Preshant; When you have said "I have read from somewhere", many list members are also members of the list where, earlier today, you read my posting. I don't think Kon is currently a member of WMTalk (if so, he hasn't posted for about a year) but many, many others are. Attribution of sources is advantageous; it brings even list members who aren't paying attention (or aren't on other apt lists) to get up to speed. It also enables people to weigh the sources. Not providing attribution means that you are close to providing rumors. I would note that most of my posts on this topic are "try it yourself" and "don't trust anybody's statements." You can't test these systems via email, or let email communications make up for testing. You'll just fail at a higher level. John Willkie ----- Original Message ----- From: "Prashant Desai" <prashant.desai@xxxxxxxxxx> To: <opendtv@xxxxxxxxxxxxx> Sent: Tuesday, May 17, 2005 2:01 PM Subject: [opendtv] Re: wm trials at different telcos using wm series > Hi Kon > > Thanks a lot for all the clarification , > > It seems whoever is using wm9 ,most of them claiming that its pretty > feature rich compared to the advanced codec MPEG-4 AVC , infact somewhere > I have read that VC1 which is just subset of wm9 deliver the same quality > video at approx half of the rate at which the mpeg-4 avc delivers the same > video with same quality ..... and also that VC1's compression mechanism > automatically gets improved with the increase of the bit rate [ encoding ] > and thus it can compress data more effectively at higher bit rates , > > I personally have never seen the quality of video for any of these MPEG-4 > AVC or wm9 or VC1 on a TV set at same rate ...... > > Are there any test results that been published by an independent entity for > these ? If there is please point me to it, since we really in dilemma > whether to go with mpeg-4 AVC or with VC1 or with proprietary wm9? > > Pls guide me on this > > Regards, > Prashant > > -----Original Message----- > From: opendtv-bounce@xxxxxxxxxxxxx [mailto:opendtv-bounce@xxxxxxxxxxxxx] On > Behalf Of Kon Wilms > Sent: Tuesday, May 17, 2005 11:25 PM > To: opendtv@xxxxxxxxxxxxx > Subject: [opendtv] Re: wm trials at different telcos using wm series > > On Tue, 2005-05-17 at 23:01 +0530, Prashant Desai wrote: > > I am curious , the list members views on the windows media server [ wm > > series ] , is it a truly a carrier grade solution > > I wouldn't call putting windows 2003 and windows media server on an off > the shelf server carrier grade. Try pulling the power plug and waiting > for the reboot and checkdisk, then having all the clients reconnect to > the box after a minute or two of downtime. > > > Does MS DRM can be used for securing/encrypting ( CAS functionality > )the > > live Broadcast TV channels ? which can be decoded on the PC or on the set > > top box > > You can't do a full-blown CAS using the off-the-shelf MS DRM rights > manager and windows media server. There is no SMS integration for one. > > DRM is file encryption and rights management. You're talking about pipe > encryption and entitlement management. These are two different things. > > > Running windows CE OS , if MS DRM cant be used for encryption of the live > > Broadcast TV channels then are there any other vendors who supports wm9 > > format > > Try any CA vendor? You're talking about pipe encryption. You don't need > to know or care what the underlying protocol is (although with the > fragmentation of large frames that windows media server/encoder does by > default, I can see this being an issue). > > > I was just surprised when I came to know that windows media server is > > capable of supporting few 10 thousands of unicast streams simultaneously > > > > At the rate of 512 KBPS to 1MBPS to 1.5 MBPS for viewing the same on > > either PC or on TV sets. > > Why? You can do this one of two ways: > > 1. You just clone the socket for every unicast connection - you're > really reading the pre-encoded source file only once. Everything else is > an issue of the bus and network stack. > 2. You load portions of the file/stream into buffered memory pools and > serve them 10,000 times. 10,000 memcpy functions wont even break a > sweat. > > Ofcourse both can go south really fast if a client has any network > issues and starts lagging to the end of the buffer (at which point you > need to drop him and restart the data flow). So you need to trade off > network reliability against the size of the buffer in seconds. > > No rocket science there.... > > Cheers > Kon > > > > > ---------------------------------------------------------------------- > You can UNSUBSCRIBE from the OpenDTV list in two ways: > > - Using the UNSUBSCRIBE command in your user configuration settings at > FreeLists.org > > - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word > unsubscribe in the subject line. > > > > ---------------------------------------------------------------------- > You can UNSUBSCRIBE from the OpenDTV list in two ways: > > - Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org > > - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word unsubscribe in the subject line. > ---------------------------------------------------------------------- You can UNSUBSCRIBE from the OpenDTV list in two ways: - Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word unsubscribe in the subject line.