>>>> If you look at the bottom bar of some browsers (eg. IE), >>>> >>> and you've fallen in to a bad trip. > > Ment to say "bad trap".
Fair enough but my point is the same. My IE well enough indicates the download speed. There is no practical point in getting too fancy. I have made the point about the important practical figures and ... well ... what more can I say? Plenty actually... but I will try to control myself. You do not need fancy metrics or to be a rocket scientist to know how fast files come down to you. If you download a file, note how long it takes. This is how long it takes. If it takes a time and it is a size then this gives a rate. If you do this a few times, or just pay more than no attention then you get to know your rough rate and that is good enough for the purpose of knowing whether it pisses you off too much or not. The mac won't lie to you about file sizes unduly (look at the figure in brackets in Get Info). Your wrist watch won't lie to you, it is too vulnerable. It knows about sledgehammers. Do this a few times and you simply know the rough average on your set up, the snail goes fast sometimes, and the snail nearly stops at other times, mostly it is somehere in the middle. Never mind metricating. My IE bar is a nice bar and it NEVER lies to me, when it says the speed is down to the bytes, I can just see it is right. The very sight of "bytes" under three digits is simple confirmation of why it is taking ages to get anything (in case you don't notice). It is not falling into trips or traps to remember and use the old analogue ways of seeing dials and figures as indicators of trends... >>>> Now, in a modern city, it is rare for anyone to get much >>>> more than 5K/sec and certainly not for long. >>>> >>> City has nothing to do with it. >>> >> >> City has nothing to do with what? > > I think you ment that backwards. You're trying to say that the > infrastructure in cities is better, so faster speeds may be obtained?. > > My point is that city or country makes little diff. A CO is a CO and > the wires only go so far from it. If they're noisy, you're screwed - > regardless of having pidgeons or cows pirching on them. > > ( CO = Central Office network switch ) Yes, I meant that city is better in infrastructure, at least over here in Australia. The copper lines, the connections and exchanges are better in the city. Things are better kept and renewed. If the lines are noisier in the bush, as they are here, then how can it make little difference? I think we need to distinguish between the average condition in the country compared to the city and if it is not the COs and the wires that go from it, then it is something else, maybe the number of them over large distances that sum little faults into bigger ones. You can't spirit away differences in performances by suddenly picking some rabbit (COs and the wires that go only so far from them) out of a hat and saying Hey Presto. >>> A decent V.90 connection is a RAW 4 to 5 KB/sec. But when you >> add V.42's bit >>> stripping and V.42bis data compression... much higher througput is easily >>> attainable --- especially over the web, where much of the traffic is easily >>> compressable ascii text (html, scripts, etc). >>> >>> When transferring raw binary data, that's mostly uncompressable, you'll see >>> the throughput rate drop to 20% higher than your raw carrier >>> speed. Why 20%? >>> Because even tho the V.42bis data compression is defeated, V.42's bit >>> stripping transmits only 8 bits per byte instead of 10 (no start and stop >>> bits). >>> >> >> I am interested in this Dan, what to do exactly to get better results than >> roughly average 3.5k per sec? > > First, use a reliable metric. Something like IPNetRouter - that > isn't succeptable to your browser's skew. > > Next, take a hard look at your phone line. Determine where the noise > is - inside your home or outside. Get that clean that up. > Eliminating the phone line noise is 90% of the battle. ...Usually to > drop brought to homes is more than just one pair of phone wires. > This means the telephone company can switch you to an alternate pair > quite easily. This can often reduce noise problems (after you've > determinate the problem is outside your home). I've said my piece about "reliable metrics". It is the last thing, not the first thing, anyone needs in this matter. Personally, I seem to have a quiet clean line now. And have no real complaints (I did not raise this subject). When it was not clean - once - I did things. I know no one at all who gets better than me in Sydney on 56K. But that may not mean a lot, I know only so many people. > > Outside vs Inside... It's amazing how much noise comes from sources > in your own home. Answering machines, fax machines, other > telephones, flourescent lights, etc... I am sure you are right. I have been thru this but in my case it makes no difference to my speeds when disconnecting everything else, no lights in day etc. I even found (I was told by others about it) a bit of exotica: a particular standard type of phone that our Telecom used to distribute had a little battery that drew extra current every 20 min or so to charge the battery and this could affect data on line... >> This V.42 business, we talking the need to get different modems here? > > V.42 Error Correction and V.42bis Data Compression are part of all > V.90 modems. But they don't "activate" until the line noise is below > an acceptable threshold. See above. So it is down to noise then. What are other folk on this list getting? What are you actually getting. And please, no fancy figures. What are folks downloading files in on average. Today, for example I have been downloading tons of stuff and it has not got much better than 2.5 K per sec. I think things are getting worse than they used to on 56K because of increasing traffic. 2.5 or 3.5 K a sec is still slow. Who amongst us all on 56K V90 are in fact getting much better generally (no remarkable incidents being relevant)? Those who are getting better, I am most interested in getting any tips from if at all relevant to my city and set up. >> PS. Practical advice: those who have a lot to do on their computers, do some >> of it while pages load in the background. Have a few things going at once >> and it is less frustrating. Eg. when checking out Ebay, open up several >> items in new windows (make them small, up in the corner) and they load up >> while you are browsing the list of auctions... No offence to those who >> already do this. It is a habit that is not always acquired even tho it may >> seem obvious. > > I do this a lot. I have both iCab and Mozilla running, each loading > perhaps a dozen pages simultaneously - while I'm chatting on IRC or > doing email or .... Definately keeps da Mac busy. > > - Dan. Thanks Dan for forwarding me a copy of your post. Best wishes... David Elmo PS. I downloaded a terrific astronomy package from ftp://ftp.ausmac.net/pub/mac/ recently (16 MB). I did it in my Transmit FTP program. FTP always starts fabulously, first few sec over 7K/sec but then reality hits and it trundles along at about 2.5 or so ... I replaced the driver's side of my front disc pads and repacked the wheel bearings on my Ford XY 4.2 Litre and the download was complete when I came in. What should I download for the passenger side Dan? -- PCI-PowerMacs is sponsored by <http://lowendmac.com/> and... Small Dog Electronics http://www.smalldog.com | Refurbished Drives | -- Sonnet & PowerLogix Upgrades - start at $169 | & CDRWs on Sale! | Support Low End Mac <http://lowendmac.com/lists/support.html> PCI-PowerMacs list info: <http://lowendmac.com/lists/pci-powermacs.shtml> --> AOL users, remove "mailto:"; Send list messages to: <mailto:[EMAIL PROTECTED]> To unsubscribe, email: <mailto:[EMAIL PROTECTED]> For digest mode, email: <mailto:[EMAIL PROTECTED]> Subscription questions: <mailto:[EMAIL PROTECTED]> Archive:<http://www.mail-archive.com/pci-powermacs%40mail.maclaunch.com/> Using a Mac? Free email & more at Applelinks! http://www.applelinks.com
