Sammy, I am beginning to come around to your way of thinking

If it's technically suppose to work, but doesn't give you a
performance boost of any kind, is it worth even talking about SLI, or
can you honestly say that the application supports SLI? All I'm
reading says that SLI is doing nothing or making FSX perform worse.
Perhaps it's a driver issue or a profile issue. I don't know. Here's
an example.

http://forums.nvidia.com/lofiversion/index.php?t22367.html

Do you have 2 video cards? Perhaps you can post some benchmarks with
SLI enabled and with SLI turned off.

I played with SLI on my computer when I first got it. I can't remember
if I got it working with FS2004 or if it crashed, but I could no
longer use multiple monitors with SLI enabled so I was no longer
interested. It's just not an acceptable tradeoff. For all practical
purposes SLI is not the way to go for flight simulator.
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

Mxsmanic wrote:

> > Mind you, the big problem comes when we have to move to IPv6 -
> > March 2010 I believe is the date set for when IPv4 runs out.
>
> And March 2020 will probably be the day that IPv6 runs out.

Wow! That is a strong statement to make. Can I ask if you have any
evidence to subtantiate this. Or, are you going to sit on the fence
with the good ole word - "probably".

Let's take your one-line statement into context.

IPv4 - 32bit, has 4,294,967,296 possible unique addresses. Due to
*mistakes* made by *engineers* (I would actually say mistakes by
laypeople, but seeing as *engineers* are in your scope, we will stick
with that for the time being).

Mistakes are relatively well known. 18million of these were reserved
for private networks, and a millon were reserved for multicast. Thus
reducing the POOL of available labels. On top of this, assignment
mistakes were made so some companies/firms were allocated an single
class and wiped out 16million straight away

IPv6 - 128bit. Available addresses, too many to work out. However, if
the world had 18 quintillon people on it, EACH person could be
allocated 18 quintillion addresses. As I am British I will use the
British method. A quintillion is
10,000,000,000,000,000,000,000,000,000,000 (forgive me if I have
slipped in one too many noughts there). We have a way to go before we
hit 18 billion people on the planet, never mind quintillion.

So pray, please tell me, how do you propose we use all this up in less
than 13 yrs?

If you are going to say something which appears on the face of it, to
be not thought out, then please explain to a mere simpleton like me how
it will be done.

I will admit I do not know the innards of the various shenigans in the
IANA. However, RFC 3041 seems to indicate that they are seriously
looking at going down the road of randomizing the latter 64bit of the
address. To do this effectively, would mean that reservation of the the
latter 64bit would not take place. Thus effectively meaning that the
pool of addresses would be taken up linearly. You could effectively
give every individual one address in the first 64bit range and let them
do what they want with the other 64bit and still not run out. If this
approach was adopted, then your assertion that address space would be
used up in 13 yrs would seem on the face of it to be mere folly.

Sorry, I know my maths ain't good but your comment does not stack up.
However, to be expected as it's just a throwaway comment. So I will
wait for your well reasoned and well thought out retort that I can
dwell on tomorrow evening.
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

On Jun 5, 12:01 pm, "Roger (K8RI)" <validaddr....DeleteThis@my.com> wrote:
> I think you and I tend to agree on most things, but we do have a bit
> of a communications problem at times<) Probably more due to there
> being several generations between us in our way of thinking and
> presenting things.

I'm absolutely fine with others having a different way of thinking.
Diversity is a good thing. I think in a way I've allowed myself to be
demonized that even those I consider friends will refer to me as an
extremist or tell me not to look at their performance reports.

I only objected to people giving a report that's awful then
summarizing as wonderful, to people being censored, and to people who
are dismissive of bugs many others are experiencing because they're
lucky enough not to have had trouble. The reaction to the release of
the SP1 patch therefore came across as staged and composed of phoney
marketing spiel, whether or not it actually was.

I was certainly colorful in my language, both in the use of expletives
and stating that certain things are madness, but that wasn't intended
as a personal attack on anyone, despite those that seem to harp on it.
There are a handful of nitwits who I don't apologize for insulting
directly after they'd directly insulted me. Even genuine apologies
were met with abuse. Hardly conducive to any further apology. What can
I say - some people get their thrills being trolls, and waiting for
you to turn your back to stick the knife in. Fortunately it's a net
forum and in the grand scheme of things makes no difference.
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

"Sammy" <syousef DeleteThis @bigpond.net.au> wrote in message
news:1181006035.893899.314270@n15g2000prd.googlegroups.com...
> On Jun 5, 10:29 am, "Mike Young" <boat042-s... DeleteThis @yahoo.com> wrote:
>> It was rhetorical. SLI is transparent to the application.
>
> Afraid not. Take a look at the links I provided.

Afraid so. Ergo, MSFS *does* take advantage of SLI.
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

"Sammy" <syousef DeleteThis @bigpond.net.au> wrote in message
news:1180999034.711759.24390@i38g2000prf.googlegroups.com...
> On Jun 5, 6:28 am, "Tilly" <nos... DeleteThis @willbeletin.com> wrote:
>> So saving my pennies for a SLI machine, and either the best SLI I can
>> afford (and get second one later when it's cheaper!) or two SLI's at
>> same time - can't make my mind up on that. Tempted to wait for DDR3
>> memory; but suspect it's too early for that yet and will be flaky as
>> hell when it comes out
>
> Just be aware at the moment MS flight sim does not support SLI.

Just what does an app need to do to support SLI?
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

"Sammy" <syousef.RemoveThis@bigpond.net.au> wrote in message
news:1181000951.733557.308460@g37g2000prf.googlegroups.com...
> On Jun 5, 9:23 am, "Mike Young" <boat042-s....RemoveThis@yahoo.com> wrote:
>> Just what does an app need to do to support SLI?
>
> My understanding is that it must make it's calls to the graphics card
> in a particular way, but I don't pretend I'm a Direct X game
> programmer.

It was rhetorical. SLI is transparent to the application.
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

"Sammy" <syousef.RemoveThis@bigpond.net.au> wrote in message
news:1181008111.580340.74340@j4g2000prf.googlegroups.com...
> If it's technically suppose to work, but doesn't give you a
> performance boost of any kind, is it worth even talking about SLI, or
> can you honestly say that the application supports SLI?

Nobody expects you to know everything. Just lighten up.

All the same, though, you need to state clearly that the previous statement
was in error. A simple "Yeah, you're right." Or the more personal, "Yeah, I
was mistaken." Or even, "I'm just too clever. MSFS doesn't support SLI,
because SLI just is." One of those will do. You wouldn't want someone making
a purchasing decision based on that earlier piece of bull, would you?
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

Sammy wrote:

> Take today. Some poor sod's going to come up against Mx's ridiculous
> assertion that the video card doesn't make a difference to flight sim.
> (This is almost true at the high end since the bottleneck at the
> moment is the CPU but the person most likely to follow the advice will
> be the one who can't afford a dedicated video card and goes with built
> in graphics, only to find the other several hundred dollars they spent
> were wasted without a decent graphics card. There's no replacement for
> a dedicated graphics card and decent DX9 cards can be had at
> reasonable prices at the moment).

This much is true. However, it is fair to say, from my own experience
(not saying that this is definative mind, just an experience) - that
the GF card did make an improvement (and I have to say it did) my
expections were somewhat dashed.

I swopped out an old AGP4 Asus 128mb FX4200i (from memory, don't quote
me) for a 256mb AGP8 7600GS (OK it's not a GT!). Now I am a sucker for
benchmarks so when the benchmarks indicated a 600% improvement I went
"way hey!!!" - to find that in real-terms, I had a 50% improvement in
frame rates in FS9.

Realistically, I am not disappointed now - I believe (like some people)
got carried away with the benchmark tests so my expectations were too
high. FS9 is working fine, it's limited to 25FPS and works fine and
dandy. Apart from the blasted DX crashes when I go ballastic with the
panning - guess that should be fixed by a more recent drivers but I
tend to avoid updating drivers unless I *really* need to - victim of
too many unstable beta releases I guess - running 93.71 at the
moment.

Got to admit the world has come on leaps and bounds with GF cards, I
remember having to buy a separate MPEG decoder card in the bad old days
- still have it around somewhere - it will be on antiques roadshow one
day!

So saving my pennies for a SLI machine, and either the best SLI I can
afford (and get second one later when it's cheaper!) or two SLI's at
same time - can't make my mind up on that. Tempted to wait for DDR3
memory; but suspect it's too early for that yet and will be flaky as
hell when it comes out
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

On Mon, 04 Jun 2007 16:37:02 -0700, Sammy <syousef.RemoveThis@bigpond.net.au>
wrote:

>On Jun 5, 6:43 am, "Tilly" <nos....RemoveThis@willbeletin.com> wrote:
>> Know he does not know any better. However, he does know how to get you
>> rattled
>
>He's gotten me rattled exactly one time, and that was when he attacked
>my fiancee's mental stability knowing almost nothing about her. That
>definitely crossed the line for me. The rest of this is just BS. I'm
>irritated that he keeps putting down the efforts of entire classes of
>people - not a single person or a single project but almost all
>computer engineers. Heck he probably couldn't even deal with the math
>to understand what these people do. It is also frustrating that he

I wonder how he'd do with "The Design and Analysis of Algorithms" or
"Digital Image Processing?" That's where I found when a PHD says "not
much math" and I say it we don't mean the same thing and I at least
have a minor in math<)
The analysis wasn't all that bad as we only went to 5 level
simultaneous equations. The image processing started easy with linear
transforms, but moved into Fourier transforms, not Fast Fourier, then
Matrix Algebra, and after that it became work.

I really doubt there are many outside the field who realize how much
mathematics is involved.

>keeps shifting the argument but as he hits the boundaries of his
>knowledge he keeps saying increasingly wacky things which are more and
>more transparent. There's a difference between these things and
>rattled.

He will milk an argument way beyond the point of believability.

I think you and I tend to agree on most things, but we do have a bit
of a communications problem at times<) Probably more due to there
being several generations between us in our way of thinking and
presenting things.


>
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

What on earth does IPv4 or 6 stand for? Am I dim or what? And why should I
care?


"Tilly" <nospam RemoveThis @willbeletin.com> wrote in message
news:1180992773.20261.0@iris.uk.clara.net...
> Mxsmanic wrote:
>
>> > Mind you, the big problem comes when we have to move to IPv6 -
>> > March 2010 I believe is the date set for when IPv4 runs out.
>>
>> And March 2020 will probably be the day that IPv6 runs out.
>
> Wow! That is a strong statement to make. Can I ask if you have any
> evidence to subtantiate this. Or, are you going to sit on the fence
> with the good ole word - "probably".
>
> Let's take your one-line statement into context.
>
> IPv4 - 32bit, has 4,294,967,296 possible unique addresses. Due to
> *mistakes* made by *engineers* (I would actually say mistakes by
> laypeople, but seeing as *engineers* are in your scope, we will stick
> with that for the time being).
>
> Mistakes are relatively well known. 18million of these were reserved
> for private networks, and a millon were reserved for multicast. Thus
> reducing the POOL of available labels. On top of this, assignment
> mistakes were made so some companies/firms were allocated an single
> class and wiped out 16million straight away
>
> IPv6 - 128bit. Available addresses, too many to work out. However, if
> the world had 18 quintillon people on it, EACH person could be
> allocated 18 quintillion addresses. As I am British I will use the
> British method. A quintillion is
> 10,000,000,000,000,000,000,000,000,000,000 (forgive me if I have
> slipped in one too many noughts there). We have a way to go before we
> hit 18 billion people on the planet, never mind quintillion.
>
> So pray, please tell me, how do you propose we use all this up in less
> than 13 yrs?
>
> If you are going to say something which appears on the face of it, to
> be not thought out, then please explain to a mere simpleton like me how
> it will be done.
>
> I will admit I do not know the innards of the various shenigans in the
> IANA. However, RFC 3041 seems to indicate that they are seriously
> looking at going down the road of randomizing the latter 64bit of the
> address. To do this effectively, would mean that reservation of the the
> latter 64bit would not take place. Thus effectively meaning that the
> pool of addresses would be taken up linearly. You could effectively
> give every individual one address in the first 64bit range and let them
> do what they want with the other 64bit and still not run out. If this
> approach was adopted, then your assertion that address space would be
> used up in 13 yrs would seem on the face of it to be mere folly.
>
> Sorry, I know my maths ain't good but your comment does not stack up.
> However, to be expected as it's just a throwaway comment. So I will
> wait for your well reasoned and well thought out retort that I can
> dwell on tomorrow evening.
>
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

This message is not archived
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

"Sammy" <syousef DeleteThis @bigpond.net.au> wrote in message
news:1181034308.651731.262010@i38g2000prf.googlegroups.com...
> On Jun 5, 1:35 pm, "Canuck" <some... DeleteThis @somewhere.com> wrote:
>> What on earth does IPv4 or 6 stand for? Am I dim or what? And why
>> should I
>> care?
>
> The very short answer is:
> Don't worry about it unless you're an IT professional or have a
> problem or are curious and want to learn about it..
>
> The slightly longer answer is:
> that IPv4 addresses are the addresses used by computers on the
> Internet. We're running out of unique addresses in the public address
> space, and once we do we can't create more IPv4 addresses

In a nutshell, IPv4 used to be referred to as simply IP, the Internet
Protocol. The distinction is made to distinguish between old-IP and new-IP,
apparently skipping middle-aged-IP. IPv6 has been underway since the
mid-90's. The changes are not limited to just the larger address space.

>(There are
> however sneaky techniques for getting around this, but they have their
> limitations and caveats to this. For reference it's called NAT -
> Network Address Translation)
>
> IPv6 is a new address system meant to eventually replace IPv4. IPv6
> has many many more addresses and can co-exist with IPv4. However all
> the existing IPv4 applications - just about all the networking apps on
> the Internet - would need to be modified to cater for the new
> addressing, and it's not trivial.

IPv4 will be tunneled in IPv6, and does not require changes in existing
apps. The two will run side by side. Initially, today, IPv6 is tunneled in
IPv4.
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking 

Some darn old flight simmer called Sammy wrote:

> Just be aware at the moment MS flight sim does not support SLI.


Yep, but my other games will
 >> Stay informed about: Sammy, I am beginning to come around to your way of thinking