Showing 1 of 342 conversations about:
it certainly has the same pcb layout, no doubt about that.
@CEE_TEE you should clarify this. This doesn’t seem right.
Not sure if it’s truly design copy, Soekris DACs also look like this one to a certain degree, the design seems to be pretty standard among R-2R implementions, if you have a valid patent on your design then nothing to worry about or almost.
I'd guess it's the question of whether it's a design-resemblance or if it's actually the same design. I'm not an expert so I'm surely interested in how this plays out.
There is only so much you can do with R2R and layout optimization can end up looking almost the same
You are right. I have done so much research and experiments on that, and figure out that it is good enough for audio playback. The most important is the purity of input signals(i.e. PLL, triggering circuits)
Yes it changed some parts' package code into 0603 and forgot to remove refdes, that why it doesnt look harmonic.
okay now, this needs to be clarified from Massdrop/Airist Audio's side.
And people talking about "valid patents" on diy designs are driving me nuts.
Yes, working on a response...thank you!
Wow that would be bad on the part of massdrop.
Unfortunately, if you don't have a patent on your design, a shout-out by the person who capitalizes on it is the best you can hope for. That's why patents exist.
I was not aware of your project until you posted here. We have been working on the R-2R DAC for a few years, and we demonstrated one when we had our first show, a New York area head-fi meetup in 2015. I have spoken to my team and they assure me that they did not copy your design.
Our DAC is a sign-magnitude R-2R ladder, and both of these are well established and standard concepts. While other designs have implemented interesting methods such as auxiliary ladders and other compensation techniques, we stuck with the basic R-2R structure. This leads to the four resistor ladders followed by current amplifiers for each ladder and differential amplifiers to merge the positive and negative halves of each channel and apply some filtering to further mitigate the high frequency switching artifacts. None of these are novel or proprietary art. As several comments have pointed out, there are only so many ways to optimize the layout of a resistor ladder, hence the similarity between many different products. I've attached an image to illustrate those similarities:
From left to right: RDAC, Soekris, Hibiki, Denafrips, Rockna
Though the chips may appear similar, programmable logic devices are simply vessels and it is the code they contain that defines their operation. The control logic for the ladder is code I wrote myself based on the concept of sign-magnitude conversion. Once flashed into the chip, the code cannot be retrieved.
By the nature of R-2R ladders, the performance of the DAC is significantly affected by the stability of the reference voltage. The two circuit blocks next to our logic device are linear series voltage regulators, using an LT6656 precision voltage reference.
For anyone interested, here is a quick read on this type of regulator with diagrams:
https://www.eetimes.com/document.asp?doc_id=1272466And one with a little more detail and background:
http://www.onmyphd.com/?p=voltage.regulators.linear.seriesAlthough it is the most visible, the top board is not the only part of the RDAC. DACs using the same delta-sigma chip are not all the same, nor are DACs using the same basic R-2R topology the same. From the side angle photos of the RDAC you can see some of the circuitry underneath for handling the power, digital input selection and decoding, as well as our digital signal processing and clocking mechanisms, etc. These circuits define the RDAC as much if not more than the top board, both in terms of sound signature and design time and effort.
Someone copied it. Yes, the others you posted are very similar in layout, but there's a difference between similar and a direct clone.
This has all of the parts in the same layout, including the spacing of the 0603 resistors matching the original layout, just with the smaller footprint. The only difference I see is the LED and current limiting resistor added.
From looking at it, there's enough superficial differences that it looks like it may have been redrawn, as opposed to directly pulled from the board files, but there's no way the layouts are that identical by coincidence.
I'm not an IP lawyer, but AFAIK, this wouldn't be patentable, as it's not a new invention. The topology is well known.
I would guess that "copyright" would be the more relevant thing in this case, if anything covers it.
I'm not sure if there's any inherent copyright on a PCB layout (There are for some things).
The thing is, that whether it's technically legal or not, directly copying someone else's design and claiming that you spent years working on it isn't cool, and it certainly appears that's what's happened with this board at least.
It's certainly possible that other parts of the design are original work, but even if that's the case, it's only right to give credit for the parts that were copied (even if no legal license or payment is required).
As much as I agree with everything you said, I would not go as far as saying you did not believe in their working for couple of years by using the word claiming, to be honest with the degree of complexity of this design, two years wouldn’t be that far fetched. To me, it’s commandable that by their design, they allow us average joes to be able to own a r-2r design, that says something about their cooperation and desire to offer a good piece of kit to use by us commons mortals. I don’t like a cheater as more as you do but their actions indicate that they want to ease our financial pain and enjoy a good resistor ladder DAC.
yes. My point is that it appears someone else did at least some portion of the work, without getting any mention.
Copyright reserved. We can send a letter to your office if you want.
You are right, we have done numbers version of that board, the latest version is with LEDs. We did sell some amount of board on Taobao, and we can upload some of our customer and see whether there is airist audio's crew. There should be 5 candidate at most, should be the quickest way to show what happens.
Regardless of the situation, I think they're trying to make money first and foremost. But yes, I'm also sure they'd like to have a great product that sells well and to the largest market possible.
I'm not sure those photos help the case here sorry, it does show that out of five way to design this type of Topology, 3 are original and 2 are the same. I really believe that it's totally OK to see what other has done while trying to find the best way to implement it, but in this case really the guy should at least have received a phone call, even if the final design of the whole DAC brings something new, if the code if different, if everything around the R2R Layout is optimized by you, maybe in the end this doesn't deserve money, but at the very least it deserve credits. It is just not possible to have a design THAT identical without studying what this guy has done. Your analogy about many designer using the same delta sigma chip not making the same DAC, that is very very true and your whole technical justification to put forward that it's in the end still your design is somehow justify as well, but it's not justify denying the Hibiki module has been used. Back to using a chip. Every engineer that incorporate a DAC chip in their design actually have to pay for that chip, and part of that is R&D, you don't just buy the Silicon, you also buy a design
I certainly would act on it then.
You should not share your customer's information. That information is private.
Yes I would have your lawyer send a letter.
Has nobody from massdrop responded to this? (Would be indicative of the times and not at all in a good way.)
If not that is truly shameful,, as are the poleoole who knowingly signed up anyway.
I suggest you hit them hard enough to wake them up if nobody has addressed this by now.
Good luck. My massdrop participation tiny though it is hangs in the balance of this issue.
Do your customers also have the gerber files? Kinda weird for them to have the exact pcb layout and claiming they are as a huge coincidence. Lmao human with enough braincells wont fall for such bullcrap. If I were you, I would suspect whoever I sent the gerber files to, most likely the pcb printing guys.
What's the position of Massdrop on all this?
Copyright, really? I don't see the (C) indicating such on the photos you've posted. Putting things into the public domain tends to undermine any intellectual property protection that might otherwise be available. But copyright wouldn't be the applicable form of protection for a device anyway.
102. Subject matter of copyright: In general28
(a) Copyright protection subsists, in accordance with this title, in original works of authorship fixed in any tangible medium of expression, now known or later developed, from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device. Works of authorship include the following categories:
(1) literary works;
(2) musical works, including any accompanying words;
(3) dramatic works, including any accompanying music;
(4) pantomimes and choreographic works;
(5) pictorial, graphic, and sculptural works;
(6) motion pictures and other audiovisual works;
(7) sound recordings; and
(8) architectural works.
(b) In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.
I agree that there is probably not much legal action here, the tech is not patentable and copyrights probably don't apply as you say, but plagiarism is certainly not ethical, ethics are more in cause here than laws
In shenzhen, you can copy a exitsted board one by one, detail by detail to a gerber file spending small amount of money even a FPGA/CPLD design. Quite a shameful truth here.
You are right, I should talk to some of them and find the one buy and copy my module.
Did you have your copyright markings, with your name and year of creation on your PCBs then?
On the ethical question, William Tse of Airist says they demonstrated their design in 2015, which sosolar starts this thread by asking whether that is a copy of his 2016 work.
Nope, if you actually paid attention to the whole thing, sosolar had started the project in late-2015. http://bbs.hifidiy.net/thread-1149252-1-1.html
There is the prove and William Tse still hasn't had anything to back his claim up.
They demonstrated quote "one" design, that just say they where working on a R2R Dac, they don't say they demonstrated "this" design. Again my understanding, correct me if I'm wrong, and it obviously depends on country, is that having a copyright mean that you have an easy to prove case in case of plagiarism, but not having a copyright doesn't mean one is not the author, just that it's harder to prove. I don't believe that by law plagiarism is permitted regardless if the author has a copyright or not, it just deny the authors of his defence but it doesn't mean you're allowed to copy!
December 15 is very late 2015. Do you really suppose the meetup that Mr. Tse mentioned was even later in 2015 than that?
"We have been working on the R-2R DAC for a few years, and we demonstrated one when we had our first show, a New York area head-fi meetup in 2015."
Legal action or not, you have to admit that selling this without thoroughly scrutinizing both designs to illustrate differences would be a 0.5Malfoy level dick move.
It could be that they had a prototype that was completely different than this one we see right here. Unless Tse proves that he used the same design with the same layout like we see right now, I would take his word with a pinch of salt.
Pierre, I took that to mean he demonstrated one example of *this* design or one very similar to it.
And you just take William Tse's word for it? How about sosolar start selling the Hibiki in 2016? When do you think he started the design work on it? Jan 2016? It might be case closed for the simple minded.
So you're saying that Hibiki copied Airist?
I'm not so sure on the copyrights not applying, but regardless, I see it more as an ethical problem. Clearly someone cribbed the design from another source. If sosolar did the design 100% from scratch, then WilliamTse or someone on his team must have copied it, whether he knows about it or not. those boards are just too similar.
They're not from the gerbers, as some of the traces are positioned differently, and some components are different packages/footprints, etc, but the layout appears to be too similar to happen purely by virtue of optimizing the same topology. I'll tell you right now, if I had drawn this board and was using 0603 resistors instead of the 0805 on sosolar's version, I would have ended up with a much smaller board, and almost inevitably, many other parts would be oriented or positioned differently.
My guess is that someone either had the PCB sources, and made the changes and tweaks as they wanted, or more likely, they just sat down with a PCB and redrew it, with all the positioning the same.
I can certainly see why people believe something "must" have been copied, but I am not so sure.
If there is an optimal design, then skilled practitioners would "discover" it independently. As YuukiHaruto said, "There is only so much you can do with R2R and layout optimization can end up looking almost the same ."
It's also entirely possible that they're sourcing that module from a 3rd party who is responsible for the cloning, (or who bought it from yet another supplier....)
No. There is no one optimal design. They could end up very similar. I wouldn't be surprised if the ladders ended out with the same layout, etc, but they won't end up with every component in the same position and orientation, even if they were working from the same schematic.
There's going to be cases where something just doesn't matter and a part could be equally well positioned vertically or horizontally, and by chance, some of those "don't matter" parts will end up different.
Well, same form factor, same connections, same component distribution, at least they changed the color.
I am not invested enough in this to pick one specific part on any of these boards, and argue for or against a 90-degree turn or the optimal orientation of any particular part, but I will say that parts that "could be equally well positioned vertically or horizontally" would be few and far between just due to the intended purpose of the finished product, and much more so considering that manufacturers will want everything to fit into a rectangle of minimal area, to minimize their expense.
I also see nothing on the Hibiki board corresponding to the Airist board's part R225, R226, D51, and D52. Like you said, there is no one optimal design, but as I see it, this is also not necessarily anything more than two nearly optimal designs that just ended up very similar.
My guess, and just a guess, they are building a R2R Dac, using well documented textbook theory and you start comparing, measuring against what's available, your design don't perform as well in some area, you look deeper and you see that a fast and easy way is to do the same thing. Even if it's the "optimal" design, Is it a reference design? Is this set of components and circuit open source and available to the public, from what I read it doesn't look like it. it's more than the same optimal design, it's the same layout, let's say the ladder is built like such and it just make sense, there still about fifty surronding components that are not only the same, they are positioned at the same place! The probability of this happening is just IMPOSSIBLE, there is absolutely no way it is just coincidence there is just too many parameters!
The almost part of that statement is the problem. These are not almost the same, they are the same. I would buy the excuse if not for the board to board connectors by the controller. The similarities there go well beyond similar optimizations to similar devices. There is no chance that two completely separate design optimizations would settle on exactly the same configuration there. There's too much left to choice or preference on connectors like that.
It's a different version of the board. This one has those components.
the board you are looking is early version without LED, that is not for sell. The version I selling is WITH LED and resistor you mentioned. Please follow links posted by @TimConor bofore, and say it again.
Based on the board here http://bbs.hifidiy.net/thread-1192897-1-1.html, it seems that the module is exactly the same as the airist ones, with those LEDs and resistor intact. This is as obvious as it gets. Airist guy don't even bother to change a single component on the board lmao.
Thank you for bringing my attention to this Airist drop. I hope that whoever did the work gets fairly compensated, and I am interested to see who that will turn out to be.
My guess is that MD will cancel this drop just to avoid even the slightest chance of getting sue for importing copy-right infringed goods. Unless they bet on sosolar not having the resource to sue.
Or think about it from a business point of view. The moment sosolar litigates the entire shipment will be held at custom until the case cleared in court right? Now MD can't deliver which will lead to unhappy customers. Right?
OK, thinking about it from a business point of view, if sosolar has a valid claim, is this the most profitable way to present it? Has anybody shopping on Massdrop been convinced to go buy a Hibiki? Or just to not buy an Airist?
"And you just take William Tse's word for it?"
Well you can say that the other way too.
I believe this is the most likely case
I'm convinced not to buy it. And I'm doubting Massdrop ethics. This is not good for the audio community, and Airist negating something that is evident without action from Massdrop is sad.
I used to do PCB design and although it does seem like somebody used the other design as a template, I agree that it's such a small part of the overall design and a pretty trivial layout. I don't know enough about patents to know whether it's too simple to patent for sure or not, but I suspect it's not protected. I don't know how many people reading this are going to understand just how simple that layout is, though. To someone who designs PCBs, it's trivial. Like.. throw it together in an hour trivial.
Well, sosolar has explained himself and the tech behind the R2R module at length over at head-fi thread, whereas Tse is doing feck all on explaining why the exact pcb layout is used and go about explaining irrelevant things that we can actually find on google by ourselves.
What I expected is either MD cancel this drop, or compensate some credit for me. For Airist Audio, I will still talk to our lawyer for further move. And for customers, I hope you would enjoy my balanced version which is using exceptional better clocking circuits and better PSU at Hongkong Exhibition 2018.
Can I hire you?
You certainly work much much faster than me, I'd like a guy that could throw this kind of thing in an hour any days, that would be gold.
Circuit board design and circuit schematics are not covered under copyright law. The courts ruled on this decisively decades ago in the Mackie vs. Behringer lawsuit, enabling Behringer to openly clone and sell other manufacturer's designs under their own brand/trademark without recourse.
Patent infringement is another matter, but R2R ladder DACs have been around for decades, and are not patentable. Something must be new technology and involve and "inventive step" to be patentable, amongst other things. And, any patent that might have existed on R2R DAC topology would've expired long ago.
Quit being armchair attorneys and enjoy the product. Nothing improper has been done here.
Does anyone here who is complaining use an Android smartphone? If so, you're being hypocritical.
To sosolar, differentiate your product via marketing, make improvements to your design to stay one step ahead of Massdrop/Airist, or come up with something new that qualifies and get a patent. This is how competition benefits us all.
Thank you for you kind remind. We still need to talk to them about this. Anyway, we have developed some other fascinating DACs to release soon.
Massdrop might be in a bit of panic now, but you really can't blame them for what happened till now. The Airist Audio team is the one that copied your design and Massdrop is just hearing about this about 24 hours back. They should be getting their facts straight and preparing a response or decision. Let us leave the judgement till we see their response.
As for me, I hope they don't cancel this drop, but, rather reach some sorts of an agreement with you. As all the big-shots at HF and SBAF seems to approve this DAC, it is also an attestation of your work. More than anything, it makes a finished R2R DAC available for us at an affordable (?) price point. But I do believe financial compensation is in order since you did part of the work in a product they are trying to market, what ever the laws of the land says about intellectual property.
While the legal aspects of it are being discussed out of curiosity, the core impropriety is that in the niche world of speciality audio products and DIY, people want credit given where credit is due. Even if they wanted to roll the product out and start selling it without mention of the Hibiki board, all they needed to do was acknowledge whatever needs to be acknowledged regarding the similarities. This isn't a billion dollar industry. Admitting that you based your little DAC on a pre-existing design isn't going to result in lawsuits and huge profit loss. If they had a semi-competent person speaking for them, it wouldn't even be a lasting issue to explain that they'd used the board (if this is true) and they won't be sharing profit with sosolar.
We're talking about a small market of moderately priced luxury goods being sold through a limited platform. The stakes aren't legal. The stakes are whether people like you.
Out of curiosity, do you make completed DACs or offer OEM sales? Has any company made a commercially available DAC using your boards?
As we are small team working on our spare time, OEM is very welcome, seriously. We also make our own DACs, ready to release this year.
Sosolar, is the price of this dac favorable or do you feel for the money you could improve on the value of this product further.
Another question, how does this particular DAC perform with the power delivery chosen by Airist?
Legalities aside, the smell from this product is enough to dismiss it IMO. There are other entry-level R2R options from respectable shops like Soekris and Metrum for not that much more $.
You got another point, 6656 using here is not suitable for ladder because it produce 50uvrms noise, while my design(6654) can go deep to 1uvpp that make the sound purer, darker. This should be a costdown point for Airist.
This is what I was afraid of. How much more costlier would it have been to implement a better power supply/delivery? Given that there is no room for upgrading this particular unit, then perhaps with a LPS?
So, he cut some more corners to save some bucks? They use cheap connectors and micro USB ports (micro USB in a desktop device LUL), and that still not enough? No wonder why it measured like shit.
That's beyond visually similar. I'm not buying it, figuratively or literally.
nope, 6654/6656 share the same footprint. It is simply a costdown(probably from 3 USD to 2USD), but 350 times noisier.
Thanks for your time on this sosolar, you've definitely confirmed my suspicions!
Damn I hope they read this comment.
I wonder if many take your same approach the cost will be reduced... just curious.
And here I was - wondering why there was so much empty space behind in the chassis (in the official pics) for a collaboration piece...
Hope things get sorted out asap!
I agree it looks very close. So what.
There are only so many ways to lay out pixels on an LCD screen, and there are only so many ways to lay out a pair of resistor ladders on a board, along with the buffers and other components required.
There is going to be a particular optimal routing of the traces to avoid noise, induction, oscillation, etc. There is only so much space that 48 SMD resistors are going to require on a PCB, which will dictate size.
For all we know, all of these boards were independently designed with extremely similar characteristics based upon widely accepted best design practices for this type of circuit using what would prove to be consensus components given the cost target.
And, if the board was essentially copied, so what. There is no intellectual property there. The circuit topology is nothing more than a commodity, and I could do that board layout from scratch in a few hours tops in a CAD application, probably less. So, nobody is being deprived of the fruits of any significant labor.
If anything, the fact that a couple of well-known and well-regarded companies implemented a clone of his design (if that is indeed a fact) could actually lend credence to his heretofore relatively unknown design and name and boost the sales of his competing product.
To the last two points...
I am not a qualified engineer. But it is not my understanding that the board is the only element of @sosolar's DAC that has been copied. The misappropriation being alleged seems far more extensive than that. This includes a prototyping and revision process that, presumably, requires far more effort and expenditure than replicating a widely understood board layout.
To date, no official attribution has been given to @sosolar for his design of the DAC in question. The extent that he is receiving any recognition is due entirely to his own efforts to call attention to the impropriety he is alleging. On the contrary, Massdrop has opted to stand behind @WilliamTse's (rather tortuous) claim that the design of the DAC is exclusive to Airist Audio.
Reasonable people can make up their own minds as to the merits of each disputant's case. I have little doubt further information will emerge one way or the other.
There is nothing immoral or illegal about researching other non-patented circuit designs and/or board layouts and using them as a basis for your design. The courts have ruled definitively about this, and this is how it is done. I suggest researching Mackie v. Behringer, or even the reverse-engineering of IBM's proprietary BIOS chip to enable PC clones, and the legality thereof. Whether you are making cars, televisions, smartphones, or litter boxes, you look at the existing technology, you determine which implementations have worked best, and you implement some clone or combination thereof, perhaps with some of your own improvements. Certain things are protected intellectual property, and certain things are not. The item in question here is not. Period.
Do you people get bent when the vast majority of audio amplifiers designed and sold over the past several decades don't credit James Bongiorno for the dual-differential, fully complementary topology?
I agree, I hope the original designer of the board is fairly compensated. And fair compensation for a non-copyrightable, non-patentable, commodity of a design completely devoid of intellectual property is exactly $0.
I'd encourage you not to take (further) offense, but you're just repeating yourself...
Reiterating basic legal theory and practice, albeit correctly, doesn't really address the salient issues. I do not expect Massdrop would have proceeded with the marketing and sale of this product if they were not receiving sound legal advice as to their potential liability in an intellectual property and/or contract dispute.
You can rest easy knowing that you've made your own position on the legalities of the matter quite clear.
I really don't have much to say about the financial compensation part as the views of the people and laws of the land differs where ever you look. But giving credit where it is deserved is common courtesy. It happens all the time even in open source software development. It is a standard people try to uphold. @WilliamTse mentions that they approached an external consultant for designing/engineering the circuit board. It could even be this person who just copied a working design off a DIY board (I am not sure about the feasibility of this, as my knowledge in this field is lacking). But, it is a bit of unfair practice considering @sosolar is planning on releasing their own fully-built devices this year.
Well, unfair is not illegal, so I will stop here.
"There is nothing immoral or illegal about researching other non-patented circuit designs and/or board layouts and using them as a basis for your design."
Yet said defendant already said it was pure coincidence they are the same, there was no research of @sosolar board/designs.
It was literally in Tse's first response lmao. "I was not aware of your project until you posted here." What he posted 10 hrs after that was just damage control and passing responsibilities onto other party, basically saying "don't blame us, blame the people we hired to clone the board, while we sit over here claiming the original work as ours and reaping the rewards."
Thanks for posting, will be curious to see what the answers are to this.
This is exactly the problem. Stating outright that it's an independent design that just "looks similar" when it clearly is a copy. There are too many variables that some things would end up different, no matter how well understood the basic circuit is.
If the design has no value, and (according to several users on this thread) could be easily thrown together in an hour or two, why copy it?
This isn't a case of similar general layout of a well known circuit topology. It's a straight up clone.
Even if it's legal, it's a scummy thing to do. Especially the part where he claims it's their own design.
If he had said "Here's a dac we implemented, based on a board design we found wherever." "we improved x,y,z, and added whatever features on the base board", etc. That would at least be the truth, and whether or not the original creator is compensated, buyers at least know who did what, and have a more realistic idea of the design practices.
I doubt that anyone who would simply copy a significant board in the design has put in a whole lot of engineering time measuring and/or revising things for the best results.
The RDAC is unrelated to your project, but if you want to discuss it further, you can email me: firstname.lastname@example.org
People with enough braincells can tell that Hibiki DAC is 100% related to the RDAC smh my head
Wow, just wow. You've not talked to sosolar yet but you already made such a statement "The RDAC is unrelated to your project"? How can you be so sure? Just because Tse said so? Or did you see irrefutable prove? I'm just curious.
As I explained, our engineer researched existing implementations and came across the Hibiki on the Chinese audio forums. He drew on this and other boards, including the Soekris and MSB implementations, as the basis for the layout. The design of the 4-layer top PCB of the RDAC goes deeper than the component layout and it is not the same design as the Hibiki.
You can read more about the development process here: