What does it take to build a PSP?

[QuantumTheory]

I'm not sure anyone here knows, but let me explain what a PSP is. A PSP stands for "Playstation portable" its a home entertainment system which you can do the following:

1) Look at photos
2) Download music, listen to music
3) Download videos, listen to video
4) Play full 3D video games

All downloaded from your computer. It also has B wi-fi, so you can connect to hotspots or a router to play games online or to browse the web.

Now, I'm curious. Who engineers these technological marvels? Is there a physicist, sitting in the background, doing calculuations on how to design it? It has a circuirt board. How do they design this machine? Is knowledge of physics/calculus nessicary?

I hope I can one day build such a machine, thanks to science.

Note: It also uses a laser to bread UMD (Universal Media Discs) and has full length feature films on these discs.

 

[Cliff_J]

The PSP is a combination of a lot of technology all blended into one device and its late arrival is an indication of how far some of those technologies needed to mature to make all of that fit into such a small device.

The feature set is a product of assembling many individual integrated circuits (IC) chips with each being a marvel of technology where a small square of very special melted beach sand (silicon) allows for an incredible number of electrical devices in close proximity that require little power to operate.

Getting all those little pieces assembled and writing software to make it all run smoothly would fall under electronic / computer /software engineering and is likely done on very specialized software.

Designing a case that can protect the circuit board and packages everything to hold the batteries and screen and buttons is more like mechanical engineering / industrial design and will likely be done on very specialized software again.

Then when you figure this is done by a team where all the compromises need to be worked out and each change tested and so on it becomes a very impressive endeavor that would require the financial resources of Sony to pull it off. It may not be on the level of our Apollo space program or the design of the 777 or A380 airplanes but it is still a big deal. The gant charts alone probably needed a dozen reams of paper. How do you calculate the costs in terms of caffine alone with the sodas and coffee that must have went into it?!

 

[exequor]

Now, I'm curious. Who engineers these technological marvels? Is there a physicist, sitting in the background, doing calculuations on how to design it? It has a circuirt board. How do they design this machine? Is knowledge of physics/calculus nessicary?

I have a similar curiosity. I often wonder how long it would take me to just design a dvd player but then I've realized that these things seem like to much because I don't have the knowledge (microelectronics, dsp, etc.) of how they work and are engineered. Once I do I know it will be as easy as shutting down my pc
.

 

[QuantumTheory]

That is very interesting. I hope to find gold in my PSP (ISn't there gold plated material in there?)

 

[chroot]

Physicists generally have nothing to do with consumer electronics design. A good number of physicists spend their time studying ways to make electrical devices smaller, faster, or consume less power.

The field of integrated circuit (IC) design can be loosely divided into design and manufacture. Physicists work on advancing the processes by which ICs are manufactured, while engineers actually design circuits which use the processes.

The people who design devices like the PSP are electrical (and computer) engineers. Hundreds of thousands of man-hours are required to engineer all the different components that go into a device like the PSP. A single person would need dozens of years of education to actually be capable of designing and building every component.

- Warren

 

[amuron]

Much of it depends on the level of integration needed. For most consumer electronics devices a vast quantity of components and or libraries are readily available, thus its not as bad as one would think. Still complex, but not beyond the realm of a few dedicated people to do (provided funding is available), combined with a lot of beer, coffee, and pizza. Now to do it all from scratch is another story entirely, but few entities can justify the costs outside of the medical arena,where they are forced to DIY due to liability issues.

If we go down the function lists.
1) Look at photos
2) Download music, listen to music
3) Download videos, listen to video
4) Play full 3D video games
5) Wifi

All pretty easy to do with off the shelf components basically an embedded pc chipset, an os, and some firmware to make it fly, and some cool packaging. I would guess 24 man months is all that would be needed if that. Now, to meet the price targets, and achieve marketability the complexity shoots through the ceiling... and much of it will be driven by marketing feedback and design changes along the way. No doubt the delay was probably caused not be technical issues, but by market driven ones causing major rip up and redo activities.

Now to design each component and function from silicon on up, it might well take tens of thousands of man months.

As far as calculus goes... its going to be behind the scenes, typically in the arena of thermal/power budgets, and the internal processes. Few firmware developers will be involved in that arena. They may however have libaries that make extensive use of numerical analysis techniques, but to them, its just a black box. Otoh, from a ground up approach, the process developers live with it pretty much constantly.

Ron

 

[QuantumTheory]

I love this thread! hope more people can respond to it!

 

[waht]

re

New chips come out very often these days with more computing power, less power consumption and less heating. It's now up to the engineer to take this new chip and implement it in a new product.

To design a new mircoprocessor takes teams of dozens of best engineers working 'round the clock for months or even years.

It is unlikey that you could do all that by yourself, but you can certainly be part of it.

 

[QuantumTheory]

New chips come out very often these days with more computing power, less power consumption and less heating. It's now up to the engineer to take this new chip and implement it in a new product.

To design a new mircoprocessor takes teams of dozens of best engineers working 'round the clock for months or even years.

It is unlikey that you could do all that by yourself, but you can certainly be part of it.

wow that is fasinating! what do you have to know to design such chips? calculus and physics?

 

[Maxwell]

wow that is fasinating! what do you have to know to design such chips? calculus and physics?

Yes, knowledge of physics and calculus is important. However, those topics are only the very fundamental beginnings of the tools engineers use. The engineers designing these devices have a background in circuit design/analysis, power electronics, semiconductor devices, digital signal processing, etc. These topics span many different fields, so it is easy to see why it would require a team of engineers from different backgrounds to assemble a PSP. And that is just the hardware -- there is the software aspect of the device as well!

The people working on the PSP are electrical and computer engineers. If you are interested in designing such devices, that is the route to take.

 

[QuantumTheory]

does it take calculus to design the PSP?

What calculus in engineering does it take to design the PSP? (Playstation portable system)?

Thanks!

 

[Pengwuino]

You've asked this question like 10 times!

 

[QuantumTheory]

You've asked this question like 10 times!

I already asked it?

 

[berkeman]

I already asked it?

https://www.physicsforums.com/showthread.php?t=74335

 

[Moonbear]

I've merged the two threads on this topic.

 

[QuantumTheory]

sorry about that!

 

[chroot]

Is there a reason why you are so concerned with the use of calculus in the design of the PSP? Homework project, perhaps?

Calculus is generally only used in the lower levels of the design, which involve analog circuits for amplifying and converting information between analog and digital forms. Calculus is also involved in the design of the individual electrical components, such as transistors and diodes, within the integrated circuits. Variants of calculus can be found in software comprising the audio and video codecs.

- Warren

 

[QuantumTheory]

Is there a reason why you are so concerned with the use of calculus in the design of the PSP? Homework project, perhaps?

Calculus is generally only used in the lower levels of the design, which involve analog circuits for amplifying and converting information between analog and digital forms. Calculus is also involved in the design of the individual electrical components, such as transistors and diodes, within the integrated circuits. Variants of calculus can be found in software comprising the audio and video codecs.

- Warren

Thanks. Actually no its not a homework project. I just ask a lot of questions and am curious about calculus's involvement in real life. I am interested to know what it is actually used for in real life. Do you explain more about audio codecs? Not sure what those are. Alot of smart cookies here, I barely know any of this, and I'm 17!

 

[QuantumTheory]

Analog and digital form. You mean like, when I push a button or when I push the D pad, this gets converted into electrical impulses ? is this what you mean?

thanks

 

[Cliff_J]

Calculus is one of the basic understandings needed to complete higher physics and engineering because the math behind calculus is the basis for the understanding of the theory. Finding a direct relation and explaining if a laplace transform or a Fourier series is what you need to study is quite a bit more difficult, but once you understand those areas you begin to apply them in new ways.

A codec is a COmpression-DECompression standard, its going to be a combination of both math and for lossy formats like MP3 or JPG a study of human senses. How much background noise or low level audio can be discarded before you can hear a difference, how much color and detail information can be discarded before you see a difference? Or, it could be better said its a question of how much can be discarded to make for an acceptable compromise of storage/speed to reduce cost or improve playtime?

Life is analog, the sun comes up and the light progressively gets brighter as the angle relative to the surface improves, water flows in a river as influenced by gravity. Now we could quantify such activities and represent the quantifications with numbers. We could then specify how we look at those numbers and further refine it into how many numbers we want to look at.

Take a person talking for example. Going back to bell we've had the ability to talk, creating air pressure variations that moved a diaphram with a coil of wire and transmit this on copper wires over distance to another diaphram with a coil of wire and magnet and have a reasonable copy of the sound transmitted. Today this could be captured by a small chip constantly examining the audio coming from the device sampling the air pressure variations and making each of those variations a number. Each number could then be processed into pulses of light sent over a thin piece of glass for miles and sent to a satellite thousands of miles in the sky and converted back again to an analog air pressure variation that would be the sound you hear on the receiving end of a telephone.

The same goes for a PSP. It has a small computer and software inside with numbers describing everything. Now it goes through many a process to convert those numbers into light/sound that you can use. Otherwise it would be about as exciting as watching the green symbols changing in the movie the Matrix, digital is a means to an end but it isn't really the end itself.

At 17 you can explore much of this in many different ways. Maybe you want to be a game designer and like to think creatively about using color and role-playing to make new challenging levels of games. Maybe you're fascinated with the math of kinematics and could go into developing a new software game engine or video chip design. Maybe analog circuits are of interest and you want to be involved in helping reduce the power consumption so the device runs cooler and uses less battery power. Hundreds of possibilities exist.

Just as there are hundreds of ways calculus is used, even if indirectly. Some brilliant people have taken the transistor valve and shrunk it to the size and cost that makes my computer possible, but a lot more people were involved in making hundreds of other items that make just my computer possible much less our conversation. Its mind boggling to consider that we've come this far from hunting-gathering, but an appreciation of such a thing is important to respecting the beauty of what we've created and the progress we (as a society) make through time.

Appreciate the beauty, but also find a outlet to use it if you desire to and don't be concerned with all the details that go into what makes your goals possible. I think the plain old light bulb overhead is a impressive invention I can purchase for a few cents, but I don't think much past the switch besides how it can provide light so I can do work. And a complex device like a PSP has a lot more to it than that!

 

[Maxwell]

What calculus in engineering does it take to design the PSP? (Playstation portable system)?

Thanks!

chroot and Cliff answered this question in much detail. Reading their responses will provide some very useful insight into how engineers use their knowledge of calculus.

If you are curious about which calculus classes engineers take, here is a general list:

Calculus 1 (Differential calculus)
Calculus 2 (Integral calculus)
Calculus 3 (Multivariable calculus)
Differential Equations

Engineering students usually finish these courses within their first two years. After they have completed these courses, the techniques are used throughout the rest of their education.

 

[QuantumTheory]

Calculus is one of the basic understandings needed to complete higher physics and engineering because the math behind calculus is the basis for the understanding of the theory. Finding a direct relation and explaining if a laplace transform or a Fourier series is what you need to study is quite a bit more difficult, but once you understand those areas you begin to apply them in new ways.

A codec is a COmpression-DECompression standard, its going to be a combination of both math and for lossy formats like MP3 or JPG a study of human senses. How much background noise or low level audio can be discarded before you can hear a difference, how much color and detail information can be discarded before you see a difference? Or, it could be better said its a question of how much can be discarded to make for an acceptable compromise of storage/speed to reduce cost or improve playtime?

Life is analog, the sun comes up and the light progressively gets brighter as the angle relative to the surface improves, water flows in a river as influenced by gravity. Now we could quantify such activities and represent the quantifications with numbers. We could then specify how we look at those numbers and further refine it into how many numbers we want to look at.

Take a person talking for example. Going back to bell we've had the ability to talk, creating air pressure variations that moved a diaphram with a coil of wire and transmit this on copper wires over distance to another diaphram with a coil of wire and magnet and have a reasonable copy of the sound transmitted. Today this could be captured by a small chip constantly examining the audio coming from the device sampling the air pressure variations and making each of those variations a number. Each number could then be processed into pulses of light sent over a thin piece of glass for miles and sent to a satellite thousands of miles in the sky and converted back again to an analog air pressure variation that would be the sound you hear on the receiving end of a telephone.

The same goes for a PSP. It has a small computer and software inside with numbers describing everything. Now it goes through many a process to convert those numbers into light/sound that you can use. Otherwise it would be about as exciting as watching the green symbols changing in the movie the Matrix, digital is a means to an end but it isn't really the end itself.

At 17 you can explore much of this in many different ways. Maybe you want to be a game designer and like to think creatively about using color and role-playing to make new challenging levels of games. Maybe you're fascinated with the math of kinematics and could go into developing a new software game engine or video chip design. Maybe analog circuits are of interest and you want to be involved in helping reduce the power consumption so the device runs cooler and uses less battery power. Hundreds of possibilities exist.

Just as there are hundreds of ways calculus is used, even if indirectly. Some brilliant people have taken the transistor valve and shrunk it to the size and cost that makes my computer possible, but a lot more people were involved in making hundreds of other items that make just my computer possible much less our conversation. Its mind boggling to consider that we've come this far from hunting-gathering, but an appreciation of such a thing is important to respecting the beauty of what we've created and the progress we (as a society) make through time.

Appreciate the beauty, but also find a outlet to use it if you desire to and don't be concerned with all the details that go into what makes your goals possible. I think the plain old light bulb overhead is a impressive invention I can purchase for a few cents, but I don't think much past the switch besides how it can provide light so I can do work. And a complex device like a PSP has a lot more to it than that!

Thakns so much. I..don't know how to respond to this.

I have an interest in technology and how it advances. Are there any good wel paying jobs in this? Do engineers make a lot of money? I am trying to find a good job that i like that makes a lot of money. Do you think there will ever be a point in time when we cannot invent anymore and technology stops?

 

[Maxwell]

I have an interest in technology and how it advances. Are there any good wel paying jobs in this? Do engineers make a lot of money? I am trying to find a good job that i like that makes a lot of money. Do you think there will ever be a point in time when we cannot invent anymore and technology stops?

There are plenty of well paying jobs in engineering. Engineers make a fair sum of money -- but it's important that you REALLY enjoy it. I can't tell you how many people start engineering school because of the potential salary, only to drop out at the end of the semester.

 

[amuron]

wow that is fasinating! what do you have to know to design such chips? calculus and physics?

Its really multidiscipinary. If you get into the nuts and bolts of chip design, calculus becomes less of a factor. If you get into process development, then chemistry becomes predominant as well as physics. As one poster alluded to earlier, calculus will be invaluable in your academic career. Out in the field, short of specific positions, other multidisciplinary areas will over shadow it.

One other thing is that short of the archetecture guys or the systems engineers, no one really has an overall concept of how all the pieces interact, and those guys don't know the ic design, or process. Process guys often times will have about zero knowledge of the ic's function, much less the end product. A friend used to work for a big name in medical equipment. He related that short of few guys out of a thousand, no one really understood the underpinnings of what they were doing. Each had a part, which they knew really well, but less than 5 really understood the system... and much of what they were doing was based upon the results of visionaries 20 years prior. In my career, I found the same thing.

It is also rare for a company to be vertically integrated. Eg, the chip designers may rarely if ever interact with the process guys, nor the hardware guys with the manufacturing guys, much less the chip designers. There are a lot of documents just thrown over the wall to the next group... this can make for a mess, but that's how the industry, and a lot of companies are organized. From a finance standpoint, it makes sense, but it creates a multitude of headaches for the engineering folks.

Ron

 

[amuron]

There are plenty of well paying jobs in engineering. Engineers make a fair sum of money -- but it's important that you REALLY enjoy it. I can't tell you how many people start engineering school because of the potential salary, only to drop out at the end of the semester.

The first few years of engineering are also simliar to medical field. Tons of hours, and combined with minimal sleep are common in the electrical and computer field. At that point, if you don't enjoy it, engineering salaries outside of process are not going to be near enough compensation.

The process guys at the fabs used to make the really big bucks, but they also had the highest stress, and probably close to zero life outside of work. Eg on call 24/7 in a lot of places, combined with the fact that an error could cost hundreds of thousands of dollars in a flash.

Ron

 

[QuantumTheory]

The first few years of engineering are also simliar to medical field. Tons of hours, and combined with minimal sleep are common in the electrical and computer field. At that point, if you don't enjoy it, engineering salaries outside of process are not going to be near enough compensation.

The process guys at the fabs used to make the really big bucks, but they also had the highest stress, and probably close to zero life outside of work. Eg on call 24/7 in a lot of places, combined with the fact that an error could cost hundreds of thousands of dollars in a flash.

Ron

wow that is rough..what job is well, easy going? LOL

 

[Pengwuino]

wow that is rough..what job is well, easy going? LOL

Be a teacher

 

[Cliff_J]

Thakns so much. I..don't know how to respond to this.

I have an interest in technology and how it advances. Are there any good wel paying jobs in this? Do engineers make a lot of money? I am trying to find a good job that i like that makes a lot of money. Do you think there will ever be a point in time when we cannot invent anymore and technology stops?

You're welcome.

The main point I was trying to convey, and others here too, is that 'technology' is a very big thing, just as engineering is a large field.

Making assumptions from your line of questioning, unless you're ready to commit to a lot of calculus and higher math (and if you're starting now we're talking 5+ years) in most any degree programs. If you can't test out of trig and algerbra and do college level calculus the 1st semester, then you'll need to complete trig/algerbra and maybe calculus intro in order. That will add a year or more to the program, because you can't take calculus based physics until you complete calculus normally. If you do a work co-op program as part of the engineering degree, you could be 7 years away from gainful employment, in a marketplace with competition coming from other countries where seasoned engineers make $8,000/year, and yes that isn't a typo that's eight thousand a year, less than a fast food job here pays.

Find an aspect of technology that you like for good reasons. Let's say video game consoles and the games are really your passion and you think that even as you mature you'll still enjoy being with it all day long. You might consider being a journalist who writes for a magazine about the technology, it might not pay super well but would allow you to travel to conventions and play with new gadgets for free.

You may want to be a game designer, a pretty rare job and you'd need some good talent for it (might not know until you try) where you could go to a school with a good arts/computer graphics department. Even if you're using the latest greatest big $$ technology you need not know the math behind the tech, you put in the correct numbers or shade in the correct color and you're set. In the CG field there are people who specialize in model creation, they do nothing but study movement of people or animals and create a CG model following that movement. Then a CG artist takes that model and makes it look correct. Get really good and who knows, maybe a job in Hollywood or taking a movie and making it a game are possible, even if it falls in the category of NBA star it has a little better safety net as graphic designers are involved in plenty of mundane tasks like brochures and promotional materials.

I'd suggest a career book from a local bookstore to try to put some focus on what you want to do. I think all of us at some point wanted to do it all, at 14 I was still sure I could be the next Kelly Johnson who's skunkworks team brought us the U2, SR71, and stealth fighter. While on advice, this is an important time make sure your SAT/ACT scores are good and choose the correct college and one that is financially workable. Having a general career choice in mind makes that process easier since some schools won't make the list. If engineering is the choice, make sure you understand what you are signing up for ahead of time.

There may be times where some aspects slow down in terms of development, but technology will always move forward. Optics was thought of as being 'finished' by some people decades ago, yet I doubt some of the recent satellite photos or astronomy advances would happen if everyone thought it was 'done'. And then to not only keep a level of progress but make it affordable, 85% of the worlds population is available to benefit from lowered costs of goods in many areas.

 

[QuantumTheory]

Thank you Cliff_J. I wrote a thread about engineer in general on the engineering in general category, please help me with it