Post Reply 
convert links
Feb. 23, 2005, 05:39 PM
Post: #16
 
ProxRocks Wrote:It's all a matter of HOW the clock pulse is USED... [...]
Okay, i apparently have a leck of knowledge here, thanks for the tutorial!
(Although i still wonder how they measure their "virtual" clock speed! *lol*)

sidki
Add Thank You Quote this message in a reply
Feb. 23, 2005, 05:54 PM
Post: #17
 
sidki3003 Wrote:
ProxRocks Wrote:It's all a matter of HOW the clock pulse is USED... [...]
Okay, i apparently have a leck of knowledge here, thanks for the tutorial!
(Although i still wonder how they measure their "virtual" clock speed! *lol*)

sidki
by comparing "instructions per second", "frames per second", "OpenGL" benchmarks, "DirectX" benchmarks, et cetera...

AMD has been stealing market share from Intel over the past several years for a reason...


AMD has the vote of this electrical engineer hands down...

Doesn't mean that I wouldn't touch a Pentium with a 10-foot poll...
Place it in the Penthouse with bells on - no wait, take the bells off... take it all off...

er, what was the question again?
Add Thank You Quote this message in a reply
Feb. 23, 2005, 07:34 PM
Post: #18
 
jUST (DANG, CAPSLOCK IS ON FOR ORCAD IN ANOTHER WINDOW, TOO LAZY TO UNLOCK)... jUST HOW DO I GO ABOUT "TIMING" HOW LONG A SPECIFIC FILTER TAKES?


PS - SORRY FOR THE CAPSLOCK... PEOPLE, GENERALLY SPEAKING, WHO always TYPE WITH THE CAPSLOCK KEY ACTIVE SHOULD BE shot...
Add Thank You Quote this message in a reply
Feb. 23, 2005, 11:27 PM
Post: #19
 
Capslock shouting:
We all get a license to shoot, should you get used to that, okay? Big Teeth


AMD vs. Intel:
Pretty sure that i'm going for the Athlon 64 now. Although i do see some pretty disturbing benchmarks with an outstanding FPU performance due to Intel's "Hyperthreading" (a.k.a. "Symmetric Multi-Threading" -- i.e. split the one real CPU into two virtual ones).


Filter benchmarking:
JD and i use the same standardized page the last two years. I emailed it to you.
Then in the filter editor: "Test" -> paste code -> "Profile".


sidki
Add Thank You Quote this message in a reply
Feb. 24, 2005, 11:57 AM
Post: #20
 
ProxRocks Wrote:the Hindu gods don't believe in Hades...
Get real....
Hindus have a place for the dead mon ami... It's called "Yam lok" or "the city of Yam"... "Yam" or "Yama" being the lord of the dead... so to speak.
Beware!
I lurk around all posts.
Wink Have fun.

<span style='font-size:8pt;line-height:100%'><i><span style='color:#0000FF'>Projekt</span> </span><span style='color:#FF0000'>D</span><span style='color:#008200'>F</span><span style='color:#8449a5'>S</span></i> : <b><span style='color:#0000FF'>projekt</span><span style='color:#FF0000'>d</span><span style='color:#008200'>f</span><span style='color:#8449a5'>s</span></b>[at]<span style='color:BLUE'>g</span><span style='color:RED'>m</span><span style='color:#ce9a31'>a</span><span style='color:BLUE'>i</span><span style='color:#008200'>l</span><span style='color:BLACK'>.</span><span style='color:#8449a5'>com </span><span style='color:BLACK'>: <u>What</u> is it?</span><br>It&#39;s ONLINE <span style='color:#FF0000'><span style='font-size:10pt;line-height:100%'>NOW</span></span>&#33;&#33;&#33;pm me for url. max 250 users. by invite only please.
Add Thank You Quote this message in a reply
Feb. 24, 2005, 11:58 AM
Post: #21
 
Quote:PEOPLE ... WHO ... TYPE WITH THE CAPSLOCK KEY ACTIVE SHOULD BE shot...
Consider it done. Wink

<span style='font-size:8pt;line-height:100%'><i><span style='color:#0000FF'>Projekt</span> </span><span style='color:#FF0000'>D</span><span style='color:#008200'>F</span><span style='color:#8449a5'>S</span></i> : <b><span style='color:#0000FF'>projekt</span><span style='color:#FF0000'>d</span><span style='color:#008200'>f</span><span style='color:#8449a5'>s</span></b>[at]<span style='color:BLUE'>g</span><span style='color:RED'>m</span><span style='color:#ce9a31'>a</span><span style='color:BLUE'>i</span><span style='color:#008200'>l</span><span style='color:BLACK'>.</span><span style='color:#8449a5'>com </span><span style='color:BLACK'>: <u>What</u> is it?</span><br>It&#39;s ONLINE <span style='color:#FF0000'><span style='font-size:10pt;line-height:100%'>NOW</span></span>&#33;&#33;&#33;pm me for url. max 250 users. by invite only please.
Add Thank You Quote this message in a reply
Feb. 24, 2005, 12:57 PM
Post: #22
 
Code:
Name = "Convert h??p:// Text to Clickable Links - v3"
Active = TRUE
Limit = 256
Match = "(h(?[^t]|[^t]?)p://)\1"
"\2"
"([< ][a-z0-9]|[.,][^a-z0-9])\3"
Replace = "<a href="http://\2">\1\2</a>\3"

Ran into a page that required a slight update...
http://finance.messages.yahoo.com/bbs?.mm=...28834&mid=45243


There are some 'benchmarks' that a H/T Intel will beat an AMD 64...
And there are some 'benchmarks' that the AMD 64 "clocks" (er, "clobbers") Intel with or without hyperthreading...

But even at that, the benchmarks are so extremely close that when considering 'bang for buck', the AMD simply cannot be beat...


ps - no Hindo gods of the underworld were harmed in the making of this post...
Add Thank You Quote this message in a reply
Feb. 25, 2005, 05:36 AM
Post: #23
 
ProxRocks;

And here I have been going through life mistakenly thinking that the Hindu "Wheel of Existance" had validity! Stupid me, thinking that they believed in reincarnation! [lol]


Oddysey

NOTICE: No post lurkers were harmed in the making of this message.

I'm no longer in the rat race - the rats won't have me!
Add Thank You Quote this message in a reply
Feb. 25, 2005, 06:19 AM
Post: #24
 
sidki3003;

As one who has been around so long that they were still sifting sand off the beach when they wanted to make solid state devices, I think I can safely offer one piece of advice, one that will fairly much agree with what's already been said here.

It truly doesn't matter what any particular benchmarks might say. They can report until the cows come home, but when fight's all over, there's only one thing that counts, and that's personal satisfaction. How do you get that, or just as importantly, how can one measure that?

Well, there's the rub. Measuring something so personal is about as impossible as herding cats, IMHO. [rolleyes] I'll tell you what I think, and you decide if it's worth adopting for yourself. I think that if I can sleep at night after making my purchase, if I don't get the heeby-jeebies and want to take the thing back the next morning, if I can pass by other computers without a second glance for at least a half-a-year or more, then I probably made a wise choice.

I don't need to tell you buy quality before features, or even before performance, you already know that. But PR is correct - bragging rights are for people who are insecure, and need to participate in pissing contests every chance they get. I'd just as soon let them brag about who's got the biggest one, and just smile as I sit back in my corner, processing away, and knowing that I've still got a pretty healthy chunk of geld left in the bank. And I still have enough horsepower under the hood to mutilate any real-world task I can find. Smile!

Oh, and for the record.... all the above has influenced my purchasing decisions for just about 20 years - I've been using AMD processors since I bought my first Compaq-clone in June of 1985. Intel's good, mighty good, but I like the price/performance ratio I get from AMD. Something about not needing to be the first kid on the block with a Radio Flyer wagon, do ya think? <_<


Oddysey

I'm no longer in the rat race - the rats won't have me!
Add Thank You Quote this message in a reply
Feb. 25, 2005, 06:40 AM
Post: #25
 
ProxRocks and sidki;

Without any sort of benchmarking, I'm "seat of the pants" certain of this - I can disable my filter, and hit a site like, say, http://www.dailyrotation.com which contains in the vicinity of 700 links. Afterwards, I re-enable my filter, and hit the same site (after exiting all browser sessions to clear the cache, and then manually checking it anyway (can't be too sure when making claims like this, right? No Expression )).

No difference. At least, not so's I can notice. And I really don't see this as a big reliance on the CPU horsepower, because it does take time for the page to load, even over a broadband connection. Yes, there is a finite amount of time taken, but if I don't subjectively see it, then it's just not worth "optimizing" for, is it?

Beyond that, I should think it's possible, ProxRocks, to use the URL field to include or exclude sites that do indeed have a great number of already-existing links. Or you can just filter on those sites that are known to screw up a good thing. <_<


Oddysey

I'm no longer in the rat race - the rats won't have me!
Add Thank You Quote this message in a reply
Feb. 25, 2005, 01:19 PM
Post: #26
 
Oddysey Wrote:Beyond that, I should think it's possible, ProxRocks, to use the URL field to include or exclude sites that do indeed have a great number of already-existing links. Or you can just filter on those sites that are known to screw up a good thing. <_<


Oddysey
Already ahead of you - "my" version is identical to the 'public post' version EXCEPT for the URL field (I kick the filter in ONLY for Yahoo Message Boards)...

Just that for "public release", I took that line out - I figure 'the masses' may be encountering those 'd@mned debunk links' on sites OTHER than Yahoo Message Boards...


But having said that, I copied the source code of a web page with literally HUNDREDS of links into the 'filter test section' of Proxo... On the computer here at the office (which has several background apps running), the "convert h??p://" filter 'profiles' at roughly 0.6ms... For comparison, I compared that to several "public" <a>...[/url] filters and noticed that such "public" filters all 'check in' at 0.6ms, 0.9ms, 1.4ms, and 2.0ms (for the "main" <a>...[/url] filters such as untangling URL's, converting JS links, handling tracking links, highlighting 'mailto:' links, et cetera)...


All in all, at 0.6ms (again, on a computer with 60+ 'processes' running at minumum), I'd say that the 'convert h??p://' filter is comparing rather nicely...
Add Thank You Quote this message in a reply
Feb. 25, 2005, 05:44 PM
Post: #27
 
Oddysey Wrote:I'll tell you what I think, and you decide if it's worth adopting for yourself.&nbsp; I think that if I can sleep at night after making my purchase, if I don't get the heeby-jeebies and want to take the thing back the next morning, if I can pass by other computers without a second glance for at least a half-a-year or more, then I probably made a wise choice.
Sure. You do know it afterwards - mostly. Personally, i try to estimate if my choice is a good one, before i shell out the money.

sidki
Add Thank You Quote this message in a reply
Feb. 25, 2005, 05:47 PM
Post: #28
 
Oddysey Wrote:No difference.&nbsp; At least, not so's I can notice.&nbsp; And I really don't see this as a big reliance on the CPU horsepower, because it does take time for the page to load, even over a broadband connection.&nbsp; Yes, there is a finite amount of time taken, but if I don't subjectively see it, then it's just not worth "optimizing" for, is it?
That depends on your config, but generally it is.

A difference in the milliseconds range while looking at single filters, can reach the seconds range when looking at the entire config. Because of potentiating effects and interdependencies.

I did drop good filters, just because they are too slow (slower than ~10ms/30k chars in my environment).

sidki
Add Thank You Quote this message in a reply
Feb. 26, 2005, 12:26 AM
Post: #29
 
sidki;
Quote:
Oddysey Wrote:I'll tell you what I think, and you decide if it's worth adopting for yourself. I think that if I can sleep at night after making my purchase, if I don't get the heeby-jeebies and want to take the thing back the next morning, if I can pass by other computers without a second glance for at least a half-a-year or more, then I probably made a wise choice.
Sure. You do know it afterwards - mostly. Personally, i try to estimate if my choice is a good one, before i shell out the money.

sidki

Quite true, that part about hindsight. :P Indeed, nothing replaces good research like...... good research. :o But you know da(r|m)n good and well that most people who buy costly items will "second-guess" themselves after the purchase. Indeed, that's why there's a federal law (in the USA) that declares a cooling off period of 3 days for certain sales. (Ask for details if you're not sure what I mean, but let's keep this thing rolling.)

That so-called 'second-guessing' is what I was referring to when I said "can you sleep at night......". That is the true test of a good purchase - it's not the moment you sign on the dotted line, it's the morning after. Big Teeth To me, that's the best way to measure "personal satisfaction".


Oddysey

I'm no longer in the rat race - the rats won't have me!
Add Thank You Quote this message in a reply
Feb. 26, 2005, 08:13 AM
Post: #30
 
ProxRocks;
Quote:..... I copied the source code of a web page with literally HUNDREDS of links into the 'filter test section' of Proxo...
Ya know, this is exactly why I stick around this joint. :P I once copied a few lines into the text box just to reduce my typing, but it never dawned on me to paste in a whole page's worth of source code. [angry]

Sadly, the test box has a limit of about 30K characters. I can't test the entirety of my favorite site, which today has 733 links. Thus the truncation to a mere 106 links, which needed 0.575 seconds to complete. I guess that puts me right where you want me, in the same ball park as you and sidki3003. Which is to say, there wasn't much benefit in optimizing as far as I did, was there. [unsure] Oh, well, it wasn't a complete waste - - at least I did learn something for my efforts. Wink


Oddysey

I'm no longer in the rat race - the rats won't have me!
Add Thank You Quote this message in a reply
Post Reply 


Forum Jump: