DVD memory storage

[Yellow Fang]

I read in a CompTIA training book that the manufacturers of DVDs decided to redefine a gigabyte decimally, i.e. as 1,000,000,000 bytes instead of 2 raised to the power 30, used by the rest of the IT industry. They also tried to redefine the 2 to the power 30 number as a gigibyte. Where do they get off? A byte is computing term after all (although it would make more sense to me to define memory in multiples of bits instead of bytes). Next they'll be saying a byte is ten bits and insisting everyone else calls their eight bit bytes octets or something.

 

[marinyork]

Where do you get these funny ideas from Yellow Fang, you always seem to come out with them. Historically there's nothing particularly wrong with what they are saying although I doubt it's the motivation.

 

[Carwash]

There is nothing new in this at all, and to be honest I'm amazed you're not already aware of it. In general, disk/disc media are sold with 'decimal' kilo-/mega-/giga- prefixes. It is the standard practice. IEC have tried to fix the confusion that results from these two parallel uses by stating that decimal units always use kilo-/mega-/giga- as it is in all other measures, and binary units use kibi-/mebi-/gibi-. This has met with mixed success... although the usage has become standard in some areas (look for the abbreviations KiB/MiB/GiB).

 

[Yellow Fang]

It's not a standard practice where I come from. We've always understood kBs, MBs and GBs to be multiples of 2. I had never even heard of a Gigibyte before. It's always been a handy coincidence to know that 2 to the 10 is about 1000 (kB), 2 to the 20 about 1000,000 (MB) and so on.

 

[Carwash]

It's not a standard practice where I come from. We've always understood kBs, MBs and GBs to be multiples of 2. I had never even heard of a Gigibyte before. It's always been a handy coincidence to know that 2 to the 10 is about 1000 (kB), 2 to the 20 about 1000,000 (MB) and so on.

Did you never wonder why for example, when you buy a 250Gb hard drive, it doesn't actually hold 250Gb?

 

[Yellow Fang]

The only thing I've wondered at so far is why my 4Gb memory stick has only about 3.7Gb on it when I stick it in my PC.

 

[Carwash]

The only thing I've wondered at so far is why my 4Gb memory stick has only about 3.7Gb on it when I stick it in my PC.

That's the same thing. See, you had noticed it really.

(Edit: It's probably partly down to filesystem overhead as well, just to be clear.)

 

[marinyork]

Anyway the other part of the history apart from SI confrontations is that 8 bit bytes have been called octets and that a byte was far from universally being 8 bits, that you for some reason apparently seem to think is the case. You've never wondered where the 8 bit byte came from? See, you'd already seen these issues in your thinking.

 

[Yellow Fang]

I'm aware there have been different sized bytes, but that was all ancient history by the time I finished college. They've all be eight bits long since I started work, although I recall seeing some protocol documents which talked about seven data bits and a parity bit, and sometimes extra stop bits as well.

It still seems pretty off to me that the DVD industry would take it upon themselves to re-define an established measurement, no matter what the IEC say. Now there is more confusion. It was fine when DVDs were just used for playing media, but now they are used as memory storage devices. Old computer systems engineers needed the powers-of-two memory sizes to work out the number of address lines and chip select lines they needed. I bet on our old 16k Spectrum ZX computer with the 48k ram pack, Clive Sinclair was giving us a bit more memory than he said he was, unlike the manufacturers of modern memory devices.

This seems like another clash between the broadcast and IT industry to me.

 

[Carwash]

It still seems pretty off to me that the DVD industry would take it upon themselves to re-define an established measurement, no matter what the IEC say.

They haven't. They're doing exactly what everyone else has been doing for years and years.

 

[marinyork]

I don't see the controversy. Other people have been doing it for years and years and years.

 

[Carwash]

I don't see the controversy.

There isn't one. He's trying to invent it.

 

[Yellow Fang]

Yes, I'm seeing it everywhere now I think previously I assumed some of the memory was inaccessible. I still think they, along with other memory storage manufacturers, are pulling a bit of a fast one. I have four memory sticks:

1. 1,010,548,736 bytes, 963MB
2. 4,003,463,168 bytes, 3.72GB
3. 1,002,160,128 bytes, 955MB
4. 4,068,872,192 bytes, 3.78GB

All cynically just over the magic round number in terms of number of bytes, yet not living up to what it said on the packet. Alright, I can't blame the DVD manufacturers if other storage device manufacturers were doing the same thing.

 

[Carwash]

I still think they, along with other memory storage manufacturers, are pulling a bit of a fast one.

I think there could be a case for that, yes. They've chosen to use the system which paints their products most favourably, rather than the one in more common usage.

 

[Yellow Fang]

It's sort of symptomatic of how fast computer technology advances. I've studied computer systems before. My last Open University module I studied in 2003 was Inside Electronic Devices, but I don't really know how modern mass storage media works any more. I would have assumed that memory sticks worked like the flash EPROMS I used to blow, which all had memory spaces of multiples of two. With modern memory devices, it seems that the simpler circuitry and presumably faster access times are not worth the relatively small amount of extra memory they have to provide. Considering how cheap they've become, that's really odd.