So you're someone who likes to buy at the component (amp, pre) level rather than module level and DIY it? Do you care if the amp or pre you buy is made up from modules inside though? There's a business opportunity for people to package my (hypothetical at this stage) modules into nice looking 'gear'.
When your amp is made up from modules inside, then upgrades aren't so difficult - just stick in an extra amp module and swap out the PSU for a slightly bigger one. Two extra channels for your amp and no wasteSo in part, the modules idea is hatched from the audiophile need toward continuous improvement.
Don't see a retail market for modular proprietary stuff. The advantage of component level is one from here, one from there. The paycheck for high end is more the bragging rites from doctors and lawyers who point to their brand name stuff. They don't replace modules, solder or adjust. A large portion of music lovers are not technical either. That leaves the DIY market.
I am not sure there is an OEM market either as the advantages of modular go away with high volume. We already have "modular". Chip amp, preamp chip, etc. A microcontroller and a few chips is all one needs in consumer grade. High end is half power supply and mechanical construction. Not friendly to snap in modules.
I'm not clear what you mean by 'proprietary' stuff here - modules can be made by anyone with a (gratis) license. Think 'ARM' style here - they do the design but rely on partners to build and sell.
True enough -they're not my intended market, at least not at the start. They won't buy because its not perceived by them to be a 'reputable brand' with a substantial marketing budget.
But think of the people who they buy from. The modular approach is designed to appeal to them too - ease of servicing. The non-technical guys and gals buy at the component or system level, from people who integrate modules.
And that's the market I'm planning to target first.
Yep, this is above the current 'consumer grade'.
This idea is not a fit with the current 'high end' market which is to a large extent (not entirely though) based on over-engineering and BS
Not quite. With a modular appraoch, your upgrade path is far clearer, be it by adding parallel amps, or changing the present for new and better ones.
Servicing is also much easier - if anyone still remembers what that dastardly deed is.
Choice is also much wider, but is especially favorable to anyone with praticular requirements, like having say 7 line inputs, 3 of which need to be balanced. Great for changing your mind later on, and converting one XLR input for a say phono RIAA stage.
But your essential evaluation, or doubt, of there being enough marketing drive to make such a system come alive is, I'm afraid, dismally correct. I also think the market will not take such an approach favorably, and will relegate it to its fad pile before even giving it a chance to prove itself.
Practically, it is scuttled by the fact that modularity AND high quality cost some serious money for some VERY serious connectors, both male and female. If you go the soldering way, with extruding pins, you disable most users to make the change themselves, but get a really good contact. If you go for the PC type card slot-in connection, you get ease of exchange but a not so good contact.
That means you need to go for some SERIOUS connectors, which provide both excellent electrical contact AND a great mechanical contact; these are available, of course, but they cost a pretty penny. And that is definitely not an advanatge in today's market of our cheapskate civilization, itself not one bit interested in having anything for longer than 6 months, in view of how many new gadgets will appear in the meanwhile.
Face it, people, quality products are on their way out, at least in the consumer market.
Yep
Yes, - the existing market most certainly won't. Been there, tried that over a decade ago. So this time I plan to do it differently - create the market for the new offering at the same time. Co-evolution if you will.
I concur - I have some ideas for connectors but haven't made a firm decision yet.
My initial thoughts are that the SATA cables used for disk drives might be a good starting point. Cat5 (RJ45) is also a possibility.
For external connectors there are some Chinese clones of Lemo which I'm starting to employ in my designs. Time will tell if they live up to the reliability and quality levels of the original.
Hi,
Have a look.
Yup. It is that.
Find the cover of the CD you want, press play.
Finding the album you want on the HDD can be very quick, much quicker than finding an actual CD in a well organised collection.
No inserting disk etc.
I must admit that my ripped collection started at my them office, were I ripped my CD's so I could listen to them in the office. This as around 10 Years ago...
What have you seen?
Yes, basically. My own is a custom build server, and it runs Media Portal (Open Source - an XBMC fork for Windows - alternatively XBMC is on Linux), however this is absolutely a GeeXboX in terms of installation and setup.
Usability is maybe a touch up on J.River, but I do recommend the inexpensive MSI and/or Asus All In One PC's with J.River for most people, especially music only use.
Get an MCE remote. It is usually still easier and quicker to use that than the touchscreen. My Media PC's have not had mouse and keyboard attached for many years, they are basically "set top boxes".
In order to help our dealers and distributors AMR is offering at cost fully build servers to dealers and distributors.
They are not available from AMR or Dealers to the public (but a few have apparently been sold anyway), but dealers are welcome to clone them, you can see one here:
AMR DP-777 Internal Photos (Uncensored) | Computer Audiophile
Wakeup from "Sleep" is pretty much instant, no slower than many CD-Players, it is important to have a "plain" windows on a dedicated machine. If the OS first has to load ten ton's of bloatware etc. that connect to the internet before they let you use the PC etc., of course it takes ages to boot or come back from sleep. Plain windows only plus Media Player software on an SSD boots up very quick and even resumes from hibernation in a jiffy.
Ciao T
Have a look.
Yup. It is that.
Find the cover of the CD you want, press play.
Finding the album you want on the HDD can be very quick, much quicker than finding an actual CD in a well organised collection.
No inserting disk etc.
I must admit that my ripped collection started at my them office, were I ripped my CD's so I could listen to them in the office. This as around 10 Years ago...
What have you seen?
Yes, basically. My own is a custom build server, and it runs Media Portal (Open Source - an XBMC fork for Windows - alternatively XBMC is on Linux), however this is absolutely a GeeXboX in terms of installation and setup.
Usability is maybe a touch up on J.River, but I do recommend the inexpensive MSI and/or Asus All In One PC's with J.River for most people, especially music only use.
Get an MCE remote. It is usually still easier and quicker to use that than the touchscreen. My Media PC's have not had mouse and keyboard attached for many years, they are basically "set top boxes".
In order to help our dealers and distributors AMR is offering at cost fully build servers to dealers and distributors.
They are not available from AMR or Dealers to the public (but a few have apparently been sold anyway), but dealers are welcome to clone them, you can see one here:
AMR DP-777 Internal Photos (Uncensored) | Computer Audiophile
Wakeup from "Sleep" is pretty much instant, no slower than many CD-Players, it is important to have a "plain" windows on a dedicated machine. If the OS first has to load ten ton's of bloatware etc. that connect to the internet before they let you use the PC etc., of course it takes ages to boot or come back from sleep. Plain windows only plus Media Player software on an SSD boots up very quick and even resumes from hibernation in a jiffy.
Ciao T
I read with interest Nelson Pass' comment in his turboF5 update to the effect that he finds nulling out the .1% 2nd "clearly" audible.
Hi,
Yet the evidence that is presented you reject.
I think of myself as someone who likes music first and prefer to spend the least amount of time and money on gear. Sadly from an early age I discovered that the gear in various got in the way, needed changing modifying etc. so I have been doing it ever since.
But I did not tell Bruno that his math was wrong. I told him that by an of hand-waving he simply excluded a key factor (and one which prior publications all the way back to Olson and Crowhurst include and stress) and so even though his math, calculations and results where correct, they where not applicable.
Yet I did not tell Dustin any such thing.
What I did say was that a low-bit DAC (or ADC) combined with noise shaping and given a specific PCM source (sample rate, wordlength) cannot given equal per sample accuracy as a "direct conversion" DAC/ADC, which incidentally may be a single bit DAC/ADC running at sufficient speed, a hybrid or a multibit DAC. In fact the claimed accuracy only "magically" materialises when many samples are averages.
In effect I objected to his claim that low bit + noiseshaping = multibit (or exceeds it). It is up to him to prove his claim.
No need to. Philips actually did 25 years ago and I am using that DAC Chip.
I do not see why the burden of proof should be mine, when the extra-ordinary claims that go against the most basic precepts of Information Theory are being made by others.
That is perfectly true. I am happy to go with the lowest distortion options available as a theoretical lower limit.
IF we scale -14dBfs to 94dB SPL and thus full scale 108dB SPL (@ listening position - so a lot more SPL at 1m) for full scale, what level of distortion would you suggest the best offer, at (say) 50Hz, 500Hz, 5KHz?
Why would anyone bother to improve an area of performance beyond what is required, especially if in doing so he introduces dis-improvements to other areas where the performance dis-improvement will be noted?
I could have given beyond 16 Bit resolution and I did. The noisefloor is no higher than quite a few pure solid state "High Resolution" options JA evaluated. It his view that he would like to see lower noise.
I would like to have lower noise, but the necessary means would impair sound quality in areas where I find it to matter.
So I leave making "perfect measuring" DAC's that have been described in comparison to my design as "sound just 'flat' and uninvolving by comparison" to others. If you prefer to buy a DAC that measures perfect, but sounds artificial and boring, I'm not stopping you, am I now?
Actually, a "20-Bit True Resolution" version (instead of the 18.5 or so Bit's we get) would not sound identical to the current design. It would sound worse.
Alas, I am not offering theories. I can only suggest that you have a listen.
Because in the end I design my stuff to be listened to.
In other words, I design it (in the case of the DP-777) for the Magazine writer who wrote: "For me, the AMR DP-777 was more than a pleasant distraction, more than just an escape from the humdrum; it provided some of the very best digital sound I’ve heard." and not for the one who observed: "I was hoping that the AMR in HD mode would offer a true high-resolution alternative to the NOS behavior, but the former seems compromised.".
Because the first writer deals what is actually important, while the second deals with what is directly observable with instrumentation but does not (or at least so far has not been shown to) correlate with what what the first writer needs or wants.
Ciao T
Yet the evidence that is presented you reject.
I think of myself as someone who likes music first and prefer to spend the least amount of time and money on gear. Sadly from an early age I discovered that the gear in various got in the way, needed changing modifying etc. so I have been doing it ever since.
But I did not tell Bruno that his math was wrong. I told him that by an of hand-waving he simply excluded a key factor (and one which prior publications all the way back to Olson and Crowhurst include and stress) and so even though his math, calculations and results where correct, they where not applicable.
Yet I did not tell Dustin any such thing.
What I did say was that a low-bit DAC (or ADC) combined with noise shaping and given a specific PCM source (sample rate, wordlength) cannot given equal per sample accuracy as a "direct conversion" DAC/ADC, which incidentally may be a single bit DAC/ADC running at sufficient speed, a hybrid or a multibit DAC. In fact the claimed accuracy only "magically" materialises when many samples are averages.
In effect I objected to his claim that low bit + noiseshaping = multibit (or exceeds it). It is up to him to prove his claim.
No need to. Philips actually did 25 years ago and I am using that DAC Chip.
I do not see why the burden of proof should be mine, when the extra-ordinary claims that go against the most basic precepts of Information Theory are being made by others.
That is perfectly true. I am happy to go with the lowest distortion options available as a theoretical lower limit.
IF we scale -14dBfs to 94dB SPL and thus full scale 108dB SPL (@ listening position - so a lot more SPL at 1m) for full scale, what level of distortion would you suggest the best offer, at (say) 50Hz, 500Hz, 5KHz?
Why would anyone bother to improve an area of performance beyond what is required, especially if in doing so he introduces dis-improvements to other areas where the performance dis-improvement will be noted?
I could have given beyond 16 Bit resolution and I did. The noisefloor is no higher than quite a few pure solid state "High Resolution" options JA evaluated. It his view that he would like to see lower noise.
I would like to have lower noise, but the necessary means would impair sound quality in areas where I find it to matter.
So I leave making "perfect measuring" DAC's that have been described in comparison to my design as "sound just 'flat' and uninvolving by comparison" to others. If you prefer to buy a DAC that measures perfect, but sounds artificial and boring, I'm not stopping you, am I now?
Actually, a "20-Bit True Resolution" version (instead of the 18.5 or so Bit's we get) would not sound identical to the current design. It would sound worse.
Alas, I am not offering theories. I can only suggest that you have a listen.
Because in the end I design my stuff to be listened to.
In other words, I design it (in the case of the DP-777) for the Magazine writer who wrote: "For me, the AMR DP-777 was more than a pleasant distraction, more than just an escape from the humdrum; it provided some of the very best digital sound I’ve heard." and not for the one who observed: "I was hoping that the AMR in HD mode would offer a true high-resolution alternative to the NOS behavior, but the former seems compromised.".
Because the first writer deals what is actually important, while the second deals with what is directly observable with instrumentation but does not (or at least so far has not been shown to) correlate with what what the first writer needs or wants.
Ciao T
Might I suggest, gentlemen, that flaws in the electronics may be masked (or hidden) by the transducer used. The most difficult energy transformations are still mechanical to electrical and vice versa I.E. microphones and speakers. Until perfect transducers are invented! many of the points raised are moot. Regards
Scott,
What I find interesting in that are a few things.
First numbers are thrown around without any reference level and context.
Without being specific about SPL's involved, frequencies etc. the number alone is meaningless.
Second, instead of a sensible fair use quote something totally different is said instead:
What must be realised is that the adjustment referred to is not a 2nd harmonic adder, but rather adjusts the overall distortion profile of the Amplifier, plus it even affects overall gain if adjusted enough.
So I would expect P3 adjustment to be audible, but not because 0.1% 2nd HD are reduced (or not), but because of other effects it has.
Ciao T
What I find interesting in that are a few things.
First numbers are thrown around without any reference level and context.
Without being specific about SPL's involved, frequencies etc. the number alone is meaningless.
Second, instead of a sensible fair use quote something totally different is said instead:
Firstwatt Website said:
What must be realised is that the adjustment referred to is not a 2nd harmonic adder, but rather adjusts the overall distortion profile of the Amplifier, plus it even affects overall gain if adjusted enough.
So I would expect P3 adjustment to be audible, but not because 0.1% 2nd HD are reduced (or not), but because of other effects it has.
Ciao T
Hi,
Have fun. At least used as Hard drive connectors they have an extremely poor reliability record. Please, someone bring SCSI and ATA back...
Ciao T
Have fun. At least used as Hard drive connectors they have an extremely poor reliability record. Please, someone bring SCSI and ATA back...
Ciao T
Not my experience. In fact, quite the opposite (which has been the source of many surprising discoveries). Certainly transducers have their flaws, but its the flaws of electronics which result in listener fatigue. I'm listening at present to some very inexpensive active speakers that I've worked on electronics for for quite some time. Improvements in the electronics are clearly noticeable behind whatever flaws they have (and being cheap, they're bound to be many and various).
Its a claim - do you have any support for it? Care to highlight which points in particular so we can debate them?
Hi,
You are too timid. The worst transducer of all is the human ear. So until a "perfect" human ear is invented many measured flaws in the electronics may be masked or hidden.
Ciao T
You are too timid. The worst transducer of all is the human ear. So until a "perfect" human ear is invented many measured flaws in the electronics may be masked or hidden.
Ciao T
Even simple measurements of harmonic and intermod distortion show much higher levels in speakers for example than in electronics. Soundstage in their speaker measurements section shows results for many speakers, from the NRC of Canada. You would be surprised at the results for some "well regarded" models. Thorsten may well have hit the nail on the head-the ear and its associated processing gear may be the most flawed device in the reproduction chain-also the most variable due to the shape of the pinnae. Regards
This reminds me of what someone said about the late Buckminster Fuller: to wit, that he was an a**hole, but was at least "our a**hole".
Yes the amount of IM produced in the human ear is astonishing. Try playing some of the Telemann on recorders that is written to specifically exploit this and makes its own bass line from the difference frequencies! (discussed iirc in Benade's cute little paperback, Horns, Strings, and Harmony).
But --- we adapt to this I think. So our judgments are not as impossible in some contexts as might be thought.
Brad
John,
This is for a different scale, but is this the correct theory on active power line filtering?
http://web.ing.puc.cl/~power/paperspdf/dixon/37a.pdf
This is for a different scale, but is this the correct theory on active power line filtering?
http://web.ing.puc.cl/~power/paperspdf/dixon/37a.pdf
Hi,
The following links give a view of this whole computer audio malarke from the position of a 60+ retired long standing audiophile and includes references to his wife's use. It may make nice reading for perspective (there are more blog posts on the subject and things not in strict sequence)...
http://diyhifisupply.com/node/707
Music Server Computer transport - Round 1! | Diy HiFi Supply
http://diyhifisupply.com/node/793
Fi For Flora: Asus EB1501 -- Honey I Shrunk the Bits, Part 2 | Diy HiFi Supply
Fi For Flora - Done! But I know What You're Thinking... | Diy HiFi Supply
Ciao T
The following links give a view of this whole computer audio malarke from the position of a 60+ retired long standing audiophile and includes references to his wife's use. It may make nice reading for perspective (there are more blog posts on the subject and things not in strict sequence)...
http://diyhifisupply.com/node/707
Music Server Computer transport - Round 1! | Diy HiFi Supply
http://diyhifisupply.com/node/793
Fi For Flora: Asus EB1501 -- Honey I Shrunk the Bits, Part 2 | Diy HiFi Supply
Fi For Flora - Done! But I know What You're Thinking... | Diy HiFi Supply
Ciao T
I've had some fails too - even straight out of the plastic - the cables themselves are just too cheap. Last one I bought was a 50cm, price 3rmb. ISTM Chinese manufacturers have a habit of reducing the Cu content of cables below the minimum needed to conduct....
ST506 too? No, let's not go there.
Agree. The module approach was always coming up every few years at Harman and elsewhere, whether in the form of a handbook of "proven" reusable circuits (so all risk could be removed from the design process and sacred schedules mantained, and the managers not lose their jobs) to the notion of reconfigurable products.
And...it can be done. The connection problem is huge, but the bigger problems with things configured from modules is --- the resulting system is never cheaper. There is simply too much redundancy, compared to a product designed to meet specific objectives.
And as for the reusable circuits, when we attempted to catalog them it became clear that the number was enormous. And then, what motivates your engineering staff anyway? To be told to do the same thing again and be happy you have a paycheck? Well in tough times this may keep people showing up in the morning and streaming out as the second hand crosses 12 at the departure hour, but it's not typically what we became engineers for.
- Status
- Not open for further replies.