The Basics Of RAM

Segfault 11 writes: "Ars Technica has Part 1 of their RAM series, which discusses the way that static and dynamic RAM work, in addition to covering memory modules. To get started, click here to learn more."

Comments

Re:memory modules

[Kickasso]

Young man, back in MY day we didn't have DIPs soldered onto the motherboard. We simply didn't have DIPs to solder onto the motherboard. For that matter, we didn't have motherboards to solder DIPs onto. We had CORE . Talk about newfangled.

ROM basic? Phew! Real men load kernels by TOGGLING the phucking SWITCHES!

(Feel free to reply with completely made up story about vacuum tubes and CRT memory :)

Re:Isn't static RAM dangerous?

[Montressor]

You dope, your cache is static RAM. L1, L2, the whole works - those are mostly SRAM, with some exceptions.

Re:Computers are for everybody

[stevelinton]

Two things. Firstly it took a long time, 50 years or so, before it was possible to operate a car with such little knowledge of its operating principles. A lot of reliability engineering was needed, and a lot of user interface standardisation (like the steering wheel, and which way it turns).

Secondly, one of the truly remarkable things about computers, and software is how much it is possible to gain from using them, even when you don't actually understand very much about how to use them, or anything about how they work, and even when they work pretty badly and unreliably (like most current software). Just think how much we'll be able to do when (a) they're built right, and (b) we all know how to use them

Re:Computers are for everybody

[stripes]

I know nothing about how my car works. I am not at all comfortable with upgrading it or maintaining it. The most complicated act I am able to perform is to put gas in it.

I'll bet you, and most other people know more about cars then that. Even ignoring the whole driving part.

You know to take it to regular mantinace, otherwise you wouldn't be able to drive it for very long. You know when the little red lights show up on the dash that you should call someone. Hopefully you know which little red lights mean "turn the engine off now now now" (hint: lack of oil can destroy an engine very very fast), if your lucky you know which ones means "you can probbably drive me ok for a bit, but you won't get me to start again". You probbably know not to turn the engine off in the middle of a drive (unless you have the car in neutral).

I'll bet you even know how to change a spare tire (if you don't, go out to your car and look, it ain't that hard, and you might rather learn now then in the rain...when your cell phone is dead...and your late for a job interview). And where the engine is. And maybe even have a general idea of what some of the engine parts look like and are for. Many people do.

On the other hand most computer owners won't know similar things. What dialog boxes mean "get help now", and what ones mean "click this away and finish what your doing"? If you open the box, what part is the CPU and what is the RAM? Are there any parts of a computer that wear out and need mantiance? (I figure it's a good idea to back up disks, and check to see if the fans still spin every once in a while)

I admit, today this scenario does not exist in as ideal a fashion as I have laid out. But that doesn't imply it cannot exist. Computers are fundamentally no different than cars - just more complicated. Thus, while it should take more work on our part, a system is able to be designed which fullfills these criteria. It is a challenge to computer scientists to create this system.

No, it's a chalange to marketing folks somewhere. A chance to get folks to spend $25 a year on a "computer oil change" millions of folks...

It's not much of a technical change. People use computers will little knolage of how the hardware works, and it only hurts them when they go to buy a new one (same as someone lack of car knolage...except there are fewer decent reviews of new computers, but you can still drag a "smart" friend to the store). Oh, plus it hurts when they forget to back up valuable data...but more from software fuckups.

Now the lack of software knolage, that hurts people all the time. Fixing that (by incresing knolage, or decresing need) is a way harder chalange. (and I have noticed a genneral increse in the amount of computer knolage the avrage person has)

Furthermore, the digital computer is so powerful a tool that it should not be withheld from the majority of society for any reason. It has the power to radically alter our entire society, and if that means the loss of comfort for the "computer elite" (which I do not believe it does) than so be it. Why should society be denied such benenfits simply because a few people prefer computers to be special hi-tech playthings? I am happy that the car was not denied society in a similiar fashion.

Who's been denying? Apple and Microsoft have been working on making computers easyer to use, and people have been learning more about them for years. Decades. They made lots of progress at first, and have slowed quite a bit. But that is because the problem is hard, and keeping your current users happy makes it harder. Not because they have some insane desire not to increse the number of potential buyers by tenfold or more.

Re:You know more than you think

[Spazmoid]

Very true, however automobiles, and their most basic processes have been a part of society for a century. What is needed to be known about such a pervasive part of society is ingrained at every turn. For instance your dad forcing you to hold the flashlight at 1AM as he futzes with changing the belts then takes it to the mechanic. That type of knowledge (subliminal if you will) has grown since the car became a commodity. Do you think that in the early days of the automobile, everyone that had one knew as much as you do about yours, which is significantly more complex, but works on the same principles? Sure, mechanincs and automobile 'geeks' of the day knew what they were doing, but there was not as much burden, as there increase in automobile usage was not nearly as exponential an increase as there is in computer usage.

Technology, computing specifically, is new to the majority of users, the know that they CAN use the computer as a tool, but they lack the basic skills and knowledge to manipulate that tool to its fullest. They lack the knowledge to proctec themselves from spammers, virii, ect. WHY DO YOU THINK WE ALL HAVE JOBS???

As time goes on, computing will become safer, and easier for the masses. At the same time the masses will gradually accumulate more of the basic knowledge to keep from annoying so much, so we will have more time for our QUAKE 15 deathmatch.


www.mp3.com/Undocumented

Re:Computers are for everybody

[FFFish]

Everyone and their dog is going to say that you know lots about your car.

What they're ignoring is the validity in what you say. Instead of focusing on your point, which I think was very well made, they're going to nitpick at the illustration you used to carry the concept.

Well, then, let's change the illustration:

I know bugger all about how my microwave works. If it went on the fritz, I wouldn't know the first thing to do. It'd hit the garbage, and I'd buy a new one. There's enough scary things in it that I *don't* want to muck about trying to repair it!

I don't know how my television works. There's a couple of lines into it, and a couple buttons on it, and it works. If anything goes wonky, it's off to the repair shop.

Same goes for the VCR. If it ate a tape, I'd recognize that that's a different sort of problem than a wobbly picture. In both cases, it'd go to a technician.

There a people who see their computer the same way I see a microwave: a scary black box that does magic.

I can't blame them for not wanting to dink around with it. After all, if you let the smoke out, it stops working!


--

Computers are for everybody

[bolan]

I know nothing about how my car works. I am not at all comfortable with upgrading it or maintaining it. The most complicated act I am able to perform is to put gas in it.

But I still use it, and need to use it, every single day. Ignorance about the internals of my car does not preclude me from using it in a valid fashion. Granted, if I were more informed about its workings I would probably be able to use it more efficiently, and more productively. But this gain is not worth the hassel I feel it would be to learn these things.

Computers are no different. They are a tool - nothing more, nothing less. The average person doesn't care how data is actually stored in their computer, or how interrupts work, or anything. They want a machine which is perfectly intuitive and can let them write reports or generate presentations or email their friends, and they do not care how these feats are accomplished.

I admit, today this scenario does not exist in as ideal a fashion as I have laid out. But that doesn't imply it cannot exist. Computers are fundamentally no different than cars - just more complicated. Thus, while it should take more work on our part, a system is able to be designed which fullfills these criteria. It is a challenge to computer scientists to create this system.

Furthermore, the digital computer is so powerful a tool that it should not be withheld from the majority of society for any reason. It has the power to radically alter our entire society, and if that means the loss of comfort for the "computer elite" (which I do not believe it does) than so be it. Why should society be denied such benenfits simply because a few people prefer computers to be special hi-tech playthings? I am happy that the car was not denied society in a similiar fashion.

Philip Quirk

Re:Computers are for everybody

[AstroJetson]

If you know to put gas in your car, you know more about cars than most people know about computers. You know, at least, that it needs gas and regular maintenance. Most folks don't even know that much about the computers they drive. If people drove their cars they way they use their computers, they'd drive until they ran out of gas. Then they would call tech support.

I'm not trying to be elitist. I agree that it would be nice to have a computer that would operate the way you'd like it to. But you admit that we aren't yet at that point. Until that day, it behooves people to learn some basics about how computers work - such as the difference between disk space and RAM.

Re:Computers are for everybody

[BigBlockMopar]

I know nothing about how my car works. I am not at all comfortable with upgrading it or maintaining it. The most complicated act I am able to perform is to put gas in it.

Let's put things into perspective for the computer industry. This is going to seem long and drawn out (aren't all my posts?), but it sheds a new light on the computer.

As a car nut, let's do a quick brief:

The first mass-produced car came out just after the turn of the last century. Almost 100 years ago. With the first affordable car came the first "backyard mechanics", people who maintain their own cars.

In those days, cars were unreliable enough that you could expect to have to fix something every thirty miles or so, and most drivers did it themselves.

Continual refinement brought us to self-adjusting hydraulic brakes, engines that self-cranked, automatically adjusted the ignition timing, and had pressurized lubrication systems by the 1930s. (Prior to that, you had to deal with clutch, throttle, cable-operated brakes, shifting, setting the ignition timing as well as steering - it took a lot of experience and theoretical understanding of how the car worked.)

The 1940s found the first "modern" cars: anyone who can drive a stickshift can drive a car from the 1940s or later, even if they are a little more temperamental. One car that I love to drive is a friend's all-original 1948 Chevrolet Business Coupe. But, even so, the thing doesn't have enough power to be able to get up to freeway speeds.

By the late 1950s, the automatic transmission, was initially available in two speeds (the GM "PowerGlide"). Bigger engines came out, allowing the new cars to cruise easily on the new Eisenhower Interstate System. By the early '60s, Chrysler had brought out their legendary TorqueFlite 727 automatic, a nearly bulletproof three-speed automatic transmission with a really neat planetary gearset design which has since been copied into most modern automatic transmissions.

In the 1970s, user features increased, drivetrain and mechanical manufacturing tolerances were increased, tire technology increased. Electronic ignition replaced points and condensors, making yet another part of the car virtually maintenance-free. But reliability still sucked, primarily because of new federally-mandated emissions controls.

The 1980s brought us EFI systems, which greatly improved reliability and driveability over the primitive emissions control systems of the 1970s. The 1974 oil crisis was working its way into automotive designs, and by 1980, the first of the modern and efficient OHC 4-cylinder engines were hitting the showroom floor. Four speed automatics were coming out, negating a big fuel-efficiency issue that kept the sales of stickshifts up. As the earlier cars were gradually retired and replaced with their more reliable equivalents, the service centers at the side of the freeways started to convert their service bays into fast food franchises.

Today, cars are easy to operate, with very few parts that require periodic adjustments. Machining technology means better fit between parts, less oil leaks, etc. And the sophistication of today's engines and engine management systems makes people feel uncomfortable about taking things into their own hands and fixing an oil leak or other small problem. Spark plugs now usually last 60,000 miles or more - and even if the backyard mechanic of today knows how to change them, they're often difficult to get at. Cars today aren't perfect, but they do represent 100 years of continual refinement and design experience.

So, where does that put us, relative to computers?

I'm going to guess that the majority of computers, running Windows 9x/NT/2K, are now at about the point that cars were at in the late 1960s. To use them really effectively and to avoid being stranded by the side of the road, you should understand how they work and keep a few tools in the trunk. But you can still toss the keys to the wife so she can do the grocery shopping.

That's very impressive: it took the car industry almost 70 years to get from the first mass-produced cars to that relative level of sophistication. With the computer industry, it's been approximately 20 years since we were typing "pr#6" on our Apple IIs to read the disk drive.

In a few years, when computer OSs are actually stable, Plug and Play works, software is always easy to install or remove, and everyone has got a system powerful enough to watch DVD movies, I'll suggest that the computer industry will have reached the level of maturity that the automobile industry is currently at.

The future? Flying cars, and computers with working voice recognition.

Re:A Top Notch Article

[grahamsz]

Yeah much as I dont like x86 assembler much, and detest the r2k assember we did at uni last year, it does give a good insight into how things come together. It's also damn handy when u need to remove irritation limitations from trial software (that last point makes it painfully obvious how few programmers do put any great depth into their programs). I was trying to help a friend of a friend with Dragon Naturally Speaking Pro (quite an expensive program i think). He's disabled and had been given it from his college but when his windows died (shock horror) he went to reinstall it and had no serial number. Yet it took me about 20 mins to reverse the installsheild script and skip the sn check... and these people are supposed to be skilled coders!? Personally i'm all for open source since it avoids all these problems - like the me having to remember asm one :)

But when I first fumbled my way onto the internet five years ago i'd have loved to have a book that just gave brief descriptions of things like: modem, ppp, ip, tcp, ftp, bbs, dns etc... since it would have made my life a lot easier. However those first two weeks trying to get an SMTP deamon runnning on a Mac so i'd receive email taught me so much about how things worked that retrospectively i'd never have skipped them for a one click sign up.

Re:hmm. they don't cover ECC RAM

[Detritus]

It's a false economy, plus at one time, Intel and other vendors were telling people that they didn't need parity or ECC RAM.

ECC RAM is more expensive and it can be hard to find. Many stores don't stock it. In some computers it can be a bit slower than regular RAM.

After having repeatedly dealt with RAM with pattern sensitivity problems, I am a firm believer in ECC RAM. Trying to diagnose and repair a computer that crashes every few days, or only when certain programs are run, can be a major headache. Is the problem the RAM, other hardware, application programs, device drivers or the flakey operating system?

Static RAM got me in the balls.

[BigBlockMopar]

Yes. Static RAM can be highly dangerous.

Especially if you're out in the middle of the woods, on a moonless night, camping with your friends. The campfire has died down, and you've just finished off a six-pack of your favorite beer. You're now heeding nature's call, staggering away from the glowing embers in the firepit...

And then BANG!, the tailgate that you just *knew* you should have closed gets you in the balls. You lie writhing on the ground for a few minutes, then regain some control over yourself as the pain subsides, start to get up and you clock your head on the back bumper, which you didn't realize you were under.

Never turn your back on a static Ram. Or any other pickup truck that you've parked in the woods on a camping trip.

Isn't static RAM dangerous?

[vertical-limit]

I know a question like this is probably old hat (or is that "red hat" :P) to you Linux veterans, but as a long-time BeOS user, I've never quite understood why people insist on using static RAM. Everyone knows that the inside of a computer need to be cool and safe -- remember that incident where Pentium chips kept melting down? Static RAM is dangerous; it can't be relied on to not burn out when it really counts.

(Why is that? Everyone know that static energy is actually a form of electricity. And electricity produces heat when it moves, thus causing the interior of the computer case to heat up. See Jager et. al -- subscription required, sorry.)

You may laugh this off as excessive worrying, but this is a serious problem. I've overclocked a few chips, and they've all died pretty quickly -- and that was without static RAM. High-powered memory zapping around would kill the chip even more quickly. And what if that chip is powering some mission critical application like a space shuttle flight or an air traffic controller?

Just my two cents,
- vl

Re:Isn't static RAM dangerous?

[curiosity]

Actually, static ram doesn't retain its data without power. That would be NVRAM (Non-Volatile RAM). DRAM simply employs capacitors which are leaky, and hence must be refreshed to maintain integrity. Static RAM needs an applied voltage, but doesn't need to be refreshed.

Re:Isn't static RAM dangerous?

[AstroJetson]

I'm not sure if you're being serious or not, but just in case you are, I'll bite.

Static RAM has nothing to do with static electricity. It's called static because if you stop clocking the chip, the contents of the memory remain valid. Dynamic, otoh, requires that you constantly refresh the the contents of the memory or else it will 'forget' what's in it.

The tradeoff is that for a given amount of memory, dynamic is much more efficient (uses fewer transistors) and therefore is smaller. Put another way, you can get more DRAM on a given die size than SRAM. Also it's cheaper. The above two factors are why computers use DRAM for main memory. However, SRAM has it's place. For one thing, it's faster than DRAM and if you must retain the contents of memory when power is not applied, SRAM uses less battery power than DRAM.

If your post was just a joke, disregard the above.

Re:Isn't static RAM dangerous?

[slashdot-me]

In my experience NVRAM (ala Dallas Semiconductor) is usually an SRAM chip with an embedded lithium battery. Flash and EEPROM are usually called "flash" and "eeprom", respectively.

Ryan

Ars Technica /box?

[ethereal]

So why hasn't this made it into my Ars Technica slashbox? It seems like the list of articles in that box stays the same for a month, and then all of the sudden there are 5 new ones in a day. Does /. not update the /boxes very often, or is the problem at the Ars end?

Re:RAM

[The Rock1699]

Quit freakin whining, these articles are the bread and butter of what this community stands for, intelligent computing practices.

Perhaps you knew the whole article already, which makes it quite simple for you to not read it. Or, better yet, since you know so much about the article, give us some good feedback and earn yourself some mod points.

Otherwise, if you want to keep crying, you can kiss my Ars.

Re:memory modules

[Detritus]

Heh!

When we asked for a ROM, they handed you a PC board covered with diodes and a pair of wire cutters.

Bonus geezer points for anyone who knows what a "schmoo chart" is.

Good article, bad web

[Animats]

Good article. I hate the web presentation, though. They spoon-feed you a page at a time, so you have to look at new ads to read the next page. And then it turns out that only the first part is on line. Ars Technica retains the bad features of print media, despite not having a print version.

At least programmers should

[Trinition]

I think every programmer should at least take a course in assembler, operating systems and computer organization. The sort of insight actually can help a programmer.

You know more than you think

[barzok]

You know your car requires oil and coolant, right? You know that when it won't turn over, you probably have a dead battery. You know that if you leave the lights on, you'll run your battery down. You know that the gas goes from the tank into the engine, mixes with air, gets ignited by a spark, and that's what makes the power, right? And I'm sure you're familiar with what happens when you move/adjust each knob, lever and pedal. There are some very basic troubleshooting processes that you know whether you know it or not.

That's a LOT more than the average computer user knows about their computer. They know that the "E" on their desktop is "the internet" and they know how to turn the system on and off. If you opened the hood of your car, I bet you could easily identify a dozen parts, and yet you claim to know knothing about how your car works. Can the average computer user even ID FOUR parts inside their computer?

I believe that the requirements for obtaining a driver's license (at least in my state) are a bit "easy" - at the very least, require that people demonstrate how to change a tire and handle one or 2 other roadside emergencies. Should all of society be denied access to computers because they aren't geeky enough? No. However, with the proliferation of viruses that prey on user stupidity/error, and the increasing toll these viruses take on the rest of the computing world, and the increasing complexity of, well, everything, I think there does need to be a "curb" for people to get over before they can play in the sandbox.

memory modules

[cthulhubob]

I don't know about this newfangled "modular memory". Back in MY day, we had the DIPs soldered onto the motherboard. And we liked it!

That's the problem with you young whippersnappers today. "Modular" this and "DDR" that. When I was kneehigh to a grasshopper, we didn't even have init! We loaded our kernels into memory by breaking into ROM basic and POKEing our 64K of RAM just the right way. Taught us a real valuable life lesson, it did.

if you cant wait for part 2....

[benshutman]

firingsquad did this article a while ago. pretty informative. http://firingsquad.gamers.com/guides/hiwmemory/def ault.asp

Re:hmm. they don't cover ECC RAM

[SirPoopsalot]

I've read all the little comments under this thread, and I want to make a statement of my own here. I am a hardware engineer. More specifically, I am a hardware engineer helping to design a Sparc64 clone. Even more specifically, I am designing the logic for the L3 cache on a Sparc64 clone. This includes (as you might imagine) extensive work with SRAMS and ECC. Enough background... I just wanted to point out that I know excatly what I am talking about. ECC RAM really isn't any different than regular RAM. And, to correct the person who said ECC RAM is redundant... you are wrong. ECC RAM is not redundant... it just keeps a 'signature' for each piece of data. Take a regular RAM (SRAM or DRAM... doesn't matter), and add a few more bits of storage space. 8 bits of storage space for each 64 bits of data, to be precise. Now you have room to hold the ECC bits. Add a little logic that implements a slightly complex algorithm (SEC-DED-S4ED Error Correcrion Code), and you have the ECC part of your RAM. The RAM itself is no different, but you've added logic that "filters" the data as it comes out of the storage array, detecting errors in the data and correcting them if possible. If you *really* want to know more about ECC, there are books galore and websites a-plenty with details on how to implement the correction and detection algorithms used for ECC RAMS. As for the person who claimed that it didn't really make any difference if you had ECC RAMS or not.... Try and find a supercomputer or high-end engineering computer that doesn't use ECC. Just try it! I'll bet my dollar against your dime that you can't. If you have a machine that is running mission critical applications or has some ungodly important function (like in a NASA shuttle or satellite) you can NOT afford to have a seg fault due to the processor fetching an instruction from memory that has one of it's bits flipped. ECC is a #1 Must Have for hard core workstations. As for your home PC... it probably doesn't matter. If I get a segfault, I simply spit and cus and restart the application or (god forbib) the whole machine. Unreal Tournament is hardly "mission critical", even when I'm kicking some serious butt, so I don't feel any need to have ECC RAM in my PC. Just my expert point of view on this. later, Sir Poopsalot