/dev/hda: Timing buffer-cache reads: 128 MB in 0.86 seconds =148.84 MB/sec Timing buffered disk reads: 64 MB in 2.60 seconds = 24.62 MB/sec
Thanks. Silly me not considering the (most likely) different hard disk we two are using.
One thing I do know is that the DMA parameter made a huge difference on my system. The original kernel that came with RH 7.3 didn't recognise my new hard disk controller, thereby disabling all optimisation options on the hard disk. That time, disk access was 10 times as slow as it is now.
Kubuntu 8.04 (kernel 2.6.24-25-generic) / KDE 3.5.10 CentOS 4.8 (kernel 2.6.9-78.0.22.ELsmp) / KDE 3.5.10 Mac OS X 10.6.1 (Snow Leopard) / iPhone 3GS (32GB black)
"Yes, Phoenix is noticeably slower in Linux than in Windows."
I don't find that to be the case. Maybe its because my hardware is only about 1.5 years old(XP1900), but Phoenix runs very quickly on Redhat 8. I launchs quickly and I see no difference in speed between it and windows. Maybe on really old hardware it makes a difference, but on modern hardware I see no difference.
On old hardware in general, people need to realize that Phoenix and Mozilla will never be fast on any platform. Considering the slowest machine you can buy these day is a 1.8GHz Emachine for example, I don't really consider it a problem either.
I have two desktops at work; one with P3-500MHz (w/ 512MB PC100 RAM) and the other with dual P3-800MHz (w/ 512MB PC133 ECC RAM). The former runs RH 8, the latter XP Pro. Interestingly, Phoenix runs quite fast on the RH system despite the hardware differences. RH itself runs much slower than XP, and that's natural. For some reason, I don't see the same difference with Phoenix. It's a nice surprise.
Kubuntu 8.04 (kernel 2.6.24-25-generic) / KDE 3.5.10 CentOS 4.8 (kernel 2.6.9-78.0.22.ELsmp) / KDE 3.5.10 Mac OS X 10.6.1 (Snow Leopard) / iPhone 3GS (32GB black)
yeah. I noticed on a friend's machine running redhat 8... p3 1.6 GHz IBM Thinkpad... I don't know his hdparm numbers... but mozilla was fast on his.
I know my machine is older and slower, but so is the other machine I tried to compare with. I was just wondering about which settings I should look at.
I've noticed that Phoenix starts up a bit slower (as of late anyway), probably because I am now using Phoenix with XFT2 enabled, so it's doing a bunch of stuff with very pretty fonts, otherwise rendering doesn't seem any different.
Edit: P.S. I'm using debian - I just upgraded to Unstable (most packages)
schapel wrote:Mozilla Linux builds use gcc 2.95 with -O optimization (low optimization), but Windows builds use Visual Studio with a high level of optimization. Developers are working on getting Linux builds to compile using gcc 3.2 with -O2 optimization. This should cause the Linux builds to be about as fast as the Windows builds.
I sure hope they will still build Phoenix using GCC 2.9x. Otherwise, I will not be able to run it on my RH 7.3 system... will I? On a side note, Intel's new compiler seems to do much better in optimisation than GCC (I read a comparison article a few months ago).
Hmmm... that depends on whether Pheonix is a static build or if it uses shared libraries, doesn't it? Anyway, the gcc 3.x series is becoming stable so everyone will switch over to it soon if they haven't already.
Intel's compiler can do much better than gcc on very specific programs, mostly numerical processing. I don't think it would be much faster on a program like Mozilla.
daihard wrote:I sure hope they will still build Phoenix using GCC 2.9x. Otherwise, I will not be able to run it on my RH 7.3 system... will I? On a side note, Intel's new compiler seems to do much better in optimisation than GCC (I read a comparison article a few months ago).
Hmmm... that depends on whether Pheonix is a static build or if it uses shared libraries, doesn't it? Anyway, the gcc 3.x series is becoming stable so everyone will switch over to it soon if they haven't already.
Looking at the size of the executable, I'd be darned if Phoenix wasn't dynamically linked with shared objects. That being the case, I suppose an app compiled with GCC 3.2 wil run on an older system if the system has the GCC 3.2 libraries?
schapel wrote:Intel's compiler can do much better than gcc on very specific programs, mostly numerical processing. I don't think it would be much faster on a program like Mozilla.
Thanks for the info. I'd be interested in trying that out myself. Like everyone else, though, I have little time to carry that out.
Kubuntu 8.04 (kernel 2.6.24-25-generic) / KDE 3.5.10 CentOS 4.8 (kernel 2.6.9-78.0.22.ELsmp) / KDE 3.5.10 Mac OS X 10.6.1 (Snow Leopard) / iPhone 3GS (32GB black)
mozillaZine is an independent Mozilla community and advocacy site. We're not affiliated or endorsed by the Mozilla Corporation but we love them just the same.