Page 5 of 22
Re: Breakthrough for the ao486 core announced - Cache
Posted: Wed Jul 15, 2020 9:53 pm
by Chris23235
Schyz wrote: ↑Mon Jul 13, 2020 2:59 pm
I thought it was just the first Duke Nukem, but it's Duke Nukem
3D! that game is not easy to move.
Dark Forces works fine as well and I always found it the most demanding of the 2.5D FPS.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 1:07 am
by NightShadowPT
FPGAzumSpass just published the latest result of his optimizations of the core... absolutely unbelievable
- Ec_hRNGWsAQqtsE.png (68.39 KiB) Viewed 11709 times
Doom is now almost 4x faster than the current official core.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 3:10 am
by PikWik
- GV8OTtP.gif (5.16 MiB) Viewed 11708 times
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 5:22 am
by FPGAzumSpass
I have to thank Sorgelig that we can even have these improvements.
He is currently spending more time then me on this core and while his commits might not be directly visible in those graphs, his changes are very important.
E.g: The last change would not have been possible without prior work and support from him.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 6:09 am
by RealLarry
Astonishing, unbelievable and fascinating. Haven't tested this optimized core yet, but followed this thread from beginning and am glad of this co-op'ed work of you. And that's why I'm a patreon supporter.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 6:24 am
by NightShadowPT
FPGAzumSpass wrote: ↑Thu Jul 16, 2020 5:22 am
I have to thank Sorgelig that we can even have these improvements.
He is currently spending more time then me on this core and while his commits might not be directly visible in those graphs, his changes are very important.
E.g: The last change would not have been possible without prior work and support from him.
Sorgelig is the unsung hero in many of these cores, but the community is aware of the debt of gratitude we owe him.
Anyway, thank you both for making improvements to this core.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 6:28 am
by NightShadowPT
It was already reported here that these "beta" cores that use cache seem to have some problems with Doom having music.
I don't know if it helps somehow, but I have also tried the following and the problem is the same. On the official core I get both sound and music (sound blaster) while on the cache cores I get sound but no music.
- Doom
- Doom II
- Heretic
- Hexen
Whatever bug there is, seems to be transversal to all versions of the engine.
And yes, I am using the exact same versions of Autoexec.bat, Config.sys and the same virtual drive.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 6:29 am
by tontonkaloun
NightShadowPT wrote: ↑Thu Jul 16, 2020 6:28 am
It was already reported here that these "beta" cores that use cache seem to have some problems with Doom having music.
I don't know if it helps somehow, but I have also tried the following and the problem is the same. On the official core I get both sound and music (sound blaster) while on the cache cores I get sound but no music.
- Doom
- Doom II
- Heretic
- Hexen
Whatever bug there is, seems to be transversal to all versions of the engine.
And yes, I am using the
exact same versions of Autoexec.bat, Config.sys and the same virtual drive.
It's the same thing for me !
No music under dos
music with win95
sound problem with other games like duke3d , Day of the Tentacle ...
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 6:49 am
by Jayde Six
To Add to the info I've got sound issues with Dark Forces and Warcraft, where the audio starts getting really choppy/distorted after like 10 mins or so.
I checked and Warcraft did it with the release core too so some sound issues (outside of the id games music issue) might not be related to these betas but weren't noticed yet because the games were unplayable before.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 10:09 am
by Chris23235
Didn't tried Warcraft, but so far I had no issues with Dark Forces, I played an hour or so.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 10:32 am
by kathleen
@Jayde Six,
I'm maybe wrong but so far I remember, AO486 is quite demanding in terms of power. Could a non strong enough power supply cause this issue ?
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 10:48 am
by Chris23235
With the ao486_Cache13_9345e87 version several new issues occured for me: I am not able to get expanded memory to work, Windows 95 gives me a protection error on startup and I get no music in X-Wing anymore.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 10:58 am
by Xevian
Aye, Cache13 98SE boots okay for me, but do get a lot of exceptions opening Exploer.. But MIDI is working! Still struggles a little, but I think that's more down to running MUNT on the ARM side of the Mister than anything!
https://youtu.be/W3YE3SZm3TM
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 11:37 am
by Jayde Six
Chris23235 wrote: ↑Thu Jul 16, 2020 10:09 am
Didn't tried Warcraft, but so far I had no issues with Dark Forces, I played an hour or so.
kathleen wrote: ↑Thu Jul 16, 2020 10:32 am
@Jayde Six,
I'm maybe wrong but so far I remember, AO486 is quite demanding in terms of power. Could a non strong enough power supply cause this issue ?
Hmmm, looks like I'm going to have to try to troubleshoot some stuff. Power supply is a good suggestion, I'll try removing the bluetooth and wifi adapters to see if it makes a difference.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 2:18 pm
by retrorepair
Would a core really put a strain on power requirements? IIRC @Sorgelig said fpga usage doesn't affect temperature which tells me it also doesn't affect power consumption.
It sounds more like something is losing sync over time.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 2:50 pm
by robng15
Chris23235 wrote: ↑Thu Jul 16, 2020 10:48 am
With the ao486_Cache13_9345e87 version several new issues occured for me: I am not able to get expanded memory to work, Windows 95 gives me a protection error on startup and I get no music in X-Wing anymore.
TRY in CONFIG.SYS
Code: Select all
DEVICE=C:\DOS\EMM386.EXE NOEMS X=A000-C7FF I=D000-EFFF
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 3:32 pm
by luishg
There is something weird with the SoundBlaster or sound implementation on the core.
For example, once you set the right SB parameters, you can enable the music for Supaplex, but at the moment you try to enable the sound effects, it halts the machine.
- jQHsZQW.png (537.47 KiB) Viewed 11734 times
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 5:42 pm
by Goingdown
luishg wrote: ↑Thu Jul 16, 2020 3:32 pm
For example, once you set the right SB parameters, you can enable the music for Supaplex, but at the moment you try to enable the sound effects, it halts the machine.
Supaplex has known issue with some timings related to soundblaster on real hardware too... There is fix / patch available somewhere.
EDIT. I think you may find if from here:
https://www.elmerproductions.com/sp/faq.html
There is also speedfix available, but I haven't tried it. I only used sndfix1.zip
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 7:15 pm
by Chris23235
robng15 wrote: ↑Thu Jul 16, 2020 2:50 pm
Chris23235 wrote: ↑Thu Jul 16, 2020 10:48 am
With the ao486_Cache13_9345e87 version several new issues occured for me: I am not able to get expanded memory to work, Windows 95 gives me a protection error on startup and I get no music in X-Wing anymore.
TRY in CONFIG.SYS
Code: Select all
DEVICE=C:\DOS\EMM386.EXE NOEMS X=A000-C7FF I=D000-EFFF
Very strange, your config.sys entry works, but now my old config.sys also works, I tried it today again and again and always ran into error messages. Win95 now boots fine also.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 7:44 pm
by SuperBabyHix
Goingdown's avatar inspired me to check, and yes, Ultima Underworld works very well with both Sound Blaster and MT-32 music. Gameplay is pretty smooth at very high detail.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 7:59 pm
by Schyz
With the speed achieved thanks to the cache there will be games that start having
compatibility issues. I wonder how hard it can be to allow for a slower performance, maybe offering different frequencies or with an option to disable caches. Is this possible in FPGA?
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 8:15 pm
by breiztiger
Core have clock option in osd
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 8:25 pm
by tontonkaloun
yes it was really an option that I expected with the increase in the speed of the core ...
A big thank you for the job and the sharing of the file!
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 8:41 pm
by Chris23235
The clock option doesn't help in some cases, Wing Commander or parts of the CD version of X-Wing now run much to fast. I don't think it is because of the general performance of the core, but because some parts of the CPU now work much faster than anything back in the day.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Thu Jul 16, 2020 11:06 pm
by thorr
Wing Commander doesn't have a software throttle built in. It was too fast on newer computers back in the day. Back then, I felt it was perfect on my 386DX33 and too fast on 486 computers. There are slowdown programs that can be used in DOS to slow it down.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Fri Jul 17, 2020 2:20 am
by Sbeehive
This is fantastic news!
Are we able to test the latest build with the L1 cache enabled? I have some vacation time and am dying to run some tests.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Fri Jul 17, 2020 2:31 am
by Jayde Six
Unplugging everything but keyboard had no effect. However Dark Forces audio glitches seem to be user error on my part, I was consolidating a couple vhds and copied a dark forces folder over that hadn't had the soundblaster settings set correctly. With Warcraft who knows, it has the same audio issues on the main release core too so it may just not like this game yet. In addition others were not able to get older revisions of the game to run at all (mine is using the latest patch) so it could be an issue with that too.
Re: Breakthrough for the ao486 core announced - Cache
Posted: Fri Jul 17, 2020 5:50 am
by lroby74
thorr wrote: ↑Thu Jul 16, 2020 11:06 pm
Wing Commander doesn't have a software throttle built in. It was too fast on newer computers back in the day. Back then, I felt it was perfect on my 386DX33 and too fast on 486 computers. There are slowdown programs that can be used in DOS to slow it down.
Keep using original Ao486 no cached core for games like this?
Re: Breakthrough for the ao486 core announced - Cache
Posted: Fri Jul 17, 2020 6:07 am
by lroby74
Chris23235 wrote: ↑Thu Jul 16, 2020 8:41 pm
The clock option doesn't help in some cases, Wing Commander or parts of the CD version of X-Wing now run much to fast. I don't think it is because of the general performance of the core, but because some parts of the CPU now work much faster than anything back in the day.
Try to set 15mhz clock and reset core..
It will become very very slow and maybe that WC will run at nice speed
Re: Breakthrough for the ao486 core announced - Cache
Posted: Sat Jul 18, 2020 6:54 am
by Shaneus
Can anyone tell me if I'm missing something re building the latest dev commit? I download it and it builds in Quartus, but there's no RBF file. There's a SOF file which I think can be converted, but it says it's time-limited or something.
Apologies if it's been brought up before, I hadn't noticed anything of it earlier.
Edit: Ignore, it was addressed earlier. Sorry for wasting everyone's time!