|
Post by Arkhan on Nov 2, 2019 21:22:34 GMT
If that's true, that might mean... we can just fix a real bug in Squirrel that exists (but works OK somehow?) and went unreported because nobody uses the damn thing in 3.21 and nothing actually needs changed besides that?
Though, the BankThing(TM) is still at large. DarkKobold said he experienced it. I have no other details. I've never experienced it in my games.
It's possible there is an actual issue when a song itself crosses the page boundary.
|
|
|
Post by gredler on Nov 2, 2019 21:39:11 GMT
It's possible there is an actual issue when a song itself crosses the page boundary. Yeah this is what I was thinking and half hoping. All of the mml2pce content we have implemented I made haphazardly while trying to make our game not muted. It could easily have issues.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 2, 2019 23:38:20 GMT
Well the PC Engine community is in a sad state of affairs right now. This whole Squirrel business is hilarious and seems to originate from the inability of PCE folks to talk to each other like normal people and cooperate, which is nothing new to be honest. It's been a long time since Elmer's HuC update broke Squirrel and we're not even sure what the problem actually is here, or if there was any issues in the first place. Nothing here is ever straightforward, it always feels like people are taking passive aggressive jabs at each other and other subtle stuff that goes unnoticed to the general public.
I don't think I've ever met a community as dysfunctional as this one and I often times feel like it's a miracle anything comes out of it. It's very easy to see why there really isn't any interest in the platform aside from the usual suspects and Gredler/DarkKobold. As for myself I'm only bothering with the PCE because of the sweet scanline interrupt and VRAM access functionalities of the console (and also because the PCE is the only console I enjoy owning hardware/software of). I'll probably just move back to NES development once my game gets done or I give up on it because quite frankly everything PCEFX related right now is a buzzkill.
|
|
|
Post by dshadoff on Nov 3, 2019 0:43:23 GMT
I've been developing things for the PC Engine for nearly 30 years now, and outside of Japan there has never been a large enough number of developers to: a) do the things that you (and some others) seem to expect. b) be called a 'community'. *Maybe*, at times, it might be called a small team.
It's a miracle that HuC exists at all, let alone exists in its current state.
If I think back, I don't recall a time when there were more than about 3, maybe 4, simultaneously-active developers contributing to open-source projects such as HuC. (Though, there have often been similar or even larger numbers of people seeking to use those tools.)
So, saying that people being angry with each other is driving people away from the community is a non-sequitur, as there were never very many people to begin with. Expectations placed on those people have always been high, which has probably caused some disagreements and ill will. And there have been passers-by who somehow think that it should be a metropolis of development, and are disappointed to find that it isn't as well-developed as they imagined it to be, because it was all built by just a few guys in their spare time.
I've seen lots of other "communities" more dysfunctional and less productive. All it takes is short tempers and high expectations.
|
|
|
Post by Arkhan on Nov 3, 2019 1:15:37 GMT
Dave is right in that there's never been a large active developer base.
You really can see just a handful of active names in the before times of TGHack/etc., and then when I came around, it was mostly myself and Rover while some people tinkered around a bit on forums or chatrooms with small things or experiments.
The one constant I've always seen is dumb horseshit like this, though. We fought this MML war in 2008 already with almost the same cast of people. I was told it was a stupid idea then, and now we're trying to get that stupid idea to work with another stupid idea so I'd say we're right on track! lol
Even in the beforetimes that I was not around for, I caught wind of "names and things we shouldn't talk about now" and got yelled at for asking about them.
So Punch isn't wrong in that regard. That level of bullshit can't help. It makes people hard-pass on this shit early on.
This shit occurs in most communities though. The C64 one has it for sure. DK showed me some next level stupidity in the SNES one.
On one hand it helps weed out the people who aren't serious enough. On the other hand, it weeds out basically everyone.
I'm pretty confused this time around. Normally I am yelling at someone or walking around with a flamethrower, but all I did here was candidly ask what was up based off of what I had heard.
Somehow that became my fantasy filled "elmer changed everything and hacked my utility" extravaganza, though.
lol.
I don't even have to do anything to receive horseshit at this point. So honestly, I am going to say this with 100% certainty, that me saying elmer should pull his head out of his ass is 100% accurate in this instance. That was not the ideal reply to what was going on and missed the goddamn point.
My reply isn't really ideal either, but I am tired of this shit so it is what it is. Stupid shit happens and people tend to say nothing, or I speak up or give a different view/advice/whatever, and then it becomes "lol that arkhan, he's such a fucker!". *mute/block/ignore/realize my advice was kinda cool at a later date and wasn't some direct attack*.
People go the distance to give what I often see is empty encouragement and praise/cheerleading, but nobody ever bothers to speak up at stupid shit.
Some of this shit with the development stuff, as smart as we all must be if we're programming this crap, is completely fucking stupid. And then people can start arguments or act like cunts, but suddenly curl up in a ball when it's coming back at them.
Kinda dumb all around.
This message was brought to you by MML.
Endorsed by Yuzo Koshiro
Distributed by Deez Nuts.
|
|
|
Post by dshadoff on Nov 3, 2019 1:28:30 GMT
Quick question:
Is squirrel broken in Uli's post-3.21 version of HuC ? How about Artemio's ?
A lot of stuff changed in those, and I am thinking that Squirrel *could* have broken in those before elmer grabbed it (unless I'm wrong and it has been tested as working on those versions). But several times in this thread, elmer's name is associated with the broken version, implying that the breakage happened under his care. This might be what is irritating him.
Since the major comparison seems to be "3.21 versus current", I'm just not clear on whether the intermediary versions have been tested. And of course, this is a crucial step in isolating what exactly is broken.
I do agree with something else you said recently (I don't recall whether it was this thread or another) - that enough major change took place after 3.21, that it should have had a major version number promotion to 4.x .
|
|
|
Post by Arkhan on Nov 3, 2019 1:43:09 GMT
Quick question: Is squirrel broken in Uli's post-3.21 version of HuC ? How about Artemio's ? Not 100% certain. I did not use either of those as nothing in my projects built cleanly when I tried, and then I said "Fuck this what kind of 3.xx update doesn't just kinda work. This is dumb.", and went back to good ol' 3.21. The one true HuC to rule them all. That is because Elmer took it upon himself on the other forum before it exploded to sort of reignite the stupid MML war, probably out of ignorance about how that was already a thing, and a soreish subject, and then heavily associated himself with post 3.21 via his fork and progress/MML alternative, complete with all of us arguing and nothing great happening. It's entirely a possibility that the breakages were caused before that and he simply assimilated them, everything was possibly made worse through all of the interactions we had, and then he also did not up the version number which takes us back to that point. Me either. It's probably a combination of sorts. It might be the case that Squirrel (outside of the _ vs __ breakage that has a workaround?) just needs that PSG_BANK thing fixed and always has. I'm still not sure why our stuff works if that is wrong. That's scary.
|
|
|
Post by turboxray on Nov 3, 2019 1:53:28 GMT
It's very easy to see why there really isn't any interest in the platform aside from the usual suspects and Gredler/DarkKobold. As for myself I'm only bothering with the PCE because of the sweet scanline interrupt and VRAM access functionalities of the console (and also because the PCE is the only console I enjoy owning hardware/software of). I'll probably just move back to NES development PC-Engine has always been just a passing curiosity except for like maybe the handful that have stuck around. There have been other mini pce communities, and mostly silos, but none of them with any real intention other than to make something (an intro, or demo, or little experimenting) and then moving on (notch on belt). It's not because of the smaller 4-5 handful the mess around with technical stuff, hack for translations, or put out some games. There's just no interest in it really. Hell, I even remember hanging out on RHDN telling people they could take their favorite NES game and enhance the graphics and sound, all the while keeping the original NES code (the nes2pce stuff)... no one cared. If I had done the same for the snes, people would be all over it like flies on shit. Even Genesis dev is similar, just larger/more interests so it looks like there's a thriving community - but besides a few members, the rest are just a revolving door. But yeah, the nesdev forum is really nice. I rarely post over there, but I read everything haha. Love seeing all the project people are working on. The problem with the nes, is that everything's been done on that system. There's a lot of interesting and new stuff that can be done on the PCE, that hasn't really be done commercially or with only little bits here and there. To me, that's the main attraction (besides being super into the system back in the day, even though I owned a snes and genesis).
|
|
|
Post by Arkhan on Nov 3, 2019 1:59:42 GMT
One thing that kills us is the lack of good tooling, and now that Kickstarting projects and shit (Thanks Watermelon cunts) is a thing, people will tend to gravitate to the Genesis/Dreamcast because it's easier and they can hit a larger mess of people who will buy literally anything that runs on their old machine.
So now we are probably even less likely to get a flood of people. We already weren't. lol
|
|
|
Post by DarkKobold on Nov 3, 2019 2:06:46 GMT
This shit occurs in most communities though. The C64 one has it for sure. DK showed me some next level stupidity in the SNES one. Just wanted to chime in - the SNESdev community is the biggest bunch of elitist fucks. If you can't program in ASM, you aren't good enough to code for their precious system.
This sound issue has been frustrating for gredler and myself, but all around, the community has been extremely helpful. I have ASM zsort thanks to touko. We have many HuC fixes thanks to elmer. We'll have SFX in our game thanks to Arkhan. We may get some extra bits and bytes in our huCard to fill, thanks to turboxray. Like, if that isn't a community effort, I don't know what is.
Yes, there is some infighting, but I'll take 10 elmer vs. Ark type feuds over the absolute shithole that is SNESdev.
|
|
|
Post by Arkhan on Nov 3, 2019 2:17:14 GMT
ok to be fair if you can't program in ASM your shit probably isn't going to work on the dumbassed SNES anyways lol.
There were some missteps that created a dumb rift in the development community. Mainly all this HuC shit not playing nice with what was already there, meaning it should've been 4.0'd.
The forum implosion didn't help, and nor did PCEngineFans coming back in a non read online state. It should've just come back readonly since this place already sprouted up.
I barely want to do 1 forum now let alone look at two.
at least during the course of this trainwreck, Koshiro released his MML crap.
VINDICATIONNNNNNNNNNNN
|
|
|
Post by spenoza on Nov 3, 2019 12:22:27 GMT
Didn’t Koshiro track to MML, though?
|
|
|
Post by spenoza on Nov 3, 2019 14:16:01 GMT
Also, this is about as close as I think we are capable of getting to a kumbaya moment. We better get back to grumpy shitposting before folks think we’ve gone soft /sarcasm
|
|
|
Post by elmer on Nov 3, 2019 20:16:11 GMT
Quick question: Is squirrel broken in Uli's post-3.21 version of HuC ? How about Artemio's ? Yes to both ... because Uli changed the old HuC single-underscore "_ax", etc variable names in order to reduce namespace pollution between HuC and asm. Also, as the guy voluntarily spending months of his spare-time trying to make significant improvements to the compiler, he modified the compiler's internal startup.asm, and Squirrel ships with a customized version of the old HuC 3.21 startup.asm which breaks Uli changes and doesn't work with Uli's v3.98. I do agree with something else you said recently (I don't recall whether it was this thread or another) - that enough major change took place after 3.21, that it should have had a major version number promotion to 4.x . Uli changed the version number from 3.21 to 3.98 ... a pretty significant bump. My fork modifies the version number to 3.99 to show that is it different to Uli's. People are either willing to make few minor changes to their code in order to take advantage of all of the work that Uli put into fixing bugs and improving the functionality of HuC ... or they're not, in which case they can keep on using HuC 3.21 with its existing-and-sometimes-known set of limitiations and bugs. I really don't understand this notion of whether the updates/changes/fixes might have somehow been more accetible if Uli has bumped the version number up to "4.00", because that still wouldn't have changed whether someone saw the benefits of using the new version as worth the cost of making some small changes to their existing code. A lot of stuff changed in those, and I am thinking that Squirrel *could* have broken in those before elmer grabbed it (unless I'm wrong and it has been tested as working on those versions). But several times in this thread, elmer's name is associated with the broken version, implying that the breakage happened under his care. This might be what is irritating him. Yes, that, and Arkhan implying in the first post as this thread that I might have childishly "deliberately broken" Squirrel on HuC because of some nefarious vendetta against either him or MML-users in general. The reality was that instead, I had actually spent a bunch of my time fixing Squirrel to make it work in both the old and new versions of HuC, without requiring future modifications of internal compiler files, and so avoiding future breakage. There is definitely an argument to make as to whether it was a step-too-far that I had the temerity to request that he add an argument to his closed-source mml2pce.exe so that it could spit out either the old single-underscore or new double-underscore variable names ... but I'll let other programmers decide if that was a horrible sin.
|
|
|
Post by Arkhan on Nov 3, 2019 20:46:25 GMT
Yes to both ... because Uli changed the old HuC single-underscore "_ax", etc variable names in order to reduce namespace pollution between HuC and asm. Also, as the guy voluntarily spending months of his spare-time trying to make significant improvements to the compiler, he modified the compiler's internal startup.asm, and Squirrel ships with a customized version of the old HuC 3.21 startup.asm which breaks Uli changes and doesn't work with Uli's v3.98. And that is our fault for operating under the assumption HuC was basically dead and going nowhere. 3.21 had saw no updates for what like 7 years? Promises of things never happened. We needed something to happen so we and others could have chiptunes, so that's what we did. It's a semantic thing. Seeing a 4.0 implies "lol there is going to be breaking changes. tread carefully!". 3.xx generally *doesn't* carry that same implication, and when the reality is that it DOES cause breakages like that, the programmers goofed. Honestly, you did yourself no favors with the way you name-dropped us in various threads and bashed MML while admitting you purposely instigated and stirred the pot because you thought it was fun or whatever. You reap what you sow man. When you do that, and somehow find a way to sort of pout about us in random threads, it paints a specific picture about your motives, and people ran with it. It wasn't a direct accusation, but more of a question based off of the way things went down and many of our assumptions. You never confirmed or denied the thought. The reality is, you kind of bullheadedly showed up and shit on a process that worked without knowing the full backstory that there was already a great MML war 6ish(7?) years before you came around. It was "been there done that" for us, and a new playground for you. I think that is the perspective you didn't understand and maybe still don't. We already had the MML works and it's not that bad vs Trackers/Samples/MML is for retards thing. It wasn't fun, and nobody wanted to do it again. Coming in and basically restarting that mess made it really hard to give a shit about HuC updates because you were being a pain in the ass. Asking to modify and add flags to our thing after being said pain in the ass didn't help. It's really not what you did. It's the way you went about it. It seemed self aggrandizing, and possibly subversive. Given that you're helping DK pretty significantly from what I understand, you too realize the usefulness of utility pipelines instead of compiler crutches, and HuC has made some progress in the right direction, I'm now operating under the assumption that you're basically self unaware of this and it's not intentional. It's just annoying. and with all that said, the bank-hop thing is still at large. Maybe DK can recreate it for it to be looked at. the interrupt/timer thing might be OK? As long as you can set it to the timer mode and changing a songs tempo (the T parameter in any song) works, it's probably not broken.
|
|