| - isoriano_ (QUIT: Quit: Leaving) (~isoriano@5.147.209.166) | 00:26 | |
| - RobDangerous (QUIT: Ping timeout: 245 seconds) (~Thunderbi@p200300ec8f3e1c00fded8ecffa28f10d.dip0.t-ipconnect.de) | 02:25 | |
| + RobDangerous (~Thunderbi@p200300ec8f3e1c00b1df07b493a316b1.dip0.t-ipconnect.de) | 08:16 | |
| - xet7 (QUIT: Remote host closed the connection) (~xet7@user/xet7) | 12:00 | |
| + xet7 (~xet7@user/xet7) | 12:01 | |
| - js (QUIT: Quit: Bridge terminating on SIGTERM) (~js@2001:470:69fc:105::232) | 12:36 | |
| - emiespo[m] (QUIT: Quit: Bridge terminating on SIGTERM) (~emiespoma@2001:470:69fc:105::92a) | 12:36 | |
| - apolkosnik[m] (QUIT: Quit: Bridge terminating on SIGTERM) (~apolkosni@2001:470:69fc:105::ff7) | 12:36 | |
| + js (~js@2001:470:69fc:105::232) | 12:39 | |
| + apolkosnik[m] (~apolkosni@2001:470:69fc:105::ff7) | 12:41 | |
| + emiespo[m] (~emiespoma@2001:470:69fc:105::92a) | 12:41 | |
| _Bnu | Ah, your new video stuff doesn't have any of the things I added for the panning, so it's probably not going to work with panning, no. I'll try to rework it. | 14:58 |
|---|---|---|
| _Bnu | I also need to change the driver back to some chip/palette type other than the ZZ9000, so it actually works on old Picasso96 again. | 14:59 |
| _Bnu | I think I'll just start over with the panning, everything kind of messed up now, lolb. | 15:01 |
| _Bnu | Yeah, somehow the RTG panning commit to drivers got orphaned or something | 15:03 |
| _Bnu | So I'll just make a new one. | 15:03 |
| mntmn | oh shit, sorry for that | 15:09 |
| mntmn | also thanks for the chip type hint, i again forgot about this | 15:09 |
| Jope | dare I make a small scandoubler vertical refresh request | 15:17 |
| Jope | must send bribes to sweden.. | 15:17 |
| mntmn | haha | 15:27 |
| _Bnu | Just sell me a recapped A4000 power supply that'll last for a few more decades and I'll consider it . . . | 15:27 |
| mntmn | btw how did that atx adapter turn out? | 15:33 |
| _Bnu | I mean, it's here, and it works... but I don't like it, lol. | 15:44 |
| _Bnu | https://cdn.discordapp.com/attachments/789807768554831902/870300771621675008/20210729_154421.jpg | 15:44 |
| _Bnu | lol | 15:55 |
| _Bnu | The grayscale "powerup" doesn't even work in the AGA version of James Pond 2. | 15:55 |
| pasik | asus powered amiga | 15:57 |
| pasik | never seen one before! | 15:57 |
| mntmn | _Bnu: i see! a4000 kinda back in business?? | 16:14 |
| mntmn | thinking if i should order another buddha from jens | 16:14 |
| mntmn | or write some autoboot code | 16:14 |
| _Bnu | Isn't the Buddha kind of crap compared to the FastATA... | 16:20 |
| _Bnu | Like, Z2 only or something. Not that you're super likely to get the up to 16MB/sec from the FastATA. | 16:20 |
| _Bnu | On the other hand, the Buddha works on everything. | 16:22 |
| _Bnu | But why would you be running a Z2 system using anything but a PiStorm anyway... got PiSCSI there... | 16:23 |
| mntmn | no i mean for the 4000 | 16:30 |
| mntmn | sorry i mean for the 3000 | 16:30 |
| mntmn | the 4000 doesn't even work | 16:30 |
| mntmn | also i'm not really concerned about the speed, as long as it's stable and i don't have to wait forever... ;3 | 16:30 |
| _Bnu | Ah yeah, not sure if the FastATA 4000 works in the A3000 or not. | 16:32 |
| _Bnu | But it's the least annoying thing I've used. | 16:32 |
| _Bnu | Okay, rebuilt my firmware and Rectfill Pattern and Text/Con-output are back to normal again... | 16:33 |
| _Bnu | I must've disabled them for some reason at some point. | 16:33 |
| _Bnu | Or maybe I added a crapton of debug output... | 16:34 |
| _Bnu | mntmn, why is the ZZ9000 slower at WriteChunkyPixels than the PiStorm... can you fix Zorro III and make it 150MB/sec bandwidth... | 16:36 |
| _Bnu | https://media.discordapp.net/attachments/836238000148578324/869949642010427442/unknown.png You can tell that I didn't add any cheat codes to RectfFill Pattern, haha. | 16:39 |
| _Bnu | https://cdn.discordapp.com/attachments/789807768554831902/870315187587129384/unknown.png And a fresh set of my good old cheat scores from my magical 50MHz 68060. | 16:42 |
| mntmn | crapton 3000 | 16:47 |
| mntmn | we can just move musashi onto the zz core 2 | 16:48 |
| mntmn | then your bandwidth problem is gone... | 16:48 |
| mntmn | we can construct a mini draco in there | 16:49 |
| _Bnu | Yeah, except then it's there for the entire rest of the Amiga instead, haha. | 16:49 |
| mntmn | well, who needs those old amiga chips anyway | 16:49 |
| _Bnu | True, true. | 16:49 |
| _Bnu | I mean, I would love to replace the Buster with something that doesn't get in the way of every expansion plan anyone ever has for the A3000/A4000... | 16:50 |
| _Bnu | A4000 is happy at 1920x1080 again. | 16:52 |
| mntmn | cool | 16:52 |
| _Bnu | No intermittent video cutouts, no weird HDD shutdown problems. | 16:52 |
| mntmn | that is nice | 16:52 |
| _Bnu | P96 3.01 is still fucking with my palette, but that's my fault for not updating it yet. | 16:53 |
| mntmn | fuxxorz | 16:57 |
| _Bnu | Also, I must've indirectly sold quite a few copies of P96 for Jens by now... you'd think he could afford sponsoring me with a lifetime license for it or something so I don't have to re-buy it, lolb. | 16:58 |
| mntmn | lol yes | 16:58 |
| mntmn | scrooge mc duck | 16:58 |
| mntmn | it's our fault because we didn't buy p96 | 16:59 |
| mntmn | i mean, the whole package | 16:59 |
| mntmn | i am really thinking atm: accelerated cpu + rtg makes sense and unaccelerated cpu + amiga chipset makes sense. in which cases do you need accelerated cpu + original chipset? for demos mostly? | 17:01 |
| mntmn | i think personally my amiga would be nice if the games used the most compatible slow cpu and workbench rtg stuff would be as accelerated as possible | 17:02 |
| _Bnu | Man, I would hate trying to play a lot of regular games on a standard 68000, haha. | 17:04 |
| _Bnu | But that's just because my first Amiga was a 1200. | 17:04 |
| mntmn | haha well ok lets say 020 or 030 | 17:04 |
| mntmn | but 060 + old games is mostly a bad combo unfortunately | 17:05 |
| mntmn | so, my a3000 works better for gaming with the default 030 than with the warpengine | 17:05 |
| _Bnu | 060 is bad for pretty much everything, yeah. It just happens to be the fastest CPU. | 17:05 |
| mntmn | the warpengine is only there to make workbench fast... which is maybe a bit nonsensical | 17:05 |
| mntmn | so if i could run workbench and intuition stuff on zz9000... | 17:05 |
| mntmn | ...my life might be improved | 17:06 |
| apolkosnik[m] | Just got to start using WHDLoad for stuff rather than ADFs from Gotek | 17:06 |
| mntmn | (at least once a month when i turn the machine on) | 17:06 |
| _Bnu | I mean I'm using WHDLoad stuff. | 17:06 |
| mntmn | me too, whdload only | 17:06 |
| mntmn | i don't even have a floppy nor gotek | 17:06 |
| mntmn | the floppy connector is E M P T Y | 17:06 |
| mntmn | i wonder what custom chip stuff amigaos/workbench really needs to live | 17:07 |
| _Bnu | All of them*. | 17:07 |
| mntmn | haha | 17:07 |
| mntmn | that's what i thought | 17:07 |
| _Bnu | Workbench can't live without IRQs, CIAs or blitter, even in RTG mode. The blitter part is of course not entirely true, you can make P96 do all of it in software instead. | 17:08 |
| mntmn | blittner | 17:08 |
| mntmn | ok interesting. | 17:08 |
| _Bnu | But from what I can tell, it doesn't work quite properly anyway. | 17:08 |
| _Bnu | Despite having been in development for 20+ years. | 17:08 |
| mntmn | schnirectory schnopus | 17:09 |
| _Bnu | But I would obviously rather have a fantasy CPU like the PiStorm instead of a 68060, haha. It's just not very feasible to run it on Zorro because it's so slow. | 17:10 |
| _Bnu | No working cache = no cache problems. | 17:10 |
| mntmn | :( | 17:10 |
| mntmn | i see! | 17:10 |
| _Bnu | I mean you could do bus mastering in Z2 mode. | 17:11 |
| _Bnu | At ~2.1MB/sec access to the entire rest of the computer. | 17:11 |
| _Bnu | Great for something that doesn't need to interact with it much at all, like a PC busboard. | 17:11 |
| _Bnu | Mega awful for something like an accelerator. | 17:12 |
| mntmn | i wonder if good enough for running workbench | 17:13 |
| mntmn | and yield to the real cpu for gamez | 17:13 |
| mntmn | so it would be a dual core amiga, a workbench/rtg one in the zz and the real metal one in... not the zz | 17:14 |
| _Bnu | Possibly, but I'm not sure how that'd work. It's not like the Amiga file system is very friendly to having two separate things accessing it at the same time... | 17:19 |
| mntmn | mhm | 17:21 |
| mntmn | fwiw USB would be a fast path from pistorm to the zz, or ethernet | 17:25 |
| mntmn | for example, zz memory could present as a mass storage device | 17:26 |
| mntmn | there is some xilinx code floating around on making zynq a USB gadget instead of a host | 17:26 |
| mntmn | (xilinx doesn't actually have any docs on using it as a host, that's why i used u-boot code for that) | 17:27 |
| mntmn | so if you plug that into pistorm you would have theoretical 60MB/s access | 17:27 |
| mntmn | you could then mmap that device and plug this into your musashi memory map | 17:28 |
| - RobDangerous (QUIT: Ping timeout: 252 seconds) (~Thunderbi@p200300ec8f3e1c00b1df07b493a316b1.dip0.t-ipconnect.de) | 17:31 | |
| apolkosnik[m] | Wow, ZorroUSB! | 17:32 |
| + RobDangerous (~Thunderbi@p200300ec8f3e1c00b1df07b493a316b1.dip0.t-ipconnect.de) | 17:37 | |
| Claude | if that stupid usb latency wouldn't be there . | 17:44 |
| mntmn | i think zz9000 has the correct signals to drive Z3 as a master, but not Z2 (OWN is not connected) | 18:00 |
| mntmn | i wonder if i could write to chipset addresses over the z3 bus... | 18:01 |
| _Bnu | Now you just have to figure out exactly how Z3 bus mastering is supposed to work, since no one ever used it... | 18:07 |
| mntmn | naja the theory is i assert BR, then buster should give me BG at some point, and then i'm allowed to run up to 8 cycles... | 18:09 |
| Claude | Z3 spec explicitly mentions chipram . but not sure about Chipset | 18:15 |
| mntmn | so i would put an address on the bus and READ or nREAD and assert FCS, right... and then wait for DOE? | 18:16 |
| mntmn | no, wait for DTACK i guess | 18:16 |
| mntmn | lulz i guess i cannot sample DTACK | 18:17 |
| mntmn | FC0,1,2 are also not connected | 18:29 |
| mntmn | > Alternate function code registers, SFC and DFC, contain 3-bit function codes. Function codes can be considered extensions of the 32-bit linear address that optionally provide as many as eight 4-Gbyte address spaces. | 18:31 |
| - RobDangerous (QUIT: Read error: Connection reset by peer) (~Thunderbi@p200300ec8f3e1c00b1df07b493a316b1.dip0.t-ipconnect.de) | 18:51 | |
| + RobDangerous (~Thunderbi@p200300ec8f3e1c00b1df07b493a316b1.dip0.t-ipconnect.de) | 18:51 | |
| _Bnu | Yeah, FC isn't used on Amiga in any useful way. | 18:53 |
| mntmn | ok | 18:57 |
| _Bnu | PC SPEAKER | 19:02 |
| apolkosnik[m] | PC LOAD LETTER | 19:04 |
| mntmn | this whole zorro stuff is mind-numbingly boring | 19:29 |
| mntmn | i will rather attempt to fiddle with autoboot | 19:29 |
| mntmn | but not today | 19:29 |
| - RobDangerous (QUIT: Ping timeout: 252 seconds) (~Thunderbi@p200300ec8f3e1c00b1df07b493a316b1.dip0.t-ipconnect.de) | 21:53 | |
| + RobDangerous (~Thunderbi@p200300ec8f3e1c00b1df07b493a316b1.dip0.t-ipconnect.de) | 22:12 | |
| - bfarmilo (QUIT: Ping timeout: 268 seconds) (~bfarmilo@202.56.40.240) | 22:18 | |
| apolkosnik[m] | Interesting that fintro-gcc is working if either data cache or instruction cache are off on my CS MK1 | 22:45 |
| apolkosnik[m] | Much higher FPS when data cache is off | 22:50 |
| + bfarmilo (~bfarmilo@202.56.40.240) | 23:54 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
