DVD-RAM is stupid

Well, the terminology is stupid even if the technology is cool.

“RAM” is not “everything that isn’t ROM”!! CD-ROMs and DVD-ROMs are accurately named because the information on them is read-only.

But “RAM” means “random-access memory”, which implies that the information within the space so designated is allocated on-the-fly (randomly) and accessed as need be, sort of a de facto work space in which one does one’s projects. RAM as we know it fits this description. “DVD-RAM” – recordable DVD media – most certainly does not. Even if it were always rewritable (and I think most of it is simply recordable on a one-time basis like CD-R), there is nothing “on-the-fly” about a storage space that must essentially be reformatted to wipe out its previous contents before putting something new in that space. And if (as I think is the case) most
“DVD-RAM” mechanisms merely write data on a one-time basis to the media, there is nothing random about it either: you choose what to put into the space, then you write it there, then there it is, forever.

By the way: do they, as of yet, support the burning of conventional 650 MB / 74 min CDs, or do most of them only read CDs and in burn mode can only burn DVD media? I think I’d swallow my annoyance at naming conventions and buy one if it could burn either 4+MB DVD or 650 MB CD media as need be.


Disable Similes in this Post

I think you’ve got it almost right.

DVD-RAM, as presently constituted is stupid.

However, once the RIAA and the movie studios give in to progress (the same sort of blocks were used to keep CD-R unavailable for quite a while), DVD-RAM will kick ass.

Cartridge-based media will go away
The 5.2GB limit will go away
Media will get cheaper
Drives will get cheaper
Speeds will improve
I have not heard of any CD-R/DVD-RAM drives, but I haven’t been looking all that hard. Or at all, really, as I’m waiting for the above improvements before I adopt the tech.

A committee is a lifeform with six or more legs and no brain.

Jo3sh (the 3 is silent, you see):

Are you saying that writable DVD’s are artificially limited to 5.2 gigabytes due to pressure from the entertainment industry?

Aren’t all cd and dvd media RAM?
That would really suck to have a Sequential-Access CD. What if you wanted to read a file stored at the outer rim of the disk? You’d have to scan through the entire contents to find it.

I think they should be called DVD-WORM (write once, read many). And rewritable CD’s should be called CD-EPROM.

And, uh, Jo3sh? You either missed or ignored the entire original post.


If you say it, mean it. If you mean it, do it.
If you do it, live it. If you live it, say it.

Joe Cool

Joe_Cool writes:

Technically, no. For CD based media, as in all disk based media, you first seek to a specified track, then read serially from the track. It’s not random access. Contrast this with true random access memories where there is one unique address that maps to a particular row and column in a memory array.

Now a related question is, can we refer to the dynamic memory in most PCs as RAM? I suggest not, technically, since they use a paging mechanism to overcome the limitations of the original Wintel architecture.

By the way, AHunter3 - I agree with you - DVDRAM is a dumb and misleading name.

Not directly, but certainly the movie studios and such feel they have an interest in slowing the development of the technology.


A committee is a lifeform with six or more legs and no brain.

Not quite true. I picked and chose. I didn’t respond to “the middle bit” as Otto might say, since it struck me as scemantic quibbling, but here goes:

RAM *is[/] pretty nondescriptive of the writability of the format. I called it DVD-R for quite a while, in parallel with CD-R(ecordable). But there were a number of standards developed by several different companies as writable DVD’s came to be (think VHS vs. Betamax), and the one that seems to be winning was dubbed DVD-RAM.


A committee is a lifeform with six or more legs and no brain.

Why yes. Yes, it is. It’s the New Math.


A committee is a lifeform with six or more legs and no brain.

FWIW, the data capacity of my wife’s DVD copy of Shakespeare In Love is 7.74GB. i have no idea if this is typical or not.


A committee is a lifeform with six or more legs and no brain.

[ Lurk Mode OFF ]

Well i am not quite sure what DVD-RAM you guys are talking about, but the DVD-RAM drive in my Mac acts exactly like a Hard Drive, with a limit of 5.2 gigs. So here DVD-RAM is a “correct” name, though i suppose they could come up with a better one.
The limit for a DVD-ROM, made by a DVD producer is 17 gigs. That is a double side, double layer disk.
And you will probably never see a combined DVD-R/CD-R because the laser they use are different sizes. The DVD laser is much much smaller, so there are more tracks on a DVD, thus more space. To make a combined DVD-R/CD-R you would need two different writing lasers which would make a rather large and cumbersome unit. Of course what you guys are talking about and what i am talking about may be something different.

Here are a couple of links to Apple’s descriptions: http://til.info.apple.com/techinfo.nsf/artnum/n60235 http://til.info.apple.com/techinfo.nsf/artnum/n58187
Hope this helps

-N
[ Lurk Mode ON ] http://til.info.apple.com/techinfo.nsf/artnum/n24451


“Oh my God! Space aliens! Don’t eat me, I have a wife and kids! Eat them.”
Homer Simpson

man i really screwed up the format at the bottom of my message :mad:

-N


“Oh my God! Space aliens! Don’t eat me, I have a wife and kids! Eat them.”
Homer Simpson

Strider,

You wrote:

You lost me there. Do you think hard disk media and RAM are equivalent? Surely you don’t believe this… However your characterization of DVDRAM as being functionally equivalent to a hard disk is probably reasonable.

RAM means Random Access Media/Memory. Compare to serial/linear access, as with tapes. Nothing says you have to read a large ammount of linear data once you get there. Because a hard-drive can seek to any given spot without having to pass the heads through all locations in between, it is random access. Ditto with CD and DVD media. And Memory is vauge, but I haven’t seen anything which suggests that only dynamic ram, or other volatile storage media applies.

Also, a weird paging scheme has nothing to do with RAM. A page just means that there is a page counter and an offset counter. Both need to be changed to point to all addresses. Neither is limited to being incremented or decremented; you can directly store any desired value. Thus even with a weird paging system, it’s still RAM.

There is a middle-ground. Imagine a series of 1K (2^10) punched cards, each with 1KB of data on it. If these cards have to be shuffled through in order, to find the correct card, but by knowing the address of the data on the card, the read/write head can position directly to the byte desired, is this random or linear access?

I ask this because everything is a variation on this. In disks, the head has to be moved, and if it goes from track 1 to track 15, it crosses 2, 3, 4, …, and 14 before getting there. And if it’s looking for sector 12, and the head gets to the track on sector 3, the device waits for the right sector to rotate around. There’s obviously a linear component to this, but it lets you skip fairly quickly to the desire element.

RAM can even have some linear elements to it, usually because of optimization. Often a read request for one byte results in between four and sixteen bytes being read, then the correct byte being pulled out and used. The requested address is random, but often sequantial reads are faster, because the memory can be told to fetch the next set of bytes rather than given a whole new address to read from.

So, if you nitpick enough, nothing is truly random access unless the heads teleport to the proper place, and all data is arranged so that individual bits can be read just as quickly from any area of the device as any other. Or, you can say that anything which doesn’t make you wait while it rewinds through 1500m of tape is random.

I tend to support the latter view.
Then, onto the issues of speed/size.

Yes, the RIAA is trying to block writable DVDs, because they allow copying movies. They make it seem like only movie pirates require this kind of storage, but then you see the many legitimate uses of VCRs and writable tapes… And this is worse, because VCR tapes were primarily video only, a writable DVD will be useful for much more than just copying movies. I have 4-5 gigabytes of old .zip files, work I’ve done for various clients, which I’d love to archive onto one DVD instead of many CDs; I’m sure many people have legitimate uses for large removable media.

The capacity will get higher. Currently I believe one layer on one side is writable, the pre-written DVDs can have two layers on each side. When writable DVDs become more common the manufacturers will have incentive to produce the higher capacity writable disks.

And, as more people buy the disks, production will increase and the economies of scale the computer industry is used to will kick in, making the disks eventually about the price of CD blanks now.

Either that, or another technology will overtake DVDs before they really take off…

btw, RAM doesn’t necessarily mean rewritable. A WORM disk can be RAM, in that you can read and write in any fashion, but of course, can only write to each area once.

WhiteNight writes:

Not true. Hard disks and CDs access data by first moving the read head to the right track, then reading all the data on that track. Neither one of these operations is random access. The read head positioning is accomplished by incrementing a stepper motor (a serial operation) and the track reading is serial, as well - the disk is spinning at a fixed rate. It’s fast, but it’s not random access. Your later paragraph seems to indicate that you understand this…

I agree that each page can be considered ramdom access, but since you can’t place machine instructions and data that crosses page boundaries, I don’t think you can consider paged memory as truely random access.

Pipelines and cache are mechanisms that can reduce the effective access time. Just because the processor chooses to fetch more data than was minimally necessary isn’t a factor in deciding whether or not a memory is random access or not.

RAM is truely random access. IC ROM, EPROM, EEPROM, and FlashROM are all random access, as well. However, by convention the acronym, RAM, has always been assumed to be Random Access (write many times / read many times) Memory.

By Jove, I think he’s got it. Consider the kind of memory that we’ve always called RAM, up until this travesty of nomenclature… Ignore Wintel memory because it has to do some hardware paging. Let’s instead, look at a 68030 computer - it’s old technology, but not complicated by sophisticated memory management schemes. The address bus is 32 bits wide which means that the address space is 4 GBytes. Now most 68030 based computers don’t have that much memory available to them, but if they did, the access time for any byte would be identical (within a few picoseconds) of any other byte. The mechanism to access each byte is, simply put it’s 32 bit address on the bus and deassert the R/W line - presto, your data appears on the data bus in the next cycle. That’s Random Access Memory and that has been Random Access Memory for the last 5 or 6 decades… why we technologists are letting some marketing weenies redefine the term now is beyond me.

No, really, you did miss the entire message of the OP. The OP said (summarized) “this technology is cool, but it has a stupid name”. Your post said that the technology, as it is currently presented, is stupid. The OP never said the technology was stupid. The OP said the technology was cool.



From an actual catalog: “Disco balls create an enchanting, dazzling effect of light shafts, adding movement and glamour to any occasion”
the Abrams’ bris was certainly memorable
O p a l C a t
www.opalcat.com

I think the head movement is a random access feature. The head can cross the disk in very little time (twice the average access latency usually), if it was reading the data in between, it’d take minutes. An example would be a record player, where the needle has to follow the single groove all the way. That’s linear access. But if you were to pick up the needle and reposition it, that would be random access. The HD skips the inbetween tracks, just like moving the needle, but because this doesn’t require a hand to lift it, this isn’t random access?

Also, there are two disk formatting methods available. One is soft sectoring, where sectors are marked by data on the drive, and to find a sector, you have to read the track data as it passes. Hard sectoring (what that hole in the hub of a floppy was for) means the drive knows right where a sector is and doesn’t have to read until that sector is under the head. The fact that the drive will usually start reading as soon as it gets to the track is an optimization issue, because if one piece of data is read, likely the ones on the same track will be read fairly soon. So the HD access is random, as far as head movement (it moves quickly to the right position, without reading intermediate data) and serial as far as the rotational latency. It’s one of those mixed cases I mentioned, but I feel it leans much more heavily towards random access.

I think that’s a bit irrelevant. Especially since the page boundaries can be changed in x86 architecture. All the page boundary system ended up doing was allowing short pointers to use less ram and be manipulated faster. And all random access means is being able to access randomly, it doesn’t guarantee a data structure will fit somewhere, or that another device will like it where it is. MS-DOS only used 1MB of RAM without modifications, does this mean that the other 63MB in my machine weren’t RAM, just because the OS, and the CPU in an older mode, didn’t choose to use it in a certain way?

Modern RAM is the same, but there are optimizations to it. EDO RAM is like regular RAM, but where extra linear accesses are cheaper. You can use EDO like regular, ignoring the benefits, or you can try to organize data to take advantage of the special features. Either way, it’s RAM, it’s just RAM with special linear access optimizations.
I maintain that there are random and linear access methods in nearly everything, but that any media where to get from X to X+N doesn’t involve actually going past N addresses is random. So for me, almost everything that’s not tape, is random access.

I did and still do disagree with the OP. This is, after all, GD, isn’t it? The name is, to me, pointless to talk about, since it’s only a tag. You could call it DVD-TINKY-WINKY, for all it matters. My point is that there are significant technological shortcomings to the format at the present time. The technology has enormous potential coolness, but it’s not there yet.

A committee is a lifeform with six or more legs and no brain.

OK. Let’s try this one more time.

WhiteNight wrote:

The head movement is not random access. If it were, the seek time for each track would be the same. It’s not, that’s why hard disk optimizers try to place frequently accessed data close to the spindle.

However, since the disk is spinning with a constant RPM, there is variant access time from when the head reaches the right track to when the relevant data is read. Again, I use disk optimization to prove my point. Defragmenting files improves performance because it cuts down on the head movement and the sector skipping.

Think of this example. Let’s say you have a huge amount of data in a single file that you want to store on a hard disk. You choose two methods of storage. The first method stores all the bytes sequentially, starting at the spindle. The second method scatters fragments of the file in random tracks and sectors on the disk. The read time for the first method will be significantly faster than the read time for the second method. If you repeated this experiment with a big chunk of true RAM, there would be no difference in read times from the sequential file and the scrambled file.

Ahh… but it’s not faster. An x86 processor, operating at a given bus frequency, will take slightly longer to read
a 10MB file from RAM than a 68K processor, operating at the same bus frequency. That’s because the x86 processor has to change pages a number of times, where the 68K processor does not.

Perhaps this is the best test of Random Access-ness. If all the data has the same access time, independent of its location in the media, it’s RAM. If access times vary according to where the data resides in the media, it’s not RAM.

Well, I for one, don’t like your definition… but I welcome you to maintain it if you want…

Well, unless you are unusually blessed, you’ve found yourself confronted with some newbie or technoklutz asking “I got an error message saying INSUFFICIENT DRIVE SPACE when I tried to install Bloatoshop. I guess I need to install more RAM, huh?” or “It said OUT OF MEMORY so I deleted a everyone’s old files to make more room”.

So time and again you have tried to explain the difference between memory and storage space. Both are expressed in KB, MB, GB units, which causes the confusion, but insufficient space means hard drive, insufficient memory means RAM.

And as if things weren’t difficult enough in this explaining venue, along comes DVD-RAM. Great. Now I’ll have to open the box itself and point to the little printed-circuit critters, I guess.

It’s a stupid name for the thing. There is no argument that supports recordable DVD as “RAM” that could not be applied (with equally convoluted logic of course) to regular DVD-ROM (“It reads it randomly, it isn’t forced to read the parts you didn’t ask for”)

Maybe I’ll get one anyway. The prospect of having 1 object the size of a CD in my CD wallet that holds the essential core of my entire music collection in MP3 format sounds awfully nice, especially since my PowerBook could read the sucker if I had it.


Disable Similes in this Post