Date: Thu, 29 Jan 2004 19:47:14 +0800 From: Raymond <[EMAIL PROTECTED]> Subject: Re: [LIB] Networking problems
At 11:39 PM 28/01/2004 -0800, you wrote:
Date: Thu, 29 Jan 2004 07:38:46 +0000 From: "Matt Hanson" <[EMAIL PROTECTED]> Subject: Re: [LIB] Networking problems
Thanks for all the responses on this. I originally intended to address everyone's kind replies to this problem tonight, but I'll just be quick for the moment..
I guess I was misinformed about the crossover cable slowing down my transfer speeds from one system networked to another. I've asked people about that for a couple years at this point, but everyone either just agreed with me, or thought going to a LAN switch would speed things up. Not understanding network cables, and after the notion got put into my head, I just thought there was something about a crossover cable that slowed things up. But the switch does indeed slow things down.
Yes, I'm using twisted pair CAT 5 cables that were pulled out of a system that was professionally installed at some point. And my crossover cable was build for me locally a couple years back. The Xircom combo, Linksys LAN card, and NIC in the desktop are all 10/100 compliant.
I tried changing all the driver settings for all NICs from 'Auto' to 100Kbps (yes >bits< is what I'm talking about) full duplex. But the tranfser rates with those set were exactly the same as without it.
That rate of 4.7Mbps was between my L70 and L100, and the L70 and the PII450 desktop. The L100 does a bit better at 6.8 Mbps. But maybe someone can check my math. I transferred 280MB of MP3s, and timed the process. The average was 476 seconds from L70 to L100. So:
280MB x 8B/b = 2240Mb / 476sec = 4.706Mbps
There's been some posts on using USB2 and Firewire on the L1x0 systems, so I just assumed that
people were getting greater than 10Mbps transfer rates out out their systems that way. But if 4.7 - 6.0 Mbits/sec is all the L100 can muster, I really don't understand Firewire or USB on these little babys.
Also, since I tested xfer rates both between Libbys and beteen each of them and the desktop with the same results, I guess components are not the problem. I guess it's just a limitation of the CPU, and what supports and augments it.
Hi Matt,
There is a logical explanation for all this and I'm almost certain the problem here is CPU power. There is a big difference between dumping digital videos through firewire at almost 30Mbps and only managing 5Mbps through a network card. Firstly, firewire cards (and, to a lesser extent, USB2.0 devices) have onboard controllers. In my card (and I assume almost all firewire cards), the firewire stream goes straight to system memory via the DMA (direct memory access, allows devices to read/write to/from main memory independant of the system processor) controller, bypassing the CPU altogether. All the CPU has to do is grab the data out of memory and shove it out to disk (the video is uncompressed and, as far as I'm aware, comes across the firewire cable in a format that is almost like uncompressed AVI anyway). In fact, if the video is already interleaved correctly, all the CPU needs to do is monitor the DMA controller until it says it has a block of data then tell the IDE controller (also on DMA) to grab data from that same block of memory - the CPU actually does the least amount of work ... one test I'm yet to do is see what the real CPU usage is when I dump digital video (I wouldn't trust the Windows CPU monitor for this sorta stuff because I don't know what the interactions are with the DMA controller but I intend on running something that doesn't hit the hard drive in the background and seeing how much I can do before it croaks). For straight copy operations on USB2.0 devices, I imagine it's a similar case although, from what I understand, USB2.0 controllers don't take quite so much load off the CPU compared to firewire controllers.
In contrast, a network card presents the CPU with network packets which must be processed, error checked, resent if corrupted and so-on as specified by the various protocols involved (TCP/IP and SMB if you're using Windows filesharing over IP for instance). The CPU needs to do a lot of work before it can unwrap the data encapsulated in a network transmission and I can well believe that the poor L70 or even the L110 will croak once you get around the 6Mbps mark.
Hope this explains a thing or two!
- Raymond
---
/~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\
| | "Does fuzzy logic tickle?" |
| ___ | "My HDD has no reverse. How do I backup?" |
| /__/ +-------------------------------------------|
| / \ a y b o t | [EMAIL PROTECTED] |
| | Need help? Visit #Windows98 on DALNet! |
| ICQ: 31756092 | Libretto IRC channel #Libretto on DALNet! |
\~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~/
************************************************************** http://libretto.basiclink.com - Libretto mailing list http://www.silverace.com/libretto/ - Archives
-------TO UNSUBSCRIBE-------
Reply to any of the list messages. The reply mail should be
addressed to: [EMAIL PROTECTED] - Then replace any text
on the message's subject line: cmd:unsubscribe
--------TO UNSUBSCRIBE DIGEST------
Do above but with this on subject line: cmd:unsubscribe digest
**************************************************************