Mantis - ISSUES

0000180: Issue not fully resolved with code list (version 5.5 to 5.6)

This existed in Cheat Engine 5.5 and was only somewhat fixed in version 5.6.

For version 5.5:
Click on the address list.
Press return.
Error shows, I assume it's trying to edit a value for something in the list when there's nothing there/nothing is selected.

For version 5.6:
Attach to any process.
Add any address.
Delete the address.
Press return, error shows, even if you click off and back.

0000174: CE 5.6: Changed addresses are not displayed correctly

I highlight this code in the disassemble window:
0082DBC7 - d8 46 18 - fadd dword ptr [esi+18]
Then, I right-click and choose "Find out what addresses this instruction accesses".
The Changed Addresses window appears.
I got back to the game, force and update, then I go back to CE.
The address is wrong (only the offset is taken into account, the register's value is ignored) therefore only question marks are displayed.
See attached screen-shot what it looks like.
NOTE: ESI is not 0 - it is: 1525DE50, so the address shown should have been 1525DE68.

0000179: Multiple Level Pointer

When manually adding a pointer when you add a new level to the pointer the first level is taken away. the offset is moved up but the address just goes away. Can reproduce this everytime

0000170: CE 5.6: Floating point values are all zero

Right-clicked on an address in the list (main CE window) and selected "Find out what accesses this address".
The appropriate dialog comes up, ALT+TAB to game, trigger update, ALT+TAB to CE and the list is populated.
This is the code I select:
004f5d2f - d9 40 2c - fld dword ptr [eax+2c]
Then I press the "More information" button.
In the next dialog, I click the little right arrow to check the floating point registers. Well, everything is zero (not updated?) - so I still can't check the FPU registers.
Thank you for fixing.

0000173: CE 5.6: Memory dissect text not visible

I am looking at the memory dissect window.
Mostly it's ok.
However. When I click on an offset, the entire row will be highlighted with dark blue, and thus the Value becomes nigh invisible.
I'd suggest light blue for highlighting the row, or, inverting the text colours for the Values as well (just like the Address colour is inverted).

0000169: CE 5.6: Insert element at address-1

In the Memory dissect window, select any address of a structure.
I pick 000A.
To insert and element, hit INS.
Note that the address of the element to be inserted is -1 ('9' in my case).
I think this is wrong as the newly inserted element should push down the highlighted address as a whole (so, the default address should be the same as the highlighted address, in my case'A') - at least CE 5.5 was working at way and I loved it about that.
For now, click ok without making any changes to the address (leave it at 4 bytes, too).
Observe your updated structure.
It's incorrect - they are not pushed forth 4 bytes as they should be.
The originally 000A offset is still 000A - and I have a new entry in the list: 0009. 0009 now overlaps with 000A because that's a 4byte integer.

0000177: CE 5.6: Cannot perform scan after pointer scan

I do a number of scans for a floating point until it's found.
Then, I do a level 5 pointer scan for that floating point value.
When I'm done, I start a new scan to look for another floating point value, only to get the following error:
Scan error:thread 1:Access violation at address 004A67DA in module 'Cheat Engine.exe'. Write of address 00000000

0000176: CE 5.6: Breakpointbug

The game hung up during a "find out what accesses this address" after a "break and trace".
I had to end the game via the task manager because I could not get back to it.
Once the CE windows was properly visible again, I saw a dialog containing this text: "breakpointbug"
What's this? What did I do wrong?
I could close CE normally, however it took quite a while for it to close itself.

0000171: CE 5.6: Stream error on Float scan

I do a new scan for a float, "unknown initial value".
I ALT+TAB to the game, wait a bit, ALT+TAB back to CE.
I do next scan, "Unchanged value".
Some 450.000.000 results.
I ALT+TAB to the game, have the value changed, ALT+TAB back to CE.
I do next scan, "Increased value".
CE hangs for a while, as it it is going to crash or something. 20-30 secs later the scan starts. I was not at the computer, because I saw it's going to take a while. When I came back, I had this error on the screen: "Scan error:thread l:Stream read error".
CE did not crash, and shows that I have about 570.000 results now. I don't know if I can trust those results though...

0000168: CE 5.6: Stop button does nothing (?)

In the Memory Viewer window I right-click a lien of code and select "Find out what addresses this instructions accesses".
The line gets a green highlight, the "Changed addresses" dialog appears with a Stop button on it and I ALT+TAB to the game.
I realize that I selected the wrong line of code, so I ALT+TAB back to CE immediately, without doing anything in the game.
I click the Stop button, but nothing happens. I keep clicking on it, nothing happens.
When I click the 'X' in the corner to close the dialog, the line of the code remains highlighted (breakpoint stays?).
I select the right code next and it's supposed to be triggered, but CE does not detect the address accessed by the code. The Stop button does not work yet again.

0000178: CE 5.6: Windows that do not store their size/position

Here's the list of such windows - pretty painful to resize and move them every time they're displayed.
- "The following opcodes accessed the selected address" (opens via context menu)
- "The following opcodes changed the selected address" (opens via context menu)
- "Trace" (opens on Break and Trace)
- "Extra info" (opens via More information button)

0000175: CE 5.6: Cannot see register values (nor copy them)

Following the bug report 0000174, I right-clicked on 00000018 and selected "Show register states" from the context menu.
This is the dialog I get (see attached screen-shot).
The dialog cannot be re-sized and CTRL+C does not work, neither (I am not sure whether it's supposed to like on the other dialogs), so I guess I am missing some data from the dialog.

0000172: CE 5.6: UI not updated when script is enabled/disabled via context menu

Add a script to the list in CE.
You can enable/disable the script using the checkbox in front of it.
This time however, use the context menu to enable/disable it.
Notice that the checkbox is not updated on the UI when you enable/disable the script via the context menu; it remains in its previous state, indicating wrong state of the script (which is enabled/disabled correctly).
Thank you for fixing.

0000167: CE 5.6: Error on next scan

I performed a series of scans for a floating point value.
I began with range, then increased, decreased, did not change, decreased, then I chose "same as first scan".
When I hit the next button, CE crashes and displays a dialog containing this text:
Scan error:thread 0:Access violation at address 004A64B6 in module 'Cheat Engine.exe'. Read of address 00000000

0000166: CE 5.6: Cannot open tables/trainers/pointerscan results

-I cannot open .ct table through the open button (throws an "unknown extension" error). Double clicking on the .ct file or autoloading it when selecting a process opens it without problem.

-I cannot open a trainer error msg:"cannot open file "[filepath]\[junk_characters_that_are_not_my_filename]""

-Pointerscan result filename corrupted. I asked to save as "C:\a\azer.ptr" I got "C:\a\??.ptr", but the scan proceeded ok. But this file cannot be reopened, even after renaming it (to "c:\a\a.ptr"->error "Cannot open "C:\a\?". The system cannot find the specified file")

-When filtering results it said "cannot open file "C:\a\?&?&?". The filename, directory name, or volume is incorrect". BTW CE froze.

0000165: Resizing columns causes shrinkage on second monitor in dual-monitor setup

I run a dual monitor setup with the primary on the right. Games run on the right monitor, and CE runs on the left so I can monitor memory in realtime. In the main window and the disassembler, if I try to resize a column by dragging its border, the column shrinks to its minimum size and will not allow me to resize it further. To fix, I have to move the CE window to the right monitor, resize the column, and move it back.

0000164: Add search for multiples of x

The search options available are excellent, but one nice addition would be the ability to search for multiples of a number. For example, if I wanted to search for the number 22, 44, 66, 88, and 110 all at once, I should be able to search for 22 and select the option (along with a secondary number determining the number of multiples) and then search.

0000162: Trainer generating in CE 5.6 is not working.

Hi!

I would like to report an issue with the trainer maker in Cheat Engine 5.6 .
I use CE for a long time to create my own trainers but for some reason,
it is not working in CE 5.6 . I have re-installed 5.5 and it was working
correctly, then I have re-installed CE 5.6 again and the error occured again.

The description of the exact error is the following:
I open up a table and start to make a trainer. Everything is fine until the
point when I would like to "Generate the trainer". Then while it should create
the exe file, I receive an error message, "Access violation at address 0050A2F6
in module 'Cheat Engine.exe'. Read of address 00000000.
(error_message_when_creating_exe.JPG is a picture about this message)

It will create the file but all of the file sizes are oddly 704 000 bytes and
when I try to start it, I get another message:
"There was an error while trying to read the trainer. Or the file got changed(like a
exe-compresser), or you're using an old unpatched version of win 95 or earlier."
(starting_corrupted_trainer.JPG is a picture about this message)

I have tried to make a trainer from all kinds of tables, with from simple freezing
and code nopping to auto-assemble scripts. None of them has been created correctly with
5.6 regardless of the fact that I have tried with tables I have made in 5.5 and with
tables from 5.6 too. I have also tried to simply open an existing working trainer (which
has been created by me in 5.5 and is working for me perfectly) and simply save it again
in 5.6 without any changes and the same error has occured. (Though I also need to mention
from some version of CE, opening an existing trainer was not working for me but it is
working in 5.6 now, however I am not able to utilize this feature until this new bug
can be terminated.)

So in summary, I doubt that the error is related to the cheat tables but in order to
provide a sample, I have attached a CE table and it contains the code which is decreasing
the number in the first step of the CE tutorial. Of course this table is working as it
should, no problems at all. I have also attached a corrupt trainer which has been made
from this table.

I have also terminated the anti-virus to see if this may be the cause but unfortunately not.
I am using the default settings in CE, I have also attached a dxdiag about my computer.
As You can see I am using WindowsXP Professional with SP3, it is a Hungarian version.
I have 4 gb ram and an Intel 8400 coreduo processor. Latest updates are installed.

If You have any suggestions that may help I will try it out as soon as I get Your reply.

Thank You and Have a Nice Day!
Geri

0000129: View pointer scan results

Currently, the pointer scan results are visible only after the scan has been completed/stopped. Is there any specific reason why the results are not currently visible during the scan?
I would like to see the results as soon as they are found - so I can start working with them right away - while the scanning can continue in the background.

0000155: No disk in drive E:

When I open Cheat Engine 5.5 and then select some process (2 out of 3 for now), a message box appear and tells me that there is no disk in Drive E:. My processes are not running from drive E:. However, when there is really a disk in drive E: (A RANDOM DISK) the program does not show any error.

0000154: Change Address window error

When opened "address change" window for pointer based address and then uncheck the "pointer" checkbox, labels will stay background.

If I uncheck and check this checkbox again, all values will be deleted and empty edits shown. It would be nice changing visible option of component, not free them. That why it will like undo.

0000054: CE 5.4.2 - Additional NOPs after conditional jump

It seems that CE inserts additional NOPs (for no apparent reason) into the code.
IMHO, these additional NOPs are not needed - they only consume CPU time.

0000066: Logging

Hi Darkbyte,
as far as I know, currently logging is not possible.
There is a game I want to debug, but piece of code (where my breakpoint is) seems to be called about a 100 times, so it's really hard to analyze each break and compare the registers one by one.
If, I could log all registers after the code has been executed (the "More info" button kind of thing in the "Find out who accessed this memory" feature), I could cross-check the registers and find out the right call conditions quickly.
I imaging three options:
"No logging" - obvious.
"Log on break" - the user sets a breakpoint and whenever it's triggered, all data is logged.
"Log on execution" - the user "flags" the instruction it wants to "watch" and every time the code is executed (regardless if there was a breakpoint or not), all data is logged. I guess there could be problems with this one (excessive amount of data logged, so a "Limit log size to x MByes" would be a good idea (the log could be automatically cleared, or, written to file and cleared - I leave this to the pros).
Alternatively, (or, at the same time ;-) ) you could implement a Log() function in the auto-assembler. Log ("") would insert a new line, Log ("Eax: ", EAX) inserts "Eax: 0x0000ABCD (43981)" to the log, and so on, to allow logging any data one might desire.

0000051: CE 5.4.2 - Cannot add pointer

I found the right address for Warlords Battlecry II - 1.04 beta patch to hack the money.

I run a pointer scan on the address.
I get one result: Battlecry II.exe+002973CC
When I double-click the result, I get this error:
Access violation at address 00404F90 in module "Cheat Engine.exe". Write of address FFFFFFF8.

In the end, instead of a pointer, a simple address is added to the table (which is already there).

I have CE 5.4.2.

0000068: Change type of multiple records

I add 20 pointers (the results of a pointer scan) to the cheat table.
Each of these are added to the table as "4 Bytes" (dunno if it's a bug, but I was scanning for a "Float" pointer - why pointers that result from a "Float" scan are added as "4 Bytes" pointers is beyond my comprehension).
Anyway, if you right-click on one entry in the cheat table, the menu pops up and you can select "Change record" -> "Type" - but only when one address is selected. When I select two or more addresses, this pop-up menu item is disabled. So, I have to change the type to "Float" manually 20 times, for each pointer separately.
I would like to request extending the "Change record" -> "Type" feature to multiple records.
Thank you.