Author Topic: Analog Devices in talks to buy Maxim  (Read 7071 times)

0 Members and 1 Guest are viewing this topic.

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Analog Devices in talks to buy Maxim
« Reply #25 on: July 13, 2020, 08:31:38 pm »
https://www.reuters.com/article/us-maxim-intg-m-a-analog-devices/chipmaker-analog-devices-to-buy-rival-maxim-for-about-21-billion-idUSKCN24E14B

21 billion? Finally I can consider Maxim again after mothballing them for over 3 decades. I have been lobbying TI, LT and ADI to buy the f***king turd for decades.

Maxim royally pissed me off in 1991 and I swore I would never consider them ever again,.... until now :clap:

Bob

To be fair, Maxim has never been a company to pay attention to the smaller companies, makers, or hobbyists. Maxim focuses on their strategic partners like Apple, Samsung, Amazon, etc. So the undercurrent of "Friends don't let friends by Maxim" has never been a concern for the company because there has never been a desire to play in the mass market.
Contrast this to TI, especially over the last 15 years, has had a strategy of "growing the tail" and going after the mass market and doing a pretty damn good job at it.

Correct. I agree. Maxim never gave a s**t about my business and never gave them a second look even though their parts were ideal in my designs. So for 30 plus years they never saw the light of day in my world and when I explained the rationale to my employer and my clients....Most were totally okay with my decisions simply because they themselves had been burned by Maxim’s selfish policies or as they called it “ “strategy” or market “focus”

So now, because of market pressure ADI are acquiring Maxim.  I applaud  that because now I can consider them as a complement to the other global players, since many of their parts are an outstanding value to me now clearly without the maxim bullsh*t factor.....like why are the MOQ and deliveries ridiculous, etc

Agreed.. Ideally, ADI buys them up and takes their portfolio as their own including ADI/LT generous policies of very rarely obsoleting a device. On the other hand, if ADI adopts Maxims obsolescence policy - the world will end.  :(

Aren't modern manufacturing processes flexible enough that it is pretty easy to bake a new batch of a chip?
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8550
  • Country: us
    • SiliconValleyGarage
Re: Analog Devices in talks to buy Maxim
« Reply #26 on: July 13, 2020, 08:48:30 pm »
https://www.reuters.com/article/us-maxim-intg-m-a-analog-devices/chipmaker-analog-devices-to-buy-rival-maxim-for-about-21-billion-idUSKCN24E14B

21 billion? Finally I can consider Maxim again after mothballing them for over 3 decades. I have been lobbying TI, LT and ADI to buy the f***king turd for decades.

Maxim royally pissed me off in 1991 and I swore I would never consider them ever again,.... until now :clap:

Bob

To be fair, Maxim has never been a company to pay attention to the smaller companies, makers, or hobbyists. Maxim focuses on their strategic partners like Apple, Samsung, Amazon, etc. So the undercurrent of "Friends don't let friends by Maxim" has never been a concern for the company because there has never been a desire to play in the mass market.
Contrast this to TI, especially over the last 15 years, has had a strategy of "growing the tail" and going after the mass market and doing a pretty damn good job at it.

Correct. I agree. Maxim never gave a s**t about my business and never gave them a second look even though their parts were ideal in my designs. So for 30 plus years they never saw the light of day in my world and when I explained the rationale to my employer and my clients....Most were totally okay with my decisions simply because they themselves had been burned by Maxim’s selfish policies or as they called it “ “strategy” or market “focus”

So now, because of market pressure ADI are acquiring Maxim.  I applaud  that because now I can consider them as a complement to the other global players, since many of their parts are an outstanding value to me now clearly without the maxim bullsh*t factor.....like why are the MOQ and deliveries ridiculous, etc

Agreed.. Ideally, ADI buys them up and takes their portfolio as their own including ADI/LT generous policies of very rarely obsoleting a device. On the other hand, if ADI adopts Maxims obsolescence policy - the world will end.  :(

Aren't modern manufacturing processes flexible enough that it is pretty easy to bake a new batch of a chip?
nope. on the contrary ...
modern processes are so finely tuned and the fabs so expensive you don't want to run old gunk in them

- for a given factory/process the cost per square millimeter is a quasi constant.
- the smaller the geometries you can run the higher the cost becomes.

so running something that was designed for 1 micrometer in a 8 nanometer fab is horrifically expensive ! becasue you are using equipment that con do precise work for course work. Tt's like buying a set of micrometers to mark a line in chalk where you are going to chop the tree with an axe ...


Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 
The following users thanked this post: SilverSolder

Offline grouchobyte

  • Regular Contributor
  • *
  • Posts: 244
  • Country: cn
Re: Analog Devices in talks to buy Maxim
« Reply #27 on: July 13, 2020, 09:06:42 pm »
In the late 1990’s i worked as a design engineer at Maxtek ( a  Tektronix and Maxim JV)
Even then.....the friction between these two tech companies was contentious eventually ending in a separation in which I was absorbed into Tektronix. Jack Gifford was the Maxim CEO back then. He micromanaged everything including stationary supplies

You can well imagine how bad the corporate culture was back then. That said, Maxim treated its small customers like crap unless they were big like apple, etc. To Maxim “size” does matter. Always did.

Hopefully ADI can integrate the products into the fold and train the employees to adhere to the ADI rules of the road. ADI did a good job with LT so my confidence is high. Their products will have a home no matter what, but those upper management assholes need to go.... imho

« Last Edit: July 13, 2020, 09:17:27 pm by grouchobyte »
 

Offline Mr. Scram

  • Super Contributor
  • ***
  • Posts: 9821
  • Country: 00
  • Display aficionado
Re: Analog Devices in talks to buy Maxim
« Reply #28 on: July 13, 2020, 09:18:37 pm »
I don't think the endless mergers are good for anyone save a few investors perhaps. However, in the case of Maxim there seems ample room for improvement. They have plenty of interesting parts on paper but little of it is more than vapour ware for most people.
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Analog Devices in talks to buy Maxim
« Reply #29 on: July 13, 2020, 09:41:35 pm »
https://www.reuters.com/article/us-maxim-intg-m-a-analog-devices/chipmaker-analog-devices-to-buy-rival-maxim-for-about-21-billion-idUSKCN24E14B

21 billion? Finally I can consider Maxim again after mothballing them for over 3 decades. I have been lobbying TI, LT and ADI to buy the f***king turd for decades.

Maxim royally pissed me off in 1991 and I swore I would never consider them ever again,.... until now :clap:

Bob

To be fair, Maxim has never been a company to pay attention to the smaller companies, makers, or hobbyists. Maxim focuses on their strategic partners like Apple, Samsung, Amazon, etc. So the undercurrent of "Friends don't let friends by Maxim" has never been a concern for the company because there has never been a desire to play in the mass market.
Contrast this to TI, especially over the last 15 years, has had a strategy of "growing the tail" and going after the mass market and doing a pretty damn good job at it.

Correct. I agree. Maxim never gave a s**t about my business and never gave them a second look even though their parts were ideal in my designs. So for 30 plus years they never saw the light of day in my world and when I explained the rationale to my employer and my clients....Most were totally okay with my decisions simply because they themselves had been burned by Maxim’s selfish policies or as they called it “ “strategy” or market “focus”

So now, because of market pressure ADI are acquiring Maxim.  I applaud  that because now I can consider them as a complement to the other global players, since many of their parts are an outstanding value to me now clearly without the maxim bullsh*t factor.....like why are the MOQ and deliveries ridiculous, etc

Agreed.. Ideally, ADI buys them up and takes their portfolio as their own including ADI/LT generous policies of very rarely obsoleting a device. On the other hand, if ADI adopts Maxims obsolescence policy - the world will end.  :(

Aren't modern manufacturing processes flexible enough that it is pretty easy to bake a new batch of a chip?
nope. on the contrary ...
modern processes are so finely tuned and the fabs so expensive you don't want to run old gunk in them

- for a given factory/process the cost per square millimeter is a quasi constant.
- the smaller the geometries you can run the higher the cost becomes.

so running something that was designed for 1 micrometer in a 8 nanometer fab is horrifically expensive ! becasue you are using equipment that con do precise work for course work. Tt's like buying a set of micrometers to mark a line in chalk where you are going to chop the tree with an axe ...

I'm not sure I understand the reasoning.  For example:  say I have an old 300dpi document that I used to print on my 300dpi laser printer in the past.  Then today, I choose to print it on my brand new 2400dpi laser printer...  the document will look the same (still 300dpi resolution), and I will not have incurred greater cost for the fact of the new printer being capable of much higher precision - paper is still paper, toner is still toner, etc.? 

This is true even if the new printer is much more expensive than the old one.  I guess if you write off part of the cost of the printer on each printout, you could get uneconomical results compared to the old, written down printer, but that seems more of an accounting technicality than a real cost?
« Last Edit: July 13, 2020, 09:44:58 pm by SilverSolder »
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
Re: Analog Devices in talks to buy Maxim
« Reply #30 on: July 13, 2020, 10:59:45 pm »
The problem is that the costs of the machines is too high to produce chips with large geometries. And it wouldn't surprise me if it turns out a 10nm fab can't do 100nm chips for some reason but free_electron knows this kind of details for sure.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online coppice

  • Super Contributor
  • ***
  • Posts: 10031
  • Country: gb
Re: Analog Devices in talks to buy Maxim
« Reply #31 on: July 13, 2020, 11:23:04 pm »
I'm not sure I understand the reasoning.  For example:  say I have an old 300dpi document that I used to print on my 300dpi laser printer in the past.  Then today, I choose to print it on my brand new 2400dpi laser printer...  the document will look the same (still 300dpi resolution), and I will not have incurred greater cost for the fact of the new printer being capable of much higher precision - paper is still paper, toner is still toner, etc.? 

This is true even if the new printer is much more expensive than the old one.  I guess if you write off part of the cost of the printer on each printout, you could get uneconomical results compared to the old, written down printer, but that seems more of an accounting technicality than a real cost?
Printing is just one part of the work of a fab. All the prcessing steps in a fab are tailored to just one or two tightly defined processes. When you design a chip it is targeted at that small number of fabs which run the relevant process. If you design a chip that is a huge hit you can often be massively constrained by the inability to add production in more fabs. Many of the end of life notices you will see, especially for long life analogue parts, relate to the fab or fabs getting so old spares are no longer available, and the parts are impractical to move to newer fabs without major re-engineering to a new process.
 
The following users thanked this post: SilverSolder

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Analog Devices in talks to buy Maxim
« Reply #33 on: July 14, 2020, 12:55:00 pm »
I still remember when Analog Devices bought PMI (Precision Monolithics Incorporated).

Fabrication processes are also dependent on suppliers and this is a major problem for old linear processes.
« Last Edit: July 14, 2020, 12:57:47 pm by David Hess »
 

Offline peter-h

  • Super Contributor
  • ***
  • Posts: 4355
  • Country: gb
  • Doing electronics since the 1960s...
Re: Analog Devices in talks to buy Maxim
« Reply #34 on: July 14, 2020, 01:11:29 pm »
The problem is that Maxim trashed and devalued their own business, through utterly inept management.

Here in the UK, they "serve" from an office in Ireland. Nothing wrong with that (corporation tax is lower there) but a year or two ago they stopped answering the phone or emails.

Yes you read that right. As a customer, you cannot phone Maxim and cannot email Maxim.

You can communicate with this owner of the universe only by support tickets on their website. No discussion possible of anything. They reply to the ticket. Even an RFQ for 10k cannot be communicated on.

Over many years I designed in loads of MAX489 MAX3089 MAX232 MAX3232 etc chips but now look at TI every time. TI are also generally cheaper, and of course any avoidance of special parts is a great idea and always was. Maxim is so difficult to deal with that we buy their parts from distis but all new products don't use them.
Z80 Z180 Z280 Z8 S8 8031 8051 H8/300 H8/500 80x86 90S1200 32F417
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Analog Devices in talks to buy Maxim
« Reply #35 on: July 14, 2020, 01:17:49 pm »
I still remember when Analog Devices bought PMI (Precision Monolithics Incorporated).

Fabrication processes are also dependent on suppliers and this is a major problem for old linear processes.


I'm really surprised that older designs can't be "emulated" on newer processes.  -  Perhaps they can...   but not without quite a bit of engineering effort that perhaps approaches a redesign - maybe like an ASIC or something?

 

Online coppice

  • Super Contributor
  • ***
  • Posts: 10031
  • Country: gb
Re: Analog Devices in talks to buy Maxim
« Reply #36 on: July 14, 2020, 01:33:49 pm »
I still remember when Analog Devices bought PMI (Precision Monolithics Incorporated).

Fabrication processes are also dependent on suppliers and this is a major problem for old linear processes.


I'm really surprised that older designs can't be "emulated" on newer processes.  -  Perhaps they can...   but not without quite a bit of engineering effort that perhaps approaches a redesign - maybe like an ASIC or something?
I'm puzzled why you would expect to be able to emulate older processes in newer fabs. They don't just keep shrinking the geometries. The chemicals used, the etching techniques, everything changes over time. You probably couldn't even find the database needed to generate a new mask set that would fit in the modern equipment.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Analog Devices in talks to buy Maxim
« Reply #37 on: July 14, 2020, 02:11:57 pm »
I still remember when Analog Devices bought PMI (Precision Monolithics Incorporated).

Fabrication processes are also dependent on suppliers and this is a major problem for old linear processes.

I'm really surprised that older designs can't be "emulated" on newer processes.  -  Perhaps they can...   but not without quite a bit of engineering effort that perhaps approaches a redesign - maybe like an ASIC or something?

Those designs are tailored to the specifics of the fabrication process.

So for instance the old NPN only processes included slow and low gain lateral PNPs made with NPN process steps but those PNPs cannot be replaced with the PNPs on a more modern complementary bipolar process if the old design relied on the 30+ volt base-emitter breakdown voltage of those old PNPs, like almost every operational amplifier with a wide differential input voltage range, including the LM324, 741, and 301A.  The "modern" replacements for those parts which are built on a "superior" complementary bipolar process are limited to about 7 volts.

If you look through the Linear Technology parts portfolio, there are cases where an older part was redesigned but they used a modified part number for the new part, so it does happen but usually the old part remains available.

 
The following users thanked this post: thm_w, SilverSolder, newbrain, duckduck

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Analog Devices in talks to buy Maxim
« Reply #38 on: July 14, 2020, 03:53:28 pm »
[...]
I'm puzzled why you would expect to be able to emulate older processes in newer fabs.
They don't just keep shrinking the geometries. The chemicals used, the etching techniques, everything changes over time. You probably couldn't even find the database needed to generate a new mask set that would fit in the modern equipment.



Too much exposure to software, I guess!  :D   

To a software engineer,  using a virtual machine to emulate an older architecture is a commonly used approach, and gives "near perfect" albeit not 100% exactly the same result as the original designs - but "good enough for Australia", even though the new technology it runs on has absolutely nothing to do with what was available back when the original was designed.

It doesn't seem beyond imagination that an older part could be emulated with a newer design, one that might have a completely different schematic diagram that suits the newer processes, yet ends up behaving like the older part at least on a "good enough for Australia" basis?


« Last Edit: July 14, 2020, 03:55:38 pm by SilverSolder »
 

Online coppice

  • Super Contributor
  • ***
  • Posts: 10031
  • Country: gb
Re: Analog Devices in talks to buy Maxim
« Reply #39 on: July 14, 2020, 04:04:03 pm »
[...]
I'm puzzled why you would expect to be able to emulate older processes in newer fabs.
They don't just keep shrinking the geometries. The chemicals used, the etching techniques, everything changes over time. You probably couldn't even find the database needed to generate a new mask set that would fit in the modern equipment.

Too much exposure to software, I guess!  :D   

To a software engineer,  using a virtual machine to emulate an older architecture is a commonly used approach, and gives "near perfect" albeit not 100% exactly the same result as the original designs - but "good enough for Australia", even though the new technology it runs on has absolutely nothing to do with what was available back when the original was designed.

It doesn't seem beyond imagination that an older part could be emulated with a newer design, one that might have a completely different schematic diagram that suits the newer processes, yet ends up behaving like the older part at least on a "good enough for Australia" basis?
If you want a new part which broadly emulates an old part, that kind of thing is being developed all the time. The original topic was sustaining production of a guaranteed compatible drop in part. A new part will generally require a whole new design around it. For example, a modern part will generally require a lower supply voltage, and have different voltage ranges on its I/O pins, etc. A broad range of more subtle differences, like tolerance to EMI, will also need to be allowed for in a new design using the new part.
 
The following users thanked this post: SilverSolder

Offline Sal Ammoniac

  • Super Contributor
  • ***
  • Posts: 1764
  • Country: us
Re: Analog Devices in talks to buy Maxim
« Reply #40 on: July 14, 2020, 04:15:22 pm »
Another one gets assimilated... damn, this is not going to end well.

Not quite yet. The merger still needs government approval, so there's still a chance it won't happen, but I wouldn't count on that given the current administration.
"That's not even wrong" -- Wolfgang Pauli
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17427
  • Country: us
  • DavidH
Re: Analog Devices in talks to buy Maxim
« Reply #41 on: July 15, 2020, 11:49:46 pm »
Another one gets assimilated... damn, this is not going to end well.

Not quite yet. The merger still needs government approval, so there's still a chance it won't happen, but I wouldn't count on that given the current administration.

It is not like there is a comparably sized third competitor to Analog Devices and Texas Instrument anyway.
 

Offline c64

  • Frequent Contributor
  • **
  • Posts: 311
  • Country: au
Re: Analog Devices in talks to buy Maxim
« Reply #42 on: July 16, 2020, 04:40:00 am »
The problem is that Maxim trashed and devalued their own business, through utterly inept management.

Here in the UK, they "serve" from an office in Ireland. Nothing wrong with that (corporation tax is lower there) but a year or two ago they stopped answering the phone or emails.

Yes you read that right. As a customer, you cannot phone Maxim and cannot email Maxim.

You can communicate with this owner of the universe only by support tickets on their website. No discussion possible of anything. They reply to the ticket. Even an RFQ for 10k cannot be communicated on.

Maybe that's why Analog want to buy them. Here in AU you cannot phone Analog and cannot email Analog. Well, you can e-mail, but don't expect any answer (at least I've never got any)
 

Online coppice

  • Super Contributor
  • ***
  • Posts: 10031
  • Country: gb
Re: Analog Devices in talks to buy Maxim
« Reply #43 on: July 16, 2020, 11:08:07 am »
Maybe that's why Analog want to buy them. Here in AU you cannot phone Analog and cannot email Analog. Well, you can e-mail, but don't expect any answer (at least I've never got any)
Its easy for people in many countries to talk to someone from Analog. Maybe you need to move to a country where the electronics industry is still strong enough to keep local representation from the major vendors. Most organisations are very much organised on country by country lines. If you have to call a representative in another country they score no brownie points for helping you, so they probably won't.
 

Offline DrG

  • Super Contributor
  • ***
  • !
  • Posts: 1199
  • Country: us
Re: Analog Devices in talks to buy Maxim
« Reply #44 on: September 24, 2020, 05:35:20 pm »
So many years ago, back in the day so-to-speak, I bought a few shares of Dallas Semiconductor, and ended up owning fewer shares of Maxim. If this goes through, I guess I will end up owning still fewer shares of AD.

The other day, I received several hundred pages of "Merger Proposal".  I read the cover to learn that in the proposal, it is 0.63 shares AD / share of Maxim. Apparently, my goal is to own only one share of the biggest, baddest and only semiconductor in existence.  :)
- Invest in science - it pays big dividends. -
 
The following users thanked this post: SeanB

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Analog Devices in talks to buy Maxim
« Reply #45 on: September 24, 2020, 05:54:00 pm »
So many years ago, back in the day so-to-speak, I bought a few shares of Dallas Semiconductor, and ended up owning fewer shares of Maxim. If this goes through, I guess I will end up owning still fewer shares of AD.

The other day, I received several hundred pages of "Merger Proposal".  I read the cover to learn that in the proposal, it is 0.63 shares AD / share of Maxim. Apparently, my goal is to own only one share of the biggest, baddest and only semiconductor in existence.  :)

It won't stop until there is just one company left - "The Firm" - that does absolutely everything!  :D
 

Online mawyatt

  • Super Contributor
  • ***
  • Posts: 4117
  • Country: us
Re: Analog Devices in talks to buy Maxim
« Reply #46 on: September 24, 2020, 06:21:57 pm »

It won't stop until there is just one company left - "The Firm" - that does absolutely everything!  :D

And that will be the end of innovation :-\

Best,
Curiosity killed the cat, also depleted my wallet!
~Wyatt Labs by Mike~
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Analog Devices in talks to buy Maxim
« Reply #47 on: September 24, 2020, 06:26:39 pm »
Too much exposure to software, I guess!  :D   

To a software engineer,  using a virtual machine to emulate an older architecture is a commonly used approach, and gives "near perfect" albeit not 100% exactly the same result as the original designs - but "good enough for Australia", even though the new technology it runs on has absolutely nothing to do with what was available back when the original was designed.

It doesn't seem beyond imagination that an older part could be emulated with a newer design, one that might have a completely different schematic diagram that suits the newer processes, yet ends up behaving like the older part at least on a "good enough for Australia" basis?

Well, you can, sort of. FPGAs and related parts can be made to emulate the function of many digital ICs, however it requires considerable effort and even then the result is often not compatible as a drop in replacement in an existing system. A new IC could be designed to replace an old IC but it's going to be an entirely new part requiring a lot of effort to design and test for what may be a small market. Perhaps a comparison that is more easily visualized, try to design a solid state replacement for a vacuum tube, in most cases it's a lot harder than it sounds. Sure you can design an entire new circuit that does what the original tube based circuit did, but it's probably going to have entirely different supply voltage and thermal requirements and it may not behave *exactly* the same under all circumstances. Designing a solid state module that plugs into a tube socket and works with all the original circuits that were made to use that tube is in most cases a difficult or impossible task. The tube might be a lot lower tech than a modern transistor but that doesn't mean that a transistor fab can crank out vacuum tubes. 
 

Online mawyatt

  • Super Contributor
  • ***
  • Posts: 4117
  • Country: us
Re: Analog Devices in talks to buy Maxim
« Reply #48 on: September 24, 2020, 07:21:52 pm »
Perhaps a comparison that is more easily visualized, try to design a solid state replacement for a vacuum tube, in most cases it's a lot harder than it sounds. Sure you can design an entire new circuit that does what the original tube based circuit did, but it's probably going to have entirely different supply voltage and thermal requirements and it may not behave *exactly* the same under all circumstances. Designing a solid state module that plugs into a tube socket and works with all the original circuits that were made to use that tube is in most cases a difficult or impossible task. The tube might be a lot lower tech than a modern transistor but that doesn't mean that a transistor fab can crank out vacuum tubes.

Some time ago I recall some folks working on a "Solid State Tube" where a vacuum tube was integrated on a silicon die. Think these were tiny triodes, with a proper cathode, plate and even a suspended filament in a sub-surface sealed vacuum. This may have been related to vacuum tubes having extremely good nuclear radiation tolerance.

Best,
Curiosity killed the cat, also depleted my wallet!
~Wyatt Labs by Mike~
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Analog Devices in talks to buy Maxim
« Reply #49 on: September 24, 2020, 09:51:18 pm »
Too much exposure to software, I guess!  :D   

To a software engineer,  using a virtual machine to emulate an older architecture is a commonly used approach, and gives "near perfect" albeit not 100% exactly the same result as the original designs - but "good enough for Australia", even though the new technology it runs on has absolutely nothing to do with what was available back when the original was designed.

It doesn't seem beyond imagination that an older part could be emulated with a newer design, one that might have a completely different schematic diagram that suits the newer processes, yet ends up behaving like the older part at least on a "good enough for Australia" basis?

Well, you can, sort of. FPGAs and related parts can be made to emulate the function of many digital ICs, however it requires considerable effort and even then the result is often not compatible as a drop in replacement in an existing system. A new IC could be designed to replace an old IC but it's going to be an entirely new part requiring a lot of effort to design and test for what may be a small market. Perhaps a comparison that is more easily visualized, try to design a solid state replacement for a vacuum tube, in most cases it's a lot harder than it sounds. Sure you can design an entire new circuit that does what the original tube based circuit did, but it's probably going to have entirely different supply voltage and thermal requirements and it may not behave *exactly* the same under all circumstances. Designing a solid state module that plugs into a tube socket and works with all the original circuits that were made to use that tube is in most cases a difficult or impossible task. The tube might be a lot lower tech than a modern transistor but that doesn't mean that a transistor fab can crank out vacuum tubes.

Back in the day, you could actually buy transistorized "tube substitutes", at least for some tubes, for TV repair purposes.  These devices were probably optimized for that particular role, so likely couldn't work in every conceivable circuit that the original tube would have worked with -  but good enough to fix the TV, and a lot longer lasting!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf