Author Topic: "Fastest"...!? Isn't this a superlative? I suppose not?! (Read 15808 times)

Saurabhdua · « **on:** April 02, 2010, 06:18:48 AM »

Hi Guys!

Isn't there a REGULATOR for WWW that can testify & ascertain the 'Frivolous' claims often made by browser makers for their product being the 'FASTEST' among breed!

Both Opera & Apple Safari claim their's to be FASTEST!!?

No lawsuit, no litigation from anyone?

This is truly uncharacteristic of the people living across Europe & US; where, being sued has become a common phenomenon!

Any inputs...?

[recovering disk space - old attachment deleted by admin]

BC_Programmer · « **Reply #1 on:** April 02, 2010, 06:47:05 AM »

It's called marketing.

Saurabhdua · « **Reply #2 on:** April 03, 2010, 09:06:26 AM »

There are laws to govern & check the 'Marketing' as well. Advertising Standards councils play a BIG tole to rein in 'Misleading' claims!

How's the IDEA on the possible REGULATOR for Online Marketing companies?

BC_Programmer · « **Reply #3 on:** April 03, 2010, 09:25:38 AM »

You have way to much time on your hands.

tsarles · « **Reply #4 on:** April 07, 2010, 02:04:28 PM »

They probably are both the fastest according to their own benchmarks wich are stacked to put them in the lead. For example, one browser may render html faster, while another may run javascript faster or whatever, and one may have a faster launch time.

All depends on what is being tested, as to how the results are skewed.

I'm pretty impressed with Chrome and Opera as far as performance

patio · « **Reply #5 on:** April 07, 2010, 06:34:05 PM »

Quote from: BC_Programmer on April 03, 2010, 09:25:38 AM

You have way to much time on your hands.

I'll go with this theory....

Lappinga · « **Reply #6 on:** April 10, 2010, 09:01:12 AM »

I read that Camino is actually the fastest but because it's optimized for Mac OS X that is why it isn't a competitor that's why it isn't renown as the fastest. Opera has kept the title for being the fastest web browser for many years now. ^_^

BC_Programmer · « **Reply #7 on:** April 10, 2010, 09:03:04 AM »

I wrote the fastest browser:

Code: [Select]


int main()
{
    return;
}

It doesn't work, but its FAST!

neelchauhan · « **Reply #8 on:** April 15, 2010, 08:08:41 AM »

They aren't really the fastest. It varies on your connnection.
They are tricking the users. The fastest is based on how fast the loading of the browser is.

BC_Programmer · « **Reply #9 on:** April 15, 2010, 11:34:56 AM »

Quote from: neelchauhan on April 15, 2010, 08:08:41 AM

The fastest is based on how fast the loading of the browser is.

Excellent! So your vote goes to BCBrowser:
I've even managed to make it even better!

Code: [Select]


int main()
{

}

kpac · « **Reply #10 on:** April 15, 2010, 11:40:20 AM »

Maybe you could make it even faster!

I know with PHP if you only have one line of code in a function, you don't need any parenthesesses.

BC_Programmer · « **Reply #11 on:** April 15, 2010, 11:45:05 AM »

Quote from: kpac on April 15, 2010, 11:40:20 AM

Maybe you could make it even faster!

I know with PHP if you only have one line of code in a function, you don't need any parenthesesses.

YES! Yes!

Code: [Select]

//credits to kpac for the massive optimization. 
//don't forget to compile with /O2 for maximum optimization
int main();

kpac · « **Reply #12 on:** April 15, 2010, 11:46:54 AM »

mr-bisquit · « **Reply #13 on:** April 17, 2010, 07:48:50 PM »

A fast browser would also be dependent on other factors:
OS, processor, memory.
What's running in the background?
What's the cpu type and available memory?
Is it conole or graphical?
How does it render graphics if it is graphical?
Does the machine it is on have a graphic card- non integrated?

Links and elinks are much faster. Console browsers.
Dillo, amaya, w3c are fast. graphical browsers.

If you can build the package yourelf, you can set the build flags and optimize it on your own.

Saurabhdua · « **Reply #14 on:** April 18, 2010, 12:07:17 AM »

So it is quite likely that the 'so called' fastest browser for Windows 7, couldn't be the one for XP & vice-versa...Right?

Cityscape · « **Reply #15 on:** April 18, 2010, 01:02:01 PM »

Quote from: tsarles on April 07, 2010, 02:04:28 PM

I'm pretty impressed with Chrome and Opera as far as performance

I'd definitely agree here. I'd say Chrome and Opera (on Windows) are the two fastest.

mr-bisquit · « **Reply #16 on:** April 18, 2010, 01:05:22 PM »

Not only that but, it could be the other way around. If the XP system has a faster processor and more memory- plus less crap running in the background- it could go faster. I believe that safari is built on webkit, which is the basis for konqueror. It is not a fast browser- konqueror- but an all purpose file browser with web browsing built in as a function. The Tiny releases use the same function with explorer.

Your best bet with any browser is to slim it down.

What was said earlier is true: optimized for that system.
If you want the best browser, build it from code.

soybean · « **Reply #17 on:** April 19, 2010, 12:01:18 PM »

Quote from: mr-bisquit on April 18, 2010, 01:05:22 PM

If you want the best browser, build it from code.

That may be true but it sure is not practical for feasible for the vast majority of users.

Quote from: Cityscape on April 18, 2010, 01:02:01 PM

I'd definitely agree here. I'd say Chrome and Opera (on Windows) are the two fastest.

I believe so, too.

patio · « **Reply #18 on:** April 19, 2010, 05:20:53 PM »

I have zero issues with FireFox's speed and will stay where i'm at...

rthompson80819 · « **Reply #19 on:** April 19, 2010, 05:47:36 PM »

This is a speed test on browsers from PC Magazine.

Quote

The standard test that tech reviewers use is the WebKit open source project's SunSpider JavaScript benchmark. But a few caveats are in order before anyone takes the results on these tests as the gospel on JavaScript performance. Even some of the most commonly called-upon JavaScript commands are not included in the tests. But they do show something about performance—anyone who's used Chrome knows it's significantly faster than IE7, and its SunSpider number is an order of magnitude faster. All that said, here are my results, using a 2-GHz Athlon AMD 64 X2-based PC with 2GB RAM, with all unnecessary processes shut down via Task Manager.

Browser      SunSpider JavaScript
Benchmark result
(in milleseconds—lower is better)
Firefox 3.6      1,405
Google Chrome 4.0      749
Internet Explorer 7      47,119
Internet Explorer 8      9,015
Internet Explorer 9 Platform Preview    1,310
Opera 10.5      577
Safari 4.0      790

As you can see, the improvement from IE7 to IE8 to IE9 is remarkable. While Chrome, Opera, and Safari still lead by a good margin, the number-two browser Firefox is now in IE's rear-view mirror.

patio · « **Reply #20 on:** April 19, 2010, 06:08:47 PM »

Not a fair comparison...IMHO

Cityscape · « **Reply #21 on:** April 19, 2010, 07:04:09 PM »

Quote from: patio on April 19, 2010, 06:08:47 PM

Not a fair comparison...IMHO

I agree.

rthompson80819 · « **Reply #22 on:** April 19, 2010, 07:21:21 PM »

Quote from: patio on April 19, 2010, 06:08:47 PM

Not a fair comparison...IMHO

How would you do it differently? I've never done any kind of speed test on browsers, but my gut feeling is that the relationships, or ratios between browser speeds, are fairly accurate.

BC_Programmer · « **Reply #23 on:** April 19, 2010, 08:11:17 PM »

First, they don't say wether IE7 is using it's compatibility mode or not, which could seriously change the results, probably bringing it right near IE8 and IE9.

Second: javascript benchmarks are meaningless. If they are going to test browser speed, they should test browser speed, not the speed of their javascript engine. Additionally, considering that all versions of IE use the jscript.dll that is part of the OS for javascript interpretation I find it hard to believe that the numbers would be so skewed.

Browser speed should be tested by disabling client-side script entirely and seeing how quickly it loads a page containing standard HTML.

Also, the sunspider test states that it is a "real-world" test, and yet, it says it doesn't test the DOM.

The main thing javascript is used for in the "real world" is for interacting WITH the DOM. therefore if one browsers DOM implementation is faster then another, this will influence the results of <real-world> browsing. it additionally states that it doesn't perform "micro" testing. and yet at the same time I witness tests that measure the speed of the bitwise operators. something I would call "micro-testing".

using a javascript speed test as a measure of a browser's speed is like measuring the speed of a game's sound subsystem to determine framerates.

mr-bisquit · « **Reply #24 on:** April 19, 2010, 09:58:36 PM »

System: P4 1280M RAM DEbian Lenny i386
Browser: Iceweasel

Code: [Select]

============================================
RESULTS (means and 95% confidence intervals)
--------------------------------------------
Total:                 11212.0ms +/- 25.4%
--------------------------------------------

  3d:                   1211.6ms +/- 13.9%
    cube:                508.8ms +/- 34.4%
    morph:               400.4ms +/- 44.3%
    raytrace:            302.4ms +/- 43.1%

  access:               1350.8ms +/- 10.8%
    binary-trees:        125.2ms +/- 53.8%
    fannkuch:            607.0ms +/- 41.9%
    nbody:               405.8ms +/- 39.1%
    nsieve:              212.8ms +/- 34.6%

  bitops:                886.0ms +/- 18.9%
    3bit-bits-in-byte:   134.4ms +/- 42.6%
    bits-in-byte:        254.6ms +/- 21.9%
    bitwise-and:         189.6ms +/- 57.1%
    nsieve-bits:         307.4ms +/- 37.0%

  controlflow:           126.4ms +/- 62.7%
    recursive:           126.4ms +/- 62.7%

  crypto:                669.6ms +/- 33.7%
    aes:                 203.8ms +/- 46.0%
    md5:                 200.4ms +/- 32.6%
    sha1:                265.4ms +/- 82.7%

  date:                 1358.0ms +/- 48.5%
    format-tofte:        821.8ms +/- 60.3%
    format-xparb:        536.2ms +/- 50.8%

  math:                 1573.4ms +/- 36.5%
    cordic:              625.4ms +/- 51.9%
    partial-sums:        632.8ms +/- 24.3%
    spectral-norm:       315.2ms +/- 73.5%

  regexp:               1072.4ms +/- 61.5%
    dna:                1072.4ms +/- 61.5%

  string:               2963.8ms +/- 32.1%
    base64:              374.2ms +/- 62.7%
    fasta:               574.6ms +/- 36.8%
    tagcloud:            462.4ms +/- 41.8%
    unpack-code:        1079.6ms +/- 53.9%
    validate-input:      473.0ms +/- 29.8%

mr-bisquit · « **Reply #25 on:** April 19, 2010, 09:59:43 PM »

Same system
Browser: arora

Code: [Select]

============================================
RESULTS (means and 95% confidence intervals)
--------------------------------------------
Total:                 29537.2ms +/- 28.3%
--------------------------------------------

  3d:                   4730.6ms +/- 64.8%
    cube:               1244.2ms +/- 52.9%
    morph:              1836.0ms +/- 55.0%
    raytrace:           1650.4ms +/- 86.3%

  access:               5426.8ms +/- 38.9%
    binary-trees:        635.6ms +/- 49.3%
    fannkuch:           2129.0ms +/- 37.1%
    nbody:              1245.2ms +/- 61.1%
    nsieve:             1417.0ms +/- 30.2%

  bitops:               4425.4ms +/- 30.8%
    3bit-bits-in-byte:  1047.0ms +/- 40.0%
    bits-in-byte:       1128.6ms +/- 33.2%
    bitwise-and:         917.0ms +/- 25.7%
    nsieve-bits:        1332.8ms +/- 26.7%

  controlflow:           652.6ms +/- 30.6%
    recursive:           652.6ms +/- 30.6%

  crypto:               2427.0ms +/- 30.9%
    aes:                 570.6ms +/- 38.6%
    md5:                 933.0ms +/- 33.1%
    sha1:                923.4ms +/- 25.7%

  date:                 2630.0ms +/- 58.3%
    format-tofte:        891.0ms +/- 51.1%
    format-xparb:       1739.0ms +/- 62.1%

  math:                 3608.0ms +/- 44.0%
    cordic:             2092.0ms +/- 62.2%
    partial-sums:        803.2ms +/- 18.3%
    spectral-norm:       712.8ms +/- 23.0%

  regexp:                973.8ms +/- 42.7%
    dna:                 973.8ms +/- 42.7%

  string:               4663.0ms +/- 37.3%
    base64:             1075.2ms +/- 35.9%
    fasta:              1052.8ms +/- 30.3%
    tagcloud:            714.6ms +/- 33.5%
    unpack-code:         961.8ms +/- 49.7%
    validate-input:      858.6ms +/- 58.4%

mr-bisquit · « **Reply #26 on:** April 19, 2010, 10:11:55 PM »

Machine specs recently given. OpenBSD 4.6 x86
Browser: Konqueror

Quote

A script on this page is causing KHTML to freeze. If it continues to run, other applications may become less responsive.
Do you want to abort the script?

mr-bisquit · « **Reply #27 on:** April 19, 2010, 10:15:26 PM »

OpenBSD, same machine
Browser: firefox

Code: [Select]

============================================
RESULTS (means and 95% confidence intervals)
--------------------------------------------
Total:                  7558.6ms +/- 2.2%
--------------------------------------------

  3d:                    899.8ms +/- 0.9%
    cube:                374.8ms +/- 0.9%
    morph:               252.0ms +/- 1.7%
    raytrace:            273.0ms +/- 2.5%

  access:               1118.4ms +/- 1.3%
    binary-trees:        259.2ms +/- 1.1%
    fannkuch:            366.0ms +/- 1.8%
    nbody:               400.4ms +/- 1.8%
    nsieve:               92.8ms +/- 1.7%

  bitops:                667.4ms +/- 2.9%
    3bit-bits-in-byte:   101.6ms +/- 0.7%
    bits-in-byte:        135.0ms +/- 1.1%
    bitwise-and:         215.8ms +/- 8.8%
    nsieve-bits:         215.0ms +/- 1.2%

  controlflow:            73.6ms +/- 1.5%
    recursive:            73.6ms +/- 1.5%

  crypto:                369.6ms +/- 5.8%
    aes:                 146.8ms +/- 14.9%
    md5:                 108.6ms +/- 0.6%
    sha1:                114.2ms +/- 0.9%

  date:                  981.2ms +/- 5.8%
    format-tofte:        399.2ms +/- 2.3%
    format-xparb:        582.0ms +/- 8.3%

  math:                  948.6ms +/- 12.1%
    cordic:              402.4ms +/- 30.0%
    partial-sums:        392.2ms +/- 3.0%
    spectral-norm:       154.0ms +/- 4.0%

  regexp:                422.4ms +/- 3.3%
    dna:                 422.4ms +/- 3.3%

  string:               2077.6ms +/- 0.9%
    base64:              236.2ms +/- 3.0%
    fasta:               611.4ms +/- 2.3%
    tagcloud:            340.8ms +/- 1.6%
    unpack-code:         561.0ms +/- 0.5%
    validate-input:      328.2ms +/- 2.9%

Cityscape · « **Reply #28 on:** April 19, 2010, 10:15:49 PM »

I absolutely hate Konqueror.

mr-bisquit · « **Reply #29 on:** April 19, 2010, 10:21:28 PM »

Same machine
FreeBSD 8 x86
browser: konqueror

Code: [Select]

A script on this page is causing KHTML to freeze. If it continues to run, other applications may become less responsive.
Do you want to abort the script?

mr-bisquit · « **Reply #30 on:** April 19, 2010, 10:22:42 PM »

Same machine
FreeBSD
Browser: Firefox

Quote

============================================
RESULTS (means and 95% confidence intervals)
--------------------------------------------
Total: 2587.0ms +/- 6.7%
--------------------------------------------

3d: 387.0ms +/- 2.5%
cube: 114.0ms +/- 5.8%
morph: 91.4ms +/- 0.7%
raytrace: 181.6ms +/- 2.8%

access: 346.8ms +/- 3.5%
binary-trees: 90.2ms +/- 1.8%
fannkuch: 159.8ms +/- 5.3%
nbody: 77.6ms +/- 8.7%
nsieve: 19.2ms +/- 2.9%

bitops: 73.8ms +/- 3.7%
3bit-bits-in-byte: 2.8ms +/- 57.9%
bits-in-byte: 14.4ms +/- 4.7%
bitwise-and: 2.8ms +/- 19.9%
nsieve-bits: 53.8ms +/- 3.0%

controlflow: 73.0ms +/- 1.2%
recursive: 73.0ms +/- 1.2%

crypto: 140.4ms +/- 5.2%
aes: 78.6ms +/- 4.3%
md5: 40.2ms +/- 4.0%
sha1: 21.6ms +/- 14.4%

date: 448.0ms +/- 22.4%
format-tofte: 237.0ms +/- 21.1%
format-xparb: 211.0ms +/- 24.0%

math: 203.4ms +/- 35.1%
cordic: 118.4ms +/- 47.9%
partial-sums: 64.8ms +/- 22.8%
spectral-norm: 20.2ms +/- 17.6%

regexp: 147.6ms +/- 22.4%
dna: 147.6ms +/- 22.4%

string: 767.0ms +/- 2.8%
base64: 42.0ms +/- 3.6%
fasta: 172.8ms +/- 3.0%
tagcloud: 216.2ms +/- 4.8%
unpack-code: 247.8ms +/- 4.6%
validate-input: 88.2ms +/- 18.5%

mr-bisquit · « **Reply #31 on:** April 19, 2010, 10:24:20 PM »

Quote from: Cityscape on April 19, 2010, 10:15:49 PM

I absolutely hate Konqueror.

Yeah, but you can see all of the nasty cookies with it.

mr-bisquit · « **Reply #32 on:** April 19, 2010, 10:25:18 PM »

See. It also depends on the OS, settings, build, etc.
The same machine gives three different outputs for the same browser. Iceweasel is a rebranded firefox.
Total: 2587.0ms +/- 6.7% Firefox35
Total: 11212.0ms +/- 25.4% Iceweasel
Total: 7558.6ms +/- 2.2% Firefox3

When you look at it, the test is even more skewed.

And I am running other processes at the same time.

soybean · « **Reply #33 on:** April 20, 2010, 08:23:42 AM »

Quote from: mr-bisquit link=topic=102791.msg700231#msg700231

When you look at it, the test is even more skewed.

How so?

mr-bisquit · « **Reply #34 on:** April 20, 2010, 10:16:12 AM »

Quote from: soybean on April 20, 2010, 08:23:42 AM

How so?

It's done only on one operating system. That's the first.
It isn't done under real world circumstances. That's the second.
It isn't done under different connection speeds. That's the third.

I'll run the same tests later on the G3 and post them.

Since my amd64 is shot to *censored*, I can't run the tests on that architecture.
I can run them on my laptop and maybe under other circumstances. Say, perhaps, a poor ppoe connection.

I don't run konqueror unsecured.

mr-bisquit · « **Reply #35 on:** April 20, 2010, 10:00:12 PM »

Debian G3/750 400mHz 384M RAM
Browser Kazahakase

Code: [Select]

============================================
RESULTS (means and 95% confidence intervals)
--------------------------------------------
Total:                 31374.8ms +/- 0.6%
--------------------------------------------

  3d:                   3856.6ms +/- 1.8%
    cube:               1418.8ms +/- 2.9%
    morph:              1393.2ms +/- 7.4%
    raytrace:           1044.6ms +/- 1.4%

  access:               4530.4ms +/- 2.4%
    binary-trees:        499.2ms +/- 8.8%
    fannkuch:           1799.8ms +/- 0.5%
    nbody:              1459.4ms +/- 5.9%
    nsieve:              772.0ms +/- 0.4%

  bitops:               3470.2ms +/- 0.4%
    3bit-bits-in-byte:   595.2ms +/- 1.0%
    bits-in-byte:        851.2ms +/- 1.7%
    bitwise-and:         862.8ms +/- 1.0%
    nsieve-bits:        1161.0ms +/- 0.4%

  controlflow:           419.0ms +/- 2.1%
    recursive:           419.0ms +/- 2.1%

  crypto:               1899.4ms +/- 0.5%
    aes:                 733.0ms +/- 1.2%
    md5:                 584.4ms +/- 1.1%
    sha1:                582.0ms +/- 2.5%

  date:                 3588.2ms +/- 1.4%
    format-tofte:       2104.6ms +/- 1.8%
    format-xparb:       1483.6ms +/- 2.0%

  math:                 3615.4ms +/- 3.4%
    cordic:             1400.4ms +/- 0.8%
    partial-sums:       1439.0ms +/- 9.1%
    spectral-norm:       776.0ms +/- 0.6%

  regexp:               1985.0ms +/- 4.4%
    dna:                1985.0ms +/- 4.4%

  string:               8010.6ms +/- 0.3%
    base64:             1017.6ms +/- 2.2%
    fasta:              2153.6ms +/- 0.6%
    tagcloud:           1509.4ms +/- 2.5%
    unpack-code:        2180.4ms +/- 0.6%
    validate-input:     1149.6ms +/- 0.9%

mr-bisquit · « **Reply #36 on:** April 20, 2010, 10:20:10 PM »

Same system PPC processor
Iceweasel

Code: [Select]

============================================
RESULTS (means and 95% confidence intervals)
--------------------------------------------
Total:                 32972.0ms +/- 19.3%
--------------------------------------------

  3d:                   3718.4ms +/- 8.2%
    cube:               1349.0ms +/- 6.6%
    morph:              1327.4ms +/- 8.5%
    raytrace:           1042.0ms +/- 11.5%

  access:               4245.2ms +/- 4.6%
    binary-trees:        466.0ms +/- 11.0%
    fannkuch:           1735.8ms +/- 0.4%
    nbody:              1304.2ms +/- 7.4%
    nsieve:              739.2ms +/- 28.2%

  bitops:               3641.0ms +/- 34.4%
    3bit-bits-in-byte:   601.2ms +/- 35.0%
    bits-in-byte:        881.0ms +/- 29.9%
    bitwise-and:         883.2ms +/- 29.2%
    nsieve-bits:        1275.6ms +/- 40.7%

  controlflow:           414.8ms +/- 44.3%
    recursive:           414.8ms +/- 44.3%

  crypto:               2014.0ms +/- 41.2%
    aes:                 773.4ms +/- 38.7%
    md5:                 623.6ms +/- 43.0%
    sha1:                617.0ms +/- 42.5%

  date:                 4237.6ms +/- 48.1%
    format-tofte:       2555.8ms +/- 48.3%
    format-xparb:       1681.8ms +/- 47.9%

  math:                 4002.8ms +/- 44.0%
    cordic:             1596.2ms +/- 46.2%
    partial-sums:       1557.4ms +/- 40.7%
    spectral-norm:       849.2ms +/- 46.2%

  regexp:               2042.8ms +/- 18.5%
    dna:                2042.8ms +/- 18.5%

  string:               8655.4ms +/- 30.1%
    base64:              947.2ms +/- 5.9%
    fasta:              2105.4ms +/- 1.1%
    tagcloud:           1777.4ms +/- 46.8%
    unpack-code:        2547.0ms +/- 47.1%
    validate-input:     1278.4ms +/- 46.7%

Computer Hope Forum

News:

Author Topic: "Fastest"...!? Isn't this a superlative? I suppose not?! (Read 15808 times)

Saurabhdua

BC_Programmer

Saurabhdua

BC_Programmer

tsarles

patio

Lappinga

BC_Programmer

neelchauhan

BC_Programmer

kpac

BC_Programmer

kpac

mr-bisquit

Saurabhdua

Cityscape

mr-bisquit

soybean

patio

rthompson80819

patio

Cityscape

rthompson80819

BC_Programmer

mr-bisquit

mr-bisquit

mr-bisquit

mr-bisquit

Cityscape

mr-bisquit

mr-bisquit

mr-bisquit

mr-bisquit

soybean

mr-bisquit

mr-bisquit

mr-bisquit