Speed Up Web Browsing


Recommended Posts

Some of you might know about this one, some might not. I read about it a long time ago, and been doing it for months now. I dont remember exactly where I read it from, but if someone knows the original link do share. This speeds up web browsing by alot, atleast for me it has. Try it but make sure you keep a backup note on the original paths and such being replaced. If it doesnt work, you can just go back to normal. If it does not work I am letting you all know right now, YOU WILL NOT BE ABLE TO GO BACK ONLINE. You have been warned, so try it with caution and have backups.

First, What youll need:

Get your hands on a good RAM Disk. I am using SuperSpeed Software's RAMDisk Plus 7.0. I got it off a friend who bought it a lil while back. There are free ones available out there though. There is the Microsoft one, which I dont recommend. And my best bet for anyone who doesnt wanna buy a RAMDisk program is ARsoft's RAMDisk, which can be found hereRAMDisk. They have stopped the program, but you can still download it.

Configuration:

I left everything default on the RAMDisk. If you are using the ARSoft one do the same, except change the size to 65MB and select emulate local hard disk. Make sure to leave the drive letter as Z.

The harder parts:

Once all that is configured, we go into the more complicated part. Get that RegEdit running (Start>Run>Regedit.exe) and surf your way over to the following...

-HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services-

From that location, youll want to find each of the following...

Afd

Netbt

Tcpip

Ipnat

In each of those change the image path to your Ramdisk, (Z:\).

Z:\<service name>

It would go as follows

Z:\afd.sys

Z:\netbt.sys

Z:\tcpip

Z:\ipnat

Once your done with that, make sure to also set your History and Cookies folder in there as well. To do that just search the Registry for anything called "Cookies", without the quotes. Change the directory of that to Z:\Cookies. In that same area search for history and change the directory to Z:\History. Search for all instances of these two files. I believe there are three in XP.

Now open Notepad and put the following script in there...

<Begin>

@echo on

Echo Starting TCP/IP Services

copy c:\windows\System32\DRIVERS\afd.sys z:\

copy c:\windows\System32\DRIVERS\netbt.sys z:\

copy c:\windows\System32\DRIVERS\tcpip.sys z:\

copy c:\windows\System32\DRIVERS\ipnat.sys z:\

net start afd

net start netbt

net start tcpip

net start ipnat

<End>

Make sure to leave out the <Begin/End> parts, as I just put those there to make it easier for some. Save that Notepad file as login.cmd and save it on your main HD, which for most of us is C:\. Then, grab the file and drag and drop it into the Startup folder in your Start>Programs menu. This way, everytime you startup your PC, XP will place those 4 files into the RAMDisk drive so that they can be used. IF THIS DOES NOT HAPPEN, REMEMBER TO KEEP A BACKUP OF THE ORIGINAL DIRECTORIES SO YOU CAN UNDO THE BEFOREHAND PROCESS.

Before you restart open a IE6/7, and select Tools>Internet Options>Settings. Adjust it to the desired size (50MB works fine) and select "Move folder" and point it at your RAMDisk. Hit ok, apply, and proceed to reboot your computer.

This helped speed up my internet browsing, and it also deletes your temporary files and such from being on the HD. Now below will be the ORIGINAL paths, in case you need to go into regedit and undo any of the things you made. Once you do that, just delete login.cmd from the startup folder and from the C:\ drive. You can then uninstall the RAMDisk, and voila, back to normal. Reboot for the changes to take effect.

c:\windows\System32\DRIVERS\afd.sys

c:\windows\System32\DRIVERS\netbt.sys

c:\windows\System32\DRIVERS\tcpip.sys

c:\windows\System32\DRIVERS\ipnat.sys

Hope it works for you, like it has for me. I in no way made this up myself, so I am not to take credit for it working or not. I just felt Id share a trick Ive been doing for some time now. Enjoy!!! :yes:

Link to comment
https://www.neowin.net/forum/topic/351721-speed-up-web-browsing/
Share on other sites

So...the genius behind this tweak is to tell the computer to write internet files onto a disk as opposed to what it normally does, which is to write files onto a...disk...

So at best, you'll get the same speed, because the computer is writing these files onto the harddrive, in a different directory. At worst, you'll lose speed because the computer is trying to write files onto your "RAMDisk".

So...the genius behind this tweak is to tell the computer to write internet files onto a disk as opposed to what it normally does, which is to write files onto a...disk...

586300428[/snapback]

Huh?

RAMDisk is a virtual "hard-drive" persay that is in your RAM. So no, its not writing it to a "disk". I dunno how much this would speed it up, but in theory, it sounds like a good idea. I always wanted to put my pagefile on a RAMdisk, so i could get a nice little paradox goin on there.

Now what you SHOULD do, is go out and get a RAMDrive.

http://www.cenatek.com/product_rocketdrive.cfm

Then install your OS onto that, THEN you could not only surf faster, but do everything faster. :)

  • 3 months later...

Well most modern browsers save cache to memory now anyways. Like Fire Fox and Opera.

So kinda pointless, unless you just have to use IE. :p

Then using hardware RAM drive is a great idea if you can afford it, sadly I can not. :(

Yes install OS on one RAM drive, browsers and other internet related appz on anther, and games on another one, page file on another one, and everything else on another one. And be sure and use 10,000 RPM hard disks in Raid. hehe That would rock!!

Hmm, I wonder if you COULD put your pagefile on the RAM drive. I got 2GB of RAM and most of it is unused, id love to try that.

586300770[/snapback]

That would defeat the point of having a Page file - because a page file, by definition, 'pages' areas of memory to the Hard disk to "

free up RAM".

Kinda stupid to page areas of one part of RAM to ... another part of RAM.

Better off just turning off paging altogether if you *really* think you have enough RAM.

Well most modern browsers save cache to memory now anyways. Like Fire Fox and Opera.

So kinda pointless, unless you just have to use IE. :p

Then using hardware RAM drive is a great idea if you can afford it, sadly I can not. :(

Yes install OS on one RAM drive, browsers and other internet related appz on anther, and games on another one, page file on another one, and everything else on another one. And be sure and use 10,000 RPM hard disks in Raid. hehe That would rock!!

586747761[/snapback]

Also a good point- since they are trying to use RAM now (since most modern system have a decent amount of RAM) that is *in theory* a good idea - until, of course, you get to a point where that RAM is needed by another resource-intensive / RAM intensive program - and then that cache is save ... again ... to a hard drive.

Not *always* a myth. When used in terms of something like compiling, it can be of a *tremendous benefit.

I have stats to prove it - in emerging an app in Gentoo I can time how long it takes to do so when using a normal compile method and when I mount a particular directory, /var/tmp/portage to a tmpfs (basically same concept as a RAM drive in Gentoo) - compile time is almost 1/3 of regular.

In a Windows environment I can agree with you - there is so much crap out there that it is *impossible* to have a system that will not page *something* at some time - and thus a RAM drive will be a waste of time unless you have something inordinately high, like 16 GB or more, and can create a small 500 MB RAM drive for use as a chance that does not get paged - but the key is to have sufficient RAM available to allow for this - and most people don't get sufficient RAM to begin with.

I completed building my machine 3.5 years ago next week, and it had 1.25 GB RAM - I took out the 256 MB and left the dual 512s for dual channel purposes, and it is 8still* running 1.0 GB - and I know people buying machines *now* that only have 1 GB.

define faster. IE loads faster b/c half of its components are in active use by Windows Explorer (which is used to display most normal desktops in XP), so it is fast to load. but as far as rendering pages, I have seen the gap / margin between Fx and IE get very tiny, especially in terms of rendering, with the 1.5 Betas. And, after loading 1.5 Bx once, the next several times I load it it loads instantaneously, as fast as IE - and sometimes, depending upon system resource usage, faster than IE.

So...the genius behind this tweak is to tell the computer to write internet files onto a disk as opposed to what it normally does, which is to write files onto a...disk...

So at best, you'll get the same speed, because the computer is writing these files onto the harddrive, in a different directory.  At worst, you'll lose speed because the computer is trying to write files onto your "RAMDisk".

586300428[/snapback]

Just for you:

Access Time for Hardisk: 8 to 10 ms (or even more)

Access Time for Ramdisk: less 70 us (microseconds) , or 0.070 ms

Transfer rates for ram are also way better for ramdisk then harddrives.

Additional:

Bigger security: all cookies and temp files are deleted when the computer gets turned off.

define faster.  IE loads faster b/c half of its components are in active use by Windows Explorer (which is used to display most normal desktops in XP), so it is fast to load.  but as far as rendering pages, I have seen the gap / margin between Fx and IE get very tiny, especially in terms of rendering, with the 1.5 Betas.  And, after loading 1.5 Bx once, the next several times I load it it loads instantaneously, as fast as IE - and sometimes, depending upon system resource usage, faster than IE.

586772612[/snapback]

faster: ie browses and loads faster.

you would be better off just disabling your page file.
Disabling the page file is a Myth

Internet Explorer is faster then Firefox in all aread except script speed. Browser Speed Comparison Test. FYI Opera is faster then both of them.

Windows uses RAM most efficiently without the overhead of a RAMDisk. Adjusting the size of your browsers cache would have more a positive effect.

  • 2 years later...
Windows uses RAM most efficiently without the overhead of a RAMDisk. Adjusting the size of your browsers cache would have more a positive effect.

Oh, really? I disagree!

(& this stuff will work for performance, and even a bit for security gains, for anyone & here is how):

------------------------

1.) Move webpage caches, history, & logging to it, for webbrowsers!

Each is easily doable in today's webbrowsers, either via their GUI interfaces, OR configuration files (or registry entries).

This avoids clutter & fragmentation on your main OS + programs bearing disk drive.

This also speeds up access to they, & largely, they are "READ" based, such as graphics (larger part) & static for HUGE periods & unchanging... best & fastest read they can get from a disk, is from an SSD!

Now, the text part? If done as I do it for security AND speed?? I use NTFS formatted disks: This speeds up their already many orders of magnitude faster pickup from disk vs. std. mechanical HDD's even more, by using compression (smaller filemass reads up faster)... & the decompress stage today is hugely offset by today fast CPU's & RAM (more than offset by gain in access/seek & loads noted above).

------------------------

2.) Same with EventLogs!

Via this registry entry (this is one example, but the FILE value alteration is the same on ANY logs near here too, like Application, System, Internet Explorer, Security, SQLServer &/or IIS ones, you name it):

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Eventlog\System

Change the FILE value to the disk & folder + filename of your choice to gain speed & yes, potentially security too (by keeping dummies in the old locations, where the new ones build on a disk that does not take an access/speed hit & works faster too, like an SSD)

OTHER APPS CAN DO THIS TOO, WinZip being an example of an app that does logging... DrWatson/DrWtsn32 is yet another, & the lits goes on.

------------------------

3.) Pagefile.sys placement

I do this on the primary/first partition of my CENATEK RocketDrive Solid-State RamDisk: Far faster access than std. mechanical disks, especially considering paging bypasses the filesytem driver & goes direct in "RAW WRITES" to the diskdrive like its own disk/partition really & F A S T E R by far alone, but, this helps it more (speed of access mainly is why, & also helping to stop fragmentation by the pagefile.sys during its bootup./shutdown grow-shrink (if needed) patterns... grow happens, shrink I have never actually seen occur, unless a user makes it happen, OR, if done by pagefile.sys wipes (possible via reg hack for security purposes)).

------------------------

THIS IS NOT SOFTWARE BASED RAMDRIVES HERE... this is solid-state, TRUE ssd (not FLASH stuff)... very fast & VERY useful, in overcoming objections to ramdisk usage (to which some points on softwares ones DO apply, & I agree, unless a system has over 4gb of memory on it... then, you CAN actually pull some cool tricks with a 32-bit OS like the boot.ini memory /exclude ranges trick & use them for a HUGE ramdisk if you have the software for it (ArSoft's FREE one, comes to mind here... unlimited 4gb 32-bit Win32 PE limited only!)

------------------------

ADDITIONALLY/BACKGROUND:

This also works for speed in information systems (i.e.- database work):

Ideas of mine took EEC SYstems/SuperSpeed.com to a FINALIST position @ Microsoft Tech-Ed 2000-2002, in the hardest category there: SQLServer Performance Enhancement.

The article for the wares by this company got reviewed excellently by Windows IT Pro magazine technical editor Mr. John Enck, & my research was used above his for instance (he only covered a fraction of what I had is why) while I wrote programs for improving their SuperCache product.

I also authored what WAS the front page article @ CENATEK in a review of their RocketDrive (was good enough that I bought one, & employed it above, & THAT is only a partial list too, of how I really use it & why) which for years was featured above many others from famous websites or magazines even... they knew who I was, because I used to put out a software based RamDisk + GUI front end tuner... they thought I was "competition" & I was not really.

APK

P.S.=> My SSD, an older CENATEK "RocketDrive" has backup powersupply of its own, unlike software based ramdisks, & I keep a UPS is why I don't worry here about losing data like logs...

There are F A S T E R units out here like it, & not "FLASH BASED" with their limited WRITE lifecycles, but faster RAM than mine (PC-133 SDRAM) & faster busses too (PCI 2.2 = 132mb/sec)...

Gigabyte's IRAM is a good example, with DDR2 RAM, & SATA 150,b/sec busses...

PLUS, an even a faster one's out there in prototype called the DDRDrive X1 (DDR RAM & PCI-Express bus using, but not in fullblown production afaik & I wish it WOULD be)...apk

Edited by APK
Talk about thread resurrection! Geez.

Necessary, per the points I posted above...

They're correct & do work for better performance (& security even) by using Solid-State Ramdisks/RamDrives!

Especially for the things noted on a Solid-State RamDrive unit (even paging), & for webpage caching, %temp/tmp% ops (which I did not list above), logging from the OS & its subsystems + apps logs (& even MIS/IS/IT related work)?

"UnToUcHaBLe"... &, proven so in fact by many sources as to how/why. Some of which were my own which corporate bodies employed to GOOD effect as noted above... others from other sources I have not listed.

Even Linus torvalds of LINUX fame is excited about them, per an article I read about the future direction of Linux in 2008 & computing on it... why?

Reduced latency!

Linus Torvalds gets excited by SSDs - Next biggest thing

http://www.theinquirer.net/gb/inquirer/new...ts-excited-ssds

I'd say ole' Linux Penguin #1 pretty much knows his stuff, despite the fact I like Windows better.

I've been saying it for years, for more than a decade in publication & online... & it works, per MS Tech Ed stuff I note above for one, & from research I've done & others have as well that is in publication or practical use (not just for IS/IT/MIS stuff, but for home users too)!

These things ARE the future of I/O, especially high-performance io. However, I hope they don't mean FLASH based ones... the limited write lifecycle is NOT as good as SDRAM/DDR/DDR2 RAM based TRUE SSD's have, by far.

APK

P.S.=> I missed posting about %temp/tmp% environmental variables placements there as well in my first post here on this toipic (another one is placing your %COMSPEC% command interpreters there, like cmd.exe for example)... it's/they're other/another one(s) that is/are outstanding & useful for performance' sake...

PLUS, lastely... these SS disks? They defrag UNBELIEVABLY FAST as well... which works out for webpage caches the best imo, as well as logging... even if the defragger's set to "aggressively consolidate freespace", especially in comparison to std. mechanical HDD's...

I also run my browsers & email apps (plus occasionally others, for "debugging purposes" & tracing) thru a great little idea for an app, especially online for security, & that is SANDBOXIE... & guess where I place its "sandbox"?

You guessed it: My Solid state Ramdisk... faster performance is why, all the way around... apk

Edited by APK
You really think it's necessary to resurrect a more than two year old thread? :|

Well, when it contains information that is "off"? Sure, why not - misinformation (even if due to stale information) is still worth correcting for.

(Don't you agree?)

APK

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • Lethal fake phone chargers are still being sold on Amazon and eBay, UK watchdog warns by Paul Hill Credit: Pexels The UK consumer rights organization, Which?, is claiming that “potentially lethal knock-off chargers” are still being sold on online marketplaces seven years after it exposed the danger of these chargers. In its latest investigation, it bought 15 USB phone chargers from several online marketplaces and found they were missing key information, meaning they cannot be legally sold in the UK. Which? bought the 15 chargers from seven online marketplaces. These were Amazon (including Amazon Haul), AliExpress, B&Q Marketplace, Debenhams Marketplace, and eBay. It said that the chargers were so badly made that anyone using them was at risk of electric shock. Over half the chargers also posed fire and explosion risks. Of the chargers purchased, one was a fake Apple USB-C 35W power adaptor charger. To confuse buyers, the box was branded with an Apple logo, but testing found it to be a fake. Further testing picked up arcing sounds after 10 seconds of use, where a current jumps between two parts of the electrical circuit, which can cause fires, explosions, or electric shock. The manufacturers of this particular charger also put modeling clay inside it to make it feel more weighty, robust, and genuine. Not all of the chargers were technically faulty; however, some were missing key packaging, markings, and documentation, meaning they can’t be sold in the UK legally. Which? said that it is now campaigning alongside a coalition of safety groups and businesses for new laws that make online marketplaces responsible for ensuring the safety of products that they choose to list on their websites. It also said the government needs to start using powers under the Product Regulation and Metrology Act, which was adopted last July, to impose safety requirements on online marketplaces via secondary legislation, but so far, there have been delays. No matter what country you are in, be sure to properly research what you are buying and only buy authentic chargers to prevent fires. You can read more about Which?’s research here.
    • Visual Studio finally gets long-awaited feature that developers will love by Usama Jawad Visual Studio Code is Microsoft's popular, lightweight, open-source code editor, it is actually Visual Studio that is the company's flagship integrated development environment (IDE). Although the IDE already offers a boatload of useful features for developers, Microsoft has finally introduced a long-requested capability that will be loved by many. While developers have already been able to create Git pull requests (PRs) directly within Visual Studio for the past couple of years, it had not been possible to review a PR without switching to the browser, until now. Microsoft revealed in December 2025 that it is working on UX that enables developers to do just that, and fast-forward to June 2026, and Visual Studio finally has native capabilities to open and inspect a PR, discuss feedback, and wrap up the review, all without switching to the browser. This integration works for both GitHub and Azure DevOps (including on-prem). Developers have access to multiple surfaces to open a PR, including Git Repository, Git Changes, and the Git menu in Visual Studio. Once you open a PR, all the important details will be immediately visible to you, from where you can navigate to various levels of granularity and branch states, depending on the reviews that you are engaged in. As you would expect, you also get a diff view that enables you to see code changes inline or side-by-side in a separate panel. You can also review commit-by-commit. Additionally, this UX fosters collaboration as you can leave comments, reply to threads, and resolve conversations easily. Naturally, you can also leverage Copilot to apply a code suggestion to fix a potential issue. When you are done, you have the ability to approve, complete, and merge the PR. This is a pretty major feature as it has been requested heavily for the past few years. You can try it out in Visual Studio 2026 version 18.7, made available here recently. Microsoft plans to enhance this experience further in future releases with comment filtering, a timeline of PR activity, and more.
    • This AdGuard Family lifetime deal is still only $15.97 by Steven Parker Today's highlighted Neowin Deal comes via our Apps + Software section, where you can get a lifetime subscription and save 90% on a lifetime AdGuard Family Plan. AdGuard is a unique program that has all the necessary features for what they claim to be "the best web experience." The software combines the an advanced ad blocker, a privacy protection module, and a parental control tool—all working in one app. This software deals with annoying ads, hides your data from a multitude of trackers, protects you from malware attacks, and even lets you restrict your kids from accessing inappropriate content. Install AdGuard and see the internet as it was supposed to be: clean and safe. Get rid of annoying banners, pop-ups & video ads once and for all Hide your data from the multitude of trackers & activity analyzers that swarm the web Avoid fraudulent and phishing website and malware attacks Protect your kids online by restricting them from accessing inappropriate & adult content Good to know Family Plan Length of access: lifetime This plan is only available to new users Redemption deadline: redeem your code within 30 days of purchase Max number of devices: 9 Access options: desktop & mobile Software version: AdGuard Family Updates included A lifetime subscription of AdGuard Family Plan normally costs $169.99, but this deal can be yours for just $15.97, that's a saving of $157.02. For full terms, specifications, and license info please click the link below. Get this AdGuard Family lifetime deal for just $15.97 (was $169.99) Although priced in U.S. dollars, this deal is available for digital purchase worldwide. As an online publication, Neowin too relies on ads for operating costs and, if you use an ad blocker, we'd appreciate being whitelisted. In addition, we have an ad-free subscription for $28 a year, which is another way to show support! Support queries If you have queries or need support for any of the Neowin Deals, please use the contact form here. Neowin Deals are managed and sold by StackCommerce who represent Neowin on an affiliate basis. Why we post these deals We post these because we earn commission on each sale so as not to rely solely on advertising, which many of our readers block. It all helps toward paying staff reporters, servers and hosting costs. So for those that keep moaning and complaining, be thankful we're still online for you to even do that. Other ways to support Neowin Whitelist Neowin by not blocking our ads Create a free member account to see fewer ads Make a donation to support our day to day running costs Subscribe to Neowin - for $14 a year, or $28 a year for an ad-free experience Disclosure: Neowin benefits from revenue of each sale made through our branded deals site powered by StackCommerce.
    • the MCT currently downloads 26200.8653, so not completely up to date.
  • Recent Achievements

    • Week One Done
      Timaximus earned a badge
      Week One Done
    • One Month Later
      Timaximus earned a badge
      One Month Later
    • Rookie
      FBSPL went up a rank
      Rookie
    • First Post
      davidbazooked earned a badge
      First Post
    • Week One Done
      davidbazooked earned a badge
      Week One Done
  • Popular Contributors

    1. 1
      +primortal
      507
    2. 2
      PsYcHoKiLLa
      179
    3. 3
      +Edouard
      160
    4. 4
      Steven P.
      83
    5. 5
      ATLien_0
      75
  • Tell a friend

    Love Neowin? Tell a friend!