1  ...  4  5  6  7  8  9  |   Order by default

Oct 28, 2003, 7:02pm

It also is a good way to reduxce the spread of STDs
and such...
<<<

The exact opposite - the less it is in the open and the more people who
are involved in it have to fear the public, the more will happen. The
public eye just does not "see" it anymore and the public conciousness is
satisfied.




[View Quote]

Oct 29, 2003, 1:09pm

Fixes the problem where caretakers of one world could change settings in
another world that would affect only them, though - eg. enable flying
for oneself.


[View Quote]

Oct 31, 2003, 6:17am

You can limit your _maximum_ visibility by using a fixed setting from
the "Visibility" menu (not "float", but like 20m, 40m, etc.). In this
case, only items withing the visibility range will be downloaded. The
maximum visibility must be set new every time you enter a world, since
it will automatically always reset to "Float".

Setting to "Float" means that the maximum visibility using the minimum
frame rate (as set in options:settng:performance) is being used, and
this can be, depending on the environment you are standing in, up to
200m.

The "minimum visibility" (options:settings:performance) and the mininum
visibility (server setting) is the range that you will see under each
circumstance - the bigger of these two settings is applied to each
individual person.

So, to download the smallest possible amount of items, you need to set
the minimum visibility to 30m (client and server) and the maximum
visibility to 30m as well (as soon as you enter a world).

The "refresh rate" in the world settings determine (in minutes) how
often items are checked if they are still actual or have changed. This
check downloads not the actual item, but only the time stamp and only if
the time stamp indicate that the item has changed since the last check,
an actual download takes place.

A senseful setting for the refresh rate is 10080 (1 week) or multiples
thereof.



[View Quote]

Nov 3, 2003, 2:09pm

10080 minutes = 1 week (7 days * 24 hours * 60 minutes)

Even when the backup is older than (refresh time), an update check is
not only much faster, but also will not require any downloads.




[View Quote]

Nov 6, 2003, 10:12pm

Not correct - when done in the right way, it looks not only identical,
but even better (since it goes up higher and has a top). :-)


[View Quote]

Feb 2, 2004, 12:36pm

I'm not impressed by a new toolbar, to say the least, nor am I
particularly fond of the lack of CPU scaling options. Let me get my
complaint clear: It's not that I dislike the CPU scaler feature, it's
great for laptops and multitasking, but I absolutely DESPISE the lack of
an option to turn it off or make it less severe. There are circumstances
where you may wish to allow AW to have all the CPU it can handle, and
now there is no way to do that.
<<<

I absolutely agree and I suggested that right away when the new method
was implemented, at least in 3 steps as "CPU relax" settings of "on",
"off" and "default" (as a medium value).

Feb 2, 2004, 12:53pm

As a strong critic of these float_around_windows, I agree 100%.

I suspect the reason to make the windows all independent of the "mother"
window is the migration to a sexed-up full screen application (like
games), in which you will then have floating windows for everything.




[View Quote]

Feb 2, 2004, 1:04pm

me too :-)


[View Quote]

Feb 4, 2004, 8:34am

IMO, 3.5 was mainly a release aimed at corporate universe owners, plus
fixing some bugs of the previous version. The practical value of 3.5 (in
terms of advantage over 3.4) is sort of limited and brings advantages
and disadvantages.



[View Quote]

Feb 4, 2004, 8:35am

> The UI has become worse in my opinion. :-\

I second that.



[View Quote]

Feb 26, 2004, 5:26pm

Has the terrain problem also be fixed in this one as in the 3.5 beta
recently? That would be wonderful :)




[View Quote]

Mar 8, 2004, 5:21pm

mainly the floating tabs and the new toolbar buttons à la E N Z O.
<<<

.... while the so-called rating symbols that have relevance exclusively
to the USA, as well as the contact list AFK symbol with relevance only
to english language, remain hardcoded in the browser?
I am not sure which was more important to have for universe owners; if I
was changing some, I had changed all. :-)


[View Quote]

Mar 8, 2004, 8:40pm

It does not matter where the product is from, but where it is supposed
to be _sold_. If I sell my game to China, they will expect me to
consider chinese issues.

It is not that AW is labelling their product as "for US use only". :-)


[View Quote]

Mar 8, 2004, 8:58pm

The product is intended for US sales. But is _available_ for the
worldwide market. Just because I make my bubblegum available in
france doesn't mean I make it compatible with french chewing procedures.
<<<

Right, that is why there is an _official_ distributor in Europe. :-)

But - I give you chance to prove your point too: where do you find any
note that the product is for use and sale in the USA only?

Mar 8, 2004, 9:09pm

Just because I make my bubblegum available in france doesn't mean I make
it compatible with french chewing procedures.
<<<

You even have to translate all descriptions into French before it will
be considered to be allowed to be made available <g>.

BTW, of the 25 universes listed on Mauz page, more than 33% are non-US,
and even when the product was once thought to be sold in USA only, the
status today is simply different. If you find out that 1/3 of your
chewing gum is bought in China, then you will tear off a leg to make it
compatible with Chinese expectations - unless you do not bother about
or do not want to have or keep any customers at all. :-)

Mar 9, 2004, 10:14am

Nope, that is not my point - my point is why using acronyms of a certain
language when universal pictograms can be used as well (and,
design-wise, look even better).

The rating system is a different thing though, since it is specific to
one country (not language) only.


[View Quote]

Mar 9, 2004, 10:19am

The light source limit derives from one of the graphic engine
implementations (DirectX, IMO it works well in opengl and software mode
- which have other problems). I doubt they will switch to another one to
solve it.


[View Quote]

Mar 14, 2004, 8:34am

There is no "better" protection, there is only "no" or "little" <g>.

When you pretend to be a AW browser, then you can download like aw
browser itself do too. One advanced method would be to have a script
checking which persons are present in a world and allow access then only
to these specific IPs, but it comes with some
disadvantages/inconveniences and would also not offer absolute security
(shared OPs, remote linking to websites, proxy useage, etc.). There is
always the question whether increased security (for whom?) beats reduced
convenience and/or rights (for whom?) and/or only offers a wrong felling
of security.



[View Quote]

Mar 14, 2004, 8:36am

Passwords (or encryption elements) are relatively easy to obtain when
you have identical data in protected and unprotected files.



[View Quote]

Mar 15, 2004, 4:35pm

To give you all an impression how long it takes to decrypt a file or
find a password, I give you some numbers.

I post this since some people seem to think that the so-called "password
crackers" are some higher beings with abilties not from this world -
which they are not, it actually takes less of a brain that a script
kiddy would need...

This only applies to retrieving a password form a zip file when you have
the file, it does NOT apply to obtaining a password in the course of
transmission (eg. by listening to the server-client data exchange).

What I needed:

a) the zipped file - password protected
b) the zipped file - without password

Usually a potential attacker can easily obtain the second by buying an
object. Now he downloads the encrypted version from the developers OP
(which usually has the same name, or the name can be obtained somehow).

What I needed to do:

Run a zipcrack program (which can be easily obtained by buying or
leeching it).

Results:

- With a password of 9 characters or less, I needed 24s (SECONDS!) to
un-encrypt the password - leaving now the whole OP open for
all-you-can-download actions, a 10 character password took no longer
than 34s. (System: P4/3500 MHz)

Conclusion:

Passwords of 10 characters or less are unsafe to a degree that you can
as well not use a password at all, while longer password are not more
safe, but only less unsafe.

It would nt take me wonder when most worlds from the "cracked" passwords
list had only short passwords, which could be obtained literally in a
matter of seconds.

Suggestions:

1) NEVER encrypt any file on your OP that exists somewhere unencrypted
already. If it does, change the original (unzipped) file and add
something or remove something, so that it is NOT IDENTICAL anymore to an
unencrypted version. Note especially that avatars.dat, ALL sequences and
ALL sounds are ALWAYS unencrypted in the cache folder of the AW browser,
so encrypting any of these files is an open invitation to casual
password-snatchers.

2) NEVER use short passwords (not here and not in any other
environment), try to use whole sentences with words that are not in a
dictionary and add some non-letter-non-digit characters.

3) As a developer, NEVER give out any files that are identical to the
protected ones on your OP, always add or remove something from the
original file before sending it to a buyer. Change the file name, too,
to make it more difficult to obtain certain model names (to prevent easy
downloading of big numbers of files from the OP).

Comment:

The near practical impossibility of (1) makes, in fact, all password
attempts pointless, as does any public building with encrypted objects,
as does any building with encrypted objects with citizens whom you do
not trust 300+%.

Additional risks include password retrieval from the network traffic
and/or obtaining it from other flaws in the system.

I'd classify the whole protection system of objects as low grade at
maximum and generally not useable to ensure the protection of
copyrighted material - and, as I have posted before, every developer
should be (and is?) aware of that. And to add, I do not see any
convenient or practical method of protection that cannot be overrun;
there ARE safe methods, but they will impose a level of inconvenience
and unpracticability that are not really justified for this environment.
What good will it do when you can protect your material and nobody will
see it anymore since nobody wants to bear the burdens of running the
software then. And the same goes to sophisticaed methods of obscuring OP
and objects names - remember, these things SHALL be downloaded, and
everything that SHALL be downloaded, WILL be downloaded (whether in the
right and lawful way, or not).

The key here is to make people sensible and aware of the work involved
in creating things - many, especially the worldowners of respected
worlds, will pay the developers for designs they get, while there will
always be a number of (mostly casual worldowners) who will leech them
from whatever source and by whatever method. And there is no point in
trying to pursue them either, due to the practical law and cost
obstacles involved.

My point is that business will continue as usual, designers build mainly
for a handful of worldowners who buy their work (and are proud that they
bought it, after all, it shows also style and class to their visitors),
and some others (without style, class, and even not admired by more than
a few even less capable pals of theirs) will continue to leech what they
can get. C'est la vie - it won't destroy our hobby and it won't drive
away developers, as long as these developers do not try to make a living
out of their work, but see it as a nice, additional, not guaranteed,
income.

Mar 15, 2004, 7:56pm

Yup, your tool is highly recommended! :-)

One thing to watch though - COB files cannot be changed by simply adding
something with a text editor, maybe you can come up with something
changing them too somehow... or developers should ALWAYS give out rwx
files only!



[View Quote]

Mar 15, 2004, 6:41pm

Well, what you describe is not the open AW approach anymore, but a
proprietary system, like others have, too.

And for such, you would not write modules to plug them into a webserver
- in fact, you would not even need a webserver anymore but can set up a
simple fileserver using a proprietary protocol.

Encoding is also not necessary when you implement a proprietary storing
format (you even cann add then knicknacks like watermarking, etc.) and
add a program through which objects, mp3, jpgs, etc. will be converted
into this format.

I am not sure, if I want this <g> - or better, I am sure I do not want
it, since the open and simple approach of AW is one of its strongest
points. AW was not meant to be a proprietary system with increased
support for copyright protection.

Why not make it really perfect and give each developer a unique key, let
them sign each sold item too and put it all on an authentication server
that keeps track how often, when and by whom it was accessed, and, at
the end of the month, writes a bill for license fees - would be fair,
secure ... and like many other things in these days, like shooting with
a cannon on a fly. :-)




[View Quote]

Apr 12, 2004, 8:55am

I don't think the tabs are a good idea either - actually I did not not
find many with the opposite view so far, but maybe they could come
forward in this thread. :-) I think the old style should be preserved as
an alternate "classic view".
The new stye is a nice thing, eg. when you have 2 screens and can then
put the tabs on one and the graphic window on a second screen, it is
nice as well when you do not use tabs very often or have a very small
screen resolution (640*480) to save space - but in a normal environment
the old style serves better and is more effective for everydays use.


[View Quote]

Apr 12, 2004, 9:19am

At 55 years of age, I learned to resize my screens to my satisfaction,
roll them up if necessary, change a bit my petty attitudes... and enjoy
the rest.
<<<

I certainly would enjoy the rest much more, when, with ongoing
development, I would not have to grow additional arms and fingers to
perform all the tasks and adjustments that were not needed before. :-)

Development of 4.0 goes toward full screen mode with the GUI painted
directly to the screen (in opposite to using the windows system),
something that has, in my very personal experience and opinion, a few
advantages and a lot of disadvantages in terms of practicability.
If the windowed mode is kept in addition, there is the need for 2 very
different GUIs anyway, it would only be in the development phase that
the split of the GUI elements has to be checked and after that, it can
be re-integrated for the windowed mode.

I am aware that one has to "learn" new things when a development takes
place - but there is quite a difference betwee the need to "learn new"
things and to "re-learn old" things, or, in an analogy, I would not
accept it either when in every new car model I buy, they rearrange the
basic functions of the buttons and switches and increase the difficulty
to handle them. :-)



[View Quote]

Apr 18, 2004, 8:24pm

There is (in AW terms) only little difference between a P3/1000, an
athlon XP2000 and a P4/2660 or P4/3200 as long as you have a decent
graphic card. In this case, I tried them out with a GF TI4400, which is
somewhat similar to the new GF5700/5750 and this is comparable to the
ATI9600 - again, in terms of AW performance, which means DX7 and no need
for antialiasing, things look different when you play the latest
DirectX9 game or run huge resolutions with 4 and more levels
antialising, then you would need, in fact, the latest products,
preferably ATI.

Whether you have 256 or 512 MB of main RAM does not really matter, and
the RAM of the graphic cards is usually not used to a higher degree than
64MB anyway.

It mainly depends whether you have antialiasing turned on or off and
whether you run 800*600 or eg. 1600*1200 screen resolution. The higher
the antialisaing level and the higher the resolution, the slower the
whole thing will get (and respectively you need a faster/newer graphic
card then).

CPU speed or type only matters when you do other things at the same time
(eg. when beoing in AW, calculate 50 digit primes while encoding a DIVx
movie and zip your whole drive - in this case you would need a very fast
CPU or even multiple ones...).

So to cut a long story short, any P3/1000+, P4/2400+, AMDT1400 or
AMDXP200+, all will work perfectly when you got at least a GF4 or Radeon
9600 graphic card.

It is more important to have your system configured and tweaked well -
both things can quite well make up for some missing 1000 MHz or RAM...




[View Quote]

Apr 19, 2004, 9:15am

I remember, the change from 1.3 ghz to XP 2600 was pretty impressive
<<<

When then 1.3 was a Duron, then the switch to a different CPU
architecture can as well have advantages - same with the jump from
Celeron to Pentium.




[View Quote]

Apr 19, 2004, 1:02pm

Looks good - btw, I have W98SE on all my machines and do, in the context
of what I am doing with the computers, not see any reason to switch to
XP (yes, I know XPpro and have administered a network with it already
<g>). :-)





[View Quote]

Apr 29, 2004, 6:02am

The blurry graphics might come from the fact that "mipmaps" now turn on
again by default - they need to be de-activated manually.

For all other issues - you summed it up good enough. :-)



[View Quote]

Apr 29, 2004, 4:53pm

Unfortunately, the tab does not remember its position anymore and turns
now up either on the main window middle left or on the main window lower
left (depending whether you closed AW with exit or X and open or closed
tabs).


[View Quote]

May 1, 2004, 1:06am

How it looks depends on many things, like screen resolution, color
depth, various settings for the graphic card etc. - So the same settings
in the AW program can look very different on 2 computers. In addition,
the texture resources play a big role, too.

I have for instance antialiasing and anisotropic filtering turned off
and mipmaps turned off - and the picture looks very crisp, clear and
colorful on a 22" monitor using 1024*768 resolution at 100Hz and 32 Bit
color depth (GFti4400) - same on a LCD ( at 60Hz) with the GFmx440 or on a
notebook using ATImobility. All of the above with DirectD8 with T&L, of
course, for Dx7, opengl and software mode, I would pick other settings.

When I turn antialiasing and/or mipmaps on, it looks more like a washed
out fuzzy picture, sort of like in there.com or in most games.
Antialisaing is, btw, not antialisaing - there are significant
differences between ATI and NVidia and also big differences between the
graphic card versions, in addition, antialiasing can be set in (at least
3) different steps as well.

However, it also depends on the resources - when most textures are
stretched from a 64*64 Pixel size, they wouldn't look too good, while
texture sources of 256*256 or 512*512 are really good (for terrain for
example I would only use 256 or 512 - 1024 do not bring enough quality
improvement anymore, while anything below 256 is quite fuzzy or has
significant block artefacts).
Of course, one has to find the right tradeoff here between texture
quality (and viewing pleasure) on one hand and the loading time on the
other, so when a world has many dialup-modem-visitors, smaller texture
resources will make more sense.

As I said in the beginning, there is a lot of things influencing the
view, and when one world looks nice/bad with specific settings, another
world might give a totally different impression. Not to speak even about
personal preferences (some people like the fuzzy view, others prefer
crisp graphics).

The "glitter" effect is, btw, caused by render
interpolation/interference of color differences between 2 pixel groups -
as such, they depend on the graphic card, the settings of the card, the
texture size (bigger sizes with high tiling frequencies will tend to
glitter from certain angles and distances, those should be used with low
tiling frequencies rather) and the color differences within the picture
(big tonal differences will tend to glitter with high tilings). Even the
actual monitor resolution (virtual and physical) will influence the
actual view.

In this regard, it is more than normal when one says "turning this or
that on or off looks good/bad" and the other says the opposite - only
when all circumstances are the same, 2 people would really see an
identical picture.



[View Quote]

1  ...  4  5  6  7  8  9  |   Order by default