Re: [hatari-devel] Preliminary Hatari 1.7.0 OSX Build Example |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/hatari-devel Archives
]
- To: hatari-devel@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [hatari-devel] Preliminary Hatari 1.7.0 OSX Build Example
- From: David Savinkoff <dsavnkff@xxxxxxxxx>
- Date: Fri, 28 Jun 2013 12:48:29 -0600 (MDT)
- Thread-index: MsflUqZxT6ok2PW47Vro+kZ9vcPBbw==
- Thread-topic: Preliminary Hatari 1.7.0 OSX Build Example
----- Eero Tamminen <oak@xxxxxxxxxxxxxx> wrote:
> Hi,
>
> On perjantai 28 kesäkuu 2013, Jerome Vernet wrote:
> > Le 28/06/13 18:42, Eero Tamminen a écrit :
> > > Only when (if) it works as well for ST/STE as the old core.
> > > There's a long way for that. Before that somebody needs
> > > to sync WinUAE CPU core from latest WinUAE to get fully
> > > working MMU, and that sync alone is a lot of work.
> >
> > And having both one in a single build ?
>
> Sure, both core builds support about the same set of CPUs and same
> set of non-CPU features.
>
> Main differences:
> - OldUAE CPU core lacks MMU emulation and support for different
> FPU types, and it's less accurate for Falcon 030.
> - WinUAE CPU core lacks some minor emulator features, isn't
> wired as well to the ST stuff and has therefore less accurate
> ST/STe 68000 emulation. It's slower too.
>
>
> - Eero
>
I've seen and heard much about the OldUAE and WinUAE cpu
cores in discussions. Hatari may be best as several emulators.
This is becoming apparent in practical reality as Hatari evolves.
Hatari must not evolve into a 'Swiss army knife' of emulators
using #ifdef and if(Machine_Falcon) at the low level when the
solution is high level design.
David