| Electronics > Projects, Designs, and Technical Stuff |
| Oscilloscope ASIC Kickstarter instead of Open scope. |
| << < (7/37) > >> |
| excitedbox:
That is why we need industry support. Hantek, MCH, Siglent, Picoscope, Voltcraft. Developing their own Asic might be out of the range of each of them individually but if they each only have to put up 5-10% they might be willing to invest. They would be the ones benefiting the most from it in the end and in the grand scheme of things is 50k no that much to invest in the future of their business. You have to remember that all their competitors now have their own chips so they are falling behind. 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. The reason they can each use the same chip is the same reason they can use the same FPGA or ADCs etc now. One company might use 1 asic for 4 "slow" channels while picoscope uses 4 asics to give you the maximum bandwidth and sampling speed on each channel. One company might decide to use a USB2.0 while another uses USB 3.0 or USB C or Ethernet, or Wireless, One uses a touch screen while the other uses knobs and buttons. All those feature differences will be controlled through the ARM processor though which is separate from the ASIC. When you look at the Asic as a complex signal capturing chip just as you would use a ADC to get your signals in another project you see that they really are not all developing the same products in the end. It is just 1 small piece of the product that will be the same. Just as in the computer industry there are only a couple CPU architectures but OEMs don´t all have the same laptops. |
| thm_w:
--- Quote from: excitedbox on January 09, 2020, 10:04:10 pm ---The Rigol 4000 or 5000 I forget which one, costs $900 for the base model but over $4000 with all upgrades. That means it contains under $500 worth of parts. Is it really justified to charge almost 1000% for the software that has no repeat cost associated with it? --- End quote --- MSO5072 costs $909, has 8GS/s stock and likely 10Gs/s capability built in. You'd be better off buying that and hacking the front end to 1GHz. The cheapest siglent is $2,900 and has 5Gs/s, so there is no comparison here in terms of raw performance. The thing is though, with that ASIC, they are using the same chip in their $10,000+ oscilloscopes. So you get payback from those sales which recovers the initial NRE of designing the original ASIC. Its the same with keysight, except they kept the ASIC in their high-end scopes for years then eventually trickled down to the low end. Rigol would probably sell you the ASIC, if you weren't building a competing product, and they'd probably charge you $500+ per piece. When you just build the low end model, and sell it for $900, you have no ability to recover those NRE costs. |
| excitedbox:
that 8gsps is on 4 channels which you cant use all of at 70mhz. to use the full bandwidth and sampling you need all the upgrades which brings the price to over 4k. That scope is an outlier and when it came out I was surprised. Rigol was celebrated for releasing that scope for such an affordable price. Rigol is not as bad as the other companies when it comes to their base models because they are hackable. But the upgrades are still over priced. You could also say Hantek makes affordable scopes so this isn`t needed. These are all excuses. Companies like Hantek have given us a good product at an affordable price which is why they don´t have the budget to develop their own chips. I for one would have no problem investing a couple hundred knowing that it will allow them to continue doing so at an even better price to performance. Think of it this way If 2000 people each paid $250, (Before people misinterpret that again. NO I don´t expect 2k people to do that) any budget manufacturer could give you 8gsps scopes for $900 like Rigol can but with all features unlocked from the start. Not having to pay another $3500 for all the upgrades. Last year that $900 MSO cost $4500 fully unlocked. That is the price difference that is possible. Also keep in mind that hacking the scope is piracy and is illegal because that is a software license they are selling you. 1. Multiple companies using the same chip spreads costs. (This seems to be the hardest for people to understand. People keep acting like this is going in only 1 product) 2. Manufacturing costs are tiny in comparison. (AMD pays less than $13 for a CPU made on the MOST expensive node while this will be one of the cheapest. A qty discount can only go so far since there are still set costs such as materials and EUV is very slow) 3. Companies offer shit software and locked down systems. 4. This chip has other uses other than Scopes. 5. These features are not set in stone. (I said in my 1st, 2nd, 3rd, 4th, 6th, and 8th post that we would need to see what makes sense.) 6. Scopes are multi year products. Meaning 1 chip will be sold over several years. (if 1% of scopes use this chip that means over 1 million units in 5-7 years.) 7. We don´t have any of the facts yet and nothing is set in stone there is no way to determine if it REALLY makes sense until you look at the math. 8. This is not going to happen through a kickstarter. The title was hyperbolic to say that these OS scope kickstarters are misplaced because they can`t compete with an ASIC. (should have been clear from my first post. Where I outlined in detail forming a foundation and going to companies for support). I know this comes off as harsh but I feel like I am talking to a wall. People keep repeating the same arguments that don´t hold up if you look at the numbers that we do have. Not one of the people posting these arguments seem to have read even the last 3 comments. They skim the first paragraph and post the first thought in their mind. I am going to have to start a new thread and maybe move to a different forum with a less shoot from the hip impulsive response. This thread has gone completely off the rails and there is nothing productive happening here. |
| thm_w:
--- Quote from: excitedbox on January 10, 2020, 12:20:10 am ---that 8gsps is on 4 channels which you cant use all of at 70mhz. to use the full bandwidth and sampling you need all the upgrades which brings the price to over 4k. That scope is an outlier and when it came out I was surprised. Rigol was celebrated for releasing that scope for such an affordable price. Rigol is not as bad as the other companies when it comes to their base models because they are hackable. But the upgrades are still over priced. You could also say Hantek makes affordable scopes so this isn`t needed. These are all excuses. Companies like Hantek have given us a good product at an affordable price which is why they don´t have the budget to develop their own chips. I for one would have no problem investing a couple hundred knowing that it will allow them to continue doing so at an even better price to performance. Think of it this way If 2000 people each paid $250, (Before people misinterpret that again. NO I don´t expect 2k people to do that) any budget manufacturer could give you 8gsps scopes for $900 like Rigol can but with all features unlocked from the start. Not having to pay another $3500 for all the upgrades. Last year that $900 MSO cost $4500 fully unlocked. That is the price difference that is possible. Also keep in mind that hacking the scope is piracy and is illegal because that is a software license they are selling you. --- End quote --- Sorry but I don't care what the upgrade costs, its not relevant to me. Publishing a binary patch is not piracy and has not been shown to be illegal, as far as I know. I disagree that any budget manufacturer can give you a 4-ch 450MHz 8GSPS scope for $900 without losing their shirts. |
| unitedatoms:
--- Quote from: T3sl4co1l on January 09, 2020, 06:26:14 pm ---But general purpose devices already exist: analog front ends, fast ADCs, SoCs and FPGAs for continuous, burst, SDR and ET sampling, and waveform analysis. Which are indeed what a lot of scopes already use, and for good reason. :) --- End quote --- From years ago what I remember about advertised breakthroughs in oscilloscope ASICs is two things: One brand advertised 50000/sec screen updates, may be meaning the good coverage of triggered sweeps ratio to missed events. Another manufacturer claimed trade mark of "Digital phosphor". That made me think that actual challenge in scope ASIC is dual-or-multiport memory with digital adders and faders per "pixel" (bin of Y magnitude). Say for Megasample wide X window you actually need full vertical Y column of "pixels" with all accumulated counts and fading. Perhaps 500x16 bit values. Before any optimization and compression that means 1GByte of multiport SRAM. That is core of ASIC. The rest are small things like interpolators, non-linearity correction by lookup, addressing routes/pipes to feed the SRAM, adders, faders, output interface to present SRAM to external CPU bus. The multiport SRAM can be turned into external normal DRAM may be, say put large amount of DRAM chips on individual buses for parallelism, as much as pincount allows. Then ASIC becomes a deserialising data sorter, updater to external DRAM, plus some DMA from the point of view of CPU. I am just theoritizing about what "universal" can there in such pure ASIC for reusability as pure digital chip. May be such chip or major part of such chip already exists in form of say obscure telecom "Bla ATM arbiter, backbone router, massive crossinterconnect Cisco Bla thing" Edit: @excitedbox. I think there is nothing negative in this thread. The idea alone is very novel and disruptive. Speaking of shooting from hip. Which is also good. Internet is fast place. Edit2: By "novel and disruptive" I mean idea of creating ASIC market for scopes. Not other related ideas of conquering existing established markets. Entering the established market is bad idea. Creating your own new market is good idea. That was some advise I heard from one successful tech in business. |
| Navigation |
| Message Index |
| Next page |
| Previous page |