|
Home > Archive > Macromedia Flash Server > April 2005 > bwcheck: is it a good solution?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
bwcheck: is it a good solution?
|
|
| Wilton Rossi 2005-04-28, 5:45 pm |
| Hi there!
I'm working with a team in the development of a
FCS-based communication suite. One of our greatest
problems now is to develop an efficient bandwidth
check solution. We've been trying different approaches
at using bwcheck.as and bwcheck.asc, which are widely
used by FCS developers.
The problem is that it seems to be inefficient in
real-world Internet-based situations with users having
different connection speeds. I'll tell you why (and
please correct me if I'm wrong). Imagine that the
first user logs in and his/her bandwidth is checked by
the server. SBeing first one, he/she will have all
bandwidth at his/her disposal. Great. Bwcheck
automatically adjusts his settings. The second one
logs in. But now a great part of the bandwidth is
being destined to the first user. So the calculation
is based on the remaining server upstream. As other
users log in things get worse. Even if they have
better connections, the automatic adjustment will
choose settings destined to slower speeds due to
decreasing server bandwidth available.
You could consider testing again every time someone
logs in. The problem in that case is that audio and
video are probably already being transmitted by all
users. So the test loses its accuracy.
This is the situation we're dealing with in our tests.
I wonder if someone was faced with similar scenarios
and was able to find a solution.
Regards,
W
__________________________________
Do you Yahoo!?
Yahoo! Small Business - Try our new resources site!
http://smallbusiness.yahoo.com/resources/
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Peldi Guilizzoni 2005-04-28, 5:45 pm |
| Hey W, the scenario you describe would only happen if the bw required to
play your FLV is in the same ballpark as your server's link to the Internet,
which is not usually the case (~200Kbps for a regular FLV, 1.5-10.0Mbit for
a regular server's connection).
Basically what I'm saying is that having even 10 people connect and
calculate their BW at the exact same time shouldn't be a problem for your
server...
As for bwcheck.as and bwcheck.asc, if you got them from my blog they're not
super-accurate, they don't have the latest fixes (which deal with modems
that automatically compress data before sending it). You should instead use
the code in the FLVPlayer.zip here: http://www.peldi.com/blog/FLVPlayer.html
Everyone who uses this bw detection code (including Amazon.com I think) has
not reported any problems with it.
Peldi
-----Original Message-----
From: flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
[mailto:flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org]
Sent: Thursday, April 28, 2005 10:00 AM
To: flashcomm-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
Subject: [FlashComm] bwcheck: is it a good solution?
Hi there!
I'm working with a team in the development of a
FCS-based communication suite. One of our greatest
problems now is to develop an efficient bandwidth
check solution. We've been trying different approaches
at using bwcheck.as and bwcheck.asc, which are widely
used by FCS developers.
The problem is that it seems to be inefficient in
real-world Internet-based situations with users having
different connection speeds. I'll tell you why (and
please correct me if I'm wrong). Imagine that the
first user logs in and his/her bandwidth is checked by
the server. SBeing first one, he/she will have all
bandwidth at his/her disposal. Great. Bwcheck
automatically adjusts his settings. The second one
logs in. But now a great part of the bandwidth is
being destined to the first user. So the calculation
is based on the remaining server upstream. As other
users log in things get worse. Even if they have
better connections, the automatic adjustment will
choose settings destined to slower speeds due to
decreasing server bandwidth available.
You could consider testing again every time someone
logs in. The problem in that case is that audio and
video are probably already being transmitted by all
users. So the test loses its accuracy.
This is the situation we're dealing with in our tests.
I wonder if someone was faced with similar scenarios
and was able to find a solution.
Regards,
W
__________________________________
Do you Yahoo!?
Yahoo! Small Business - Try our new resources site!
http://smallbusiness.yahoo.com/resources/
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Wilton Rossi 2005-04-28, 5:45 pm |
| Hello Peldi,
Thanks a lot for your fast response. By the way, your
book is being of great help to our team.
But I think I wasn't able to describe my situation
properly. The application we're working on is a 6-seat
audio/video meeting room. That application also has
features such as whiteboard, chat, text editor, file
transfer etc. We're testing it using Influxis as our
host with a 1.25 Mbps connection.
Our tests indicate that first user is granted a
greater share of the server's bandwith, while those
who log in later have their tests impaired by server
bandwidth already destined to previous users. At least
that's our assumption.
Under those circumstances, is it reasonable to have
all users re-check whenever a new user logs in as you
proposed? Do you think it'll work?
Regards,
W
--- Peldi Guilizzoni <gguilizzoni-14osZcCZf762oZ/6fjIToQ@public.gmane.org>
wrote:
> Hey W, the scenario you describe would only happen
> if the bw required to
> play your FLV is in the same ballpark as your
> server's link to the Internet,
> which is not usually the case (~200Kbps for a
> regular FLV, 1.5-10.0Mbit for
> a regular server's connection).
>
> Basically what I'm saying is that having even 10
> people connect and
> calculate their BW at the exact same time shouldn't
> be a problem for your
> server...
>
> As for bwcheck.as and bwcheck.asc, if you got them
> from my blog they're not
> super-accurate, they don't have the latest fixes
> (which deal with modems
> that automatically compress data before sending it).
> You should instead use
> the code in the FLVPlayer.zip here:
> http://www.peldi.com/blog/FLVPlayer.html
>
> Everyone who uses this bw detection code (including
> Amazon.com I think) has
> not reported any problems with it.
>
> Peldi
>
> -----Original Message-----
> From: flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
> [mailto:flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org]
> Sent: Thursday, April 28, 2005 10:00 AM
> To: flashcomm-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
> Subject: [FlashComm] bwcheck: is it a good solution?
>
> Hi there!
>
> I'm working with a team in the development of a
> FCS-based communication suite. One of our greatest
> problems now is to develop an efficient bandwidth
> check solution. We've been trying different
> approaches
> at using bwcheck.as and bwcheck.asc, which are
> widely
> used by FCS developers.
>
> The problem is that it seems to be inefficient in
> real-world Internet-based situations with users
> having
> different connection speeds. I'll tell you why (and
> please correct me if I'm wrong). Imagine that the
> first user logs in and his/her bandwidth is checked
> by
> the server. SBeing first one, he/she will have all
> bandwidth at his/her disposal. Great. Bwcheck
> automatically adjusts his settings. The second one
> logs in. But now a great part of the bandwidth is
> being destined to the first user. So the calculation
> is based on the remaining server upstream. As other
> users log in things get worse. Even if they have
> better connections, the automatic adjustment will
> choose settings destined to slower speeds due to
> decreasing server bandwidth available.
>
> You could consider testing again every time someone
> logs in. The problem in that case is that audio and
> video are probably already being transmitted by all
> users. So the test loses its accuracy.
>
> This is the situation we're dealing with in our
> tests.
> I wonder if someone was faced with similar scenarios
> and was able to find a solution.
>
> Regards,
>
> W
>
>
>
> __________________________________
> Do you Yahoo!?
> Yahoo! Small Business - Try our new resources site!
> http://smallbusiness.yahoo.com/resources/
>
>
=-----------------------------------------------------------
> Supported by Fig Leaf Software -
> http://www.figleaf.com
>
=-----------------------------------------------------------
>
> To change your subscription options or search the
> archive:
>
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
>
=-----------------------------------------------------------
> Supported by Fig Leaf Software -
> http://www.figleaf.com
>
=-----------------------------------------------------------
>
> To change your subscription options or search the
> archive:
>
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
________________________________________
__________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Peldi Guilizzoni 2005-04-28, 5:45 pm |
| Ha! You tickled my memory. I think that the bwcheck.asc code has a bug in
which it doesn't reset the payload properly after the first person uses it.
This would result in what you're seeing. You should definitely move to the
bw checker code in the FLVPlayer.zip, all bugs have been fixed in that one.
Peldi
-----Original Message-----
From: flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
[mailto:flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org]
Sent: Thursday, April 28, 2005 11:07 AM
To: FlashComm Mailing List
Subject: RE: [FlashComm] bwcheck: is it a good solution?
Hello Peldi,
Thanks a lot for your fast response. By the way, your
book is being of great help to our team.
But I think I wasn't able to describe my situation
properly. The application we're working on is a 6-seat
audio/video meeting room. That application also has
features such as whiteboard, chat, text editor, file
transfer etc. We're testing it using Influxis as our
host with a 1.25 Mbps connection.
Our tests indicate that first user is granted a
greater share of the server's bandwith, while those
who log in later have their tests impaired by server
bandwidth already destined to previous users. At least
that's our assumption.
Under those circumstances, is it reasonable to have
all users re-check whenever a new user logs in as you
proposed? Do you think it'll work?
Regards,
W
--- Peldi Guilizzoni <gguilizzoni-14osZcCZf762oZ/6fjIToQ@public.gmane.org>
wrote:
> Hey W, the scenario you describe would only happen
> if the bw required to
> play your FLV is in the same ballpark as your
> server's link to the Internet,
> which is not usually the case (~200Kbps for a
> regular FLV, 1.5-10.0Mbit for
> a regular server's connection).
>
> Basically what I'm saying is that having even 10
> people connect and
> calculate their BW at the exact same time shouldn't
> be a problem for your
> server...
>
> As for bwcheck.as and bwcheck.asc, if you got them
> from my blog they're not
> super-accurate, they don't have the latest fixes
> (which deal with modems
> that automatically compress data before sending it).
> You should instead use
> the code in the FLVPlayer.zip here:
> http://www.peldi.com/blog/FLVPlayer.html
>
> Everyone who uses this bw detection code (including
> Amazon.com I think) has
> not reported any problems with it.
>
> Peldi
>
> -----Original Message-----
> From: flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
> [mailto:flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org]
> Sent: Thursday, April 28, 2005 10:00 AM
> To: flashcomm-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
> Subject: [FlashComm] bwcheck: is it a good solution?
>
> Hi there!
>
> I'm working with a team in the development of a
> FCS-based communication suite. One of our greatest
> problems now is to develop an efficient bandwidth
> check solution. We've been trying different
> approaches
> at using bwcheck.as and bwcheck.asc, which are
> widely
> used by FCS developers.
>
> The problem is that it seems to be inefficient in
> real-world Internet-based situations with users
> having
> different connection speeds. I'll tell you why (and
> please correct me if I'm wrong). Imagine that the
> first user logs in and his/her bandwidth is checked
> by
> the server. SBeing first one, he/she will have all
> bandwidth at his/her disposal. Great. Bwcheck
> automatically adjusts his settings. The second one
> logs in. But now a great part of the bandwidth is
> being destined to the first user. So the calculation
> is based on the remaining server upstream. As other
> users log in things get worse. Even if they have
> better connections, the automatic adjustment will
> choose settings destined to slower speeds due to
> decreasing server bandwidth available.
>
> You could consider testing again every time someone
> logs in. The problem in that case is that audio and
> video are probably already being transmitted by all
> users. So the test loses its accuracy.
>
> This is the situation we're dealing with in our
> tests.
> I wonder if someone was faced with similar scenarios
> and was able to find a solution.
>
> Regards,
>
> W
>
>
>
> __________________________________
> Do you Yahoo!?
> Yahoo! Small Business - Try our new resources site!
> http://smallbusiness.yahoo.com/resources/
>
>
=-----------------------------------------------------------
> Supported by Fig Leaf Software -
> http://www.figleaf.com
>
=-----------------------------------------------------------
>
> To change your subscription options or search the
> archive:
>
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
>
=-----------------------------------------------------------
> Supported by Fig Leaf Software -
> http://www.figleaf.com
>
=-----------------------------------------------------------
>
> To change your subscription options or search the
> archive:
>
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
________________________________________
__________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Andy LeMay 2005-04-28, 5:45 pm |
|
> ------------Original Message------------
> From: Peldi Guilizzoni <gguilizzoni-14osZcCZf762oZ/6fjIToQ@public.gmane.org>
> Everyone who uses this bw detection code (including Amazon.com I think) h=
as
> not reported any problems with it.
I can confirm this, at least as far as Amazon is concerned. The bugfix wor=
ked wonders for that project.
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Stefan Richter 2005-04-29, 2:45 am |
| I have tried to combine the flvplayer code with the one from Peldi's blog
but failed miserably.
It would be very cool if someone could release a version which
A) fixes the aforementioned bugs and
B) also detects upload speed reliably
The flvplayer code only tests download speed which is great for video
delivery but not so good for apps that use live webcam feeds or recording
apps.
Peldi, is there any chance of getting an updated, 2 way bandwidth checker
from you?
If not, do you have a fix for the payload bug which so far has been unknown
to me... :-\
Thanks
Stefan
> -----Original Message-----
> From: flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
> [mailto:flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org] On Behalf Of
> Peldi Guilizzoni
> Sent: 28 April 2005 19:47
> To: FlashComm Mailing List; FlashComm Mailing List
> Subject: RE: [FlashComm] bwcheck: is it a good solution?
>
> Ha! You tickled my memory. I think that the bwcheck.asc code
> has a bug in which it doesn't reset the payload properly
> after the first person uses it.
> This would result in what you're seeing. You should
> definitely move to the bw checker code in the FLVPlayer.zip,
> all bugs have been fixed in that one.
>
> Peldi
>
> -----Original Message-----
> From: flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
> [mailto:flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org]
> Sent: Thursday, April 28, 2005 11:07 AM
> To: FlashComm Mailing List
> Subject: RE: [FlashComm] bwcheck: is it a good solution?
>
> Hello Peldi,
>
> Thanks a lot for your fast response. By the way, your book is
> being of great help to our team.
>
> But I think I wasn't able to describe my situation properly.
> The application we're working on is a 6-seat audio/video
> meeting room. That application also has features such as
> whiteboard, chat, text editor, file transfer etc. We're
> testing it using Influxis as our host with a 1.25 Mbps connection.
>
> Our tests indicate that first user is granted a greater share
> of the server's bandwith, while those who log in later have
> their tests impaired by server bandwidth already destined to
> previous users. At least that's our assumption.
>
> Under those circumstances, is it reasonable to have all users
> re-check whenever a new user logs in as you proposed? Do you
> think it'll work?
>
> Regards,
>
> W
>
> --- Peldi Guilizzoni <gguilizzoni-14osZcCZf762oZ/6fjIToQ@public.gmane.org>
> wrote:
> bw required
> link to the
> regular FLV,
> problem for
> sending it).
> develop an
> widely used
> is checked
> settings.
> bandwidth is being
> was able to
> =-----------------------------------------------------------
> =-----------------------------------------------------------
> http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
> =-----------------------------------------------------------
> =-----------------------------------------------------------
> http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
> ________________________________________
__________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
>
> =-----------------------------------------------------------
> Supported by Fig Leaf Software - http://www.figleaf.com
> =-----------------------------------------------------------
>
> To change your subscription options or search the archive:
> http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
> =-----------------------------------------------------------
> Supported by Fig Leaf Software - http://www.figleaf.com
> =-----------------------------------------------------------
>
> To change your subscription options or search the archive:
> http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Sean Moran 2005-04-29, 5:45 pm |
|
Along the same lines, any good examples on how to use this code with a
CDN like VitalStream?
The bw check in there component failed miserably on a dialup connection
- consistently reported connection speeds of over 300kbps!
TIA,
Sean
On Apr 29, 2005, at 1:54 AM, Stefan Richter wrote:
>
> I have tried to combine the flvplayer code with the one from Peldi's
> blog
> but failed miserably.
> It would be very cool if someone could release a version which
> A) fixes the aforementioned bugs and
> B) also detects upload speed reliably
>
> The flvplayer code only tests download speed which is great for video
> delivery but not so good for apps that use live webcam feeds or
> recording
> apps.
>
> Peldi, is there any chance of getting an updated, 2 way bandwidth
> checker
> from you?
>
> If not, do you have a fix for the payload bug which so far has been
> unknown
> to me... :-\
>
> Thanks
>
> Stefan
>
>
>
>
>
> =-----------------------------------------------------------
> Supported by Fig Leaf Software - http://www.figleaf.com
> =-----------------------------------------------------------
>
> To change your subscription options or search the archive:
> http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Brian Lesser 2005-04-29, 5:45 pm |
| Hi Sean,
http://flash-communications.net/sourceCode/index.html
Under chapter 17 there is a file called ConnectionTester.
It may require significant adaptation for your needs and the latest
version has not been extensively tested. Still, you may find it useful.
Yours truly,
-Brian
Sean Moran wrote:
[vbcol=seagreen]
>
> Along the same lines, any good examples on how to use this code with a
> CDN like VitalStream?
> The bw check in there component failed miserably on a dialup
> connection - consistently reported connection speeds of over 300kbps!
>
> TIA,
> Sean
>
> On Apr 29, 2005, at 1:54 AM, Stefan Richter wrote:
>
--
________________________________________
______________________________
Brian Lesser
Assistant Director, Teaching and Technology Support
Computing and Communications Services
Ryerson University
350 Victoria St.
Toronto, Ontario Phone: (416) 979-5000 ext. 6835
M5B 2K3 Fax: (416) 979-5220
Office: AB48D E-mail: blesser-6s6ziW1YCwCw5LPnMra/2Q@public.gmane.org
(Enter through LB66) Web: http://www.ryerson.ca/~blesser
________________________________________
______________________________
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Sean Moran 2005-04-29, 5:45 pm |
|
Thanks Brian...hadn't made it that far in the book yet :-)
Sean
On Apr 29, 2005, at 9:46 AM, Brian Lesser wrote:
>
> Hi Sean,
> http://flash-communications.net/sourceCode/index.html
> Under chapter 17 there is a file called ConnectionTester.
> It may require significant adaptation for your needs and the latest
> version has not been extensively tested. Still, you may find it
> useful.
> Yours truly,
> -Brian
>
>
>
> Sean Moran wrote:
>
>
>
> --
> ________________________________________
______________________________
> Brian Lesser
> Assistant Director, Teaching and Technology Support
> Computing and Communications Services
> Ryerson University
> 350 Victoria St.
> Toronto, Ontario Phone: (416) 979-5000 ext. 6835
> M5B 2K3 Fax: (416) 979-5220
> Office: AB48D E-mail: blesser-6s6ziW1YCwCw5LPnMra/2Q@public.gmane.org
> (Enter through LB66) Web: http://www.ryerson.ca/~blesser
> ________________________________________
______________________________
>
>
> =-----------------------------------------------------------
> Supported by Fig Leaf Software - http://www.figleaf.com
> =-----------------------------------------------------------
>
> To change your subscription options or search the archive:
> http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Wilton Rossi 2005-04-29, 5:45 pm |
| Hello Brian,
I'm ready to have a go at ConnectionTester. But I
think one file is missing. ConnectionTester.asc
(inside pfcs/components) makes reference to
RMILatencyTester.asc. That file should be in the same
folder, but it isn't. I couldn't find it anywhere in
the zipped file.
Regards,
Wilton
--- Brian Lesser <blesser-6s6ziW1YCwCw5LPnMra/2Q@public.gmane.org> wrote:
> Hi Sean,
>
http://flash-communications.net/sourceCode/index.html
> Under chapter 17 there is a file called
> ConnectionTester.
> It may require significant adaptation for your needs
> and the latest
> version has not been extensively tested. Still, you
> may find it useful.
> Yours truly,
> -Brian
>
>
>
> Sean Moran wrote:
>
> use this code with a
> on a dialup
> speeds of over 300kbps!
> the one from Peldi's
> version which
> which is great for video
> webcam feeds or
> 2 way bandwidth
> which so far has been
> On Behalf Of
> List
> solution?
> bwcheck.asc code
> properly
> should
> FLVPlayer.zip,
> solution?
> your book is
> situation properly.
> audio/video
> such as
> etc. We're
> 1.25 Mbps connection.
> greater share
> later have
> destined to
> have all users
> proposed? Do you
> <gguilizzoni-14osZcCZf762oZ/6fjIToQ@public.gmane.org>
> happen if the
> your server's
> (~200Kbps for a
> connection).
> 10 people connect and
> shouldn't be a
> them from my blog
> latest fixes (which
> data before
> (including Amazon.com I
> [mailto:flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org]
> solution?
> FCS-based
> problems now is to
> trying different
> which are
>
=== message truncated ===
________________________________________
__________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Michael 2005-04-29, 5:45 pm |
| I won't really be able to look at this until tonight. Let put our =
meeting
off until next week so I have a chance to play with your content a bit =
and
get it up on the site.=20
I'm heading to Palm Springs next Friday afternoon -- hopefully before
traffic starts to get bad. Friday morning would be fine. Otherwise we =
should
make it some evening.
-- Mike
-----Original Message-----
From: flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org
[mailto:flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org] On Behalf Of Wilton =
Rossi
Sent: Friday, April 29, 2005 11:00 AM
To: FlashComm Mailing List
Subject: Re: [FlashComm] bwcheck: is it a good solution?
Hello Brian,
I'm ready to have a go at ConnectionTester. But I
think one file is missing. ConnectionTester.asc
(inside pfcs/components) makes reference to
RMILatencyTester.asc. That file should be in the same
folder, but it isn't. I couldn't find it anywhere in
the zipped file.
Regards,
Wilton
--- Brian Lesser <blesser-6s6ziW1YCwCw5LPnMra/2Q@public.gmane.org> wrote:
> Hi Sean,
>
http://flash-communications.net/sourceCode/index.html
> Under chapter 17 there is a file called
> ConnectionTester.
> It may require significant adaptation for your needs
> and the latest=20
> version has not been extensively tested. Still, you
> may find it useful.
> Yours truly,
> -Brian
>=20
>=20
>=20
> Sean Moran wrote:
>=20
> use this code with a=20
> on a dialup=20
> speeds of over 300kbps!
> the one from Peldi's=20
> version which
> which is great for video
> webcam feeds or=20
> 2 way bandwidth=20
> which so far has been=20
> On Behalf Of
> List
> solution?
> bwcheck.asc code
> properly
> should
> FLVPlayer.zip,
> solution?
> your book is
> situation properly.
> audio/video
> such as
> etc. We're
> 1.25 Mbps connection.
> greater share
> later have
> destined to
> have all users
> proposed? Do you
> <gguilizzoni-14osZcCZf762oZ/6fjIToQ@public.gmane.org>
> happen if the
> your server's
> (~200Kbps for a
> connection).
> 10 people connect and
> shouldn't be a
> them from my blog
> latest fixes (which
> data before
> (including Amazon.com I
> [mailto:flashcomm-bounces-1Ss2GqJETD3yZ38Mhd3e/9ZfFG6BLHNm@public.gmane.org]
> solution?
> FCS-based
> problems now is to
> trying different
> which are
>=20
=3D=3D=3D message truncated =3D=3D=3D
________________________________________
__________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around=20
http://mail.yahoo.com=20
=3D-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=3D-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
| |
| Brian Lesser 2005-04-29, 5:45 pm |
| Acck!!
Here's a file from my machine at work. When I get home I'll check that
its the latest version and update the zip file.
Thanks,
-Brian
Wilton Rossi wrote:
>Hello Brian,
>
>I'm ready to have a go at ConnectionTester. But I
>think one file is missing. ConnectionTester.asc
>(inside pfcs/components) makes reference to
>RMILatencyTester.asc. That file should be in the same
>folder, but it isn't. I couldn't find it anywhere in
>the zipped file.
>
>Regards,
>
>Wilton
>
>--- Brian Lesser <blesser-6s6ziW1YCwCw5LPnMra/2Q@public.gmane.org> wrote:
>
>
>
>http://flash-communications.net/sourceCode/index.html
>
>
>=== message truncated ===
>
>
> ________________________________________
__________
>Do You Yahoo!?
>Tired of spam? Yahoo! Mail has the best spam protection around
>http://mail.yahoo.com
>
>=-----------------------------------------------------------
>Supported by Fig Leaf Software - http://www.figleaf.com
>=-----------------------------------------------------------
>
>To change your subscription options or search the archive:
>http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
>
--
________________________________________
______________________________
Brian Lesser
Assistant Director, Teaching and Technology Support
Computing and Communications Services
Ryerson University
350 Victoria St.
Toronto, Ontario Phone: (416) 979-5000 ext. 6835
M5B 2K3 Fax: (416) 979-5220
Office: AB48D E-mail: blesser-6s6ziW1YCwCw5LPnMra/2Q@public.gmane.org
(Enter through LB66) Web: http://www.ryerson.ca/~blesser
________________________________________
______________________________
| |
| Wilton Rossi 2005-04-29, 5:45 pm |
| Brian,
Thanks a lot! ConnectionTester is now working and
we're already trying to adapt it to our applications.
Regards,
Wilton
--- Brian Lesser <blesser-6s6ziW1YCwCw5LPnMra/2Q@public.gmane.org> wrote:
> Acck!!
> Here's a file from my machine at work. When I get
> home I'll check that
> its the latest version and update the zip file.
> Thanks,
> -Brian
>
> Wilton Rossi wrote:
>
> same
> in
>
>http://flash-communications.net/sourceCode/index.html
> needs
> you
> wrote:
> updated,
>
>
>
=== message truncated ===> function
RMILatencyTester(owner, client){
> this.owner = owner;
> this.client = client;
> this.messageCount = 0;
> this.maxMessages = 5;
> this.data = [];
> this.sendTime = (new Date).getTime();
> this.client.call(this.owner.path +
> "/getClientTime", this);
> }
>
> RMILatencyTester.prototype.numericCompare =
> function(aObj, bObj){
> var a = aObj.dt;
> var b = bObj.dt;
> if (a < b) return -1;
> else if (a > b) return 1;
> else return 0;
> }
>
> RMILatencyTester.prototype.onResult =
> function(clientDateTime){
> this.receiveTime = (new Date).getTime();
> //trace("RMILatencyTester.prototype.onResult> " +
> this.receiveTime);
> this.messageCount++;
> this.data.push({rt:this.receiveTime,
> dt:this.receiveTime -
> this.sendTime,
> ct:clientDateTime});
> if(this.messageCount < this.maxMessages){
> this.sendTime = (new Date).getTime();
> this.client.call(this.owner.path +
> "/getClientTime", this);
> }
> else {
> this.data.sort(this.numericCompare);
> var median =
> this.data[Math.floor(this.data.length/2)];
> var upperLimit = median * 1.5;
> var sum = 0;
> var csSum = 0;
> var n = 0;
> for (var i =0; i < this.data.length; i++){
> var dt = this.data[i].dt;
> if (dt > upperLimit) continue;
> sum += dt;
> csSum += clientDateTime.getTime() -
> this.receiveTime + dt/2 ;
> //trace(clientDateTime.getTime() + " - " +
> this.receiveTime + " + " + dt/2);
> n++;
> }
> this.owner.onClientLatency(this.client, sum/n,
> csSum/n);
> }
> }
>
> trace("Loaded:
> pfcs/components/RMILatencyTester.asc");
>
>
=-----------------------------------------------------------
> Supported by Fig Leaf Software -
> http://www.figleaf.com
>
=-----------------------------------------------------------
>
> To change your subscription options or search the
> archive:
>
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
>
________________________________________
__________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
|
|
|
|
|