Right now (unless some massive breakthrough happens), the only way you can toy with homebrew is on 3.0.0 and this kind of reminds me of the PS3 when it comes to having to run a specific firmware to run. Are there any lessons that can be borrowed from the PS3 to expedite homebrew on switch?
- Thread starter boscocraftXD
- Start date
- Views 3,217
- Replies 15
First I dont have much knowledge, so its just my opinion!
I don't think they are any similar... It is based on the Nvidia Shield TV chip so its ARM like most mobile devices and I think it will be way better to code for the Switch because it supports Vulkan, Cuda cores.......
Also the GPU is considerably better than the PS3, but the CPU, I have no idea, I have the Nvidia shield tablet which uses a older chip with an even crapier CPU and the CPU power is not up to the GPU, its very weak and I'm still to see any game running better than a PS3 top graphics\physics game even though Nvidia claims the GPU is better than a X360 and PS3.
ps: GPU is better than PS3\X360, but the CPU I don't have a clue if it will be as powerful as the PS3 (but I'm thinking its still a bit far from the PS3 CPU performance). However I think it will surely be way easier to code for this and if using Vulkan API + Cuda well implemented might help increasing performance for emulators (but I have no clue, it's just my opinion and thinking that in PC we use Cuda processing to increase performance of the program
).
I don't think they are any similar... It is based on the Nvidia Shield TV chip so its ARM like most mobile devices and I think it will be way better to code for the Switch because it supports Vulkan, Cuda cores.......
Also the GPU is considerably better than the PS3, but the CPU, I have no idea, I have the Nvidia shield tablet which uses a older chip with an even crapier CPU and the CPU power is not up to the GPU, its very weak and I'm still to see any game running better than a PS3 top graphics\physics game even though Nvidia claims the GPU is better than a X360 and PS3.
ps: GPU is better than PS3\X360, but the CPU I don't have a clue if it will be as powerful as the PS3 (but I'm thinking its still a bit far from the PS3 CPU performance). However I think it will surely be way easier to code for this and if using Vulkan API + Cuda well implemented might help increasing performance for emulators (but I have no clue, it's just my opinion and thinking that in PC we use Cuda processing to increase performance of the program
).
Last edited by guily6669,
The scenarios are too different.
The ps3 has full cfw as we have the signing key along with newer firmware and the ability to downgrade an exploitable console with hardware.
It was also possible to resign games for older firmware.
The ps3 has full cfw as we have the signing key along with newer firmware and the ability to downgrade an exploitable console with hardware.
It was also possible to resign games for older firmware.
Yeah, but thinking on the long future I'm thinking since this is one of the most powerful portable gaming devices on the planet it will or at least should interest a lot of people into exploiting it.
I'm thinking a few years from now we will be able to get something like a custom FW and full hardware access.
ps: As of now I think it will be too hard to extract a lot of performance from the device
.
Last edited by guily6669,
First I dont have much knowledge, so its just my opinion!
I don't think they are any similar... It is based on the Nvidia Shield TV chip so its ARM like most mobile devices and I think it will be way better to code for the Switch because it supports Vulkan, Cuda cores.......
Also the GPU is considerably better than the PS3, but the CPU, I have no idea, I have the Nvidia shield tablet which uses a older chip with an even crapier CPU and the CPU power is not up to the GPU, its very weak and I'm still to see any game running better than a PS3 top graphics\physics game even though Nvidia claims the GPU is better than a X360 and PS3.
ps: GPU is better than PS3\X360, but the CPU I don't have a clue if it will be as powerful as the PS3 (but I'm thinking its still a bit far from the PS3 CPU performance). However I think it will surely be way easier to code for this and if using Vulkan API + Cuda well implemented might help increasing performance for emulators (but I have no clue, it's just my opinion and thinking that in PC we use Cuda processing to increase performance of the program
Misunderstood the OP's Question here I think.
OP wasn't asking if the hardware was similar, he was asking about fragmentation within the Homebrew scene and potential lessons we can learn from the PS3.
The PS3 required you to be on a particular firmware to run any user code. The Switch is in a similar position. As it stands due to the fact that we have an exploit in 3.0.0 I suspect that this will be the entry point that the majority of Devs will be working on.
Code/firmware/hardware revision levels is nothing new. It is hardly ideal but it has been worked with in the past. We are already seeing the "well it is not going to be any good" notions surfacing -- I am reminded of the discovery of JTAG for the 360, it happened mid year and did not get worked up into something capable of doing commercial games until around Christmas. We saw loads of people stop holding off (it was a good year for games) around September/October, and then there was not a perfectly timed banwave but a good one. It would be a few years before RGH appeared so the amount of whining that went on as people sat there with their banned 360s (and apparently online was absolutely essential for a lot of people thus them deeming them worthless, sadly I never got it together to buy all those banned 360s for nothing).
I have previously pondered what a homebrew/hacking scene needs.
http://gbatemp.net/threads/buy-switch-now-or-hold-off.482813/#post-7551355
Still if you want to defy the odds then
Interesting hardware... it is not really. It is a sort of high end* phone/tablet with a control pad, and not even a great one in the case of the joycons. On the flip side it is not a complete turn off.
*in terms of grunt it is up there, however in terms of architecture it appears it is a sort of "money no object" project about 3 years ago. This is not quite as bad as it was for the 3ds (a really odd 3d setup) and DS (a completely custom 3d setup that you had to learn to handle) but I will note it as part of this.
Power wise I am not seeing it break out of the 16 bit and earlier rut for emulation (mainly as PC has barely decided to drag itself out of that rut), I can see N64, PS1 and gamecube stuff happening in the longer term but straight recompiles is a harder prospect. To that end nobody is likely to keep their switch on an older firmware (more later) for that, and may even look back at their PSP or a raspberry pi or something. If you can get all the usual suspects from that era (offerings from sega, nintendo, atari, maybe amiga and then start on the edge cases) going in reasonable speeds and resolutions (the screen should handle it well enough), possibly with the perk of local two player if you reckon the joycons will manage it you will gain some interest.
Interesting commercial games... despite what fanboys are gushing it is not there and the upcoming releases are pretty barren, as is the commitment from the big pubs and devs, I shall await the turn of the tide on Mario like it did for Zelda (we saw the cracks from the start, I am not sure it will end up as another Sunshine but it ain't far off). The library is not abysmal but yeah... oh and we are paying for online this time as well. If you can get ROM hacking going on for it, and also provide the option to play new games (I am not seeing any great black box type setups preventing you from virtualising the newer loaders, this is not to say you won't have to comb through the code and look for every check and trip they might employ but if a handful of people can do it for the 360 against Microsoft you can certainly take on the clowns at Nintendo's security department) it will help.
Software. Get those libraries, interpreters and compilers in -- as great as the GBA and DS homebrew scenes were you have to remember they were truly good for the time, and as soon as something better (in this case IOS about 2/3 of the way through the DS) came along they were dropped hard. I like messing around with C and assembly but it is not a terribly practical way to write cool applications you just want to have, and while I massively respect those doing the lovingly hand crafted emulators for the GBA and DS when the option arose to half arsed port something written entirely in generic C and patch a bit to get it running then that became the dominant method and one used by most.
Libraries start with SDL which should be perfectly doable (the DS was not quite there, PSP did fine though and the switch should surely be able to handle it), you almost certainly want python, lua will doubtless turn up before long. Java then... the best I can see happening is a workable version of android gets ported and with that comes Java, without that then you could probably get something resembling a real version of Java going on (compared to http://gbatemp.net/threads/java-for-ds.57263/ we saw on the DS).
From there you can go for the higher level game makers, things like unity and whatever else.
Likewise if you want to port the usual suspects from http://osgameclones.com/ then do it to it. Alternatively figure out the sorts of libraries such things will need and start on them, or at the very least figure out if any of the code there will need redoing (for quake on the DS it was noted that the DS' floating point options were... not great and that held things back a bit). Maybe also check the popular mods to see if they have a chance of running.
Media is a good thing to have. Don't know if you can get a mplayer port + basic frontend all handling scene style videos (AVI XVID + MP4 H264 + MKV H264 + whatever the cretins in the anime world are doing this week (is that 10 bit nonsense still going on?), though subs to vsfilter grade are a good choice), maybe also an ad blocking youtube player, but that would be my baseline.
Probably pie in the sky but get a serviceable web browser going on and things will be good. You live in a good time for it as web development got nerfed in favour of mobile phones and tablets compared to the DS era which caught the rise of javascript and CSS (+ "rich media" + flash + urgh...). If you can get browser web dev to the point where the high end software development options mean web devs can fart out fun little games aimed at joycons or whatever then that will not hurt.
Mentioned in the early link but money changed the game entirely. Barring an android port I am not sure what you can get going on for a paid money shop, and adverts would be an odd one to see. If you want to get a download service going on then know most of your theoretical competition (which is to say all the other open source handhelds and such devices) will tend to have such a service.
Frankly I imagine I am going to be looking back at the 3ds fondly for this one (and I did not care for the 3ds) -- give me a nice writeup of your hacks and I will enjoy reading it but I am here for software wherever it comes from, not for the sake of hacking a console.
I have previously pondered what a homebrew/hacking scene needs.
http://gbatemp.net/threads/buy-switch-now-or-hold-off.482813/#post-7551355
Still if you want to defy the odds then
Interesting hardware... it is not really. It is a sort of high end* phone/tablet with a control pad, and not even a great one in the case of the joycons. On the flip side it is not a complete turn off.
*in terms of grunt it is up there, however in terms of architecture it appears it is a sort of "money no object" project about 3 years ago. This is not quite as bad as it was for the 3ds (a really odd 3d setup) and DS (a completely custom 3d setup that you had to learn to handle) but I will note it as part of this.
Power wise I am not seeing it break out of the 16 bit and earlier rut for emulation (mainly as PC has barely decided to drag itself out of that rut), I can see N64, PS1 and gamecube stuff happening in the longer term but straight recompiles is a harder prospect. To that end nobody is likely to keep their switch on an older firmware (more later) for that, and may even look back at their PSP or a raspberry pi or something. If you can get all the usual suspects from that era (offerings from sega, nintendo, atari, maybe amiga and then start on the edge cases) going in reasonable speeds and resolutions (the screen should handle it well enough), possibly with the perk of local two player if you reckon the joycons will manage it you will gain some interest.
Interesting commercial games... despite what fanboys are gushing it is not there and the upcoming releases are pretty barren, as is the commitment from the big pubs and devs, I shall await the turn of the tide on Mario like it did for Zelda (we saw the cracks from the start, I am not sure it will end up as another Sunshine but it ain't far off). The library is not abysmal but yeah... oh and we are paying for online this time as well. If you can get ROM hacking going on for it, and also provide the option to play new games (I am not seeing any great black box type setups preventing you from virtualising the newer loaders, this is not to say you won't have to comb through the code and look for every check and trip they might employ but if a handful of people can do it for the 360 against Microsoft you can certainly take on the clowns at Nintendo's security department) it will help.
Software. Get those libraries, interpreters and compilers in -- as great as the GBA and DS homebrew scenes were you have to remember they were truly good for the time, and as soon as something better (in this case IOS about 2/3 of the way through the DS) came along they were dropped hard. I like messing around with C and assembly but it is not a terribly practical way to write cool applications you just want to have, and while I massively respect those doing the lovingly hand crafted emulators for the GBA and DS when the option arose to half arsed port something written entirely in generic C and patch a bit to get it running then that became the dominant method and one used by most.
Libraries start with SDL which should be perfectly doable (the DS was not quite there, PSP did fine though and the switch should surely be able to handle it), you almost certainly want python, lua will doubtless turn up before long. Java then... the best I can see happening is a workable version of android gets ported and with that comes Java, without that then you could probably get something resembling a real version of Java going on (compared to http://gbatemp.net/threads/java-for-ds.57263/ we saw on the DS).
From there you can go for the higher level game makers, things like unity and whatever else.
Likewise if you want to port the usual suspects from http://osgameclones.com/ then do it to it. Alternatively figure out the sorts of libraries such things will need and start on them, or at the very least figure out if any of the code there will need redoing (for quake on the DS it was noted that the DS' floating point options were... not great and that held things back a bit). Maybe also check the popular mods to see if they have a chance of running.
Media is a good thing to have. Don't know if you can get a mplayer port + basic frontend all handling scene style videos (AVI XVID + MP4 H264 + MKV H264 + whatever the cretins in the anime world are doing this week (is that 10 bit nonsense still going on?), though subs to vsfilter grade are a good choice), maybe also an ad blocking youtube player, but that would be my baseline.
Probably pie in the sky but get a serviceable web browser going on and things will be good. You live in a good time for it as web development got nerfed in favour of mobile phones and tablets compared to the DS era which caught the rise of javascript and CSS (+ "rich media" + flash + urgh...). If you can get browser web dev to the point where the high end software development options mean web devs can fart out fun little games aimed at joycons or whatever then that will not hurt.
Mentioned in the early link but money changed the game entirely. Barring an android port I am not sure what you can get going on for a paid money shop, and adverts would be an odd one to see. If you want to get a download service going on then know most of your theoretical competition (which is to say all the other open source handhelds and such devices) will tend to have such a service.
Frankly I imagine I am going to be looking back at the 3ds fondly for this one (and I did not care for the 3ds) -- give me a nice writeup of your hacks and I will enjoy reading it but I am here for software wherever it comes from, not for the sake of hacking a console.
no because both systems are completely different its like asking if there is anything your iron can do to get bbc radio 2 lol
I know that the architecture is different. That's not what I'm asking. I'm asking if there are things we can learn from it. Things to advance homebrew. On PS3, they figured out how to mod games, how to pirate, and how to modify system files to unlock locked features (for instance, PS2 backwards Compatability, other OS, and many high level systems.) Flog is one such higher feature that is "locked" and as it stands right now, we have basic access on one specific version. On PS3, if you don't run 3.5.5 firmware, you have to get a device that many people probably know called an E3 flasher. Why haven't people thought of making a hardware mod to downgrade firmware, and install CFW... cuz Wii u needed cfw to run many homebrew apps. Something tells me that the switch will need a modified update package in order to get homebrew...
--------------------- MERGED ---------------------------
Precisely what I was asking. Thanks for clarifying it.
Indeed. I don't have a switch, but I want to try and help by looking back at times when it was successful. I look for lessons, things that we can learn from. And by learning from our past, we can influence our future...
At least... that's how I see it...
I'm not even wasted yet. Give me a few and I'll respond to that with a "WOAH".
I would advise not going through my post history if that rates as a wall.
the rule would be common sense, do not update console and wait for the scene to release its findings.
a lot of people are rushing to install updates when ever any group says its found an exploit for xyz firmware.
these people need to be patient because these "hacks" are nothing to be excited about, its not running fully pledged homebrew such as emulators or backup loaders.
so it makes no sense why people are updating.
once the system is fully exploited and we start seeing homebrew, then people should look at updating the console, chances are if a higher firmware is hacked, then it could be possible to do so on lower firmwares.
Similar threads
-
- Portal
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
@
Xdqwerty:
something about ea reverse engineering the genesis so they could make their games there+1 -
-
@
BakerMan:
if they get permission from 4j (the studio who helped put minecraft on consoles), they could put the console edition tutorial worlds in the game, as seeds+1
and i mean all of them (if not, maybe a few highlights, like TU1, TU31, and TU69) -
-
-
-
-
-
-
-
