Homebrew Official Retroarch WiiU (wip.)

jacobsson

Well-Known Member
Member
Joined
Oct 30, 2019
Messages
165
Trophies
0
Age
38
XP
769
Country
Sweden
No, i can´t tell more about it. Because a) this is dangerous. Nobody guarantees WiiU won´t malfunction. b and b) It´s hard to do. Who can compile their own rpx? Exactly. Not a beginner´s task.
c) You still couldn´t load @ max-speed when you only use a hdd. An SSD is required & strongly recommended (SSD with many iops & of a good brand).

Increasing the transfer speed could greatly reduce loading times of larger games, this is not at all about hitting USB 2.0 max speeds, it's about not having extreme load times, hence a nicer experience using RA. Just look at Street Fighter 3: 3rd strike, 1 min load time (most people prob thinks their console has frozen by now!).
As of right now running more advanced roms in RA from a hard drive is on the verge of "unplayable" for many people.

Personally, whenever I share stuff that could potentially harm the hardware I always put up a disclaimer, but hey we can't force you you to share the solution with us. Don't get me wrong, you've done plenty of neat stuff for the scene and I'm very great full all devs still in what little is left of the WiiU-scene putting in their hard work!

I'm optimistic to the fact that there is a solution out there for increased write-speeds, even if it's "dangerous" it can always be refined, I mean isn't that what this is all about?

I don't think building your own RetroArch .rpx is above most peoples heads in here, basically you need a devkit from the download archive and some environmental paths set in your OS.

I'd love helping you making a guide for this patch/mod if you ever change your mind about sharing your findings!
 
Last edited by jacobsson,

depaul

Well-Known Member
Member
Joined
May 21, 2014
Messages
1,293
Trophies
0
XP
2,952
Country
France
No, i can´t tell more about it. Because a) this is dangerous. Nobody guarantees WiiU won´t malfunction. b and b) It´s hard to do. Who can compile their own rpx? Exactly. Not a beginner´s task.
c) You still couldn´t load @ max-speed when you only use a hdd. An SSD is required & strongly recommended (SSD with many iops & of a good brand).
Hi The chosen. Please excuse my question but even if dangerous we have the tool 'Dumpling' that can read encrypted HDD at full speed (20MB/s) and I had no issue so far.. So we can asume that HDD reading isn't too risky (?).
 
  • Like
Reactions: jacobsson

jacobsson

Well-Known Member
Member
Joined
Oct 30, 2019
Messages
165
Trophies
0
Age
38
XP
769
Country
Sweden
Hi The chosen. Please excuse my question but even if dangerous we have the tool 'Dumpling' that can read encrypted HDD at full speed (20MB/s) and I had no issue so far.. So we can asume that HDD reading isn't too risky (?).

I think you make an excellent point!
I've reached out to Lia (the author of Dumpling) to see he thinks.
 
  • Like
Reactions: depaul and Masana
D

Deleted User

Guest
Patience guys. I´m working on things, too.Currently, i learn to understand how loadiine works. my understanding is, every file is being unencrypted, so that every other machine can run it.WiiU natively uses WUP-format (for discs) though which is encrypted.
All i said, is WiiU by standard encrypts every file on USB with an AES-128 bit cbc encryption (on a posix filesystem with a 6-kbyte-blocksize), which is a heavy encryption-standard considered to be secure (assuming you don´t have the key). As we have the keys, we can use the key & feed it that mechanism & it´s output an unencrypted file.
Currently, i didn´t get my loadiine-setup working. I dumped some games, but i just got a new ssd to try this. I´m using an Intel-SSD to try, what happens if you modify necessary codelines in loadiine-program & run it on WiiU with unencrypted files.
Typically alot of loading-issues should vanish. Everything should also be more smooth (assuming it did have i/o-issues before because of heavy encryption/drm).
But i only have 2 games to try it.
 

CORE

3:16
Member
Joined
Jul 15, 2018
Messages
1,176
Trophies
1
XP
2,067
Country
United Kingdom
@TheChosen hows things running for you regarding Loadiine?

I have tryed with a Portable 2.5 OCZ SSD and USB 3.0 Connection in combination with MochaFAT32 and results where not that impressive to say the least regarding couple of Builds that I tryed.

Have or Are you working on Loadiine?
 
  • Like
Reactions: depaul

n00bsaib0t

Well-Known Member
Member
Joined
Feb 12, 2015
Messages
287
Trophies
0
Age
38
Location
Phoenix, AZ
XP
1,150
Country
United States
How fast is Retroarch on the Wii U for SNES, Genesis, CPS2 and NeoGeo?
If you mean how well do they run, full speed with no issues. If you mean how fast do they load, not super fast. I personally don’t think it’s an issue since I’m not game hopping but some do. Expect 10-30 seconds depending on the size of the rom. But once they load they run great.
 
  • Like
Reactions: CORE and Nemix77
Joined
Apr 19, 2015
Messages
1,023
Trophies
1
Location
Stuck in the PowerPC
Website
heyquark.com
XP
3,909
Country
Australia
Just so you all know, TheChosen has a history of spreading technical misinformation despite numerous attempts by homebrew developers to clear things up. I'm sure they'll quote this post and say I don't understand something, it's what they always used to do when misunderstanding how interrupts work.
A few quick notes:
- Aside from libfat, filesystem encryption and access is handled entirely by the IOSU, not RetroArch. The OS functions take a path to a file, then the IOSU returns unencrypted data. Anything between those two steps is hard-coded in IOSU and we have no control over it. libfat asks for raw disk sectors instead of files, and then parses out the filesystem itself, but none of the FAT filesystems used have any encryption - libfat improvements are likely where the most potential lies.
- The Wii U has hardware AES acceleration, for the IOSU only. I believe this is used for encrypted filesystem access.
- The encryption is not what's slowing down the Wii U's filesystem access - as evidenced by libfat and jacobsson's recent USB patches having comparable speeds.
- If there was patches available to significantly improve speeds, we'd happily merge them upstream to have them in the next batch of nightlies - no need to patch every core. This doesn't even make sense when you know how RetroArch is built - you only need to change it once to affect every core you build.
- Filesystem access isn't inherently dangerous. If we started writing files to the NAND I'd be worried, but ultimately you can unplug a USB drive and the console will be entirely unaffected.
 

jacobsson

Well-Known Member
Member
Joined
Oct 30, 2019
Messages
165
Trophies
0
Age
38
XP
769
Country
Sweden
@QuarkTheAwesome thank you for the clarification!
Regarding the earlier mentioned, it made me very confused when there was mostly talk and no PoC to be shown, or any evidence of something working. I'm glad that we can get back on track again.
I really wish a knew more about filesystems so that I could contribute, but one is way over my head.

I have a RA related question for you: since I don't own a gamepad/tablet and many others are dying, I trying to figure out why the gamepad is auto assigned to user 1 even though it's not even connected to the system.

The problem with this is that RA has been coded to only allow player 1 (index 1) to toggle the menu, which you can't if you don't have a tablet.

A dirty workaround for those w/o gamepads, mod the retroarch.cfg as following:
FIND: input_player1_joypad_index = "0"
REPLACE WITH: input_player1_joypad_index = "5"

FIND: input_player2_joypad_index = "1"
REPLACE WITH: input_player2_joypad_index = "0"

FIND: input_player3_joypad_index = "2"
REPLACE WITH: input_player3_joypad_index = "1"

FIND: input_player4_joypad_index = "3"
REPLACE WITH: input_player4_joypad_index = "2"

FIND: input_player5_joypad_index = "4"
REPLACE WITH: input_player5_joypad_index = "3"

This will make the first non-gamepad type controller to be assigned index 1, hence be able to toggle the menu.

Do you know where in the code the gamepad gets initated to auto-allocated to index 1? I'd love to just fix this properly so if it's not connected is shouldn't populate the first index if possible.

As always, thanks man you're awesome as usual!
 
Last edited by jacobsson,
Joined
Apr 19, 2015
Messages
1,023
Trophies
1
Location
Stuck in the PowerPC
Website
heyquark.com
XP
3,909
Country
Australia
@QuarkTheAwesome thank you for the clarification!
Regarding the earlier mentioned, it made me very confused when there was mostly talk and no PoC to be shown, or any evidence of something working. I'm glad that we can get back on track again.
I really wish a knew more about filesystems so that I could contribute, but one is way over my head.

I have a RA related question for you: since I don't own a gamepad/tablet and many others are dying, I trying to figure out why the gamepad is auto assigned to user 1 even though it's not even connected to the system.

The problem with this is that RA has been coded to only allow player 1 (index 1) to toggle the menu, which you can't if you don't have a tablet.

A dirty workaround for those w/o gamepads, mod the retroarch.cfg as following:
FIND: input_player1_joypad_index = "0"
REPLACE WITH: input_player1_joypad_index = "5"

FIND: input_player2_joypad_index = "1"
REPLACE WITH: input_player2_joypad_index = "0"

FIND: input_player3_joypad_index = "2"
REPLACE WITH: input_player3_joypad_index = "1"

FIND: input_player4_joypad_index = "3"
REPLACE WITH: input_player4_joypad_index = "2"

FIND: input_player5_joypad_index = "4"
REPLACE WITH: input_player5_joypad_index = "3"

This will make the first non-gamepad type controller to be assigned index 1, hence be able to toggle the menu.

Do you know where in the code the gamepad gets initated to auto-allocated to index 1? I'd love to just fix this properly so if it's not connected is shouldn't populate the first index if possible.

As always, thanks man you're awesome as usual!

The Wii U's kinda built around assuming there's a Gamepad all the time (shoutouts to System Settings) but I did manage to detect it being missing in SDL, it was just kinda odd. I might have a look into that though, should definitely be able to fix something up - assuming RetroArch cleanly deals with controllers ending up with different indexes!
 
  • Like
Reactions: Zense and jacobsson

jacobsson

Well-Known Member
Member
Joined
Oct 30, 2019
Messages
165
Trophies
0
Age
38
XP
769
Country
Sweden
The Wii U's kinda built around assuming there's a Gamepad all the time (shoutouts to System Settings) but I did manage to detect it being missing in SDL, it was just kinda odd. I might have a look into that though, should definitely be able to fix something up - assuming RetroArch cleanly deals with controllers ending up with different indexes!
Cool! Yeah I can see that most stuff origin centralized around the gamepad. I'll have another look tonight to see if I can get a better understanding too.

I must say so far I've had a blast joining this community!
 

n00bsaib0t

Well-Known Member
Member
Joined
Feb 12, 2015
Messages
287
Trophies
0
Age
38
Location
Phoenix, AZ
XP
1,150
Country
United States
Is it possible to make it where, say, a Game Pad and a Pro controller can both be the Player 1 controller? It's a minor inconvenience, yeah, but still kinda annoying when I'm playing on TV and want to use a Pro controller then the next day I want to use the Game Pad off TV, I have to use the one I set as my Player 1 in the settings to switch it to the other. I know some official games are set up this way where the Game Pad and the first detected controller are both Player 1. Or maybe a Switch style menu where you press L+R (or some other combination) on the controller you want to use every time you launch it?
 
  • Like
Reactions: september796
D

Deleted User

Guest
@QuarkTheAwesome: sorry, but what you said is simply wrong.

a) It IS dangerous to unlock USB, for several reasons (remember WiiUs dying "suddenly" after using "GXLoader" for months"?) - yeah we had several cases here of DOA WiiUs because of using this tool- and i told it was dangerous before.

USB 2.0 is a dangerous technology because it´s directly related to your cpu, i know 15 years ago when first Mainbaords with USB 2 were made, cpus died when you killed your USB or you fried your cpu when your USB got overloaded, you surely don`t know it.. It has to do with "V-Core-Drop". You ever heard that word`? to me it seems you don`t know what that is.

What happens if you overload your USB is => your vdrop falls. If you load USB low, your vdrop isn`t running in dangerous levels. but it may happen that if you run certain instructions on the cpu (which max it out to an out-of-TDP-standard level such as AVX) that your USB 2.0 load is so high that you have a very high vcore-drop. And that means your cpu can be damaged if you load it heavily.Nintendo was smart choosing USB 2 for this security measure. i told you 1 year ago.

b) The WiiU can potentially overheat and/or the VRM can be damaged if you unlock the USB without taking measures and testing it carefully. As you know, not every device, which got overheated will work again. I know cpus, which didn´t survive overheating. And that was a Core 2 Quad-cpu from Intel! So there is a certain risk going with it,which is why i don´t post here how it´s fully done. Everybody does it on their own risk and a such damaged WiiU cannot be repaired.

c) Currently unlocking USB ONLY works with one program. and that is "loadiine". If you patch it (see my other thread, which was the reason i created this thread btw since i told you, Nintendo used a custom USB-driver, which tells to not use the full bandwidth).

This can now be changed, when you delete the "256" in each rpx/rpl-file and change one codeline. This is known since a few months, yeah. But nobody ever reported back. So they didn`t follow it further.

I came, knew it was a thing and then i found that thread and i am doing that now. It takes time to do so though. Converting all games to Loadiine (game with no encryption/DRM) takes a lot of time. I just got a new Intel-SSD with very high IOPS in order to test this.

d) Btw:: I know pretty much anything. I even know why "Super Mario 3D World" runs "badly" in "Loadiine".

Wanna know the answer? Nintendo implemented "software-triggers" in it (because they were again smart, made a double-DRM-protection, if first AES 128-bit is broken, you still have software-DRM-triggers, which means you have certain code which checks "how long does my execution take on original hardware?" and then reports back if the result is not like on original hardware and triggers a "key-chain" Which can then lead to...well certain behaviour you told the program to do when it detects a DRM-breach. E.g. "wait xxx seconds and then try again"

Why do i know about this?

Simple. Nintendo made similar DRM-protection on Wii-games like "New Super Mario Bros Wii".

And finally e) I said, theoretically you could unlock all programs like that (e.g. Online-games, you would have to patch the necessary online-rpx-library). The problem is: It´s not known yet how to do this on "Retroarch" or other programs. Because you need to change 2 things in each RPX-file in order for it to work. a) DRM has to be shut-off which means you delete one code-line. And b) You tell the USB to have no limit (=0 as a number instead of the 256 max limit of block-size/chunk etc written as a standard).

But to me it seems you still haven´t understand what the WiiU really is. It`s time to wake up, dude.

And no- right now we have to assume- Nintendo did the most-stupid thing with WiiU and that means they went they software-DRM route. Which means AES 128 bit is fully running in software on the PPC-cores (In Cafe OS).

Which would also perfectly explain why the OS itself is so slow. Because you get the exact same result if you run AES 128 bit-Software encryption on say a standard PC with some HDD attached to it. You even ran a Software-AES-encryption on a non-hardware-accellerated cpu? You saw how a Windows can run if you do that?

Your OS-speed will be limited to 1-2 Mbytes/second. E.g. if you use Truecrypt/Veracrypt and the like with a stone-old cpu which has no hw-accelleration.

The WiiU is a custom machine. It`s cpu can run 20 Gbyte/second. But: If you load it with AES 128 bit, you will never be able to use it with those numbers. Which is very similar to what EA currently does (Denuvo/Origin/etc) all costs a lot of cpu-time.


So what this means is: The WiiU´s cpu uses 95% of ressources on DRM.

Just that simple. And only 5% is usable for the OS.

If- and that´s a big IF- we some day could patch out ALL AES-RPX-files on WiiU- and replace them with patched files and circumvent all necessary OS-libraries which check for all that DRM-chunk in there- the same way how "Loadiine" was patched and it works now- delivering amazing performance- then you could have a smooth OS. The processor would instantly load the data instead of having to wait for the DRM to be finished.


And one last point which is a bit off-topic but i want to mention it here: I also found out more reasons why EA hates Nintendo now. Well it is quite simple:

EA wanted to handle some trigger-based DRM (called Denuvo and Origin is the system which it runs on) on WiiU. Which would have ment they have the full force of what can be done in terms of DRM on WiiU.

But Nintendo created their Software-based AES 128-bit solution and thus the system is "incompatible" with Origin/Denuvo.

Yes, guys. That´s the whole story behind Nintendo´s "EA-love-hate-championship". EA hated Nintendo because they choose to use the software-AES-solution for DRM (which taxes a processor to 90% easily) instead of using EA`s own solution called "Origin" (which is nothing different than all games having software-triggers which detect what you do in the game and if your game is running original or if it´s a copy). Origin is also an account-based system but that´s just basic DRM.

So it was simply a DRM-war. Which DRM has won? It was Origin. Nintendo lost the fight. Crazy times, huh?

Well it was quite easy for me to find this all out. I just needed to compare how Origin/Denuvo works. Et Voila. The comparison is perfect.

The point is: You cannot run a trigger based DRM-system such as "Denuvo" on another AES 128 bit-software-encryption based DRM.

You only can have one without your cpu being maxed out.

--------------------- MERGED ---------------------------

@QuarkTheAwesome: sorry, but what you said is simply wrong.

a) It IS dangerous to unlock USB, for several reasons (remember WiiUs dying "suddenly" after using "GXLoader" for months"?) - yeah we had several cases here of DOA WiiUs because of using this tool- and i told it was dangerous before.

USB 2.0 is a dangerous technology because it´s directly related to your cpu, i know 15 years ago when first Mainbaords with USB 2 were made, cpus died when you killed your USB or you fried your cpu when your USB got overloaded, you surely don`t know it.. It has to do with "V-Core-Drop". You ever heard that word`? to me it seems you don`t know what that is.

What happens if you overload your USB is => your vdrop falls. If you load USB low, your vdrop isn`t running in dangerous levels. but it may happen that if you run certain instructions on the cpu (which max it out to an out-of-TDP-standard level such as AVX) that your USB 2.0 load is so high that you have a very high vcore-drop. And that means your cpu can be damaged if you load it heavily.Nintendo was smart choosing USB 2 for this security measure. i told you 1 year ago.

b) The WiiU can potentially overheat (because of too small ventilator-size) and/or the VRM can be damaged if you unlock the USB without taking measures and testing it carefully. As you know, not every device, which got overheated will work again. I know cpus, which didn´t survive overheating. And that was a Core 2 Quad-cpu from Intel! So there is a certain risk going with it,which is why i don´t post here how it´s fully done. Everybody does it on their own risk and a such damaged WiiU cannot be repaired.

c) Currently unlocking USB ONLY works with one program. and that is "loadiine". If you patch it (see my other thread, which was the reason i created this thread btw since i told you, Nintendo used a custom USB-driver, which tells to not use the full bandwidth).

This can now be changed, when you delete the "256" in each rpx/rpl-file and change one codeline. This is known since a few months, yeah. But nobody ever reported back. So they didn`t follow it further.

I came, knew it was a thing and then i found that thread and i am doing that now. It takes time to do so though. Converting all games to Loadiine (game with no encryption/DRM) takes a lot of time. I just got a new Intel-SSD with very high IOPS in order to test this.

d) Btw:: I know pretty much anything. I even know why "Super Mario 3D World" runs "badly" in "Loadiine".

Wanna know the answer? Nintendo implemented "software-triggers" in it (because they were again smart, made a double-DRM-protection, if first AES 128-bit is broken, you still have software-DRM-triggers, which means you have certain code which checks "how long does my execution take on original hardware?" and then reports back if the result is not like on original hardware and triggers a "key-chain" Which can then lead to...well certain behaviour you told the program to do when it detects a DRM-breach. E.g. "wait xxx seconds and then try again"

Why do i know about this?

Simple. Nintendo made similar DRM-protection on Wii-games like "New Super Mario Bros Wii".

And finally e) I said, theoretically you could unlock all programs like that (e.g. Online-games, you would have to patch the necessary online-rpx-library). The problem is: It´s not known yet how to do this on "Retroarch" or other programs. Because you need to change 2 things in each RPX-file in order for it to work. a) DRM has to be shut-off which means you delete one code-line. And b) You tell the USB to have no limit (=0 as a number instead of the 256 max limit of block-size/chunk etc written as a standard).

But to me it seems you still haven´t understand what the WiiU really is. It`s time to wake up, dude.

And no- right now we have to assume- Nintendo did the most-stupid thing with WiiU and that means they went they software-DRM route. Which means AES 128 bit is fully running in software on the PPC-cores (In Cafe OS).

Which would also perfectly explain why the OS itself is so slow. Because you get the exact same result if you run AES 128 bit-Software encryption on say a standard PC with some HDD attached to it. You even ran a Software-AES-encryption on a non-hardware-accellerated cpu? You saw how a Windows can run if you do that?

Your OS-speed will be limited to 1-2 Mbytes/second. E.g. if you use Truecrypt/Veracrypt and the like with a stone-old cpu which has no hw-accelleration.

The WiiU is a custom machine. It`s cpu can run 20 Gbyte/second. But: If you load it with AES 128 bit, you will never be able to use it with those numbers. Which is very similar to what EA currently does (Denuvo/Origin/etc) all costs a lot of cpu-time.


So what this means is: The WiiU´s cpu uses 95% of ressources on DRM.

Just that simple. And only 5% is usable for the OS.

If- and that´s a big IF- we some day could patch out ALL AES-RPX-files on WiiU- and replace them with patched files and circumvent all necessary OS-libraries which check for all that DRM-chunk in there- the same way how "Loadiine" was patched and it works now- delivering amazing performance- then you could have a smooth OS. The processor would instantly load the data instead of having to wait for the DRM to be finished.


And one last point which is a bit off-topic but i want to mention it here: I also found out more reasons why EA hates Nintendo now. Well it is quite simple:

EA wanted to handle some trigger-based DRM (called Denuvo and Origin is the system which it runs on) on WiiU. Which would have ment they have the full force of what can be done in terms of DRM on WiiU.

But Nintendo created their Software-based AES 128-bit solution and thus the system is "incompatible" with Origin/Denuvo.

Yes, guys. That´s the whole story behind Nintendo´s "EA-love-hate-championship". EA hated Nintendo because they choose to use the software-AES-solution for DRM (which taxes a processor to 90% easily) instead of using EA`s own solution called "Origin" (which is nothing different than all games having software-triggers which detect what you do in the game and if your game is running original or if it´s a copy). Origin is also an account-based system but that´s just basic DRM.

So it was simply a DRM-war. Which DRM has won? It was Origin. Nintendo lost the fight. Crazy times, huh?

Well it was quite easy for me to find this all out. I just needed to compare how Origin/Denuvo works. Et Voila. The comparison is perfect.

The point is: You cannot run a trigger based DRM-system such as "Denuvo" on another AES 128 bit-software-encryption based DRM.

You only can have one without your cpu being maxed out.

So what this means is Nintendo simply went for the cheapest solution for DRM.

And that is buying some simple software to create an AES 128-bit key. We have to assume now the WiiU doesn´t have any sort of hardware-accelleration for AES. It´s all done in a software-solution instead (which is why you call it "RTOS" = Real-time-operating-system, you only use such things in trains and buses and planes, but usually not in a console).

And that´s why they also gave the hardware itself a key. Because this software running in "Cafe OS" is reading those keys constantly before doing anything.

But if you don`t believe me. Hey, you`re free to try it out on your own.

I know it must be frustrating to not understand how it all really works.
But i´m an expert in this. i did this the last 10 years. I learned it all. And to be honest. To learn how Denuvo works or why EA has choosen Denuvo and Origin is quite interesting stuff to learn.

Denuvo is quite similar in working to Steam`s "Anti-Vac" btw.

If you ask: Why did they went with a software-solution for AES?

Simple. You cannot crack software-based AES 128bit. Nobody ever has broken software-based AES 128bit-encryption with Truecrypt/Veracrypt (allways assuming you don`t have hardware-accelleration). It takes centuries to calculate the necessary key.

But you see? Nintendo did one big mistake. And that was putting the key right there in the hardware. They should have known it´s only a matter of time until somebody will be able to detect that key in the hardware...

And once you gain that key, you can decrypt everything. And that´s exatcly how WiiU`s security (partly) was broken. I mean until today so far what´s going on...

And what they did with "Loadiine" is quite simple: If you know the keys of a software-based encryption, you can write a program with the necessary encrpytion-algorithm. Then input the key and voilá it will output you the unencrypted files.

That´s not magic. That´s quite simple how it works.
 
D

Deleted User

Guest
Why did my post got doubled? lol. I swear sometimes this site`s script is messed up.

^^Btw: I`m not taking it personal. I just wanted to clear a few things. and yeah i know that i made a few misconceptions in the past and said some (not so correct) stuff, because my knowledge based on some differing concept, that the WiiU uses different tech than people here said..

I thought- 6 months ago- the WiiU uses an ARM-cpu for security only (which is what TheAwesomeQuark talked about).

But sorry, the code is now directly lying in front of us. And it`s software based AES (128 bit cnc encrypted) only. You can directly look at what it does if you use that special tool which reverse-engineers the code.

All results currently rely on that reverse-engineering. Without that NSA-like tool we wouldn´t have come thar far.

Thanks to that tool more and more was found out and we know exacly how the OS works now.

We just haven´t circumvented all DRM/bullshit right now. But that is only a matter of time.

And we need a lot of time. Because it´s really hard work. Since the WiiU is (and we all knew it- deep in our hearts) a very, very very complex machine. To understand it takes years.

No sane person would invent something such as the "WiiU" except Satoru Iwata. And i have theories why he even came to invent that thing...But he is gone and we can´t ask him. And even if we did he wouldn´t answer us, because it´s still kept under NDA.

But i have faith, that me- together with the others can make it. We can unleash the Power of the WiiU. Maybe until end of 2020 it´s ready.

Like i allways say: Step by step we get to the goal. Now we have USB-accelleration to speed up games and loading. Maybe in another 6 months we find ways to patch the RPX, which controls the network-speed. And then we could even download at full USB/Wifi-Speed as well.

And Miiverse will come back as well. Part of it (the offline part) is actually allready running again. The part of the characters walking in the "Mii Plaza", telling you what you play, what others are online and what they play.

Thanks to Corona i also have more time than ever to look after my hobbies.
 

jacobsson

Well-Known Member
Member
Joined
Oct 30, 2019
Messages
165
Trophies
0
Age
38
XP
769
Country
Sweden
@TheChosen Thanks a lot for sharing those stories and information, it's interesting to say at least! I can't wait to see what this leads up to.

There is just one thing that I can't really understand right now that we should really be asking our selves when it comes to RetroArch (absolutely not neglecting anything of your prior reply):
How come an app like Dumplings can achieve speeds of 20MB/s (~160Mbit/s) when transferring data between HDD and SD card, while RA needs almost a minute to load a 30MB rom from SD or HDD? I know the roms are in zip archives mostly, is that a factor?

EDIT: I guess I need to find a core that accepts both zipped and unzipped roms and compare the load times.

EDIT2:
Street Fighter Alpha 2, Zipped, SNES 9X: 27s
Street Fighter Alpha 2, Unzipped, SNES 9X: 26s

Loading of the core seems to be a huge time factor here. I guess I'm running in circles
 
Last edited by jacobsson,

Sonic Angel Knight

Well-Known Member
Member
Joined
May 27, 2016
Messages
14,399
Trophies
1
Location
New York
XP
12,928
Country
United States
Not to interrupt the story about loading speed, but I had some questions about the performance of retroarch. Just was curious if someone can list all the options that affect the performance. I was playing super mario RPG using snes9x latest core last night and even though It runs at 60FPS, i try to speed up to skip cutscenes and it bearly makes much impact. Not to mention using genesis plus gx core, that I can't play 2 player sonic 2 because it also runs slow compared to single player.

If anyone knows what options that effect performance, that would be nice. :ninja:
 
  • Like
Reactions: Zense and jacobsson

depaul

Well-Known Member
Member
Joined
May 21, 2014
Messages
1,293
Trophies
0
XP
2,952
Country
France
@TheChosen Thanks a lot for sharing those stories and information, it's interesting to say at least! I can't wait to see what this leads up to.

There is just one thing that I can't really understand right now that we should really be asking our selves when it comes to RetroArch (absolutely not neglecting anything of your prior reply):
How come an app like Dumplings can achieve speeds of 20MB/s (~160Mbit/s) when transferring data between HDD and SD card, while RA needs almost a minute to load a 30MB rom from SD or HDD? I know the roms are in zip archives mostly, is that a factor?

EDIT: I guess I need to find a core that accepts both zipped and unzipped roms and compare the load times.
Glad to inform you @jacobsson that all uncompressed roms (SNES, SEGA CD, GBA ...) will load almost instantly if you apply the workaround (remove 'info', 'assets' and 'media' directories). If you don't remove those directories RA will take almost 10s before you load each rom.

So for WiiU USB : uncompressed roms will start almost instantly (2-3s at most) if you apply the workaround. Only compressed roms (Arcade, CPS, Neogeo) will suffer from long loading time.

@Sonic Angel Knight I'd suggest to use old Snes9x builds. They are usually faster.
 
Last edited by depaul,
  • Like
Reactions: Zense

jacobsson

Well-Known Member
Member
Joined
Oct 30, 2019
Messages
165
Trophies
0
Age
38
XP
769
Country
Sweden
@depaul thanks for the info! Do you happen to know roughly when this started to be a problem? I'd like to take a look at the code and see what's going on, but it'd help greatly with some kind of reference so that I can find what commit might have caused it. Have a great day!
 
Last edited by jacobsson,

depaul

Well-Known Member
Member
Joined
May 21, 2014
Messages
1,293
Trophies
0
XP
2,952
Country
France
Thanks @jacobsson. Retroarch has been slow to load even very small roms. Ploggy showed an old video and the community had the idea to delete those folders and it worked.

17s to load a game? yikes! :P I remember it being alot faster than that. Take alook at this vid from 2016..

Its pretty much instant.

Don't know what could have changed over the years for the load speed to become so slow but there's definitely something going on?
 
  • Like
Reactions: jacobsson

Site & Scene News

Popular threads in this forum

General chit-chat
Help Users
  • No one is chatting at the moment.
    K3Nv2 @ K3Nv2: Least they got head in the end