Back to the main page

Mailing List Logs for ShadowRN

Message no. 1
From: Jan-bart van Beek <flake@***.NL>
Subject: Some question on memory.
Date: Mon, 11 Sep 1995 14:49:13 +0200
One of my players has recently designed a charater with skill-wires and
was looking for a way too have a lot of skills available at any time.
And we are talking a lot he has bought over 1200 mp of skillsofts.

One of his plans was to buy a wrist computer and put all of the 1200 mp
into the thing and download it to his skill-wire when he needed it.
Can he do this, can skillwires direcly acces external memory. I would
think so. It can access the data on a skillsoft chip, so why not the data
in a standard optical computer chip.
What would be the system load delay time, or should you use the IO speed
of the computer. What is the standard IO speed of a computer.

Another question that came up was, why cyber memory should cost so
godawful lot of essence. I believe a one gigapulse optical chip is no
bigger then a thumbnail, so why does it cost aproximatly 3.3 essence
points to install. Couldn't you just such a chip in a cybereye or some
other more bulky body part, where it's presence wouldn't interfere with
the human biological and aural functions.





--------------------------------------------------------------
| Beware of what you ask for you may recieve it |
--------------------------------------------------------------

**** The Cornflake Killer Strikes again ****
Message no. 2
From: Marc A Renouf <jormung@*****.UMICH.EDU>
Subject: Re: Some question on memory.
Date: Mon, 11 Sep 1995 13:32:50 -0400
On Mon, 11 Sep 1995, Jan-bart van Beek wrote:

> Another question that came up was, why cyber memory should cost so
> godawful lot of essence. I believe a one gigapulse optical chip is no
> bigger then a thumbnail, so why does it cost aproximatly 3.3 essence
> points to install. Couldn't you just such a chip in a cybereye or some
> other more bulky body part, where it's presence wouldn't interfere with
> the human biological and aural functions.

Technically speaking you are corect. The actual storage space
*is* fairly small. However, you can't really isolate it from the brain's
functions because that's its whole purpose. Especially when you
encorporate things like datasoft links and encephalons, you get into the
realm where the computational processes *must* be integrated with the
brain or the thing doesn't work.
Additionally, you have all of the hardware designed to read and
write to the memory, which will also take up space. I don't know about
you, but I would think that a laser and some spinning optical disks in my
head would get downright obnoxious as far as spacial use was concerned.

Marc
Message no. 3
From: Dustin Wood <cukoo@*****.NET>
Subject: Re: Some question on memory.
Date: Mon, 11 Sep 1995 18:10:58 -0700
>On Mon, 11 Sep 1995, Jan-bart van Beek wrote:
>
>> Another question that came up was, why cyber memory should cost so
>> godawful lot of essence.
>
> Technically speaking you are corect. The actual storage space
>*is* fairly small. However, you can't really isolate it from the brain's
>functions because that's its whole purpose. Especially when you
>encorporate things like datasoft links and encephalons, you get into the
>realm where the computational processes *must* be integrated with the
>brain or the thing doesn't work.
> Additionally, you have all of the hardware designed to read and
>write to the memory, which will also take up space. I don't know about
>you, but I would think that a laser and some spinning optical disks in my
>head would get downright obnoxious as far as spacial use was concerned.
>
Although this seems like sound rational to me, I doubt that is what FASA or
the SR designers had in mind, I think they were probably looking more toward
game balance, especially where storing skillsofts is concerned. I do
however have another question..... According to Shadowtech, a standard
Datajack has a DFR of 25mps, in order to work effectively, the DFR on a jack
connected to a set of skillwires for an even semi-active skill or any skill
requiring any sort of quick reaction would have to be horrendous. I think
the rational here is a little weak.

In order for a chipjack to work effectively for skillsofts, it has a
horrendous DFR, so in theory, they could apply the same technology to the
DFR for Datajacks right? It is only a I/O for data unless it is connected
to other hardware.
Message no. 4
From: Marc A Renouf <jormung@*****.UMICH.EDU>
Subject: Re: Some question on memory.
Date: Tue, 12 Sep 1995 00:13:19 -0400
On Mon, 11 Sep 1995, Dustin Wood wrote:

> In order for a chipjack to work effectively for skillsofts, it has a
> horrendous DFR, so in theory, they could apply the same technology to the
> DFR for Datajacks right? It is only a I/O for data unless it is connected
> to other hardware.

I think the difference here lies in Data Flow Rate (DFR) versus
Softlink System Load Delay (SSLD). As far as I know, the data for
skillsofts does not "flow" per se. It's like having a program loaded
into your computer's RAM. Now in the case of a datajack, consider the
following. Your computer may be horrendously fast (a Pentium running at
89.9998 MHz :) but if your modem is only at 300 baud, you're screwed. It
was only until recently that exceedingly fast communication between
computers caught up with a computer's intrinsic processing speeds. Thus,
it may be that technology in SRII is at that same stage (albeit at a
higher point in the tech curve). So the datajack is the same way. The
softlink is analogous to a program already running in RAM, while the DFR
of the datajack is purely an I/O device limitation.

Marc
Message no. 5
From: Dustin Wood <cukoo@*****.NET>
Subject: Re: Some question on memory.
Date: Mon, 11 Sep 1995 21:26:45 -0700
>On Mon, 11 Sep 1995, Dustin Wood wrote:
>
>> In order for a chipjack to work effectively for skillsofts, it has a
>> horrendous DFR, so in theory, they could apply the same technology to the
>> DFR for Datajacks right? It is only a I/O for data unless it is connected
>> to other hardware.
>
> I think the difference here lies in Data Flow Rate (DFR) versus
>Softlink System Load Delay (SSLD). As far as I know, the data for
>skillsofts does not "flow" per se. It's like having a program loaded
>into your computer's RAM. Now in the case of a datajack, consider the
>following. Your computer may be horrendously fast (a Pentium running at
>89.9998 MHz :) but if your modem is only at 300 baud, you're screwed. It
>was only until recently that exceedingly fast communication between
>computers caught up with a computer's intrinsic processing speeds. Thus,
>it may be that technology in SRII is at that same stage (albeit at a
>higher point in the tech curve). So the datajack is the same way. The
>softlink is analogous to a program already running in RAM, while the DFR
>of the datajack is purely an I/O device limitation.
>
My point lies in the fact that both are from external sources. Why could
information flow into your head faster from the matrix than from a chip
plugged into your skull? This is saying that the character doesn't have
headware memory, in which case we'll say he has a cranial cyberdeck. Even a
fully internal system, using headware memory as storage memory, uses a load
speed before he can run a utility.

Even though he must force the matrix to execute his utility (using the
systems security code) the program is still running on his deck.
Message no. 6
From: Marc A Renouf <jormung@*****.UMICH.EDU>
Subject: Re: Some question on memory.
Date: Tue, 12 Sep 1995 00:55:51 -0400
On Mon, 11 Sep 1995, Dustin Wood wrote:

> My point lies in the fact that both are from external sources.

But they're not. The way I view Softlink System Load Delay is as
the time it takes the softlink system to sift through the garbage and load
the appropriate information from the chip into the skillwire system's RAM

> Why could information flow into your head faster from the matrix than
> from a chip plugged into your skull?

That's the point. It doesn't Skillsoft information is already
loaded and is ready on demand. Data from the Matrix via a datajack has
to be processed (as an I/O function, coding, decoding, et cetera) and
thus will be slower. Just how much slower is the only question.

> This is saying that the character doesn't have
> headware memory, in which case we'll say he has a cranial cyberdeck.

Why would you have a cranial cyberdeck without headware memory?
What would be the point?

> Even a fully internal system, using headware memory as storage memory,
> uses a load speed before he can run a utility.
> Even though he must force the matrix to execute his utility (using the
> systems security code) the program is still running on his deck.

Exactly. The program is running on *his* deck. so when he needs
to laod a program from storage to active memory, he is stuck with the
System Load Delay (SLD). Note that this is similar to but not the same
as the Softlink System Load Delay (SSLD). In either case, however, it
merely reflects the time it takes to take the program and get it ready to
run.
It think what you're getting hung up on is the datajack. All the
datajack is for is to let info flow back and forth. If you make the
assumption that even the slowest datajacks are capable of carrying
information fast enough for a simsense feed, then the only thing your
datajack limits is how fast you can upload or download raw data to the
Matrix. In this case, it is directly analogous to a modem. The faster
the modem, the more info you can grab in the same amount of time. It
doesn't affect how fast your programs run at all.
Am I making any sense?

Marc
Message no. 7
From: Dustin Wood <cukoo@*****.NET>
Subject: Re: Some question on memory.
Date: Mon, 11 Sep 1995 22:09:58 -0700
>On Mon, 11 Sep 1995, Dustin Wood wrote:
>
Let's try a different analogy. If you compare a program, say an attack
program, to another program, this one written to run on skillwires through a
skillsoft system, say firearms, both have to be interpreted by the hardware
(processors), right? And have to have room in which to work (RAM) in one
case the ram is active memory, in the other it is a chip (or headware
memory). In the case of a chip, all that data still has to flow through the
chipjack.
Message no. 8
From: Gurth <gurth@******.NL>
Subject: Re: Some question on memory.
Date: Tue, 12 Sep 1995 10:50:17 +0200
>One of his plans was to buy a wrist computer and put all of the 1200 mp
>into the thing and download it to his skill-wire when he needed it.
[snip]
>What would be the system load delay time, or should you use the IO speed
>of the computer. What is the standard IO speed of a computer.

I ran across a similar case some time ago, and I wish I could remember how
we solved it... Let's see, it was a character with skillwires nor chipjack,
but with a datajack. He gets a Linguasoft from somewhere, and wants to run
it. Look it up in SRII, the only chips you can plug into a datajack are
Datasofts (though if anyone ever tries to do that I'll probably tell them
"no" anyway). What I think we did was say that he could plug the chip into
his pocket computer, and plug the computer into his datajack, but we gave
him the SLD every time he wanted to access the chip, instead of only when
you just plug in a new chip.

--
Gurth@******.nl - Gurth@***.nl - http://www.xs4all.nl/~gurth/index.html
And a sense of adventure
-> Unofficial Shadowrun Guru & NERPS Project Leader <-
-----BEGIN GEEK CODE BLOCK-----
Version 3.1:
GAT/! d-(dpu) s:- !a>? C+(++)@ U P L E? W(++) N o? K- w+ O V? PS+ PE Y PGP-
t(+) 5 X R+++>$ tv+(++) b+@ DI? D+ G(++) e h! !r(--) y?
------END GEEK CODE BLOCK------
Message no. 9
From: Axel Strack <strack@***.TU-FREIBERG.DE>
Subject: Re: Some question on memory.
Date: Tue, 12 Sep 1995 13:36:17 MESZ
>
> >On Mon, 11 Sep 1995, Dustin Wood wrote:
>
>(* a lot of interesting stuff sniped *)

As far as i remember there is a passage in the book which says the chipjack is
a specialized datajack (or the the datajack is more general than the chipjack).
It should be somwhere in the equipment description part of the book. If that`s
true it could mean that the datajack has a broader bandwidth than the chipjack
which would account for the lower transferspeed. But i`m only guessing ...

Kid Flash
strack@*******.tu-freiberg.de
Message no. 10
From: Marc A Renouf <jormung@*****.UMICH.EDU>
Subject: Re: Some question on memory.
Date: Tue, 12 Sep 1995 12:40:30 -0400
On Mon, 11 Sep 1995, Dustin Wood wrote:

> Let's try a different analogy. If you compare a program, say an attack
> program, to another program, this one written to run on skillwires through a
> skillsoft system, say firearms, both have to be interpreted by the hardware
> (processors), right? And have to have room in which to work (RAM) in one
> case the ram is active memory, in the other it is a chip (or headware
> memory). In the case of a chip, all that data still has to flow through the
> chipjack.

No. The data doesn't "flow" through the chipjack. It's just
sitting there. That's like saying that data "flows" through your disk
drive. The chip is only storage. The actual part that is the "RAM" in
this case is either your skillwire system in the case of active softs or
your encephalon or datasoft link in the case of knowsofts. The chipjack
itself is merely a receptacle to read the data off the chip and keep it
safe during use. It doesn't do any real computation of its own. This is
important. The chip is not *running* the program. The skillwires are.
To illustrate the point, a softlink III by itself is totally
useless. Without skillwires (for active skills) or and encephalon or
datasoft link (for knowledge skills) you have no way to access the data
on the chip in a useful fashion. A display link lets you view raw chip
data in your field of view, but that's it. So the softlink itself is
basically just a floppy-drive in your head.
The confusion may result from the fact that originally, the
encephalon was not part of the overall equipment in SRII. Thus, when
Shadowtech came out, they said that the old chipjacks had a rudimentary
"level 0" encephalon built in. That's probably why they cost more
Essence than a 1-port softlink. But either way, its that 0-level
encephalon that's actually doing the work, not the chip. When you stick
your brand new CD-ROM game into your CD-drive, the CD is not running the
program. Your computer is. Skillsofts work the same way. The chip is
not doing the work, the skillwires or encephalon are.
Does *this* make any sense?

Marc
Message no. 11
From: Dustin Wood <cukoo@*****.NET>
Subject: Re: Some question on memory.
Date: Tue, 12 Sep 1995 15:48:55 -0700
>On Mon, 11 Sep 1995, Dustin Wood wrote:
>
>> Let's try a different analogy. If you compare a program, say an attack
>> program, to another program, this one written to run on skillwires through a
>> skillsoft system, say firearms, both have to be interpreted by the hardware
>> (processors), right? And have to have room in which to work (RAM) in one
>> case the ram is active memory, in the other it is a chip (or headware
>> memory). In the case of a chip, all that data still has to flow through the
>> chipjack.
>
> The confusion may result from the fact that originally, the
>encephalon was not part of the overall equipment in SRII. Thus, when
>Shadowtech came out, they said that the old chipjacks had a rudimentary
>"level 0" encephalon built in. That's probably why they cost more
>Essence than a 1-port softlink. But either way, its that 0-level
>encephalon that's actually doing the work, not the chip. When you stick
>your brand new CD-ROM game into your CD-drive, the CD is not running the
>program. Your computer is. Skillsofts work the same way. The chip is
>not doing the work, the skillwires or encephalon are.
> Does *this* make any sense?
>
OK, I think I've got it. You have probably just explained to me what others
before you have tried and failed, and thanks for explaining in laymans terms.
Message no. 12
From: Marc A Renouf <jormung@*****.UMICH.EDU>
Subject: Re: Some question on memory.
Date: Tue, 12 Sep 1995 20:25:10 -0400
On Tue, 12 Sep 1995, Dustin Wood wrote:

> OK, I think I've got it. You have probably just explained to me what others
> before you have tried and failed, and thanks for explaining in laymans terms.

No problem. If you have anymore questions, feel free to ask.
*Someone* on this list is bound to be able to explain whatever it is you
are asking.

Marc
Message no. 13
From: Kelly Martin <kelly@*******.BLOOMINGTON.IN.US>
Subject: Re: Some question on memory.
Date: Tue, 12 Sep 1995 21:53:30 EST5
"Axel" == Axel Strack <strack@***.TU-FREIBERG.DE> writes:

Axel> If that`s true it could mean that the datajack has a broader
Axel> bandwidth than the chipjack which would account for the lower
Axel> transferspeed. But i`m only guessing ...

in general, the broader the bandwidth of a data channel the _faster_
you can jam data thru it. that's why ethernet is 10 megabits/second,
and a void phone line is at best some 20 kilobits/second: the phone
line is 22 kHz and the ethernet cable 20 MHz (or thereabouts).

k.
--
kelly martin <kelly@*******.bloomington.in.us>

A duck is a duck. A dog is a dog. And a cat is a person.
-- The Dick Van Dyke Show

Further Reading

If you enjoyed reading about Some question on memory., you may also be interested in:

Disclaimer

These messages were posted a long time ago on a mailing list far, far away. The copyright to their contents probably lies with the original authors of the individual messages, but since they were published in an electronic forum that anyone could subscribe to, and the logs were available to subscribers and most likely non-subscribers as well, it's felt that re-publishing them here is a kind of public service.

Thwap!