Post Reply 
Proxomitron reduces RWIN to 32768
Nov. 28, 2005, 07:04 AM
Post: #106
 
I changed the recommended byte from 04 to 00 and the speed test still lists the RWIN as 32768, even after rebooting the computer.
Add Thank You Quote this message in a reply
Nov. 28, 2005, 01:53 PM
Post: #107
 
Siamesecat Wrote:I changed the recommended byte from 04 to 00 and the speed test still lists the RWIN as 32768, even after rebooting the computer.

There are a several possibilities.
Windows is set to use 32768. At least one person has already seen this.
Something else is sending 32768, http://www.speedguide.net/faq_in_q.php?c...=98&qid=23 .
You missed or forgot to save the change. ;-)

HTH,
--
JJoe
Add Thank You Quote this message in a reply
Nov. 28, 2005, 02:22 PM
Post: #108
 
JJoe Wrote:
Peakaboo Wrote:Having no experience with this it took me a few trials to get it but I finally did. The hard part was finding the corresponding byte in the version of proxo I was using 4.3. Finally found it I think @ c101.

It might be my imagination but after making the hexedit change, and firing up the edited version, web pages seem to be loading quicker - further testing on this... Smile!

Finally, I agree with you on duty of those who alter the program - to make a notation and to not distribute the altered exe. This is the least we can do in memory of SRL.

Umm. I appreciate the enthusiasm but you might want to change that back. ;-)

I think that part of 4.3 has something to do with text display.
I'm not sure what you are looking for is in 4.3.

Take a look at 4.5 june instead.
Starting at d57c you should see
00 80 00 00 6a 04
the 04 is what you would change to 00
00 80 00 00 is the 32768 part.
You can find the same string in the Proxopatchered 4.5 EXEs.

Changing the value is easy.
Adding the disclaimer is a little more... adventurous.
But a big part of Proxomitron is learning.

Have fun,
--
JJoe

Thanks JJoe,

it is fun to learn.

I appreciate your feedback.

What I changed was this:

0000C0FB E400 in al,0
0000C0FD 800000 add b.[bx+si],0 ; ' '
0000C100 6A00 push w.0 ; ' '
0000C102 8D55E4 lea dx,[di+0E4]

I changed c101 from 04 to 00
c101 in 4.3 looks awfully similar to d581 in 45j and the surrounding code to me.

Guess I'll run Kye-U's patch and see if it changes something different in 4.3.
Add Thank You Quote this message in a reply
Nov. 28, 2005, 03:30 PM
Post: #109
 
Peakaboo Wrote:Thanks JJoe,

it is fun to learn.

I appreciate your feedback.

What I changed was this:

0000C0FB E400 in al,0
0000C0FD 800000 add b.[bx+si],0 ; ' '
0000C100 6A00 push w.0 ; ' '
0000C102 8D55E4 lea dx,[di+0E4]

I changed c101 from 04 to 00
c101 in 4.3 looks awfully similar to d581 in 45j and the surrounding code to me.

Guess I'll run Kye-U's patch and see if it changes something different in 4.3.

My mistake.
I opened the wrong exe, duh.
I don't even have 4.3 on this machine.
I do have ProxN3, tho.
Hey! There's a 3 in there.
It was a long day.

Looks good but I've really only looked at 4.5.

Sorry,
--
JJoe
Add Thank You Quote this message in a reply
Nov. 28, 2005, 05:00 PM
Post: #110
 
Kye-U Wrote:JJoe and z12, let me know if there's anything you want me to add/change/remove in the patcher.
Well I have many of the same objections I have with the Japanese patches and had with Henk's first efforts.
Which isn't surprising considering
http://groups.yahoo.com/group/prox-list/message/16469
http://groups.yahoo.com/group/prox-list/message/16471

Looks like your patch doesn't add a disclaimer or even change the date stamp.
To download the patch from UOPF, they had to agree to terms but people will share the patch.
If their memory is good, they'll remember they patched the exe but people *do* distribute patched Proxomitron EXEs.

I think people who patch programs *or* write programs that patch programs have a responsibility to protect the authors by adding a disclaimer to the Program. Some people insisted that an "acceptable" patch be available *before* the review was posted. Some in the group understood but did not agree.
So I realize that others don't feel the way I do.

Seems to me that there are some questions to consider:
Does your patch enable a behaviour that you do not approve of?
Should you care?
What would be your opinion of your patch, if Proxomitron was your Program?
What would be your opinion of your patch, if you were Henk?
Are *you* sure the alteration made is OK?
Who are these people that discovered and reviewed this?
Why aren't all their names or nicks made available?
How many were there?
What are their qualifications to do such things?
http://www.google.com/search?q=Proxomitron+RWIN yet nobody posted the link?
Does such a patch belong at UOPF?
Does having it at UOPF imply legitimacy?

And as I said before:
Not that it is any of my business,
--
JJoe
Add Thank You Quote this message in a reply
Nov. 28, 2005, 09:30 PM
Post: #111
 
JJoe, I agree with all your points, and overall, I agree with your train of thought.

I will take the patch off the Internet as I do not have the time right now to add a little disclaimer, and all credits.

It's very difficult, keeping both sides happy. Perhaps it's better if individuals were able to hex-edit Proxomitron themselves to avoid all this trouble.

Maybe I'll just take it off for good? I'll probably leave it up to you and z12 to deciding on how to release the patch, and what the patch should do, since you guys have more information than I do, and I do mean this. The link to the patch generator is:

http://digilander.libero.it/zeusoft/zpmaker.htm

Many thanks to you and z12 for all your research, and to our Japanese Proxomitron-enthusiasts for discovering this and finding a solution.
Visit this user's website
Add Thank You Quote this message in a reply
Nov. 29, 2005, 08:02 AM
Post: #112
 
Quote:Something else is sending 32768, http://www.speedguide.net/faq_in_q.php?c...=98&qid=23 .
The speed test page makes it sound as if one could make a change in the RWIN manually. How would I do that? I was unable to find RWIN in the registry.
Add Thank You Quote this message in a reply
Nov. 29, 2005, 04:33 PM
Post: #113
 
Siamesecat Wrote:
Quote:Something else is sending 32768, http://www.speedguide.net/faq_in_q.php?c...=98&qid=23 .
The speed test page makes it sound as if one could make a change in the RWIN manually. How would I do that? I was unable to find RWIN in the registry.
Assuming Windows has control you should be able to change RWIN.
You may have to add a value to the registry, however.
http://www.dslreports.com/faq/tweaks/5.+RWIN#911

There are utilities like Dr. TCP and TCP Optimizer that can do this for you.
http://www.dslreports.com/drtcp
http://www.dslreports.com/faq/578
http://www.speedguide.net/downloads.php

I have some experience with Dr. TCP on Win98 but have not used TCP Optimizer.

HTH,
--
JJoe
Add Thank You Quote this message in a reply
Nov. 29, 2005, 08:14 PM
Post: #114
 
JJoe,
Thanks for the info. I have TCP Optimizer, so I tried the optimal setting. It made no difference. I wonder what is limiting the RWIN? I did save the Prox after changing the byte. My download speed is pretty fast anyway, so it is no big deal, but that is bugging me a bit.
Add Thank You Quote this message in a reply
Nov. 30, 2005, 04:49 AM
Post: #115
 
Siamesecat Wrote:JJoe,
Thanks for the info. I have TCP Optimizer, so I tried the optimal setting. It made no difference. I wonder what is limiting the RWIN? I did save the Prox after changing the byte.
Well lets start at the start.
Without Proxomitron, RWIN is what?

If RWIN without Proxomitron is 32768, what all is between your browser and speedguide?
Is the IP shown at speedguide your computer's?
Mark has reported that his ISP is setting his RWIN.
http://groups.yahoo.com/group/prox-list/message/22407
Modems, routers, another proxy... would be suspects.

Siamesecat Wrote:My download speed is pretty fast anyway, so it is no big deal,
That is the part that people will overlook.

Siamesecat Wrote:but that is bugging me a bit.
But I can understand that.

--
JJoe
Add Thank You Quote this message in a reply
Nov. 30, 2005, 05:12 AM
Post: #116
 
Kye-U Wrote:It's very difficult, keeping both sides happy.

Don't I know it. ;-)
Not that I'm upset with anybody. :-)

As far as I know we only really looked at ProxN45.
So it made sense to limit the patches to 4.5.
Fortunately, PatchMaker patches require the original EXE.
PatchMaker is also the app suggested by Scott.
http://www.clickteam.com/English/patch_maker.htm

Unfortunately, ProxoPatchered EXE's make the patch creator wonder if ?he is doing the right thing.
A PatchMaker patch can't patch the latest ProxoPatchered EXE.
So a new patch that contained Henk's changes would be needed for the original exe...
http://www.xs4all.nl/~vsetten/prox/nomis.html Wrote:No, these patchers do not change any of the actual program code. They are not a form of hacking, which indeed would be not permissible. All they do is change some of the interface resources that are called by the actual program. The latter is not modified in any way.
Also, people would no longer visit http://www.xs4all.nl/~vsetten/prox/ to get the patch.
It feels like Henk should get a chance to protest first but he hasn't answered mail. :-(

Kye-U Wrote:Perhaps it's better if individuals were able to hex-edit Proxomitron themselves to avoid all this trouble.

I like that idea. One of the great things about Proxomitron is the way it teaches.
Once you learn that little bit you needed to know to do that thing you wanted to do,
you may see that you need to know just a little more and a little more and a...

The Hex Editor should take care of the Date Stamp.
If they want to do "just a little more", ;-)
Resource Hacker isn't that tough.

http://www.angusj.com/resourcehacker/
Open the altered EXE with Resource Hacker.
Proxomitron's "About" screen is in folder 147 of Dialogs.
Drag the "Dialog - 147" window up, so you can work on it.
Grab the bottom of the "Dialog - 147" window and drag it way down.
Right-click in the open field you just created and select "Insert Control"
Click on "Label" in the "Control Editor".
Add your text where it says "Caption".
--"The value at 0xd581 has been changed to 00. It was 04." maybe?
Click OK.
Now make it pretty.
--Arrange things so you see all sides of the new "Control".
--Grab the little squares of the new Control to stretch it.
--Move the Control to its new home.
--Grab the bottom of the Dialog and move it back up some.
Click "Compile Script".
Save your work.

Think that's about it, anyway.

Kye-U Wrote:Maybe I'll just take it off for good? I'll probably leave it up to you and z12 to deciding on how to release the patch, and what the patch should do, since you guys have more information than I do, and I do mean this. The link to the patch generator is:

http://digilander.libero.it/zeusoft/zpmaker.htm

Whether or not to host a patch at UOPF is your call, of course. It just seems to me that, for now, having it off-site is best. People may question why they have to get the file elsewhere and then keep an eye on things. I find it unlikely any named person will ever "guarantee" the hack but with time we may be able to say it has been tested and deemed safe. count_b has written that he knows of no problems. FWIW, we may consider that this hack has a year of experience behind it. Even then, an ok hack is still a hack.

It looks to me like pmaker can be of use but we should limit patches to ProxN4.5, at least until somebody has a chance to look at code.

Which EXEs have this RWIN problem?

Kye-U Wrote:Many thanks to you and z12 for all your research, and to our Japanese Proxomitron-enthusiasts for discovering this and finding a solution.

People who don't use Proxomitron helped.
People who are unknown to me helped.
I thank them all but we still don't know the whole story.

--
JJoe
Add Thank You Quote this message in a reply
Nov. 30, 2005, 07:19 AM
Post: #117
 
Quote:Without Proxomitron, RWIN is what?
Without Prox, the RWIN is 256960. Are there any other bytes in Prox which should be changed (and whose offset is not affected by Proxopatcher or Resource Hacker)?
Add Thank You Quote this message in a reply
Nov. 30, 2005, 10:40 AM
Post: #118
 
Kye-U Wrote:I will take the patch off the Internet as I do not have the time right now to add a little disclaimer, and all credits.

Bum! I tried downloading it last night and it had already gone. Still stuck with a restricted RWIN. :-( And no, I've never used a hex editor.

Kevin
Quote this message in a reply
Nov. 30, 2005, 12:45 PM
Post: #119
 
Kevin,

I hadn't used a hex editor before, but I found it dead easy. I used xvid32 which is free.

Bill.
Add Thank You Quote this message in a reply
Nov. 30, 2005, 03:31 PM
Post: #120
 
Siamesecat Wrote:
Quote:Without Proxomitron, RWIN is what?
Without Prox, the RWIN is 256960. Are there any other bytes in Prox which should be changed (and whose offset is not affected by Proxopatcher or Resource Hacker)?
Changing the one byte has done it so far...

Is Proxomitron using another Proxy?

(I've actually done what I'm about to ask, so don't take it the wrong way. ;-)
Are you sure the exe you hacked is the one you are using?

Use the patch at rapidshare to hack an original exe and see if that's ok.

Let us know what happens,
--
JJoe
Add Thank You Quote this message in a reply
Post Reply 


Forum Jump: