Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
foxPEP and TenFourFoxPEP - For A Faster Firefox / TenFourFox
- Thread starter z970
- Start date
- Sort by reaction score
My scroll bar moving more fluidly on my screen is not a subjective observation ?
I may have an unpopular and minority opinion, but I'm really no fan of smooth scrolling. The fact that WebGL enabled makes this enormously smoother confirms that smooth scrolling is hitting the CPU. If it wasn't then WebGL enabled wouldn't make a difference in this at least.
The way my brain comprehends web pages is for me to read, scroll down in one jump and then continue reading. Repeat as necessary. I'm not reading while scrolling. So, smooth scrolling is nothing I need. I'd rather have the browser jump down quickly and be able to continue with what I'm doing then paint while I'm scrolling.
But hey, people all have different likes and different ways of doing things.
I have smooth scrolling specifically turned off for the reasons @eyoungren stated, it's not a desired feature for me.
But smooth scrolling, the feature, is in all forms explicitly turned off...
And scrolling is nonetheless smoother on slower CPUs. If they were being hit, then they would choke.
Based on my work in development, I'd say the tile HxW, frame rate, and distinct lack of smooth scrolling have more to do with the motion advancements than WebGL. ...Although this is not necessarily all-conclusive.
Last edited:
I think there might have been a misunderstanding here when it comes to "smooth" scrolling. What the UOC Patch aims to do is not enabling "smooth scroll" like the one you get for example on stock Internet Explorer. That one is plain annoying to me as well. But the aim is to make scrolling smoother when you drag the scrollbar up and down with the mouse cursor. Like in the good old days when mice did not have a scrolling wheel.
I have a mouse without a scrolling wheel, and so I drag the scrollbar in order to scroll the pages, or when I don't feel like, I just use the Page Down key on the keyboard. But when I drag the scrollbar, I find annoying to see it moving in a stuttering fashion or even stopping when some heavy content is loading. So, my efforts were and are directed to make that as much smooth as possible, on old hardware. Then, anyone's preference is different, but I hope this helped to make things more clear regarding the kind of "smooth scrolling" I was talking about.
Indeed, the "smooth scrolling" option is turned off by default in the UOC Patch, as well as the PowerUOC one.
EDIT: Incidentally, at least on my machine, regardless of whether or not webGL is enabled, even without the UOC Patch, scrolling affects the CPU usage. But even on very slow hardware, such as a Celeron Mendocino, enabling it, alongside the patch, makes scrolling smoother, without making the browser unusable. I don't know if it offloads it from the CPU or not, but at least achieves the result I am looking for.
I have a mouse without a scrolling wheel, and so I drag the scrollbar in order to scroll the pages, or when I don't feel like, I just use the Page Down key on the keyboard. But when I drag the scrollbar, I find annoying to see it moving in a stuttering fashion or even stopping when some heavy content is loading. So, my efforts were and are directed to make that as much smooth as possible, on old hardware. Then, anyone's preference is different, but I hope this helped to make things more clear regarding the kind of "smooth scrolling" I was talking about.
Indeed, the "smooth scrolling" option is turned off by default in the UOC Patch, as well as the PowerUOC one.
EDIT: Incidentally, at least on my machine, regardless of whether or not webGL is enabled, even without the UOC Patch, scrolling affects the CPU usage. But even on very slow hardware, such as a Celeron Mendocino, enabling it, alongside the patch, makes scrolling smoother, without making the browser unusable. I don't know if it offloads it from the CPU or not, but at least achieves the result I am looking for.
Last edited:
Got any benchmarks supporting that hypothesis?
Why not get a 2011 i3 system for $60 that will run circles around a G5 (or netbook), in addition to being significantly faster than any Core 2 Duo as well? And I'm not even mentioning that it's miles ahead of the G5 in terms of energy consumption.
Then you get crappy windows or get stuck on Linux, why not use rock solid leopard?
If you just want to surf the web (which people using a Chromebook probably do), the OS doesn't really matter. Windows and Linux are rock-solid too, and, unlike Leopard, allow running current versions of several web browsers. Heck, you can run Chromium OS on a PC if you're keen on the Chromebook experience. The point I'm trying to make is that for a person who's just looking for an inexpensive machine to surf the web on, there are much better (as in faster, more secure, more energy-efficient) options than wasting money on 13-year-old hardware running a 12-year-old OS.
Last edited:
Guys, when you say "G5", please be more specific. Else it's an insult to every G5 Quad out there! Comparing it to a dirty-ass Core 2 Duo... ew!
Also, i3 has all the Intel ME troubles, don't forget! Even Core 2 Duo, which (often? always?) has removable Intel ME (unlike any Intel processor that came later, like the i3), it doesn't work with a lot of OSes once it's removed.
On PowerUOC and the prefs file, I can't comment yet. But I can already say the important thing here is that these efforts are taking place at all, and that's what matters. Eyoungren's efforts, UOC efforts, UOC PPC efforts etc.... It's good that all this is coming and being discussed.
Else it's an insult to every G5 Quad out there! Comparing it to a dirty-ass Core 2 Duo... ew! Also, i3 has all the Intel ME troubles, don't forget! Even Core 2 Duo, which (often? always?) has removable Intel ME (unlike any Intel processor that came later, like the i3), it doesn't work with a lot of OSes once it's removed.
On PowerUOC and the prefs file, I can't comment yet. But I can already say the important thing here is that these efforts are taking place at all, and that's what matters. Eyoungren's efforts, UOC efforts, UOC PPC efforts etc.... It's good that all this is coming and being discussed.
Guys, when you say "G5", please be more specific.
Also, i3 has all the Intel ME troubles, don't forget! Even Core 2 Duo, which (often? always?) has removable Intel ME (unlike any Intel processor that came later, like the i3), it doesn't work with a lot of OSes once it's removed.
Fair enough - a 3 GHz Core 2 Quad Extreme on a 945 chipset has no ME and still packs a punch. But let's go back on topic
Last edited:
Number IV is here!
This release brings tidal waves of improvements to privacy, security, and performance. Namely when playing video, or rendering large amounts of images at once.
(More details in the Changelog.)
For all still on the fence, now is an excellent time to try!
This release brings tidal waves of improvements to privacy, security, and performance. Namely when playing video, or rendering large amounts of images at once.
(More details in the Changelog.)
For all still on the fence, now is an excellent time to try!
I'd prefer a 4 GHz 2-processor 44-core 196-threaded POWER9 Talos II with 240MB of L3 cache.Fair enough - a 3 GHz Core 2 Quad Extreme on a 945 chipset has no ME and still packs a punch. But let's go back on topic
Also no ME.Not that the processor clock speed is an indicator of processor speed per se, since processor cycles count more in some processor architectures and algorithms than others (as is the case between a server-class RISC processor like the G5 vs. a cheaper desktop-class CISC processor like the Core 2 Duo line), but I concur, we can save this talk perhaps for some other time in a separate topic.
Just updated my prefs file. I timed TFF load times which went 30 - 24- 16 seconds before UOC and then 21, 23 and 24, so an average improvement of 8 tenths of one second. With the prefs I'm certainly getting more consistent load times anyhow.
Haven't tried anything else nor have I tried it with AF. One unexpected improvement is that the rich text editor is a bit snappier. Why? I have no idea but with UOC prefs there is less typing lag. Scrolling with the trackpad does also seem to be somewhat smoother. Yeah I know - anecdotal. Whatevs
Testing machine is a powerbook 12" 867Mhz 10.4.11.
Haven't tried anything else nor have I tried it with AF. One unexpected improvement is that the rich text editor is a bit snappier. Why? I have no idea but with UOC prefs there is less typing lag. Scrolling with the trackpad does also seem to be somewhat smoother. Yeah I know - anecdotal. Whatevs
Testing machine is a powerbook 12" 867Mhz 10.4.11.
Last edited:
Version IV revision 2 is out.
This one fixes a couple of things, further improves scrolling smoothness, and improves video playback.
More in the Changelog.
This one fixes a couple of things, further improves scrolling smoothness, and improves video playback.
More in the Changelog.
Just wanted to say thank you for making this. I'm typing this post from my Sawtooth and I can tell the difference between using PowerUOC and the default TFF preferences file.
Installed it on my cube (admittedly not much difference with a 450MHz G4) and on my 1 GHz TiBook. I have a better G4 PowerBook with an SSD that I need to really try it on.
My Sawtooth has an upgraded CPU and graphics, so my results may be different from someone who has a Sawtooth with stock hardware. Scrolling is a lot smoother on PowerUOC. Youtube playback is smoother too with the latest version of PowerUOC. The video still stutters a bit, but it's watchable. Youtube is a slide show for me under stock TFF.
I highly recommend www.tonvid.com for YouTube videos. The site's a lot lighterweight, so videos will play even smoother.
Great work @z970mp and contributors! PowerUOC speeds up stock TenFourFox in a MAJOR way. I found I can re-enable Firefox sync again- I had to disable it on my G4s because the initial sync on each run would bog the system down for a few minutes after app startup.
Page load is much zippier and scrolling is smoother. Also “modern” websites like the ones I build in my day job are now usable on a PowerPC - including CSS animations/transitions, HTML5 canvas and sites with heavy use of JS. Even the web inspector has become usable, which is a big step up for a G4!
Thank you again for supporting the PowerPC platform and community!
Page load is much zippier and scrolling is smoother. Also “modern” websites like the ones I build in my day job are now usable on a PowerPC - including CSS animations/transitions, HTML5 canvas and sites with heavy use of JS. Even the web inspector has become usable, which is a big step up for a G4!
Thank you again for supporting the PowerPC platform and community!
@Dronecatcher @RhianB The base UOC Patch actually has a set of recommended system requirements for maximum effectiveness. I opted to omit this strategy because I didn't think it necessary, which may not be the case after all.
As both of you have been trying it on lower-spec iBooks and PowerBooks and reported rather uneventful changes, maybe I've been amiss in my "one-size-fits-all" stance. Yeah sure, it will improve any (Mozilla) browsing experience on any device, at least a little bit, but to leverage all that it contains may actually require better machines, and for good reason.
I do solidly think there is a difference in it compared to already tweaked prefs (loading times aren't everything it offers), but while I've tried it on lower-spec machines, I haven't done extensive testing on lower-spec machines. Therefore, I believe there may have been a disconnect when there have been disagreements regarding its effectiveness on low-end hardware.
But for video playback on G5s, I can definitely confirm it essentially enables 720p HD video when streaming from YouTube (or Tonvid). Stock prefs present a technically watchable but very choppy experience when streaming 720p HD video, while PowerUOC smooths everything over, effectively putting it in about the same framerate class as 480p. Which is mainly because PowerUOC contains media preferences that force incoming video streams to be hardware-accelerated, which is just one of its many underlying technologies.
And I believe this should be stated... If you do not feel a difference between PowerUOC and a preexisting tweaked preferences file, although I personally do not share your experience in my own tests, please feel free to use whatever makes browsing the Web in its entirety a better experience for you on whatever able hardware available. PowerUOC's whole goal here is to make life easier, not generate cortisol.
Thank you.
As both of you have been trying it on lower-spec iBooks and PowerBooks and reported rather uneventful changes, maybe I've been amiss in my "one-size-fits-all" stance. Yeah sure, it will improve any (Mozilla) browsing experience on any device, at least a little bit, but to leverage all that it contains may actually require better machines, and for good reason.
I do solidly think there is a difference in it compared to already tweaked prefs (loading times aren't everything it offers), but while I've tried it on lower-spec machines, I haven't done extensive testing on lower-spec machines. Therefore, I believe there may have been a disconnect when there have been disagreements regarding its effectiveness on low-end hardware.
But for video playback on G5s, I can definitely confirm it essentially enables 720p HD video when streaming from YouTube (or Tonvid). Stock prefs present a technically watchable but very choppy experience when streaming 720p HD video, while PowerUOC smooths everything over, effectively putting it in about the same framerate class as 480p. Which is mainly because PowerUOC contains media preferences that force incoming video streams to be hardware-accelerated, which is just one of its many underlying technologies.
And I believe this should be stated... If you do not feel a difference between PowerUOC and a preexisting tweaked preferences file, although I personally do not share your experience in my own tests, please feel free to use whatever makes browsing the Web in its entirety a better experience for you on whatever able hardware available. PowerUOC's whole goal here is to make life easier, not generate cortisol.
Thank you.
Last edited:
And what is that reason? As the original patch was written with Pentiums in mind, why would there be a preference for G5 against G4?
But presumably have done extensive testing on higher spec machines...so what is that testing so at least we can follow that path and compare?
Stuff like multithreaded decoding, for example. Although there is an improvement with it present on Pentium IIIs and G4s, I believe it takes greater effect on multi-threaded machines, like G5s, and HT P4s.
Greater numbers of maximum server connections... Higher framerates... etc.
Of course, these are just guesses as to why it might have less pronounced effects on lower-end hardware than more ablebodied configurations, as reported here.
Could you rephrase that?
Last edited: