> ... we built out a pretty significant portfolio of these applications solutions and they make it very easy for the customer to adopt our devices into their systems...
Very disappointing that on the topic of Software/Toolchain it's more of the same business decisions of just copying the Xilinx/Altera paradigm.
Since they seemed rather non-agressive towards Yosys/SymbiFlow they made me quite hopeful that Lattice would be the fpga players that would kickstart open-source toolchain.
I am tired of being forced to use software that vendors treat as a cost centre. I want to buy your hardware, I want the people that I work with to buy your hw.
I am tired of vendors forcing a graphical programing toolchain that hides complexity behind half baked software. And judging from the Altera/Xilinx forums I know I cannot be alone in this.
The secret sauce is in the software. The hardware isn’t that revolutionary. On top of that, fpga vendors have this habit of outsourcing every little thing. This makes it super difficult for them to embrace open source give the IP liabilities.
The secret sauce may be in the software, but the software they sell you is a 1998 edition of Borland C++, and we're all screaming at them to please just make their hardware an LLVM target already.
I had fun years ago with FPGA's, first on a digilent nexus 3 and then a terasic de0 nano. I kinda gave up as the IDE's sucked and sucked harder if you were on anything other than windows. I spent hours getting Quartus running on Ubuntu, something to do with the draconian licensing software. I also had issues with ISE on Linux and never bothered to move on to Vivaldo for the same reason, didn't want the headache.
Unfortunately, 10G serdes and >50kLUTs are table stakes to run the applications I have in mind...I think we're a long way from officially supported open toolchains for medium-size FPGAs.
If we could get similar momentum we have with gcc and LLVM behind projects like yosys and nextpnr, FPGAs would be much more useful and likely be a larger market overall.
I'm willing to personally donate 10k a year to projects like these to liberate myself from crappy vendor tools that I can't improve.
And what really annoys me about Lattice in particular is that they shut down their community forum and now there is basically no way to get any form of support when using a free license.
A while ago I found a bug in Lattice Radiant that produced broken PLL configurations. I reproduced this and confirmed that it doesn't happen in their iCECube software, but had no desire to argue with their support about why they should even look at it when I'm not paying $$$ for the software...
I permanently switched to the open toolchain and never had any serious issues since...
While others keep their bitstreams closed and forums shut down, QuickLogic has opened theirs. They opened up the arch information needed for the open source tools to correctly support the architecture, which yields a better product for users. QuickLogic also has a forum that covers the open source tools. No need to pay to play. https://forum.quicklogic.com/
Hi Brian, welcome to HN and thank you for joining this discussion.
I will admit that QuickLogic was not a household name for me but the SoCs you have do tick a lot of boxes for me. I have now ordered a Qomu dev kit to evaluate, I am looking forward to see how the workflow/toolchain are like.
A final question, when trying to order a QuickFeather from Europe, I get a 'restricted availability' message that Mouser is unable to deliver the dev kit in my EU country due to 'government regulations'.
In the past I had to go through a few hurdles/forms when using a specific chip for a medical device because it apparently was also used in some form of US defense application. Is that something EU buyers should expect from Quicklogic products in general? It's not a show stopper but it would be good to know ahead of designing anything.
The entire EDA stack from front to back is so ridiculously bad. System Verilog, UVM and the ecosystem that works on it are equally terrible. Just depressing all around.
Once we have a high quality "LLVM of digital logic", we have a great foundation for better high-level tools. It's just a matter of funding and motivation. In my dreams some Chinese FPGA upstart decides to invest huge in the open tools rather than negotiate with the current toolchain vendors...
Have you read the verilog spec? Half the commercial vendors haven’t implemented large portions of the spec or have implemented it differently from each other.
I’d love to have the momentum around open source EDA tools. At this point I’m not hopeful though.
I guess I'm not seeing from that quote how they're anti-Yosys/SymbiFlow. Are you saying that they just aren't as helpful towards the open source FPGA tools as they could be? Lattice still seems to be more open towards these kinds of open source tools than Xilinx/Altera by far. Yes, Lattice should probably be more proactive in actually working with the open source developers, but baby steps, I guess.
As someone with breaking into the hardware space with a software background, I've acquired a newfound appreciation for the amount of work that gets poured into GCC/LLVM toolchains.
Open implementations which acquire enough traction become open standards. I'm doing block-level CPU simulation work for my dissertation, and there's pretty much a wall of abstraction where my work has to stop before it explodes with licensing + compatibility landmines. A rock solid tooling ecosystem I could target for FPGAs and then eventual fabrication would be the dream.
I recently discovered that Xilinx has lots of small R&D things I github. I had the opportunity to work with the team behind FINN (https://xilinx.github.io/finn/) and they're trying to do everything, put everything they develop, on github. I will praise this in every future meeting I have with any Xilinx rep.
I feel this is a mentality change, albeit slow, that is starting to happen. We still don't have source code for Intel MKL, or all of OpenVINO and it bit us in the ass several times, so much I'm really trying to find ANY OTHER SOLUTION and pay the perf hit, if I can own and debug it. I'm not even asking for github or even publicly available tarballs, but 'I paid I want the source and I want to be able to look inside to be autonomous'. Make it OSS without community involvement, for all I care, but dammit let your paying customers look and debug. Let us automatize, let us own the tech.
> I feel this is a mentality change, albeit slow, that is starting to happen.
Nah, at best it could be a rediscovery that took an inexplicable amount of time in the wake of the UNIX wars and dotcom bubble. If you look at paper titles from the 70s, when things were advancing at breakneck speed, you'll see "Open" commonly prefixing proposals - even stuff coming out of IBM (sadly looks over at PS/2 doorstop). Though, I think it is far more likely that businesses have developed a strategy that involves extracting as much free labor possible without actually generating free software: OpenSource® <insert Intel jingle>.
I am somewhat sympathetic to their problem though, the early developments occurred in a very different environment - involving companies that couldn't have existed as they did without state sponsorship.
I wonder how much that matters, and how uniformly it applies (primarily on the dimension of industry), when institutional inertia isn't perfectly aligned with employee preference. For example: Microsoft employees... lots of mixed signals. Supposedly their love of opensource has resulted in a Microsoft that no longer plots the destruction of free software, and everybody can rest assured that the bad old days are over. But when they openly protest DoD contract work, they suddenly find themselves impotent to stem the flow of HoloLens units going to the Army. Obviously there are a lot of different things at play - even within that single example, but you can find the same pattern all over the place. So I don't put a lot of weight on the employee sensibility side of the scale... despite the fact that it may diminish, to some degree, the accomplishments of the graybeards - who I venerate more than most.
I know what you mean and... I don't know. I work in a big company, in an industry where I ask myself moral questions quite often and I know if I'm faced with or if I learn of a sketchy or immoral thing, I'll be sure to talk loudly about it, and feel like I could be heard. But if it were to have no effect, I'd try to vote with my feet asap, but there's also kids to feed etc. I'd sure be less motivated and would be looking for an out. Bye bye Highly invested collaborator, hello 9-5-fuck-you-pay-me.
I wasn't saying one team would steer the ship, but rather that the culture of opensource also lives inside Xilinx and I'll take any progress I can. Especially on such emergent tech and after the pains of closed-source AI toolkits and even inside Xilinx, the pain of using their DPU/VITIS-AI offer: oh you want to do X? Too bad, the IP can't be changed or touched (and is or will even be encrypted? Woa).
That would be the absolute free market proposed solution for addressing social externalities in the absence of regulation: companies that make employees unhappy are rendered less competitive due to higher churn and a need for above market wage levels to act as a perpetual retention bonus. I expect that the more readily an industry can convert novel thinking into economic advantage, the greater the strength of that phenomenon, and the more challenging it is to find a global minimum for both individual and organization prosperity.
My last 9-5, many years ago, involved writing logistics software in an industry that: didn't present moral dilemmas, didn't especially reward creativity, and certainly didn't leave one feeling any real sense of accomplishment - outside of occasionally conquering another rung on the corporate ladder. It did, however, regularly present interesting problems - so I was pretty happy getting paid to solve non-contrived puzzles. But as I assumed more responsibility I surrendered more independence, and it eventually got to the point were it was a job - so I asked for a pay bump large enough to compensate the loss of intellectually challenging problems. I found myself reading IRS guidance for independent contractors shortly thereafter.
Over the years I've considered paths that would eventually necessitate hiring, but I haven't been able to come up with a solution for the problem of balanced interests. One that would have been able to retain my younger self while also maximizing utility. I'm not sure if such a financial instrument exists, one that incentivizes the execution of long term objectives in the company's interest, doesn't collect dead weight silent partners, doesn't leave anyone starving, and doesn't end in a coup.
I'm with you here. Not found the solution yet. Though there's a strange thing going on lately with lots of my generation is leaving to do some public service mission, either encouraged by the company (military 'réserve' is what we call it around here, and I hear the missions are really great, and the influx of so much expertise and good will from individuals is - from what I'm hearing - received extremely well) or just f'in leaving to help solve a societal problem with whatever tools they can bring to the table (be it IT for a non-profit, setting video conf hw/sw, doing personal support, raiding ebay for cheapest most-solid HW for servers or other missions, connecting PR people to worthy causes to pour their cash in, teaching all kinds of basic modern skills like opsec, itsec, backups, event organisation tools, group communication tools, ...). Huge pay cut but I see it a lot, even more since COVID-19. I may be in a bubble but I was surprised.
I know a couple people in their late 20s who recently enlisted, which is very odd and makes me suspect that recruiters are selling it as a sort of pause button for adult responsibilities. I joined shortly after 9/11 and ended up fighting in Fallujah... I wouldn't recommend it to anyone I liked. After I got out I was living hand to mouth for a little over a year - which might have been the single most influential hardship in my life. If I were to do any real charity work (instead of a tax advantaged hobby project) it would be teaching personal finance, the need for it can't be overstated - as the financial system actively punishes the unwary. I would not relish the task of brush-up for certification though...
One of the paths to me, to steer things in a way, is hiring, and even just internships. I'm REALLY upfront with the kinds of things we build and sell, and am trying to be systematically the 'first interviewer' they meet of my entity in bigco. And I go: shiny toys! But for uses that some may find hard to live with... Hot tech challenges! But most days are still filles with a lot of process and bureaucracy... Mission critical systems! But it means being beyond reproach so no slacking on allllll the boring tasks, especially documents and reviews (not 'just' code!) and tests. We do some R&D but we're an operational entity at heart. We're a leader in sales/installs in the industry, but that means you should read the Innovator's Dilemma. We work 'very hard' when we work but we like our family lives.
I try to distill our culture (actual and the one I strive) as a first image of Corp X. And I always try to steer those who don't feel they'd fit in, but have a great CV or story, to other parts of the corp or to places I know would fit better.
It all starts at first impressions and those who want to define the future of the company should invest energy in hiring and mentoring, mentoring and mentoring again. 'no we don't do that here' at the coffee machine, with a smile. 'oh I wouldn't do that, that's a breach of your customer's trust, if you want I can come to the next meetings with you or help you craft a written answer with the proper people in Cc... Yes that sounds like being an annoying busy body but the help seems usually very appreciated.
You aren't alone. Most people who work with FPGAs hate the toolchains.
The entire FPGA industry and to a larger extent the entire semi industry in plagued by this problem.
This is a deep cultural issue.
Look for example at NVidia and their dumb secrecy policy around drivers (just the first example on top of my head, examples abound).
Secrecy and making your customers pay for tools is dyed in the DNA of these people and it is going to take a very long time for them to understand the economic advantage of openness.
It took MSFT almost 30 years to finally understand the lesson that giving away dev tools for free (e.g. Visual Studio) was a competitive boost.
It's been like this for over 20 years. I was trying to persuade people on comp.arch.fpga 20 years ago that we should band together and produce some open tools for Xilinx, and I was eager to do it myself.
People warned me that Xilinx were legally very aggressive and that I could lose everything I owned. Nobody wanted to get involved then. At one point an academic paper about alternative programming techniques on Xilinx devices withdrew some of their work, out of legal concerns. So I kept out of it too.
In fact, I quit working with FPGAs for a long time (it had been my main job, developing and using hardware compilers at a higher level than VHDL/Verilog), so that my previous use of Xilinx and Altera software might become a distant memory and I wouldn't feel that the licensing restrictions accompanying their old software would constrain reverse engineering on a new generation.
I was intending to eventually reverse engineer some of the devices from first principles without using their software. We know that can be done to some extent just by programming bitstreams. Timing and power models are more difficult to reverse engineer, but not impossible and I had some measurement ideas, and I had in mind to additionally etch and photograph the chips as well, or perhaps use focused X-ray tomography. But that never happened; life happened instead.
Times have changed. The FPGA patents from that era have expired. You can build a decent device without infringing, in principle. There's a rapidly growing movement in open source hardware on FPGAs and even ASICs. And people have been reversing Xilinx among others, more confidently than the people of 20 years ago were willing to.
For today I have two words: QuickLogic and OpenFPGA. QuickLogic devices may be small compared with the high end, but they are explicitly supporting open source toolchains, better than Lattice's passive tolerance. This is a breath of fresh air, and I hope they are successful. OpenFPGA is actual open source FPGAs.
"I am tired of vendors forcing a graphical programming toolchain that hides complexity behind half baked software."
Amen. The size of these closed source toolchains is often enormous. I wonder how many folks are not content with using relatively large Windows point-and-click software to program relatively small form factor hardware.
Hey, guys. Looks like I am late to the show with my comment here, but I am pushing 20 years in FPGA dev experience and hoping that maybe I can share some perspective on what seems to be a reoccurring theme on HN:
Why aren’t there any open FPGAs or at least more Open FPGA tools? Why are vendor FPGA tools so huge, terrible and hard to use? Open FPGA tools are better and the way to go! etc.
In one of Dr. Cutress’ questions he states and then asks:
> … they're just hard to develop for. You need to know how to use them before you use them , which sounds like the wrong way to learn how to code! What exactly is Lattice doing to kind of ease that transition for people who may understand software, but are kind of new to the hardware?
Why is this? What can Lattice do?
It’s because one does not “code” an FPGA, they design a microchip and that process involves a lot more than writing code. It is pretty hard to do right. FPGA is on the spectrum of VLSI microchip implementation technologies. That’s why they compare FPGA’s to ASICs in the interview.
So in general the progression is FPGA -> GA (Gate Array) -> Standard Cell ASIC -> Full Custom ASIC
FPGA development is microchip development. It requires a great deal of tacit knowledge, experience and a background in electrical engineering and digital design and a number of other skills to do well.
In order to develop open tools for these devices most of the devs in the community would have to have a lot of this knowledge and experience and most if not all of the FPGA vendor’s IP on all of their devices in order to work on an open tool chain. In the end an open tool chain will do little to alleviate all the pains of the chip design process anyway so it really isn’t worth the effort to make open tools. It is way easier to just use the vendor’s tools once you know what you are doing. Most of the tools have gotten much better in the last 20 years.
That is why you will not see open tools for anything but the smallest and simplest devices that can be reverse engineered.
IMO, the best thing lattice can do for someone who doesn’t have any experience with the technology is to point them to the Lattice Partners program to find an expert to help.
I am probably not making any friends here by saying this, but I welcome any counterpoints we might argue together here.
Thanks for the perspective. For someone who used to study EE / basic digital design but took the software route as a career, do you have suggestions on FPGA projects to contribute to on the side akin to the software world’s ability to contribute to an endless number of open source projects? I would love to chip away at learning how to work with anything more complex than the academic Xilinx boards we used in school years ago.
It's a decent overview of Lattice's business and some of their applications. Nothing too technical. A shame Ian didn't ask about open source toolchains.
I started watching Ian this year and have really enjoyed his reporting. I would like to learn more about how he learned this complicated industry… he’s so knowledgeable!
10 years of pattern recognition and probing questions to the point of being annoying enough that companies have to drag out expert engineer XYZ every time I attend a meeting but then I end up striking a rapport with that engineer who tells me what really goes on :)
I was surprised to hear lotsa talk about FPGAs and basically nothing about HDMI or silicon video tech.
Turns out Lattice bought Silicon Image in 2015, then turned around and sold off the HDMI portion of the business to another company I've never heard of, which was again sold - to Analog Devices.
I like the direction Lattice is going. The bigger FPGA developers Xilinx and Intel are going for bigger, denser, faster, more ultra-ram, and as many LUTs as the chip can handle; Lattice is going for smaller, cheaper, and more accessible. I'm very surprised to hear they're going after low power edge AI, but it's niche markets like that where FPGAs shine!
Basically just a marketing fluff interview. Would love to see lattice ditch their shitty tooling and dedicate time to developing yosys and nextpnr. Nextpnr especially misses crucial features.
I don't want my comment to detract from what I'm thinking was a pretty good interview. I really didn't like the CEO's response to this question though:
> One of the things with FPGAs - I get a lot of feedback for is that they're just hard to develop for. You need to know how to use them before you use them, which sounds like the wrong way to learn how to code! What exactly is Lattice doing to kind of ease that transition for people who may understand software, but are kind of new to the hardware?
If you're talking about the words, it's a transcription of how he talks. People rarely talk in full sentences and if you actually read how someone talks it's very hard to follow. I cleaned it up a lot, but still wanted to keep some of that nuance in. There's a video version of this on youtube, and try and just listen to the words as if they were written, and leave in all the 'so the thing is' and 'you know's.
I'm actually not referring to the transcriptions and it's a small thing, but it really does detract from the overall quality of the article a bit.
But I really don't blame you at all. When I write for myself, I try very hard to self-edit, but I still leave all kinds of small silly little grammar/spelling mistakes in.
I really did like the way you setup the interview and the questions you ask were pretty on point, so don't take this negative feedback too strongly.
51 comments
[ 3.0 ms ] story [ 109 ms ] threadVery disappointing that on the topic of Software/Toolchain it's more of the same business decisions of just copying the Xilinx/Altera paradigm.
Since they seemed rather non-agressive towards Yosys/SymbiFlow they made me quite hopeful that Lattice would be the fpga players that would kickstart open-source toolchain.
I am tired of being forced to use software that vendors treat as a cost centre. I want to buy your hardware, I want the people that I work with to buy your hw.
I am tired of vendors forcing a graphical programing toolchain that hides complexity behind half baked software. And judging from the Altera/Xilinx forums I know I cannot be alone in this.
I'll take my hardware without the sauce, thank you!
I wish!
Open your stupid bitstream formats FFS.
Linux is supported.
I'm willing to personally donate 10k a year to projects like these to liberate myself from crappy vendor tools that I can't improve.
A while ago I found a bug in Lattice Radiant that produced broken PLL configurations. I reproduced this and confirmed that it doesn't happen in their iCECube software, but had no desire to argue with their support about why they should even look at it when I'm not paying $$$ for the software...
I permanently switched to the open toolchain and never had any serious issues since...
I will admit that QuickLogic was not a household name for me but the SoCs you have do tick a lot of boxes for me. I have now ordered a Qomu dev kit to evaluate, I am looking forward to see how the workflow/toolchain are like.
A final question, when trying to order a QuickFeather from Europe, I get a 'restricted availability' message that Mouser is unable to deliver the dev kit in my EU country due to 'government regulations'.
In the past I had to go through a few hurdles/forms when using a specific chip for a medical device because it apparently was also used in some form of US defense application. Is that something EU buyers should expect from Quicklogic products in general? It's not a show stopper but it would be good to know ahead of designing anything.
I’d love to have the momentum around open source EDA tools. At this point I’m not hopeful though.
Open implementations which acquire enough traction become open standards. I'm doing block-level CPU simulation work for my dissertation, and there's pretty much a wall of abstraction where my work has to stop before it explodes with licensing + compatibility landmines. A rock solid tooling ecosystem I could target for FPGAs and then eventual fabrication would be the dream.
I feel this is a mentality change, albeit slow, that is starting to happen. We still don't have source code for Intel MKL, or all of OpenVINO and it bit us in the ass several times, so much I'm really trying to find ANY OTHER SOLUTION and pay the perf hit, if I can own and debug it. I'm not even asking for github or even publicly available tarballs, but 'I paid I want the source and I want to be able to look inside to be autonomous'. Make it OSS without community involvement, for all I care, but dammit let your paying customers look and debug. Let us automatize, let us own the tech.
Nah, at best it could be a rediscovery that took an inexplicable amount of time in the wake of the UNIX wars and dotcom bubble. If you look at paper titles from the 70s, when things were advancing at breakneck speed, you'll see "Open" commonly prefixing proposals - even stuff coming out of IBM (sadly looks over at PS/2 doorstop). Though, I think it is far more likely that businesses have developed a strategy that involves extracting as much free labor possible without actually generating free software: OpenSource® <insert Intel jingle>.
I am somewhat sympathetic to their problem though, the early developments occurred in a very different environment - involving companies that couldn't have existed as they did without state sponsorship.
I wasn't saying one team would steer the ship, but rather that the culture of opensource also lives inside Xilinx and I'll take any progress I can. Especially on such emergent tech and after the pains of closed-source AI toolkits and even inside Xilinx, the pain of using their DPU/VITIS-AI offer: oh you want to do X? Too bad, the IP can't be changed or touched (and is or will even be encrypted? Woa).
My last 9-5, many years ago, involved writing logistics software in an industry that: didn't present moral dilemmas, didn't especially reward creativity, and certainly didn't leave one feeling any real sense of accomplishment - outside of occasionally conquering another rung on the corporate ladder. It did, however, regularly present interesting problems - so I was pretty happy getting paid to solve non-contrived puzzles. But as I assumed more responsibility I surrendered more independence, and it eventually got to the point were it was a job - so I asked for a pay bump large enough to compensate the loss of intellectually challenging problems. I found myself reading IRS guidance for independent contractors shortly thereafter.
Over the years I've considered paths that would eventually necessitate hiring, but I haven't been able to come up with a solution for the problem of balanced interests. One that would have been able to retain my younger self while also maximizing utility. I'm not sure if such a financial instrument exists, one that incentivizes the execution of long term objectives in the company's interest, doesn't collect dead weight silent partners, doesn't leave anyone starving, and doesn't end in a coup.
I try to distill our culture (actual and the one I strive) as a first image of Corp X. And I always try to steer those who don't feel they'd fit in, but have a great CV or story, to other parts of the corp or to places I know would fit better.
It all starts at first impressions and those who want to define the future of the company should invest energy in hiring and mentoring, mentoring and mentoring again. 'no we don't do that here' at the coffee machine, with a smile. 'oh I wouldn't do that, that's a breach of your customer's trust, if you want I can come to the next meetings with you or help you craft a written answer with the proper people in Cc... Yes that sounds like being an annoying busy body but the help seems usually very appreciated.
You aren't alone. Most people who work with FPGAs hate the toolchains.
The entire FPGA industry and to a larger extent the entire semi industry in plagued by this problem.
This is a deep cultural issue.
Look for example at NVidia and their dumb secrecy policy around drivers (just the first example on top of my head, examples abound).
Secrecy and making your customers pay for tools is dyed in the DNA of these people and it is going to take a very long time for them to understand the economic advantage of openness.
It took MSFT almost 30 years to finally understand the lesson that giving away dev tools for free (e.g. Visual Studio) was a competitive boost.
People warned me that Xilinx were legally very aggressive and that I could lose everything I owned. Nobody wanted to get involved then. At one point an academic paper about alternative programming techniques on Xilinx devices withdrew some of their work, out of legal concerns. So I kept out of it too.
In fact, I quit working with FPGAs for a long time (it had been my main job, developing and using hardware compilers at a higher level than VHDL/Verilog), so that my previous use of Xilinx and Altera software might become a distant memory and I wouldn't feel that the licensing restrictions accompanying their old software would constrain reverse engineering on a new generation.
I was intending to eventually reverse engineer some of the devices from first principles without using their software. We know that can be done to some extent just by programming bitstreams. Timing and power models are more difficult to reverse engineer, but not impossible and I had some measurement ideas, and I had in mind to additionally etch and photograph the chips as well, or perhaps use focused X-ray tomography. But that never happened; life happened instead.
Times have changed. The FPGA patents from that era have expired. You can build a decent device without infringing, in principle. There's a rapidly growing movement in open source hardware on FPGAs and even ASICs. And people have been reversing Xilinx among others, more confidently than the people of 20 years ago were willing to.
For today I have two words: QuickLogic and OpenFPGA. QuickLogic devices may be small compared with the high end, but they are explicitly supporting open source toolchains, better than Lattice's passive tolerance. This is a breath of fresh air, and I hope they are successful. OpenFPGA is actual open source FPGAs.
How well does their FPGAs work and how much can you do with the open source tools (e.g. timing driven synthesis)?
Amen. The size of these closed source toolchains is often enormous. I wonder how many folks are not content with using relatively large Windows point-and-click software to program relatively small form factor hardware.
Why aren’t there any open FPGAs or at least more Open FPGA tools? Why are vendor FPGA tools so huge, terrible and hard to use? Open FPGA tools are better and the way to go! etc.
In one of Dr. Cutress’ questions he states and then asks: > … they're just hard to develop for. You need to know how to use them before you use them , which sounds like the wrong way to learn how to code! What exactly is Lattice doing to kind of ease that transition for people who may understand software, but are kind of new to the hardware?
Why is this? What can Lattice do?
It’s because one does not “code” an FPGA, they design a microchip and that process involves a lot more than writing code. It is pretty hard to do right. FPGA is on the spectrum of VLSI microchip implementation technologies. That’s why they compare FPGA’s to ASICs in the interview.
Don’t believe me? Look at this article: https://www.tutorialspoint.com/vlsi_design/vlsi_design_fpga_...
So in general the progression is FPGA -> GA (Gate Array) -> Standard Cell ASIC -> Full Custom ASIC
FPGA development is microchip development. It requires a great deal of tacit knowledge, experience and a background in electrical engineering and digital design and a number of other skills to do well.
In order to develop open tools for these devices most of the devs in the community would have to have a lot of this knowledge and experience and most if not all of the FPGA vendor’s IP on all of their devices in order to work on an open tool chain. In the end an open tool chain will do little to alleviate all the pains of the chip design process anyway so it really isn’t worth the effort to make open tools. It is way easier to just use the vendor’s tools once you know what you are doing. Most of the tools have gotten much better in the last 20 years.
That is why you will not see open tools for anything but the smallest and simplest devices that can be reverse engineered.
IMO, the best thing lattice can do for someone who doesn’t have any experience with the technology is to point them to the Lattice Partners program to find an expert to help.
I am probably not making any friends here by saying this, but I welcome any counterpoints we might argue together here.
It's a decent overview of Lattice's business and some of their applications. Nothing too technical. A shame Ian didn't ask about open source toolchains.
Turns out Lattice bought Silicon Image in 2015, then turned around and sold off the HDMI portion of the business to another company I've never heard of, which was again sold - to Analog Devices.
Huh. Silly me for not paying attention.
> One of the things with FPGAs - I get a lot of feedback for is that they're just hard to develop for. You need to know how to use them before you use them, which sounds like the wrong way to learn how to code! What exactly is Lattice doing to kind of ease that transition for people who may understand software, but are kind of new to the hardware?
But I really don't blame you at all. When I write for myself, I try very hard to self-edit, but I still leave all kinds of small silly little grammar/spelling mistakes in.
I really did like the way you setup the interview and the questions you ask were pretty on point, so don't take this negative feedback too strongly.