TCP Optimizer v3.0.3


Recommended Posts

I'm curious too is this even useful on Windows 7, I kinda doubted it.

But, looking at that page, the Vista/7 patch below it says it "sets both netsh auto tunning levels", I'm assuming the optimizer does the same now, so, maybe there can be some improvement done by it? .... Waits for others to test it first.

I'm curious too is this even useful on Windows 7, I kinda doubted it.

But, looking at that page, the Vista/7 patch below it says it "sets both netsh auto tunning levels", I'm assuming the optimizer does the same now, so, maybe there can be some improvement done by it? .... Waits for others to test it first.

I'm using it.

So far so good, at least it didn't slow my speed.

Feels snappier a more consistent speed maybe on my PC only or perhaps placebo :p

Anyway I don't think its harmful.

It generates a backup of your previous setting so you can roll-back.

If you want you can even set a System Restore checkpoint before trying it out.

Most of the stuff it tweaks for 7 is described here:

http://www.speedguide.net/read_articles.php?id=2574

So far used it on my desktop and laptop.

There really isn't any need for this anymore with Vista and 7. Both those OSes automatically change TCP settings based on your usage.

For some games that use TCP, both Vista and Windows 7 still send TCP ack's every 2 packages so several games can drastically reduce ping times by changing that to every package. It can lower ping (or what's show as server ms reply times) times by half or even more.

But you don't need an application to do that, it's easy to change with a registry modification. Or a small script that is dedicated for only that and nothing else.

For some games that use TCP, both Vista and Windows 7 still send TCP ack's every 2 packages so several games can drastically reduce ping times by changing that to every package. It can lower ping (or what's show as server ms reply times) times by half or even more.

But you don't need an application to do that, it's easy to change with a registry modification. Or a small script that is dedicated for only that and nothing else.

Sure if you're on like 56k then you might notice a difference. But I'm still going to stick with the difference will be so marginal that you wouldn't even know its there, or there will be no difference whatsoever. Drastic reduction? Highly doubt it.

The entire CTCP stack on Vista / 7 is self-tuning, ie it configures itself automatically when its needed, and as a result pretty much any registry edits is in 99.9999% of the time, detrimental.

I installed this for testing purposes and my speed went up almost 2MB.

I have a 19MBPS connection but due to the quality of the line in my street, I could only get around 14, now its over 16, so I guess it works for me in my circumstance!

Edit: I moved the slider to 19 btw.

For some games that use TCP, both Vista and Windows 7 still send TCP ack's every 2 packages so several games can drastically reduce ping times by changing that to every package. It can lower ping (or what's show as server ms reply times) times by half or even more.

But you don't need an application to do that, it's easy to change with a registry modification. Or a small script that is dedicated for only that and nothing else.

But changing the setting so that Windows sends an ACK response for every packet would actually leave less "room" for the game data, slowing it down.

It's set to 2 by default for better bandwidth allocation, lowering it increases the amount of packets sent.

I installed this for testing purposes and my speed went up almost 2MB.

I have a 19MBPS connection but due to the quality of the line in my street, I could only get around 14, now its over 16, so I guess it works for me in my circumstance!

Edit: I moved the slider to 19 btw.

I take it you tested before, then made the changes and tested after. Your not just compairing to an old test?

Sure if you're on like 56k then you might notice a difference. But I'm still going to stick with the difference will be so marginal that you wouldn't even know its there, or there will be no difference whatsoever. Drastic reduction? Highly doubt it.

It does work, and it does it on ADSL and better connections even. You can check the feedback on a script that is made as a WoW mod, but works for other games that use TCP.

http://www.wowinterface.com/downloads/info13581-LeatrixLatencyFix.html#changelog

But changing the setting so that Windows sends an ACK response for every packet would actually leave less "room" for the game data, slowing it down. It's set to 2 by default for better bandwidth allocation, lowering it increases the amount of packets sent.

That would be a problem if it wasn't a game, which I was talking about, where you want faster response time not better bandwidth.

And as I said, to change this, you don't need the application, it's just one parameter. The script above does it a bit easier for the user.

I would like to add that in general, I would never fiddle with the default settings as the program above does. I just changed TCP Ack frequency, because it improved my MMO experience (in 2 games that I ran back then). Currently, I'm not playing any games that would improve due to this change, so at the moment I don't need to change anything from default.

...

That would be a problem if it wasn't a game, which I was talking about, where you want faster response time not better bandwidth.

And as I said, to change this, you don't need the application, it's just one parameter. The script above does it a bit easier for the user.

Except the actual response time is the same, it's just talking to the remote system more often.

It does, and it does it on ADSL and better connections even. You can check the feedback on a script that is made as a WoW mod, but works for other games that use TCP.

http://www.wowinterf....html#changelog

Ah. How about you actually research that script before talking about it huh? Well since you actually have no knowledge on how Windows network stack works, and what the ACK frequency actually does, and like talking about placebo effects, I'm going to simply quote a blues post on the very same "feature":

No he's right, you do not lower the amount of ACKs, you increase it.

Packets will still take exactly the same amount of time between you or your sever, the only real effect changing this registry entry is the latency calculations will think there's lower latency than there actually is.

Your computer only THINKS the response time is better since its talking to the WoW servers more often. HOWEVER the lag is still there, since you haven't magically increased your bandwidth / speed. Your actual response time / ping, as a result, is still EXACTLY the same.

Of a random sidenote, you'll notice that the recommended values for ACK INCREASE as your speed increases:

1 GigaBit: TcpAckFrequency = 13 (RWIN = 64 KByte)

100 MegaBit: TcpAckFrequency = 5 (RWIN = 17 KByte)

10 MegaBit: TcpAckFrequency = 2 (RWIN = 8 KByte)

Ah. How about you actually research that script before talking about it huh? Well since you actually have no knowledge on how Windows network stack works, and what the ACK frequency actually does, and like talking about placebo effects, I'm going to simply quote a blues post on the very same "feature":

Ah, a comeback, you must have been busy searching for info after your first post, as you didn't seem to know this info then even though it was clear what it was about. Yet you still only post a quote from a "blue" without any technical information. Who is this Blizzard support guy and what is his technical knowledge of TCP? You are just assuming a Blue post means it's good. For sure Blizzard don't want people to change this is because it does increase traffic to servers and possible congestion.

Granted I'm not an expert, but I've tested it and it seemed to work fine. If it was a placebo effect, I can accept that, if I get a real and detailed technical explanation from a person that is an expert on these matter on why it does not work. Your responses mean nothing, as you probably don't know anything about it.

Ah, a comeback, you must have been busy searching for info after your first post, as you didn't seem to know this info then even though it was clear what it was about. Yet you still only post a quote from a "blue" without any technical information. Who is this Blizzard support guy and what is his technical knowledge of TCP? You are just assuming a Blue post means it's good. For sure Blizzard don't want people to change this is because it does increase traffic to servers and possible congestion.

Actually I went and looked through the source code of your amazing "mod" (it's simply in VB and you can open it in Notepad) did a very quick google search to see what blizzard had to say on the matter and found that.

And sorry that I don't sit here waiting for you to reply.

Granted I'm not an expert, but I've tested it and it seemed to work fine. If it was a placebo effect, I can accept that, if I get a real and detailed technical explanation from a person that is an expert on these matter on why it does not work. Your responses mean nothing, as you probably don't know anything about it.

Go check MSDN then.

I mean just logically think about it for a second. If such a small registry edit can yield, as you claim, a massive performance gain. Why does blizzard spends hours and hours on enhancements to WoW's networking when they can just change a registry value? I mean considering their entire game requires a constant connection to the server, and that a lower response time means better gameplay. Wouldn't blizzard want that? Or are they just happy with people suffering with high pings, while their developers laugh all day.

Why do other game makers spend hours on it too? I mean sure it may affect stuff out of the game, but it isn't hard to remove a registry edit and that can simply be coded in to be done when the game exits.

Hell, why did Microsoft do major changes to Windows TCP stack from XP to Vista? I guess all their experience, knowledge and ability comes to nothing. Because they still haven't figured out that changing a value from 2 to 1 yields "massive benefits"

What are you going to tell me next? That increasing the number of half-opened connections in XP magically increases your internet speed as well? I mean I heard tons of people back in the XP days (and even some today) who seem to believe that...

Someone tell me how this will actually speed up things. A standard Windows install here seems perfectly capable of maxing out both my LAN and my 10Mbit internet connection. If it couldn't at least do the latter, then it would be ridiculous to the point where Windows is a product so defective that it should be banned from the market for being ****.

I doubt this does anything for ADSL, in the UK at least. BT only give out speed increases per 0.5mbit here for every sync range you're in. e.g. for a set range of 4544 - 5119 and I sync 4600 it would be 4.0mbps. If I sync 5100 it would still be 4.0mbps, it won't jump to the next increment unless I sync 5120. And seeing this is all done on the router, TCP Optimizer wouldn't do much on Windows as I already get the full 477-480kb/sec in my downloads

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

    • No registered users viewing this page.
  • Posts

    • For the purpose that it was built for, it’s a great machine. It’s okay to own multiple machines, it’s okay for machines to be different. If every computer was the same, they’d be boring af.
    • OpenAI is rolling out a major upgrade to ChatGPT memory by Pradeep Viswanathan OpenAI is rolling out a major upgrade to ChatGPT's memory, making the system more capable, current, and scalable across long-term use. Memory allows ChatGPT to remember useful details about users, including their preferences, projects, and constraints. Instead of starting every conversation from scratch, ChatGPT can use this context to provide more relevant responses in future chats. OpenAI first launched saved memories in February 2024. That feature allowed users to explicitly ask ChatGPT to save information into its memory, such as travel plans or writing preferences. However, this system had limits because it depended heavily on users giving clear instructions to remember something. Additionally, saved memories could become stale over time. In April 2025, OpenAI expanded memory by allowing ChatGPT to reference past chat context outside the saved memories list. This was powered by a background process called “dreaming,” which automatically curates memories from chat history. This made ChatGPT better at learning from natural conversation without requiring users to manually save every detail. Today, OpenAI announced a more capable and compute-efficient memory architecture built on top of dreaming. This new system improves ChatGPT’s ability to carry forward useful context, follow user preferences, and remain accurate as time passes. According to OpenAI’s internal evaluations, the new system improves factual recall from 67.9% in 2025 to 82.8% in 2026. Preference adherence improves from 55.3% to 71.3%, while accuracy over time improves from 52.2% to 75.1%. The best part of this new system is a new memory summary page where users can review ChatGPT's memories. Users can even update details, correct information, or give instructions on what topics ChatGPT should bring up and when. This new, improved memory system is available to ChatGPT Plus and Pro users in the US starting today. It will roll out to more countries, as well as Free and Go users, in the coming weeks.
    • I work for a video production company in Australia. The camera operators shoot footage and then pass the SD card over to the editors. Much easier than handing over the entire camera. Plus, on a busy day you can hand off the SD card and then pop another in for the next shoot. Or, you might have used multiple SD cards because you need the extra space for a long shoot. I also use USB cables and wifi for transferring footage, but in many cases an SD card reader is the easiest method.
  • Recent Achievements

    • Week One Done
      Dr Jared Dental Studio earned a badge
      Week One Done
    • Week One Done
      RG INVESTMENT GROUP earned a badge
      Week One Done
    • Very Popular
      The Norwegian Drone Pilot earned a badge
      Very Popular
    • Very Popular
      s0nic69 earned a badge
      Very Popular
    • Collaborator
      Asgardi earned a badge
      Collaborator
  • Popular Contributors

    1. 1
      +primortal
      472
    2. 2
      PsYcHoKiLLa
      247
    3. 3
      Skyfrog
      79
    4. 4
      FloatingFatMan
      67
    5. 5
      Michael Scrip
      59
  • Tell a friend

    Love Neowin? Tell a friend!