Gecko dotNET Bugs and Requests

[dcx2]

Mathew - I did ten MEM2 searches and I had no disconnection problems.  I even tried pausing the game and doing unknown-equal searches so that way it would query every address in memory.  I'm going to leave MemView Auto-Update on all night to see what happens.

Mokuro - Until you provide me with the instructions I need to reproduce your bug, I can't help you.  And with that attitude, why should I?  No one else has problems with breakpoints, so as far as I know, the app is not what needs fixed.   ::)

[Mokuro]

Will provide you with more information regarding the issue I have with breakpoint later today. Because I still can't access the breakpoint even though I told you that I had it fixed when it's not.  :-\ I've never had problems using BP with Wiird except that Wiird crashes and freezes several times on me.

Edit: My bad. I should have asked you nicely. I do appreciate the time you spend fixing the bugs. :)

[Romaap]

Mokuro, are you sure you have set the breakpoint on the right bit-alignment?

[hetoan2]

Mokuro, are you sure you have set the breakpoint on the right bit-alignment?

Thats what i was thinking. If its a 32 bit ram write (most are), then you have to have the address end in 0, 4, 8, or C.

I'm almost positive this is just a noob error. I occasionally even click on the wrong bit when moving it over to the BP tab.

You should probably stay away from exact matches... it helps to prevent user error.

[James0x57]

Some games BP's crash the game on use, shortly after use, or on next use. I really don't think it's a GUI problem though; for me WiiRD GUI has the same result.

[wiiztec]

I've also experienced the problem of Gecko.NET lagging Super Mario Galaxy to a nearly a complete stop when applying codes, I've tried all the hooktypes and none of them fix the problem yet with WiiRd I experience no such problem, I can't remember if I tried with "pause before sending" unchecked though

[Mokuro]

Alright. I just checked BPNext box and pressed pause and pressed "set" and guess what? It WORKS.  :eek: Do I have to do this every time I want to use breakpoint ?

and thank you guys for all your efforts. Keep it up.  :cool:

[dcx2]

wiiztec - the original SMG?  No problems with SMG2?  What codes are you using?

Pause While Sending was designed to overcome a problem where repeated application of C2 codes can cause games to crash.  This is a symptom I first observed with WiiRDGUI and Ghost Squad, though I have seen it in other games too, like Okami.  All it does is pause; send cheats; unpause.  In theory you could do the same thing with WiiRDGUI (and in fact that is what I had to do before I started working on Gecko.NET)

I don't see why Pause While Sending would cause any slowdowns...

---

mokuro - WiiRDGUI's next button isn't very reliable; it pauses the game, waits for a few milliseconds, and then unpauses the game.  Sometimes, you will see two frames go by after only a single click.  This was a pain in the ass when trying to find timers using different-by-1 searches, so I had to find some way to advance the game one frame at a time.

Before Gecko.NET, this involved setting an Execute Breakpoint on 800018A8 (the first instruction in the code handler, which *should* execute once per frame).  This led to a lot of switching back and forth between the Breakpoint tab and the Code Search tab.

Gecko.NET's BPNext streamlines this process.  It makes the Next button set an Execute Breakpoint on 800018A8.  That's why I asked you to try BPNext; because if you can use BPNext then there is no problem with Breakpoints.

BPNext should have nothing to do with the actual act of setting a breakpoint.  You need to provide information.  Are you setting a read, write, read/write, or execute breakpoint?  On what address?  Is "Exact" checked?  Do you have any Breakpoint Conditions?

[Mokuro]

Never mind. I'm the one who needs to be fixed. I didn't know the real purpose of the breakpoint function until yesterday when I hacked some flash games on FaceBook using Cheat Engine. The BP in Gecko.Net is working like a charm. I no longer need to check the BPNext box anymore. dcx , thank you so much for your support for this great app. Keep it up.

:D

[dcx2]

Glad to hear that your issue has been resolved.  Happy hacking!

In case you aren't aware of how the Step buttons work...

Step Into always go to the next instruction, and this includes going "into" a function call (bl/bctrl)

Step Over will usually go to the next instruction, except for function calls; it will instead go "over" the function call to the instruction after it, allowing the function to do all of its work without forcing you to Step through all of it.

Step Out will repeatedly Step Over until it gets to the blr; it goes "out" of the function to the caller.  I plan on working on an update to Step Out that will hopefully avoid the "Step Over repeatedly" by analyzing the stack frame for the LR Save Word, or using the value in the LR directly if we're in a leaf function that has not created a stack frame.

[wiiztec]

I don't see why Pause While Sending would cause any slowdowns...

That's not what you said before

This is weird, because I've hacked both Galaxies with this app...Do you have Pause While Sending checked?  That should be the only thing that can cause that sort of issue...

wiiztec - the original SMG?  No problems with SMG2?  What codes are you using?

Pause While Sending was designed to overcome a problem where repeated application of C2 codes can cause games to crash.  This is a symptom I first observed with WiiRDGUI and Ghost Squad, though I have seen it in other games too, like Okami.  All it does is pause; send cheats; unpause.  In theory you could do the same thing with WiiRDGUI (and in fact that is what I had to do before I started working on Gecko.NET)

Yes Super Mario Galaxy not Super Mario Galaxy 2 i'm never vague about what I'm talking about, It doesn't matter what codes I use it's pressing the send to game button that causes it, I am aware of the purpose of the pause before sending checkbox, back when I used WiiRd I would usually pause before sending codes or ASM instructions.

[dcx2]

The reason I said it could be Pause While Sending is because that should be the only difference between how WiiRDGUI and Gecko.NET send cheats.  Several weeks ago, I noticed there was a difference in how they create the cheat stream, but I fixed that.

Have you tried sending codes with no codes checked?

[dcx2]

I moved your post, Panda.

That's weird because I know I tried testing it out to make sure it works...check the GDNdebug.log that's in the same folder as the app.  Do you see any exception code text at the bottom?

What type of search are you doing?  Unknown, specific?  32-, 16-, 8-bit, or Single?

Are you sorting search columns?  r87 had a bug where the search MUST be sorted by address ascending (the default); any other sort will cause a crash...but that's when a search begins.

r88 is just released, that might fix your problem.  I fixed the sorting bug, and added search comparison groups.

[dcx2]

Not to be rude, Panda...but please quit posting bug reports in the release thread.  The release thread is meant to be for detailed explanations of the new features that are being added, and the occasional encouraging reply of "this feature rocks".

Also, please provide more details so that I can replicate your bug.

[Panda On Smack]

yeah, misunderstanding there dcx, browser went a bit funny and it didn't show my post so I resubmitted.

anyway, here is the error when I debug with VS:



I'm on Win 7 64 if that helps.

I have fixed it by downloading Ionic.Zip.Reduced.dll and putting it in my Gecko dNet directory.