Benchmark vs Waterfox on Waterfox Homepage

[stravinsky]

@ tribaljet :

1. AFAIK MC doesnt have an Intel CPU. So he cant really build specifically for Intel using ICC.
2. ICC is not free. The complete/paid version has the complete optimisations. The demo/free doesnt have all. So it would be similar to MSVC
3.Performance gains would probably be very limited.
4.you can try cross-compiling PM in linux with GCC4.7 for windows. That will definitely give you much much more control over optimisations. And you can build with march=native. So if you have an ivy bridge, you will get a browser built specifically using ivy-bridge instruction set.
Of course, it will still perform like PM, but you will get the satisfaction of using the fastest possible browser based on mozilla code.
5.Other possible PGO+SSE3+ICC build FF clones are :
Lawliet firefox and PCX firefox.
6. MC believes AMD processors are faster than Intel. So in any case he wont build for Intel.

[tribaljet]

@ tribaljet :

1. AFAIK MC doesnt have an Intel CPU. So he cant really build specifically for Intel using ICC.
2. ICC is not free. The complete/paid version has the complete optimisations. The demo/free doesnt have all. So it would be similar to MSVC
3.Performance gains would probably be very limited.
4.you can try cross-compiling PM in linux with GCC4.7 for windows. That will definitely give you much much more control over optimisations. And you can build with march=native. So if you have an ivy bridge, you will get a browser built specifically using ivy-bridge instruction set.
Of course, it will still perform like PM, but you will get the satisfaction of using the fastest possible browser based on mozilla code.
5.Other possible PGO+SSE3+ICC build FF clones are :
Lawliet firefox and PCX firefox.
6. MC believes AMD processors are faster than Intel. So in any case he wont build for Intel.

Thank you for your reply stravinsky.

Moonchild can build with ICC, but not having an Intel CPU does prevent him from testing performance improvements, which is why a tester's group could be made.

I wasn't aware that ICC was paid, and if it follows similar brand products, licensing fees might be expensive. The fact that the free version lacks the complete optimization set is quite a letdown, unfortunately.

I do wonder whether performance gains would be that limited, even putting aside specific content that would benefit most from it.

The instruction set update of Ivy Bridge was smaller than expected over the previous generation, and most seem to improve over cryptography and little else, which is understandably moot for browser performance.

What do you think of Lawliet Firefox, PCX Firefox and Waterfox when compared to Pale Moon?

AMD had a good run during the Athlon 64 era, almost close to Intel's market share at the time, but ever since Intel released Core 2 Duo, AMD has been on a downslide ever since, up to the point AMD publicly stated it no longer competes with Intel in terms of performance, so that alone should point out that AMD simply isn't up to scratch, regardless of personal opinions

[tribaljet]

I spent the past few hours testing the latest Waterfox 14.0.1 to get a better understanding of how both Pale Moon and Waterfox work.

First of all, I should mention I ran both on a 2nd generation Intel Core i7 (therefore, with fully supported instruction set), with 8GB of system memory and Windows 7 (x64), on a mechanical drive.

The first thing I noticed was that Waterfox felt smoother and more responsive than Pale Moon, smoothness being the relevant point to take from this, as responsiveness was different but the gap was rather narrow. In terms of actual browsing performance, it did feel ever so slightly faster than Pale Moon, but not in a significant way, which puzzled me as such differences weren't to be expected, so I then decided to run a couple benchmarks (as unsubstancial as they are to me), which then showed marked improvements across the board, but as Moonchild already mentioned (and like hardware manufacturers also do), results might tend to be skewed as to favor benchmarking rather than real world usage.

Usability wise, I might (special focus on "might") be able to use Waterfox without much hassle if coming from Firefox, but coming from Pale Moon simply made Waterfox awful to use, Pale Moon is pure delight when compared to official Firefox and all unofficial builds. While some might like it, the "load tab as it is made active" is something that doesn't work for me, the hidden (yet manually enabled) tab save made me restless at every client shutdown, and the Waterfox installer didn't have any new feature, so it felt like it had a facelift just for the sake of it, just to name a few things.

All things considered, I'm happily returning to my own personal and familiar Pale Moon *pets his customized client*, hoping that improvements might be on the cards in order to put to proper use to the still untapped hardware I'm running on the foreseeable future.

[megaman]

What do you think of Lawliet Firefox, PCX Firefox and Waterfox when compared to Pale Moon?

Lawlietfox uses the PCX PGO, so it is PCX under the hood as well.
What surprised me is that the Yahoo! Mail was loading as fast as it does to me in Chromium-based browsers and Opera, astonishing, but it seems to render too fast that it actually starts to choke and instead uses pauses, noticeable in the loading bar.

WF and PM would still seem unfair in making a comparison until they are both on the same versions, each version has performance fixes in Fx, or not, maybe just tweaks that make them faster in some perspective.

[Hoggyy]

I say experimental build as in beta build for a future stable build for the majority of users, as AMD users are a minority.
EDIT: Beta build for an Intel build, as for all intents and purposes, the AMD build would effectively be the regular build currently available.
/quote]

Nah - the AMD version should be first! hehe As I understand it, the Fx code itself is slightly skewed towards Intel in the first place. And besides - it seems MC might have an AMD anyways.

6. MC believes AMD processors are faster than Intel. So in any case he wont build for Intel.

That's nice to hear! I read that they are for graphics anyways, which webpages are highly composed of these days. Which is why I bought an AMD machine - plus they're cheaper than Intel. I've read that raw processing power is better on the Intel side - but they're just SO damn expensive for what you get.

I also went for AMD because I read that their processors can be easily replaced with better/faster versions most of the time. I'm sick of Intel with all their socket changes, ensuring that people can't do direct upgrades. I'm just hoping that will be possible with my new laptop here.


I really wish I could make a contributed build, and may try to do that eventually.. The problem is that I have physical and cognitive limitations - I forget easily and get confused easily. Nevertheless - I may want to try that soon. I'd hope to get enough experience to even build my own Gargoyle firmware for my router at some point (I REALLY want to change how often it writes to the flash.).

BTW - is the MSVC compiler free? That could make a big difference for me.

[tribaljet]

That's nice to hear! I read that they are for graphics anyways, which webpages are highly composed of these days. Which is why I bought an AMD machine - plus they're cheaper than Intel. I've read that raw processing power is better on the Intel side - but they're just SO damn expensive for what you get.

I also went for AMD because I read that their processors can be easily replaced with better/faster versions most of the time. I'm sick of Intel with all their socket changes, ensuring that people can't do direct upgrades. I'm just hoping that will be possible with my new laptop here.


I really wish I could make a contributed build, and may try to do that eventually.. The problem is that I have physical and cognitive limitations - I forget easily and get confused easily. Nevertheless - I may want to try that soon. I'd hope to get enough experience to even build my own Gargoyle firmware for my router at some point (I REALLY want to change how often it writes to the flash.).

BTW - is the MSVC compiler free? That could make a big difference for me.

Do note that "Nah - the AMD version should be first! hehe As I understand it, the Fx code itself is slightly skewed towards Intel in the first place. And besides - it seems MC might have an AMD anyways." was not said by me at all, therefore has no place under text quoted by myself. I'm sure it must've been a mistake, but I just want to prevent things getting taken out of context

AMD CPUs have both single and multithreaded performance inferior to Intel CPUs, except on the most absurdly (special focus on absurdly) multithreaded scenarios, which exclude all consumer usage, browsers included, which is quite relevant when considering that browsers are barely dual threaded. The "you get what you pay for" motto applies strongly here, despite Intel having higher price points than what they should be selling their products at.

About AMD's easier (read, cheaper) upgrade cycles, that's a rather simple situation, and one that doesn't translate into savings, as each AMD stepping is so incremental that the user ends up replacing hardware enough times to match the seldom yet substantial Intel upgrade cycle, so it all ends up on the same level, money wise.

There is a free version of MSVC, so that's something you can get without any hassle and tinker away

[Hoggyy]

Yeah - that was a mistake.. I tried to do a multi-quote, but it apparently didn't work out so well and I forgot to check the preview... Naturally. (I call those 'fibro moments' -- kinda like 'senior moments', but for those with Fibromyalgia... I have those VERY frequently! )

It would have been nice to be able to afford a top of the line Intel, but when going on disability, for that money I could only afford a top-of-the-line AMD. :-/ At least I think this machine (in sig) is among the best AMD's... I lost touch with what was going on in the CPU/APU world, so I had to do a crash-course to decide what laptop to get.

And yes - I really WISH that Fx could get their act together and get into mutli-thead/multi-cpu territory! Maybe one of these centuries.. I thought MC did a little bit of that with moving the add-on stuff (or something) to a separate process. Maybe MC could work on that himself since Mozilla seems to be taking their sweet old time on that. That would also further set PM apart. However, I fear that that could be a monumental undertaking.

[Hoggyy]

About AMD's easier (read, cheaper) upgrade cycles, that's a rather simple situation, and one that doesn't translate into savings, as each AMD stepping is so incremental that the user ends up replacing hardware enough times to match the seldom yet substantial Intel upgrade cycle, so it all ends up on the same level, money wise.

I would certainly wait for a few steps to go by first. Otherwise it wouldn't be worth it. I'm hoping that this processor could eventually be replaced with an 8-core APU, but I'm not sure if laptops are included in those simple-processor-replacement scenarios.

[tribaljet]

I wholeheartedly agree with both things you said, Hoggyy, in regards to both improved multithreading on Firefox, as well as taking larger steps when doing AMD upgrades

About mobile CPU replacements, it tends to depend on how modular the laptop model is from the start, but most times it's not worth the asking price for replacement units IMHO.

[megaman]

And yes - I really WISH that Fx could get their act together and get into mutli-thead/multi-cpu territory! Maybe one of these centuries.. I thought MC did a little bit of that with moving the add-on stuff (or something) to a separate process. Maybe MC could work on that himself since Mozilla seems to be taking their sweet old time on that. That would also further set PM apart. However, I fear that that could be a monumental undertaking.

Mozilla is working with a possibility like that on Project Snappy, and it might end up making different threads for several things.

AMD's latest offerings are still very performing. Like Intel, they actually are getting too fast for applications requirements. The only thing holding back applications was the HDD, processor is peaking too much. This is my speculation and AMD's new vision, ever since they called it quits on competing with Intel.