From sean at practicalweb.co.uk Mon Jan 7 17:46:21 2008 From: sean at practicalweb.co.uk (Sean Burlington) Date: Mon, 07 Jan 2008 17:46:21 +0000 Subject: [sIFR3-dev] sifr3 not showing up in IE7 Message-ID: <4782656D.60507@practicalweb.co.uk> Hi, I seem to be having a similar problem to a previous poster I've done a bit of digging in the code and it seems that in my version of IE7 flash isn't being detected I've altered line 452 of sifr.js to add an error report : if(!stop) try { axo = new ActiveXObject('ShockwaveFlash.ShockwaveFlash'); } catch(e) {alert(e.description)} I get the error message "Automation server can't create object" sifr works fine on IE6 FF and safari -- Sean Burlington PracticalWeb Ltd - websites that work for you From sean at practicalweb.co.uk Mon Jan 7 18:57:08 2008 From: sean at practicalweb.co.uk (Sean Burlington) Date: Mon, 07 Jan 2008 18:57:08 +0000 Subject: [sIFR3-dev] sifr3 not showing up in IE7 In-Reply-To: <4782656D.60507@practicalweb.co.uk> References: <4782656D.60507@practicalweb.co.uk> Message-ID: <47827604.2040202@practicalweb.co.uk> Sean Burlington wrote: > Hi, > > I seem to be having a similar problem to a previous poster > never mind... It turned out to be a corrupt install of flash player I've uninstalled and reinstalled the flash player and now all is as it should be. For reference this required downloading an uninstaller from the adobe website -- Sean Burlington PracticalWeb Ltd - websites that work for you From mark at novemberborn.net Mon Jan 7 19:30:54 2008 From: mark at novemberborn.net (Mark Wubben) Date: Mon, 7 Jan 2008 20:30:54 +0100 Subject: [sIFR3-dev] sifr3 not showing up in IE7 In-Reply-To: <47827604.2040202@practicalweb.co.uk> References: <4782656D.60507@practicalweb.co.uk> <47827604.2040202@practicalweb.co.uk> Message-ID: <70151DAC-620A-44AF-B3B9-4D0A82E34CCC@novemberborn.net> Hi Sean, Just curious, did you have problems viewing other websites using Flash? sIFR uses the same detection technique as SWFObject, so I'd assume it goes wrong on more sites? Regards, On Jan 7, 2008, at 19:57, Sean Burlington wrote: > Sean Burlington wrote: >> Hi, >> >> I seem to be having a similar problem to a previous poster >> > > never mind... > > It turned out to be a corrupt install of flash player > > I've uninstalled and reinstalled the flash player and now all is as it > should be. > > For reference this required downloading an uninstaller from the adobe > website > > -- > > Sean Burlington > > PracticalWeb Ltd - websites that work for you > _______________________________________________ > sifr3-dev mailing list > sifr3-dev at lists.novemberborn.net > http://lists.novemberborn.net/mailman/listinfo/sifr3-dev -- Mark From sean at practicalweb.co.uk Mon Jan 7 19:47:44 2008 From: sean at practicalweb.co.uk (Sean Burlington) Date: Mon, 07 Jan 2008 19:47:44 +0000 Subject: [sIFR3-dev] sifr3 not showing up in IE7 In-Reply-To: <70151DAC-620A-44AF-B3B9-4D0A82E34CCC@novemberborn.net> References: <4782656D.60507@practicalweb.co.uk> <47827604.2040202@practicalweb.co.uk> <70151DAC-620A-44AF-B3B9-4D0A82E34CCC@novemberborn.net> Message-ID: <478281E0.3040107@practicalweb.co.uk> Mark Wubben wrote: > Hi Sean, > > Just curious, did you have problems viewing other websites using Flash? > sIFR uses the same detection technique as SWFObject, so I'd assume it > goes wrong on more sites? > Not that I noticed - and there was flash running on the same page that sIFR was failing on! It was only the detection that failed - I'm not sure if many sites don't run detection scripts ... sIFR worked OK too if I disabled the browser check! It must have been a fairly recent problem though as I had sIFR2 working on the same machine until recently This is the closest to an explanation I can find http://blog.deconcept.com/swfobject/forum/discussion/350/flash-detection-fails-in-ie6-after-installing-flash-cs3/ Though flash was still working properly after I installed flash CS3 - possibly the problem arose only after rebooting... -- Sean From mark at novemberborn.net Mon Jan 7 19:57:43 2008 From: mark at novemberborn.net (Mark Wubben) Date: Mon, 7 Jan 2008 20:57:43 +0100 Subject: [sIFR3-dev] sifr3 not showing up in IE7 In-Reply-To: <478281E0.3040107@practicalweb.co.uk> References: <4782656D.60507@practicalweb.co.uk> <47827604.2040202@practicalweb.co.uk> <70151DAC-620A-44AF-B3B9-4D0A82E34CCC@novemberborn.net> <478281E0.3040107@practicalweb.co.uk> Message-ID: <9687E28D-0EE0-4EB8-9C9E-B8A3358495A5@novemberborn.net> On Jan 7, 2008, at 20:47, Sean Burlington wrote: > This is the closest to an explanation I can find > > http://blog.deconcept.com/swfobject/forum/discussion/350/flash-detection-fails-in-ie6-after-installing-flash-cs3/ > > Though flash was still working properly after I installed flash CS3 - > possibly the problem arose only after rebooting... The reboot sounds plausible, and it sure sounds like the problem you've been having! Thanks for digging this up. -- Mark From wbowley at gmail.com Tue Jan 22 02:34:01 2008 From: wbowley at gmail.com (Warren Bowley) Date: Mon, 21 Jan 2008 21:34:01 -0500 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 Message-ID: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> I first want to say thank you for the continued development of sIFR. It is a great addition to any website. I do have one problem that has been plaguing me and I can't seem to figure out exactly what the problem is. I'm running the latest version as of 01/21/08 (v370) on a client website at http://www.edgewaterhomes.net. The problem I have is that when the sIFR transformation is applied to the menu links at the top of hte page, the CPU usage tends to hover around 50% in both IE7 and FireFox. If I comment out the links, there are no issues with sIFR taking up CPU resources. Even if I only show one link, I still have the problem. I don't believe I am overusing sIFR on the page, so I'm not sure what the issue is. All of the links are contained within a
tag as recommended, and the CSS elements for links are applied in the .replace command at the bottom of the page. Can anyone shed some light on what the problem could be? I've been chasing this all day long to no avail. Thanks -- Warren From mark at novemberborn.net Tue Jan 22 11:02:51 2008 From: mark at novemberborn.net (Mark Wubben) Date: Tue, 22 Jan 2008 12:02:51 +0100 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> Message-ID: <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> Oh wow, that's pretty bad indeed! I'm spotting another issue as well, which is that the Flash movie isn't scaling along with page zoom. Could you make a standalone version of the site, zip it and mail it to me, so I can look more closely into what's going on? Regards, On Jan 22, 2008, at 03:34, Warren Bowley wrote: > I first want to say thank you for the continued development of sIFR. > It is a great addition to any website. I do have one problem that has > been plaguing me and I can't seem to figure out exactly what the > problem is. I'm running the latest version as of 01/21/08 (v370) on a > client website at http://www.edgewaterhomes.net. The problem I have > is that when the sIFR transformation is applied to the menu links at > the top of hte page, the CPU usage tends to hover around 50% in both > IE7 and FireFox. > > If I comment out the links, there are no issues with sIFR taking up > CPU resources. Even if I only show one link, I still have the > problem. I don't believe I am overusing sIFR on the page, so I'm not > sure what the issue is. All of the links are contained within a
> tag as recommended, and the CSS elements for links are applied in the > .replace command at the bottom of the page. > > Can anyone shed some light on what the problem could be? I've been > chasing this all day long to no avail. > > Thanks > -- Warren > _______________________________________________ > sifr3-dev mailing list > sifr3-dev at lists.novemberborn.net > http://lists.novemberborn.net/mailman/listinfo/sifr3-dev -- Mark From arsart at gmail.com Tue Jan 22 19:36:38 2008 From: arsart at gmail.com (Arsart) Date: Tue, 22 Jan 2008 22:36:38 +0300 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> Message-ID: Hi all, Having same issues with CPU and mostly with browser freeze on Mac OS Firefox. 80% of time after a sudden attempt to close a tab with site on sIFR3beta firefox freezes until complete Force Quite. Same freeze can happens when site loads and I switch to other tab :\ Arseny sIFRussian:) On 22/01/2008, Mark Wubben wrote: > Oh wow, that's pretty bad indeed! > > I'm spotting another issue as well, which is that the Flash movie > isn't scaling along with page zoom. Could you make a standalone > version of the site, zip it and mail it to me, so I can look more > closely into what's going on? > > Regards, > > On Jan 22, 2008, at 03:34, Warren Bowley wrote: > > > I first want to say thank you for the continued development of sIFR. > > It is a great addition to any website. I do have one problem that has > > been plaguing me and I can't seem to figure out exactly what the > > problem is. I'm running the latest version as of 01/21/08 (v370) on a > > client website at http://www.edgewaterhomes.net. The problem I have > > is that when the sIFR transformation is applied to the menu links at > > the top of hte page, the CPU usage tends to hover around 50% in both > > IE7 and FireFox. > > > > If I comment out the links, there are no issues with sIFR taking up > > CPU resources. Even if I only show one link, I still have the > > problem. I don't believe I am overusing sIFR on the page, so I'm not > > sure what the issue is. All of the links are contained within a
> > tag as recommended, and the CSS elements for links are applied in the > > .replace command at the bottom of the page. > > > > Can anyone shed some light on what the problem could be? I've been > > chasing this all day long to no avail. > > > > Thanks > > -- Warren > > _______________________________________________ > > sifr3-dev mailing list > > sifr3-dev at lists.novemberborn.net > > http://lists.novemberborn.net/mailman/listinfo/sifr3-dev > > -- > Mark > > _______________________________________________ > sifr3-dev mailing list > sifr3-dev at lists.novemberborn.net > http://lists.novemberborn.net/mailman/listinfo/sifr3-dev > -- Arseny Vesnin Profile: http://www.linkedin.com/in/arsart Owner: http://designcollector.ru http://designcentre.ru From mark at novemberborn.net Tue Jan 22 20:25:13 2008 From: mark at novemberborn.net (Mark Wubben) Date: Tue, 22 Jan 2008 21:25:13 +0100 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> Message-ID: <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> Warren sent me a test case, and if I remove the overflow:hidden; declarations in it, the CPU usage dropped significantly. It was still large, but then there are about 20 Flash movies on the page. This would indicate that CSS has an effect. Perhaps you (Warren et Arseny) could play around with the CSS and see if it matters? Properties I'd believe to affect CPU usage are float, overflow and position. I'll see what I can do to lower the amount of CPU a sIFR Flash movie requires when it's idle. On Jan 22, 2008, at 20:36, Arsart wrote: > Hi all, > > Having same issues with CPU and mostly with browser freeze on Mac OS > Firefox. 80% of time after a sudden attempt to close a tab with site > on sIFR3beta firefox freezes until complete Force Quite. Same freeze > can happens when site loads and I switch to other tab :\ > > Arseny > sIFRussian:) > > On 22/01/2008, Mark Wubben wrote: >> Oh wow, that's pretty bad indeed! >> >> I'm spotting another issue as well, which is that the Flash movie >> isn't scaling along with page zoom. Could you make a standalone >> version of the site, zip it and mail it to me, so I can look more >> closely into what's going on? >> >> Regards, >> >> On Jan 22, 2008, at 03:34, Warren Bowley wrote: >> >>> I first want to say thank you for the continued development of sIFR. >>> It is a great addition to any website. I do have one problem that >>> has >>> been plaguing me and I can't seem to figure out exactly what the >>> problem is. I'm running the latest version as of 01/21/08 (v370) >>> on a >>> client website at http://www.edgewaterhomes.net. The problem I have >>> is that when the sIFR transformation is applied to the menu links at >>> the top of hte page, the CPU usage tends to hover around 50% in both >>> IE7 and FireFox. >>> >>> If I comment out the links, there are no issues with sIFR taking up >>> CPU resources. Even if I only show one link, I still have the >>> problem. I don't believe I am overusing sIFR on the page, so I'm >>> not >>> sure what the issue is. All of the links are contained within a >>>
>>> tag as recommended, and the CSS elements for links are applied in >>> the >>> .replace command at the bottom of the page. >>> >>> Can anyone shed some light on what the problem could be? I've been >>> chasing this all day long to no avail. >>> >>> Thanks >>> -- Warren >>> _______________________________________________ >>> sifr3-dev mailing list >>> sifr3-dev at lists.novemberborn.net >>> http://lists.novemberborn.net/mailman/listinfo/sifr3-dev >> >> -- >> Mark >> >> _______________________________________________ >> sifr3-dev mailing list >> sifr3-dev at lists.novemberborn.net >> http://lists.novemberborn.net/mailman/listinfo/sifr3-dev >> > > > -- > Arseny Vesnin > Profile: http://www.linkedin.com/in/arsart > Owner: http://designcollector.ru http://designcentre.ru -- Mark From arsart at gmail.com Tue Jan 22 20:53:49 2008 From: arsart at gmail.com (Arsart) Date: Tue, 22 Jan 2008 23:53:49 +0300 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> Message-ID: Yes I also have nearly 20 sIFRs for a page. But what declaration you mean? In .sIFR-alternate? It leads an issue... On 22/01/2008, Mark Wubben wrote: > Warren sent me a test case, and if I remove the overflow:hidden; > declarations in it, the CPU usage dropped significantly. It was still > large, but then there are about 20 Flash movies on the page. This > would indicate that CSS has an effect. > > Perhaps you (Warren et Arseny) could play around with the CSS and see > if it matters? Properties I'd believe to affect CPU usage are float, > overflow and position. > > I'll see what I can do to lower the amount of CPU a sIFR Flash movie > requires when it's idle. > > On Jan 22, 2008, at 20:36, Arsart wrote: > > > Hi all, > > > > Having same issues with CPU and mostly with browser freeze on Mac OS > > Firefox. 80% of time after a sudden attempt to close a tab with site > > on sIFR3beta firefox freezes until complete Force Quite. Same freeze > > can happens when site loads and I switch to other tab :\ > > > > Arseny > > sIFRussian:) > > > > On 22/01/2008, Mark Wubben wrote: > >> Oh wow, that's pretty bad indeed! > >> > >> I'm spotting another issue as well, which is that the Flash movie > >> isn't scaling along with page zoom. Could you make a standalone > >> version of the site, zip it and mail it to me, so I can look more > >> closely into what's going on? > >> > >> Regards, > >> > >> On Jan 22, 2008, at 03:34, Warren Bowley wrote: > >> > >>> I first want to say thank you for the continued development of sIFR. > >>> It is a great addition to any website. I do have one problem that > >>> has > >>> been plaguing me and I can't seem to figure out exactly what the > >>> problem is. I'm running the latest version as of 01/21/08 (v370) > >>> on a > >>> client website at http://www.edgewaterhomes.net. The problem I have > >>> is that when the sIFR transformation is applied to the menu links at > >>> the top of hte page, the CPU usage tends to hover around 50% in both > >>> IE7 and FireFox. > >>> > >>> If I comment out the links, there are no issues with sIFR taking up > >>> CPU resources. Even if I only show one link, I still have the > >>> problem. I don't believe I am overusing sIFR on the page, so I'm > >>> not > >>> sure what the issue is. All of the links are contained within a > >>>
> >>> tag as recommended, and the CSS elements for links are applied in > >>> the > >>> .replace command at the bottom of the page. > >>> > >>> Can anyone shed some light on what the problem could be? I've been > >>> chasing this all day long to no avail. > >>> > >>> Thanks > >>> -- Warren > >>> _______________________________________________ > >>> sifr3-dev mailing list > >>> sifr3-dev at lists.novemberborn.net > >>> http://lists.novemberborn.net/mailman/listinfo/sifr3-dev > >> > >> -- > >> Mark > >> > >> _______________________________________________ > >> sifr3-dev mailing list > >> sifr3-dev at lists.novemberborn.net > >> http://lists.novemberborn.net/mailman/listinfo/sifr3-dev > >> > > > > > > -- > > Arseny Vesnin > > Profile: http://www.linkedin.com/in/arsart > > Owner: http://designcollector.ru http://designcentre.ru > > -- > Mark > > -- Arseny Vesnin Profile: http://www.linkedin.com/in/arsart Owner: http://designcollector.ru http://designcentre.ru From wbowley at gmail.com Tue Jan 22 21:25:53 2008 From: wbowley at gmail.com (Warren Bowley) Date: Tue, 22 Jan 2008 16:25:53 -0500 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> Message-ID: <8eb514fe0801221325w228ed636x77444ad348da21a@mail.gmail.com> I removed any references to overflow in my .css, but removing float and position aren't really an option as the pay layout basically relies on these to position elements correctly. Does the number of elements converted on the page make that much of a difference in CPU usage? I looked at the Louis Logan example (http://www.louislogan.com/default.asp) and it barely makes my laptop sweat. From the looks of it, there are a number of elements that are converted and does appear to use float elements. I'm just trying to see what I'm doing wrong here as I really do want to use sIFR for the menus and such. On Jan 22, 2008 3:25 PM, Mark Wubben wrote: > Warren sent me a test case, and if I remove the overflow:hidden; > declarations in it, the CPU usage dropped significantly. It was still > large, but then there are about 20 Flash movies on the page. This > would indicate that CSS has an effect. > > Perhaps you (Warren et Arseny) could play around with the CSS and see > if it matters? Properties I'd believe to affect CPU usage are float, > overflow and position. > > I'll see what I can do to lower the amount of CPU a sIFR Flash movie > requires when it's idle. From wbowley at gmail.com Wed Jan 23 03:33:25 2008 From: wbowley at gmail.com (Warren Bowley) Date: Tue, 22 Jan 2008 22:33:25 -0500 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> Message-ID: <8eb514fe0801221933l16c305eaje23d9fc980745bdd@mail.gmail.com> I've found some more oddities. I spent most of the day tracking down the issue and found the high CPU usage to be caused by the menu I had on the page. There are 8 menu entries each separated with a dot character. Each of these entries was in it's own DIV having a basic sIFR class applied to it to render the text. Nothing really out of the ordinary. You can see it at: http://www.edgewaterhomes.net/sIFR1.aspx Let the page load and watch the CPU usage. It will spike between 10-50% with the page sitting idle. Mark had made a suggestion that float, overflow and position may be affecting CPU usage, so I tried a number of combinations to get the layout right, but really didn't get anywhere. I started removing some of the elements to see if there was a threshold at which point it would trigger the CPU usage and thought I had something, but it really wasn't conclusive. Removing some of the "dots" between the menu entries lowered CPU usage, but you could still see it jump up and down between 3% and 6%, sometimes up to 10%. The last option was to try and consolidate things. I started by taking the "dots" and putting them into the same DIV tag as the menu entries. This is where I found one problem. If I had the following:
HOME ?
The hyperlink wouldn't be recognized. I could move my mouse over it, but it wasn't recognized as a link. If I changed it to:
HOME ?
It worked fine. It appears as though if you put a hyperlink inside a DIV tag where the transformation is applied, and also have another character in that DIV tag, the hyperlink does not work. I couldn't tell if this was by design (ie: I shouldn't do that), or if it's a bug. Maybe someone can tell me. Anyhow, after changing all of the links to have the "dot" within the actual link, there is no more high CPU usage. I don't know why. The good page is located at: http://www.edgewaterhomes.net/sIFR2.aspx Does anybody have any ideas? I'm glad I found a workaround for the issue as I really do like sIFR, I just want to know what I was doing wrong so that I don't get caught in the same situation in the future. Most clients wouldn't really notice the CPU usage, but I couldn't put that out there knowing that something just wasn't right. I hope the cause can be found and that this helps anybody else having a high CPU usage issue. On Jan 22, 2008 3:25 PM, Mark Wubben wrote: > Warren sent me a test case, and if I remove the overflow:hidden; > declarations in it, the CPU usage dropped significantly. It was still > large, but then there are about 20 Flash movies on the page. This > would indicate that CSS has an effect. > > Perhaps you (Warren et Arseny) could play around with the CSS and see > if it matters? Properties I'd believe to affect CPU usage are float, > overflow and position. > > I'll see what I can do to lower the amount of CPU a sIFR Flash movie > requires when it's idle. > > > On Jan 22, 2008, at 20:36, Arsart wrote: > > > Hi all, > > > > Having same issues with CPU and mostly with browser freeze on Mac OS > > Firefox. 80% of time after a sudden attempt to close a tab with site > > on sIFR3beta firefox freezes until complete Force Quite. Same freeze > > can happens when site loads and I switch to other tab :\ > > > > Arseny > > sIFRussian:) > > > > On 22/01/2008, Mark Wubben wrote: > >> Oh wow, that's pretty bad indeed! > >> > >> I'm spotting another issue as well, which is that the Flash movie > >> isn't scaling along with page zoom. Could you make a standalone > >> version of the site, zip it and mail it to me, so I can look more > >> closely into what's going on? > >> > >> Regards, > >> > >> On Jan 22, 2008, at 03:34, Warren Bowley wrote: > >> > >>> I first want to say thank you for the continued development of sIFR. > >>> It is a great addition to any website. I do have one problem that > >>> has > >>> been plaguing me and I can't seem to figure out exactly what the > >>> problem is. I'm running the latest version as of 01/21/08 (v370) > >>> on a > >>> client website at http://www.edgewaterhomes.net. The problem I have > >>> is that when the sIFR transformation is applied to the menu links at > >>> the top of hte page, the CPU usage tends to hover around 50% in both > >>> IE7 and FireFox. > >>> > >>> If I comment out the links, there are no issues with sIFR taking up > >>> CPU resources. Even if I only show one link, I still have the > >>> problem. I don't believe I am overusing sIFR on the page, so I'm > >>> not > >>> sure what the issue is. All of the links are contained within a > >>>
> >>> tag as recommended, and the CSS elements for links are applied in > >>> the > >>> .replace command at the bottom of the page. > >>> > >>> Can anyone shed some light on what the problem could be? I've been > >>> chasing this all day long to no avail. > >>> > >>> Thanks > >>> -- Warren > >>> _______________________________________________ > >>> sifr3-dev mailing list > >>> sifr3-dev at lists.novemberborn.net > >>> http://lists.novemberborn.net/mailman/listinfo/sifr3-dev > >> > >> -- > >> Mark > >> > >> _______________________________________________ > >> sifr3-dev mailing list > >> sifr3-dev at lists.novemberborn.net > >> http://lists.novemberborn.net/mailman/listinfo/sifr3-dev > >> > > > > > > -- > > Arseny Vesnin > > Profile: http://www.linkedin.com/in/arsart > > Owner: http://designcollector.ru http://designcentre.ru > > -- > Mark > > From wbowley at gmail.com Wed Jan 23 04:33:12 2008 From: wbowley at gmail.com (Warren Bowley) Date: Tue, 22 Jan 2008 23:33:12 -0500 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: <8eb514fe0801221933l16c305eaje23d9fc980745bdd@mail.gmail.com> References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> <8eb514fe0801221933l16c305eaje23d9fc980745bdd@mail.gmail.com> Message-ID: <8eb514fe0801222033v2be23120tecce8b1fd4033409@mail.gmail.com> *sigh* I thought I had the problem solved only to get the same CPU usage spikes with a different set of sIFR elements on the main page. If I avoid formatting specific elements I can stop the situation from happening. There has to be some explanation for it, but I'm too tired right now to start digging into it. More fun tomorrow. On Jan 22, 2008 10:33 PM, Warren Bowley wrote: > I've found some more oddities. I spent most of the day tracking down > the issue and found the high CPU usage to be caused by the menu I had > on the page. There are 8 menu entries each separated with a dot > character. Each of these entries was in it's own DIV having a basic > sIFR class applied to it to render the text. Nothing really out of > the ordinary. You can see it at: From mark at novemberborn.net Wed Jan 23 07:05:20 2008 From: mark at novemberborn.net (Mark Wubben) Date: Wed, 23 Jan 2008 08:05:20 +0100 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> Message-ID: <8C5517A2-CD06-4D22-86FF-39226E52E7FE@novemberborn.net> On Jan 22, 2008, at 21:53, Arsart wrote: > Yes I also have nearly 20 sIFRs for a page. But what declaration you > mean? In .sIFR-alternate? It leads an issue... In the CSS for the page itself, not particularly the sIFR styles. -- Mark From mark at novemberborn.net Wed Jan 23 07:06:37 2008 From: mark at novemberborn.net (Mark Wubben) Date: Wed, 23 Jan 2008 08:06:37 +0100 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: <8eb514fe0801221325w228ed636x77444ad348da21a@mail.gmail.com> References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> <8eb514fe0801221325w228ed636x77444ad348da21a@mail.gmail.com> Message-ID: <513E3D4A-5B1A-493C-BDB9-952FD5B95292@novemberborn.net> On Jan 22, 2008, at 22:25, Warren Bowley wrote: > I removed any references to overflow in my .css, but removing float > and position aren't really an option as the pay layout basically > relies on these to position elements correctly. Yes, but give it a shot just to see if layout affects CPU usage. > Does the number of elements converted on the page make that much of > a difference in CPU > usage? I looked at the Louis Logan example > (http://www.louislogan.com/default.asp) and it barely makes my laptop > sweat. From the looks of it, there are a number of elements that are > converted and does appear to use float elements. Well it's bound to have an effect. sIFR movies aren't doing much, so it shouldn't have a huge effect, but more Flash will mean more CPU. -- Mark From mark at novemberborn.net Wed Jan 23 07:18:24 2008 From: mark at novemberborn.net (Mark Wubben) Date: Wed, 23 Jan 2008 08:18:24 +0100 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: <8eb514fe0801222033v2be23120tecce8b1fd4033409@mail.gmail.com> References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> <8eb514fe0801221933l16c305eaje23d9fc980745bdd@mail.gmail.com> <8eb514fe0801222033v2be23120tecce8b1fd4033409@mail.gmail.com> Message-ID: On Jan 23, 2008, at 05:33, Warren Bowley wrote: > I thought I had the problem solved only to get the same CPU usage > spikes with a different set of sIFR elements on the main page. If I > avoid formatting specific elements I can stop the situation from > happening. There has to be some explanation for it, but I'm too tired > right now to start digging into it. More fun tomorrow. I'm afraid it'll be hard to pinpoint this to any specific thing on the page. It seems very much to be related to layout and the browser itself. Minimizing the amount of Flash on the page, and making the layout less complex should help. By the way, why are you specifying a width and height on the elements being replaced? In my own testing I also moved the sIFR.replace() statements to sifr-config, rather than having them embedded in the page itself. That shouldn't matter for the CPU issue, but you never know. (It's a best practice, too.) -- Mark From wbowley at gmail.com Wed Jan 23 13:30:54 2008 From: wbowley at gmail.com (Warren Bowley) Date: Wed, 23 Jan 2008 08:30:54 -0500 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> <8eb514fe0801221933l16c305eaje23d9fc980745bdd@mail.gmail.com> <8eb514fe0801222033v2be23120tecce8b1fd4033409@mail.gmail.com> Message-ID: <8eb514fe0801230530i5a0ad7d3hf0b42cbf524a3f9f@mail.gmail.com> I want the layout to look the same with or without sIFR which requires changing the width and heights so that it's visually appealing. I'll try working on the layout and moving the replace into the config files as well. Thanks for the explanations and tips! On Jan 23, 2008 2:18 AM, Mark Wubben wrote: > By the way, why are you specifying a width and height on the elements > being replaced? In my own testing I also moved the sIFR.replace() > statements to sifr-config, rather than having them embedded in the > page itself. That shouldn't matter for the CPU issue, but you never > know. (It's a best practice, too.) From wbowley at gmail.com Wed Jan 23 20:02:56 2008 From: wbowley at gmail.com (Warren Bowley) Date: Wed, 23 Jan 2008 15:02:56 -0500 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> <8eb514fe0801221933l16c305eaje23d9fc980745bdd@mail.gmail.com> <8eb514fe0801222033v2be23120tecce8b1fd4033409@mail.gmail.com> Message-ID: <8eb514fe0801231202p320430co9bca1d4c9b999b3@mail.gmail.com> I've gone ahead and taken your advice and removed most of the flash elements from the page. I still have sIFR working on the menu entries at the top and headings for various pages, but have converted the rest to images. I don't get the high CPU usage and can keep the layout (although I have removed some of the things you've suggested). I'll try to keep the usage of sIFR down so as to avoid this sort of thing in the future. I did find something really interesting. I have another machine with Windows XP Professional, a trial download of Adobe Flash CS3, and Internet Explorer version 6 and it had NO CPU issues at all. I could load the bad page and it wouldn't cycle like it did on the laptop (running XP Home with IE7 and no Flash CS3). I thought that was a little odd. There are enough variables between the two machines that it isn't going to be worth my time trying to figure out what made the difference. I do appreciate all the help and advice you've given me on this issue. Keep up the great work on sIFR! On Jan 23, 2008 2:18 AM, Mark Wubben wrote: > I'm afraid it'll be hard to pinpoint this to any specific thing on the > page. It seems very much to be related to layout and the browser > itself. Minimizing the amount of Flash on the page, and making the > layout less complex should help. > > By the way, why are you specifying a width and height on the elements > being replaced? In my own testing I also moved the sIFR.replace() > statements to sifr-config, rather than having them embedded in the > page itself. That shouldn't matter for the CPU issue, but you never > know. (It's a best practice, too.) From mark at novemberborn.net Thu Jan 24 19:21:02 2008 From: mark at novemberborn.net (Mark Wubben) Date: Thu, 24 Jan 2008 20:21:02 +0100 Subject: [sIFR3-dev] High CPU usage with sIFR3 v370 In-Reply-To: <8eb514fe0801231202p320430co9bca1d4c9b999b3@mail.gmail.com> References: <8eb514fe0801211834y219b6633v5c3489b41a84aee9@mail.gmail.com> <3AD2AD6D-726D-4C6E-92C2-DEBF679B3111@novemberborn.net> <3CD2561A-15F5-4CE0-97C8-93474378D7AF@novemberborn.net> <8eb514fe0801221933l16c305eaje23d9fc980745bdd@mail.gmail.com> <8eb514fe0801222033v2be23120tecce8b1fd4033409@mail.gmail.com> <8eb514fe0801231202p320430co9bca1d4c9b999b3@mail.gmail.com> Message-ID: <37E3710C-C96F-405B-A19A-344C94C078C7@novemberborn.net> Okay then, glad it's more or less resolved. These issues are quite tricky indeed? On Jan 23, 2008, at 21:02, Warren Bowley wrote: > I've gone ahead and taken your advice and removed most of the flash > elements from the page. I still have sIFR working on the menu entries > at the top and headings for various pages, but have converted the rest > to images. I don't get the high CPU usage and can keep the layout > (although I have removed some of the things you've suggested). I'll > try to keep the usage of sIFR down so as to avoid this sort of thing > in the future. > > I did find something really interesting. I have another machine with > Windows XP Professional, a trial download of Adobe Flash CS3, and > Internet Explorer version 6 and it had NO CPU issues at all. I could > load the bad page and it wouldn't cycle like it did on the laptop > (running XP Home with IE7 and no Flash CS3). > > I thought that was a little odd. There are enough variables between > the two machines that it isn't going to be worth my time trying to > figure out what made the difference. > > I do appreciate all the help and advice you've given me on this issue. > Keep up the great work on sIFR! > > > On Jan 23, 2008 2:18 AM, Mark Wubben wrote: >> I'm afraid it'll be hard to pinpoint this to any specific thing on >> the >> page. It seems very much to be related to layout and the browser >> itself. Minimizing the amount of Flash on the page, and making the >> layout less complex should help. >> >> By the way, why are you specifying a width and height on the elements >> being replaced? In my own testing I also moved the sIFR.replace() >> statements to sifr-config, rather than having them embedded in the >> page itself. That shouldn't matter for the CPU issue, but you never >> know. (It's a best practice, too.) -- Mark