Never tried getting my Revo online, but the mobile phone connection via infra-red idea sounds just stupid, especially on a bus. It would be marginally more useful, though, had I a mobile phone...

Quoting ilovemydualg4:
btw.... with the phantom ppp connection... what was the isp? ... i've had same problem
What, disconnect after about 15 minutes, the first time each day? An odd co-incidence - my ISP is ntlworld, or whatever they are now after having been bought up (my hostmask is now *.ntli.co). Maybe they've done the same thing to their PPP server?

Apologies for not replying thus far to any of you on the matter. I did compose two different replies but was never happy with either. The short of the matter is that I get a good deal with my ISP and have no plans to change at the moment. Also, the problem seems to only occur every few days at the moment, so I'm not sure what's triggering something to change. A simple possible solution is to leave off connecting AIM, IRC and ICQ until the first connection has died, but then, what if it isn't going to die?

I'm guessing that the actual solution lies in the modem script. I've looked at it, and don't forsee myself learning modem scriptish any time soon, but the only part that I think needs changing is either the modem's init string, or maybe one of the other command strings. A Google search for the problem yielded nothing, so I guess I'll just have to live with it. Way too lazy and unmotivated to learn modem strings and spend a few decades screwing with them until it works :)

Thanks anyhow.

fill in the values and give it a few shots, if you really feal it is a modem script error, try making a new one with that

quote:

---

Originally posted by Rednivek:
MacGenius is a fraud!

---

as i said, i don't want to start a war here, especially when i'm on vacation but...

heheh that was funny

quote:

---

Originally posted by Rednivek:
MacGenius is a fraud!

---

Thanks anyhow

quote:

---

Originally posted by Rednivek:
MacGenius is a fraud!

---

With a name like that he had to be.

He is not

Thankfully, though, Rednivek gets to be spared this most odd of punishments.

But yes, this entire genius/fraud war is getting tiresome. Give over.

One comes to expect a certain... geniosity from someone with Genius in their name, especially one starting their own Q&A. If you can't live up to what seems promised, then coming across as pretentious seems inevitable.

As I said previously, we're all in this together - we all each have questions to ask, and answers to give to others. Your response to the help I offered regarding that T-shirt ASCII art won't win you any favours - maybe the question was worded such that it implied that an answer was not needed (I'm not too good on communication), but getting angry wasn't the best way to explain that I was mistaken there.

Anyhow, if you truly are a Mac genius, perhaps you'll be happy to field a few nastier questions about the innards of Mac OS. For starters, why do Control Strip modules never call waitNextEvent() when in operation? In fact, why can waitNextEvent() never be called by the kernel during menu/move/scrolling operations? Actually, that isn't entirely true. You see, I run Power Windows, and with live window dragging, the extension's code itself yields time to other apps, such that all apps are running except the one whose window is being dragged, which has made a blocking function call to handle window drag.

Why do Finder copy operations (to floppy discs especially) tie up the CPU? I'm sure I have DMA disc access (hell, BBC Micros have DMA disc access AFAIK), because Disk Copy can write a disc image out to disc and use nearest 0% CPU time. Yet, the Finder seems to be using programmed I/O. And if not, what is it sat there doing?

The same is true of Windows, though, with file operations freezing the current folder window or, with WebDAV access, the entire desktop (and this is a P4 1.4 GHz on a LAN).

And how can I hack an extension with ResEdit to choose what icon it shows on the screen during start-up (it's showing the wrong one)?

Finally, why does the progress bar during start-up get to about 50-80% and then suddenly jump to 100%? What does that last part represent, that my Mac does not have running? Or can it just not add up?

There, have fun with those :)

quote:

---

Originally posted by MacGenius:
Anyway, Rednivek should give over because he started this fight.

---

yeah red, you should give over. Why would you argue with a child anyways?

I assumet hat the bar aproximates so much time per each item. Libraries may take less time? Also, it may jsut be shity aproximation, or it may use winutes�.

When you copy it, the finder may verify everything that you copy, as you go, checking each byte after it is writen, taking up cycles. Disc copy verifies at the end of the imaging

http://docs.info.apple.com/article2.html?artnum=106928&SaveKCWindowURL=http%3A%2F%2Fkbase.info.apple.com%2Fcgi-bin%2FWebObjects%2Fkbase.woa%2Fwa%2FSaveKCToHomePage&searchMode= Expert&kbhost=kbase.info.apple.com&showButton=false&randomValue=100&showSurvey=true&sessionID=anonymous|132560077

(scarily long. I'll see if there's a shorter link)

http://makeashorterlink.com/?H21215131

As I recal, for the icons, open the extension in res edit, then look in the icon resources...
Sure. The default BNDL resource will have an entry corresponding to the extension itself, which maps to the icon to be displayed. However, the BNDL resource only means anything to the Finder (which inserts that creator code/file type->icon mapping into the Desktop Database) it seems, there must be an OS call somewhere for showINITicon (handle *icon) or something, and that isn't giving the OS the extension icon, but a different one, and it looks inconsistent in my marching icons line :)

I assume that the bar aproximates so much time per each item.
It is a common failing across all of Mac OS 8 upwards, as far as I can tell, that the bar only ever reaches about 50-70% (I'd give a more precise figure, were I on my home Mac and content to reboot for the sake of getting it), and then jumps to 100%. But why? It is as if there is some other process that's assigned that portion of the progress bar, and unless your Mac has it installed and running, the progress bar skips it.

When you copy it, the finder may verify everything that you copy, as you go, checking each byte after it is writen, taking up cycles. Disc copy verifies at the end of the imaging
The Finder should be saying, write this disc cluster, and tell me when you're done. Then, read this disc cluster in, and tell me when you're done. All the time that it is waiting for the drive, it should hand the CPU back. The comparison shouldn't take long on a 200 MHz machine.

Now, what happens is that the Mac gets very lagged out, from the Finder doing something that I don't know what. Maybe it is, indeed, the comparison after all, but I never thought it would be that slow. Certainly in Windows, the Explorer/My Computer windows involved can just stop responding altogether, which is, I think, the result of programmed I/O being used. Yet, why does the copying not run in a separate thread so that the interface is freed up? Just Microsoft's stupidity, really.

Actually, the worst drive I have on the Mac is the Zip drive, as SCSI access (Zip, scanner) 0wns the entire machine during data transfers. The data available event, I assume it is, is lower level even that usr VBL tasks, such that MP3 playback even stops during SCSI transfers. Grr.

Even so, I don't really know how Finder copying works in that sense, or why there is a lag, just my guesses.

The program someone posted a URL to for making new modem scripts looks unhelpful as it is for people with technical modem knowledge - I do have one command string in the manual to me knowledge, but whether I can build a script from that, I don't know. Might give that one in your article a try, though.

I'll see. Yesterday the connection dropped while I was gone - it was left downloading, which seems to prevent the dropping I've been noticing, but something else seemed to have gone wrong. Insufficient evidence, at the moment.

http://www.clarkwoodsoftware.com/peekaboo/

it tells you all sorts of stuff about what tasks are running, how much CPU time they're using, how long they've been running etc. and you can use it to kill them individually.

in mac os x, copying huge files to my external firewire drive from my 500mhz ibook showes no delay in speed, next time i have the drive handy, i'll check with top to see how the cycles are being used.

Zip drives btw. ew! cpu-time sucking data losing overheating psu flaming things. ;)
Before you get too harsh, consider the alternatives. CD-RW etc are not suitable for my usage, which is transport of changing data between different machines, and it is also what the university uses, so MO or LS-120 won't help me either. They don't suck CPU time on PCs in themselves (though as I said, Explorer is poor in that sense), it seems to be a problem with SCSI on the Mac in general, as the same problem affects my scanner.

And my Zip drive works just fine, no problems with it ever.

- uilleann

Edit:
Well, the suggestion on that page was stupid - the 56k V.36 script it told me to use (yay, go TomeViewer) script drops my modem speed down to 36 kbit/sec. However, I also found a 56k V.90 script, which makes more sense. Now, when I tried the V.36 script, I did not get disconnected, and with both that and the V.90 script, connecting to AIM and ICQ seems faster (as if the modem were more responsive). It also seems that, with the V.90 script, upload traffic is not hogging the connection like it used to (might be because I'm getting slightly lower transfer rates for the same connected speed (50667)).

The above is all pure speculation and weak evidence, I'll have to give it a week or so before coming to any conclusions. One thing the new one did seem to do was hang up the modem incorrectly, thus causing the next connections to be virtually unusable until I power cycled the modem. On the other hand, the V.36 script (and hopefully the V.90) disconnects me when the OS shuts down, which the Pace script failed to do (and an AppleScript to do that was causing shut downs to stall).

I might have to learn the script language and splice the two together, maybe even look up a modem command reference and manually tune the init strings, which look like random gibberish : )

Time will tell...

Dave is real great, but I think it's best if you can enable appletalk on your windows box, if it is win2k server or win nt. otherwise, dave is nice (though you can obviously use hotline/ftp or some other crosh platform filesharring protocol if it is just a quick few transfers)

...remove splitters and ... line integrity test
No splitters, and given I get a consistent rate of 5.7k/sec transfer, I think the line is just fine, even after my dad hit it with a spade *grins* (wasn't put down deep enough).

You could blow away [the prefs]
Already been down that route.

As for blaming the problem on the ISP... It would be interesting to know what the ISP timeout is if it's exactly 15 minutes
Who says I am? I've considered it as being a problem with either my ISP or my OS. And yes, my timeout is 15 minutes, but that should not, and does not, apply when I am using the connection. I think you've missed the point - it was always the first connection each evening, and is now the first connection of every fourth evening or so. The disconnect might well be the idle time being triggered, but why then is my ISP thinking I've been idle ever since I connected? I think the problem was still present at about 14 minutes after connection even when my ISP's limit was only five minutes, though.

I'm guessing that Zwilnik is the closest with that post about changing modem script, and thus far not enough time has gone by to be able to tell. I want to evaluate performance on this script first (v.90), before returning to the v.34 to see whether I get disconnected on that one (I did get disconnected on the v.90 script). If the v.34 works as stated by Apple, I'll have to dissect them both and ananlyse, compare and merge the modem init strings to find something that works. But, seeing as that would take forever (because I can only get one test result every few days) I might not bother. I'm not that desperate, just wanted to see if anyone had the answer, especially MacGenius.

As far as tracking down the cause of the problem, don't rule anything out. I once knew a beta tester who always had the game she was testing crash at regular intervals, and no-one else could get it to crash. It turned out to be her refrigerator, which was spiking the mains and causing her computer to crash (rather than the game itself).

- uilleann

quote:

---

Originally posted by uilleann:
They don't suck CPU time on PCs in themselves (though as I said, Explorer is poor in that sense),

---

It makes your entire machine slow, or just Explorer? No app can make an entire PC slow per se (pre-emptive multitasking), unless it is tying up system resources that other apps need to run. I can't think of any right now (even hard disc usage should be shared out equally amongst apps), so maybe Windows multitasking isn't as good as it should be? The one obvious example of slowdown, though, is RAM shortage.

As for my Mac, the problem with the Zip drive isn't the drive, it seems, but the SCSI system itself, running on very low-level interrupt. My reasoning for this is that SCSI data transfer halts MP3 playback, and MP3 playback runs on a low enough event (VBL task) that it only stops during paging (not sure if that's due to paging being a lower-level event, or just through the hard disc being tied up).

As for your machine, I recall version_lili complaining that her scanner would mess up MP3 playback, and hers was also connected to the parallel port, so maybe parallel-port implementation is just poor in Windows. Allegedly, the paging system in UNIX even gives the CPU over to other apps when it's waiting for a page of another app to be read in, so why would any OS get in a tangle just transferring data via the parallel port? It's presumably programmed I/O, though, so maybe the OS is compelled to give the I/O-using app lots of CPU time else buffer underruns (in the drive) or overflows (in the I/O controller) would occur.

quote:

---

Originally posted by uilleann:
Time Synchroniser (sorry, Synchronizer) is running, set for every three days; Software Update is set to manual. However, the last time the clock was synchronised was 2:28:28 am today, and no disconnect occurred today; I still harbour doubts that TCP could be the cause of this. I'll wait until I next get disconnected, though, and see if the disconnect time matches the synchronise time (if I remember, that is).

- uilleann

PS Hrm, I guess its my... not modesty, more negative than that, that is concerned that this problem has been blown way out of proportion given that its nearly taken the thread over, that I shouldn't have the... right(?) to do that? Even so, MacGenius seems to have vanished from here altogether...hrm

---

personally i dont think that the time sync could really do it. does anyone pick up the phone exactly at that time (a bit obvious there but...) do you have call waiting? i never would have thought of that if i hadn't jsut seen the sg-1 when thor comes to get sg-1 to help the azguard destroy the replicators, because they "need some one dummer"

Nope. The line is mine, I tell you, all mine, harrrr hahaha hahahah ha ha *coughing fit*

quote:

---

Originally posted by uilleann:
Quoting ilovemydualg4:
[b]does anyone pick up the phone exactly at that time

Nope. The line is mine, I tell you, all mine, harrrr hahaha hahahah ha ha *coughing fit*

- uilleann[/B]

---

we are controlling the broadcast. For the next hour, we will control everything that you see and hear. We control the horizontal, the vertical and the diaganol. We can diluge you with thousands of images, or sharpen one to crystal clarity......
or something like that

quote:

---

Originally posted by uilleann:
Heh, when I start seeing things down the telephone line, I'll know I've really lost it...

- uilleann

---

you mean you don't see them yet?
They are there every time i try to dial up and it says it is the wrong username and or password. Then i spray some poison down the line, and i connect fine....
I think I have temporarily killed them but they have respawned before

antioed:
Oops :-) Oh, and what are you anti- anyhow? Anti-Oxford English Dictionary? :P

Well, I would like to thank most of you for awnsering questions here while I was doing a computer job for somebody and had no time to awnser questions. (Some of you won't mind that anyhow.) But, I would like memphisto to know that his post informing Rednivek to give over, since why argue with a child anyway? Well mphisto, I am sure you did this to make fun of me that I am a childish person. But...

I AM ONLY 12 YEARS OLD ANYWAY! So this was NOT an insult, actually. Heh Heh!

So ya want proof?
Ok, here you go...

To the not-so-official geekculture desktop pattern!

http://www.geekculture.com/ultimatebb/Forum16/HTML/000072.html

Hrm... I'm rofling with all the other laughter-induced doubled-up-ees in here, truly...

One comes to expect a certain... geniosity from someone with Genius in their name, especially one starting their own Q&A. If you can't live up to what seems promised, then coming across as pretentious seems inevitable.

As I said previously, we're all in this together - we all each have questions to ask, and answers to give to others. Your response to the help I offered regarding that T-shirt ASCII art won't win you any favours -maybe the question was worded such that it implied that an answer was not needed (I'm not too good on communication), but getting angry wasn't the best way to explain that I was mistaken there.

Oh uileann, you really shouldn't get so easily offended

Anyhow, if you truly are a Mac genius, perhaps you'll be happy to field a few nastier questions about the innards of Mac OS. For starters, why do Control Strip modules never call waitNextEvent() when inoperation? In fact, why can waitNextEvent() never be called by the kernel during menu/move/scrolling operations? Actually, that isn't entirely true. You see, I run Power Windows, and with live window dragging, the extension's code itself yields time to other apps, such that all apps are running except the one whose window is being dragged, which has made a blocking function call to handle window drag.

Now wait a second - Mac OS 9 has no kernel, and Mac OS X had not CS. So what is exactly going on here?

Why do Finder copy operations (to floppy discs especially) tie up the CPU? I'm sure I have DMA disc access (hell, BBC Micros have DMA disc access AFAIK), because Disk Copy can write a disc image out to disc and use nearest 0% CPU time. Yet, the Finder seems to be using programmed I/O. And if not, what is it sat there doing?

A bug in OS 9. When the Mac first came out, there was no HD. So the Finder was programmed so that the floppy drive is above priorty than other processes. Unfortunealy, this escalated into Mac OS 9 from System 1.0 and is rearing its ugly head when you do any floppy disk related stuff. This is not so in Mac OS X, which better supports mulitasking and the FD is passe anyway. If you move the cursor on an older (this was on my Mac Portable 68000) Mac while there is a DISKOP in progress, the cursor jumps. Other copy operations don't tie up the CPU at all, except burning a CD or DVD, which require lots of CPU power.

The same is true of Windows, though, with file operations freezing the current folder window or, with WebDAV access, the entire desktop (and this is a P4 1.4 GHz on a LAN).

That's just the sluggish performance of Windows. It does that whenever you do something that requires CPU power.

And how can I hack an extension with ResEdit to choose what icon it shows on the screen during start-up (it's showing the wrong one)?

Open the extension, open the icl8 resource, and open the icon that appears at start-up. Then hack it up

Finally, why does the progress bar during start-up get to about 50-80% and then suddenly jump to 100%? What does that last part represent, that my Mac does not have running? Or can it just not add up?

That happens on all Macs, not just yours. Interestingly enough, when I upgraded to OS 9.2.1, it went away. One of the last versions of OS 9, and only then do the programmers remember to fix up minor bugs like this?

There, have fun with those

I had loads of fun, thank you

- uilleann