That's weird. Looking at about:buildconfig, they targeted i686 (needed for some atomic instructions, the -march parameter to GCC). Unless you have a different version of mine, an RPM. Check the URL about:buildconfig.
</silence>
Well, the name of a *pentium4* file is just a name. A remaining (Pentium 4-) install of the same FF 45.9.0-2, with the same profile, reports: target i386-pc-os2-emx, -march=pentium4.
Illegal instruction, *pentium4* DBG files, -march=pentium4: no typo. Duh...
As for SeaMonkey, did you try disabling add-ons? Some, especially chatzilla, cause high CPU usage.
It's probably that VSYNC'ish (the worst of both worlds) bug, causes a 100% CPU load while browsing about:blank, becomes a honest ~95% CPU load when a faster T42 is used, and becomes 20%'ish of one of ones multiple cores. So it qualifies as broken.
SM is the most important browser over here. I don't really use FF. The version of FF>=SM. If FF is stable, then I'll upgrade to the matching version of SM too. Normally I would not have downgraded FF and wait for a new version, but installing Pentium 4 OS/2 software has no use at all.
A Pentium III's FF 45.5.0 is not 100% useless. Some add-ons/plug-ins/extensions require such a version. It will take a long while, but after several minutes you should be able to download an on-topicv Youtube video, which wasn't possible with FF38/SM2.35 tools anymore.
The good news is that dmik is finally getting the tests working and running so hopefully we'll get more stable browser.
Certainly not "we" anymore!
I'm not one of the happy few, which they are targeting by explicit and implicit requirements, based on whatever they are using (including eCS 2.x DE/EN or AOS EN). The bad news is that not everyone shares your standards. They have the freedom of policy to use e.g. -march=pentium4, which is also a great way to reduce the size of the OS/2 community while pretending to write a "FF for OS/2".
There's no need to guess, but maybe a Pentium 4-target was their way to try to solve 0.001% of the "performance issues", which could be the same issue as my 95%-100% CPU load. With such a CPU load no version of FF45 should pass a test.
FWIW, FF 48.0.2 works quite fine with the Pentium III CPU in question. The last non-ESR version of FF which supports a Pentium III. So any performance issue is OS/2 specific. As such the hardware isn't the problem, and OS/2 demands less resources than WinXP.
<silence>