Slashdot Mirror
Playing Music Slows Vista Network Performance?
An anonymous reader writes "Over the months since Vista's release, there has been no doubt about the reduced level of network performance experienced compared to Windows XP. However, some users over at the 2CPU forums have discovered an unexplained connection with audio playback resulting in a cap at approximately 5%-10% of total network throughput. Whenever any audio is being sent to a sound card (even, several users report, while paused), network performance is instantly reduced. As soon as the audio is stopped, the throughput begins to climb to its expected speed. It's a tough one for users — what do you pick, sound or speed? So much for multi-tasking."
Its not that Vista's network performance drops when an audio file is played. Rather, Vista uses part of the network bandwidth to phone home to Redmond. This is pure speculation, but it logically makes sense considering MS's track record.
I love it when my possessions spy on me!
I actually have a 10% faster internet connection! All I need to do is to upgrade to Linux.
"Luke, you've switched off your targeting computer, what's wrong?"
"This is FUD of the highest quality. I'm sitting now with my Vista edition listening to some Iron Maiden's mpr (DRM free of course) and using Windows Media Player 11. I'm measuring the network speed by sending a 34 GB files to the server with the player working and without it. The dspeed is being measured using BMST (Bandwidth Meter Speed Test). No difference at all."
/. and any sheep will just believe it without further inverstigation..."
Right. YOU don't seem to be having the problem, so there is no problem.
"Of course you can write anything you want negatve about MS in
Troll.
"Gold still represents the ultimate form of payment in the world." - Alan Greenspan, 1999
maybe that preview button wiould help.
or perhaps posting in your native tounge?
loose jumbles of english sounding words do not make paragraphs.
-- Sig under construction...
iPods don't play MP3's, they play AAC files.
1984 was not supposed to be an instruction manual.
I never understand why people complain about MS's costs for upgrades... then go and buy a Mac, which is an expensive hardware dongle for pretty much a BSD variant. Its like complaining about the performance of a Ferrari's radio, and instead of replacing the radio, buying a Chevy Aveo or some other subcompact (paying $100k for it), saying its better than the Ferrari because its made by Chevy and ignoring everything else that matters in a car (0-60 times, crash tests, horsepower, handling, and so on).
If people don't like MS, go Linux or BSD and keep the same hardware. Most Linux distros made in the past couple years work with almost any hardware you can throw at them, and can do 90-95% of what Windows can do.
I just don't get the rationality of replacing a PC with another overpriced and underpowered PC (the new 2007 iMac video chipsets sell for $50-$85 if you want to buy a PCI-E card with one) with an Apple logo, and a certificate in the TPM chip (so MacOS can boot.)
If people want better than PC hardware, there is always Sun, or IBM. Apple's hardware is a joke (and a bad one at that) compared to true high end offerings.
I think its the same neurological pathway that gets people to buy Macs as it is for people to care more about Paris Hilton's escapades in jail than if the bridges in their city will collapse in the next several years.
Lets see, I can play an mp3 while transferring a file at 120 mega bytes per second over an iSCSI interface hooked up to a pair of 28 disk fiber filers. I can transfer to a SATA RAID from a SATA RAID at 60 mega bytes a second. Playing an MP3 in media player or a movie, even a 1080P movie, causes no effect at all on the network speed during the copy (a 8gb ISO file). I'm pretty sure 1000 mega bits per second is equal to 120 mega bytes a second. If I can use my iSCSI NAS at 100% and play a 1080P video at the same time then I'm guessing just about anything would work without the described behavior.
I would assume spending 5 minutes running media player and performance monitor together to see if the story is true wouldn't be too much trouble. That might might make slashdot editors talk to a dirty vista OS user, let's just assume if the news is negative then the news must be true.
Why not just spend 1 interrupt to transfer the whole 8 to 20meg of data to ram in the ONE GO, then play the mp3 from ram.
Surely the 20meg is cheaper than 50000 switches/second. Since many lame apps use lots of ram, 20meg is hardly a killer.
Friggin use a 16-32meg buffer, thats less than 25cents of ram. Any programmer that just a lame ass OO input class that depends on the language/OS to do
its prebuffering is as lazy as a VBasic programmer. Get of your ass, write the 50 lines of code it takes to make a buffer pre loader.
Lesson one, never trust the OS to do all the magic for you, as your code could be written for any OS but if not then any model of the one OS.
Yes, all windows caches are dumb, they are not content dynamic in nature, they treat all data equally which is wrong, caching a dvd is different to caching
a txt file or an mp3 file. Some require more , some require little.
Liberty freedom are no1, not dicks in suits.