95 % efficiency SMPS

[dougy83]

it happens if more people knew how simple programmers were for the chips these days they would not be so reliant on opportunistic jerks reaching for wallets.

have a look on the microchip forums for people having problems using home built programmers with crappy free software (e.g. ic-prog, ponyprog, etc.).

 

[Sceadwian]

dougy, I bet 99.9% of the time it's user error. AVR or PIC serial programmers there are plenty of people that have no clue what they're doing.

 

[Ubergeek63]

have a look on the microchip forums for people having problems using home built programmers with crappy free software (e.g. ic-prog, ponyprog, etc.).

depends on who's crappy parts you are using. 8051's are crappy but flash magic does a wonderful job very easilly programming them. Numerous NXP ARM parts program with it just as easilly - but they are NOT crappy parts.

Point is, marketting twits want you to think everything is hard to do so they can sell you crap. Problem is you generally only find out it IS crap after your money is gone.

Dan

 

[Thunderchild]

looks like I need to learn to program and I need to start from the very begining, any no nonesense books anyone can recomend ? I've been burnt twice already

 

[Sceadwian]

For the basics don't buy books, learn online. There is more information online than is contained in many common books, and you don't have to pay for them. Or better yet, use a library like I do, then you don't have to pay for anything =)

 

[dougy83]

depends on who's crappy parts you are using. 8051's are crappy

What? All 8051's are crappy?? So ATMEL, AD, Dallas, infineon, silicon labs, STMicro, TI are all making crappy parts? Maybe you should inform them. I would assume that in most of the cases of parts not programming properly, it is the fault of the programming device or the programming software; the micro would not be to blame.

dougy, I bet 99.9% of the time it's user error. AVR or PIC serial programmers there are plenty of people that have no clue what they're doing.

Well, either I'm in the top 0.1% (I know I'm good), or I don't know my hand from my foot (also possible).

 

[Ubergeek63]

What? All 8051's are crappy?? So ATMEL, AD, Dallas, infineon, silicon labs, STMicro, TI are all making crappy parts? Maybe you should inform them. I would assume that in most of the cases of parts not programming properly, it is the fault of the programming device or the programming software; the micro would not be to blame.


Well, either I'm in the top 0.1% (I know I'm good), or I don't know my hand from my foot (also possible).

LOL well yes they are all crap-they are based on the 8032, based on the 8080, based on the 8008, based on the 4004... all intel crap. a wise man, one of the origionaters of BBS software, once said, in an intel/motorola spat, "intel can't achitect their way out of a paper bag".

They were only worth using until recently because they were the most cost effective on the market. Now that apps are getting bigger they are jammed up by the same things that intel scammed us with in the early PCs-segment registers, and all the rest of the crap that goes with using a 8 bit processor in a 32 bit application.

Dan

 

[Sceadwian]

The actual silicon designs have changed dramatically Ubergeek. 8051 nowdays only means generally code compatible (excluding the plethora of peripherals available) There is not much thats is 'crap' about them, and they're drop in replacement for older 8051 is the main deal, often with more advanced features.

 

[Ubergeek63]

The actual silicon designs have changed dramatically Ubergeek. 8051 nowdays only means generally code compatible (excluding the plethora of peripherals available) There is not much thats is 'crap' about them, and they're drop in replacement for older 8051 is the main deal, often with more advanced features.

actually i design aound them all the time and am well aware of the pros and cons. The overall architecture sucks and the only reason you do not notice it is because you are not likely to be fighting with the achitecture, the compiler is.

But... even the best compiler can not hide the performance hit that occurs when the chip runs out of memory and has to start bank switching. And only the lack of exposure to good architecture could allow the worship of the 80xx familly to continue.

Now that $1 ARMs exist, there is no reason for lesser parts to be learned about since they all pale in comparison performance wise and cost the same or more. Therefore, they will eventually die off as whitegoods workhorses and hense be an incredible waste of time to learn professionally.

Dan

 

[Sceadwian]

They are still used... widely, that is one reason to learn then. If it was all about architecture in the first place someone would have burned the X86 line to the ground years ago. The real world doesn't work like that though you as an engineer already know that.

You SAY they'll eventually die off, but almost every home PC regardless of it's power and instruction set additions is still based off a design that is 30 years old.

What about ladder logic, same thing. It's been obsolete for how long now, yet many PLCs still use it. It's entire existence is based on the simple fact that the people that were REALLY good at relay logic couldn't program or develope for the new micro controllers that were coming out, so engineers had to engineer around that. You can't just snap your fingers put the worlds most advanced architecture on the ground floor and expect the real world to just 'deal with it'

How many devices are there in the real world already that are still growing in popularity and sales even though their basic architecture is obsolete? Oh yeah, ALL of them.

 

[Ubergeek63]

They are still used... widely, that is one reason to learn then. If it was all about architecture in the first place someone would have burned the X86 line to the ground years ago. The real world doesn't work like that though you as an engineer already know that.

You SAY they'll eventually die off, but almost every home PC regardless of it's power and instruction set additions is still based off a design that is 30 years old.

What about ladder logic, same thing. It's been obsolete for how long now, yet many PLCs still use it. It's entire existence is based on the simple fact that the people that were REALLY good at relay logic couldn't program or develope for the new micro controllers that were coming out, so engineers had to engineer around that. You can't just snap your fingers put the worlds most advanced architecture on the ground floor and expect the real world to just 'deal with it'

How many devices are there in the real world already that are still growing in popularity and sales even though their basic architecture is obsolete? Oh yeah, ALL of them.

I never said that they were not still used. I said that new designs do not generally use them as economics dictates otherwise. Any company that wants to stay in business will use what is the most economical since if they don't their competitors will out do them easily.

In the long run for someone looking to make a profession of it, it makes no sense to learn 8051s.

As to PCs, while there is a silicon and power hogging "compatibility mode", they have not been the same since the pentiums came out with 32 bit native mode, though they kept the ridiculous assembly nomenclature so it looked similar. Even so it is still basically a single accumulator crap architecture in a 32 bit system.

Powerful architectures do not have accumulators, they have general purpose registers 32 bits wide, between 8 and 32 of them. If you are familiar with your pet 8 bit assembly code just think how much less juggling you would have to do if you had 32 accumulators at your beck and call!

Dan

 

[Sceadwian]

Maybe a profession in new designs ubergeek, but not ones based on existing designs. I think you missed the point of my last post, when the circuit is working no one wants to change it, so they keep using 'obsolete' parts, even new obsolete parts like the 8051. In order to get rid of 8051's you'd have to redesign the whole thing from the ground up. Problem is 8051's WORK just fine in their given applications. If you so much as introduce one single new bug in your 'superior' design entire industries will abandon your products for an 'inferior' product that does what it's supposed to.

 

[Ubergeek63]

Maybe a profession in new designs ubergeek, but not ones based on existing designs. I think you missed the point of my last post, when the circuit is working no one wants to change it, so they keep using 'obsolete' parts, even new obsolete parts like the 8051. In order to get rid of 8051's you'd have to redesign the whole thing from the ground up. Problem is 8051's WORK just fine in their given applications. If you so much as introduce one single new bug in your 'superior' design entire industries will abandon your products for an 'inferior' product that does what it's supposed to.

true... but that is not what happens on new white goods designs, only on the small niche stuff and often not even then.

We have a quote out now for one that the customer has 5-10 years worth of old CPU stock and if their customer buys the change they asked for it will cost less to throw away the old 8032 parts that we do not have the compiler for for a completely new design, than it would to get the compiler and such to reprogram the crappy thing.

BTW the 8051s have only gone uneconomical in that past couple years. I was not meaning to imply that it was going to happen in the next month or even year, only that it would be more profitable to learn the up and coming instead of the old and tired.

Dan