep everything digital down the path as long as possible when there are gains to be made literally everywhere else.

Again, this comes down to commercial viability.

Yes, of course it's possible to put the DAC in the speaker housing so it is as close as possible to the speaker elements (tweeter, woofer, mid-range, whatever) but that A.) increases the speaker cost and B.) reduces compatibility with existing A/V equipment that have standard analog speaker outputs.

And can Joe Consumer hear the difference whether the DAC is in the speaker a foot away from the tweeter or if the DAC is in the A/V receiver (or TV set) a few feet farther away?

Then take for example a 7.1 speaker array. That's 8 discrete DACs requiring separate physical connections (digital audio signal transport and some sort of power line for the built-in amplifier). Sure it's possible but not likely going to appeal to Joe and Jane Consumer.

In a similar way, it's fairly easy/inexpensive to send lossy digital audio wirelessly via Bluetooth Audio to consumer earbuds. Is it possible to send 192kHz/24-bit lossless audio wirelessly? Sure, there are probably prototypes sitting in labs somewhere but it's not going to be at a price point appealing to Joe Consumer. The DACs will probably be too expensive and battery life might be a problem. And are those little tiny earbud drivers going to reflect the benefit of pure lossless audio in the typical listening conditions that Joe Consumer will be using them? Probably not.

There are lots of things that are theoretically possible with existing technologies but most simply don't have the practicality or wide appeal necessary to make it in the consumer marketplace. It's clear that not everyone gets that whether it be here at TPU or elsewhere.
 
Last edited:
Joined
Dec 8, 2020
Messages
2,777 (2.23/day)
PowerDAC-S - ECdesigns - A power DAC example (NO AMP), as I said its new, and there are not yet many of them. Its still backwards to promote full analogue over minimal analogue.
If in anyway analogue was better, then why are we using DSP's to process audio, and not an analogue equivalent (ASP)?

Original: Wadia Digital – Power DAC / Digital Amplifier | HFA - The Independent Source for Audio Equipment Reviews (hifi-advice.com)

----

Reiteration of the image I posted from Texas Instruments, in relation to Class D, where we can see the above is an improvement, and my idea another (P-DAC in the speaker).

1663933887274.png

Notice the 'Digital Domain' reference in both (Wadia). As you said speakers need analogue, but the above is as close as you will get to lossless digital speakers.

----

Attenuator (electronics) - Wikipedia "An attenuator is an electronic device that reduces the power of a signal without appreciably distorting its waveform."

====

Forgot to mention, the digital amp I am suggesting will work with PCM signals, more of a DSP, in the same way your media player can turn up-down PCM.
If I turn up PCM from my media player before its sent out, the volume and amount of watts consumed go up, same as an amp.

Also, it's possible to essentially double the volume without any amp, if you cloned 2 channels into 4 and then merged them back together.

2 x 8 dB into 4 x 8 dB, back into 2 x 16 dB, no amp at all.
 
Last edited:
Joined
Dec 8, 2020
Messages
2,777 (2.23/day)
Another way to describe my 'Smart Speaker' design:

PCM @ 20 dB >> Digital amp [PCM DSP] >> 20 db (no user change) >> PowerDAC >> 40w Analogue signal (analogue is power) >> Speaker driver (about 2cm away from the P-DAC, internal circuit)
PCM @ 20 dB >> Digital amp [PCM DSP] >> 40 db (user change) >> PowerDAC >> 80w Analogue signal (analogue is power) >> Speaker driver (about 2cm away from the P-DAC, internal circuit)

Pure signal into the speaker, the watts is an example.

----

Now lets say the speaker can only do 50 watts.

PCM @ 20 dB >> Digital amp [PCM DSP] >> 40 db (user change) >> PowerDAC [Reduce digital signal to match max power output] >> 50w Analogue signal.

----

Note, that the cost and requirements to produce good PowerDAC's in this design is reduced, as its mono (per speaker).

Smart Speakers.png

The ultimate edition would be, 0 Ohms.

====

E1DA PowerDAC V2.1 – Linsoul Audio (Headphones)
 
Last edited:
Joined
Dec 8, 2020
Messages
2,777 (2.23/day)
Last note, it would be a good idea that each speaker has a light indicator, which lights up (or changes colour) when then unit reaches maximum stated power.
Going above this, wont strictly cause an issue, however the input signal will be ducked to match to output stats (no volume increase).

Green: Maximum power output is now active, Amber: In ducking/compensate mode.
 
Last edited:
Joined
Jan 28, 2021
Messages
845 (0.71/day)
You seem to be getting hung up on this lossless digital domain thing that doesn't really exist and that this proprietary technology not being an amp when in fact it still is.

There is no such thing as lossless digital once you start reconstructing the audio in the digital to analog process, the very beginning of the process of the DAC is lossy not just the analog output. You can preserve the converted binary data forever as long as the bits are intact but thats not audio, you need a reconstruction filter and while thats digital its just a mathmatical approximation and best effort to create the original audio signal. Thats why DACs have different topologies and some DACs let you alter their filter, there is no perfect / lossless way to do this.

The PowerDAC or any similar approach is still an amplifier working with an analog signal. It doesn't matter how closely you couple the digital to analog conversion and preamplifer to whatever amplifier topology and circuit you are using, its still all the same thing. Their approach is unique and the name makes sense but ultimately its all the same processes just packaged differently.

If in anyway analogue was better, then why are we using DSP's to process audio, and not an analogue equivalent (ASP)?
You mean like an EQ?

DSP audio processing makes sense in form of room correction like Dirac because you are leveraging the advantages of the digital aspects of digital audio. DSP in this case is just a active EQ that is changing the amplitude of certain frequencies to compensate for room modes and reflections. The binary representation of audio you are working with is known and you only changing that binary representation based the software you are running. How the digital audio is reconstructed into analog sound is not changed just the tonality.
The ultimate edition would be, 0 Ohms.
If its 0 ohms its not making sound. The ohm rating is just the numerical representation of (work) of load the transducer represents to whatever its providing it with the power it needs to do work (make sound).
 
Joined
Dec 26, 2019
Messages
342 (0.21/day)
I think ole' Ferry is on to something.
And why would you care about Average Sleepy Joe Consumer when there's Early Adopters in a High-End Market (where ASJC is not invited) - with trickle-down effects *eventually*?
I'd say that the idea of pure digital is mesmerizing. There is an intentional distorting algorithm on EVERYTHING digital. Our audio formats or on kernel-level, I dare not say - but it's there. Making us stupid(er).

And those that say "over-processing" hurts the audio.. then you don't remember what audio should sound like. I have broken every rule in the book (including the "standard model"). There's not much you can tell me that isn't possible. You guys always re-iterate and recycle the same old "knowledge" without testing shit yourselves. How can you ever make something new? What if "the Master of Masterers" would be intentionally simplificating his own theorem so that you can never make as much money or cred as he is? Then you pass on this pseudo-max-knowledge to the next generation, and a perfect circlejerk of a lie is created and adamanted. Beware of that this is how it is.
// My 2 psychosises & the Copyright Adaptive Samplerate-Mafia
 
Joined
Dec 8, 2020
Messages
2,777 (2.23/day)
I don't think you follow at all, you keep reverting back to how it needs to be analogue, if all analogue was better than minimal why do PowerDAC do a better job without an amp as such?

You should read more about the efficiency of Class D and PWM.

Analogue is so 1940's, and also resistance effects the ability to pass power at a set rating, if you don't believe me make 100 Ohm speakers or coil up some cable.

If I remember correctly car stereo setups have low resistance, due to the fact they run from a battery with limited charge, power loss is bad.


Analogue is a power signal in a circuit, power + resistance = ?

The Complete Guide To Speaker Impedance (2Ω, 4Ω, 8Ω & More)
Speaker_impedance.pdf (kevinchant.com)

You also can't measure optical in ohms, not conductive.


Imagine a digital PCM amp increasing a signal to say 100 db as an example, and it costs £10, and uses 5v (1-2w for the amp), without loss.
Now the same but an A/B amp increasing a signal to say 100 db, but costs £500, and uses 150w, lossy.


TDI Smart speakers (Turbo digital injected), lol.

====

Given a PCM amp is essentially a DSP and not really an amp, there can be decoders (DTS, Dolby, other), and enhancers (DTS, Dolby, other).
You only worry about the speaker itself with the Smart Speaker design, add more or buy different ones.

The PCM amp should also carry one internal PowerDAC for headphones.

----

Current consumer level SPDIF is 15 channels @ 192k-24b max, on HDA that would be 15 x 96k-24b.
The PCM amp is also a digital optical channel router, since each speaker is mono.


1664287359729.png
 
Last edited:
Status
Not open for further replies.
Top