PDA

View Full Version : Manufacture for Christmas Selling



erik_blevins
09-11-2007, 07:41 AM
1 Does anyone know how to manufacture a these componets?
2 Where do we buy·the components·to assemble and resell?
3 Are·nintendo video games still copyrighted?
4 Who's interesed in putting together a game from scratch?
5 How do we connect old Nintendo games to the Hydra?
6 How would be go about getting the copyrights on a game?
7 Can we sell them like we saw in the mall small game packages of 30-100 games in each package at Christmas time?
8 Finally, I'd like to know about the engineering books available for the manufacture of microchips.

We should be able to make a, super nintendo,·64 bit if we know what were doing for cheap still. I think the Hydra is just a stepping stone of 32 bit.

JT Cook
09-11-2007, 08:12 AM
Give it up. You can't connect NES games to the Hydra, the copyrights to the games will continue to be renewed by Nintendo, the 30-100 games in kiosks sold in malls are illegal pirated version, and you will never be able to manufacture anything cheap enough to sell to make money off of.

erik_blevins
09-11-2007, 11:25 AM
Sounds like you have personal experience with copyrights of the old games. Please enlighten us how we can save our time and stay on the right side of the law? What Personal experience have you had in contacting the Nintedo companies? If you'd rather we sell American games we could go with Jaguar my personal favorite from Atari, or a Xbox rendition. Why should we give up when there is power in numbers and numbers speak volumes. When we havn't begun to try. Please share with us your wisdom and knowledge.

JT Cook
09-11-2007, 08:40 PM
I have never done any license deals or have even ever contacted any companies. Really the only thing you need to make this happen is money. Lots and and lots of money. If you have lots and lots of money, then you need to track down the owner of the copyright. After you track down the copyright owner, then you need to strike a deal where you pay them lots and lots of money to use their copyright.

After that you need more money to manufacture the devices.

Then you need to find people to buy all these devices you have poured lots and lots and lots of money into and hope you didn't just make a huge mistake.

potatohead
09-12-2007, 02:11 AM
IMHO, the power in numbers works against you in this case.

If only some people would be interested, then it's not worth the hassle. If a LOT of people are interested, then they need to introduce their own product for max revenue.

Licensing costs money, and there is always the chance that the resulting product could be ugly, damage the property in question, etc...

End result is a lot of time and effort is required to reach any kind of agreement, if said agreement is even possible. Some projects I've watched succeed are limited, non-profit, historical value kinds of things. It was more about the art than dollars.

Nobody is gonna license you anything that might actually mean dollars! Again, look at it from their perspective. If you make the case for demand, they can just say, "thanks" and fill it however they want to.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness! (http://propeller.wikispaces.com/)

Spork Frog
09-12-2007, 02:56 AM
If I'm reading/understanding your first post correctly, Erik, it sounds to me like you wish to manufacture your own 100-in-one Nintendo games console, as were used to be commonly found in mall kiosks. I don't advise this, for several reasons--

1. Yes, Nintendo ROMS are still copyrighted.
2. The units sold in stores were all pirated games, and there was a big news thing a year or two ago about people getting in big trouble for selling them. They were quite illegal.
3. Just to make this clear, just because the Hydra supports Nintendo and compatible gamepads makes it in other way a recreation of a Nintendo. The controllers are all it supports- not the games, not anything. I believe Nintendo controllers were selected originally for their price and availability (don't quote me on this), but that's it. THE HYDRA IS IN NO WAY A NINTENDO. I seriously doubt it being even able to run a NES emulator (though admittently that would be cool.)

Of course, none of this at all stops you from writing your own games and making units with them to sell. Though as was stated previously, make sure your market is there so you don't just end up in large amounts of debt.

erik_blevins
09-12-2007, 03:54 AM
No I'm not saying skip the copyrights I'm saying call nintendo ask to work with their skimatic and ask if you can have a list of games. For a fraction of the price we are selling them will go to thier copyrights. People don't have copyrights to prevent people from making money, but to get more money.
We have the knowledge we have the tools we should be able to understand and build our own nintendo from scratch. If not design a much more compressed computer chip to do the same thing. It'll work faster be cheaper and you'll probably make alot of money at it.
If we get enough people interested I think we could do it for next Christmas. If not this christmas. I bet Atari will love to do something like that if Nintendo doesn't. We could redo XBOX in a compact version, I'm sure Bill Gates doesn't care. He's got enough money. Just some ideas. The best thing for our economy is to do all the game systems and see which one will give use the most profit.
Since it is copywritten I think we'll have an easier time manufacturing the same specifications if not better ones.

erik_blevins
09-12-2007, 04:10 AM
I'd like to reply to Grasshopper Potato Head. Yes, your right it would be like that, if you make a case for demand then they will fill in the demand. It sounds like you have the marketing experience so why don't you tell us how to avoid that.
I don't see why it would not sell. It is obviously a cheaper substitute for the larger version and a greater player value. They should sell them. Tell them to go ahead I challenge them. All I'm saying is that if you want to make money you need to be positive and push forward to see how far we get. "Rome wasn't built in a day" they say. It will take us some time to find all the obsticales and hoops corporate monsters want us to jump through. We don't know for sure that the answer is no do we? If we tell ourselves no before we get start what's the answer going to be? How about yes and when we hear no we keep saying yes and can I ask you why, does it have to do with the percentage your copyright is receiving, maybe you should consider why after you heard about the piracy why your company saw it too small a profit margin to pursue. Get ahead of these people and answer their questions before they ask them. We will make it a goal to find the answer, yes.
You said yourself there is a demand people have been caught selling pirated video games, now all we need is to give them a percentage of the profit to the copywrite holder. Some one already has done it so can we and we can do it again this time legeally, we know there is a market and it's small enough bigger companies don't care about it. Gentlemen what we have here is a nitch market! Many companies start out just like this, people pull their hair out tring to find Nitch Markets. We will do it and we will make money at it. Not much money view it as a hobby and be suprised by the results. Think of the knowledge we will gain from such a joint venture even if we are unsuccessful people will be knocking our doors down with job offers. It's a win win situation.

potatohead
09-12-2007, 04:29 AM
I think it's a significant and growing niche. I also think your point on the challenge is well taken.

One of the problems to be faced is marketing and building to older demographics. Nobody cares about older gamers (yet!). Should that change, there will be progress.

IMHO, probably one of the better ways to avoid the problem is to do some very serious market research. If that demand is there, and a product can be produced at a good cost, that's worth managing for them, maybe it would fly. That research is necessary to really address, "I think it would sell" -vs- "I would buy one!"

Somebody here should go ask Kurt Vendel (found on Atari Age) about this stuff. The recent Atari Flashback II is a great example of how this stuff can work --and that console contained home brew games to boot!

It also comes down to having to sell too. Everybody wants a WII, it's new, gets the buzz, etc... An older school effort would be a lot of fun, particularly for young, old and very casual gamers. But, so is the WII! Why bother segmenting one's market, when one does not have to? If something is simply not available, then those wanting to fill that need will either need to build their own (that's us), pirate, go nostalgic, or just settle for what is out there.

Say somebody fills that need with the cheap and old school effort. That's a sale right?

What if they filled it with a new model? Still a sale right?

The difference between the two is reinforcing an existing product that's not yet run the life cycle curve, and opportunity cost in that maybe the cheaper one is all that is ever sold, or selling it lowers expectations and value in the market. (That last one is just huge!)

more later: got a call!

Ok, on that matter of demand, there is also distribution and such. The established products have these things in their favor, where new products either need to consume some of those resources (shelf space), or evolve new ones. Getting all that done, or justifying the amount of resources, costs and returns is not easy. IMHO, this is highly likely why this stuff remains a PITA.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness! (http://propeller.wikispaces.com/)

Post Edited (potatohead) : 9/11/2007 10:04:55 PM GMT

erik_blevins
09-12-2007, 06:12 AM
·· You'd be a great politician you said very much without saying a thing.
We are technitians not polititians, how do we do market research is missing from all the technical books I went to school for. The exact mechenism has been misplaced please enligthen us how.
··· Several books leave valuable information out of the books and call themselves Textbooks. I get more information from reading Jules Verne than reading 1,000 pages of Textbooks. So polititians are writting technical books now? No wonder the Amercian economy is being passed up by nemerous countries.
··· You want me to walk around and knock on doors asking if it would sell to waste my time doing this instead of building as a hobbie. I've got an idea I'll be right behind you. Maybe, you should write a marketing plan. Do you know what that intails, maybe you could get a small buisness loan through the small buisness loan department of America. So can we get back to what were suppose to be talking about?
·· Just say you don't want to do it and leave me alone. That so far is my oppinion of forums waste my time. I could have had a dozen conversations in real time about something that is going to get somewhere. If you don't have a clue just say so!
··· How old are you you barely grasp marketing?

JT Cook
09-12-2007, 09:30 AM
erik_blevins said...

·· Just say you don't want to do it and leave me alone. That so far is my oppinion of forums waste my time. I could have had a dozen conversations in real time about something that is going to get somewhere. If you don't have a clue just say so!
··· How old are you you barely grasp marketing?
I would have to agree, you are wasting your time. Whether anyone here has a grasp on marketing is really not relevent to this forum since it is a programming/electronics forum. It is also not relevent because you don't even understand fully what needs to be done to get such a product licensed, created, and sold;·and I myself would wonder how old you are. Infact if I were to guess, I would guess that you would want everyone else to do the work on this, and you take the credit and cash on it.

potatohead
09-12-2007, 09:45 AM
I've been involved in some product launch efforts. Also some marketing on a level where retail distributors are concerned.

I kept it really general just to highlight some areas of concern, that's all. Figured you might read that and go do some digging. Better to do that digging now, rather than later.

For what it's worth, this forum is not where you are going to find some of the information you need. In that, I totally agree. Also, for what it's worth, if you consider some of the above "politics", congrats! You've made one of many realizations necessary to actually research, market, produce and distribute your product.

Finally, should you get it up and going, I want one! I think it's great personally. More power to you.

Again, go do some research on the Flashback II dedicated console, produced by Curt Vendel's team. It's an excellent case study, perfectly applicable to what you are wanting to do.

Edit: (sorry, but I think it is better that you read the following now, rather than later.)

From the business persons perspective!

Techies are a dime a dozen. Really great techies are far fewer, but are expensive, opinionated, and generally problematic.

Sales people are a dime a dozen as well. Of course, really great sales people are fewer too, and expensive, but not so problematic.

One who can cross both bridges is priceless!

I am not as technical as I used to be. I am considerably better at sales, relationship management, project management and have the kind of people skills that say, "trust me, I'm not going to do you wrong" when sitting across the table from prospects in need of technology solutions, consulting, etc...

Want to know why I made that change?

(hope so)

About 15 years ago, I got into a very interesting discussion with an insurance salesman. He was a top notch technical lead, who happened to be on the leading edge of what we all know now to be the outsourcing movement. Years in college, skills up the wazoo, brilliant --unemployable at 40. Why?

Expensive, problematic, opinionated.

He bagged it, does the stuff on a hobby level, and makes his living doing deals day after day.

At the end of the day, technical stuff can be had for cheaper, there is always somebody better and technology is always invalidating skills aquired. In fact, there are tons of people (of which I am one, from time to time) who sell technology that can very easily reduce people count, while at the same time improving overall production. Helps to have outsource capability too.

Do I like it?

No.

However, I did begin to acquire and cultivate the people side of things. That's your politics. No matter what happens technology wise, there are always people needed to interface with people. Managing them, planning for them, rendering specialized services for them, etc...

The current rate of technology adoption more or less puts owning these services in house out of reach for a growing number of companies. So, they outsource a lot of this, and that's where I come in. I can deal with people, shake hands, make deals and engage in politics, while at the same time getting behind the tech and making it do things other people want to do, but would rather pay somebody to do, for them.

I'm writing this, not to blow my own horn, but to relate some experience I had in my 20's that ended up being quite valuable. Perhaps you are there and might learn from it. Perhaps not....

Here is your take away:

At the end of the day, products happen because of the politics. There are lots of great designs. New ones are realized every day. Why don't we see them? Why doesn't the best tech win on it's merits?

Politics my friend!

Those really great people, that can form relationships, make deals, convey and deliver trust, get it done. What ever they are backing is what will be realized on a greater scale. It is the marketing people, sales people, business development people that go do the lunches, get to know the right folks, rope in investors, etc... that make things really happen.

At the bottom are technical people and their skills. They are necessary, but always kept under some level of control.

Want to do your own thing?

(and I hope you do)

Go to work on the business end of things. Start networking, take some people to lunch, get to know those who have control over the means and methods by which your ideas can be empowered. Save your dollars, live cheap and make regular investments in your self, and who you know and most importantly --why you need to know them!

I really didn't mean to piss in your Cheerios. Sorry for that.

If you are pissed, you are pissed. I might as well do some good for that. So get really pissed --pissed enough to actually consider what you are reading here from that Potatohead guy. Then go act on it.

Thank me in 10 years, just like I did that insurance salesman, who flipped my world upside down and in one conversation set me straight on how things really go in this world.

Peace.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness! (http://propeller.wikispaces.com/)

Post Edited (potatohead) : 9/12/2007 6:15:21 AM GMT

Oldbitcollector (Jeff)
09-12-2007, 08:40 PM
I'd suggest that you contact Jeri Ellsworth (No I won't give out her email address) but she isn't that hard to locate. en.wikipedia.org/wiki/Jeri_Ellsworth (http://en.wikipedia.org/wiki/Jeri_Ellsworth) She has some experience in creating a retro gaming platform having designed the Commodore DTV joystick. I can assure you that you have an uphill route to go if you go this direction.

Yes, it can be done, but it isn't easy, and certainly nothing you will get to market by this Christmas.

Oldbitcollector

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Buttons . . . check. Dials . . . check. Switches . . . check. Little colored lights . . . check.

— Calvin, of 'Cavin and Hobbes.

jimmysmith
01-01-2013, 05:21 PM
The copyrights to the games will continue to be renewed by Nintendo very year.Copy rights are much important in these days due more legal issues about incorporates.I will try to answer these questions latter when i will have some spare time.
Does anyone know how to manufacture a these componets?
2 Where do we buy·the components·to assemble and resell?
3 Are·nintendo video games still copyrighted?
4 Who's interesed in putting together a game from scratch?
5 How do we connect old Nintendo games to the Hydra?
6 How would be go about getting the copyrights on a game?
7 Can we sell them like we saw in the mall small game packages of 30-100 games in each package at Christmas time?
8 Finally, I'd like to know about the engineering books available for the manufacture of microchips.

cgracey
01-01-2013, 09:36 PM
8 Finally, I'd like to know about the engineering books available for the manufacture of microchips.

This book taught me everything I needed to know to design chips:

98336

This other book was very good about teaching concentric layout practice, which is needed to get analog accuracy:

98337

All you need to practice with is a SPICE program with waveform viewer and some SPICE models for the technology you are working in. It helps to have a nice schematic editor to generate the SPICE files from, too. We use this package:

http://www.tannereda.com/t-spice-pro

Since those books were written, much of the effort in designing chips has moved away from the bottom-level details and into using proven cell sets for constructing huge logic via synthesis from Verilog and VHDL. If you want your chip to do anything special with its I/O pins, you can still benefit by designing those circuits from scratch. The modern design flow does not even favor looking at the actual layout, anymore, only the general floorplan.

idbruce
01-02-2013, 01:12 AM
Thanks for the info Chip.

I imagine those books would be very interesting, providing I could possibly understand a fraction of the content.

cgracey
01-02-2013, 02:13 AM
Thanks for the info Chip.

I imagine those books would be very interesting, providing I could possibly understand a fraction of the content.

Trust me. If you read them with sincere interest, you would learn just like I did. It's a fascinating subject, anyway, so it's not hard to pick up. And, there's really just a finite base of things you need to think about, so it's not like you can't achieve some degree of mastery on your own in a few short years. I'm sure it helps if you have some goal in mind, though, like I did.

Sometimes I think about how a LOT of people are engaged in software development and, as a result, there is a lot of good software written, somewhat by the sheer force of the number of people involved. Algorithms are commonly shared and many people have a good knowledge of best practices. As you move up from software through hardware, and into chip design, there are far fewer participants and the air is a lot thinner. It's not that it's any more difficult - to the contrary, it's a more finite discipline than software, so you can do a more exact job of getting things clean and perfected. And there are still good resources available for learning, especially in the form of university class power-point presentations. But, because there are fewer people involved in chip design, you are not likely to have more than a few contacts a year with other chip designers. At least, that's been my experience. I'm sure, though, that if chip design were more accessible and of interest to more people, chip design know-how could become just as common as software know-how is now.

KeithE
01-02-2013, 03:23 AM
Trust me. If you read them with sincere interest, you would learn just like I did. It's a fascinating subject, anyway, so it's not hard to pick up. And, there's really just a finite base of things you need to think about, so it's not like you can't achieve some degree of mastery on your own in a few short years. I'm sure it helps if you have some goal in mind, though, like I did.

Sometimes I think about how a LOT of people are engaged in software development and, as a result, there is a lot of good software written, somewhat by the sheer force of the number of people involved. Algorithms are commonly shared and many people have a good knowledge of best practices. As you move up from software through hardware, and into chip design, there are far fewer participants and the air is a lot thinner. It's not that it's any more difficult - to the contrary, it's a more finite discipline than software, so you can do a more exact job of getting things clean and perfected. And there are still good resources available for learning, especially in the form of university class power-point presentations. But, because there are fewer people involved in chip design, you are not likely to have more than a few contacts a year with other chip designers. At least, that's been my experience. I'm sure, though, that if chip design were more accessible and of interest to more people, chip design know-how could become just as common as software know-how is now.

Chip it is frustrating. I think that one fundamental problem is tools - e.g. try to get a decent free Verilog simulator. You can probably point to a couple that limp along, but they are so inferior to the "professional" tools. Then try to get a simulator that can run both VHDL and Verilog, or one that can run SystemVerilog. And then try to find all of the other tools that you would need just for basic digital design. (e.g. linters/cdc checkers/isolation checkers, synthesizers, place and route, formal equivalence checking, test insertion and pattern generation, static timing analysis, power simulation, formal property verification, package design,...) As far as I know the EDA companies don't want anyone providing easy access to these - e.g. can't rent them out by the hour in the cloud. There's also the high cost if you actually want to get a chip built, and then just getting access to the leading edge processes can be difficult.

As far as the ratio of hardware to software engineers - I work at a somewhat large semiconductor company and I heard that 80% of the engineers there are software engineers. In my group the majority of these software engineers were actually EE majors since it seems to be one of the best ways to get a strong signal processing background.

cgracey
01-02-2013, 04:24 AM
Chip it is frustrating. I think that one fundamental problem is tools - e.g. try to get a decent free Verilog simulator. You can probably point to a couple that limp along, but they are so inferior to the "professional" tools. Then try to get a simulator that can run both VHDL and Verilog, or one that can run SystemVerilog. And then try to find all of the other tools that you would need just for basic digital design. (e.g. linters/cdc checkers/isolation checkers, synthesizers, place and route, formal equivalence checking, test insertion and pattern generation, static timing analysis, power simulation, formal property verification, package design,...) As far as I know the EDA companies don't want anyone providing easy access to these - e.g. can't rent them out by the hour in the cloud. There's also the high cost if you actually want to get a chip built, and then just getting access to the leading edge processes can be difficult.

As far as the ratio of hardware to software engineers - I work at a somewhat large semiconductor company and I heard that 80% of the engineers there are software engineers. In my group the majority of these software engineers were actually EE majors since it seems to be one of the best ways to get a strong signal processing background.

But you don't need to bother with most of those things in your first paragraph. They are implementation details that you aren't even concerned about in your Verilog code. And the way you test your Verilog code is by running it on FPGA's. Only by running it at (near) real-time speed can you get a 'feel' for how your design works. Why bother simulating it when you can actually run it? All that synthesis, test insertion, layout, power checking, etc. can be done for a fee by people who have access to those expensive tools and know how to use them; it's just a one-time process and occurs only after your Verilog work is done. If you're wanting to design complex logic, you are kind of stuck with that process. If you are working in .35um technology and have lower-complexity digital and custom analog, you can fully design everything yourself from scratch, using only SPICE simulation, while using Verilog and FPGA's to prove the logic concept.

KeithE
01-02-2013, 05:08 AM
Chip - your code might be small compared to what we work with, so I guess this depends on what one is planning to build. I guess that you're thinking about some small scale design work for educational purposes, versus a typical Silicon Valley chip design. For the latter you're going to want to learn about all of these tools eventually though.

Edited to say - it might be interesting if you could give a $$$ estimate of what it would take for someone starting with nothing to get a chip built and somewhat tested. Or another take would be how much money someone would need to ask for if they wanted to startup a fabless semiconductor company.

This may all seem a bit random but some things from my perspective... When it takes ~1 day or more to get your code into an FPGA, then in my view high-quality simulation is required. Also with an FPGA only it's not clear how you are getting coverage metrics. And why spend time finding bugs using dynamic sims that could be found with static checks provided by some of the tools that I mentioned? (e.g. hooked up the wrong clock to a synchronizer, or forgot an isolation gate which isn't present in the FPGA.) Also it's nice to have an automatic regression system - especially when you're working with a team of designers. There are even special systems you can buy for accelerating sims of synthesizable code - e.g. Palladium that don't use FPGAs. The main advantage being the time to build. Having sims with dump files it potentially saves some work having to instrument your code - what I mean is wiring things out to pins or to something like ChipScope. We definitely could not do without FPGAs and of course we use them for verification and software co-development. The FPGAs don't emulate everything though - e.g. clock gating support is quite primitive. Also I disagree that you can avoid thinking about implementation details until the design work is done, but we use a different flow than you. Keep in mind that we work in 40 nm and below, and have to come out with new generation of chips every year or two. And we have customers that are buying tens of millions of chips who grill us on the whole process. This drives us to work in certain ways - e.g. we always put in test logic. You cannot tapeout a chip without going through a huge checklist that inquires about such things. (Is this reminding you of working with lawyers? It does suck a bit of joy out of life ;-)

Also I have to say that gate level simulations are an absolute must. If you asked me 10 years ago I wouldn't have believed it due to formal equivalence checking and static timing analysis, but some of the things done for testability and power savings can really cause problems and I've seen that gate sims are a good way to catch these. You may not run into these issues.

idbruce
01-02-2013, 06:37 PM
@KeithE

From what I have read and as indicated by your own words, it sounds to me like you are frustrated with your area of expertise.

According to my interpretation, Chip provided information in effort to be helpful to those that might be interested in learning, as I am. Considering that you specifically quoted his response to me, I think you pulled his response way out of context. In other words, he was responding to:


I imagine those books would be very interesting, providing I could possibly understand a fraction of the content.

Firstly, I never said that I wanted to do this professionally, I just said that I imagined that the books would be an interesting read, and secondly, I don't believe that Chip's response was meant to encourage me to attempt the production of an IC, instead I think his response was meant to encourage learning.

I read about and dabble in quite a few different subjects that interest me, but that doesn't mean that I want to become an engineer in those specific fields. And pertaining to the money and expense side of the topic, there is an old saying:


It takes money to make money.

Author Unknown

However, I must say that I found your response interesting and informative, I just think you should have provided your point of view in a different manner, than pulling his response to me out of context.

Bruce

KeithE
01-02-2013, 06:59 PM
@Bruce - I see what you're saying. If you want to learn some basics about layout but not actually fab chips there are some games that you might find interesting. You may find some better examples than this with a little searching but here is one - http://www.zachtronicsindustries.com/play-kohctpyktop/

If you're more interested in digital logic design at a higher level like what Chip is doing in Altera FPGAs, then this is quite doable on a hobby budget. There's a book "The Elements of Computing Systems: Building a Modern Computer from First Principles" that you could use in conjunction with something like a DE0 nano. So if you buy one to try out the Prop2, you could also use it to make your own processor.

ctwardell
01-02-2013, 07:06 PM
There's a book "The Elements of Computing Systems: Building a Modern Computer from First Principles" that you could use in conjunction with something like a DE0 nano. So if you buy one to try out the Prop2, you could also use it to make your own processor.

Thanks for the book tip Keith, now my nano may have something to do when the P2 hits the street.

C.W.

idbruce
01-02-2013, 07:30 PM
@KiethE

Thanks for the additional information.

Bruce

cgracey
01-02-2013, 07:41 PM
Edited to say - it might be interesting if you could give a $$$ estimate of what it would take for someone starting with nothing to get a chip built and somewhat tested. Or another take would be how much money someone would need to ask for if they wanted to startup a fabless semiconductor company.

Well, a decent FPGA setup today costs only $600. I think the Tanner EDA IC design tool set that we use costs about $25k. I remember it costing $5k back in the 1990's. A shuttle run for 0.35um CMOS, which is a nice analog/digital process, with low leakage and practical 3.3V throughout, runs about $15k from AustriaMicrosystems. There are maybe 300 design rules to follow for that process. A final mask set costs about $60k. The 8" wafers run about $900/each. That's pretty much how we made the Propeller chip that we've been selling since 2006.

A leading-edge 40nm process is quite a bit more expensive, as you know. I believe a shuttle run costs $150k and the mask set is ~$3M. This is probably not something that you want to design custom silicon for, as there are 3k+ design rules to follow and you're much safer using proven cells available from the foundry. The design work is done at a much higher level than you would engage in at 0.35um. You'll need tools that cost over $1M and people that know how to operate them. Much of what they'll do is independent of your own design efforts in Verilog. I've heard it said that you need to be able to sell at least $10M of such a chip to make the investment worthwhile.

idbruce
01-02-2013, 07:58 PM
Sounds pretty expensive. Not the average do-it-yourselfer kind of stuff.

http://www.backyardsemiconductor.com

:)