Author Topic: Oscilloscope ASIC Kickstarter instead of Open scope.  (Read 19190 times)

0 Members and 1 Guest are viewing this topic.

Online nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #75 on: January 11, 2020, 11:32:40 pm »
That will just drive people away.
Off-topic rants by the OP also do that. If you want a thread to to stay on-topic then stay on-topic yourself. Lead by example.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17428
  • Country: us
  • DavidH
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #76 on: January 12, 2020, 02:03:04 am »
David Hess- Keysight, Tektronix, Rigol, all have their own ASICs. Did you even read the first post? You picked single pieces of information to base your argument on and didn´t look at the whole picture.

You not only have to compete with existing designs which have the advantage of economy of scale allowing a fully custom ASIC but existing programmable logic based designs which did *not* justify an ASIC.  Based on HP and Tektronix using the same fully custom ICs for such long periods of time that it held back features and performance, I think ASICs were marginal even for them except at the high end.

Is there something that you cannot accomplish economically with an FPGA now?  I have looked into this problem in detail before and with a small or perhaps no loss of performance, I am sure that an economical FPGA based design is feasible now if the hardware and software is partitioned correctly.  Most of the flaws in commercial designs seem to be on the software side.
 
The following users thanked this post: ogden

Offline excitedboxTopic starter

  • Regular Contributor
  • *
  • Posts: 104
  • Country: de
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #77 on: January 12, 2020, 06:35:46 am »
Xfab has open source ip on this site and also includes a handy little cost calculator that shows just how cheap it is for raw dies. Obviously you gotta package them too but it is pretty cool looking at all the microscope pictures of the transistor stacks.
https://www.xfab.com


You are right that software is a major expense because salaries stack up quickly. That is why the 2 prong approach with a open source scope framework. The only reason those Chinese Logic Analyzers are so cheap is because they stole the software fro that other project.

Right now a lot of projects are not viable because of the high cost for fast ADCs and DSPs etc. Each chip is not just associated with a manufacturing cost right now but also with a profit margin, marketing budget, and operating expenses for the manufacturer and that reduces the profit you can make on your product.

You do bring up some valid points. Rigols 7 in 1 scope would probably not have been possible without a custom asic so it allows them to add features that wouldn´t exist otherwise. But that is also a plus for competition. As an opensource project we don´t have to compete with anything since it is not about making money. Part of the "profit" is helping people access equipment they couldn´t otherwise. Right now a 2ch1Gsps ADC costs over $400 that is not a cost many consumer products can carry. If that chip comes down in price we might see a whole host of new products on the market.

An FPGA is just not an efficient way of doing things. Yes it is great when you need to customize a circuit but a chip is already 70% interconnects. When you add all that switching it becomes even more.even Intel eAsics which have a bunch of blocks and only the top layer is customize robs you of one out of 6 layers. With each layer you build the yield drops substantially too driving costs up more.


We need to focus on finding out what is on one of these custom Asics besides a DSP and ADC and what modules manufacturers are currently implementing on the FPGA. Only then can we see if it is worth it and what it would take to make it reality. Everything else is useless right now. None of these arguments are real without having real world facts.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29811
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #78 on: January 12, 2020, 07:52:27 am »
Right now a 2ch1Gsps ADC costs over $400 that is not a cost many consumer products can carry.
:-//
How can that be when something just like that is the basis for a SDS2000X-E series DSO retailing @ $620 ?
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 
The following users thanked this post: wraper

Offline EEVblog

  • Administrator
  • *****
  • Posts: 39026
  • Country: au
    • EEVblog
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #79 on: January 13, 2020, 08:13:04 am »
They are holding back innovation by keeping the test equipment out of the reach of students, researchers, small companies, and individuals. We need to put a stop to it.

Wow, are you serious?
Do you have any clue how much scope you can buy for under $400 these days?
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 39026
  • Country: au
    • EEVblog
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #80 on: January 13, 2020, 10:22:36 am »
There are also virtually no scopes that use a single SoC solution for the whole thing.  You get the ADC, the memory controller/triggering, the sample memory, and then the SoC running the display and stuff.  We've got good availability on SoCs for the application software, ADCs (to some degree), and memory.... so it's the triggering/channel timing/memory controller ASIC that's the real missing link in my mind.

This is why many scopes use a the Xilinx Zynq as it combines an ARM applications processor and an FPGA. More than powerful enough for decent performance entry level scopes.

 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 39026
  • Country: au
    • EEVblog
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #81 on: January 13, 2020, 10:28:28 am »
I think ASIC on Kickstarter is possible, but only after some meaningful transparent history of lesser successes.

And only if you show a working prototype first, like the whole working scope built from an FPGA and ADC's etc. Then you can say "We need $X million to make this cheaper with an ASIC. You can't just say "Give us money to make a scope ASIC and we'll work out the details later".
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 39026
  • Country: au
    • EEVblog
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #82 on: January 13, 2020, 10:33:19 am »
As technology gets better signals have gotten faster and there are more of them. People used to be excited about 20MHZ analog scopes but now that is considered a useless toy scope.

Yes, but now you can get hacked 200MHz 4 channel scopes for under $400.
Anything much higher bandwidth and you reach the usable limit of passive probing technology.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 39026
  • Country: au
    • EEVblog
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #83 on: January 13, 2020, 10:34:13 am »
I've unlocked this thread because I think it's hugely interesting, please play nice.
 
The following users thanked this post: thm_w, tautech, ogden

Offline EEVblog

  • Administrator
  • *****
  • Posts: 39026
  • Country: au
    • EEVblog
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #84 on: January 13, 2020, 10:38:01 am »
The first thing you need to do before even considering such an ASIC is to fully itemise the BOM cost for current designs and find out how much you can actually save with this magical custom ASIC. You might very well find it's simply not worth it and doesn't offer enough advantage.
And I'm talking BOM cost down to every last screw and knob.
If you haven't done that then the whole thing is a folly.
 
The following users thanked this post: oPossum, george.b, schmitt trigger

Offline EEVblog

  • Administrator
  • *****
  • Posts: 39026
  • Country: au
    • EEVblog
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #85 on: January 13, 2020, 10:41:27 am »
I am also not gonna spend hours each day arguing with people over which country has the bigger chip industry or the legality of exploiting DRM restrictions on software licenses.

That right there is your problem. You can simply chose to ignore such comments and not respond to them. This is how to forum works, you can direct the discussion by only talking about what you want to talk about and ignoring everything else.
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #86 on: January 13, 2020, 11:05:48 am »
So what would be a BOM cost estimation for the first version:

1. 2 channels (ie. parts similar to the Rigol 1054) - simple AFE and the 1x 8bit ADC (500M/1G samples/s)
2. an FPGA capable enough to process the data and push the data into say 4x64MB sdrams (thus 64bit data access, max 128MB/sec write speed)
3. 4x 64MB sdrams
4. a high speed USB interface (an MCU or embedded into the above FPGA)
5. 1GHz clock generator
6. voltage regulators
7. R,C,L passives
8. 10x10cm 4 layer pcb
9. 2x input BNC, 1x usb and 1x power connector (5V/0.5A).
« Last Edit: January 13, 2020, 11:13:01 am by imo »
Readers discretion is advised..
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 39026
  • Country: au
    • EEVblog
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #87 on: January 13, 2020, 11:09:14 am »
So what would be a BOM cost estimation for the first version:

1. 2 channels (ie. parts similar to the Rigol 1054) - simple AFE and the 1x 8bit ADC (500M/1G samples/s)
2. an FPGA capable enough to process the data and push the data into say 4x64MB sdrams (thus 64bit data access, max 128MB/sec write speed)
3. a high speed USB interface (an MCU or embedded into the above FPGA)
4. 1GHz clock generator
5. voltage regulators
6. 10x10cm 4 layer pcb
7. 2x input BNC, 1x usb, 1x power connector.

That's not a bench oscilloscope. If oyu want to revolutise the bench scope market and get the likes of the big payers to buy in then there has to be substantial cost saving on a proper conventional bench oscilloscope product. Not a bare bones hobby PCB only scope.
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 5570
  • Country: va
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #88 on: January 13, 2020, 11:17:36 am »
.. If oyu want to revolutise the bench scope market and get the likes of the big payers to buy in then there has to be substantial cost saving on a proper conventional bench oscilloscope product. Not a bare bones hobby PCB only scope.
I don't want. I want to know what would be a BOM cost "estimation" for the above hobby PCB :)
PS: 1k quantity for example.
« Last Edit: January 13, 2020, 11:35:30 am by imo »
Readers discretion is advised..
 

Offline aheid

  • Regular Contributor
  • *
  • Posts: 245
  • Country: no
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #89 on: January 13, 2020, 12:18:33 pm »
I want to know what would be a BOM cost "estimation" for the above hobby PCB :)

For reference, the ADC used in the Siglent SDS-1104x (it uses one per pair of channels) is, AFAIK, the HMCAD1511. The per-channel VGA is AD8370 and per-channel ADC driver is ADA4932-1. For two channels, you're looking at about $70 just between those parts.
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11714
  • Country: my
  • reassessing directives...
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #90 on: January 13, 2020, 12:21:08 pm »
I've unlocked this thread because I think it's hugely interesting, please play nice.
and i think you are wise in doing so ;D its not nice when mod can reply without letting the defendent draw their opinion. as someone said everybody has a right to their own opinion... i dont care if this thread is locked or not (i have no slight knowledge in ASIC) but it keeps popping out with admin's replies made me WTH. btw i think one of OP's main point is this...

...use the same chip in a $450 scope as their $45k scope.

but imho, it will take someone with more knowledge in this area to make a start. simply saying "hey come on lets make this!" thread will not go anywhere. when someone starts with something usefull, the rest will join in. maybe for the start, doing in FPGA maybe? ASIC is too much of a take, as many others mentioned about the cost etc etc.. but as i said, i dont really care much its not my league, i already got a decent high end scope from used market. so if anything i have to say... lets scour the feabay for this kind of gems... ymmv. peace no war carry on gais.
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15799
  • Country: fr
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #91 on: January 13, 2020, 12:45:59 pm »
Yup yup yup. Wise thing obviously would be to start with a prototype design with off-the-shelf parts (ADCs, FPGA, etc.) Once there's something working, there can be a subsequent phase of trying to simplify it as much as possible. Then once it's there, going to an ASIC (with all the pitfalls we talked about) could be considered. But going for it right from the start would be a mistake, and as Dave mentioned, would make the crowdfunding campaign hard anyway, as such an ambitious goal without any prototype to show? Not gonna work.


 
The following users thanked this post: NiHaoMike

Offline ogden

  • Super Contributor
  • ***
  • Posts: 3731
  • Country: lv
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #92 on: January 13, 2020, 12:49:04 pm »
For reference, the ADC used in the Siglent SDS-1104x (it uses one per pair of channels) is, AFAIK, the HMCAD1511. The per-channel VGA is AD8370 and per-channel ADC driver is ADA4932-1. For two channels, you're looking at about $70 just between those parts.
Right. Obviously much better idea than scope ASIC is ADC Kickstarter project :-DD
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11714
  • Country: my
  • reassessing directives...
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #93 on: January 13, 2020, 02:25:15 pm »
For reference, the ADC used in the Siglent SDS-1104x (it uses one per pair of channels) is, AFAIK, the HMCAD1511. The per-channel VGA is AD8370 and per-channel ADC driver is ADA4932-1. For two channels, you're looking at about $70 just between those parts.
one maybe can get cheaper 50% discount at best by buying in say 1000pcs qty, but $35K? only for that? maybe we are looking at $100K (if using FPGA uneducated guess) excluding headache to find and deal with assembler and mold maker (enclosure) its time for a new brand name company, maybe Riglent? or Siglol? or more precisely... SickLoL.. jk..

...you can get tens of GHz of BW using sampling scope technology, or massive array of 1Gsps ADCs to achieve realtime. This is the secret sauce part of mainstream players.
lets forget sampling scope which can be done with KSps or MSps ADC at tens of GHz BW, its different beast people have done it at cheap. but when you said mainstream players are using massive array of ADCs? i dont think so. i believe its another custom ASIC in one package, maybe massive array they are, but they are in one package, and in one IC... (pictured below, unluckily blocked by a heatsink)

4. Equally important is analog frontend. That's the difference between a good scope and a cheap scope. With the same ADC, an R&S scope is quieter than a Rigol, here's the difference....
Thus, I recommend to start from the easy part, AFE, then S/H. Leave the professional people do the digital stuff, and they will trickle down from consumer stuff quickly and cheaply.
if you refer to this thread, Lecroy DDA-5005 DSO Teardown AFE for high end (50 ohm) scope is quite simple, i guess anyone can do with a free EM simulator and proper PCB materials... but once the signal goes into the ADC/DSP/ASIC/whatever it is, thats where the massive array of money is... there are few of them, from ADC/AFE to DSP/ASIC to RAM down the line, those the unobtaniums...



ps: i once tried to figure out the possibility of diy high sampling rate DSO, 10x whats the venerable DS1054Z can, using their brilliant idea of interleaving slower ADCs. the first study is to look at ADC cost in digikey, luckily that alone made me teared the idea into pieces without further wasting my time.... yeah agreed, let the man do their job.. let me tinker with arduino...
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 

Offline OwO

  • Super Contributor
  • ***
  • Posts: 1250
  • Country: cn
  • RF Engineer.
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #94 on: January 13, 2020, 02:38:02 pm »
the first study is to look at ADC cost in digikey,
That's where the mistake is. I know for a fact from working with a startup that FPGA prices at distributors are typically over 10x what you get direct from Xilinx at even small batch quantities, and ADCs may be similar.
Email: OwOwOwOwO123@outlook.com
 
The following users thanked this post: thm_w, iMo

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15799
  • Country: fr
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #95 on: January 13, 2020, 02:57:04 pm »
the first study is to look at ADC cost in digikey,
That's where the mistake is. I know for a fact from working with a startup that FPGA prices at distributors are typically over 10x what you get direct from Xilinx at even small batch quantities, and ADCs may be similar.

Yes... but can you deal with Xilinx or AD directly if you're a small fish (regardless of batch quantities)?
 

Offline OwO

  • Super Contributor
  • ***
  • Posts: 1250
  • Country: cn
  • RF Engineer.
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #96 on: January 13, 2020, 03:02:55 pm »
ADI probably not, but in my experience Chinese chip vendors are more open to talking to random small businesses, so I'm going to investigate that direction first.
Email: OwOwOwOwO123@outlook.com
 

Offline Mechatrommer

  • Super Contributor
  • ***
  • Posts: 11714
  • Country: my
  • reassessing directives...
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #97 on: January 13, 2020, 03:04:58 pm »
the first study is to look at ADC cost in digikey,
That's where the mistake is. I know for a fact from working with a startup that FPGA prices at distributors are typically over 10x what you get direct from Xilinx at even small batch quantities, and ADCs may be similar.
i'm glad if you can share where we can get GSps ADC and parts etc at 1/10X of digikey, it will surely make diy/os designs bloom.
Nature: Evolution and the Illusion of Randomness (Stephen L. Talbott): Its now indisputable that... organisms “expertise” contextualizes its genome, and its nonsense to say that these powers are under the control of the genome being contextualized - Barbara McClintock
 

Offline ogden

  • Super Contributor
  • ***
  • Posts: 3731
  • Country: lv
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #98 on: January 13, 2020, 03:20:01 pm »
Yes... but can you deal with Xilinx or AD directly if you're a small fish (regardless of batch quantities)?
Define small fish :D AFAIK no matter you are small or big fish - quarterly volume and ability to commit to your agreements matters. If interested - ask: https://www.xilinx.com/about/contact/contact-sales.html
 

Offline aheid

  • Regular Contributor
  • *
  • Posts: 245
  • Country: no
Re: Oscilloscope ASIC Kickstarter instead of Open scope.
« Reply #99 on: January 13, 2020, 03:21:00 pm »
i'm glad if you can share where we can get GSps ADC and parts etc at 1/10X of digikey, it will surely make diy/os designs bloom.

Sarcasm aside, I'd be interested in any such results. I tried searching on LSCS and such but couldn't find any.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf