UMD Format

Discuss the development of new homebrew software, tools and libraries.

Moderators: cheriff, TyRaNiD

User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

Klendathu wrote:Measure the total surface of a dvd (only the part where you can write, not the border or the hole in the middle.) you'll find that a DVD stores roughly 42MB per square centimeter.
Did you base this calculation on a single-layer or dual-layer disc? A dual layer disc has rougly twice the storage density than a single layer disc.
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
Klendathu
Posts: 20
Joined: Mon Feb 14, 2005 10:12 pm

Post by Klendathu »

It is based on a 4.7GB DVD-R.
Dual layer does not increase density itself, it "merely" creates multiple layers with same density [double the surface so to speak]. I also supposed the 1.8GB to be single layer.

But it is well in the realms of the possible that games like lumnies are single layer umds, with the same data density as a dvd (wavelenght of laser as an indicator). effectively reducing maximum size to around 870MB

870MB are more than enough for such games as well as for movies encoded in mpeg-4 at full psp resolution. the 1.8 GB thing would then only be a promotion stunt. [remember Dreamcast "1.2GB" discs?] The resulting dual layer UMDs would then be even less accessible outside Sony HQ.
All in all delivering a cheap copy lock by easy means. something which sony seems to like if you remember the PS1 CDs the DVD integration in PS2 or the Blueray discs for PS3. They were typical optical discs which were easy to copy with no read protection and simple authentification tricks. but the required equipment was not widely available.

my guess is that they outfitted a minidisc drive with a standard DVD [single or double layer option scenario] or HD-DVD laser [always single layer scenario]. the only reason the numbers don't match perfectly is due to types of error correction.

on an interesting side note. play ridge racer and suspend to ram after the race starts but before the music starts. when you turn the psp back on, the game will continue instantly but the second the psp accesses the drive to start playing the music the psp will freeze until the disc is authenticated (?). after that the the race will go on but the music will only resume after the disc found the right track (halfway through the first tunnel if you play highway 776). this looks to me like classic IDE behaviour.
one more indication that sony just threw together already existant hardware instead of developing something truely new
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

Klendathu wrote: Dual layer does not increase density itself, it "merely" creates multiple layers with same density [double the surface so to speak].
Yes, but since the layers occupy the same 2D space, the density of the disc, expressed as bits per cm², becomes doubled.
Klendathu wrote: [remember Dreamcast "1.2GB" discs?]
What I don't remember is anyone referring to the as "1.2GB". The GD stands for GigaDisc, implying a capacity of roughly one gigabyte. Since it uses double density, compared to a regular CD-ROM, a capacity of 2*0.66 = 1.32 GB could be obtained, but because there was also space reserved on the disc for a normal single-density region, the area available for the high-density region is less than a full CD, resulting in a slightly lower capacity. I'm not sure what the exact maximum capacity is, but it should be around 1GB at least.

Klendathu wrote: my guess is that they outfitted a minidisc drive with a standard DVD [single or double layer option scenario] or HD-DVD laser [always single layer scenario]. the only reason the numbers don't match perfectly is due to types of error correction.
Yup. Using standard components is the thing to do in order to keep the cost down. The GD drive in the Dreamcast which you mentioned was a standard ATAPI CD-ROM mechanism with custom firmware.

The wavelength specified for the UMD laser suggests a regular DVD laser rather than HD-DVD (which has a wavelength of 405 nm).
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
Klendathu
Posts: 20
Joined: Mon Feb 14, 2005 10:12 pm

Post by Klendathu »

Yes, but since the layers occupy the same 2D space, the density of the disc, expressed as bits per cm², becomes doubled.
NO, two layers of a disc do not occupie the same space. Pieces of paper do not magically disappear when stacked [If only]
just because you can't see them with the eye does not mean they aren't there. therefore not only the ammount of data doubles but also the surface (though it is layered and appears the same size to you)

TWO LAYERS = DOUBLE DATA = SAME DENSITY!!!
X/Y=Z
2X/2Y=Z
100.000X/100.000Y=Z
Guest

Post by Guest »

Klendathu,

I am confused. Could you please explain the flavor of physics that you use to determine this ? Or is your frame of reference that of one layer being just above the event horizon of a black hole, whilst the other layer has fallen into the the aforementioned event horizon ?

Given 1 cm^2, how is the amount of information in a single layer the same density as that in a dual layer ?

Your sheet of paper analogy is wrong, and MC is correct. From a 2D point of view, your analogy of a stack of paper results in an increased density of information (or potential information) per cm^2 as a function of the number of sheets in the stack.
ooPo
Site Admin
Posts: 2023
Joined: Sat Jan 17, 2004 9:56 am
Location: Canada
Contact:

Post by ooPo »

There's no increase per cm^3.

Dual layer, not dual density.

You're both talking about different methods. Though only one would specifically apply to the dreamcast I guess...
Klendathu
Posts: 20
Joined: Mon Feb 14, 2005 10:12 pm

Post by Klendathu »

From human standpiont a dual layer dvd has twice the data desity of a single layer. because we humans see only the disc and the fact that the ammount of data has doubled:

X= Ammount of data
Y= Space
Z= density

human Point of View:
2X/Y=2Z

mc can only see the disc, the amount of data and thus calculates the density incorrectly.


My standpoint is from the technology point of view. I am the laser burning holes into the ground. my wavelength is fixed. It is also fixed how dense I can pack the data on any layer. I have however another option and that is to change my focusing point. if I (the laser) do this, the first layer disappears, and another one comes into my view. Humans might notice that the layers of a dvd are NOT two dimensional, they are above each other like floors of a building. I can wirte in each floor with the exact same density. Since these floors are so minute humans are given the impression that they are two dimensional, when in fact they are not. they are two two-dimensional discs stacked upon each ther. therefore the density remains the same YET THE SURFACE DOUBLES.

the laser can see the same thing an icehockeyplayer can. Both the surface of the ice AND the markings below.

2x/2Y=Z

Image

Learn from the picture that a DVD9 simply has twice the surface. to write on. the laser does not magically improve on its ability to pack data denser together.
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

[quote="ooPo"]There's no increase per cm^3.[/qoute]

Precisely. The density increases only because we're considering the surface area of the disc, not the volume. Nobody has measured the thickness of the UMD disc, or calculated the number of bits/cm³, so the facts that dual layer discs are thicker is largely irrelevant to the calculations.
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
ooPo
Site Admin
Posts: 2023
Joined: Sat Jan 17, 2004 9:56 am
Location: Canada
Contact:

Post by ooPo »

'From human standpiont a dual layer dvd has twice the data desity of a single layer.'

Ahh, now we're getting into the realm of semantics and exploring definitions.

Definately correct, though. :)
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

Klendathu wrote: Learn from the picture that a DVD9 simply has twice the surface.
The thing though is that layer L0 is no longer on the surface of the disc after you glue it together, and so does not contribute to its area. If you take a ruler and measuer the radius of a dual-layer disc, you will get the same measurement as for a single-layer disc. The area of a circle is pi*r², so the area is also the same. Only if you take the disc apart again, giving you two circles, will you be able to measure a larger area. And I don't think you did that with the UMD disc, did you?
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
ooPo
Site Admin
Posts: 2023
Joined: Sat Jan 17, 2004 9:56 am
Location: Canada
Contact:

Post by ooPo »

With hopes of returning this back to the topic at hand:

http://darkwatcher.psxfanatics.com/cons ... cal_v2.htm
GD-ROM

In 1998, Sega released the Dreamcast in Japan. The console used it’s own proprietary disc format designed by Yamaha. These “Gigabyte Discs” were capable of storing 1228.8 megabytes (1.2GB) of data. Although they resemble normal CDs, the GD-ROM contains 4 tracks. The first 2 tracks of the inner ring conformed to the normal “Yellow Book” CD-ROM track, and a “Red Book” audio track. These tracks could hold up to 35MB of data. It is from here where the disc strayed from the norm. A 3rd track contained no data, and was used as a separator. The 4th track on the outer ring stored the actual game data. This outer ring is a high density area that packs the pits and lands closer then on a regular CD. This allows the disc to hold more information, but needs a special drive to read them. The drive is basically built out of normal CD drive components, but has the ability to read at two different speeds (For both the lower density inner ring, and higher density outer ring). This provided the Dreamcast with an extra added means of copyright protection. However the higher density areas were sensitive to scratches, dust, and fingerprints.
It seems that in this specific case the actual density of the layer was increased, not the number of layers...

...but then, don't believe everything you read on the internet.
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

Klendathu wrote: mc can only see the disc, the amount of data and thus calculates the density incorrectly.


My standpoint is from the technology point of view. I am the laser burning holes into the ground.
No, you are the guy measuring that an UMD is 22.66 cm² large, and from this computing that the UMD stores 79.4 MB per cm². No matter how many layers the UMD has, your measurement will still give you 22.66cm².

And you really don't have to explain how a multi-layer disc work, everybody knows this already.

'Nuff said.
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
Klendathu
Posts: 20
Joined: Mon Feb 14, 2005 10:12 pm

Post by Klendathu »

if it had a second layer the "surface that could be written to" would increase to 44cm². since it would be two thin layers of 22 cm² above each other. according to sony 1.8 Gig is maximum size. but suddenly with a two layer 1.8 setup, the density would not be that different from a regular dvd. so all sony would have to do is to hook up a dvd lens to a MD drive and go.
this would greatly improve the chance of hustling some PC compatible device together.
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

ooPo wrote: It seems that in this specific case the actual density of the layer was increased, not the number of layers...
For the GD-ROM, this was the case yes. As I said, a normal CD-ROM drive mechanism was used. The principle is somewhat like a HD floppy. CD-ROM drives are able to read in much larger speeds than nominal, but also spin the disc faster. This is how you get e.g. a "48X" drive. You spin the disc at 48X the speed, and read in 48X the speed. If you instead spin the disc at 24X but still read at 48X, you will read twice the amount of information per length of data track. Of course the disc would have to be mastered this way as well. This works because CD-ROM is a very old standard which has lots of margins not really needed with modern production standards. The same trick would probably not work with a DVD.
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

Also, you can't make a multi-layer CD-ROM as the data layer is non-transparent.
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
ooPo
Site Admin
Posts: 2023
Joined: Sat Jan 17, 2004 9:56 am
Location: Canada
Contact:

Post by ooPo »

I can't imagine any transparent layer being useful to lasers.
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

No, but semi-transparent are.
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
ooPo
Site Admin
Posts: 2023
Joined: Sat Jan 17, 2004 9:56 am
Location: Canada
Contact:

Post by ooPo »

So you can make a multi-layer cdrom, then?
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

Nope. Because the data layer is non-transparent. I.e. it's fully opaque. Short memory?
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
ooPo
Site Admin
Posts: 2023
Joined: Sat Jan 17, 2004 9:56 am
Location: Canada
Contact:

Post by ooPo »

Not at all. I just figured the layer would at least be reflective.

I guess that's why psx games are on totally black discs.
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

Of course. All things which are not either entirely transparent (which, I trust, has now been clarified that the layer isn't :) or completely black (which not even the "black" PS CD:s are) are reflective to some extent. As for the "black" CDs, they have a film on them that filters out all wavelengths _except_ the one used by the laser. I've used CD-R:s with such film (from Traxdata) with my Dreamcast, so they work just like regular CD:s. The laser won't even notice that the film is there.
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

Just to cover all bases, since today seems to be "misunderstand everything-day", it is of course also possible for an object to be non-reflective if it is, say, 50% transparent and 50% black, i.e. 50% of the light is let though, 50% is absorbed, and 0% is reflected. Not that this has anything with CDs to do...
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
ooPo
Site Admin
Posts: 2023
Joined: Sat Jan 17, 2004 9:56 am
Location: Canada
Contact:

Post by ooPo »

So you can put only 50% of the data on that layer, while 50% of the laser burns into the label on the disc?
Victor
Posts: 15
Joined: Wed Jan 26, 2005 12:59 pm

Post by Victor »

like this?
Image
pixel
Posts: 791
Joined: Fri Jan 30, 2004 11:43 pm

Post by pixel »

Ho god :) Do these burners still exist ? :P It was some funky yamaha, wasn't it ?
pixel: A mischievous magical spirit associated with screen displays. The computer industry has frequently borrowed from mythology. Witness the sprites in computer graphics, the demons in artificial intelligence and the trolls in the marketing department.
Klendathu
Posts: 20
Joined: Mon Feb 14, 2005 10:12 pm

Post by Klendathu »

If you took a close look at the picture I posted earlier you would see that there is a difference between the layer which contains the information and the layer that reflects the beam.
one is still 100% reflective (but only for the laser wavelength)
the other is roughly semi reflective.
the layer which stores the data is in any case transparent.

if you took a cd which is "totally reflective" to your eyes and hold it into the sun you would also notice that some light passes through. the same goes for the black or blue discs. your eyes are tuned to frequencies ranging from 360nm to 680nm wavelength. so they will catch some light that can pass through the cd. within the boundaries of the Laser less light passes through. so a CD can still be "reflective" for the laser and look transparent to you or vice versa. but if you put less coating on top of it the sun too will start to damage the cd quicker.
jsut rung your own trials to see what I mean

and by the way, you can boil a cd without damaging it.
User avatar
mc
Posts: 211
Joined: Wed Jan 12, 2005 7:32 am
Location: Linköping

Post by mc »

Klendathu wrote:and by the way, you can boil a cd without damaging it.
But does it make a tasty soup? :-)
Flying at a high speed
Having the courage
Getting over crisis
I rescue the people
Klendathu
Posts: 20
Joined: Mon Feb 14, 2005 10:12 pm

Post by Klendathu »

nope, the effect is quite dissatisfactory
F9zDark
Posts: 127
Joined: Sat Apr 02, 2005 11:34 am

Post by F9zDark »

Well I took apart the UMD Sampler disc that came with PSP. First thing I noticed was that the metal plate in the center of the UMD disc, is rather permanent (perhaps I didn't try hard enough to remove it).

Secondly, the disc is too small to fit into a DVD rom drive. So I aborted the test (don't want to ruin my DVD rom).

Now the only viable reason I see for using UMD on the user level is music and movie data for the PSP. Memsticks are just too expensive (for me anyway) to make the PSP a viable game, movie and music venue. (even though I did put 5 of my favorite songs on the 32mb duo that comes with it).

But on the flip side UMDs are not commercially available for the end user and probably not a plausible medium to burn songs to. 1.8 gigs is a hell of alot of space to fill with songs, especially if it is not rewritable...
DBG
Posts: 12
Joined: Mon Mar 28, 2005 9:08 am

Post by DBG »

Well this discussion has been all over the place, so I don't mind mucking it up a bit myself =). I have to believe a large reason for using UMD is marketing. Yes, it is also small and optical, but it will never be a consumer format, as Sony has already stated. I'm going to pop out a UMD disc (will probably trash Spiderman and the Sampler) and do my best to mount it on a spare drive and mess with the POT enough until I can read something or it smokes... I'll post a follow up if anything comes out of it ;).
Post Reply