Breakthrough for the ao486 core announced - Cache

JimBobMcGee
Posts: 12
Joined: Thu Jul 30, 2020 6:16 am
Been thanked: 2 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by JimBobMcGee »

ZigZag wrote: Sun Aug 09, 2020 1:28 pm With the new MiSTer files for cache 41 my monitor says "signal out of range" (set at stock 800x600@60). Even main menu isn't displaying.
You might have to set AO486 to variable sync. I did that for mine and it worked perfectly after that.

To the previous posts about overwriting the Main Mister file, that also fixed my booting errors.
jordi
Posts: 238
Joined: Thu Jun 11, 2020 10:11 am
Has thanked: 95 times
Been thanked: 80 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by jordi »

Q47 wrote: Sat Aug 08, 2020 9:23 pm
jordi wrote: Sat Aug 08, 2020 9:00 pm I've seen this strange scroll artifacts on 2d games

See right part of the screen:
https://www.youtube.com/watch?v=fVMoYMJb3hk

It's strange as it only happens on horizontal scrolling.
This is just hw/coding limitation of that era. Video under explains this on NES. Why you see it is simply because your screen doesn't do overscan anymore as CRTs did back in the day.

https://www.youtube.com/watch?v=wfrNnwJrujw
oh that's strange, but you could see any walkthrough and the effect might appear but less visible.
It's like in your mario video, it has an effect but it is almost unnoticeable.
https://www.youtube.com/watch?v=iKHUOKVzo0Q

Then it might be interesting to be able to cut the right part of the screen.
ZigZag
Posts: 99
Joined: Sun Jun 28, 2020 7:05 am
Has thanked: 1 time
Been thanked: 5 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by ZigZag »

The "out of sync" issue with cache 41 wasn't limited to ao486 core - the actual menu screen wasn't showing after turning the MiSTer on so I couldn't see anything to actually access ao486 or anything else! Changing back to the old MiSTer file sorts the menu out, but cache 41 complains it can't see any boot devices.

EDIT: Oh yeah, forgot to say I have latest bios files. Thanks for the replies.
User avatar
tontonkaloun
Posts: 354
Joined: Sun May 24, 2020 7:38 pm
Has thanked: 152 times
Been thanked: 52 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by tontonkaloun »

ZigZag wrote: Sun Aug 09, 2020 8:40 pm The "out of sync" issue with cache 41 wasn't limited to ao486 core - the actual menu screen wasn't showing after turning the MiSTer on so I couldn't see anything to actually access ao486 or anything else! Changing back to the old MiSTer file sorts the menu out, but cache 41 complains it can't see any boot devices.

EDIT: Oh yeah, forgot to say I have latest bios files. Thanks for the replies.

For me, you have not put the correct main Mister file on your SD, it is in the archive with the core file.
The black screen is that it is missing or corrupt.
The Core that does not boot to HD is that it is not the correct version.
ZigZag
Posts: 99
Joined: Sun Jun 28, 2020 7:05 am
Has thanked: 1 time
Been thanked: 5 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by ZigZag »

tontonkaloun wrote: Sun Aug 09, 2020 9:05 pm
ZigZag wrote: Sun Aug 09, 2020 8:40 pm The "out of sync" issue with cache 41 wasn't limited to ao486 core - the actual menu screen wasn't showing after turning the MiSTer on so I couldn't see anything to actually access ao486 or anything else! Changing back to the old MiSTer file sorts the menu out, but cache 41 complains it can't see any boot devices.

EDIT: Oh yeah, forgot to say I have latest bios files. Thanks for the replies.

For me, you have not put the correct main Mister file on your SD, it is in the archive with the core file.
The black screen is that it is missing or corrupt.
The Core that does not boot to HD is that it is not the correct version.
Yep, should've trusted myself, I'm quite familiar with this kind of thing. Just got out of my bed to triple check I hadn't made any errors (like the OCD twat that I am). I extracted the files again, copied the MiSTer file from the archive over again (having renamed the old one as backup - again). No menu screen, input out if sync. Had to remote connect to mister & revert to the backup (so I guess its just screen output that's not working, at least when set to 800x600@60). No idea if the menu works with other screen resolutions, might try that tomorrow (later today I mean).
User avatar
tontonkaloun
Posts: 354
Joined: Sun May 24, 2020 7:38 pm
Has thanked: 152 times
Been thanked: 52 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by tontonkaloun »

ZigZag wrote: Mon Aug 10, 2020 1:31 am
tontonkaloun wrote: Sun Aug 09, 2020 9:05 pm
ZigZag wrote: Sun Aug 09, 2020 8:40 pm The "out of sync" issue with cache 41 wasn't limited to ao486 core - the actual menu screen wasn't showing after turning the MiSTer on so I couldn't see anything to actually access ao486 or anything else! Changing back to the old MiSTer file sorts the menu out, but cache 41 complains it can't see any boot devices.

EDIT: Oh yeah, forgot to say I have latest bios files. Thanks for the replies.

For me, you have not put the correct main Mister file on your SD, it is in the archive with the core file.
The black screen is that it is missing or corrupt.
The Core that does not boot to HD is that it is not the correct version.
Yep, should've trusted myself, I'm quite familiar with this kind of thing. Just got out of my bed to triple check I hadn't made any errors (like the OCD twat that I am). I extracted the files again, copied the MiSTer file from the archive over again (having renamed the old one as backup - again). No menu screen, input out if sync. Had to remote connect to mister & revert to the backup (so I guess its just screen output that's not working, at least when set to 800x600@60). No idea if the menu works with other screen resolutions, might try that tomorrow (later today I mean).
To follow your idea, I tested your resolution and it works for me
We are all the same, I would have got up too !! :lol:
ZigZag
Posts: 99
Joined: Sun Jun 28, 2020 7:05 am
Has thanked: 1 time
Been thanked: 5 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by ZigZag »

Maybe it's something to do with me having activated the VGA scaler in order to be able to use filters. If this had the internal filters I'd just use HQ2X. This monitor is prety damn robust when it comes to displaying signals (I did have it set for 1600x1200@60 over VGA but the ao486 core looks nicer with 800x600). It's the HP LP2065 monitor. Unfortunately my HDMI is bust or I'd try it on the TV. I still can using VGA but it means arseing around with the cables! I'll just stick to cache 37 for now since I managed to install 98lite on it and it wasn't working properly in cache 40. Hopefully whatever changed that is causing the MiSTer to mess with my monitor sync will get fixed in a later release.
SavageX
Posts: 5
Joined: Thu Jul 30, 2020 7:20 pm

Re: Breakthrough for the ao486 core announced - Cache

Unread post by SavageX »

Hmmm... for some people Cache 37 seems to be more stable than later iterations (at least judging from e.g. viewtopic.php?f=13&t=918 saying "(cache 37 tested -40 failed)"). I think versions beyond Cache 37 are including the last round of Cache performance tweaks and I wonder if that has interesting side-effects...
IAmParadox
Posts: 59
Joined: Fri Jun 26, 2020 6:48 am
Been thanked: 2 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by IAmParadox »

ZigZag wrote: Mon Aug 10, 2020 7:45 am Maybe it's something to do with me having activated the VGA scaler in order to be able to use filters. If this had the internal filters I'd just use HQ2X. This monitor is prety damn robust when it comes to displaying signals (I did have it set for 1600x1200@60 over VGA but the ao486 core looks nicer with 800x600). It's the HP LP2065 monitor. Unfortunately my HDMI is bust or I'd try it on the TV. I still can using VGA but it means arseing around with the cables! I'll just stick to cache 37 for now since I managed to install 98lite on it and it wasn't working properly in cache 40. Hopefully whatever changed that is causing the MiSTer to mess with my monitor sync will get fixed in a later release.
Mister.ini
vsync_adjust=0

worked for me.
User avatar
Caldor
Top Contributor
Posts: 930
Joined: Sat Jul 25, 2020 11:20 am
Has thanked: 112 times
Been thanked: 111 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by Caldor »

I made a build of the latest updates:
ao486_caldor_cache42b.7z
(2.05 MiB) Downloaded 171 times
The MiSTer main got updated with the Sorgelig comment: ao486: fix midi flag. (its currently 8 hours old)

AO486 got updated with these comments:
temp commit with low slack. - 18 hours ago
uart: fixes in mpu mode. - 12 hours ago
uart: don't use fifo in mpu mode. - 10 hours ago
Move 100MHz to debug menu. - 10 hours ago

I think Sorg is getting ready to make an official release soon. Much of this seems to be fixing and polishing minor things.

A feature I think could be nice to see, would be an option to support two GPU bioses. Mainly because Game Wizard 32 Pro works with the original VGA bios that the release is using, but not with the new VGA bios on the dev branch or the new SVGA/VESA bios. I have not found any other reason to sometimes want to use the old VGA bios, but since its the case with GW32Pro, I am guessing it might also be the case with other things.
AntialiasedPixel
Posts: 44
Joined: Mon May 25, 2020 2:17 am
Been thanked: 2 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by AntialiasedPixel »

Caldor wrote: Tue Aug 11, 2020 10:49 pm I made a build of the latest updates:
ao486_caldor_cache42b.7z
I can't seem to get this one to boot up, but I haven't tried the beta core since one of the early ones. I've tried booting from floppy images, various vhds (including the freedos 1.3 one), swapping the floppy/hdd boot order and it always finds no boot devices. Any ideas on what I could be missing?
tlaloc
Posts: 26
Joined: Wed May 27, 2020 6:59 pm
Has thanked: 1 time

Re: Breakthrough for the ao486 core announced - Cache

Unread post by tlaloc »

AntialiasedPixel wrote: Wed Aug 12, 2020 2:38 am
Caldor wrote: Tue Aug 11, 2020 10:49 pm I made a build of the latest updates:
ao486_caldor_cache42b.7z
I can't seem to get this one to boot up, but I haven't tried the beta core since one of the early ones. I've tried booting from floppy images, various vhds (including the freedos 1.3 one), swapping the floppy/hdd boot order and it always finds no boot devices. Any ideas on what I could be missing?
It looks like the last two commits are getting strange errors related to UART, I noticed MIDI wasn't switching between MUNT & FluidSynth as well the last functioning commit (i.e. 'ao486_Cache41_9c63fe4.zip' [Add simulation framework for modelsim for both caches and cpu test])

The people that know better mention it in this 'Why OpenCore Plus UART?' thread:
viewtopic.php?p=7691#p7691

For now, if you want to hear really funky MIDI sounds coming out the MiSTer, download either of the two most recent commits and switch between MUNT and Fluidsynth... making sure to toggle between the "UART BAUD' speed of 31250 and the usual 38400 that the [now outdated?] MIDI-Link guide recommends specifically for ao486 (conflicting with a recent commit 06092fc, which I believe says we are to now use 31250 like the one recommended for the minimig - as mentioned in the below guide?):
https://github.com/MiSTer-devel/MidiLink_MiSTer


So yeah, if you want funky MIDI thrown in dissarray, pickup the following two most recent commits 100432e & cf9e20f:
ao486_Cache42_100432e.zip
ao486_Cache43_cf9e20f.zip

Otherwise, stick with the tried and true commit 9c63fe4/'ao486_Cache41_9c63fe4.zip' to be able to switch it up without hearing the MIDI all mixed up, which can be fun for a minute ;-)
User avatar
Caldor
Top Contributor
Posts: 930
Joined: Sat Jul 25, 2020 11:20 am
Has thanked: 112 times
Been thanked: 111 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by Caldor »

AntialiasedPixel wrote: Wed Aug 12, 2020 2:38 am
Caldor wrote: Tue Aug 11, 2020 10:49 pm I made a build of the latest updates:
ao486_caldor_cache42b.7z
I can't seem to get this one to boot up, but I haven't tried the beta core since one of the early ones. I've tried booting from floppy images, various vhds (including the freedos 1.3 one), swapping the floppy/hdd boot order and it always finds no boot devices. Any ideas on what I could be missing?
Boots for me, but remember to use the bioses, the new mister file and the rbf core file.
tlaloc wrote: Wed Aug 12, 2020 7:45 am
AntialiasedPixel wrote: Wed Aug 12, 2020 2:38 am
Caldor wrote: Tue Aug 11, 2020 10:49 pm I made a build of the latest updates:
ao486_caldor_cache42b.7z
I can't seem to get this one to boot up, but I haven't tried the beta core since one of the early ones. I've tried booting from floppy images, various vhds (including the freedos 1.3 one), swapping the floppy/hdd boot order and it always finds no boot devices. Any ideas on what I could be missing?
It looks like the last two commits are getting strange errors related to UART, I noticed MIDI wasn't switching between MUNT & FluidSynth as well the last functioning commit (i.e. 'ao486_Cache41_9c63fe4.zip' [Add simulation framework for modelsim for both caches and cpu test])

The people that know better mention it in this 'Why OpenCore Plus UART?' thread:
viewtopic.php?p=7691#p7691

For now, if you want to hear really funky MIDI sounds coming out the MiSTer, download either of the two most recent commits and switch between MUNT and Fluidsynth... making sure to toggle between the "UART BAUD' speed of 31250 and the usual 38400 that the [now outdated?] MIDI-Link guide recommends specifically for ao486 (conflicting with a recent commit 06092fc, which I believe says we are to now use 31250 like the one recommended for the minimig - as mentioned in the below guide?):
https://github.com/MiSTer-devel/MidiLink_MiSTer


So yeah, if you want funky MIDI thrown in dissarray, pickup the following two most recent commits 100432e & cf9e20f:
ao486_Cache42_100432e.zip
ao486_Cache43_cf9e20f.zip

Otherwise, stick with the tried and true commit 9c63fe4/'ao486_Cache41_9c63fe4.zip' to be able to switch it up without hearing the MIDI all mixed up, which can be fun for a minute ;-)
Cache43 was released after my post, so it most likely has whatever I have put in my zip and more.
jordi
Posts: 238
Joined: Thu Jun 11, 2020 10:11 am
Has thanked: 95 times
Been thanked: 80 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by jordi »

cache43 looks to have broken keyboard input. Doing a 'keyb sp' does not fix it.
I didn't updated again the Mister file, I'm not sure if it's required again.
AntialiasedPixel
Posts: 44
Joined: Mon May 25, 2020 2:17 am
Been thanked: 2 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by AntialiasedPixel »

Found the issue I was having. Switched ftp apps recently and in the config it had a setting to transfer files with no extension as ascii. Used Binary for "MiSTer" file and it's working now.
tlaloc
Posts: 26
Joined: Wed May 27, 2020 6:59 pm
Has thanked: 1 time

Re: Breakthrough for the ao486 core announced - Cache

Unread post by tlaloc »

How does one go about dismounting a drive in the ao486 core, btw?

I've just been switching to non bootables as a workaround, any guidance would be welcomed.
flynnsbit
Top Contributor
Posts: 552
Joined: Sun May 24, 2020 8:07 pm
Has thanked: 183 times
Been thanked: 308 times
Contact:

Re: Breakthrough for the ao486 core announced - Cache

Unread post by flynnsbit »

open the mounting screen for it and then hit backspace. It probably needs a hint on that screen to hit backspace to unmount. Let me see what I can do.
User avatar
Goingdown
Posts: 38
Joined: Mon May 25, 2020 6:58 am
Has thanked: 5 times
Been thanked: 4 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by Goingdown »

tlaloc wrote: Wed Aug 12, 2020 6:16 pm How does one go about dismounting a drive in the ao486 core, btw
In the disk selection screen, press backspace.

It is same in all cores.

Edit. I am too slow
tlaloc
Posts: 26
Joined: Wed May 27, 2020 6:59 pm
Has thanked: 1 time

Re: Breakthrough for the ao486 core announced - Cache

Unread post by tlaloc »

flynnsbit wrote: Wed Aug 12, 2020 6:19 pm open the mounting screen for it and then hit backspace. It probably needs a hint on that screen to hit backspace to unmount. Let me see what I can do.
Ah, makes sense. Thank You for that bit of insight.
I vaguely remember reading (or perhaps, hearing) that on one of the many tutorials I've scoured.
It would be nice to see some scrolling text 'courtesy reminder on how to dismount' added to the mount menus globally, as it isn't readily apparent what to do in order to dismount when in that screen.

Once again, thank you for your time and consideration.
User avatar
Caldor
Top Contributor
Posts: 930
Joined: Sat Jul 25, 2020 11:20 am
Has thanked: 112 times
Been thanked: 111 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by Caldor »

tlaloc wrote: Wed Aug 12, 2020 9:44 pm
flynnsbit wrote: Wed Aug 12, 2020 6:19 pm open the mounting screen for it and then hit backspace. It probably needs a hint on that screen to hit backspace to unmount. Let me see what I can do.
Ah, makes sense. Thank You for that bit of insight.
I vaguely remember reading (or perhaps, hearing) that on one of the many tutorials I've scoured.
It would be nice to see some scrolling text 'courtesy reminder on how to dismount' added to the mount menus globally, as it isn't readily apparent what to do in order to dismount when in that screen.

Once again, thank you for your time and consideration.
Yeah, I was thinking this as well. Its pretty obscure right now, being something you have to find out reading the wiki or asking on forums. I tried reading all that scrolling text and it was not mentioned at all.
User avatar
kathleen
Top Contributor
Posts: 414
Joined: Fri Jun 26, 2020 4:23 am
Location: Belgium
Has thanked: 226 times
Been thanked: 136 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by kathleen »

Was mentioned here by Cebion the 26th of July : viewtopic.php?f=13&t=680&p=6436&hilit=backspace#p6436

かすりん

User avatar
Caldor
Top Contributor
Posts: 930
Joined: Sat Jul 25, 2020 11:20 am
Has thanked: 112 times
Been thanked: 111 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by Caldor »

kathleen wrote: Thu Aug 13, 2020 4:20 am Was mentioned here by Cebion the 26th of July : viewtopic.php?f=13&t=680&p=6436&hilit=backspace#p6436
Yes, that is how I found out. Still a lot more helpful if its something that is explained in the menu, rather than on some forums.
User avatar
jrronimo
Posts: 79
Joined: Sat Jul 04, 2020 6:03 pm
Has thanked: 33 times
Been thanked: 11 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by jrronimo »

Caldor wrote: Thu Aug 13, 2020 12:19 pm
kathleen wrote: Thu Aug 13, 2020 4:20 am Was mentioned here by Cebion the 26th of July : viewtopic.php?f=13&t=680&p=6436&hilit=backspace#p6436
Yes, that is how I found out. Still a lot more helpful if its something that is explained in the menu, rather than on some forums.
Especially on page 10 of a 22 page thread! 🤣 It could be as simple as, in the Win+F12 menu, saying "Press Backspace to Unmount". Seems like that would be useful across all cores.
thorr
Top Contributor
Posts: 1136
Joined: Mon Jul 06, 2020 9:37 pm
Has thanked: 562 times
Been thanked: 259 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by thorr »

I finally got around to trying this. I have been waiting because so many improvements have been coming and I wanted to let it mature a bit before taking the plunge. Well, I have to say that I am thoroughly impressed! A great benchmark for me is the Future Crew Second Reality demo. I played this demo so many times on my 386DX33, my friend's 486DX33 and eventually my 486DX2-66 although my memory of it on the DX2-66 is foggy. On MiSTer the performance is for sure better than the 486DX33, and maybe even better than the 486DX2-66. The way to tell is how far the sword comes out and how fast the screen spins during the piano part of the song. I don't remember the sword ever coming all the way out and flying off the screen, but it does on the MiSTer, and the spinning part is faster than I ever remember seeing it.
A few things I noticed that don't work right for some reason:
- I tried Doom the shareware version and the right mouse button is supposed to move you forward, but it is strafing instead.
- The Future Crew demo called "Panic" crashes near the beginning. I remember this demo being finicky even back in the day. It might be the EMM386 I have loaded causing it. (Edit: I read the readme for it and it suggested trying different sound settings, and turning off the sound makes it not crash, but there is no sound of course).
- The Incredible Machine 2 demo doesn't have any picture but I can hear the music.
- When I select 60Hz in the menu, my VGA monitor goes out of sync range. On Variable, it works fine but the screen shakes a little bit.

This weekend I am going to build my MT-32 and hook up my SC-55 MK2. Can't wait to get some great games going. Thanks for the great work!
ZigZag
Posts: 99
Joined: Sun Jun 28, 2020 7:05 am
Has thanked: 1 time
Been thanked: 5 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by ZigZag »

Glad to see I'm not the only person with monitor sync issues. I find cache version 40 runs fine for me using normal MiSTer file if set at 90mhz. Anything later than that I can't get to work because of monitor synch issues with the MiSTer file. That includes other cores & even main menu, tested with multiple resolutions & vsync/scaling options but with VGA output only as my HDMI is fried. I've ordered a new DE10 Nano so I was hoping that'd solve some of the sync issues (if the HDMI got fatally zapped by static electricity who knows what other weird damage could've been done?).
Schyz
Posts: 63
Joined: Sun Jul 12, 2020 4:53 pm
Has thanked: 18 times
Been thanked: 13 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by Schyz »

ZigZag wrote: Fri Aug 14, 2020 9:12 am Glad to see I'm not the only person with monitor sync issues.
I'm using a PC CRT monitor with my MiSTer, and these are my findings:
- VGA Scaler set to "on" solves a lot of my problems
- Changes on speed will cause issues unless I restart the core, so every time I want a different speed I set it to that speed, save settings as default, exit the core back to the main menu and then start the core again.
- 15 MHz still causes the monitor to report wrong frequencies, but 30, 56 and 90 work fine.
- As a rule of thumb, I would recommend to restart the core as described above for any config change, not only speed, for stability.
User avatar
NightShadowPT
Posts: 208
Joined: Mon May 25, 2020 9:56 am
Has thanked: 5 times
Been thanked: 9 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by NightShadowPT »

Where have you guys downloaded the compiled core from? I'm looking forward to try the new improvements, but don't have the tools to compile the core myself...
User avatar
SuperBabyHix
Posts: 152
Joined: Sun May 24, 2020 8:26 pm
Has thanked: 44 times
Been thanked: 34 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by SuperBabyHix »

NightShadowPT wrote: Fri Aug 14, 2020 11:00 am Where have you guys downloaded the compiled core from? I'm looking forward to try the new improvements, but don't have the tools to compile the core myself...
First page of this thread. viewtopic.php?p=5058#p5058
User avatar
NightShadowPT
Posts: 208
Joined: Mon May 25, 2020 9:56 am
Has thanked: 5 times
Been thanked: 9 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by NightShadowPT »

SuperBabyHix wrote: Fri Aug 14, 2020 12:05 pm First page of this thread. viewtopic.php?p=5058#p5058
:lol: :lol: :lol:

Didn't notice the post was being updated ... silly me. Thanks
User avatar
tontonkaloun
Posts: 354
Joined: Sun May 24, 2020 7:38 pm
Has thanked: 152 times
Been thanked: 52 times

Re: Breakthrough for the ao486 core announced - Cache

Unread post by tontonkaloun »

the newest:
Attachments
ao486_20200814.rar
(1.78 MiB) Downloaded 186 times
Post Reply