#391 Progress of the 3.5.0 patch

Slēgta
Krock atvēra 2 gadi atpakaļ · 21 komentāri
Krock komentēja 2 gadi atpakaļ

Official release date: 1 Mar (pre-download 27 Feb)

Patch completion/script unlock: 4 Mar ¹

Progress so far (updated incrementally):

  • DONE - Locate the "check functions"
  • DONE - Locate the service start function
  • DONE - Locate & patch the manufacturer string ("Wine", new method)
  • DONE - Write patch skeleton
  • DONE - Anti-abuse quirk
  • DONE - Check for new launcher key
  • DONE - Secondary "low CPU" patch

Tasks after official launch:

  • DONE - Dump "correct" values from a Windows installation
  • DONE - Check for newly added logging servers
  • DONE - Publish patch for testing
  • TODO - Test that patch (crowdsourcing)

Note: The CN counterpart is done by @y0soro and will have the same changes.

Please only answer to this issue if you have questions or helpful inputs. Note that development is taking place in the doctor branch until there is a patch available for testing.

¹) Might also take longer. New code is an unknown time factor.

Official release date: 1 Mar (pre-download 27 Feb) Patch completion/script unlock: 4 Mar ¹ **Progress so far (updated incrementally):** * DONE - Locate the "check functions" * DONE - Locate the service start function * DONE - Locate & patch the manufacturer string ("Wine", new method) * DONE - Write patch skeleton * DONE - Anti-abuse quirk * DONE - Check for new launcher key * DONE - Secondary "low CPU" patch **Tasks after official launch:** * DONE - Dump "correct" values from a Windows installation * DONE - Check for newly added logging servers * DONE - Publish patch for testing * TODO - Test that patch (crowdsourcing) Note: The CN counterpart is done by @y0soro and will have the same changes. Please only answer to this issue if you have questions or helpful inputs. Note that development is taking place in the `doctor` branch until there is a patch available for testing. ¹) Might also take longer. New code is an unknown time factor.
Krock komentēja 2 gadi atpakaļ
Īpašnieks

Testing patches for 3.5.0 are done

As usual, please use only testing accounts until the patch is confirmed working. If everything goes well, patch unlock will be on 4 March.

Please test the following patches:

  • patch.sh
  • patch_anti_logincrash.sh
    • Likely starting from Friday (server time), this patch will avoid increased CPU usage on Linux kernels with timer frequencies <1000 Hz (see #195 for details).
#### Testing patches for 3.5.0 are done As usual, please **use only testing accounts** until the patch is confirmed working. If everything goes well, patch unlock will be on 4 March. Please test the following patches: * `patch.sh` * `patch_anti_logincrash.sh` * Likely starting from Friday (server time), this patch will avoid increased CPU usage on Linux kernels with timer frequencies <1000 Hz (see #195 for details).
ekz48 komentēja 2 gadi atpakaļ

Looks like at least for me the game is stuck at "Checking updates"...

But two minutes of "doing something" the game was loaded.

Passed some dailies on second account. Looks like the game works without issues

R7-3800X/RX6600XT/Lutris Proton 7-33/Gentoo

Looks like at least for me the game is stuck at "Checking updates"... But two minutes of "doing something" the game was loaded. Passed some dailies on second account. Looks like the game works without issues R7-3800X/RX6600XT/Lutris Proton 7-33/Gentoo
Nadats komentēja 2 gadi atpakaļ

One unusual thing I could notice this time is that for the game refuses to start entirely without "patch_anti_logincrash.sh" applied. The application just quietly exits before a window is even shown.

Aside from that I've encountered no issues so far but my testing was very short.

One unusual thing I could notice this time is that for the game refuses to start entirely without "patch_anti_logincrash.sh" applied. The application just quietly exits before a window is even shown. Aside from that I've encountered no issues so far but my testing was very short.

Played for around ~1 hour. One freeze encountered when accessing Paimon menu, didn't happened again with subsequent gameplays.

Thank you very much!

Played for around ~1 hour. One freeze encountered when accessing Paimon menu, didn't happened again with subsequent gameplays. Thank you very much!
cybik komentēja 2 gadi atpakaļ

Patch Test Log

Updated throughout the week.

Bench: 8700k/3060Ti(525.85.05)/32gb RAM, Ubu by Pop 22.04.

Trash account test

Target: Patched by hand, Proton-GE 7-49, manual run.

  • Log in, Play etc, Daylies, Wishing, Leylines, Netplay
  • TCG
  • No auto-ban after instant / a few minutes / ~1hr / ~4hr / daily reset

Main account test

Target: As above

  • Log in, Play etc, Daylies, Wishing, Leylines, Netplay, TCG
  • No auto-ban after instant / a few minutes / ~1hr / ~4hr / daily reset

Verdict

Looks clear.

### Patch Test Log Updated throughout the week. **Bench:** 8700k/3060Ti(525.85.05)/32gb RAM, Ubu by Pop 22.04. #### Trash account test *Target*: Patched by hand, Proton-GE 7-49, manual run. * [x] Log in, Play etc, Daylies, Wishing, Leylines, Netplay * [ ] TCG * [x] No auto-ban after instant / a few minutes / ~1hr / ~4hr / daily reset #### Main account test *Target*: As above * [x] Log in, Play etc, Daylies, Wishing, Leylines, Netplay, TCG * [x] No auto-ban after instant / a few minutes / ~1hr / ~4hr / daily reset #### Verdict Looks clear.
saturn komentēja 2 gadi atpakaļ

Tested on a test account with both patches applied for about an hour. Made a couple wishes, did some overworld exploration, tested the in-game browser. Everything working great so far.

4790k, GTX 1070 (525.89.02), arch linux, wine lutris-ge 7.35, dxvk 2.1

Tested on a test account with both patches applied for about an hour. Made a couple wishes, did some overworld exploration, tested the in-game browser. Everything working great so far. 4790k, GTX 1070 (525.89.02), arch linux, wine lutris-ge 7.35, dxvk 2.1
Ro.R komentēja 2 gadi atpakaļ

Played for couple of hours too. Not any problem to report.

I performed both patches

AMD Ryzen 7 2700X + GeForce GTX 1060
openSUSE Tumbleweed kernel 6.1.8-1-default
Lutris with wine lutris-fs-hack-7.2
Played for couple of hours too. Not any problem to report. I performed both patches AMD Ryzen 7 2700X + GeForce GTX 1060 openSUSE Tumbleweed kernel 6.1.8-1-default Lutris with wine lutris-fs-hack-7.2
shenmo komentēja 2 gadi atpakaļ

Failed to load the game Error code: -9006 Genshin CN official

             ............                shenmo@shenmo-PC 
         .';;;;;.       .,;,.            ---------------- 
      .,;;;;;;;.       ';;;;;;;.         OS: Deepin 20.8 x86_64 
    .;::::::::'     .,::;;,''''',.       Host: ASUS TUF Gaming A15 FA506QM_FA506QM 1.0 
   ,'.::::::::    .;;'.          ';      Kernel: 5.18.17-amd64-desktop-community-hwe 
  ;'  'cccccc,   ,' :: '..        .:     Uptime: 23 mins 
 ,,    :ccccc.  ;: .c, '' :.       ,;    Packages: 2649 (dpkg) 
.l.     cllll' ., .lc  :; .l'       l.   Shell: bash 5.0.3 
.c       :lllc  ;cl:  .l' .ll.      :'   Resolution: 1920x1080 
.l        'looc. .   ,o:  'oo'      c,   DE: Deepin 20.8 
.o.         .:ool::coc'  .ooo'      o.   WM: KWin 
 ::            .....   .;dddo      ;c    Theme: deepin [GTK2/3] 
  l:...            .';lddddo.     ,o     Icons: blossom-light [GTK2/3] 
   lxxxxxdoolllodxxxxxxxxxc      :l      Terminal: deepin-terminal 
    ,dxxxxxxxxxxxxxxxxxxl.     'o,       CPU: AMD Ryzen 7 5800H with Radeon Graphics (16) @ 3.200GHz 
      ,dkkkkkkkkkkkkko;.    .;o;         GPU: NVIDIA GeForce RTX 3060 Mobile / Max-Q 
        .;okkkkkdl;.    .,cl:.           GPU: AMD ATI 06:00.0 Cezanne                                                         
            .,:cccccccc:,.               Memory: 5552MiB / 15402MiB   
Failed to load the game Error code: -9006 Genshin CN official ``` ............ shenmo@shenmo-PC .';;;;;. .,;,. ---------------- .,;;;;;;;. ';;;;;;;. OS: Deepin 20.8 x86_64 .;::::::::' .,::;;,''''',. Host: ASUS TUF Gaming A15 FA506QM_FA506QM 1.0 ,'.:::::::: .;;'. '; Kernel: 5.18.17-amd64-desktop-community-hwe ;' 'cccccc, ,' :: '.. .: Uptime: 23 mins ,, :ccccc. ;: .c, '' :. ,; Packages: 2649 (dpkg) .l. cllll' ., .lc :; .l' l. Shell: bash 5.0.3 .c :lllc ;cl: .l' .ll. :' Resolution: 1920x1080 .l 'looc. . ,o: 'oo' c, DE: Deepin 20.8 .o. .:ool::coc' .ooo' o. WM: KWin :: ..... .;dddo ;c Theme: deepin [GTK2/3] l:... .';lddddo. ,o Icons: blossom-light [GTK2/3] lxxxxxdoolllodxxxxxxxxxc :l Terminal: deepin-terminal ,dxxxxxxxxxxxxxxxxxxl. 'o, CPU: AMD Ryzen 7 5800H with Radeon Graphics (16) @ 3.200GHz ,dkkkkkkkkkkkkko;. .;o; GPU: NVIDIA GeForce RTX 3060 Mobile / Max-Q .;okkkkkdl;. .,cl:. GPU: AMD ATI 06:00.0 Cezanne .,:cccccccc:,. Memory: 5552MiB / 15402MiB ```
shenmo komentēja 2 gadi atpakaļ

The screenshot here

The screenshot here
Hy.S komentēja 2 gadi atpakaļ

I played through the new Archon Quest using y0soro's CN patch. Everything works! The game hung a few times and I had to force quit, but it didn't matter. The overall experience was great!

                    'c.          hongyushi@HySs-MacBook-Air.local 
                 ,xNMM.          -------------------------------- 
               .OMMMMo           OS: macOS 13.2 22D49 arm64 
               OMMM0,            Host: MacBookAir10,1 
     .;loddo:' loolloddol;.      Kernel: 22.3.0 
   cKMMMMMMMMMMNWMMMMMMMMMM0:    Uptime: 7 days, 14 hours, 19 mins 
 .KMMMMMMMMMMMMMMMMMMMMMMMWd.    Packages: 211 (brew) 
 XMMMMMMMMMMMMMMMMMMMMMMMX.      Shell: zsh 5.8.1 
;MMMMMMMMMMMMMMMMMMMMMMMM:       Resolution: 1440x900@2x 
:MMMMMMMMMMMMMMMMMMMMMMMM:       DE: Aqua 
.MMMMMMMMMMMMMMMMMMMMMMMMX.      WM: Quartz Compositor 
 kMMMMMMMMMMMMMMMMMMMMMMMMWd.    WM Theme: Blue (Light) 
 .XMMMMMMMMMMMMMMMMMMMMMMMMMMk   Terminal: Apple_Terminal 
  .XMMMMMMMMMMMMMMMMMMMMMMMMK.   Terminal Font: SFMono-Regular 
    kMMMMMMMMMMMMMMMMMMMMMMd     CPU: Apple M1 
     ;KMMMMMMMWXXWMMMMMMMk.      GPU: Apple M1 
       .cooc,.    .,coo:.        Memory: 2431MiB / 16384MiB 

I used a patched version of crossover-wine (based on wine 7.7), with a patched version of dxvk (based on DXVK-macOS 1.10.3-47). To bypass the transform feedback issue on macOS, I edited globalgamemanagers in YuanShen_Data using UABE (by disabling GPU-skining, credit goes to Shain, I also used his dxvk patch).

Since I don't know how and why the dxvk patch works, I won't report this to the macOS related issue for now.

The game runs at 2880x1800 resolution and the framerate varies from 24fps to 30fps. For more details, please refer to the screenshots.

I played through the new Archon Quest using y0soro's CN patch. Everything works! The game hung a few times and I had to force quit, but it didn't matter. The overall experience was great! ``` 'c. hongyushi@HySs-MacBook-Air.local ,xNMM. -------------------------------- .OMMMMo OS: macOS 13.2 22D49 arm64 OMMM0, Host: MacBookAir10,1 .;loddo:' loolloddol;. Kernel: 22.3.0 cKMMMMMMMMMMNWMMMMMMMMMM0: Uptime: 7 days, 14 hours, 19 mins .KMMMMMMMMMMMMMMMMMMMMMMMWd. Packages: 211 (brew) XMMMMMMMMMMMMMMMMMMMMMMMX. Shell: zsh 5.8.1 ;MMMMMMMMMMMMMMMMMMMMMMMM: Resolution: 1440x900@2x :MMMMMMMMMMMMMMMMMMMMMMMM: DE: Aqua .MMMMMMMMMMMMMMMMMMMMMMMMX. WM: Quartz Compositor kMMMMMMMMMMMMMMMMMMMMMMMMWd. WM Theme: Blue (Light) .XMMMMMMMMMMMMMMMMMMMMMMMMMMk Terminal: Apple_Terminal .XMMMMMMMMMMMMMMMMMMMMMMMMK. Terminal Font: SFMono-Regular kMMMMMMMMMMMMMMMMMMMMMMd CPU: Apple M1 ;KMMMMMMMWXXWMMMMMMMk. GPU: Apple M1 .cooc,. .,coo:. Memory: 2431MiB / 16384MiB ``` I used a patched version of crossover-wine (based on wine 7.7), with a patched version of dxvk (based on DXVK-macOS 1.10.3-47). To bypass the transform feedback issue on macOS, I edited `globalgamemanagers` in `YuanShen_Data` using UABE (by disabling GPU-skining, *credit goes to Shain, I also used his dxvk patch*). **Since I don't know how and why the dxvk patch works, I won't report this to the macOS related issue for now.** The game runs at 2880x1800 resolution and the framerate varies from 24fps to 30fps. For more details, please refer to the screenshots.
cybik komentēja 2 gadi atpakaļ

@Hy.S O.O

Sorry krock but this is amazing.

@Hy.S O.O Sorry krock but this is amazing.

Played for about half an hour yesterday, no ban after a daily rollover today.

Haven't encountered any issues in gameplay either.

However, remember to write down your account details. Username field got reset this update too. So if you don't log in that account often, it might catch you by surprise (as it did me). You might want to start the game before running ./update_g.sh to check the credentials.

Also, thank you very much for the patch.

Played for about half an hour yesterday, no ban after a daily rollover today. Haven't encountered any issues in gameplay either. However, **remember to write down your account details**. Username field got reset this update too. So if you don't log in that account often, it might catch you by surprise (as it did me). You might want to start the game before running `./update_g.sh` to check the credentials. Also, thank you very much for the patch.
3Shain komentēja 2 gadi atpakaļ

A couple of days ago I had successfully made GI run on macOS by patching globalgamemanagers and dxvk although it's yet to be public. Several folks (@Hy.S included) tested it on various models of Mac (Intel/M1/M2 all included) as well as systems (Big Sur/Monterey/Ventura) and so far everything works great, especially the experience on M1/M2 Max is fantastic with Retina enabled. A basic M1 model can also have a decent experience.

However, we still encounter some glitches (not to be investigated here but just for your information)

  • M1/M2 users have huge fps drop when entering Dragonspine
  • Randomly frozen screen (but background music continues) when playing dialogues

All tests are done on CN server using y0soro's patches.

I do have a plan for the launcher for macOS.

A couple of days ago I had successfully made GI run on macOS by patching `globalgamemanagers` and dxvk although it's yet to be public. Several folks (@Hy.S included) tested it on various models of Mac (Intel/M1/M2 all included) as well as systems (Big Sur/Monterey/Ventura) and so far everything works great, especially the experience on M1/M2 Max is fantastic with Retina enabled. A basic M1 model can also have a decent experience. However, we still encounter some glitches (not to be investigated here but just for your information) * M1/M2 users have huge fps drop when entering Dragonspine * Randomly frozen screen (but background music continues) when playing dialogues All tests are done on CN server using y0soro's patches. I do have a plan for the launcher for macOS.

i use bash "...../patch.sh it appear

=== !! UNTESTED PATCH. CHECK FOR BANS USING A TRASH ACCOUNT !! === === REPORT RESULTS TO THE MAIN REPOSITORY. THANK YOU. ===

If you would like to test this patch, modify this script and remove the line below this one.

......

sorry , I fix it . i delete the "exit 0" it work thank

i use bash "...../patch.sh it appear === !! UNTESTED PATCH. CHECK FOR BANS USING A TRASH ACCOUNT !! === === REPORT RESULTS TO THE MAIN REPOSITORY. THANK YOU. === If you would like to test this patch, modify this script and remove the line below this one. ...... sorry , I fix it . i delete the "exit 0" it work thank
Krock komentēja 2 gadi atpakaļ
Īpašnieks

@shenmo The error code -9006 stands for "I/O error". Please check whether the game directory is writable by the current user (non-root). See TROUBLESHOOTING.md for solutions and sanity checks.

@Hy.S / @3Shain This is an impressive achievement. Would you (or another user) please be so nice to write down instructions (separate issue or Pull Request) for INSTALLATION.md on how to get Genshin to run on MacOS (M1+)? I can imagine that there are many more people interested in it.

@shenmo The error code -9006 stands for "I/O error". Please check whether the game directory is writable by the current user (non-root). See [TROUBLESHOOTING.md](src/master/TROUBLESHOOTING.md#an-error-code-appears) for solutions and sanity checks. @Hy.S / @3Shain This is an impressive achievement. Would you (or another user) please be so nice to write down instructions (separate issue or Pull Request) for INSTALLATION.md on how to get Genshin to run on MacOS (M1+)? I can imagine that there are many more people interested in it.
infine komentēja 2 gadi atpakaļ

Same as @Nadats. Game quietly died before showing anything without the antilogincrash patch.

Working fine with it.

Same as @Nadats. Game quietly died before showing anything without the antilogincrash patch. Working fine with it.
HK416 komentēja 2 gadi atpakaļ

Playing on Ubuntu for approximately 5 hours hours with my main Account (yes I know...) since yesterday. No problems whatsoever. Performance is good and no kick, no crash, no ban.

Playing on Ubuntu for approximately 5 hours hours with my main Account (yes I know...) since yesterday. No problems whatsoever. Performance is good and no kick, no crash, no ban.
3Shain komentēja 2 gadi atpakaļ

@Krock Will be public soon. Although it's complicated since we need custom build of dxvk (and possibly custom build of wine for 8G-Ram users) so I'm not really interested in writing full instructions within short time, but I can provide prebuilt binaries (honestly I'm really keen on providing a ready-to-use launcher, of course it will be open source somewhere so anyone can check how it works)

@Krock Will be public soon. Although it's complicated since we need custom build of dxvk (and possibly custom build of wine for 8G-Ram users) so I'm not really interested in writing full instructions within short time, but I can provide prebuilt binaries (honestly I'm really keen on providing a ready-to-use launcher, of course it will be open source somewhere so anyone can check how it works)
shenmo komentēja 2 gadi atpakaļ

fixed

fixed

Tested on a trash account, I'm using a custom Linux kernel 6.2.1-tkg-bmq, and I did not realize the high CPU usage issue until I noticed my CPU temperature was way higher than what it was in 3.4. It seems using esync fixes that problem, but fsync causes near-constant 100% CPU. I read up on the CPU issue above, so I understand now why that's the case.

Tested on a trash account, I'm using a custom Linux kernel 6.2.1-tkg-bmq, and I did not realize the high CPU usage issue until I noticed my CPU temperature was way higher than what it was in 3.4. It seems using esync fixes that problem, but fsync causes near-constant 100% CPU. I read up on the CPU issue above, so I understand now why that's the case.
Krock komentēja 2 gadi atpakaļ
Īpašnieks

Thank you for all test reports. No further issues found. The patch script is now unlocked.

Thank you for all test reports. No further issues found. The patch script is now unlocked.
Pierakstieties, lai pievienotos šai sarunai.
Nav atskaites punktu
Nav atbildīgā
Notiek ielāde...
Atcelt
Saglabāt
Vēl nav satura.